Locus Bonus: The Relationship of the Roman Villa to its Environment in the Vicinity of Rome

Locus Bonus The RelaTionship of The Roman Villa To iTs enViRonmenT in The ViciniTy of Rome EEva-Maria viitanEn acadEMic dissErtation to bE publicly discussEd, by duE pErMission of thE faculty of arts at thE univErsity of hElsinki in auditoriuM Xv, on thE 2nd of octobEr, 2010 at 10 o’clock hElsinki 2010 © Eeva-Maria Viitanen ISBN 978-952-92-7923-4 (nid.) ISBN 978-952-10-6450-0 (PDF) PDF version available at: http://ethesis.helsinki.i/ Helsinki University Print Helsinki, 2010 Cover: photo by Eeva-Maria Viitanen, illustration Jaana Mellanen contEnts abstract iii acknowlEdgEMEnts v list of figurEs, tablEs and platEs vii 1 studying thE roMan villa and its EnvironMEnt 1 1 3 6 6 8 10 11 1.1 introduction 1.2 dEfining thE villa 1.3 thE roMan villa in classical studiEs Origin and Development of the Villa Villa Typologies Role of the Villa in the Historical Studies 1.4 thEorEtical and MEthodological considErations 2 archaEological MatErial and writtEn sourcEs 2.1 rEsEarch history of thE roMan caMpagna 2.2 fiEldwork MEthodology Excavation Survey 2.3 archaEological MatErial Settlement Sites from Surveys and Excavations The Sites Reclassiied Chronological Considerations 2.4 writtEn sourcEs Ancient Literature Inscriptions 2.5 conclusions 3 gEology and roMan villas 3.1 background 3.2 gEology of thE roMan caMpagna 3.3 thE changing landscapE of thE roMan caMpagna 3.4 writtEn sourcEs for thE usE of gEological rEsourcEs 3.5 archaEology of building MatErials 3.6 intEgrating thE EvidEncE Avoiding the alluvium? Favoring boundary zones? Looking for building stone? 3.7 conclusions 4 soils and roMan villas 4.1 background 4.2 soils of thE roMan caMpagna 4.3 land Evaluation of soils for anciEnt agricultural purposEs 4.4 writtEn sourcEs for agriculturE 4.5 archaEological EvidEncE for agriculturE 4.6 palynological EvidEncE for agricultural activitiEs 4.7 intEgrating thE EvidEncE Site type distribution and quality of soils Changes in site selection over time? Changes in types of production? 4.8 conclusions 15 15 18 18 19 21 21 25 28 33 33 35 37 38 38 40 42 44 47 50 50 51 53 54 55 55 56 59 62 69 72 73 73 77 79 82 i 5 watEr and roMan villas 5.1 background 5.2 watEr rEsourcEs in thE roMan caMpagna 5.3 writtEn sourcEs on thE watEr rEsourcEs and watEr usE 5.4 archaEological EvidEncE for watEr rEsourcEs and watEr usE 5.5 intEgrating thE EvidEncE Healthy site equals high and dry site? Malaria? Water supply for the villas? Water for irrigation and display? 5.6 conclusions 6 tErrain and roMan villas 6.1 background 6.2 landscapE and tErrain in thE roMan caMpagna 6.3 writtEn sourcEs for landscapE and tErrain 6.4 archaEological EvidEncE for sitE sElEction and tErrain typEs 6.5 intEgrating thE EvidEncE Mid-slope on an eminence? Aspect and orientation 6.6 conclusions 7 visibility, viEwability and roMan villas 7.1 background 7.2 visibility and viEwability in thE roMan caMpagna 7.3 writtEn sourcEs for visibility and viEwability 7.4 archaEological EvidEncE for sitE sElEction, visibility and viEwability 7.5 intEgrating thE EvidEncE The “open” villa? Experiencing the view? The viewable villa? 7.6 conclusions 8 roads, towns, villagEs and roMan villas 8.1 background 8.2 writtEn sourcEs on habitation cEntErs and transportation 8.3 archaEological EvidEncE for roads, villagEs and towns 8.4 intEgrating thE EvidEncE Near a good road, but not directly on one? Centuriations? Towns, villages and road stations? Neighborhoods and communities? 8.5 conclusions 9 finding thE idEal location Modelling the ideal location Villas in ideal locations 83 83 84 87 94 100 100 104 105 106 107 108 108 110 112 114 118 118 120 123 124 124 126 128 130 134 134 136 138 142 143 143 145 148 151 151 153 155 158 161 162 162 164 10 conclusion 167 appEndiX i sitE cataloguE 169 appEndiX ii tablE of datEd sitEs 195 appEndiX iii rEMains rElatEd to agricultural production 216 appEndiX iv watEr installations 222 bibliography 234 platEs 249 ii abstract EEva-Maria viitanEn: locus bonus – thE rElationship of thE roMan villa to its EnvironMEnt in thE vicinity of roME The aim of this study is to examine the relationship of the ancient Roman villa to its environment. The villa was an important feature of the Roman countryside intended both for agricultural production and for leisure from the 2nd century BC onwards. It has also often been treated in Roman literature; manuals of Roman agriculture give explicit instructions on how to select an ideal location for an estate as well as where to place the buildings. The ideal location was a moderate slope facing east or south in a healthy area and good neighborhood, near good water resources and fertile soils. A road or a navigable river or the sea was needed for transportation of produce and for reaching the estate easily. In addition, a market for selling the produce, a town or a village, should have been nearby. These recommendations are often cited in research literature, but have never really been examined in detail. The research area is ca. 500 km² in the surroundings of the city of Rome, a key area for the development of the villa. The materials used consist of archaeological settlement sites collected from published survey reports, literary and epigraphical evidence as well as environmental data. The sites include villas as well as all other settlement sites from the 7th century BC to 5th century AD to examine development and changes in the tradition of site selection. Geographical Information Systems were used to analyze the archaeological and environmental data based on the hypotheses derived from the written sources. Six aspects of location are examined: geology, soils, water resources, terrain, visibility/ viewability and relationship to roads and habitation centers. Geology is not explicitly mentioned as a criterion for site selection by the Roman authors, but it was important for inding building materials as well as for the stability of the buildings. The analyses show that the large villas established in the 2nd century BC tend to be located close to sources of building stones. Fertile soils were, on the other hand, very important. The productive areas were sought even in the period of the densest settlement in the 1st century AD. Even the poorest zones were used intensively during the same period. The Roman region is rich in water, both rainfall and groundwater. Water was also important for pleasure as well as for display of wealth and status. Excessive water, however, could be a problem and the research area was riddled with malaria in the early modern period. Although the Roman region may already have been troubled by malaria in ancient times, the settlement density testiies that the disease could have not been as devastating as it was later. A certain kind of terrain was sought over very long periods, showing the strength and the success of the tradition. A small spur or ridge shoulder with an open area in front of the site was selected for most buildings. Slopes facing east or south are not very common in the area, but the southern ones were sought. The eastern aspect might even have been avoided in order to catch heat-reducing breezes in the summer and to miss the cold winter winds blowing from the east. Visibility was studied both as the opportunity for views from the site as well as the viewability of the building in its surroundings. The most popular villa resorts in the region are located on the slopes visible from Rome as well as from almost all other parts of the research area. A villa visibly part of a high society resort served the social and political aspirations of the owner. Being in the villa and views both internal and external created the sense of isolation and privacy. The area has a very dense road network ensuring good connectivity from almost anywhere in the region. The area of best visibility/viewability, dense settlement and most burials by roads coincide, creating a neighborhood for the area. When the various qualities are combined, the ideal locations featuring the most qualities cover nearly a quarter of the research area and more than half of all the settlement sites are located in them. The ideal location was based on centuries of practical experience and rationalized by the literary tradition. iii abstrakti EEva-Maria viitanEn: locus bonus – rooMalaisEn villan suhdE yMpäristöönsä rooMaa yMpäröivällä MaasEudulla Tutkimuksen tavoitteena on analysoida roomalaisen villan suhdetta ympäristöönsä. Villa oli roomalaisen maaseutuelämän keskiössä sekä maatalouden että virkistäytymisen kannalta 100-luvulta eKr. alkaen. Roomalaiset maatalousoppaat antavat ohjeita sekä maatilan valitsemiseen että rakennusten sijoittamiseen. Ihanteellinen paikka oli loiva etelän- tai idänpuoleinen rinne terveellisellä alueella, hyvässä naapurustossa. Tie, purjehduskelpoinen joki tai meri tarvittiin tuotteiden kuljettamiseen sekä tilalle kulkemiseen. Lisäksi maatilan tuli sijaita kaupungin tai kylän lähellä, jotta tuotteita voitiin myydä. Nämä ohjeet mainitaan usein tutukimuskirjallisuudessa, mutta niitä ei ole koskaan tutkittu yksityiskohtaisesti. Tutkimusalue kattaa noin 500 km2 Rooman kaupunkia ympäröivää maaseutua, joka oli keskeinen alue villan kehityksen kannalta. Aineisto koostuu arkeologisten inventointien julkaisuista kerätyistä asuinpaikoista, antiikin kirjallisuudesta ja piirtokirjoituksista sekä ympäristöä käsittelevästä tiedosta. Tutkimuksessa kaikki mahdolliset asuinpaikat villojen lisäksi 600-luvulta eKr. 400-luvulle jKr. otettiin huomioon, jotta voitiin tutkia paikanvalinnan ajallisia muutoksia. Paikkatietojärjestelmiä käytettiin hyväksi arkeologisen ja ympärististötiedon kirjallisuuden pohjalta muotoiltujen hypoteesien analyyseissa. Tutkittavana oli kuusi paikan ominaisuutta: geologia, maaperä, vesi, maanpinnan muodot, näkyvyys sekä suhteet teihin ja asutuskeskuksiin. Geologiaa ei mainita paikanvalinnan kriteereissä, mutta se oli merkittävä rakennusmateriaalien hankkimisen sekä rakennuksen vakauden kannalta. Suuret, 100-luvulla eKr. perustetut villat sijaitsevatkin usein lähellä rakennuskivien hankinta-alueita. Hedelmällinen maaperä oli sen sijaan hyvin tärkeä kriteeri. Maatila pyrittiin hankkimaan hyvältä viljelyalueelta jopa suurimman asutustiheyden aikana 1. vuosisadalla jKr., jolloin myös kaikki epäsuotuisimmat paikat olivat käytössä. Veden saanti on Rooman alueella kohtuullisen helppoa, sillä alueen sademäärä on melko korkea ja pohjavettä on saatavilla melko helposti. Vesi oli tärkeää myös varallisuuden ja sosiaalisen aseman osoittamisessa. Liiallinen vesi saattoi toisaalta olla ongelma, kuten aluetta uudella ajalla piinannut malaria osoittaa. Malaria on saattanut vaivata aluetta jo antiikin aikana, mutta asutuksen tiheys osoittaa että sen vaikutus ei ollut yhtä suuri kuin myöhemmin. Tietyt maanpinnan muodot, harjanteiden reunat ja niiltä ulkonevat niemekkeet, olivat suosittuja hyvin pitkän aikaa osoittaen tradition toimivuutta ja kestävyyttä. Rakennusten eteen piti jäädä avointa tilaa. Etelän- ja idänpuoleiset rinteet eivät ole alueella kovin yleisiä, mutta rakennukset pyrittiin silti sijoittamaan etelään viettävälle rinteelle. Itää sen sijaan vältettiin, koska kesän kuumuutta vilvoittavat tuulet eivät itäpuolella saavuttaneet rakennuksia ja talven kylmät tuulet puhalsivat usein idästä. Näkyvyyttä tutkittiin sekä rakennuksista näkyvinä maisemina että rakennuksen näkyvyytenä ympäristöönsä. Suosituimmat villa-alueet sijaitsivat rinteillä, jotka näkyivät Roomasta asti sekä suurimpaan osaan muuta kaupunkia ympäröivää maaseutua. Villa näkyvällä paikalla eliitin suosimissa kohteissa vahvisti omistajansa sosiaalista ja poliittista asemaa. Villassa oleskelu ja sen näköalat sekä rakennuksen ulkopuolelle että sen sisällä antoivat kuitenkin tunteen yksityisyydestä. Rooman alueen tieverkko oli hyvin tiheä ja varmisti hyvät yhteydet lähes koko tutkimusalueelta joka puolelle. Paras näkyvyys, tihein asutus sekä suurin määrä teiden varsille tehtyjä hautoja osuvat kaikki samalle alueelle, joka muodostaa Rooman alueen keskeisen hyvän naapuruston. Lopuksi tutkitut ominaisuudet yhdistettiin ja eniten ominaisuuksia omaavia paikkoja voidaan pitää ihanteellisina. Nämä kattavat noin neljäsosan tutkimusalueesta ja puolet kaikista asuinpaikoista sijaitsee näillä alueilla. Ihanteellisen paikan käsite perustui vuosisatojen käytännölliseen kokemukseen, joka kiteytettiin maatalousoppaiden ohjeisiin. iv acknowlEdgEMEnts Finding an interesting research problem is often a dificult process and sometimes the choice can almost come as a surprise. Classical archaeology did not interest me at all when I started to study archaeology in 1986. In fact, I thought it was very dull indeed: why study something that appeared to be already completely known and understood? My preconceptions of the discipline already started to change during an enlightening visit to Switzerland – ancient Helvetia – in the summer of 1987, but a full love affair began when I irst climbed the Gianicolo hill to the Institutum Romanum Finlandiae in Rome and looked out over the Eternal City very early one March morning in 1988. I thank Ari Siiriäinen from the Department of Archaeology at the University of Helsinki for providing me an opportunity to participate in the introductory course for ancient studies at the Institute. He certainly knew the impact the place could have on an innocent young student. Unfortunately, Ari passed away before he could see the end of the academic journey he set in motion, but I like to think that he would be pleased to see that the topic concerns the relationship of people and environment, the main subject of much of his own work. During that month in Rome in 1988 I also met Eva Margareta Steinby for the irst time. The meeting was unforgettable, at least for me, and I was thrilled to be accepted for the three month long course under her direction at the Institute in the spring of 1993. The theme of the course was the social associations of public building in ancient Rome, and she suggested I study two towns in the vicinity of Rome and the Romans who were associated with them. That ledgling seminar paper on Tibur and Tusculum developed later into the topic of my doctoral dissertation and Margareta became my supervisor. Her gentle encouragement, strong support and many fruitful discussions among other things have been of enormous value to my work and I cannot ever thank her enough for her wonderful example and for all the work she has done! One of the participants on that course in 1993 was Janne Ikäheimo from the University of Oulu. We had already met at the Palatine East excavations the year before and a lifelong friendship had begun. Janne was far more dynamic in his academic pursuits than I have ever been and consequently, could function as my supervisor during the last part of the work. I am grateful for his unwavering support and friendship – they have always been very important and highly valued! My external examiners, Elizabeth Fentress and Nicola Terrenato, are warmly thanked for reading and contemplating my dissertation. Their comments were invaluable and made the inal work so much better. Without the Institutum Romanum Finlandiae, very little of this project might have been realized and I also feel privileged to have been able call that pink Renaissance villa home on many occasions. Long stays in 1997, 1998–1999 and 2006–2007 were vital for collecting and checking the material as well as for writing the work. The upper terrace also happens to provide a magniicent view of my entire research area and many an hour has been spent there contemplating various research problems. All the directors of the institute during those v times, Päivi Setälä, Christer Bruun, Mika Kajava and Kaj Sandberg, offered their help and support when it was needed and this is gratefully acknowledged. The one constant igure at the Institute, Simo Örmä, has also always made his vast knowledge on all things Roman (whether ancient or modern) available. In addition, his patient assistance on practicalities of all kinds has been invaluable. Rome introduced me also to many other persons who generously gave me their time and expertise when I most needed it: Antonia Arnoldus-Huyzendveld, J. Rasmus Brandt, Jesper Carlsen, Maria Grazia Granino Cecere, Jochen Griesbach, Pia Guldager Bilde, Elizabeth Macaulay Lewis, Birte Poulsen, Saskia Stevens and Massimiliano Valenti. Most of my waking hours in Rome were spent in some of the most wonderful libraries in the world, particularly at the American Academy in Rome, the British School at Rome, the Deutches Archäologisches Institut and the École française de Rome, and their staffs are fondly remembered for all their help. The stays in Rome were also funded by the Foundation Institutum Romanum Finlandiae (the Wihuri Foundation Scholarship 1998–1999, the Cultural Scholarship 2006) as well as by the Villa Lanten Ystävät ry. (1997). The Alfred Kordelin Foundation gave me my irst research grant in 1996 as well as a number of others over the years (1999, 2001 and 2009). The Finnish Cultural Foundation and the University of Helsinki also funded my research. I am deeply grateful for all these instances of their faith in my work. Rome has been important for my work, but large parts of it took place in Finland. The Department of Archaeology at the University of Helsinki under Ari Siiriäinen and after his retirement, under Mika Lavento, and the ever changing staff have been of great help in various stages of my research. The discussions during graduate student seminars, over lunches and coffee breaks have meant a lot to me. The Institutum Classicum at the University of Helsinki and its staff have also always been helpful and supported my work – I would like to thank particularly Heikki Solin for his assistance in acquiring one of the basic sources for my work. The academic world is often perceived as an ivory tower isolated from the rest of the world, but thankfully, the tower is inhabited by a great number of wonderful people. Kalle Korhonen has been a dear friend for many years and an invaluable aid in all things related to Latin and Greek. Margot Stout Whiting did so much more than just correct my English. All linguistic errors are, of course, my own! Jaana Mellanen, a friend in music and archaeology, drew the Roman landowner and his Greek architect for the cover. To my friends and colleagues I am grateful for all the help, support, wonderful company and general maintenance of my mental health: Sanna Aro-Valjus, Nina Heiska, Vesa-Pekka Herva, Maija Holappa, Teemu Immonen, Jesse Keskiaho, Harri Kiiskinen, Anu Koponen, Laura Nissinen, Marika Räsänen, Kirsi Salonen, Sirkku Viitanen-Vanamo, Heini Ynnilä. My dear parents must have felt some concern over the exotic choice of profession of their youngest child, but nevertheless they always supported my choice – sometimes also inancially. My father passed away at the beginning of the research project and I am very sorry he never got to see it inished. My mother has always strongly encouraged my academic pursuits and I hope she is proud of me inally achieving one of its benchmarks. I dedicate this book to my parents with all my respect and love. Helsinki September 2010 Eeva-Maria Viitanen vi list of figurEs, tablEs and platEs figurEs Fig. 2.1 Distribution of Class 1 to 4 sites in the research area. a) Class 1, b) Class 2, c) Class 3 and d) Class 4. Fig. 2.2 Distribution of sites inhabited only before the Late Republican period. Fig. 2.3 Distribution of a) artiicial platforms and b) pottery inds. In b black = sites with one or two pottery types and white = sites with three or more types. Fig. 2.4 Distribution of dated (black) and not dated (white) sites. Fig. 2.5 Distribution of sites in each period. a) Archaic, b) Early Republican, c) Middle Republican, d) 2nd century BC, e) 1st century AD and f) 4th and 5th centuries AD. In a–c white = sites inhabited only before the Late Republican period. In d–e white = new sites. Fig. 3.1 Source areas (white) of a) tuff, b) peperino, c) travertine and d) lava with a 5 km wide buffer zone (dark grey) and distribution of sites (black) where the stone type has been used. Fig. 3.2 Boundaries of geological formations (white) with a 100 m wide buffer zone (dark grey) and Class 1 sites (black). Fig. 4.1 Areas best suitable (dark grey) for growing a) Emmer wheat, b) other wheat varieties and c) barley. d) Areas recommended for modern grain cultivation. Fig. 4.2 Areas best suitable (dark grey) for growing a) grapes and b) olives. Fig. 4.3 Areas recommended for modern cultivation (dark grey) of a) orchards, b) pastures, c) brush and d) trees. Fig. 4.4 and Table 4.6 Distribution of best soils (dark grey) for various crops and number of sites by class and date. Fig. 4.5 Distribution of best soils (dark grey) and a) sites inhabited only before the 2nd century BC, b) sites established in the 2nd century BC, c) sites established in the 1st century AD and d) sites occupied during the 4th–5th centuries AD. Fig. 5.1 Rivers, streams, and lakes in the research area. River basins outlined with black thick line, watersheds with white line. Fig. 5.2 Public aqueducts in the Roman region. Fig. 5.3 Early sites (white) and those established in the 2nd century BC (black) with 100 m a.s.l. contour. Fig. 6.1 and Table 6.4 Direction of the main axis of sites and the features the orientation follows. Fig. 6.2 Number and direction of slopes around the settlement sites. Fig. 6.3 and Table 6.5 Most common view directions and view widths. Fig. 6.4 and Table 6.6 Names for parts of hill, typical site locations and numbers of sites on these. Spur sites by class and date. Fig. 6.5 Areas receiving most sunlight during the day on a) March 21, b) June 21, c) September 21 and d) December 21. Darker areas receive more light. Fig. 8.1 Road networks in the research area with bridges and crossroads. Fig. 8.2 Distribution of settlement sites compared to a 500 m wide buffer zone around the main roads. a) Classes 1 (white) and 2 (black). b) Classes 3 (white) and 4 (black). Fig. 8.3 Distribution of settlement sites compared to a 2 km wide buffer zone around the main roads. Secondary roads also indicated. a) Classes 1 (white) and 2 (black). b) Classes 3 (white) and 4 (black). Fig. 8.4 Distribution of settlement sites compared to habitation centers (towns in dark grey, villages in light grey). a) Archaic, b) Early Republican, d) Middle Republican and d) 2nd century BC. Fig. 8.5 Distribution of votive deposits (large grey) compared to roads, habitation centers (black) and crossroads (white). Fig. 9.1 a) Geological formations used as building materials. b) Areas at a distance greater than 100 m from open water. c) Areas within 1 km of a known spring. d) Almost level to moderately sloping (1–12%) ground. e) Areas within 3 km from habitation centers. list of tablEs Table 2.1 The deinition of “site” in various surveys in Italy. Table 2.2 Conventional dates used for building techniques and pottery types. vii Table 2.3 Comparison of dates on surveyed and excavated sites. Excavation dates from De Franceschini 2005. x = evidence from survey only, sett = settled, aban = abandoned. i–ii, etc. refer to building phases. Table 2.4 Summary of dated sites. Table 2.5 Summary of different inscriptions and sites. Uncertain cases in brackets. ID = instrumentum domesticum. Table 3.1 Geological formations in the research area. Number of all sites 1,941, of Early sites 279 and of sites established in the Late Republican period 184. Note that one site can be located on more than one formation. Table 3.2 Quarries and pottery production sites in the Roman region. Map numbers refer to Plate II.2. * = outside research area. Table 3.3 Distribution of sites by class and date over one or more geological formations, boundary zones and soft/hard formations. Table 4.1 Soil types found in the research area. Map numbers refer to Plate III.1. Table 4.2 Soil quality requirements for major Roman crops and how they were modelled. Table 4.3 References in written sources to cultivation and animal husbandry in the Roman region. Table 4.4 Agricultural activities related to speciic villas mentioned in the written sources. Table 4.5 Summary of classes, dated sites and remains related to agricultural production. >1 = more than one feature; PR = pars rustica. Fig. 4.4 and Table 4.6 Distribution of best soils (dark grey) for various crops and number of sites by class and date. Table 4.7 Classes of sites at Tibur (areas I, II and IV) and Tusculum. Table 5.1 Monthly and annual temperatures and rainfall in Tivoli and Rome 1921–1965. Based on Ventriglia 1990a, Tables 3.8, 3.11 (Rome, 1921–1965); Ventriglia 1990b, Tables 3.2, 3.4 (Tivoli, 1935–1965). Table 5.2 Climate changes in the Mediterranean area from the Archaic through the early Medieval period. Based on Ortolani and Pagliuca 1994; 1995; 1996; 2003; Caiazza et. al. 1999; Molinaro et al. 2001, Fig. 10. Table 5.3 Public aqueducts running through the research area. Table 5.4 Consumption of aqueduct water in quinaria according to Frontinus (aq. 76–88). No attempt has been made to correct the data which added up do not match the total reported by Frontinus. * = The amount of water from the Aqua Claudia and the Anio Novus arriving in Rome is presented as a total, hence the countryside igures are also totals. ** = Aqua Alsietina is on the right bank of the Tiber. Table 5.5 Summary of data presented in Appendix IV with comparison to excavated sites. Uncertain cases in brackets. * = only private aqueducts. Table 5.6 Sizes of cisterns found in classes of sites. Table 6.1 Elevations in the research area and number of sites on each height zone by class and date. Table 6.2 a) Aspects in the research area. b) Aspects of the settlement sites by class and date. Note that most sites are open to both main directions. Table 6.3 Attributes used to describe terrain for settlement sites. Fig. 6.1 and Table 6.4 Direction of the main axis of sites and the features the orientation follows. Fig. 6.3 and Table 6.5 Most common view directions and view widths. Fig. 6.4 and Table 6.6 Names for parts of hill, typical site locations and numbers of sites on these. Spur sites by class and date. Table 6.7 Sun altitude and azimuth during solstices and equinoxes. Table 9.1 Qualities used for modelling the ideal location. Table 9.2 Number of sites located in good and excellent areas by class and date. list of platEs Plate I.1 The Roman region with towns and main roads. The research area divided into survey areas in black. The limit of the Roman Campagna in red. Plate I.2 Symbols used for indicating sites in three survey publications: left Tibur (Mari 1991), center Ficulea (Quilici and Quilici Gigli 1993) and right Ager Tusculanus (Valenti 2003). Plate II.1 Geological formations in the research area. Descriptions in Table 3.1. Plate II.2 Quarries and pottery production sites in the Roman region. Surveyed sites numbered. Small symbols for sites on the geological map, Carta storica 1988, Talbert 2000, Petracca and Vigna 1985 and De Franceschini 2005. Plate III.1 Distribution of soil types in the research area. Descriptions in Table 4.1. Plate III.2 Land use in the Roman region according to written sources. See Plate I.1 for explanation of other map symbols and colors. Plate IV.1 Distribution of remains related to agricultural production in the Roman region. See Appendix III for a catalogue of sites. Plate V.1 Distribution of best soils (dark grey), settlement sites and roads in the northwestern part of the research area. Table features numbers of sites. Class 1 = black, Class 2 = red, Class 3 = green, Class 4 = yellow. Square = remains related to agricultural production. Sabatine paleosol in brown. Plate V.2 Distribution of best soils (dark grey), settlement sites and main roads in the central part of the research area. See Plate V.1 for explanation of symbols. Plate V.3 Distribution of best soils (dark grey), settlement sites and roads in the southwestern part of the research area. See Plate V.1 viii for explanation of symbols. Plate VI.1 Wells in the research area: pozzi romani in blue, settlement sites with wells in red, other wells in yellow. Groundwater depth on the black contours in meters below ground level. Plate VI.2 Springs in the research area. Seep areas in dark grey, spring lines in grey. Settlement sites with springs in green, other spring sites in red. Plate VII.1 Aqueducts and water channels in the research area. Public aqueducts in red, other channels in purple. Class 1 sites with water channels in red, other settlement sites in yellow, other channel sites in blue. Plate VII.2 Cisterns found in the research area. Plate VIII.1 Distribution of basins, nymphaea and baths in the research area. Plate VIII.2 Landscape units based on geological divisions in the research area. Plate IX.1 Elevations in the research area. a) Digital elevation model. b) Ortho map of the DEM. c) Elevations divided into zones. d) Analytical hillshade. Plate IX.2 Aspects in the research area. Southerly aspects in red, northerly ones in blue. Plate X.1 Terrain types and sites divided by class and date. Valley bottoms in orange, ridge shoulder in green, ridge crests in black, steep slopes in yellow and lat areas in blue-grey. Plate X.2 Cumulative viewsheds from a) ridge crests and b) river valleys. Red = best visibility, green second best and blue mediocre or poor. Ridge crests and rivers in black lines. Plate XI.1 Villa of the Quintilii on the Via Appia. Ground plan with windows and view directions. (Based on Ricci 2000.) Plate XI.2 Villa of the Quintilii on the Via Appia. View from the caldarium (complex E–L) towards northeast–east. (Photo: EMV.) Plate XII.1 Views from the Villa Sette Bassi. a) Ground plan with windows and view directions. (Based on Coarelli 1993, ig. on p. 149.) b) The surroundings of the Villa of the Quintilii and Sette Bassi. Plate XII.2 Views from the Villa Adriana. a) Ground plan with contour curves. (Based on Ricotti 2001.) b) The surroundings of Villa Adriana. Plate XIII.1 Even the largest villas blend in with the landscape today. The artiicial platform of the Villa of the Quinctilii Vari near Tibur – ca. 5 ha in area – can be seen under the olive trees in the middle of the picture starting from the white church on the right. (Photo: EMV.) Plate XIII.2 Cumulative viewsheds from a) the Via Salaria, b) the Via Nomentana, c) the Via Tiburtina and d) the Via Praenestina Red = best visibility, green second best and blue mediocre or poor. Plate XIV.1 Cumulative viewsheds from a) the Via Labicana, b) the Via Latina, c) the Via Appia and d) the Via Ardeatina. Red = best visibility, green second best and blue mediocre or poor. Plate XIV.2 Cumulative viewsheds from a) all the main roads and b) the secondary road network. Red = best visibility, green second best and blue mediocre or poor. Plate XV.1 Cumulative viewsheds from the burials in the research area. Red = best visibility, green second best and blue mediocre or poor. Plate XV.2 Cumulative viewsheds from a) Class 1 and b) Class 3 settlement sites. Red = best visibility, green second best and blue mediocre or poor. Plate XVI.1 Views from a settlement site a) in the plateau (Ficulea site 9a, Class 3, view towards north) and b) on the slopes (Tibur IV site 224, Class 1, view towards northwest–north. (Photos: EMV.) Plate XVI.2 Cumulative viewsheds from Class 1 sites located around a) Tusculum and b) Tibur. Red = best visibility, green second best and blue mediocre or poor. Plate XVII.1 View from the Gianicolo hill in Rome towards modern Frascati. Note that the early modern villas can be easily discerned. (Photo: EMV.) Plate XVII.2 Towns, villages and road stations in the Roman region. Ancient names used if known. Circles indicate 10 and 15 km radii around Rome. Plate XVIII.1 Single burials and groups of burials or cemeteries in the research area. ER = Early Republican, LR = Late Republican, Imp = Imperial. Plate XVIII.2 Distribution of Class 1 (red), 2 (black) and 3 (purple) sites compared to a 500 m wide buffer zone (dark grey) around crossroads. Plate XIX.1 Centuriations suggested for the research area. a) Collatia–Gabii. b) Bovillae–Tusculum (red) and Campi Tiberiani (blue). Base lines in black. Plate XIX.2 Composite map of the distributions of all qualities divided into ive classes. Table features the amounts of area for the ive classes. Plate XX.1 Distribution of settlement sites compared to the qualities of location. a) Class 1, b) Class 2, c) Class 3 and d) Class 4. See Plate XIX.2 for explanation of colors. Plate XX.2 Distribution of new 2nd century BC sites in yellow. Class 1 in red and Class 2 in black. 100 m contour in black line. Plate XXI.1 Distribution of dated sites against qualities of location. a) Archaic, b) Early Republic, c) Middle Republic, d) 2nd century BC, e) 1st century AD and f) 4th and 5th centuries AD. In a–c black = sites settled only in early periods, in d–e yellow = new sites. iX 1 studying thE roMan villa and its EnvironMEnt 1.1 introduction In the 1st century AD, the geographer Strabo described the area of Tusculum as “adorned by the plantings and villas encircling it.”1 The villas, “magniicently devised royal palaces,” were the central feature of the countryside in ancient times, and even today, the Roman villa is one of the most common inds in the archaeological landscapes of central Italy and elsewhere in the Roman world. It is perhaps the most common type of site found in surface surveys and probably thousands of sites have also been excavated. The villa is also a phenomenon well known from written sources: it is described in manuals of agriculture and in many other kinds of texts. Despite – or maybe because of – its commonness and the abundance of material, there still remain many unanswered questions and many topics to research. Several seminal studies were done in the late 19th and early 20th centuries on villa architecture and as well as on its economic aspects.2 General books on Roman domestic architecture and also speciically on Roman villas have also been published starting in the 1950’s.3 Excavation reports on single villa sites have always been published, but few have had such far-reaching effects as the one concerning the Setteinestre Villa in modern Tuscany excavated in the 1970’s and 1980’s.4 The interpretation of the data and its contextualization in Roman economic and social history remains even today one of the most comprehensive analyses of the villa system. The last decades have seen a gradual increase in all aspects of villa studies. New analyses of various types of architectural forms, decorative materials, social use of space in and around the villa as well as economic associations of the villa have been published.5 The subject of this study is a topic that still has rarely been examined: the relationship of the villa to its environment, both natural and man-made. In ancient sources, this topic is discussed when the authors give instructions on selecting a suitable location for a villa and the estate in general or, more rarely, as descriptions of the surroundings of a villa. Cato, in his agricultural handbook, goes as far as saying that a good location s more important to the villa than its other qualities.6 In modern research, the literary evidence is carefully cited and a 1 5,3,12; translation H. L. Jones (Loeb Classical Library). For villa architecture, see, e.g., Rostovzeff 1904; 1911; Swoboda 1918; Carrington 1931; Drerup 1959. For villa economy, see, e.g., Rostovzeff 1926; Carrington 1931; Day 1932. The Roman agricultural economy from the point of view of agricultural methods and land holding patterns were dealt by, e.g., White 1970; Kuziščin 1984; Neeve 1984; Lewit 1991. 3 Mansuelli 1958; McKay 1975; Percival 1976; Mielsch 1987. 4 Carandini 1985a; 1988. 5 Architecture: Rossiter 1978; Lafon 2001a; Romizzi 2001. Sculpture: Neudecker 1988. Use of space: Smith 1997; Adams 2006; 2008; Griesbach 2007. Economy: Marzano 2007. 6 Cato agr. 1,5: Instrumenti ne magni siet, loco bono siet. 2 chaptEr 1 few archaeological villas itting the advice given are mentioned,7 but the incompatible cases are ignored. Systematic studies on how well these instructions were or could be regarded in any region of Italy have not been conducted, although some aspects of the relationship of the villa to nature have been previously explored.8 The work at hand addresses the question in one region of central Italy, the surroundings of Rome, which was important for the economy of the city of Rome9 and also functioned as favored holiday resorts.10 The aim is to use written, archaeological and environmental evidence to examine what kinds of locations were commonly chosen for building villas, how well these locations match the instructions given in written sources as well as what beneits or problems the environments offered. The area chosen for this study consists of approximately 500 km² of countryside on the left back of the river Tiber northeast, east and south of Rome (Plate I.1). Villas can be found everywhere in Roman Italy and it would have been possible to choose among many areas when considering the selection of the study area. However, some central zones feature limitations, such as missing environmental data for the Bay of Naples, where the eruptions of Mount Vesuvius in AD 79 and afterwards have changed the milieu considerably. Many of the other interesting areas of Italy have been insuficiently surveyed archaeologically or published only partially.11 Large tracts of the surroundings of Rome have been archaeologically surveyed and published in the past four decades, thus providing easy access to a relatively uniform data set.12 Many different kinds of sites have also been excavated and this material supplements the survey data. The region is also well known from written sources: as one of the central areas of the Roman world, it features frequently in ancient literature and the epigraphic record is also one of the largest in the whole of Italy. Environmentally, the Roman region consists of a plateau lanked by two mountain ranges, thus offering varying milieus for analysis. The archaeological survey areas do not reach the Tyrrhenian coastline, but remain in the inland area, thus excluding the study of maritime and coastal villas. In addition, it is also important to note that one of the key authors who described a suitable location for a villa, Marcus Porcius Cato, hails from the area. Cato’s treatise is the earliest description of a villa and thus has its roots irmly in central Italy in general and probably speciically in the surroundings of Rome. The availability of material is also an important question and as the Roman region is also a central area today, the environmental materials needed for this study are available.13 In ancient times, the research area was regarded as part of a region called suburbium that probably extended far beyond it.14 Administratively, small sections of the studied region were under the city of Rome, whose limits are relatively well known,15 but mostly it was covered by territories of several small towns, the limits of which are generally poorly known.16 The 7 E.g., McKay 1975, 100–101; Carandini 1988, 52–53; Schneider 1995, 73–76; Perkins 1999; Mayer 2005, 149–163; Marzano 2007, 155–171. 8 Schneider 1995. 9 Morley 1996; De Seña 2003; 2005; Witcher 2005a. 10 Champlin 1982; Mayer 2005. 11 Cf. the catalogue in Marzano 2007, where most sites are in Lazio and the majority of the Lazio sites are in the surroundings of Rome. 12 See below Chapter 2 for a detailed description of archaeological data and written sources. The surveys used are: De Rossi 1967 (Tellenae); 1979 (Bovillae); Giuliani 1966 (Tibur II); 1970 (Tibur I); Mari 1983a (Tibur III); 1991 (Tibur IV); Quilici 1974a (Collatia); Quilici and Quilici Gigli 1986 (Fidenae); 1993 (Ficulea); Valenti 2003 (Ager Tusculanus). Four surveys adjacent to these areas have also been published (De Rossi 1970; Muzzioli 1970; Pala 1976; Quilici and Quilici Gigli 1980). I decided to leave these four out, as the chosen areas form a a large, uniform surveyed region covering 28 x 27 km. 13 Sometimes the quality is better for the Roman region than for other areas of Italy, cf. Perkins 1999 and Goodchild 2007 who had to conduct their analyses without, e.g., a detailed soil map. 14 Champlin 1982; Agusta-Boularot 1998; Lafon 2001b; Mayer 2005, 43–59. 15 Panciera 1999. 16 Tusculum has been treated extensively on many occasions: Grossi Gondi 1908, 35–39; Valenti 2003, 64–66; 2 introduction regions near certain ancient towns, such as Tibur and Tusculum, can be easily associated with them, but large tracts of the research area cannot be connected to town territories. The modern situation is administratively even more complicated, but the general geographical area of the surroundings of Rome does have a modern term. The region is today often called Campagna Romana (“Roman countryside”) in Italian and this consists generally of the gently undulating plain stretching 20–30 km outside the city of Rome (Plate I.1). Its limits in the northeast are the pre-Apennine hills and the Alban Hills in the southeast. The Campagna Romana also naturally covers the western bank of the Tiber.17 The term “Roman Campagna” is thus used for the research area of this work.18 1.2 dEfining thE villa Before discussing the aims and methods of this work, it is necessary to take a look at previous research on the Roman villa in order to place the current project in its context. The survey also underlines some of the main trends in research approaches, their beneits and problems, as well as gaps they have left. The research history of the Roman Campagna, including discussion of the methods used, can be found in Chapter 2. One of the irst tasks, though, is to discuss the deinition of the Latin term villa and to also look at its use in research literature. The deinitions of the word villa are based on written sources and this is not surprising considering the fairly poor preservation of archaeological remains. Texts offer detailed information on the ancient thinking of what a villa was and what its manifestations and its uses were. They also offer the opportunity to study the changes in the term over time, at least from the 3rd/2nd century BC until Late Antiquity. The relative abundance of written sources gives the chance to try to understand the concepts used by the ancients themselves and even to try to apply these to the archaeological realm.19 The most vivid ancient discussion on the meaning of the term can be found in Varro’s de re rustica in the form of a conversation between Varro and his friends in the villa publica in Rome towards the end of the Republican period. This text forms the basis for most modern discussions on the matter.20 “Villa” denoted a building in the countryside or sometimes the whole estate,21 but it is also clear that not all buildings in the countryside were villae. The whole estate was called praedium, fundus or, more rarely, rus, but the most common practice was perhaps using a noun derived from the name of the town where the villa was located, e.g., (praedium) Tusculanum. Agriculture or other productive activities were probably involved more often than not, but a division of the villae into those dedicated only or mostly to productive activities (villa rustica) and to those used mostly for recreational purposes (villa urbana) is often presented. These terms can also, of course, mean different parts of the same villa. The productive activities were often commercial, i.e., cash-crops such as grapes and olives were grown to sell surplus in the markets. These activities also commonly required investments in land and equipment, as Hernández Martínez 2007, 47–68. The evidence for the territory of Tibur is more scarce, e.g., Mari 1983a, 24–25 and 1991, 24–25. The limits of the other towns have not been discussed, probably because too little is known of them. 17 Scotoni 1993. 18 The legendary British archaeologist, Thomas Ashby, already used the term for the same area in the late 19th and early 20th century (e.g., Ashby 1927). 19 Cf. Morley 2004, 26–30. 20 Varro rust. 3,2,3–17. E.g., Percival 1976, 12–15; Mielsch 1987, 7–8; Carandini 1988, 44–51; Gros 2001, 265–267; Romizzi 2001, 29–32. 21 Cf. OLD s.v. villa. 3 chaptEr 1 well as sustenance through the waiting period until the estate became productive. In addition, the villa is most often regarded as the country residence of the members of the Roman elite, most commonly those of senatorial or equestrian rank. The meaning of the term also changed when the Roman socio-economic situation changed. The rustic farmhouse of the 2nd century BC developed into the luxurious and mostly residential villa described from the late 1st century BC onwards. The villa was also used as a tool for self-promotion for political purposes in a manner similar to the town house, domus.22 It also had an important role in the cultural pursuits of the Roman elite as a place for contemplation and study. The villa was the center for the economic, political and social activities of the Roman elite in the countryside. These deinitions derived from written sources have often been illustrated with archaeological remains.23 The application of the literary term to the archaeological material seems so easy and uncomplicated that it has not been questioned and concrete images of what certain types of villa looked like emerge from the archaeological data. The most famous archetypes of a simple villa rustica are perhaps those excavated in the late 19th century in the areas buried by the AD 79 eruption of Vesuvius.24 Examples of the villa urbana can be found in abundance in the same region, but also very prominently in the surroundings of Rome in the huge complexes such as the villa at Tivoli owned by Emperor Hadrian.25 Using these kinds of extreme cases has made the differences seem obvious, but it is also clear from both excavation and survey results that a great range of buildings existed in the Roman countryside. The productive farm buildings vary between very small and modest to large and sophisticated units and the same applies to the large and luxurious residential complexes.26 The written sources describe establishments of both productive and residential type and, certainly, some archaeological sites it the descriptions, but what about the rest of the sites? Some of the attributes derived from written sources used to deine the term villa are hard or impossible to ind in archaeological data. One of the more important deining factors is the villa’s connection to the Roman elite. The written sources were produced by members of the Roman economic and social elite and the texts were intended to be read by their equals. The sources deine the most prominent features of the Roman countryside in terms of the Roman aristocracy, from the point of view of the political and economic elite. Little is known of what the others outside the senatorial and equestrian ranks thought of the countryside.27 Other groups with economic means, such as rich freedmen, certainly wanted to own and did own country residences and these were also called villae.28 If the term villa should be reserved for the buildings owned by the Roman elite, what should the others be called? The identity and social status of the owner of an archaeological site is known with certainty only extremely rarely and even then it represents only one period in the history of the building. Did a building stop being a villa if it was bought by someone other than a member of the elite? Whether sites were controlled by landowners or tenants cannot be distinguished archaeologically. Some words, such as tugurium or casa,29 have been offered as terms for the “other” buildings in the countryside, although the written sources are vague on what these other buildings were and how they could be deined. On closer examination, most passages referring 22 23 24 25 26 27 28 29 4 E.g., Hales 2003. E.g., McKay 1975, 100–135; D’Arms 1984; Gros 2001, chapters 8–12. E.g., Carrington 1931; Rossiter 1978, passim; D’Arms 1984. E.g., D’Arms 1984; Romizzi 2001, passim; De Franceschini 2005, passim. Respectively Rossiter 1978 and Romizzi 2001. E.g., MacMullen 1974; Garnsey 1979; Dommelen 1993. E.g., Cic. leg. 3,13,30 concerning a libertus owning a villa at Tusculum. E.g., Romizzi 2001, 30. introduction to tugurium describe buildings in foreign countries or in the remote past of the Romans.30 Casa is also problematic because it is not very clearly deined as a rural building; it also has urban connections. In rural contexts, casa is also used in association with foreign countries as well as with the distant past, e.g., the hut of Romulus on the Palatine Hill in Rome is casa Romuli.31 Neither word was used very clearly to mean rural buildings in the context of Roman culture of the Late Republican or Imperial periods. This also accentuates the chronological distribution of the word villa as the written sources start from the 3rd/2nd century BC, but it cannot be known for certain whether the word was used or what it might have meant in the preceding periods. The later writers also applied their contemporary vocabulary to the past, using the word villa for some country houses, e.g., Livy from chapter 2 or from around the beginning of the Republic onwards. Varro’s discussion on the meaning of the term villa also makes it clear that although the country houses of the Roman ancestors were very different from the ones of the 1st century BC, they should nevertheless be regarded as villae.32 The terms used for settlement sites in archaeological publications show variation and sensitivity to chronological as well as to regional contexts. Villa is quite uniformly used for the richest settlement sites of the Republican and Imperial eras: a certain amount of inds or combinations of inds qualify the site as a villa or villa rustica.33 When all the sites in a surveyed region are generally poor, few sites are called villa or even villa rustica.34 The situation becomes more problematic when poorer sites are described and classiied. These are sometimes called casa rustica35 or fattoria36– thus the use of Latin nomenclature has not been even attempted. Chronological differences can also be detected: an Archaic or early Republican site is rarely called villa, casa or even farm. The terms used tend to be neutral, such as sito antico (“ancient site”). Using ancient terminology for modern research would be ideal, but it remains uncertain if these terms can be understood properly considering the nature of the sources on their content and use. In addition, the modern associations of “villa” in most languages are to luxurious country residences, mansions, vacation retreats, etc.37 In this study, the term villa is used generally in accordance with the traditional manner: it is taken to mean a rural building displaying signs of wealth, such as large size and expensive decorations. The descriptions of the buildings in written and archaeological sources it together reasonably well in this category. Luxurious residential complexes can be found side by side with smaller productive units with well-appointed living quarters and baths. It should be noted that villa is used more as a modern archaeological term for a certain type of site in the surroundings of Rome than what was perhaps intended as a villa by Romans themselves. The use as an archaeological rather than a historical term is also the reason why I exclude one of the factors, ownership from the deinition.38 30 Searches were made in the Brepolis Library of Latin Texts A collection (for more information see http://www. brepolis.net/) for tuguri* and casa*. 31 E.g., Val. Max. 4,4,11; Vitr. 2,1,5. 32 A thorough investigation of the ancient texts concerning terms used for rural buildings, similar to those done in the context of the room names of the Roman domus (e.g., Leach 1997; Riggsby 1997; Nissinen 2010), would be very useful. 33 E.g., Kahane et al. 1968; Dyson 1978; Potter 1979; Ikeguchi 2000, 8–11. For a more detailed discussion of the concept of site, see Chapter 2.3. 34 Cf. Yntema 1993. 35 “Rural house;” e.g., Quilici 1974a; De Rossi 1979. 36 “Farm;” e.g., De Rossi 1979. 37 E.g., Longman Webster College Dictionary (1984) s.v. villa. The same applies to the most common translation of both the Latin and English villa to Finnish, “huvila” (although it should be noted that Pitkäranta 2001 gives also the meanings of manor/mansion, country house and farm). Nomenclature used for the archaeological sites in the Roman Campagna is discussed in Chapter 2.3. 38 In the analysis Chapters 3–9 I do not often use the term “villa” since the discussion mostly concerns not the building, but rather the plot it was built upon and the area around it. 5 chaptEr 1 A villa, deined as a large country house, was perhaps the most important feature of the Roman countryside, but it did not exist in a vacuum. The majority of the settlement sites are either earlier than the attested villas or poorer in inds or smaller in size. It is important to examine whether the villas and these other sites were different in other respects, particularly with regard to the main question of this study, the relationship of the villa to its environment. The poor and early sites are called simply “settlement sites” or “possible settlement sites” when the interpretation is uncertain, as no good evidence for using Latin nomenclature exists. The villa as known based on the written sources and some archaeological sites is the focus of the analyses, but it is equally important to realize that it was not the only type of building and/ or estate in the countryside. 1.3 thE roMan villa in classical studiEs The Roman countryside and the villa have been objects of countless studies and despite the great number of archaeological remains, the main emphasis of these has been on historical sources. Archaeology has been used perhaps more to corroborate or visualize the results derived from the texts. The studies can be divided roughly into three categories. The irst concerns the buildings themselves and their origins, typologies and functions. The second category deals with the economy of the Roman countryside, with landownership, investment and agriculture and the role of the Roman villa in them. The third category is slightly more heterogeneous, concerning general aspects of the Roman world which could be classiied under Roman social history: the social and ideological uses of the countryside, studies on Roman otium or civilized leisure so closely connected to the concept of the villa. The following discussion concerns irst the origins and development of the villa, then the typologies and, lastly, the economic and social aspects of the villa. Origin and Development of the Villa The earliest contemporary sources concerning the Roman villa available can be dated to the early 2nd century BC.39 In the plays of Plautus, the term villa was used as a self-evident and understandable phenomenon of the time. Roughly contemporary to Plautus is Cato’s handbook of agriculture written before 149 BC. What happened before that is dificult to know based on written sources. Accounts of Roman history written in the Late Republican or Imperial periods have been used to reconstruct the social and economic development of Rome starting from the Archaic period and to explain the birth of the villa. Archaeological data has also been used in this process, but even excavation often reveals only a little of the earlier phases of the building as the later structures cover the older ones almost completely. Two main versions for the development of Roman agriculture and rural settlement have been delineated and they treat the origin of the villa in slightly different ways. The irst one is based on a slow process moving from fairly primitive subsistence farming and animal husbandry on small farms towards commercial and investment agriculture using slaves as a workforce.40 According to this model, the change from subsistence to commercial agriculture can be dated to the 3rd–2nd century BC, i.e., the period slightly prior to the writings of Cato. The Catonian villa rustica was a fairly small affair with only a small slave workforce. The 39 Villas are mentioned in works of Roman history in connection to fairly early events, but as these, e.g. Livy, were written much later, these references cannot be taken as accurate descriptions of the material culture of the times. 40 Torelli 1990. 6 introduction earliest archaeological examples of the villa would be the artiicial platforms constructed using polygonal technique. These are mostly found in the more remote areas where they survived later rebuilding. The models for the development of both agriculture and architecture originated outside Italy; hellenized Southern Italy and Punic North Africa were the sources. The second also model presumes the independent peasant as the farmer of the Early and mid-Republican period.41 The differences appear in the later development. Rome’s expansion in the Mediterranean world, particularly in the period after the Second Punic War, was based on an army formed of free peasants, who could not work their lands while abroad on long war campaigns. They fell into debt and had to abandon their farms. As a result, the rural population moved to the cities in search of work and livelihood. War brought booty, slaves and opportunities for enrichment for those who received them. This wealth had to be invested somehow and the lands of the free peasants in trouble could be easily bought and developed by the persons beneitting from the spoils of war. The villa, then, was a new invention targeted on commercial agriculture. The agriculture described by Cato, Varro and Columella often requires substantial investment in planting and production equipment as well as long waiting periods before proits could be realized. Commercial agriculture could not be regarded as suitable for subsistence farming. Villa architecture is not treated very much in this theory, but the models for the reforms in agriculture and architecture are again derived from Hellenistic or Punic spheres. The most recent attempt at exploring the origins of the villa is based on the excavations of the Auditorium Villa in Rome in the 1990’s.42 The building has a very long history, starting from the Archaic period, and its earlier phases are also fairly well preserved. What is special about this site is the rebuilding of the early small farm as a large and luxurious complex with what are probably separate living quarter and productive parts towards the end of the 6th century BC. In comparison with other sites of the period, the built area is enormous and it remains very large among its peers until the 1st century BC, when the villa becomes quite normal in size among the many other large country houses. It has been suggested that the Auditorium Villa was in fact a country residence for the head of a Roman elite family, who thus asserted his right over the landscape inhabited by members of his clan. Such residences would have been relatively rare which is why they probably have not been found before. The other farms were small in size and there would not have been intermediary forms between the very large and the small. The possible model for these large residences might have been the slightly earlier Etruscan elite palaces, such as Murlo. It is also claimed that there were no Catonian small or medium-sized villae rusticae in the 3rd–2nd centuries BC.43 The development into the 1st century BC villa would have been much faster and the changes would have happened closer to this boom period than suggested before. The economic explanation echoes the common model of war booty invested in land, as external funds were needed to establish the new large villas and that the owners would have represented a much more heterogeneous group of persons than before. The main novelty of this new theory is the suggestion that the villa took its ideological and architectural models mainly from the home ground. The beginning of the villa would already date to the Early Republican period, if not even further back in time, in contrast to the standard 3rd century BC theory. Archaeological evidence for the Hellenistic and Punic models has not really been satisfyingly produced and the idea of eastern luxury as a model for the residential villa is based mostly on ancient literature which can often be regarded as 41 Toynbee 1965 is the originator and countless shorter references exist, e.g., Marcone 1997, 123–149. Terrenato 2001. For the excavation results, see now Carandini et al. 2006. 43 See Mari 2005 for analysis of early Catonian villas. The dispersed nature of the settlement in this period is not denied by Terrenato, but the sites known cannot be regarded as villas. 42 7 chaptEr 1 propagandistic and political in its aims. The economic model opposes the slow process of growth from simple and small to complex and large, but it its the other main model, i.e., external funds were needed and these were used to develop landed property. The time of these changes is pushed even later than has been presumed before, from the 2nd century BC to the late 1st century BC. The new model is still based on evidence from only one site44 and its relevance to the interpretation of the settlement patterns in central Italy still needs to be established. As can be seen from the discussion above, the origins of the villa still remain very much under debate. The later architectural development of the villa is generally agreed upon. The buildings grow more and more substantial and the living quarters gain precedence over the productive parts. This is a phenomenon frowned upon by the Late Republican and Early Imperial authors as a decadent imitation of the eastern luxury of oriental and Hellenistic rulers and aristocracy. The humble and hardworking Roman farmer and his modest farm were replaced by opulence, marble, paintings, sculpture and leisure. The archaeological evidence seems to conirm this by producing countless large villa complexes with baths, libraries, elaborate gardens and other features, but few – or even maybe no – facilities for agricultural production.45 The same applies to the later economic development of the phenomenon. Relatively few literary sources exist for the Early Empire and what happens in the following centuries in Italy has been debated hotly. At the beginning of the 20th century, a theory was already presented according to which the Roman provinces outside Italy grew in importance with the production of the old staples of commercial villa agriculture, olives and wine. The process started in the 1st century AD. The competition from the provinces resulted in a crisis in Italian agriculture which led to changes and, in many cases, the demise of the old villas as unproitable enterprises.46 The process was fueled by the later political disturbances which made the countryside unsafe and the farmers more dependent on the protection offered by rural elites.47 The archaeological surveys and extended excavation record have revealed a picture much more complicated than has been thought. A great deal of regional variation in the survival of the villae and their productive activity in Italy as well as in the provinces has been observed.48 If inding an origin for the villa has been hard, then charting its later development and possible end is an even more confusing endeavor, emphasizing the importance of local contexts. Villa Typologies The architecture of the villa is described in excavation publications, but relatively few studies on the types of buildings and building complexes have been conducted. The lack of good data can explain this partly, as relatively few villas have been excavated extensively or the results properly published. Work has also often concentrated more on the living quarters than other parts of the building. In addition, in many cases, the structures have only barely survived above loor level due to plowing and sometimes to robbing of building materials. The areas outside the built environment have often been left unstudied, leaving the building in a kind of a vacuum, unconnected even to its most immediate surroundings. Roman wall paintings and 44 Cf. Becker 2005; 2006. E.g., Romizzi 2001, 37–40; Gros 2001, chapters 7–10. 46 Rostovzeff 1926, with further argument in Carandini 1988, 267–285; summaries in, e.g., Vera 1995a or Marcone 1997, 151–156. 47 Dyson 2003, 89. 48 E.g., Vera 1995a; 1995b; Dyson 2003, 89–106; Majbom Madsen 2003. 45 8 introduction mosaics depicting landscapes have also been used as sources for the appearance and styles of intact, whole buildings, which are sometimes quite hard to imagine based on the scanty remains.49 Many of the buildings identiied as villas in the paintings consist of long porticoes and they are often located by the sea. The stereotypical image of a maritime villa is probably based on reality, but using it as a basis for reconstruction is somewhat problematic. A tendency to separate productive farms from the non-productive vacation residencies can be detected in the treatments of villa architecture. Typology of the productive farms is based on size and functional variations of the buildings and a possibility of an upper loor. The simplest houses feature a roughly rectangular ground plan with much internal variation. The functional parts consist of living quarters, productive parts, and sometimes even a bath. The larger houses have living quarters and productive parts arranged around a central courtyard or sometimes axially. The rectangular or square form is still common. The living quarters, production and processing areas can also be arranged in separate buildings with yards, alleys or corridors between them. This kind of layout can be found in connection to large and luxurious sites. It has also been suggested that “agricultural factories” existed where luxurious living quarters for the owner did not exist, but instead only small, undecorated rooms for the slave who took care of the farm.50 The typology of residential villas has been treated in more detail. Types of villas are based on the main architectural space(s) observed in the ground plans of the buildings: the atrium villa, the peristyle villa, the portico villa.51 In the atrium villas, the rooms are arranged around an atrium similarly to a Roman town house with a possible garden (hortus) behind the atrium. Although the type could be considered Italic or typically Roman compared to the others, these are not very commonly found. In the peristyle villa, the main rooms are arranged round a closed courtyard surrounded by colonnades on all sides. The portico villa features main rooms arranged in a row and fronted by a portico and it tends not to be very common in Italy, but is found in great numbers in the Roman provinces. A further type could be the complex buildings starting to appear from the 2nd century AD onwards, the so-called “pavilion” villas. Many of them are hard to put into any one category as they often feature all three basic elements of the ground plans; atria, peristyles and porticoes. The variety of ground plans found in these villas is accentuated by the fairly large number of subtypes. The models for all types but the atrium villa are derived from Hellenistic architecture. Macedonian, Alexandrian and Near Eastern palaces built between the 4th and 1st centuries BC are cited as inspirations for the architecture in accordance with what is known from written sources.52 Such Italic complexes as the Etruscan palace of Murlo are not mentioned even though that particular building is undoubtedly a peristyle and much earlier than any of the Hellenistic examples. What is striking in both the typologies of the productive and residential villas is the great variety of ground plans and how a very large number of them do not really conform to any of the proposed types. Some of the “types,” such as the pavilion villa, are in reality collections of varying ground plans which only vaguely resemble each other. Many times the main type seems to apply only to one small part of the building leaving the rest more or less unclassiied. Many aspects of the building have not been discussed, e.g., the effect the terrain had on where and how the various building parts were placed.53 Maritime and coastal villas 49 50 51 52 53 E.g., Rostovzeff 1904; Thagaard Loft 2004. Carrington 1931; Rossiter 1978. Swoboda 1918; Romizzi 2001. E.g., Nielsen 1994; Kutbay 1998. Cf. Tessaro Pinamonti 1984; Zarmakoupi 2005; 2006. 9 chaptEr 1 have been studied in greater detail as the only topographically deined group.54 Determining the function of the rooms in villas is very dificult and the poor preservation of the buildings makes it very dificult to apply such methods of analyzing use of space, as has been done in the case of the Roman domus. The ground plans, access between spaces, decorations, as well as inds, have been taken as points of departure and the result is a renewed picture of how the Romans used their houses socially.55 Some attempts have been made to study social use of space in Roman provincial villas56 and to recognize villae suburbanae by analyzing the number of entertainment spaces.57 The most comprehensive analysis of use of space in a single building is that made of the Setteinestre Villa in Tuscany,58 where areas intended for the use of the family and guests could be recognized by analyzing the patterns of doors and access from one area to another.59 The dificulties of analysis also apply to studying decoration and recovered artifacts in relation to space types and access. For example, the vast amount of sculpture found in villas has been recontextualized and the signiicance of statue collections has been discussed,60 but further study is mostly impossible as the exact original locations of the statues are not known. Role of the Villa in the Historical Studies Ancient literature testiies that the villa was an important part of the economy of the Roman countryside. The prevailing models of the Roman economy have for a long time been based on the dichotomy of the consumer city and the producer countryside. Landed property was probably the most important source of wealth and income and as such, the villa is central to the discussion of landownership and real estate business in the Roman world. The villa was the center for agricultural production, but probably also included other types of production. The variety of productive activities of the villa and, speciically, of the aristocratic landowner, is related to two further models of the Roman economy: the adherents to the primitivist model maintain that economic rationality was not known, the scale of the economy was small and that landowners had few other interests beyond agriculture. The opposing, modernist/substantivist theory supposes some kind of economic rationality, and interest in manufacture and trade on the part of the aristocracy.61 The dichotomy of consumer city versus producer countryside has also been of considerable importance in the past and only quite recently has the consumerism of the countryside been recognized.62 The villa has also entered the discussion in connection to economic rationality, such as calculations of cost eficiency and productivity.63 Productivity has been analyzed in the context of, e.g., demography, organization and use of slave labor.64 54 Lafon 2001a. Wallace-Hadrill 1994; Grahame 2000; Hales 2003; Allison 2004; Berg 2010. 56 Smith 1997. 57 Adams 2006; 2008. In Campania (2006), this method could be regarded as applicable due to the better preservation of the buildings, but in the Roman region (2008), the results are less convincing as the buildings are so poorly preserved. 58 Carandini 1988, 109–224. 59 Some of the interpretations have been criticized since, e.g., the identiication of slave quarters around the rustic courtyard (Marzano 2007, 125–153). 60 Neudecker 1988. 61 Morley 2004, 33–50. See also Witcher 2006b, 40–42 for a brief, but effective summary of the interpretations concerning the Roman countryside, particularly the economic ones. 62 Morley 1996 for both a synthesis of previous research as well as a new interpretation. For consumerism in the countryside, see also De Seña 2003; 2005; Martins 2003; Witcher 2005a. 63 E.g., Tchernia 1986; Carandini 1988, 235–285. 64 E.g., Carandini 1988, 287–338; Jongman 1988; Rosaio 1994; Scheidel 1994; Carlsen 1995; Witcher 2005a; Goodchild 2007, particularly 298–384; Marzano 2007, 125–153. 55 10 introduction Different types of agricultural production, e.g., grain cultivation, wine and olives, animal husbandry, ish cultivation, etc. have also been analyzed.65 Studies of Roman owners of landed property in the countryside concern both economic and social history. Ancient literature and inscriptions have been used to recognize landowners and to study the attitudes towards land, the economy of landowning and patterns of landholding.66 The studies concentrate on senatorial landowners as they are the ones mentioned most often in the texts and can be interpreted most securely as landowners. The data concerning landowners and their activities in the Bay of Naples area has been woven into a historical narrative of changing social, economic and cultural conditions of the area.67 The area is one of the most popular villa resorts in the whole of Italy, which therefore provides quite a unique point of view. No similar works could be (or have been since) written on any other area in Roman Italy. The studies of the social and ideological dimension of villas are a relatively recent phenomenon. Most of these tend to be based on historical sources and do not necessarily explore the archaeological side of villas. The attitudes of the Romans to nature inside and outside their villas – gardens, agriculture and the surrounding natural world – and the ideas of idyllic nature and its relationship to life in the villas have been studied recently.68 The conclusions of this study accentuate the owner’s private pleasure in the villa, but since then, the importance of the villa for the social promotion of the owner has been argued for.69 Having a villa with the right address was important and even the owner’s memory remained associated with the buildings for a long time, even after his death. 1.4 thEorEtical and MEthodological considErations The purpose of this study is, as stated, to examine the relationship of the villa to its environment. This is done by comparing the physical characteristics of the settlement site to its surroundings. The aim is to ind out whether the sites feature speciic qualities repeatedly or whether their characteristics correspond with what can be found in their surroundings in general. The main questions are where the villas were placed and why these locations were chosen. Locational analysis70 is still relatively rarely applied to studies of the Roman countryside despite the great number of surveys providing the archaeological material and often also including analyses of the environment. In archaeology, locational studies are based on site catchment analysis introduced to the discipline in the 1970’s and used since then in a wide variety of ways.71 The original intent of the method was to analyze the resource potential of the surroundings of an archaeological site for gaining insight into the land use and 65 For general works on Roman agricultural economy, see above note 2. A valuable survey of the uses of archaeology in the study of Roman economy can be found in Storey 1999, 222–231. Grain: Spurr 1986. Wine and olives: Tchernia 1986; Amouretti and Brun 1993; Mattingly 1996. Animal Husbandry: Rinkewitz 1984; MacKinnon 2004. Fish: Higginbotham 1997; Marzano 2007, 47–81. 66 E.g., Rawson 1976; Shatzman 1976; Treggiari 1979; Wiseman 1987, 367–370; Neeve 1990; Kehoe 1993; 1997; Andermahr 1998. 67 D’Arms 1970. 68 Schneider 1995. 69 Bodel 1997; Hales 2003. 70 E.g., Haggett et al. 1977. 71 Vita-Finzi and Higgs 1970 being one of the seminal applications, see also Chisholm 1962 for an introduction to many of the key concepts. For reviews including more current research trends in landscape archaeology in general and in the Roman world in particular, see Kvamme 1999; Anschuetz et al. 2001; Witcher 2006b. 11 chaptEr 1 livelihood of its inhabitants. The approach became popular in prehistoric archaeology and many methods were adopted and developed in the following decades. The development of the Geographical Information Systems (GIS) increased the use of spatial analysis techniques and approaches.72 The economic aspects were mostly considered when Roman landscapes were irst analyzed on the island of Hvar on the Croatian coast in one of the seminal studies in the early 1990’s using GIS.73 Analysis of site distributions and landscapes has since extended towards the ideological realm, studying religious landscapes and landscapes of power, among other topics.74 Locational analysis can also be used for administrative purposes, e.g., to help manage cultural heritage or to create predictive models to ind new sites and map areas of archaeological potential.75 In this study, various characteristics of the site locations are analyzed and then compared to the general physical environment.76 In the beginning, the intention was also to look at the immediate surroundings of the sites, e.g., with the help of a buffer zone around the site, but this was quickly abandoned as even relatively small buffer zones would have covered most of the research area, rendering comparison more or less useless – in this respect, this work does not strictly represent site catchment analysis. The relative number of sites featuring certain qualities, e.g., geological formation or soil type, is compared to the amount of that quality in the area. If more sites are located on the zones with that quality than can be expected based on the extent of these zones, then it can be said that there is a tendency to favor the quality when sites are selected. The qualities or aspects of the physical environment analyzed are geology (Chapter 3), soils (Chapter 4), water (Chapter 5), terrain (including altitude, slope gradient, aspect and geomorphology; Chapter 6), visibility (both view from and visibility of the site in its surroundings; Chapter 7) and roads, towns and villages in the vicinity (Chapter 8). These were selected based on reading the ancient literature, most importantly the agronomists’ work, but also other texts.77 The site distributions were analyzed based on the types of sites and their period of use. The purpose of the building, whether intended and used mostly for productive purposes or as a country residence designed for leisure and rest, is a signiicant factor in the process of site selection. Consequently, it is necessary to compare the locations of the different kinds of sites. In order to see possible changes in site selection over a long period of time, the locations of the dated sites were analyzed in connection to all the aspects. The sites have commonly been used for long periods of time and often also before they became villas. Dispersed settlement began in the Roman region in the Archaic period and the traditions of site selection were formed long before the villa, as known from Roman literature, developed. These traditions and changes in them can be traced by comparing the choices made in different periods of time. Each aspect is discussed irst separately in order to explore its signiicance and to understand what it could offer for the ancient settlement. However, this isolation is not complete as the aspects are intrinsically connected to each other: soil formation depends largely on the geology and terrain, visibility of terrain, etc. Some features of the aspects can also be discussed in connection to the others. Each aspect also leads to studying other topics related to the social and economic uses of the countryside: geology leads to building materials, soils to agriculture, water and terrain to salubriousness, visibility and roads/towns/ E.g., Hodder and Orton 1976; Hietala and Larson 1984; Lock and Stančič 1995; Conolly and Lake 2006. Gaffney and Stančič 1991. A similar approach can be found in Perkins 1999 analyzing the Etruscan and Roman settlement in Tuscany. 74 E.g, Ashmore and Knapp 1999; Thomas 2001; Fleming 2006; Sterry 2007; Lake 2007. 75 See Rua 2009 for an example of a predictive model for mostly administrative purposes. 76 All the analyses were conducted with the raster-based GIS program Idrisi with visualizations exported to a graphics program to create the inal images. Materials are described in detail in Chapters 2–8. 77 Cf. Perkins 1999, Goodchild 2007, 121–179 and Rua 2009 for similar selections. 72 73 12 introduction villages to the ideological and social meaning of the country estate as well as community. In the end, the various aspects are discussed as a whole, as parts of the continuous landscape of the Roman Campagna, combining the natural environment with the way people perceived it (Chapter 9). It might be asked whether this approach suits the Roman materials and analyzing Roman culture. Locational analysis has been mostly used in connection to prehistoric cultures whereas Roman society is a complex, historical entity with fairly advanced agricultural techniques as well as varying motives for selecting sites for habitation. Quantifying material and seeking trends might also seem unnecessary in the light of ancient literature which explicitly states what should be sought in buying an estate and building a villa. The details provided by the ancient literature are compelling, but it should be remembered that the written sources do not represent the entire Roman society. Material culture provides additional information and sometimes can even contradict the literary evidence. In addition, ancient literature does not reach very far back in time, only until the 3rd/2nd century BC, and archaeological material is the main source for the periods before this. The instructions given by Cato for site selection in the 2nd century BC were probably not wholly created by Cato himself, but were also based on a long period of observation, practical experience and common knowledge by previous generations. What makes the whole process interesting is comparison of the data sets (Chapter 2): ancient literature gives an idea of what the ancient perception of how things were might have been and checking the archaeological material against the environmental data can lead to the same results, but can also offer other solutions.78 The analyses also aim at contextualization of the sites. In previous studies, the villa is mostly treated as isolated from its surroundings and even from other similar sites. Villa architecture is discussed based on two-dimensional ground plans which commonly do not feature much information concerning the terrain of the site even though this is signiicant to the way the buildings were designed and constructed. At its worst, even the various building parts are discussed separately and not from the point of view of the whole.79 Another typical way of discussing villas is using some well-known and well-preserved buildings to represent a relatively poorly deined group(s) of other sites.80 One of the typical cases is the Villa dei Misteri located just outside Pompeii, used frequently as a prime example of early villa architecture.81 The archaeological source material is vast and growing constantly; consequently it is harder and harder to grasp the whole of the Roman villa even in such a limited area as central Italy. Despite the dificulty, quantifying the not so well-preserved and well-known sites behind the exemplum case would make the arguments more persuasive and reliable.82 Sometimes it is necessary to use only the well-known cases because the other sites simply do not offer similar material. For example, studying the views from the villa is not possible if the walls have not been suficiently well preserved and the places of openings in them are not known. But even in these cases, it should be emphasized that the examples could very well be unique and not represent the common situation. The villas were an integral part of the Roman landscape, deined as an entity consisting of the natural environment and the human actions and perceptions related to it. They should also 78 Cf. Storey 1999. See, e.g., Förtsch 1993 for a discussion on archaeological examples of spaces mentioned by Pliny the Younger in his letters. This corresponds with the way Pliny writes of the various rooms, as they are described isolated from each other, island-like (Riggsby 1998). 80 Cf. Pucci 1994, 60–62. 81 E.g., Mielsch 1987, 30–41; Schneider 1995, 74. 82 E.g., De Franceschini 2005 sets out to delineate the history and development of the villa in the Roman region by analyzing a sample of one hundred excavated sites inside the modern Comune di Roma without discussing what her sample represents. Romizzi 2001 discusses the typology of villa architecture with a similar premise; the representativeness of her sample is not discussed. 79 13 chaptEr 1 be studied as part of the landscape, not separated from it. The Romans might have not had a speciic word for landscape,83 but they were very aware of the inluence of visual signs, of how the material culture inluenced the human mind and behavior.84 Psychological, philosophical and anthropological studies have pointed out that humans are shaped psychologically by their environment, that no human experience and existence could exist without “being-inthe-world.”85 The landscape was not merely a passive background for human actions nor did it set too strict constraints for them,86 it took an active part in the actions that also might have changed it. By analyzing the places where Romans chose to live in the area where their knowledge concerning land use was created, it is hopefully possible to gain some insight into how they were being-in-the-world. 83 See Cosgrove 1984, 1–68 for the development of the concept in early modern times. See Appleton 1975 and 1990 for analyses of landscapes in art. Latin prospectus is perhaps the closest equivalent, but it is perhaps most often used in the meaning of “view.” 84 Zanker 1987 on the emperor Augustus’s visual propaganda is the classic work. See also Zanker 1979 for the effect that villas had on the design and decoration of Pompeian houses. For later works on various aspects of Roman social life and its connection to materiality, see, e.g., Wallace-Hadrill 1994; Favro 1996; Hales 2003. 85 E.g., Gibson 1979; Ingold 2000. 86 Knapp and Ashmore 1999, 2. 14 2 archaEological MatErial and writtEn sourcEs This study uses materials collected and published by other researchers in many disciplines: archaeologists, philologists, historians, geologists, palaeobotanists and zooarchaeologists. To be able to use these sources, it is important to understand the basic methods of collection and representation, whether it is an archaeological site list, a geological map, a text edition of Cicero’s letters or a corpus publication of inscriptions. Each category of data has its strong and weak points and these collectively have an effect on the outcome of this study. This chapter concentrates on the main source materials, archaeology and written sources, including a brief history of research concerning the Roman region. The environmental sources are discussed in the relevant chapters. 2.1 rEsEarch history of thE roMan caMpagna The Roman Campagna has been an important research area in central Italy from the Renaissance period onwards. Its location around the city of Rome (Plate I.1), abundant literary references and the richness of archaeological inds have all contributed to the curiosity of antiquarians and researchers alike and made sure that the area has always received much attention. The earliest descriptions of ancient monuments, maps, plans and drawings were published in the 15th century starting a steady low of reports and research on the area. All of these documents create a considerable bibliography of which only glimpses can be offered here.87 The earliest notes on discoveries of ancient buildings and artifacts date to the medieval period and are mostly sporadic references to accidental inds while building in the area.88 These notes contain mostly information about speciic objects, architectural elements or sculpture, and their ind places. One example of such notes is the discovery of columns of Parian marble while constructing the monastery of Grottaferrata in 1020.89 The monastery is built on top of an ancient villa and in this case, the reference also provides contextual information. Accidental excavation and discovery continues, of course, even today. Starting from the 15th century, excavations were also made more consciously. At irst, the purpose of digging was to ind sculpture, precious stones and other materials to use in the building and decoration of palaces, villas and the museum collections of the Renaissance and Baroque aristocracy. The places prowled were often sites where the buildings remained visible above ground. The Roman Campagna was one of the prime areas for such ruins and 87 Each of the survey publications presents a full research history with a more detailed bibliography than can be presented here: Giuliani 1966; 1970; De Rossi 1967; 1979; Quilici 1974a; Mari 1983a; 1991; Quilici and Quilici Gigli 1986; 1993; Valenti 2003. 88 The history of the early excavations and discoveries has been partially collected in Lanciani 1902–1912 and mentioned passim in many modern publications. 89 Valenti 2003, 28, nota 13. chaptEr 2 consequently such digging activity.90 Knowledge of previous inds probably directed the excavation and sites known to have produced inds in the past were excavated repeatedly.91 The contexts of the inds did not really seem to matter to most; the collector was not interested in the ind spot, whether it could be said that the statue originated in, e.g., the villa of Cicero or some other ancient celebrity. This seems to have been the case despite the enthusiasm with which the ancient literature was studied; the connection between the text and the site was not necessarily very concrete.92 However, some exceptions to this rule existed as some famous buildings mentioned in texts were actively sought, e.g., Cicero’s or Lucullus’s famous villas in Tusculum. In these cases, excavation was used in an almost modern manner to test the hypotheses based on written sources.93 The data set was also supplemented by drawings and maps. Some of the irst surviving maps of the area were drawn in the mid-16th century.94 The best maps are of great signiicance in locating sites; they also often offer information on the historical place names and help place discoveries on the correct estate, if not more accurately. Publications of drawings and etchings depicting romantic landscapes of ruins also offer visual information on the remains and their condition although it has to remembered that not all were faithful renderings of what could be seen. Many descriptions of the whole Italian territory as well as of its regions were also written. These offer the irst collections of literary references concerning the area as well as descriptions of the antiquities found there.95 In general, the descriptions are based on ancient texts and are mostly attempts to locate buildings mentioned in them. One of the sites mentioned most often is the emperor Hadrian’s Villa Adriana near Tivoli, which aroused interest quite early. The search for the sites of the villas of Lucullus and Cicero in Tusculum still continues today with equally uncertain results as produced in the 17th century.96 The 18th and 19th centuries saw the birth of more precise and encyclopedia-like volumes, especially on the history and antiquities of the region.97 Rome and Italy were important targets of the Grand Tour and in the early 19th century, many travelogues of journeys in the area were published. These also often include sporadic information on ancient remains as well as drawings.98 The source material for the Tivoli region presents a peculiarity not found elsewhere: local histories written by local persons starting from the 16th century.99 These early volumes already catalogue all the later known signiicant ancient monuments and historical sources. The enthusiastic local work has continued in the 20th century with the journal Atti e Memorie della Società Tiburtina di Storia e d’Arte, published regularly since 1921. Local histories are otherwise relatively rare, and similar activity occurs only in the region of Tusculum, where the irst local archaeological and historical study was published in the early 20th century.100 The difference between these two regions is interesting to note. Tusculum has a far better 90 Cf. Romizzi 2001, 13–14. E.g., Villa of the Quintilii (Tellenae site U7) by the Via Appia was called Statuario (from Italian statua, “statue”) as a reference to the abundance of inds from the area (Schädler 1998, 29, nota 1). 92 Weiss 1969, 90–130. 93 Ehrlich 2002, 55–58; cf. also Weiss 1969 for more general Italian cases. 94 Eufrosino della Volpaia’s Il Paese di Roma/Mappa della Campagna Romana from 1547 (Ashby 1914; Frutaz 1972 I, 20–22; 1972 II, tav. 26). 95 E.g., Clüver 1624; Holstenius 1666; Kircher 1671. 96 Cf. McCracken 1935; 1942; Ehrlich 2002, 55–58; Valenti 2003, 78, 85–87. 97 E.g., Volpi 1745; Nibby 1848–1849. 98 E.g., Uggeri 1800–1830; Nibby 1819. 99 Nicodemi ca. 1585–1589 [1926]; Zappi ca. 1572–1583 [1920]; Del Re 1611; Cabral and Del Re 1779; Bulgarini 1848. 100 Grossi Gondi 1908 was preceded, however, by Mattei 1711 and followed later by Borda 1943; 1958; Devoti 1978; 1981. 91 16 sourcEs documented and illustrious past than Tivoli, but it has not inspired the locals to such great extent. One reason for this can be found in the habitants and the land use in the early modern period. The western slopes of Tusculum around modern Frascati were appropriated by the Papal court and Roman aristocracy in deliberate emulation of the ancient Roman villeggiatura.101 They took care of the recording and narration of the history of their chosen region in their own way and this did not probably entail such local pride as in the case of Tivoli. Only one early modern villa was built in Tivoli, Villa d’Este, and the locals could call the Roman history and the ruins their own and write about them in a very different way. A more scientiic approach to studying ancient topography developed in the late 19th century and more systematic documentation of the ancient monuments began in the Roman region. The area began to change very rapidly in this period and there was much building activity which was accompanied sometimes – but not very often – by excavations, which added to the survey data collected. The aim of the studies was to collect as much as possible of what was still visible. The researchers were both Italian, such as Roberto Lanciani and Giuseppe Tomassetti, and foreigners, such as Thomas Ashby, whose survey work covered almost the entire Roman Campagna and still remains the basic publication in some areas.102 These descriptions include ancient, medieval and modern sources as well as brief notes on the archaeology of the sites. Ashby also photographed the area with great enthusiasm, which makes his archive at the British School at Rome invaluable. The expansion of Rome in the early and mid-20th century did not leave a great record of archaeological investigations. Some excavation results, but not much else, were published before the 1960’s, when the work on the Carta archeologica d’Italia began in earnest. The idea for an archeological map of Italy was conceived almost a century earlier, in the 1880’s, as an attempt to document the wealth of archaeological heritage in Italy.103 In the Roman Campagna, most of the surveys have been conducted by the Università degli Studi di Roma “La Sapienza.”104 Each survey generally deals with one sheet of the 1:25,000 topographical map of Italy. Most of the resulting reports have been published in the Forma Italiae and Latium Vetus series.105 The publications present the collected results of research in archives and previous publications as well as of the actual ieldwork. They comprise the basic collection of archaeological information on the Roman Campagna, which has since been added to by the publication of excavation reports and some further surveys. Today, the most active research and publication occur in the northern parts of the Roman Campagna. The area has been resurveyed in cooperation between the Soprintendenza Archeologica di Roma and the Università degli Studi di Roma “La Sapienza,” but the results still remain mostly unpublished.106 Many of the excavations of the area are published in 101 Ehrlich 2002, 49–50. E.g., Ashby 1902; 1906; 1907; 1910; Lanciani 1909; Tomassetti 1910–1926; Buonocore 1997–2002; Lilli 2001. 103 Cambi and Terrenato 1994, 25–27; Quilici and Quilici Gigli 2004, 63–66. 104 The three archaeological superintendencies involved in the administration, protection and research of archaeological remains in the research area (two responsible for the Comune di Roma, one for the state and one for the Comune di Roma itself, and one responsible for Lazio) have not been directly responsible for much survey work. The superintendency of the Comune di Roma maintains the Carta storica archeologica monumentale e paesistica del suburbio e dell’agro romano in a GIS system (see also preface by Eugenio La Rocca in De Franceschini 2005) which I consulted in 2006. No complete archives, collections of all sites or databases for the whole area exist as far as I know. 105 The Latium Vetus series was the result of a project inanced by the Consiglio Nazionale delle Ricerche (CNR) and its Centro di studio per l’archeologia etrusco-italica (Quilici and Quilici Gigli 1986, 9; 1993, 9). 106 Cf. Carafa 2000; Pergola et al. 2003; Di Gennaro et al. 2004. The reason for the resurveys is obvious: the need for data from better controlled circumstances. As the ind material from the old surveys has not been preserved apart from the Latium Vetus surveys pottery material such a reexamination of the material as is going on with the data from the South Etruria Survey cannot be fully conducted (e.g., Patterson 2004; Patterson et al. 2004). This also means that, e.g., new knowledge of pottery typologies and dates cannot be applied to the old data. 102 17 chaptEr 2 the Bullettino della Commissione Archeologica Comunale di Roma or other journals. New work in the Tivoli area under the Sopintendenza per i Beni Archeologici per il Lazio is also regularly published. The eastern area is also relatively well documented, but perhaps not as thoroughly as the northern sector. The southern zone is more problematic as hardly any of the new excavations have been published and apparently no new surveys to complement those made in the 1960’s and 1970’s have been conducted. Archaeological activity in the Roman Campagna has been considerable for many centuries. The next task is to look at how these studies were conducted and how their results were presented. 2.2 fiEldwork MEthodology Excavation The irst excavations were not real excavations, but more accidental discoveries of archaeological material while digging for other purposes. This continued to be the case when the sites were dug speciically for precious materials and artifacts. One might need a permit from the Papal authorities to dig, but there were no requirements to produce documentation or reports. The excavation techniques did not develop greatly until the late 19th century, when more attention began to be given to stratigraphy and contexts. It remains unclear how well these concepts were applied in the Roman Campagna as the excavation reports very rarely discuss the methods used. This is a common phenomenon for all of Italian classical archaeology where the discussion on method only began in earnest in the 1970’s.107 The method applied has been described as strictly stratigraphic108 or then as scavo dell’attenzione,109 not strictly stratigraphic. No real rules or deinitions were outlined for the scavo dell’attenzione providing great liberty for applying different techniques to digging and documentation. The quality of the results depends solely on the knowledge and diligence of the researcher him/herself.110 When excavating buildings, only limited areas were dug and these in narrow trenches along the uncovered structures. Digging consisted of “liberating” the buildings from the layers covering them, not paying very much attention to these destruction layers and the information they might have contained on the abandonment and/or destruction of the buildings. The stratigraphic excavation was generally used when digging below ancient loor or ground levels. Natural interfaces of layers were followed if the director of the excavation considered that useful, but otherwise arbitrary levels were probably used quite commonly. The reason cited for this was the use of unskilled labor.111 Few plans or sections were drawn during the excavation and often only the ground plans and elevations of the building itself were drawn. Little attention was paid to the inds except for those which might be considered signiicant for the dating of the building, i.e., brick stamps or stamped pottery. The decorative elements were another valued category, e.g., mosaics, wall paintings 107 Manacorda 1982; Barbanera 1998, 152–154. Some of the irst applications of the stratigraphical method in the 1970’s took place at a villa excavation at Setteinestre. In addition, the pioneer work of Nino Lamboglia in northern Italy should be mentioned. However, these were great exceptions to the general rule. See, e.g., Barbanera 1998, 170–173 and in the same volume Terrenato 1998, 178–181. 109 “Attentive excavation.” D’Agostino 1991, 52; Barbanera 1998, 68–69. 110 Barbanera 1998, 68–69. 111 Manacorda 1982. 108 18 sourcEs and sculpture. Modern excavations commonly follow the techniques of stratigraphic or single context excavation, but very little information is still given on the methods applied. Another indication of the classical archaeologist’s apparent lack of interest in developing new methods is buildings archaeological analysis. Until the 1990’s, the description and autopsy of buildings were based on traditional observation presented as an argued narrative. The stratigraphic analysis of buildings began to be formulated only in the 1990’s by Italian Medieval archaeologists and since then the method has worked its way into classical archaeology, e.g., in Pompeii.112 It can be claimed that the new method does not bring anything radically new to the actual process of analysis of buildings, but its merit is in making the process more systematic than before. It also offers ways of visualizing the argumentation and better chances of verifying and evaluating the results. Only approximately 5% of the ca. 2,000 settlement sites found in the surveys of the research area have been excavated and the majority of these fairly recently in the area to the north and northeast of Rome.113 Many are rescue excavations and the structures have been revealed only partially, naturally concentrating on the areas that may be destroyed. Excavations covering the whole site or even most of it have been rare.114 The complete excavations of two sites on the Via Gabina conducted by Rice University in the late 1970’s and early 1980’s have produced very interesting results.115 Apart from the Villa Adriana, only a few of the large villas have been excavated systematically. The recent work in Villa dei Quintilii by the Via Appia has not yet been published fully116 and the same applies to the recent work at Frascati on several large villas there. The published reports are often very short and concentrate on the structures. In addition, the diagnostic pottery and the most important other inds are published. The situation observed in the Roman Campagna also applies to whole of central Italy. The greatest exceptions to this rule are perhaps the villas destroyed by the eruption of Vesuvius in AD 79, many of which have been mostly or totally excavated and which have yielded a great amount of important evidence.117 Survey Most of the sites are known only through archaeological surveys. The early survey works consist of fairly random observations of visible remains and collections of historical notes. The same applies to the descriptions of sites in travel diaries. The writers had a fairly limited knowledge of archaeological remains, but the historical sources are quoted quite accurately and the collections can be extensive. Very often the writers aimed at creating lists of known landowners from literary sources and inscriptions as well as sometimes including local legends. Medieval and modern place names were already used to connect ancient remains to known historical persons in these early treatises. Very little is known of the way the ieldwork was conducted, but maps attached to the texts are often accurate as are the descriptions of Parenti 1988a; 1988b. Parenti’s work is truly signiicant as it is used as a model for Medieval buildings archaeology also in, e.g., Finland and Sweden (e.g., Ratilainen 2001; Eriksdotter 2005). For Pompeii, see, e.g., Coarelli and Pesando 2006. 113 Cf. De Franceschini 2005, which catalogues one hundred excavated sites in the area of the Comune di Roma. 114 Very large scale excavations have been conducted in connection to major building projects in the Roman region recently, but most of these have yet to be fully published. See, e.g., Di Manzano 2001; Musco et al. 2001; Gioia and Volpe 2004. I would also like to thank Dr. Jochen Griesbach for showing me around the enormous Porta di Roma excavation area in the summer of 2002; cf. Di Gennaro et al. 2004 for some preliminary results. 115 Cotton 1979; Oliver-Smith and Widrig 1981; Widrig 1980; 1983a; 1983b; 1987. The most complete data is now available through the website, The Via Gabina Villas. Sites 10, 11 and 13 (http://viagabina.rice.edu/). 116 Ricci 1998 is the irst volume. 117 For a list of sites, see D’Arms 1984. In addition, see De Caro 1994 for a more recent approach on the excavation of a modest villa rustica site. 112 19 chaptEr 2 the buildings and this would permit the assumption that the writers observed the remains themselves. The basic methods and contents of topographical studies were more or less established towards the end of the 19th century.118 What is perhaps slightly surprising to note is that the Italian tradition regarding the research related to studying and interpreting regional data or as it is called in Italian topograia antica (“ancient topography”) is an almost independent discipline compared to archaeology. According to this view, archaeology comprises excavation and art historical studies and the studies of ancient topography are to be separated from these.119 The contents of the discipline can be described the same as those of the English term “landscape archaeology,” which has been translated into Italian as archeologia del paesaggio (“archaeology of the landscape”).120 The two Italian terms mean more or less the same judging by the recent handbooks using both terms in their titles: Introduzione all’archeologia del paesaggio and Introduzione alla topograia antica.121 The separation of landscape archaeology from the other forms of classical archaeology has not prevented archaeologists from participating in both activities. Lanciani might have not studied ancient art history,122 but he certainly excavated, surveyed as well as studied archives and inscriptions. Today, Italy is perhaps one of the few countries in the world – or perhaps even the only one? – with institutions and professors dedicated solely to ancient topography (professore di topograia antica). Despite this specialization, relatively little seems to have changed since the late 19th century as even the modern Italian surveys starting in the late 1960’s were extensive in character. They were conducted over relatively large areas by one person only.123 Naturally, the professional knowledge of the surveyor has increased and the work is more systematic than before. Nevertheless, in inal publications only a little thought and space is given to the way(s) in which the surveys have been carried out or circumstances. This is somewhat surprising considering the many publications of methodological literature on survey, how to process material and how to present results.124 In addition, even the work conducted by the British School at Rome in South Etruria, the beginning of modern archaeological survey in the Mediterranean area, seems to have had very little, if any, inluence on the way the Italians themselves conducted their surveys.125 This lack of interest in methodology is perhaps embedded in the Italian tradition of not discussing the “how it was done,” but rather only the “what was found.” The principle is expressed most clearly by two extremely experienced ieldworkers and respected professors, Lorenzo Quilici and Stefania Quilici Gigli,126 in their recent book Introduzione alla ricerca topograica (2004). Of the 186 pages of text, they give ca. 10 to the methodology of ield survey. In general, they dismiss the past and current lively discussion on how to conduct surveys in such a way that the results are comparable and reliable as unnecessary since each real situation will dictate the methods used and consequently no “rigid rules and norms” are See also Terrenato 1996 for a discussion of ieldwork methods in surveys of the central Italian area. Cf. Barbanera 1998, 90; Quilici and Quilici Gigli 2004, 9. 120 Cf. Cambi and Terrenato 1994; Cambi 2003. 121 Respectively, Cambi and Terrenato 1994; Quilici and Quilici Gigli 2004. 122 According to Barbanera 1998, 90 this would perhaps have earned him an entry in the Enciclopedia dell’Arte Antica. 123 In the most recent surveys in the Roman region, the extensive character remains, but the area surveyed by one person has diminished from the 10 x 10 km map sheet to the extent of the late 19th century estates. 124 Cf. Belvedere 1994; Cambi and Terrenato 1994. Terrenato’s own survey work in Etruria is an exception to the general Italian rule, e.g., Terrenato 1992; 1996, 2000; 2004. 125 Kahane et al. 1968; Potter 1979. 126 Respectively in the Università di Bologna and Seconda Università di Napoli. 118 119 20 sourcEs needed. This is certainly true as there will always be ad hoc applications to even the most strictly deined method. What is more serious is their failure to understand that some basic rules to work with are needed. The rules of stratigraphic excavation can be applied according to speciic situations, but the basic rules remain the same and the results of excavations can be evaluated based on the knowledge that these basic norms have been appreciated and applied. This kind of agreement on basic survey techniques is still lacking and to be able to evaluate the results of the work, it is necessary to know how it was done. This is something that is still almost completely missing, even in the most recent of the survey reports on the Roman Campagna, conducted in the late 1990’s.127 2.3 archaEological MatErial One of the reasons for selecting the research area was the fact that large tracts of it have been archaeologically surveyed and published in the past 40 years. The original survey publications were intended as basic surveys of the archaeology of the Roman region and were meant to be used for administrative purposes, for protection and conservation, as a basis for further research and, simply, as documents of the rapidly vanishing archaeology of the area. The survey publications contain much of the prior archival and publication data and this has been supplemented with more recent publications of surveys and excavations.128 Acquainting myself with the area and the material included not only studying relevant literature, but also visits to see some of the sites that still survive the continuing expansion of modern habitation in the surroundings of Rome. These visits were directed mostly to the northern and southern sections of the research area as these still contain many accessible zones. Some new sites were found in the process, but, more importantly, an impression of the changing topography and landscape of the area was gained. The emphasis of this section is on the settlement sites although, naturally, a large number of various other types of sites were also found in the surveys. These will be described and discussed in the following chapters in connection to the aspect they are related to, e.g., cisterns and aqueducts in Chapter 5 on water resources. Settlement Sites from Surveys and Excavations Most of the publications give little space to methodological issues, but some information can be extracted reading between the lines and what follows is a description of basic survey methodology as it can be reconstructed. The area covered by one survey is generally one 1:25,000 topographical map sheet, ca. 10 x 10 km in extent, but this varies according to topographical factors, e.g., the eastern limit of the Fidenae survey is the river Tiber and not the edge of the map sheet some kilometers to the west. The naming of the survey area and publication depends on the major ancient town/village which existed in the area and in such a way the surveys can be perceived to be connected to the ancient territories of these towns. The clearest case is Tibur, whose survey more or less follows the known extent of the ancient 127 Cf. Valenti 2003. In the end, I decided not to include the new sites featured on the Carta storica archeologica monumentale e paesistica del suburbio e dell’agro romano (1988) in the territory of the modern Comune di Roma in this study. A total of 111 new sites for the whole research area were recognized, but the amount of data for each was minimal – usually the entry consisted of merely villa or area di frammenti ittili (“area of ceramic fragments”) with no dates. Considering the number of sites with more data and the fact that the analyses presented in Chapters 3 to 8 concern dated sites, it seemed unnecessary to include this material. 128 21 chaptEr 2 town’s territory. The ieldwork is conducted by one or two persons. In addition to ieldwork, some of the researchers have used aerial photographs to recognize roads and aqueducts, often very poorly visible in the ield.129 Very little information is given on what was done in general, e.g., what sort of plans for the visits to the ield were made, how many times they were visited, how much the time was used at every site, or even how much time was used for the whole survey. In some reports, the surveyors maintain that they had sometimes great problems with access to areas for various reasons. The landowners are often reluctant to let anyone on their premises, which in the case of the Collatia survey, led to very brief visits to many sites, but which ones is not mentioned in the site descriptions.130 Military areas are naturally inaccessible and their extent also cannot be published. Visibility problems are also mentioned sometimes.131 As the surveys were conducted in and after the 1960’s, many areas, especially by the main roads and closer to the center of Rome, had already been destroyed or covered by modern building activity. In the hill regions, particularly in the Alban Hills, observation is also dificult due to extensive wooded areas. As these inaccessible or non-visible areas are not marked in the inal distribution maps, it is dificult to know what the observed sites actually represent and whether the gaps in the distribution are accurate or possibly caused by other reasons. This is very problematic in the southwestern part of the research area as very little new research concerning it has been published since the survey in the 1960’s.132 On the other hand, in the areas where recent, largescale excavations have taken place, the presence of previously observed settlement sites has been veriied and the sites found missing in surface survey are roads or burials.133 The large number of sites and the great density of inds make the database for this study fairly reliable. The published reports irst offer a general introduction and a synthesis of the various chronological periods. Then a catalogue of sites follows which lists one by one all the sites observed. Their distribution and location is marked on the 1:25,000 topographical maps either as the true extent of the site or as symbols (Plate II.1). In most cases, the site areas have been indicated with raster symbols which probably give an idea of the true extent and shape of the scatter area.134 Clearly discernible structures, such as villa platforms, are often also marked on the maps as lines or hatched areas with precise limits.135 Some of the publications use combinations of representations, e.g., symbols for small sites (squares of varying sizes, circles, asterisks, numerals) and larger sites with clear boundaries.136 Most of the symbols 129 E.g., Quilici 1974a, passim. Quilici 1974a, 11 nota 1. In a January 2002 conversation with Dott. Massimiliano Valenti, surveyor of Ager Tusculanus, he mentioned that it took many years to gain access to some of the private areas in his research area. 131 E.g., Quilici and Quilici Gigli 1986, 10 (with a map of the urbanized areas showing which zones could be surveyed); Valenti 2003, 24. 132 Even the Carta storica archeologica monumentale e paesistica del suburbio e dell’agro romano (1988) map features only a few new sites in addition to those known in the 1960’s. Distribution maps for one zone slightly outside the research area (Bedini 1979; 1984) do point towards a much greater density of habitation than what is visible at the moment. In addition, personal experience has shown that each visit to the area has produced possible new settlement sites. 133 E.g., Musco et al. 2001, passim; Di Gennaro et al. 2004, 102–105. 134 A total of 1,060 sites or 55% of all sites. The raster symbols, shading with dots or hatching with lines, vary in size, shape and direction in the Tibur and Bovillae areas. On the Tellenae map, the site extent and orientation are almost always the same and this makes them look more like symbols. Rasters were used to indicate Archaic to Late Republican sites in the Collatia, Fidenae and Ficulea publications. 135 These were used in all publications in a uniform way. 136 Combinations were used in the Latium Vetus publications (Quilici and Quilici Gigli 1986; 1993), as well as in the Collatia volume of Forma Italiae (Quilici 1974a). The Ager Tusculanus map (Valenti 2003) shows ind points with dots and/or numbers, which in some cases are united into a villa platform. The digitized sites in this study are polygons instead of points, giving some indication of the size of the site. 130 22 sourcEs are unhelpful when, e.g., the orientation or the size of the site need to be determined.137 The majority of the material – 70% – does give an idea of size and orientation. What has been included in the catalogues varies slightly over time. Some of the earlier works present very few Archaic or earlier sites, e.g., the Bovillae survey features no sites prior to the 2nd century BC, but the later ones are often all-inclusive, recording pre-Roman, Roman and Late Antique sites; some even list Medieval remains. The heading for the description of each site is what could be described as an interpretation of the observations: villa, house, tomb, road, aqueduct, etc. These categories vary, but based on comparisons of actual site descriptions, the practical contents of the sites are often very similar. If no interpretation can be given, then the heading is simply area di frammenti ittili (“area of ceramic fragments”), ruderi (“ruins”) or some other general, descriptive category. The concept of “site” is discussed in some of the publications, but not to great length. Perhaps revealingly, the discussion is present in the three publications dealing with the lattest and also most plowed regions of the area.138 Most sites in these landscapes are scatters of pottery and building debris, which makes their observation and interpretation harder and the researchers perhaps felt the need to argue their cases better than when the sites are wellpreserved with clear building remains. The discussion concentrates on the interpretation of the sites and the arguments for these interpretations, not so much on what has been regarded as a site in general. The form of the descriptions is normally the same: irst, a general location expressed as distance and direction from a nearby landmark, e.g., a large modern house or church. No geographical coordinates are used, despite the fact that they could have been derived relatively easily from the topographical maps used as background. The general location is followed by a description of the observations: structures, artifacts, and their distribution in the area, and, in the best case, also a size of the site. No quantitative information is given on the inds, i.e., no sherd counts. Nor has the information been quantiied by the scholars themselves. Research history, if the site has any, is recapped as accurately as can be through references to research literature and archival documents. The descriptions end in a discussion of chronology and the main building phases for sites which have building remains. This is preceded by a discussion of the historical sources, if such can be connected to the site. The text is accompanied by drawings, plans and photographs. Although the survey publications have been produced in a relatively similar manner, using the material for identifying and classifying settlement sites is not unproblematic.139 The deinitions of sites are not given and some of the types of information generally used for classifying sites, such as size or other quantiiable data, were not uniformly available. The array of interpretative categories used for settlement sites in the publications is large: villa, villa urbana, villa rustica, villula, casa rustica, casa antica, fattoria (“rural house”), aggregato rustic (“rural cluster”), sito rustico (“rural site”), insediamento rustico (“rural settlement”), aggregato (“cluster”). In addition, a number of descriptive types are also used: area dei frammenti ittili (“area of ceramic fragments”), area dei frammenti ceramici (“area of pottery fragments”), area di materiale edilizia (“area of building debris”), sito antico (“ancient site”), ruderi (“ruins”), fabbricato/fabbricati (“building/s”) as well as combinations of interpretations, e.g., villa/tomba (“villa/tomb”). After the initial collection of the material, an analysis of the site descriptions was conducted to ind – if possible – the logic for assigning the interpretative categories. The 137 A total of 584 sites/30% of all sites. Squares were used to indicate Imperial period sites without platforms in the Collatia, Fidenae, and Ficulea publications. In the region of Tusculum, many sites with uncertain location were indicated only by numbers or small circles. 138 De Rossi 1967, 11–12; Quilici 1974a, 27–28, 33–34, 51; Mari 1983a, 32–33, 35. 139 For a general need to reclassify the sites from various surveys, see Ikeguchi 2000, 8–11. 23 chaptEr 2 Table 2.1 The deinition of “site” in various surveys in Italy. Area/Site type Architecture Decoration Pottery Tile Size platform, tower, other wall paintings, mosaics, stone revetments wall paintings, mosaics, stone revetments modest, cocciopesto, spicatum none yes yes Cosa (Dyson 1978) Major villa Well-to-do villa minor architecture Small villa Small villa/house site Via Gabina (Kahane and Ward-Perkins 1972) none none Villa platform, cistern, debris less debris some debris wall paintings, mosaics, stone revetments less variation in decorations none many classes Small villa/Farm Undeined Ager Veientanus (Kahane et al. 1968) Villa cistern, mausoleum, bath, paved road, architectural elements “modest” none wall paintings, mosaics, stone revetments marble, bw mosaics none ine wares Villa bath, architectural elements Small farm building debris Small scatters/huts, shacks Molise (Lloyd and Barker 1981) Very Large Debris Scatter none wall paintings, stone revetments, glass mosaics wall paintings, stone revetments, bw mosaics none yes Small farm Smaller site Etruria (Potter 1979) Large Debris Scatter brick or concrete walls, street layout brick, stone or concrete walls; associated graves none none Medium Debris Scatters Small Debris Scatters Rieti (Coccia and Mattingly 1992; 1995) Villa Farm Farmstead Probable occupation sites Biferno (Barker et al. 1978; Barker 1995) Villa/large site building materials Farmstead/small site stone rubble Possible 3rd type Gubbio (Malone and Stoddart 1994) Grade 1: House/Farm/Villa Grade 2: Farm Grade 3: Shed Liri Valley (Hayes and Martini 1994) Villa Major site/small villa Minor site/work-a-day farm Scatter/secondary building, manure/detritus on a ield? Sicily (Fentress et al. 1990) Villa Large farm Small farm/sherd scatter Northern Campania (Arthur 1991) yes yes yes 100–300 m² 3500 m² 1,000–1,400 m² yes 100–300 m² mosaics, plaster, marble ine, coarse occasional painted plaster ine, coarse none none narrow range poor range yes yes 250,000 m2 2,500–10,000 m2 400–2,500 m2 < 400 m2 dolia yes yes yes yes > 2,000 m2 < 2,000 m2 2,000–5,000 m2 marble, mosaic rubble various ine wares less common coarse wares yes 1,500–7,500 m² yes < 1,500 m² yes yes yes yes yes yes large area small area very small area building evidence (platform) plaster, marble, mosaics dense scatter, gentle southward slope less dense scatter ine wares various coarse wares yes > 2,200 m² 1,800–2,200 m² 1,200–1,800 m² undifferentiated, sparsely scattered coarse wares yes small yes 4,000–6,400 m2 2,500–9,600 m2 1,000–2,500 m2 building stone, box tiles, columns building stone sometimes some stone columns glass, metal etc. mosaics Farm Pottery scatter/small house, secondary building, temporary activity area limits discernible > 7,500 m² mosaics yes stone or brick built, clear functional areas, differentiated residential areas stone or brick built, productive and residential parts Villa/villa maritima fewer classes yes wall paintings, mosaics, stone revetments wall paintings, mosaics, stone revetments reasons for describing a site as a villa proved out to be generally fairly uniform. The site must be large in extent, in the best of cases exhibit remains of buildings, present a certain amount of precious decorative materials, such as remains of marble riveting, pieces of mosaics and wall paintings, as well as varying classes of pottery from ine wares to amphorae. Other types feature fewer of the same elements ending with area dei frammenti ittili/ceramici, or scatters of just coarse pottery and tiles. The deinitions and categories of sites were compared to others used in surveys in various parts of Italy and the range of sites as well as differences 24 sourcEs between categories were very similar (Table 2.1). Two to four categories of settlement sites were commonly used based on the size, number and types of inds. The rich sites are most often described as villas, followed by farms, farmsteads and possible settlement sites. The Sites Reclassiied For classifying the sites, the data concerning each one, including the pottery/tile scatters, was tabulated using a simple yes/no principle and, then, based on presence or absence of categories of inds, the 1,941 sites were divided into four main classes. There were 50 ind categories which ranged from building parts and building techniques to pottery and stone types used.140 In addition, the original interpretative category used by the surveyor was taken into consideration. The main class descriptions are as follows:141 Class 1, 333 sites, 17%: clearly identiiable building parts (platform, foundations, living quarters, bath), inds related to production (production parts, torcularia, millstones), waterrelated structures (cisterns, pipes, channels, basins, pools), many decorative elements (architectural elements, sculpture, wall and loor decorations), luxurious building materials (marble, travertine), varying types of pottery from ine wares to amphorae, other inds (burials, inscriptions). Of these, 328 were originally deined as certain or probable villas or villae rusticae. The remaining ive sites were deined as casa rustica, aggregato, area dei frammenti ittili, area di materiale edilizia and sito antico. (Fig. 2.1a.) Class 2, 459 sites, 24%: remains of walls or single building parts (platforms usually), some inds related to production, varying decorative and building materials, varying pottery types. Of these 413 were originally deined as certain or probable villas or villae rusticae. The remaining 46 sites were area dei frammenti ittili (22), sito rustico (6), casa rustica (5), fattoria (3), insediamento (2), area di materiale edilizia (2), aggregato rustico (2), ruderi (2), sito antico (1) and piattaforma (1). (Fig. 2.1b.) Class 3, 449 sites, 23%: no clear building parts, some inds related to production, some decorative elements, various building materials and pottery types. Even for this class most sites were deined originally as certain or probable villas or villae rusticae (334). The remaining 115 sites were deined as area dei frammenti ittili (47), casa rustica (28), aggregato rustico (13), sito rustico (9), area di materiale edilizia (7), fabbricato/ruderi (6), fattoria (3), insediamento (1) and sito antico (1). (Fig. 2.1c.) Class 4, 700 sites, 36%: pottery and tile scatters, sometimes building materials, no decorative elements. Of these, 301 were originally deined as certain or probable villas or villae rusticae. The remaining 399 sites were casa rustica (199), area dei frammenti ittili/ ceramici (154), sito rustico (17), aggregato rustico (13), area di materiale edilizia (6), fabbricato/sito antico/ruderi (5), insediamento (3), casa antica (1) and fattoria (1). (Fig. 2.1d.)142 140 A) Building parts (platform, foundation, cryptoportico, living quarters, bath, walls/spaces). B) Installations and inds related to production (productive part, service areas, torcularium, mill, other types of production). C) Water installations (cistern, well, water channel, sewer, pool, basin, nymphaeum). D) Building techniques (polygonal, opus quadratum, opus incertum/quasireticulatum, opus reticulatum, opus mixtum, opus listatum/vittatum, mortar, brick and tile). E) Decorative materials (wall plaster, stucco, mosaic, opus sectile, opus signinum, opus spicatum, cocciopesto, architectural elements, sculpture). F) Stone types (marble, travertine, limestone, tuff, peperino, lava, other). G) Pottery and glass (impasto, black gloss, terra sigillata, African Red Slip, thin-walled ware, African Cookware, ollae perforatae, plain/cookware, dolia, amphora, lamp). H) Other (inscriptions, coins, cippi, votive artifacts, burials, attribution to owners). 141 The site catalogue in Appendix I includes more detailed information. 142 The distribution of Class 4 sites is heavily concentrated in the Collatia area and the southern edge towards the Bovillae and Tusculum zones is especially clear. They both feature few scatters of pottery/tile, possibly for the reasons related to differences in site formation. 25 chaptEr 2 Fig. 2.1 Distribution of Class 1 to 4 sites in the research area. a) Class 1, b) Class 2, c) Class 3 and d) Class 4. The selected sites do not include the ones used exclusively before the Middle Republican period (279 sites) or some of the very poorly known ruins (termed ruderi, 85 sites). The sites inhabited only in earlier periods tend to be pottery and tile scatters as these two ind types are present at 205 and 227 sites respectively. The early sites form an important group when their locations are compared with the sites that are established or continue to be inhabited later (Fig. 2.2). The original interpretations clearly refer to settlement sites and the amount of uncertainty naturally increases for sites with fewer inds. Sometimes the original survey data has been added to by later publications or even excavation material, e.g., the Collatia site 386, which was found as a simple scatter of pottery, tiles and some building debris, but by excavation proved to have been a small villa site inhabited for several centuries.143 Thus, it was deemed necessary to also include the simple artifact scatters in the possible settlement sites. Many of these can be also remains of other activities, such as burials, temporary buildings, rustic buildings with no residential use, etc. Furthermore, sites recorded as villas in the late 19th 143 Kahane and Ward Perkins 1972, site 11; Collatia site 386; see also above note 115. 26 sourcEs or early 20th century but later completely destroyed have been included since the original interpretation was made based on well-preserved inds and can be thus regarded as reliable. Most of the data is derived from surface surveys, but some excavated sites are also known. A catalogue of one hundred excavated villas in the area of the modern city of Rome was recently published and this collection is used as a comparison to the survey sites.144 The comparison offers an opportunity to see how some aspects of the sites are visible in materials extracted in different ways. In addition, differences in survey inds can be seen depending on the Fig. 2.2 Distribution of sites inhabited only area in which they were found. The hilly before the Late Republican period. areas at the edges of the research area near modern Tivoli and Frascati145 tend to be spotted with fairly well preserved large artiicial villa platforms with quantities of building parts. These platforms have been used either for modern agriculture (Tivoli) or as foundations for modern buildings (Frascati). (Fig. 2.3a.) In the lower areas, the sites tend to be scatters of settlement debris with fewer remains of buildings. These sites are partially destroyed by modern plowing and other agricultural activities. The most notable difference compared to the higher sites with preserved building remains is in the pottery evidence which is more abundant in the disturbed lower sites, e.g., the distribution for sites that feature three or more types of pottery (Fig. 2.3b).146 The deinitions of site classes thus varies slightly depending on their location: e.g., half of the Class 1 and 2 sites with platforms featuring no pottery are located in the higher Tibur IV or Tusculum survey areas and, vice versa, almost all of the Class 1 and 2 sites featuring 3 or more pottery types but no platform are in the lower areas. The differing formation processes of the sites also need to be taken into consideration when classifying them. Changes in the use of the site are dificult to document. Most of the sites have been interpreted as settlements and this has been recently criticized,147 but other interpretations are hard to come by and not backed by relevant data. Building remains with decorations, identiiable functional parts such as heating or water pipes, oil or wine presses, etc. provide reliable material for recognizing the use of the buildings and the site. Even the presence of burials cannot necessarily indicate function as a cemetery, as settlement sites regularly feature burials related to the occupation. It is also relatively common to ind Late Antique or later burials in abandoned villas, but these are found in excavations and not in surveys.148 The villas could also be used for other purposes at the same time it was inhabited, such as road stations, but, again, this kind of use might not be obvious in survey material. Votive inds could be interpreted as shrines or sanctuaries, but the excavations of the Auditorium Villa in Rome 144 De Franceschini 2005; 43 of the sites catalogued there are located in the research area and they are all also part of the survey material. 145 Survey areas of Tibur II and IV, Tusculum, eastern part of Bovillae. 146 The neat cluster in the eastern zone shows the extent of the 1960’s Via Gabina survey by the British School at Rome (Kahane and Ward Perkins 1972) and the potential of more detailed work in that area. 147 Witcher 2006b, 60–63. 148 E.g., Di Gennaro and Griesbach 2003; Dyson 2003, 96; Guldager Bilde 2004; Sfameni 2004, 359. 27 chaptEr 2 Fig. 2.3 Distribution of a) artiicial platforms and b) pottery inds. In b black = sites with one or two pottery types and white = sites with three or more types. have indicated that such inds could also belong to private or semi-private religious activity at a settlement site.149 In one excavated instance, a change in use has been detected: the site Collatia 392 was used as a villa until the 2nd–3rd century AD after which a huge granary was built beside it and the house was abandoned.150 In the vast majority of cases, it would seem justiied to deem the sites as having functioned as settlement sites based on the presence of a variety of domestic pottery and building debris. Chronological Considerations Another important aspect is dating of the sites and variations in the inds also inluence the dates and their reliability. On the slope areas sites, with building parts are dated using building techniques. Pottery evidence is more prevalent in the lower areas and these dates are most often based on ine wares. (Table 2.2; Appendices I–II). Diagnostic inds can be dated quite accurately, but also non-diagnostic sherds of, e.g., ine wares can be dated relatively accurately. Many of the coarse and local wares are Table 2.2 Conventional dates used for building still poorly known and cannot be used techniques and pottery types. very eficiently for dating purposes. The dating of the settlement sites in this study Material Type Date follows these principles and the date Building technique opus quadratum 2nd century BC or earlier given in the survey publications. In many opus reticulatum 1st century BC – 1st century AD opus mixtum 1st–2nd century AD cases, the date is given only generally opus vittatum/listatum 2nd century AD or later as “Republican” or “Imperial” and the brick according to stamps former is considered to be between the Pottery bucchero Archaic 4th and 1st centuries BC. The general black gloss Republican “Imperial” date is taken as ranging from thin-walled 1st century BC – 1st century AD terra sigillata 1st century BC – 2nd century AD the 1st to 3rd centuries AD. Comparison African Red Slip 1st century AD until Late Antiquity between the dates provided by the survey and excavation inds of 43 sites (Table 149 150 D’Alessio and Di Giuseppe 2005. See above note 115. 28 sourcEs Table 2.3 Comparison of dates on surveyed and excavated sites. Excavation dates from De Franceschini 2005. x = evidence from survey only, sett = settled, aban = abandoned. i–ii, etc., refer to building phases. Site Fidenae 78 Fidenae 83 Fidenae 139 Fidenae 141 Fidenae 163 Fidenae 182a Fidenae 250 Ficulea 7b Ficulea 129a Ficulea 144 Ficulea 172 Ficulea 185a Ficulea 192 Fidenae 197 Ficulea 201a Ficulea 398b–c Ficulea 455a Ficulea 473a Ficulea 479 Ficulea 487a Ficulea 489a Ficulea 538–539 Ficulea 536a Ficulea 528c Ficulea 604a Ficulea 601 Collatia 4d Collatia 242 Collatia 284 Collatia 394a Collatia 386 Collatia 454b Collatia 663b Collatia 646 Collatia 649 Collatia 679a Collatia 776 Tellenae 97a Tellenae U7 Bovillae 48 Bovillae 101 Bovillae 103 Bovillae 123 7th 6th 5th 4th 3rd 2nd 1st 1st 2nd 3rd 4th 5th x x x x x x x x x x x x sett? x x ii x sett x sett? x x sett? x sett? x x x x x x x x x iii sett sett x sett sett? x aban? sett sett x x sett sett sett ii sett sett sett sett x x sett sett x x iii sett sett sett iii ii sett x sett sett? sett? i i i?–ii sett–i i–ii i i i–ii x i–ii i i–ii i sett–i i i i–ii i i–ii x ii sett–i i sett? iii ii i sett ii? iv sett sett x ii sett x sett? x ii? iii sett sett x ii sett ii ii x destr iii ii x x sett ii sett iii ii sett i x sett? sett sett x sett sett i–ii i? i? ii i sett ii i ii i i ii i ii ii x ii i i–ii sett ii i–ii ii x aban i ii x iii sett i ii sett ii ii? x i i i i? i? i x sett i x x sett–i i–ii x x sett ii x sett sett sett iii sett sett sett ii ii sett sett sett x i i i x ii sett sett ii iv sett x x sett? x x sett? x i? sett i x x i x x x x x x x x x x x x sett i x x x x x sett? x sett? i i x i sett x sett? ii i x ii sett i iv sett ii sett i i–ii ii iii i–ii i–ii iii ii 6th 7th sett? sett sett? 2.3) reveals that excavation data covers the intense periods of building, i.e., Late Republican and Imperial very well, but that for the earlier periods, excavation does not seem to provide much more data than survey. Late periods are also better caught by excavation. Resurveys also tend to lengthen the occupation periods towards Late Antiquity.151 Approximately 70% of the sites have been dated and the intensively worked northwestern area clearly offers the best material in this respect (Fig. 2.4). The publication time of the surveys is also of relevance as the earlier ones tend to include only Late Republican and later inds. Thus, the northern part of the area published in the 1980’s and 1990’s features almost all of the sites used in earlier periods (e.g., the 454 sites reported to have been occupied in the Archaic period, Fig. 2.5a). The only exception to this pattern is the Ager Tusculanus survey published in 2003 where only a few early sites are published. 151 Cf. the site catalogue for Pergola et al. 2003. 29 chaptEr 2 Fig. 2.4 Distribution of dated (black)and not dated (white) sites. The reason for this can perhaps be found in the modern land use of the area: most of the surveyed region around modern Frascati is heavily built upon or consisting of cultivated slopes where only a few plowed scatter-type sites can be found. The evidence for early use is small amounts of pottery or tiles found in artifact scatters and these kinds of sites are not common in the higher areas. The lack of evidence does not necessarily mean lack of occupation in these regions in the pre-Late Republican periods. In addition to the early periods, the late occupation of the sites is also often present as pottery in excavated sites or in artifact scatters. The intense building and habitation of the Late Republican and Early Imperial period dominates the ind repertoire and inding traces of Late Antique occupation in surface survey can be dificult. More detailed recent ieldwork has often revealed longer occupation periods than was observed in earlier surface survey. As most of the new work has been conducted in the northern section of the research area, the distributions of both early and late sites tend also to concentrate in the north, but less markedly than the early ones (Fig. 2.5a–c, f). Gaps in occupation are also often dificult to recognize and verify. Some sites feature inds from the Archaic/Early Republican periods and then signs of later occupation, e.g., from the Early Imperial period. What happened in between is often unclear and continuity cannot be assumed. Excavated material also yields surprisingly few answers regarding questions 30 sourcEs Fig. 2.5 Distribution of sites in each period. a) Archaic, b) Early Republican, c) Middle Republican. d) 2nd century BC, e) 1st century AD and f) 4th and 5th centuries AD. In a–c white = sites inhabited only before the Late Republican period. In d–e white = new sites. 31 chaptEr 2 Table 2.4 Summary of dated sites. By date Prehistoric Archaic Early Republican Middle Republican 2nd century BC 1st century BC 1st century AD 2nd century AD 3rd century AD 4th century AD 5th century AD Settled New Continue End By class Date No 82 454 306 359 410 522 1,171 1,117 985 141 81 82 403 79 146 184 78 502 15 2 2 1 49 221 179 142 395 433 1,083 978 138 79 0 31 211 116 209 7 77 71 125 840 61 0 Class 1 Class 2 Class 3 Class 4 Total Early sites 281 325 273 506 1,385 270 52 134 176 194 556 8 of continuity of or gaps in site occupation. Early inds tend to be mixed in the later activity layers and later structures also destroy or cover signs of previous occupation. The commonly small excavation areas most certainly identify the dominant building phases. In addition, it cannot be known whether the buildings were abandoned between building phases even with excavation data: new structures either add to or even completely replace previous ones, but little can be said about continuity between the phases. Even the complete replacement of earlier structures could mean merely a need or desire to rebuild without indicating changes in ownership or periods of abandonment. Building remains of the later phases tend to be found on top of each other. In some cases, the earlier habitation is located near the later one, but the distributions of inds do not overlap. This normally applies to traces of occupation of the Archaic to Middle Republican periods compared to Late Republican/Early Imperial habitation. The scatters are commonly in such a short distance from each other that they are described under the same heading in the survey publications. Approximately one hundred such sites (5%) have been identiied in the research area, but the number could be larger if all variations in the distribution of differently dated materials could be taken into account. This can signify change in selection criteria for building sites or simply a need or a desire to avoid earlier debris or even possibly visible structures. The number of these sites is so small, though, that it is dificult to draw more conclusions. Some observations on the development of settlement based on the dated sites can be offered (Table 2.4; Fig. 2.5). The beginning of the dispersed settlement is commonly dated 32 sourcEs to the Archaic period, but as the survey publications include only a few prehistoric sites, this cannot be veriied.152 In the earlier periods, more luctuations in the settlement can be seen as new sites are established and old ones abandoned during each phase. Most new sites are established in the 2nd century BC and in the 1st century AD and most sites are abandoned at the end of the Archaic period and in the 3rd century AD.153 Settlement is at its thinnest density during the Early and Middle Republican periods and reaches its greatest density during the 1st century AD. The distribution of sites is more dificult to interpret as the southern part is virtually empty until the 2nd century BC, but the luctuations of settlement until the 2nd century BC are clear in the central region covered by the Collatia survey. 2.4 writtEn sourcEs The second main group of sources consists of ancient Latin and Greek literature and inscriptions found in the research area. The texts were studied for three purposes: general information about Roman villas and country life, speciic information (history, environment, etc.) concerning the research area and the surroundings of Rome, as well as for persons connected to the research area. The literary references have most often been used to identify landowners and these names have been connected to archaeological remains in the area. This activity started with the earliest work on the ancient history of the Roman region, particularly in relation to Tibur and Tusculum, the places most often mentioned in literature.154 The names of landowners were then connected to archaeological remains, sometimes based on linking evidence such as medieval (and later) place names and inscriptions,155 but very often also based only on the great size and richness of the ruin. The geographic accuracy of literary sources remains at the level of region or town, not that of an individual site. Most of the literary citations used in this study concerning, e.g., the history of the area, have been found in various publications: Real-Encyclopädie der classischen Alterthumswissenschaft, survey reports, local and more general histories, etc. These references were collected and their contents checked. This collection was supplemented by searches in databases of Latin texts.156 In addition to place names, searches were also conducted for general terms important to the study, such as villa, fundus, praedium, hortus, prospectus, etc., in order to gain a better idea of how the words were used. The discussion below is limited to the most important authors and their works concerning the Roman villa and the central argument of this study. Ancient Literature The most important sources concerning the Roman villa are the writings of three authors: Marcus Porcius Cato, Marcus Terentius Varro and Lucius Junius Columella. Each wrote a handbook of agriculture and for each of them, the basic unit of rural life was the villa. 152 See Chapter 8 for more detail. The latter igure may be exaggerated because of the deinition of the general “Imperial” date, i.e., 1st through 3rd centuries AD. 154 E.g., Del Re 1611; Cabral and Del 1779 for Tibur; Mattei 1711 for Tusculum. 155 E.g., Champlin 1993; Calzolari 1994; Del Lungo 1996. For most complete collections of senatorial landowners, see Shatzman 1976 and Andermahr 1998. 156 Brepolis Library of Latin Texts A database and The Roman Law Library (http://webu2.upmf-grenoble.fr/Haiti/ Cours/Ak/). Greek texts were browsed using indices and translations, usually from the Loeb collection, were consulted. 153 33 chaptEr 2 These three books also form a chronological continuum from the 2nd century BC (Cato) through 1st century BC (Varro) to the 1st century AD (Columella). The writings of Rutilius Taurus Aemilianus Palladius in the 4th century AD can be added to the sequence, although the text reads more like a calendar with tasks listed for each season and month than a proper handbook. Each of the earlier treatises is intended as a useful handbook although their intended readerships and scopes may be slightly different. The books are also different in content and style: Cato writes brief and sparse prose, Varro uses wordy dialogues and Columella offers lengthy, detailed instructions. The most important parts with regard to the aims of this study are the ones concerning the selection of a location for a farm and the site for the buildings on it. The information is provided in the beginning of the treatise157 and both Varro and Columella quote Cato on many topics. The agricultural handbooks naturally very much concentrate on the productive and economic side of country life. The cultural and social side of the Roman rural experience can be read from the letters of Cicero and Pliny the Younger. Cicero’s texts are very important for the research area as he owned a villa in Tusculum and wrote a lot about the building as well as about the happenings in his own and his friends’ villas. Late Republican villa life is known almost exclusively from Cicero’s descriptions. Pliny’s villa descriptions concentrate on other areas: Tifernum Tiberinum in Etruria, a seaside villa at Laurentum on the coast of Latium, and the ancestral (and other) villas near Lago di Como in northern Italy. He wrote in the 1st and 2nd century AD forming thus a chronological continuum with Cicero. Pliny’s letters also include the most detailed literary architectural descriptions of a Roman villa.158 The letters of the 4th century AD politician and author Quintus Aurelius Symmachus also discuss country life and even refer to the research area as he had a villa in Tibur. Seneca the Younger also uses villas in his philosophical and other writings as metaphors and examples of various things. Many other authors also mention villas and they are often the stages for many events in Roman history. These texts offer glimpses of the topic from the point of view of individuals, which have been generalized due to the lack of comparative material. Latin poetry is also important in its descriptions of places, persons and buildings. Almost all the Augustan and Early Imperial poets sing the praises of Tibur in their work, Horace particularly so, as he had a villa in the Sabine country near Tibur. Horace writes a great deal about the pleasures of simple country life. Statius describes two buildings in his silvae collection: the Tiburtine villa of Manilius Vopiscus and the Surrentine villa of Pollius Felix.159 Poetic language and imagery are often repetitive and use topoi to portray various themes, but despite the conventions of writing, poems also relect the habits and customs of the era and are useful as sources.160 Roman literature offers many important pieces to combine with the archaeological data, but it also has some limitations. Most of the texts come from the Late Republican and Early Imperial period and – similarly to the archaeological material – the earlier and later periods are relatively poorly known. The works on Roman history have been written in later periods and even though they sometimes mention villas in connection even to very early events, it could be that the later material culture is portrayed as part of an earlier period. Plautus’s plays are some of the earliest texts in Latin, from the 3rd century BC, and there the term villa is used to mean a country house. Cato’s text from the next century describes a fully developed phenomenon, but does not discuss its history. Varro discusses the history and changes in country life, including the deinition of villa, in his work and the conclusion is that the 157 158 159 160 In Cato chapters 1,1–1,5; in Varro 1,6–1,7, 1,12–1,13 and 1,16; in Columella chapters 1,2–1,6. For the most recent attempts at reconstructions of these, see Förtsch 1993. Respectively silv. 1,3 and 2,2. For an analysis of the poets and their relationship to the countryside around Rome, see Mayer 2005, 149–219. 34 sourcEs ancestral villae are villae although they were very different in character to the villae in Varro’s time.161 Changes in material culture were perceived and given signiicance by the Romans. Literary texts are also produced by a very small section of the Roman population, by aristocratic Romans and other persons involved in the same cultural and economic sphere. The opinions, attitudes and customs are of the aristocracy and the wealthy in general. The archaeological material, on the other hand, reveals a wide range of settlement sites from simple artifact scatters/houses/sheds to the emperor Hadrian’s huge villa in Tibur. It is obvious that the countryside was also inhabited by members of the non-elite population, but their point of view is not in any way represented in the ancient literature. In addition, the image given is often based on the writings of one person and thus relects the values, tastes and customs of that person. It is also important to remember to look at each author’s work in the wider context of the period he represents. What can also be relevant considering the topic of this study, selection of the site, is the origin of the main authors and their knowledge of the central Italian environment. Cato’s family originated in Tusculum and he grew up in the Sabine country, near Reate.162 His treatise cites few other authors and the text is based on Cato’s own experiences and knowledge of agriculture. Cato is irmly rooted in central Italy and the same can be said of Varro, whose family came from Reate.163 He cites many earlier Greek authors as well as Cato and other Latin authors whose works have not survived, and also acknowledges personal communications.164 Varro’s approach is perhaps more general than Cato’s, but also irmly based on his central Italian experience and knowledge. Columella’s life is not as well known as the other writers. He came from southern Spain, the province of Baetica, and lived in Latium having landed property in the vicinity of Rome.165 Columella also uses a wide selection of sources, but his own experiences both in Spain and Italy are important as well. The information the authors give might have been intended to be valuable in most locations, but central Italy and the surroundings of Rome played a major part in the formation of the data.166 Inscriptions The Roman epigraphic habit has left behind a great deal of written sources that offer information on different aspects of country life and their ind places can sometimes be pinpointed on a map. The collection used here is based on the survey publications, which have been supplemented with more recent publications, should such be available.167 Some of the texts are related to urban activities in towns and villages of the area, such as those recording construction and repair of public buildings. These offer names and ofices of local magistrates as well as give an indication of the towns and villages that existed in the area. Votive and honorary inscriptions offer an impression of public and private religious activity and also reveal names of individuals interested in the sanctuaries/deities of the area as well as those who were otherwise important to the community. Building inscriptions from various contexts, both public and private, can, in the best cases, reveal functions of buildings in towns and in the 161 Varro rust. 3, particularly chapters 1 and 2. RE XXII,1 Porcius 9. 163 RE Suppl. VI Terentius 84. 164 Varro rust. 1,1. 165 RE X Iunius 104; Colum. 3,9,2 for the locations of the properties. 166 Most of the other main authors were also either of central Italian origin or lived and/or owned land in the Roman region. 167 New editions of CIL VI and XIV are being prepared, but neither could be used here. Some new data as well as good advice was given to me by editor of CIL XIV, prof. Maria Grazia Granino Cecere, whose help is gratefully acknowledged. 162 35 chaptEr 2 Table 2.5 Summary of different inscriptions and sites. Uncertain cases in brackets. ID = instrumentum domesticum. Sites Settlement Burial Town/Village Public building Other Uncertain Unknown Total Building Honorary Votive Burial 19 (6) 5 (2) 4 (1) 31 (12) 0 8 (3) 9 (5) 76 (29) 29 (5) 7 (4) 14 (5) 33 (12) 1 15 (11) 10 (8) 109 (45) 23 (4) 6 (4) 3 29 (5) 2 (1) 8 (1) 8 79 (15) 315 (52) 427 (42) 41 (12) 24 (8) 8 (2) 173 (27) 68 (9) 1,056 (152) Lead pipe Brick stamp ID 70 (2) 3 2 12 3 11 6 107 (2) 1,043 105 68 42 10 71 277 1,616 94 31 15 3 3 5 1 152 Statue Other Unknown/False Total 36 0 0 3 (1) 0 2 (1) 10 51 (2) 16 11 3 8 1 12 4 55 60/37 27/2 12/0 25/5 6/0 34/2 25/0 189/46 1,742 (69) 624 (52) 162 (18) 215 (38) 34 (3) 341 (43) 418 (22) 3,536 (245) countryside. Most of the inscriptions found in the surroundings of Rome derive from burials and they range from simple mentions of names of the deceased and of those who set up the commemorative monument to elaborate lists of ofices for the members of the upper echelons of Roman and local society. The most common inscription type is the brick stamp, which is often very important for dating the buildings and giving an indication of the use periods of the site. Pottery stamps on black gloss, terra sigillata, amphorae and dolia serve much the same purpose. (Table 2.5.) Although thousands of inscriptions have been found in the Roman countryside, as a source material they are perhaps more problematic than literature. Inscriptions have been recorded and collected from very early times, which means that the original ind contexts were commonly not recorded at all. Some of the text types are dificult to understand and to interpret without a context, e.g., honorary texts could be set up either in a public or private context and this knowledge is signiicant for their interpretation. Knowing the type of monument in connection to burial inscriptions would add meaningfully to understanding the texts. In some cases, information on the original ind spot is known, e.g., very often the name of a vineyard where the ind was made is known. Unfortunately, the exact (and sometimes even the general) location of the vineyard has often been forgotten with time, changing ownership, etc., and can no longer be found any more very accurately. Inscriptions can also be moved around relatively easily and great collections have been gathered in the Renaissance and Baroque palaces and villas of Rome and its surroundings. If the inscriptions’ movement history is not well recorded, it can be dificult to know where it was actually found versus where it was only stored and displayed. In villa studies, inscriptions have also been used in a similar manner as literary references, to identify landowners or to connect other data to archaeological sites.168 The indication of landowning is often indirect information: burials refer to landed property of the deceased or his/her family in the area; private votive and honorary inscriptions found outside cities probably come from villas; names in the genitive case on lead water pipes indicate water rights of landowners in the ind place or somewhere nearby. Public inscriptions, such as public honors or votive texts in large sanctuaries, have also been used to interpret a possible estate in the area – the person honored or the one setting up the stone was important to or interested in the area and thus could very well also have been a proprietor in the region.169 All of these methods of attributing ownership can be regarded as uncertain.170 The lack of contextual data diminishes the possibility of interpreting the text correctly. Some general trends detected in well-documented cases can perhaps be used to help interpretation. The large burial monuments of the Late Republic and Early Imperial periods tend to be placed by the main roads to achieve prominence for the deceased and his family. The 168 169 170 E.g., Shatzman 1976; Andermahr 1998; Valenti 2003, 66–91. Andermahr 1998, 14–26. E.g., Bruun 2000. 36 sourcEs family certainly owned the burial plot, but this does not necessarily indicate an estate nearby. During the 2nd century AD and later, the trend was to build burial monuments near the villa buildings and this could be considered a possibility of interpreting the later burial inscriptions as part of the estate.171 For lead pipes, the archaeological context is also signiicant, as they are often found in certain settlement contexts and names in the genitive could relatively safely be interpreted as owners, but other possibilities also exist. Public/semi-public baths set up by private citizens on their estates are known. Aqueducts ran across the countryside and clusters of lead pipes could also indicate a castellum plumbeum or other point of distribution of water – the names in the genitive on pipes could again indicate landownership, but as the end point of the pipe is unknown, it cannot be regarded as a certain interpretation.172 In addition, honorary inscriptions found in the countryside without a public building nearby could be interpreted as private and thus an indication of landownership. Each case has to be considered separately within its epigraphic and archaeological context. 2.5 conclusions In the past decades, much has been written about the problems of survey inds and how to interpret them. Survey methods have been discussed perhaps the most: What are the most reliable and representative ways of gathering data? Issues affecting the results and how to best take them into account have also been discussed at length.173 The survey material used here is produced with very traditional methods and also the publications are missing many types of data that would be regarded necessary for reliable interpretations in most recent work. It is obvious that the archaeological material features many problematic aspects, but at the same time, it is the most complete and, in cases of destroyed sites, the only material available. What should be remembered is that the southwestern part of the research area has been poorly studied and that the earlier periods are covered only in the eastern and northern parts. What is perhaps lost in quality of the data is partially reclaimed by the extent and quantity of the material. The research area covers a larger extent of Rome’s surroundings than has recently been taken under consideration. The number of sites is also very large and it is probably a representative sample of all settlement sites in the area. The sheer quantity evens out some of the errors caused by the problems of data collection. The information provided by archaeological data is complemented by the written sources and environmental data which also naturally have their limitations. The period best covered by archaeology and written sources is between the 1st century BC and the 4th century AD. The 2nd century BC is already relatively poorly known and periods before it as well as those after the 3rd century AD, are even more uncertain. The three data sets intertwine, representing the same phenomena from different points of view. 171 Griesbach 2005. Bruun 2003 with references to earlier literature. 173 Witcher 2006b offers a good survey of most relevant literature despite the fact his emphasis is on the theoretical issues related to archaeological surface survey. 172 37 3 gEology and roMan villas 3.1 background Geology determines many other environmental features. Topography, soils and water resources all depend mostly on the basic geology. Despite its fundamental nature, geology did not play a major part in the advice for selecting a good villa site. On the contrary, geological aspects are not mentioned at all in the classic passages describing an ideal location for a villa.174 It is understandable that quality of soils or water resources were more relevant factors in the deinition of a good farm site, but geology did matter when building materials for the villa were needed,175 and when a suitable building site was sought.176 If the villa needed to be built from the ground up, a builder was required and the advice given by Cato for drawing contracts deine the obligations of each party.177 The building contractor was responsible for the walls and other building parts, but the owner was expected to provide the construction materials including their transport to the site: stone, timber, lime, sand, water, straw and earth. The scrupulous landowner naturally tried to cut the costs of both acquisition and transportation by using local products, preferably from the estate itself.178 Calculation of material cost was advised for all projects, even for such modest ones as construction of fences, and one of the deciding factors for choosing a building technique was the availability and price of stone and sand.179 The basic building materials are not mentioned in connection to large buildings since the emphasis is on luxurious marbles, metals and other costly or rare materials,180 but depending on the size of the project, the basic costs might have been considerable.181 Good building stone, clay pits, metal ores as well as volcanic sand for mortar from the estate were also considered good sources of income.182 174 E.g., Cato agr. 1,3; Colum. 1,2,3;1,9,9. Colum. 1,2,4. 176 Pallad. 1,8,2. See also below note 183 on foundations. 177 Cato agr. 14,1–3. 178 Varro rust. 1,16,3. 179 Colum. 9,1,2;11,3,2; Vitr. 1,2,8. 180 References on the luxury of villas are many, but even the longest passages on villas do not mention basic building stones, e.g., Cicero’s letter to his brother Quintus describing the construction works at the latter’s villa in Arpinum (ad Q.fr. 3,1) or Pliny the Younger’s descriptions of his two villas (epist. 2,17 and 5,6). 181 Estimates for the costs of building villas have not been presented, but cf. DeLaine 1997 for the building of the Baths of Caracalla in Rome. Shatzman 1976, 11–46 presents some igures for property values of Republican senators and Duncan Jones 1974, 17–32 analyses the economics of Pliny the Younger including the estates and buildings he had. Duncan Jones 1977, 124–126 discusses construction prices for mostly public buildings and roads (private building on p. 125), lists of prices in Italy on pp. 157–162. Diocletian’s Price Edict (in Frank 1940) does not refer to any other building materials except for timber (XII,1), but lists fairly high daily wages for builders (VII,1). 182 Varro rust. 1,2,22–23. 175 gEology Another aspect of importance for building is the laying of solid foundations. Roman construction was, of course, able to deal with poorer qualities of sites using piles and other methods of foundation, but, nevertheless, a solid base for a building was preferable.183 The ancient methods for inding a solid base, if it was not readily visible, are not known, but they were probably similar to the ones used until drilling became possible. One method used until quite recently was to dig down until a solid base was found. Its hardness was tested by dropping a heavy rock on it and listening to the sound the base emitted. If the sound was not sharp and dry and/or if the stone sank, it was necessary to dig more. Hard and compacted stone has the best load bearing capacity for the heavy stone and mortar architecture used by the Romans. On the other hand, semi-hard rock as well as dry sand, gravel and clay can also be considered as good terrain for building even quite large complexes.184 The relationship of Roman settlement and its geological environment has been studied only a little in Italy. The distribution of sixteen villa sites in Basilicata was compared to the road system, elevation contours, geology and modern distribution of forests. Almost all sites were placed on Plio-Pleistocene clay as compared to the more unstable Early Miocene lysch.185 Usually the sites were close to – or on the boundaries of – two geological zones, clay and limestone. The clay is more easily cultivable than the lysch or limestone. The lysch zone might also have been forested in antiquity as it is today,186 which would have provided other resources such as lumber, charcoal and pastures.187 Settlement on the Salento peninsula and in Murge, both in Apulia, was also studied with respect to geology. The Salento study was based on the distribution of modern villages and they were noted to be situated within a 0.5 km distance from boundaries of two geological outcrops. The contact zones offer better chances of inding ground water and the different geological zones provide possibilities for different agricultural uses. In the Murge region, the settlements of possible Roman origin were found to be in locations at 10 km distance from the sea, within easy reach of the Roman main road, and in pockets of Quaternary and Holocene deposits situated within an earlier limestone terrain. These sites provided better water resources as well as possibilities for rotational crops.188 In these studies, geology emerges as a signiicant environmental background factor affecting other resources, particularly agricultural and hydrological ones, but it is not studied as a resource itself. The third study was of Southern Etruria and the main question of the locational analysis concerned the agricultural productivity of the area. Modern land use and the distribution of Roman sites were compared to the geological formations and it was noted that both were most commonly located on volcanic deposits – not perhaps surprising considering that the study area is mostly covered with volcanic geology. Alluvial deposits, sand and conglomerate were deemed not preferred for, based on their poorer agricultural potential.189 In the research area, the archaeological survey reports always provide a brief description of the geology of the area, but this information is generally not used in any way in the analysis of the sites or in the syntheses. The geology of the site is sometimes mentioned as, for example, 183 Vitruvius treats foundations in connection to towns (1,5,1), temples (3,4,1–2) and theaters (5,3) and recommends solid ground or digging until solid base is found. Chapter 6,8,1 deals with the stability of the building and the importance of good foundations. Colum. 1,5,9–10 gives advice on how to build a solid foundation on a slope. See also Pallad. 1,8,2. 184 Giuliani 1990, 121–123. 185 Clay interspersed with layers of sandstone and limestone. 186 Forests in the lysch zone may also the mean that the sites there have not been found due to poor visibility. This was not mentioned by the authors. 187 Buck and Small 1980. 188 Mørch 1995. 189 Goodchild 2007, 146–150. Perkins 1999, 48–50 also discusses geology briely in relation to his study area in Etruria: Pleistocene and limestone/marl formations were sought after, alluvium avoided. The reasons for these tendencies were harder to grasp, but better agricultural potential, particularly for vineyards, is one possibility. 39 chaptEr 3 in the case of the Villa Centroni.190 It was built on crest of a lava spur using local lava stone and volcanic sand was dug from under the building for use in mortar.191 The use of geological formations as a resource for building material in Rome and its vicinity has been studied, but only from the point of view of the city of Rome or other urban centers.192 In the following, the geology of the Roman Campagna will be described and analyzed from the point of view of building activity in the countryside. The irst three sections provide the basic data from geological studies, ancient sources and archaeological research and this will be analyzed and discussed in the inal section. 3.2 gEology of thE roMan caMpagna The Roman region is characterized geologically by two main features: volcanic formations originating from two different centers and the limestone slopes of the Apennines. Alluvial deposits in the river valleys and lacustrine basins could be considered a third major feature as they are chronologically different from the two other formations (Table 3.1; Plate II.1, nos. 2–3).193 The oldest geological formations in the region are the pre-Apennine limestone hills, Monti Praenestini and Monti Tiburtini, in the northeastern corner of the research area. The deposits are either fairly compact white or whitish limestone or off-white, greenish or grayish marly limestone or marlstone. Flintstone and lint nodules also occur. The next major deposition phase occurs in a marine environment and these sediments can be found in the north central part of the research area deposited over the lower slopes of the Monti Cornicolani. The deposits are clay, sand and marl with plenty of faunal remains and the layer can be as thick as 800 meters. Later marine phases did not leave great deposits in the region. Some luvial deposits with the same characteristics can be found in the northwestern part of the area. Formations of volcanic origin cover 70% of the research area. (Table 3.1; Plate II.1, nos. 4–5, 6–10.) The volcanic activity in the Roman region originates from two centers: one is located to the northwest of Rome in the Monti Sabatini and it represents the earliest phase of activity starting 600,000 years ago. The second center is located in the Alban Hills southeast of Rome and its activity lasted until 20,000 years ago. Most of the deposits in the research area originate from the Alban Hills, but some deposits from the Sabatine complex occur in the northwestern and northern parts. The deposition of volcanic ash and lava created a whole new landscape subsequently changed by erosion. Eruptive activity from both centers deposited various materials according to the type of eruption and distance from eruptive centers: tuff, volcanic ash sand and gravel, lava, as well as various other pyroclastic materials. The tuff is deposited by pyroclastic lows or by air fall ash. They are generally grey, yellow or red in color with a granular or stratiied structure. Many are compacted or have experienced changes that have made them more lithoid; many incoherent beds also occur. The last phase of activity created deposits of peperino stone, a distinctive grey-based tuff/breccia “peppered” with smaller and larger inclusions of varying volcanic origin as well as with bits 190 Bovillae site 101; Di Matteo 2002; 2003b. Cf. Ricciardi 2005 on volcanic sand shafts illed with 2nd century AD materials right next to the villa found in the Via Grotta Perfetta to the south of Rome. 192 Cf. Lugli 1957, passim; DeLaine 1995; 2001; Jackson et al. 2005; 2007; Jackson and Marra 2006; Giampaolo et al. 2008. 193 The following description is based on the geological map of Italy, Carta geologica d’Italia, foglio 150 to scale 1:100,000 (1967) and more recent geological literature concerning the area: Cosentino et al. 1993; Amanti et al. 1995; Arnoldus-Huyzendveld et al. 1997; Funiciello and Parotto 2001; Molinaro et al. 2001; Funiciello et al. 2008. The volcanic activity in the area has been recently much discussed and the information provided by one of the contributions, Karner et al. 2001a, has been of great importance in clarifying formation sequences and nomenclature. 191 40 gEology Table 3.1 Geological formations in the research area. Number of all sites 1,941, of Early sites 279 and of sites established in the Late Republican period 184. Note that one site can be located on more than one formation. Map Formation type 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Appennine Limestone Limestone, White Limestone, Compact Limestone, Marly Limestone, White Limestone, Mixed Flint Marlstone Marine Clay Fluvial/Eolic Eolic (?) Sand Sand, Yellow-Brown Fluvial Sand/Gravel Sand, Yellow Puddingha Sabatine Paleosol Sabatine Tuff Tuff, Grey incoherent Tuff, Palatino Tuff, Yellow Via Tiberina Tuff, Fidenae Alban Paleosol Alban lava Lava, Capo di Bove Lava, Vallerano Lava, Fosso Benzone Alban tuff Tuff, Aniene Tuff, Villa Senni Peperino Tuff, Bagni di Albule Alban pozzolana Pozzolana, grey Pozzolana, red Alban miscellaneous Scoria Lapilli Scoria/Lapilli Lacustrine Cones/Lapilli Scoria/Lapilli, Dark Lava/Scoria Travertine Alluvial Alluvium Lacustrine, craters Terra Rossa Miscellaneous Lapilli, pedological changes Colluvium Area ha Area % Sites Sites % Early Early % LR LR % 1,179 367 261 193 155 127 45 33 1,109 328 86 69 66 42 65 1,535 2,199 1,551 375 189 84 465 69 1,705 50 19 13,751 5,677 5,661 1,857 556 3,478 8,839 3,478 3,146 1,868 498 390 171 119 47 53 2,361 5,936 5,636 300 1,380 84 52 32 2.5 0.8 0.6 0.4 0.3 0.3 0.1 0.1 2.3 1.5 0.2 0.1 0.1 0.9 0.1 3.2 4.6 3.3 0.8 0.4 0.2 1.0 0.1 3.6 0.1 0.0 28.8 11.9 11.9 3.9 1.2 7.3 18.5 7.3 6,8 3.9 1.0 0.8 0.4 0.3 0.1 0.3 5.0 11.8 11.2 0.6 2.9 0.2 0.1 0.1 14 0 8 3 2 1 0 0 71 17 5 5 3 1 3 121 176 116 26 17 17 23 85 84 0 1 786 351 289 112 34 613 415 198 188 124 18 15 10 5 0 16 75 176 158 18 37 2 1 1 0.7 0.0 0.4 0.2 0.1 0.1 0.0 0.0 3.7 0.9 0.3 0.3 0.2 0.1 0.2 6.2 9.1 6.0 1.3 0.9 0.9 1.2 4.4 4.3 0.0 0.1 40.5 18.1 14.9 5.8 1.8 31.6 21.4 10.2 9.7 6.4 0.9 0.8 0.5 0.3 0.0 0.8 3.9 9.1 8.1 0.9 1.9 0.1 0.1 0.1 0 0 0 0 0 0 0 0 32 4 3 1 0 0 0 35 62 46 3 8 5 0 2 2 0 0 88 70 12 6 0 72 24 48 7 1 1 2 3 0 0 0 2 25 23 2 4 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 11.5 1.4 1.1 0.4 0.0 0.0 0.0 12.5 22.2 16.5 1.1 2.9 1.8 0.0 0.7 0,7 0.0 0.0 31.5 25.1 4.3 2.2 0.0 25.8 8.6 17.2 2.5 0.4 0.4 0.7 1.1 0.0 0.0 0.0 0.7 9.0 8.2 0.7 1.4 0.0 0.0 0.0 4 0 3 1 0 0 0 0 9 3 0 1 1 1 0 13 22 14 5 1 2 7 12 12 0 0 86 43 15 18 10 40 18 22 28 17 2 1 2 2 0 4 9 2 16 2 4 1 1 0 2.2 0.0 1.6 0.5 0.0 0.0 0.0 0.0 4.9 1.6 0.0 0.5 0.5 0.5 0.0 7.1 12.0 7.6 2.7 0.5 1.1 3.8 6.5 6.5 0.0 0.0 46.7 23.4 8.2 9.8 5.4 21.7 9.8 12.0 15.2 9.2 1.1 0.5 1.1 1.1 0.0 2.2 4.9 1.1 8.7 1.1 2.2 0.5 0.5 0.0 Permeability Hard/ Soft good non-perm. non-perm. medium medium non-perm. non-perm. non-perm. H H H H H H H S poor non-perm. medium non-perm. poor poor S S S S H S good medium good medium poor S H H H S medium medium medium H H H poor poor medium medium H S H H medium good S S medium poor good medium good good medium good S S H S S S H H medium medium medium S S S poor good S S of carbonates, around large and small craters. Lava deposits are found close to the craters, radiating outwards in long and narrow strips and they consist invariably of hard stone. They form distinct ridges in the topography as softer materials have been eroded from around them; the most famous ridge is probably Capo di Bove southwest of Rome on which the Via Appia runs. Two major deposits of volcanic ash (pozzolana) in red, violet and black sand and gravel can be found forming wide rings around the Alban Hills. The static periods between eruptive phases were sometimes quite long and the volcanic deposits were exposed to erosion and pedogenetic processes producing paleosols. The beds are clay, sand and gravel, but they also feature luvio-lacustrine materials with plenty of faunal and loral remains as well as remains 41 chaptEr 3 of early humans and Paleolithic stone tools. The most recent geological formations were deposited during the last 130,000 years. (Table 3.1; Plate II.1, nos. 11–14.) The most signiicant of these is the extensive travertine area in the northeastern part of the research area. Travertine is a biochemical limestone formed around and, because of, thermal water springs in the area. The stone is porous and contains plenty of loral and faunal remains. The slightly rising area north and west of the travertine and partially on limestone is characterized by an extensive Terra rossa soil formation. The limestones have eroded to form ine soils, which are red or ochre clay and sand, sometimes with volcanic inclusions. The last 10,000 years have seen the formation of mostly sediments related to movement and deposition of material by water. The two main rivers in the area, the Tiber and Aniene, have deposited most of the alluvium. The last glacial phase caused a great lowering of the sea level, which in turn forced the Tiber to carve a deeper channel for itself in order to be able to low to the sea. When the sea levels rose again, the valley was illed with deposits which are still continuing to form. The alluvial deposits are clay, sand and gravel. Recent lacustrine sediments, clay and sand, can be found in the small craters formed in the last volcanic phase and later illed with water. Variations in the geology of the area are great, ranging from loose alluvial sediments to hard lava. Many of the deposits can be used with success for various construction needs. The lithic tuff is easy to cut and in contact with air, will harden and become fairly resistant to erosion. In moist conditions, volcanic sand reacts with calcium hydroxide and harden to a rock-like condition. Lava is a good building stone for structures requiring great durability. Limestone and travertine can be used in two ways: the more durable kinds for building and the others for burning lime.194 Before looking into the written sources and archaeological evidence for their extraction and use in ancient building activities, a short discussion on the possible changes in the geological landscape after the Roman period is necessary. 3.3 thE changing landscapE of thE roMan caMpagna The geological formation of the Roman Campagna began over 200 million years ago and continues today. Geological processes are slow, lasting hundreds of thousands if not millions of years, but the catastrophic nature of volcanic activity shows that rapid changes can also occur. Almost 90% of the formations were deposited before the Holocene began 10,000 years ago. The earlier beds are thick and even 2,000 years of erosion have not erased them – most of the geology seen by the Romans must have been the same as can be observed today. The deposition of alluvium in river valleys, however, is a continuing process that may have changed the landscape considerably. The irst studies of river valley ills in the Mediterranean area divided them into two chronological phases, called simply Older and Younger Fills. The former was deposited during the Pleistocene period and was later cut by the latter whose formation was related to the extensive looding of the period AD 300–1500.195 This theory has been discussed and criticized for various reasons in the past decades and current research points towards a good deal of regional variation for reasons ranging from general climatic changes to land use and human inluence.196 Unfortunately, no detailed geological study of the development of the river valleys in the Holocene in the research area has been conducted. The closest 194 195 196 DeLaine 1995; 2001; Arnoldus-Huyzendveld et al. 1997, 13–14; Molinaro et al. 2001, 60–67; Jackson et al. 2005. Vita-Finzi 1969. A good survey on the discussion in general can be found in Brown and Ellis 1995, 45–51. 42 gEology parallel comes from Etruria where four river valleys were analyzed. The results show great variation between river basins even in that small area. Despite problems of generalization, some common occurrences can be observed. Roman period river channels were probably in most cases braided and changing rapidly, unlike the modern rivers running in fairly deeply incised channels. Most of the alluvial deposits over possibly Roman deposits seem to be from medieval and later times.197 Archaeological and paleo-environmental studies of the Tiber river valley in Rome and towards the mouth of the river indicate similar developments. The sea level was lower than the current one by one meter from the Archaic to 2nd century BC after which it began to rise, reaching a level of one meter above the current sea level by AD 400. This development would have been matched by the river valleys with lower beds when the sea level was lower and rising valley loors when the sea level was higher. 198 The effects of this process can be seen, e.g., in the Auditorium Villa located in northern Rome on the Tiber lood plain. The villa probably suffered from continuously rising lood levels starting from the 4th–3rd centuries BC and was perhaps inally abandoned because of looding. It was later covered by 1–1.5 meters of post-Roman alluvium. 199 Recent archaeological and geological studies in the eastern parts of Rome also indicate processes of erosion and deposition. In the river valleys, archaeological sites, such as roads and agricultural installations were discovered under sediment layers varying from 0.8 to 5 meters in thickness. The intense Roman land use of the area was found to be based on careful soil preservation techniques indicating problems with erosion and land degradation, but also active measures taken to control them.200 Fairly rapid changes in the environment can also be seen in a more limited study along the Via Nomentana, where the valley below the excavated area was silted up and turned into a swamp by Late Antiquity.201 Varying periods of deposition and erosion related to a cyclic climatic system have been recorded in Italy with corresponding phases elsewhere in Europe and North Africa. The beginning of one cycle occurred in the Archaic period lasting until 520 BC. The climate was very similar to the current one with a relatively high degree of pedogenetic activity. This was followed by a colder and rainier period, with deposition of alluvium, lasting until 350 BC. The Middle Republican, Late Republican and Imperial periods until AD 150 were much like today with increased pedogenetic activity. The rest of the Imperial period until AD 350 was very hot and dry with little pedogenetic activity. The Late Antique climate was again similar to the current one but with more erosion until AD 500. After this, the cycle started again with a cold and rainy period ending AD 750.202 Similar cyclicity is also indicated by a comparison of geological and historical information on loods in central Italy: there were few loods during the Republican period until the 2nd century BC as well as again after AD 200. Major looding occurred in the period of perhaps most intense land use with the warm and relatively wet climate.203 Thus, it would seem most likely that the Roman period was preceded and succeeded by deposition of alluvium, but the period itself would probably have been one of erosion owing to climatic conditions as well as to intense land use. The implications of this development for the geology are relatively small as most of the formations observable today were exposed whether more or less alluvium had been deposited. The lower edges of the river valleys might have been different with formations slightly more exposed, if the valleys have been 197 Brown and Ellis 1995. See also Potter 1976 for data and interpretations. Arnoldus-Huyzendveld 1995. But see also Lambeck et al. 2004 for more moderate igures of changes in sea level. 199 Arnoldus-Huyzendveld 1995. 200 Quilici and Quilici Gigli 1993, 335; Molinaro et al. 2001; Pracchia 2001. 201 Di Manzano 2001. 202 Molinaro et al. 2001. 203 Giraudi 2005. 198 43 chaptEr 3 as deep and wide as today, or less exposed, if the valleys have been shallower and narrower. This might also have resulted in changes in topography as well as have implications for the retrieval of archaeological evidence. The geological landscape was well known to the Romans and its products used with increasing intensity as shown by written sources discussed in the next section. 3.4 writtEn sourcEs for thE usE of gEological rEsourcEs Many of the rocks found in the Roman area are well-suited for building purposes and this was discovered early on by the people living in the area. Most of the stones, and sometimes also the locations they were extracted from, are described in texts. They were used locally and in Rome for private and public building, although most of the texts do not refer to speciic buildings. The following combines ancient and modern descriptions of the stones as well as some archaeological data concerning the quarries. The dates for the use of stone come from Rome, which is the only place where the topic has been studied. Tuff was considered a soft building stone with problems of eroding if not properly covered.204 Five types of tuff from the region are mentioned by name in the ancient sources: Ruber, Fidenae, Pallens, Alba and Gabii.205 Three of the names also indicate the source as Fidenae, Alba and Gabii were extracted from the areas of the ancient towns with the same names. Ruber and Pallens are more dificult cases, but the former has often been connected to the red stone found on the west bank of the Tiber near Saxa Rubra and Grottarossa. Ruber could also refer to the red stone quarried from the banks of the Aniene mentioned by Strabo.206 The rock surface weathers to reddish brown which could explain Strabo’s description. However, the color of the freshly cut stone is not commonly red, but rather a pale grayish-yellow. This in turn has been associated with Vitruvius’s description of Pallens, “pale, wan, pallid” tuff. Considering the popularity of the stone in Vitruvius’s time and the fact that he is the only one to mention that type of tuff, it would seem plausible to call the Aniene tuff Pallens.207 The Fidenae tuff is traditionally considered to be the irst to be imported to Rome after the city was conquered in 426 BC. It is a red, moderately well-lithiied tuff with inclusions of black scoria and lava giving it its modern name, red tuff with black scoria (It. tufo rosso a scorie nere). It is soft, easy to cut and so it has been regarded as suitable for unsophisticated stone work.208 It is found on the hill of Villa Spada on the Via Salaria and the area east of it. The quarries can be placed with relative certainty on the northern part of the hill, where analysis of old maps and photographs has revealed a probable location.209 204 Plin. nat. 36,166; Vitr. 2,7,2; 2,7,5. Tuff of the Roman region is referred to as lapis or saxa, whereas tofus/tophus is used for stones of other areas, e.g., Vitruvius (2,7,1) uses it for Campanian red and black stones as well as for white stones in Umbria, Picenum and Venetia. The chapter in Pliny the Elder’s Naturalis Historia describing tofus (36,166–167) mentions Fidenae and Alban stones (as well as Umbrian and Venetian), but as “other soft stones,” not explicitly as tofus. Cf. Pallad. 1,8,2 where lapis and tofus are also listed separately. 205 Plin. nat. 36,167; Strab. 5,3,10–11; Tac. ann. 15,43; Vitr. 2,7,1; 2,7,5. 206 5,3,11. 207 Cf. Calci and Mari 2003, 184. 208 The treatises on building stones used in Rome are old, but unfortunately, no recent reviews or studies on the subject exist. Recent archaeological excavations in Rome reaching early levels can also change the dates for the beginning and end of use of the various stones. For more modern surveys of building stones written by geologists, see De Rita and Giampaolo 2005; Giampaolo et al. 2008. For the use of Fidenae tuff, see Frank 1924, 16–17, 21–22; Blake 1947, 26–27; Lugli 1957, 184–186, 235–255, 302–304; Claridge 1998, 37; Jackson et al. 2005. 209 Quilici and Quilici Gigli 1986: 158–161 site 51. The Fidenae area is today otherwise almost completely transformed by a large modern suburb with adjacent roads. 44 gEology The use of the Alban and Gabii stones in Rome starts in the 3rd–2nd century BC.210 Both of these are called peperino in modern terminology and the name derives from the appearance of the stones: grey matrix peppered with black and white inclusions.211 Peperino is well-lithiied and thus harder than the other types of tuff. It does not weather as fast or as easily and consequently it was used for load bearing structures as well as for features not protected by roofs. Gabii tuff has been regarded as the poorer of the two in quality by most modern researchers. Gabii and Alban stones were also regarded as ire-resistant and Tacitus mentions that after the AD 64 ire of Rome many buildings were erected using these stones for the sake of this quality.212 This characteristic has recently been veriied by scientiic tests which show that tuff does not have a great thermal expansion and as a result does not fracture as easily as, e.g., limestone.213 The quarries for the Gabii stone are situated on the edges of the eponymous town close to the south bank of the Aniene, which made it easier and cheaper to transport to Rome – and so possibly also more popular to use – compared to the Alban stone which had to be hauled by road.214 Alban peperino was mostly quarried from the area below the modern town of Marino, southeast of Rome.215 As mentioned above, the Pallens tuff referred to in ancient sources is probably the same as the Aniene tuff quarried from the banks of the river. Its use began towards the end of the 2nd century BC and it was the most popular volcanic rock used in the Augustan period. Aniene is a lithoid tuff with a grayish-brown yellow color which has given it its modern name, “tawny tuff” (It. tufo lionato). It is fairly hard compared to the other stones and this has been often regarded as the reason for its fairly late extraction, i.e., only development of stone quarrying techniques enabled its use.216 The quarries along the river are situated on both banks and they are the largest ancient tuff quarries in the Roman region.217 The use of the stone in Rome in large quantities was made easy by transport on the river, as pointed out by Strabo.218 In addition to tuff, Pliny the Elder 219 mentions a Tusculan stone, silex Tusculanum, otherwise not known from ancient sources. It has not been discussed by modern researchers until quite recently – possibly because the stone type has not been used or recognized in Rome. Silex usually refers to hard stones which could mean, e.g., lava or lint.220 The quarries east of ancient Tusculum produce a grayish-brown tuff, a compact and hard stone commonly known as sperone.221 This could very well be the silex mentioned by Pliny. Its quarries lie east of the research area. In addition to rocks, another volcanic product widely used in construction is sand and gravel or pozzolana. It was used for mortar from the 2nd century BC onwards to reduce the amount of lime needed.222 The ancient sources mention two pozzolanas: pulvis Puteolanus 210 See Karner et al. 2001b, 390–391 for the suggestion that the early use of peperino in Rome would have been of the local Tufo del Palatino or some type of Cappellaccio, both similar to peperino. 211 Sperone is also a term used for the Gabii stone (e.g., Frank 1924, 24–25; Lugli 1957, 184–186) but geologically, this stone type should be connected to a different formation located a few kilometers south of Gabii. 212 ann. 15,43. 213 Frank 1924, 24–26; Blake 1947, 34–39; Lugli 1957, 184–186, 302–306; Jackson et al. 2005. 214 Frank 1924, 24–25 based on Strab. 5,3,11. 215 Lugli 1957, 302–304 on the Alban quarries and 306–308 on the Gabii quarries. See Dal vulcano all’uomo 2003 for the Marino quarries and use of both peperino and sperone in ancient and modern times. 216 Frank 1924, 14–16, 26–28; Blake 1947, 29–34; Lugli 1957, 171–172, 186–186, 253–255, 309–310; Jackson et al. 2005. 217 Quilici 1974a, 62–82 site 4 and 143–169 site 55 for the main quarries. 218 5,3,11. 219 nat. 36,135. 220 OLD s.v. silex. 221 Jackson et al. 2005. 222 Plin. nat. 36,175. 45 chaptEr 3 from the Campi Flegrei region in Campania and then, as a more general type, harena fossica or “pit sand.”223 The irst is a pale yellow, ine powder which differs considerably from the coarse, red or grey sand of the Roman region. The two main pozzolana formations in the research area are the earlier red and the later grey or violet – and these are probably what is meant by “pit sand” in written sources. The red sand is better in quality as it is iner and more homogeneous, allowing more surface for the chemical reaction with calcium hydroxide, the basic requirement for its use in mortar. The best known quarries in the Roman region are all situated outside the research area: San Paolo fuori Le Mura by the Tiber, Casale della Caffarella on the Via Appia and Grotta Perfetta on the Via Ardeatina. All of these are very close to Rome and also by a river or a main road for easy transportation. Some differences in the methods of extraction of the various deposits exist: closer to the Tiber, tunnelling is required whereas closer to the Aniene, extraction from open quarries is also possible.224 Travertine was regarded a very good building stone due to its durability and its only poor quality seems to have been its susceptibility to ire, i.e., the stone calciies when burned.225 Its use in Rome began towards the end of the 2nd century BC, but locally it was probably already in use earlier and the beginning of quarrying might have started by the early 2nd century BC. It is a fairly ine biochemical limestone with a normally pleasant creamy color. The main source for travertine is the large deposit below Tivoli and some ancient quarries have been recognized despite the extensive modern quarrying. As mentioned by Strabo, the river allowed for easy transportation to Rome.226 It has been suggested that some of the stone might have been transported on the Via Tiburtina at the end of the 1st century BC.227 The limestone from the slopes of the Apennines could also have been used for building purposes. However, little direct evidence for extensive exploitation of limestone as a building material either in written sources or in archaeological inds exists. Lime was also needed in large amounts for mortar and limestone could have been burned for lime. Again, little evidence for extensive use of the available resources for this activity can be found.228 The reason might be the dificulty of transportation as the limestone slopes lie above the Aniene escarpment and would have required hauling the lime down the slope for shipping.229 Apart from the stone and sand used in building, a few other uses for geological resources can be found in written sources concerning the research area. Clay suitable for brick and pottery production is found, but little evidence for their use exists.230 Vitruvius mentions a Tiburtine brick which was possibly used in pavements similar to opus spicatum’s herringbone pattern.231 Production of brick and tile for local purposes starting from the 2nd century BC, has been suggested based on brick stamps of Q. Pandusinus/Pantisinus232 which have been found around the outcrop of marine clay on the northern edge of the research area.233 Bricks stamped by L. Ar(i?)lenus Chrestus, dated to 1st and/or 2nd centuries AD, as well as his burial 223 Plin. nat. 36,175; Vitr. 2,4,1–3; 2,6,1–2; 2,6,5–6. Cf. also Pallad. 1,10,1–2. Frank 1924, 36–38; Blake 1947, 31–34; Lugli 1957, 399–400; Molinaro et al. 2001, 60–67; Jackson et al. 2007. 225 Plin. nat. 36,167; Vitr. 2,7,1–2. 226 5,3,7; 5,3,11. 227 Frank 1924, 32–33; Blake 1947, 44–48; Lugli 1957, 319–321; Mari 1983a, 361–370 site 380; 2004. 228 Pallad. 1,10,3 mentions both limestone and travertine in connection to making lime. 229 Cf. DeLaine 1995, 559–560. The remains of lime kilns in the research area almost all date after the settlement site was abandoned or to modern times. 230 Clay could also be used for building, e.g., internal walls in pisé de terre technique, but this technique has not been found in the villa excavations in the research area. 231 Vitr. 7,1,4. 232 Steinby 1981, Nr. 181, 182. Found at sites Tibur III 18, 35, 42, 45, 47, 189, 218; Ficulea 405, 539. 233 Mari 1983a, 39. 224 46 gEology inscription have also been found in the area.234 It has also been suggested that the brick stamp M. TVLI found in Tusculum may have been the result of Cicero’s production of brick for his private use.235 Some production and kiln sites have been found in the research area (Plate II.2), but clay pits are not known.236 Seneca mentions a Tiburtinus calix237 which, from the general context, can be understood to be of clay. The recent discovery of a pottery production site in Tibur with products ranging from ine wares to cooking pots gives a whole new perspective to that reference. The workshop produced pottery from the 2nd century BC until the beginning of the 1st century AD. Its products were also exported as some have been discovered in Spain. The clay used displays volcanic materials, but seem to bear more similarities with those found in Segni and Palestrina.238 3.5 archaEology of building MatErials Most of the building stones and their quarries were introduced in the previous section. Only one further extraction site remains to be mentioned. Below Tibur lies another small quarry not mentioned by ancient authors. It is situated on the south bank of the Aniene and the stone extracted was a tuff of the Tufo di Bagni Albule formation. The stone was used locally at Tibur for public buildings from the late 3rd century BC onwards and its use diminished only towards the Imperial period when large ashlars were not so commonly used any longer.239 The quarries reported in the archaeological surveys have been plotted in Plate II.2 and listed in Table 3.2. The undated quarry sites indicated on the geological map have also been added to the distribution map to give a wider view of the possible extraction areas. Little can be said of the relationship between the quarries and settlement sites. In some cases, habitation possibly ended because the quarry expanded to cover the buildings with waste or partly destroyed them.240 On the hill of Fidenae, the site Fidenae 73 was inally almost completely surrounded by quarries, connected to the main part of the hill only by a narrow tongue of land. The quarrying was probably stopped due to the continued use of the villa. The stones found in the area were used in Rome, but their local use in the countryside is relatively poorly known.241 The site descriptions mention the general stone types found, 234 The burial is possibly at site Tibur III 24 = Ficulea 348. Wasters of brick/tile have been found at the same site, probably indicating a production area. The burial inscription is published, e.g., in Solin 1975, 72 Nr. 117. The brick stamps are CIL XV 2379 and Steinby 1981, Nr. 99 and Nr. 146, found at sites Tibur III 45, 217. Another possible local brick production site may have existed east of the research area, at site Tibur II 149, based on wasters and a number of brick stamps of the Caecilii family in the Empolitana valley (CIL XV 2381a, 2381b, 2382a, 2382b). These stamps have also been found at sites Tibur II 182, 207. Gliozzo and Filippi 2005 also suggest a production site near the conluence of the Tiber and the Aniene. 235 CIL XV 2277. Manacorda 1985, 102; Coarelli 1993, 117–118. Bricks with the same stamp have also been found at Ariccia, Terracina and at the site Tibur III 249 = Ficulea 573 (and possibly also somewhere in Guidonia; Moscetti 2002, 71 Nr. 65 I). At the last site, another version of possibly the same stamp was found recently – Moscetti 1999a, 130 Nr. 28c.1: M. Tulli M.l. (cf. also Moscetti 2002, 71 Nr. 65 I–VII). The distribution pattern and knowledge of Cicero’s properties do not match when production for private use is considered. Private production in the Tusculum area has also been suggested for the Asinii Polliones based on brick stamp CIL XV 2231 (Dessau in CIL XV pp. 45, 457). 236 See also Petracca and Vigna 1985 for inds of kilns and wasters in the area of the modern city of Rome (Plate II.2). See Table 3.2 for sites in the research area with ceramics were possibly produced. 237 epist. 119,3. 238 Leotta 1993; 1995; 1997; 1998; 1999, 41–46; Olcese 1997. 239 Mari 1991, 116–123 site 63; 1992. 240 Sites Fidenae 74; Tibur III 35; Collatia 4; destruction of agricultural installations is mentioned in connection to Collatia 57 (Calci and Mari 2003, 184), but the villa remained in use and probably had a connection to river transport. 241 Jones 1963, 138–143 is one of the few attempts to study the use of building materials in the Ager Capenas. 47 chaptEr 3 Table 3.2 Quarries and pottery production sites in the Roman region. Map numbers refer to Plate II.2. * = outside research area. Map Site Rock type Comments Map 1 2 3 4 5 6 7 8 9 10–11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Fidenae 11 Fidenae 14 Fidenae 17 Fidenae 19 Fidenae 20 Fidenae 22 Fidenae 26 Fidenae 39 Fidenae 44 Fidenae 51–52 Fidenae 74 Fidenae 81 Fidenae 156 Fidenae 160 Fidenae 234 Ficulea F.1 Ficulea 26* Ficulea 445 Ficulea 485 Ficulea 487 Ficulea 488 Ficulea 492 Ficulea 496 Ficulea 504 Ficulea 540 Tibur III 152 Tibur III 258 Tibur III 278 Tibur III 380 Tibur III 381 Tibur III 386 tuff tuff tuff tuff tuff tuff tuff tuff tuff tuff tuff tuff breccia tuff tuff tuff tuff/pozzolana tuff tuff tuff tuff tuff tuff tuff? tuff travertine tuff tuff travertine travertine travertine for aeration? stone blocks stone blocks stone blocks 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 Map 65 66 67 68 69 70 71 72 73 74 75 76 Site Fidenae 78 Fidenae 88 Fidenae 191 Ficulea 201a Ficulea 371b Tibur III 24=Ficulea 348 Tibur III 99 Tibur III 156 Tibur I 151 Tibur II 149* Collatia 67 Collatia 157 near a villa near a villa near a villa near a villa near a villa Date pottery 1st through 3rd centuries AD? 4th century AD? Imperial? Site Rock type Tibur III 411 Tibur IV 18c* Tibur IV 41 Tibur IV 63 Tibur IV 83 Tibur IV 84 Tibur IV 150 Tibur IV 223* Tibur II 33b* Tibur II 150* Collatia 4 Collatia 22 Collatia 32 Collatia 34 Collatia 43 Collatia 55 Collatia 70 Collatia 71 Collatia 72 Collatia 80e Collatia 82c Collatia 84 Collatia 89b Collatia 297b Collatia 313b Collatia 449c Collatia 820b Collatia 832 Bovillae 101 Bovillae 189 Bovillae 299 Tusculum 57 travertine limestone limestone?/tuff? tuff tuff pozzolana pozzolana lava tuff/pozzolana? limestone?/tuff? tuff tuff tuff tuff pozzolana tuff tuff tuff tuff tuff tuff tuff tuff tuff tuff tuff peperino tuff pozzolana/lava peperino peperino pozzolana Comments near a villa stone blocks for Villa Adriana for Villa Adriana? for Villa Adriana? water control? stone blocks near a villa underneath a villa Finds wasters?, pottery? wasters, pottery? furnace, pottery, bricks? furnaces, ixtures for clay processing, bricks furnace, wasters; bricks iglina, wasters, brick/tile?; Arlennus/Arilenus brickstamps furnace, wasters, pottery, column bricks iglina?, wasters? iglina, pottery, lamps, matrices wasters, brick/tile?; gens Caecilia? wasters, furnace wasters but the information is not always very detailed, e.g., the most common building stone, tuff, is not necessarily given a speciic type. Thus, distribution of the use of, e.g., the Aniene tuff cannot be provided. Peperino is easy to recognize and so its distribution map is also more reliable. The distributions for the use of these two stone types (Fig. 3.1a–b) show that tuff (as indicating all types of tuff) is used in the northern and central parts and peperino in the central and southern parts of the research area. The occurrences of peperino in the area north of the Aniene can also be explained by the Palatino type tuff found in that area as it can resemble peperino a great deal.242 This distribution pattern is somewhat surprising considering the good quality of peperino and its common use in various architectural elements, not only in wall facings or in opus caementicium. In some cases, use of local stone, tuff or different types of 242 See above note 210. 48 gEology Fig. 3.1 Source areas (white) of a) tuff, b) peperino, c) travertine and d) lava with a 5 km wide buffer zone (dark grey) and distribution of sites (black) where the stone type has been used. limestone, is speciically mentioned, but they remain few in number.243 Travertine was also used more in the northern and central parts, closer to its source, but it has also reached areas further south (Fig. 3.1c). Limestone is used, again, locally in the northeastern area, particularly in the Tiburtine region. Some of the limestone could also be travertine mistaken for limestone. Lava (often called selce) is found in the central and southern area and its use is most abundant in these areas (Fig. 3.1d), but it is also frequently found north of the Aniene. Lava is mentioned in connection to the opus caementicium of cisterns, both as caementa and in the facings. Based on this information on the distribution of recognized building materials, it is fairly clear that local stone types were mainly used for construction in the research area. The stones travelled less than 10 km from their source areas: peperino from the southern part of the area did not normally cross the Aniene, 15–17 km away from the known quarries. Tuff spread on 243 Sites Fidenae 11, 24, 31, 48 (all these located on the hill of Villa Spada, the source of Fidenae tuff) and 83; Tibur III 76, 83, 176 (conglomerate), 257 (from quarry Tibur III 258), 278 (quarry next to building); Tibur IV 37, 40, 44, 65, 225, Collatia 280, 386; Bovillae 335. 49 chaptEr 3 both sides of the river, but not much beyond a distance of 10 km. This also applied to stone types commonly used in architectural elements, as peperino was used in the area south of the Aniene and travertine in the northern zone. This its well with the Vitruvian advice to select a site where materials are easily available and to design the building in such a way that these materials it the plan. As a result, the cost of construction can be reduced. The only stone type that needed to be imported was marble and its distribution shows a different pattern as it is spread almost equally in all parts of the research area.244 3.6 intEgrating thE EvidEncE The Roman Campagna afforded many opportunities both for the landowner’s own or even commercial exploitation of geological resources. Were they considered when selecting sites for villas? In order to answer this question, three approaches were adopted. First, the distribution of the sites was compared to the geological formations in general to see if some formations were avoided or preferred. Secondly, the distribution of the sites was compared to the boundaries of geological formations based on the discoveries of previous analyses of settlement sites and geology in southern Italy. And inally, the distribution of the sites was compared to the distribution of stones which could be used as building material. In each case, the basic assumption is that if geological factors did not matter in selecting a location for a building, then the relative number of the sites on each formation type should be more or less the same as the extent of that formation. This means, e.g., that if the formation covers 20% of the area and 20% of the sites are located on it, the selection was random. On the other hand, if 40% of the sites were on that same formation type, then it was favored for settlement sites. Avoiding the alluvium? The irst task is to compare the sites to the geological formations in general. The four largest formations cover over half of the research area: grey pozzolana, Villa Senni tuff, Aniene tuff, and alluvium (Table 3.1). The two irst create a wide belt across the research area from southwest to northeast. The ifth largest formation is the red pozzolana, but all others, apart from the travertine, are relatively small in extent and many also exist only in small patches in geographically distinct zones. The settlement sites are quite evenly distributed on the geological formations as the relative numbers of sites found on each formation type correspond well with the proportion of the formations in the research area. No clear patterns of preference or avoidance of some formation(s) over other(s) can be detected. However, two minor anomalies should be noted. First, the alluvium seems to have been avoided as fewer sites are located on the alluvium than could be expected based on its extent. The lack of sites could be partially explained by the post-antique accumulation of soil covering traces of human activity. Recent excavations covering vast areas have shown that most archaeological sites in the valley bottoms are related to roads, channelling of water and sewage, and possibly also to some productive activities, but no settlement remains have been found.245 This is probably true for most of the research area, where the river valleys are It has to be remembered that marble is also an easily visible ind in surface surveys and, being a precious material, tends to be recorded more carefully than tuff, peperino or even travertine and limestone. The most detailed data on marble types found in the Ficulea/Tibur III region can be found published in the site descriptions by Eugenio Moscetti in the Notiziari archeologici of the Annali dell’Associazione Nomentana di Storia e Archeologia starting from 1994. Cf. Clarke 2003 for a study of marble inds in the South Etruria survey area. 245 Di Manzano 2001; Musco et al. 2001. 244 50 gEology mostly small and narrow. The extensive valley of the Tiber, however, is probably a different matter and few archaeological sites have been found there. The discovery of the Auditorium Villa in Rome very close to the Tiber under post-Roman alluvium could point towards the possibility of buried settlements in the river valley.246 Written sources also refer to villas and other settlements directly in the river valleys and near them. Pliny the Younger mentions how a lood of the Aniene in AD 105 caused great destruction to the buildings on its banks.247 Risk of loods might have discouraged habitation near the river channels, but obviously it was not enough to prevent it completely. It thus seems likely that more sites could be found in the major river valleys, but elsewhere it is equally likely that few, if any, settlement sites were located directly on alluvial deposits. The second anomaly to be noted is the preference for the Sabatine paleosol in the northern part of the research area. It features twice the number of sites that its extent would lead one to expect. The deposit covers the crests of the large ridges east and south of Fidenae and the reason for its preference is not very clear. The phenomenon could be related to the types of soils it produces as these are very good for agricultural purposes. The sites in the area are also mostly of Classes 2 and 3 (62 sites), i.e., possibly more agricultural in character than the Class 1 sites. In addition, a large number of Class 4 sites (44), i.e., small scatters of brick/tile and pottery, have been found. The general character of the settlement in the region seems to it well with the desirable agricultural situation. Thus, the reason for the preference of this formation is perhaps not directly related to its geological characteristics, but rather to the potential agricultural use of the soils derived from them. The last matter to be discussed in this section are the sites established in the Late Republican period. They appear in the greatest numbers in the northeastern and southeastern parts of the research area. The main patterns to be seen are the already familiar avoidance of the alluvium and popularity of the Sabatine paleosol. Other minor patterns also emerge: slightly more Late Republican sites than could be expected are located on peperino and Bagni di Tivoli tuff and slightly less sites than expected on grey pozzolana. The reasons for these can be derived from the characteristics of the formations, and they will be discussed further below. Favoring boundary zones? From the previous discussion, it is clear that mere comparison of distribution of sites and geological formations yields little information on what regions were preferred or avoided. The second approach is based on the observations of villas being located close to formation boundaries for both obtaining ground water as well as for various resources. The geology of the research area is deined by small units, particularly in the northern and southern parts. These zones of varying geology are separated by a belt of large, uniform deposits (grey pozzolana/ Villa Senni tuff/travertine) running from southwest to northeast. Thus, at irst sight, it would seem likely that most sites would be naturally close to formation borders wherever they are located. A closer examination of the distribution reveals that almost two thirds of the sites are located entirely on one formation and the rest on two or more formations (Table 3.3). Class 3 and 4 sites are slightly more frequent on single formations than Class 1 and 2 sites, but the difference is not very large. The higher proportion could be partially explained by the small size of the Class 3 and 4 sites.248 The distribution of the Class 4 sites is also most dense in the 246 Arnoldus-Huyzendveld 1995; Pisani Sartorio 1995; Carandini et al. 1997. See also Pavese 2004 for a discussion on legal disputes over alluvial lands with few references to actual buildings in these areas. The same applies to texts concerning land surveying, e.g., Campbell 2000, passim. 247 Plin. nat. 3,54 (Tiber); Plin. epist. 8,17 (Aniene). 248 I thank Dr. Janne Ikäheimo for this observation, although verifying it is more dificult due to the lack of reliable information on site sizes. 51 chaptEr 3 Table 3.3 Distribution of sites by class and date over one or more geological formations, boundary zones and soft/hard formations. Site type Class 1 Class 2 Class 3 Class 4 Early Late Republican Total One formation 333 459 449 700 278 184 231 313 352 585 227 120 >2 Sites % formations 69 68 78 84 82 65 102 146 97 115 51 64 Boundary Sites % Exp. 56% 232 314 279 410 204 134 70 68 62 59 73 73 Soft Sites % Hard Exp. 69% 178 276 282 473 180 71 54 60 63 68 65 59 Exp. 31% 95 96 98 150 65 49 & Sites % Soft Hard Sites % 29 21 22 21 23 41 60 87 69 77 33 64 18 19 15 11 12 53 grey pozzolana/Villa Senni tuff belt (323 of the 698 sites, 46%). Conversely, the Class 1 and 2 sites are found slightly more frequently in locations where more than one type of geological formation are present, i.e., they are often situated close to formation boundaries. However, the number of such sites seemed fairly small compared to the number of formations boundaries. In order to examine this aspect further, the boundaries of the formations were extracted from the geological map.249 Then a buffer zone of 100 meters was created on both sides of the border so that the areas close to formation boundaries would be covered and the distribution of the sites could be more easily compared to the boundary zones. A 100-meter buffer zone covers 56% of the whole research area.250 (Fig. 3.2.) When the distribution of all sites is compared to the boundary zones, the numbers correspond with the expected: 56% of the sites (1,078) have more than 50% of their area inside the buffer zone. Approximately 64% (1,235) of all sites have some part of their area within the buffer zone. Some differences appear when classes of sites are compared: Class 1 and 2 sites are found slightly more frequently on or very close to the boundary zones than the other two. The tendency of large sites to be located close to boundary zones is fairly distinct, particularly in the northeastern part of the research area. The reasons for choosing the boundary zones might be the same as the ones cited for southern Italy: trying to ind better access to ground water as well as for exploiting other resources afforded by a different geology. This would also suit well with the predominantly limestone-based geology of the northeastern zone as the southern Italian examples feature similar conditions. One of the most important qualities of a formation for gaining better access to ground water is its permeability, i.e., how well water can travel through rock or sediment. The volcanic deposits are, for the Fig. 3.2 Boundaries of geological formations greater part, mediocre to good in this regard (white) with a 100 m wide buffer zone (dark because of their porosity (tuff, pozzolana, grey) and Class 1 sites (black). scoria, lapilli) or discontinuities (lava, 249 This was done by using the Filter module of Idrisi with Sobel Edge Detector function and 3x3 ilter size. The result was compared to the vector ile containing the digitized edges and these were found to match perfectly. The edges of the research area became visible on the boundary map and these were checked for sites for each calculated statistic. Only very few sites are situated on the edges of the research area. 250 The relatively short and arbitrary distance of 100 m was used since larger buffer zones would have covered most of the area – a 200 m wide buffer zone would cover 75% of the research area. The studies in southern Italy used a buffer zone of 500 m, but their research areas were generally much larger than this one. 52 gEology peperino, lithoid tuff). Most discontinuities are between layers of different materials on top of each other. The limestone geology varies more; some are practically impermeable (e.g., white, hard limestone) and some very permeable (e.g., travertine). Discontinuities tend to be vertical. The clay and sand/gravel/clay are practically impermeable. The Holocene alluvium and other water-related formations are mediocre in permeability.251 (Table 3.1.) No clear correlation between formation boundaries and ground water resources can be seen in the Roman area, but nevertheless, the boundary zone was checked for permeability and possible preferences for more permeable formations. Almost half of the area (49%) is mediocre in permeability and, in addition, 18% is very permeable. The central and southern parts are mostly of mediocre quality. Both the very permeable and practically impermeable formations can be found in the northern part of the area. The boundary zone is consistent with the general area as its relative proportions of permeability match those of the whole area. In other words, no tendencies towards choosing permeable geology for sites, at least in the boundary zones, can be seen. Looking for building stone? The geological formations can be divided into compacted and loose ones. The division, of course, is never absolutely accurate, as the volcanic formations can feature considerable internal variation. Despite this, a rough reclassiication can be made. Of the volcanic formations, many types of tuff (Aniene, yellow, Palatino type, Fidenae), sperone, peperino and lava are hard. Limestone and travertine are also hard rocks. The largest loose volcanic formations are grey and red pozzolana, but also the incoherent Villa Senni tuff covers large areas. Other loose formations consist of the clay and sand as well as Holocene alluvium. Approximately 70% of the research area is covered by loose formations and the rest is hard stone. The harder rocks are found in the Aniene valley (Aniene tuff), in the northeast (limestone, travertine) as well as in the southern parts (lava and peperino). The loose formations form a wide belt between these zones. Their distribution is reminiscent of the boundary zones: the hard formations are often featured in smaller patches or long stretches which have many boundaries and the loose formations (pozzolana) often create large and uniform areas. The distribution of sites was then compared to soft and hard formations and in general, the sites are located according to expectations based on extent of the formations. The only anomaly is the slightly lower number of Class 1 sites on loose formations. (Table 3.3.) When the dated sites are considered, a slight tendency to avoid the harder rocks can be noted for the sites settled before the Late Republican period. Vice versa, the sites established around 100 BC tend to be located on hard rocks. The latter sites are also fairly commonly situated in the northeastern and southeastern areas which feature more hard rocks. The Class 1 sites are also found in the boundary zones where one of the most common combinations of formations is the Aniene tuff with either red or grey pozzolana. The banks of the Aniene consist almost exclusively of a possibly ideal combination of building materials, i.e., the Aniene tuff and red pozzolana. These two are also in general the most common types of formations when sites in the boundary zone and on only one formation type are concerned. In addition, of the most common combinations of formations, 19 out of 24 feature at least one formation type which could be used as building material. The Late Republican sites form the core of the distribution of the otium or leisure villas on the slopes around Tibur and Tusculum. The buildings are generally large and require many building materials. Considering the preference for using local materials, it seems likely that, 251 This classiication is based on Ventriglia 1990a, 183–185 and Ventriglia 1990b, 183–188, as well as on the maps in each volume. Cf. also Vitr. 8,1 for a description on the qualities of water and how to ind it in different geological formations. 53 chaptEr 3 when possible, sites where hard rocks and/or pozzolana suitable for mortar were available would have been selected. The distances to the known major production centers are, however, not very long anywhere in the area. Another advantage the hard rocks could provide is a solid foundation for a large building. Many of the large villas have been built on sites where the slope is fairly steep and the foundations of the buildings, i.e. their platforms, are extensive. It would have been beneicial for the building project to be based on solid geology. In this regard, however, even the mostly incoherent formations, such as the Villa Senni tuff,252 can provide a good base. It should be noted, however, that the areas of hard stones are also often elevated from their surroundings, forming high spurs which were preferred for building villas. Acquiring building stone or having a solid foundation were probably added bonuses to the other qualities of the site, but not necessarily actively sought. Selling building material(s) to other landowners in the vicinity could have been a source of extra income. Very little is known of who controlled the quarrying of basic building stones even in the main production centers: written sources describing building materials mention no owners and no epigraphic evidence exists. The fact that many of the stones were used extensively for public building could indicate public ownership. The closest comparison comes possibly from brick and tile production which was organized mostly as a private enterprise despite the later virtual monopoly of the imperial family. The quarries could have been similar, partly in private hands and partly under the control of the imperial family.253 The distribution of known small quarries shows that minor production could have been more common and more signiicant than perhaps thought. 3.7 conclusions The Roman agronomists offer little direct advice concerning the geology of the ideal villa location and the results of the analysis relect this disinterest. Some local variation could be seen in the analysis of site distributions with regard to geological formations. The alluvial formations of the major river valleys were avoided, the fruitful Sabatine paleosol was preferred, probably due its agricultural potential, and in the limestone areas, the boundary zones were preferred, possibly for the better availability of ground water. Geology in itself was not signiicant, but rather the qualities derived from it. Other passages and texts hint at the usefulness of certain geological formations, e.g., of those that could be used as building materials or that could offer solid foundations for buildings. The distributions of the use of various stone types indicate that building materials used most commonly in the Roman region were strictly local, not spreading much further than 10 km from the source. However, the useful materials were fairly evenly distributed in the region and there was probably little need to consider their availability speciically when selecting the location for a building. Quarrying and selling suitable building materials could also be an additional resource for the estate and could supplement its income. It could have been an important addition to the versatility recommended by the agronomists for the estates near large towns and cities. 252 Arnoldus-Huyzendveld et al. 1997, 14. For reviews and recent studies on various aspects of brick production, see Bruun 2005. The ownership of quarries is very briely discussed in Quilici and Quilici Gigli 1986, 404, nota 231. Torelli 1980 concerns the involvement of an Italian private family, the Cossutii, in the marble trade. See also Lega 2003 for possible distribution/sales of marble products from the Villa Barco Borghese = Tusculum site 426. Some references in Digest also suggest private ownership (10,3,19; 18,1,77; 23,3,32; 24,3,7, 13–14; 24,3,8). 253 54 4 soils and roMan villas 4.1 background The formation of soils is strongly related to the basic geology and geomorphology of the area. The weathering of rocks produces the main body of soil and the terrain of the area determines its thickness and texture, i.e., grain sizes. Soil is one of the key elements when agricultural production is considered. Its qualities, particularly fertility, determine what crops can be cultivated. Other factors, such as water resources, climate in general and even human effort can, of course, also make a poor soil productive. Consequently, soil is one of the fundamental features when the locations of settlements of an agricultural society are studied. Varro mentions knowledge of soil as the most important thing in agricultural know-how.254 The good quality of soil was also considered one of the deciding factors when selecting a location for a farm.255 In addition, diversity of soils on the estate was also recommended for making versatile production and self-suficiency possible.256 Despite this, studying the relationship of Roman settlements to the soil and agricultural resources surrounding them has not been seen as of great import. Soil is not mentioned in the survey reports for the research area and so it has been used even less than the geological information. The reason for this disinterest might be the lack of more detailed soil surveys until recently.257 Some studies on the relationship of soil and settlement have recently been done in the central Italian region. The irst major study concerns three areas in central and southern Italy: the Pontine region in southern Lazio, the Salento Peninsula in Puglia and Sibaritide in Calabria.258 The approach adopted was geoarchaeological and the aim was to map the areas and to evaluate their suitability for agricultural production in the irst millennium BC, i.e., from the Bronze Age to the Roman period. The method used was the land evaluation system developed by the Food and Agricultural Organization of the United Nations. Another important aim was to develop this methodology for archaeological purposes. The results of the land evaluation were compared to available archaeological information and palynological data was used to check the archaeological as well as land evaluation results. What is interesting in this study is that modern criteria were not applied directly to ancient conditions. Instead, the differences of the ancient perception of the environment as well as the different requirements of the ancient crops were taken into consideration. In the end, some conclusions were drawn concerning the relationship of settlement and soil. The comparison of agricultural suitability and archaeologically attested use of the area seem to point towards intensive use of even quite 254 255 256 257 258 rust. 1,5,3–4. E.g., Cato agr. 1,2; Colum. 1,3,1; Varro rust. 1,7,5. E.g., Cato agr. 1,7; Colum. 1,2,3–5. Cf. Perkins 1999. Joolen 2003. chaptEr 4 agriculturally unsuitable areas, such as the Murge area on the Salento Peninsula, during the Archaic and Roman periods. The second study is connected to the Tiber Valley Project of the British School at Rome and the aim was to model agricultural production in the river valley.259 The analysis was hampered by lack of a detailed soil map as only a 1:1,000,000 scale map was available. In order to evaluate the agricultural potential of the area and the relationship of the settlement with the resources, a similar approach to the one presented above was adopted. The suitability of soil for agricultural production was modelled for wheat, olives and vineyards based on data on slope, aspect, distance to water and the general productive potential of the geology of the area. The results were also compared to the distribution of the known sites and the densest site distribution in the area coincided with the most suitable agricultural lands in South Etruria, north of Rome and thus also in the region closest to the city – and perhaps also the most intensively surveyed area in the Tiber Valley. The approach adopted here is similar to both of these studies in methodology and questions. At the beginning of the chapter, the soils in the research area are described and then their agricultural suitability for some of the main crops of the Roman period is evaluated based on the land evaluation system. Then, the literary, archaeological and palynological evidence for agricultural activities for the Roman area is presented. In the last part of the chapter, the suitability of soil for varying agricultural purposes and the evidence for known land use are compared. Although soils have a greater importance to growing crops, animal husbandry is also taken into consideration as another integral part of agricultural production. 4.2 soils of thE roMan caMpagna A detailed soil survey was conducted in the territory of the modern Comune di Roma in the 1990’s.260 The scale of the soil map is 1:50,000, which is suficiently detailed for the purposes of this study.261 The surveyed region can be divided roughly into three main landscape units based on general relief (morphology and slope), lithology and soil characteristics. The most common landscape type consists of fairly level or very gently sloping ridges in the plateau area incised by the river valleys. The second unit consists of river valleys and their ills; small depressions can also be found in the plateau area. Slope areas are the third type and they can mostly be found in the river valleys. Each environment produces its characteristic soils.262 The main soil groups featured in each landscape unit are described in the following; more detailed properties of each soil type can be found in Table 4.1. Everyone of the 17 soil types featured in the research area is described briely and their other standard nomenclature and descriptions are given. 259 Goodchild 2003; 2006; 2007. Arnoldus-Huyzendveld 2003 (also available in Funiciello et al. 2008). Due to the kindness of one of the directors of the project, Dr. Antonia Arnoldus-Huyzendveld, I was already able to use this material before its publication in 2003. For this, I owe her my heartfelt thanks. The map of 2003 replaced an earlier soil map published in 1959 (cf. Frutaz 1972, tav. XC). 261 The main problem with this data is its limited extent: almost half of the research area remains outside the map. No attempt has been made to extrapolate soil data outside the surveyed area despite the fact this would have been partly possible. Most of the areas outside the surveyed region feature different geology or have much steeper slopes, which would make extrapolation somewhat uncertain. Thus, it was decided to use the soil survey map as such and make the analyses only in the area it covers. 262 Various ways of classifying soils have been created and the system used in this connection is the Food and Agriculture Organization’s (FAO) World Reference Base for Soil Resources (1998). In addition, FitzPatrick 1983, FAO 1993 and Arnoldus-Huyzendveld 2003 have been used. 260 56 soils Table 4.1 Soil types found in the research area. Map numbers refer to Plate III.1. Map Type Area ha Area % Description Depth Texture Drainage Slope Yield Land use 16 deep volcanic soils on plateau tops medium-ine well or somewhat nearly level or excessively very gently 4,305 18 not very deep volcanic soils 40–80 medium-ine on plateau tops cm well to somwehat nearly level or excessively very gently Luvisols 2,001 8 deep reddish volcanic soils on plateau tops, paleosols deep medium-ine – ine well or moderately nearly level or well very gently high grain, grapes, orchard, pasture 4 Luvisols 1,353 6 deep volcanic soils on shallow depressions on plateau tops deep medium-ine well very gently or nearly level high grain 5 Luvisols 450 2 non-calcareous soils of valley bottoms deep or ine – thin medium moderately well to level/ nearly level somewhat poorly high grain deep volcanic soils on plateau tops, dark organic horizon deep volcanic soils on shallow depressions on plateau tops usually medium-ine deep well to moderately very gently or well nearly level high grain, grapes, orchard, pasture very gently or nearly level high grain high grain high grain 1 Luvisols 2 Luvisols/ Phaeozems 3 3,922 6 Phaeozems 1,388 6 7 Phaeozems 277 1 8 Phaeozems 575 2 9 Phaeozems/ Luvisols 1,516 6 deep deep medium-ine well non-calcareous soils of valley bottoms deep medium or medium-ine well or moderately level well soils of minor valleys in the volcanic landscape deep medium – medium-ine medium/ grain, grapes, high orchard, pasture medium/ grapes, grain, high pasture well very gently or nearly level not very deep volcanic soils 50 cm medium – mediumon plateau tops coarse medium not very deep soils on mediumvolcanic slopes, often rocky thin –coarse somewhat excessively very gently medium/ grapes, grain, high pasture somewhat excessively strongly or moderately steep brush, medium grapes, trees somewhat excessively strongly or moderately steep brush, medium grapes, trees 10 Andosols 2,227 9 11 Andosols 1,061 4 12 Cambisols 831 3 not very deep soils on volcanic slopes, often rocky 13 Cambisols 976 4 calcareous soils of the valley bottoms thin medium – medium-ine deep or medium-ine thin moderately well to level somewhat poorly high grain low brush, trees 14 Cambisols/ Andosols 95 <1 not very deep soils on volcanic slopes, rocky deep medium somewhat excessively or excessively 15 Cambisols/ Calcisols 65 <1 soils on pre-volcanic slopes deep medium/ine well or moderately strongly or well moderately steep medium/ grain, pasture high 16 Leptosols 2,404 10 shallow, rocky soils on travertine plateaus thin medium-ine somewhat excessively very gently medium/ pasture, brush low 17 Arenosols/ Cambisols 1,179 5 soils on pre-volcanic slopes, often rocky thin mediumcoarse – coarse excessively moderately steep or steep medium/ trees, brush, low pasture steep to extremely steep The irst landscape unit, the ridges in the plateau area, is mostly covered by Luvisols (Plate III.1, nos. 1–5), which are the most common soil type found in several varieties covering 25% of the research area. In addition to ridges in the plateau area, they can be found in small depressions of the plateaus (no. 4) as well as in some river valleys (no. 5). Luvisols are typical of lat or gently sloping lands in Mediterranean climates, where clear dry and wet seasons occur. They are generally fertile due to mixed mineralogy and high nutrient content. They are porous, well-aerated and well-drained, i.e., they do not become water-logged, but are capable of good moisture storage. Luvisols are often connected to Phaeozems (5% of the area), which are also typical of lat or gently sloping land. They are featured on the ridges in the plateau area (Plate III.1, no. 6) as well as in their shallow depressions (no. 7) and river valleys (no. 8). Their distribution in the research area is concentrated in its central and southern parts. Phaeozems are formed in steppe environments with clear dry seasons, rainfall concentrating in spring and early summer and with short, hot summers. They are generally rich in nutrients with a humus-rich surface horizon, porous and well-aerated. These two soil types, Luvisols and Phaeozems, are associated so closely that they cannot be separated into two types. In the research area, they cover 4% of the whole (part of no. 2 on the plateaus and 9 in river valleys). The next largest groups are Andosols and Cambisols, both with 3.5% coverage. They are featured on the ridges in the plateau area in the central and southern parts (Plate III.1, 57 chaptEr 4 no. 10 Andosol), but are more common on slopes (no. 11 Andosol, no. 12 Cambisol, no. 14 Andosol–Cambisol association) and valleys (no. 13 Cambisol). Andosols are typical of recent volcanic deposits containing plenty of volcanic glass. They are generally rich in nutrients and have good moisture storage capacity. Cambisols are also recent soils forming most commonly on lat or gently sloping land in moist climates. They are generally fertile with good water-holding capacity and yet drain easily. Cambisols also feature in association with Calcisols which have a high calcium carbonate (CaCO3) content (no. 15). These are formed in Mediterranean climate, and are potentially fertile soils with good water-holding capacity and drain easily. This soil type can be found on some of the slopes in the northwestern part of the area. In general, no extrapolations of the data were made, but one further soil type found in the areas outside the boundaries of the Comune di Roma can be added: Leptosols or Rendzinas. These are formed over hard, calcareous rocks and are often shallow and rocky. They can be fertile, but also quite dry and susceptible to erosion due to often existing on steep slopes. These can be found in the northeastern part of the research area: the travertine formation below (Plate III.1, no. 16) and the limestone slopes around Tibur (no. 17).263 Before turning to land evaluation analysis, it is still necessary to discuss the relationship between modern and ancient soils. The soils described are found in the area today, but were the Roman period soils similar to them? Soil is an incoherent element of the landscape and as such, it undergoes constant evolution and change. Both natural and human processes cause changes in its composition, thickness and other qualities. The natural processes, effected by water, wind and animals, tend to be slow, taking hundreds or even thousands of years, though sometimes they can be fast and, at least locally, catastrophic, e.g., landslides. Displacement of soil due to the steepness of slopes, saturation by water, lack of vegetation, movement of animals, etc., is probably the most common change occurring in soils.264 Most of the research area features gentle relief which points towards slight changes due to steepness of slopes. The relief has changed little and the underlying geology remains the same, so the basic qualities of the soil types as observed today would probably be the same as those in the Roman period. Changes in climate can also cause variation in rates of soil genesis and the Roman period was one of a heightened rate of pedogenetic activity (cf. Chapter 3.2). This could mean that the situation in Roman times was more favorable than today with thicker soils.265 The comparison of buried soils of Pompeii and modern soil proiles also showed that their characteristics were much the same.266 Human activity can change topography and soil cover very rapidly and extensively. Archaeological ieldwork conducted in the eastern part of the research area revealed evidence for concerns for land degradation caused by intensive land use possibly already in the Late Republican period.267 Modern land use has somewhat changed the area, although no signs of major changes to the soil cover can be detected. The soil survey used indicates very clearly all the built-up regions in the 1990’s where observation of soil cover has been impossible. It seems safe to assume that the qualities of the observed soil types would have been very similar in ancient times.268 263 Cf. also the 1:250,000 Soil Regions of Italy map and notes (Costantini et al. 2004). Its unit 16.4 corresponds with the Apennine reliefs on limestone and intra-mountain plains. 264 Shiel 1999. 265 Arnoldus-Huyzendveld 2004, 462; Volpe and Arnoldus-Huyzendveld 2005, 61. 266 Foss 1988. 267 Pracchia 2001, 300–305. 268 E.g., Joolen (2003) does not even mention the question of change of soils over time. For evaluation of the soils and their sustainable use in modern times, see Arnoldus-Huyzendveld 2008; few of them require protective measures. 58 soils 4.3 land Evaluation of soils for anciEnt agricultural purposEs The soils of the Roman Campagna are generally fertile and of high or medium productivity. They are good for most agricultural purposes from orchards, vineyards and grain to supplying wood and brush. This pertains particularly to the ridges in the plateau area, where only a few limitations for sustained agricultural use of these soils exist. Despite this universal suitability, I considered it interesting to test the soils for some of the main crops of the Roman period with the land use requirements outlined in the study of central and southern Italian areas mentioned above.269 The crops are emmer wheat (Triticum dicoccum), other wheat varieties (Triticum sp.), barley (Hordeum vulgare) and millet (Panicum miliaceum), which represent basic subsistence farming. In addition to these, the two most important commercial (as well as partially subsistence farming) crops, grapes (Vitis vitifera) and olives (Olea europea), are discussed. The land quality requirements for each crop are listed in Table 4.2. Some of these qualities were not speciied in the soil survey data, most importantly soil structure. Considering the generally ine to medium-ine textures and the good drainage capacities of the soils in the area, a fairly irm structure with quite a lot of porosity would seem likely. However, in spite of this general conclusion, structure had to be left out of the analysis. The evaluation was executed by creating Boolean theme maps of each quality for each suitability category (suitable, marginally and not at all suitable) and then adding all properties for each category (see Table 4.2 for how the qualities were modelled). The end result was a series of three maps with the best suited areas marked with the highest results, e.g., for emmer wheat the best suited areas are those with four of the ive qualities present. The results of the evaluation show that the area is well-suited for subsistence farming. Emmer wheat is perhaps the best it as 20% of the area is well or averagely suited (four or three qualities of the ive used in the evaluation) for growing it. These are distributed fairly equally across the research area with larger areas in the northwest and southwest and less in the central part (Fig. 4.1a). The result for other wheat varieties is similar, with the main emphasis in the slightly higher regions (16% of the area; Fig. 4.1c). Of the soil qualities best suited for growing barley, loose structure and excessive dryness do not match many parts of the area. The criteria for marginally suitable soils are found most frequently in the central part (three out of the three qualities used; Fig. 4.1d). Thus, it would seem fair to say that the central area and parts of the southwest are marginally suitable for growing barley. For millet, only three of the ive qualities listed are found in the area. Only small amounts of sandy or loose soils are present in the Roman Campagna. The three qualities found (well drained, fertile and non-calcareous) match most of the soils, so that it is possible to say that the area is mostly marginally suited for growing millet. In respect to subsistence farming, it can be said that the research area is quite well-suited for growing grain. A comparison of the results obtained from the land evaluation with the suggested uses of soils mentioned in the soil survey data conirms this view: apart from some soil types covering slopes (nos. 11, 12, 14) and probably the calcareous regions (nos. 16, 17), the rest of the area is recommended for modern grain cultivation (40%; Fig. 4.1d). The two main commercial crops, grapes and olives, are discussed next. The qualities for soils suitable for grapes match large parts of the area, only the sandy texture is mostly missing (Table 4.2). The areas best suited for vineyards (four or ive of the six qualities listed) are distributed fairly evenly across the area, only the northwestern region can be described 269 Joolen 2003, 122–127. 59 chaptEr 4 Table 4.2 Soil quality requirements for major Roman crops and how they were modelled. Crop Suitable Model Emmer wheat At least 30 cm thick Triticum dicoccum Clay or sandy clay Moderately to poorly drained (wet) Firmly structured Deep = 80 cm or deeper Clay or sandy clay Moderately to well drained Not available Marginally fertile Non-calcareous Wheat varieties Deeper soils Clayey soils (Moderately) irm(ly) structured Fertile Moderately to welldrained Open and elevated situations Plains Calcareous (marls) Relatively warm Marginally suitable Model Model Thin = 60 cm or less Marshy Loamy Somewhat poorly drained Sandy soils Excessively drained (dry) Loose Excessively drained Mediocre to fertile Fertile Fertile Very fertile Very fertile Non calcareous Calcareous Calcareous Very calcareous Very calcareous Medium salinity Not available Thin soils Thin = 60 cm or less Loamy soils Loamy Sandy soils None found Deep = 80 cm or deeper Clayey Loamy soils Not suitable Thin (less than 30 cm) None found Not available Not available Loose Not available Fertile Well to excessively drained Above/below 75 m a.s.l. Nearly level or very slightly sloping Calcareous Aspect east and/ or west All Infertile Infertile Somewhat poorly drained Wetter Moderately well drained Poorly drained Hill slopes Strongly sloping to steep Calcareous Non-calcareous Calcareous Cold Non-calcareous Aspect northwest – northeast Barley Thin and deeper Hordeum vulgare Very fertile (very rich) Very fertile “Lean”; when replenished Mediocre fertile with nutrients Somewhat excessively Well-drained Well or moderately to excessively drained well drained Infertile Infertile Excessively drained (dry) Poorly drained (wet) Somewhat poorly drained Loose Not available Loamy and clayey soils Irrigated sand Loamy soils Fertile None found Well or moderately well drained Fertile Firmly structured (compact) Clayey soils Excessively drained (dry) Unfertile Loose Not available Firmly structured Not available Firmly structured Not available Non-calcareous Non-calcareous Calcareous Calcareous Calcareous Calcareous Liable to fog Loamy soils Not available Thin = less than 80 cm Loamy Poorly drained None found Clayey soils Excessively drained (dry) Very fertile Clayey Somewhat excessively to excessively drained Very fertile Bitter or brackish Not available Icy cold or burning hot Aspect northwest – northeast or southeast – southwest Millet Panicum miliaceum Well-drained Grapes Deeper Sandy soils Moderately drained Rather fertile Deep = 80 cm or deeper None found Well or moderately well drained Mediocre fertile Elevation 51–300 m Slightly elevated plain a.s.l., nearly level or very gently sloping Relatively warm Olives Poorly drained Fertile Thin Loamy Somewhat poorly drained Fertile “Fat” or infertile Plain or hillside Clayey Somewhat excessively to excessively drained Fertile Elevation below 51 m a.s.l. Aspect east and/ or west Thin and deeper soils All Loamy soils Loamy Well-drained (moist) Well or moderately well drained Clayey soils Clayey Somewhat Excessively drained (dry) excessively to excessively drained Chalk mixed with coarse sand Fertile Fertile Fertile Moderately inclined Strongly sloping to steep Steep slopes with thin soils Calcareous Calcareous Calcareous None found Fertile Moderately to extremely steep, thin soils Calcareous Sandy soils None found Poorly drained (wet); Muddy or marshy Somewhat poorly drained Gravelly soils None found Infertile (lean) Unfertile Rocky terrain Non-calcareous Non-calcareous Bare Warm Aspect east and/ or west Not available Aspect northwest – Very warm or very cold northeast or southeast – southwest as slightly less-suitable (14%; Fig. 4.2a). For olives, the result is similar. Of the qualities for best suitability, only chalk mixed with coarse sand is not found in the area (Table 4.2). The best-suited soils (ive or six qualities of the seven used) can be found in the whole area apart from the northeast (Fig. 4.2b), which in turn its well with the marginally suitable soils. 60 soils Fig. 4.1 Areas best suitable (dark grey) for growing a) Emmer wheat, b) other wheat varieties and c) barley. d) Areas recommended for modern grain cultivation. Comparison with the modern recommendations for land use shows a good it: all of the ridges in the plateau area are deemed good for vineyards (34%). No recommendations for growing olives were given. As the modern recommendations have so far proven to be a good match for the land evaluation, it seems fair to look at the other types of land uses mentioned in the soil survey data. Orchards are recommended for some of the best agricultural soils in the whole area, deep volcanic soils rich in organic material situated on the ridges in the plateau area (15%; Fig. 4.3a). These are almost completely missing from the central part of the research area. Good pastures would be possible in almost the whole area apart from the valleys and steep slopes (37%; Fig. 4.3b). Even the worst soils would be suitable for brush and trees. These can be found in the regions with steeper slopes, usually in the river valleys of the northwestern, central and southwestern parts of the research area. The calcareous stones of the northeast would also be suitable for brush and/or trees (respectively 12% and 7%; Fig. 4.3c–d). The area of the soil survey can thus be deemed very good or good for agricultural activities. The part of the research area outside the surveyed region can mostly be considered 61 chaptEr 4 Fig. 4.2 Areas best suitable (dark grey) for growing a) grapes and b) olives. of equally good quality. Most of the outside area is volcanic in origin – in fact, all of the calcareous regions are included in the survey area by extrapolation – and the slopes are also mostly equally gentle. A different geology is found on the northern/northeastern edge, where marine clay, Terra rossa (Chromic Luvisols), as well some other minor groups of pre-volcanic formations dominate. These could probably be connected to class G of the soil survey, i.e., soils of the pre-volcanic slopes. The volcanic areas can be regarded as good for most agricultural purposes, even though the slopes get steeper, particularly towards southeast. The steeper relief would possibly make these areas less suitable for wheat, grapes and olives.270 The pre-volcanic areas tend to be of medium or low productivity and mostly suitable for pasture, brush, trees or grain growing. Possibilities, or even suitability, for use does not necessarily mean that the use was realized. In order to ind out how the Romans used the area and how their land use corresponded with the possibilities afforded by it, it is necessary to look at the literary and archaeological evidence. 4.4 writtEn sourcEs for agriculturE In ancient times, the Roman Campagna was a large tract of countryside surrounding the largest city of the period. Despite this close relationship, the economic and demographic potential and signiicance of the area have only quite recently been recognized and studied relatively little.271 Most of these studies concentrate on the role of Rome’s immediate hinterland in feeding the population both in the city and in the country. On the other hand, the area has also been regarded as an intellectual space with little or no economic or productive signiicance.272 270 Cf. though, Carocci 1988, 433–571 for Medieval records of cultivated crops, their fairly high yields in the Tiburtine region as well as great changes in types of production over time. Comparison of two maps of land use, Carta Topograica dell’Agro Romano e territori limitroi (1880; Frutaz 1972, Tav. 379–380) and Carta dell’utilizzazione del suolo (1960), shows remarkably little change apart from the diminishing amounts of gardens and vineyards in the immediate vicinity of central Rome. 271 Carandini 1985b; Kolendo 1995; Purcell 1995; Morley 1996; De Seña 2003; 2005; Goodchild 2007; Witcher 2005a. 272 Champlin 1982; Mayer 2005. 62 soils Fig. 4.3 Areas recommended for modern cultivation (dark grey) of a) orchards, b) pastures, c) brush and d) trees. The written sources support both interpretative models. The economic aspects are referred to fairly often and the descriptions of personal experiences, such as Cicero writing about his villa at Tusculum, consider more the intellectual and recreational aspects. These both will be discussed in the following, starting with general information concerning the potential productivity of the area and its products. The main written sources for agricultural production in the Roman area are the three agronomical treatises written by Cato, Varro and Columella. An important fourth source is Pliny the Elder’s Natural History. All mention the general region or speciic places and their agricultural produce.273 Most of the ancient authors comment on the suitability of the area for cultivation. Latium and especially the area around Rome were mostly considered fertile and as having good production possibilities.274 Cicero, however, compares the Roman Campagna 273 For Roman agriculture in general, see White 1970; Flach 1990; Marcone 1997. Particularly for the Roman area, see Bussi and Vandelli 1985; Kolendo 1995; Morley 1996. 274 Cic. Flacc. 71; Procop. Goth 2,3,10; Strab. 5,3,5; 5,3,7. Cf. White 1970, 72 for an interpretation that the Latium vetus was worn out and probably uninhabited. This is based on the modern appearance of the Roman Campagna, which was rather empty until the 1950’s. Archaeological research has since proved that this interpretation is 63 chaptEr 4 to the abundance of Campania and inds the former area poor in comparison.275 Variation in the quality of soils in the area is also recognized. Varro considers soils close to Tibur thin and only of medium quality.276 The area most famous for its infertility is the ager Pupiniensis located on the plateau between modern Frascati and Rome.277 Its soils were considered thin and the area in general unhealthy.278 Strabo, on the other hand, considers roughly the same area on both banks of the Aniene to be fruitful.279 He also praises Tusculum and the foothills of the Alban Hills.280 The soils may be fertile, but their value could also diminish due to neglect, as Pliny the Elder mentions as having happened in the vicinity of Rome.281 The story of Remmius Palaemon and how he turned a barely productive farm in the Nomentan region to a highly productive vineyard with plenty of hard work is also indicative of awareness of the need to maintain the good qualities of soil. Despite the somewhat contradictory notes on fertility, the general picture seems to be that of fairly good potential for agricultural production. This general fertility made it possible to plant all sorts of crops and the references to various agricultural products in the area range from barley to lowers.282 The surroundings of Rome are not noted for any particular type of product. This its well with recommendations for versatility of production on a farm close to the city. One of the most famous passages concerning suburban production is that of Cato’s recommendation for planting the suburban farm as ingeniously as possible.283 Similar advice can also be found in other agricultural treatises.284 Cato’s general ranking of types of agricultural production285 also values variety. The preferred crops were grapes, watered garden, osier-bed, olives, pasture, grain, trees, trees with vines or orchard and mast grove. Half of these represent intensive agriculture, but pasture, osier-beds and trees were also very important. Wine, vegetables and fruit are the products mentioned most often and concern all parts of the Roman Campagna, whereas olives and olive oil are mentioned only in connection with Tibur and the Sabine region. Easily perishable products such as vegetables and fruit were considered best suited for cultivation in areas near markets and these are attested in the ancient sources. Rome also had a great need for lowers and some speciic types of lowers, roses and violets, are mentioned in connection to the towns around the city.286 Grain cultivation is mentioned twice; once probably in relation to the plateau area between Rome and Tibur and once just in general for the vicinity of Rome.287 The closest preferred grain growing areas were Etruria and Campania, but some subsistence farming would have been necessary even in completely wrong; cf. Quilici 1974b. 275 leg. agr. 2,96. 276 rust. 1,9,6. 277 E.g., Liv. 26,9,12. 278 Colum. 1,4,2–3; Varro rust. 1,9,5. 279 5,3,11 could refer to the Campi Tiberiani situated between Rome and Tibur mentioned by Lib.Col. II p.254, p.255 and p. 258, but see also note 528. In addition, Cato frg. 57 (Peter) mentions a Campus Tiburtinus which could also be the same area. 280 Strab. 5,3,12. 281 nat. 14,48–52. 282 See Table 4.3 for products and references, Plate III.2 for distribution; cf. Morley 1996, 83–107. 283 agr. 8,2. 284 E.g., Cato agr. 7,1; Colum. 3,2,1 (particularly for edible grapes); Varro rust. 1,16,3 (lowers and other products). 285 Cato agr. 1,7, Plin. nat. 18, 29–31 and Varro rust. 1,7,9 refer to Cato. In Varro rust. 1,7,10, Scrofa ranks meadows and vineyards as his favorites. 286 Mart. 9,60 (Tibur, Tusculum, Praeneste, Sabina, Nomentum); Plin. nat. 21,16 (Praeneste); 21,20 (Praeneste); 21,27 (Tusculum); cf. also Varro rust. 1,16,3 (lowers on suburban estates). 287 Cato fr. 57 (Peter); Procop. Goth 2,3,8. 64 soils Table 4.3 References in written sources to cultivation and animal husbandry in the Roman region. Product Area Reference Product Area Reference Almonds Alba Plin. nat. 15,90 Olives Sabina Apples Rome? Ath. Deip. 3,82 Hor. sat. 2,4,70–1; Iuv. 11,64–76; Prop. 4,7,81–6 Plin. nat. 16,138 Hor. epist. 1,16,1–4; Mart. 4,4,10; Strab. 5,3,1 Mart. 7,28,1–4; CIL XIV 3677 Crustumerium? Isid. orig. 17,7,67 Cattle Laurentum Plin. epist. 2,17,1–3 Tibur Cherries Rome Tibur Chestnuts Rome Plin. nat. 16,138 Chickens Tibur Iuv. 11,64–76 Figs Laurentum Plin. epist. 2,17,15 Sheep Laurentum Plin. epist. 2,17,1–3 Iuv. 11,64–76; Mart. 7,80,11 Fruit in general Mulberries Rome Ath. Deip. 3,75e Tibur Tibur Colum. 5,10,11; Plin. nat. 15,70 Tusculum Liv. 27,4,11 Tusculum Plut. Caes. 41 Pastio villatica Alba Varro rust. 3,2,17 Nomentum Tibur Bees Falerii Varro rust. 13,16,10–1 Laurentum Mart. 13,42; 10,94 Colum. 10,137–9; Hor. carm. 1,7,12–4; Plin. nat. 17,120; Prop. 4,7,81–2; Sil. 4,224–5 Plin. epist. 2,17,15 Ostia Plin. nat. 15,97 Rome Plin. nat. 15,97 Wild boar Plin. nat. 15,97 Nuts Tusculum Praeneste Peach Tusculum Plin. nat. 16,138 Pears Crustumerium Plin. nat. 15,53; Colum. 5,10,18 Asparagus Tibur Iuv. 11,64–76 Edible bulbs Praeneste Plin. nat. 19,97 Cabbage Birds Cato agr. 8,2; 51,54; 133,2; 143,3; Macr. Sat. 3,18,5; Plin. nat. 15,90; 17,96 Fish Game in general Wood Ostia Varro rust. 3,2,7–13 Rome Ath. Deip. 14,654d Sabina Varro rust. 3,2,15; 3,4,2 Tusculum Varro rust. 3,4,3; 3,5,8 Laurentum Mart. 9,48; 10,45; Varro rust. 3,13,2 Ostia Varro rust. 3,2,7–13 Antium Ath. Deip. 6,224c Laurentum Mart. 10,37 Tibur Macr. Sat. 7,16,15 Tusculum Varro rust. 3,3,8 Aricia? Stat. silv. 4,4,12–9 Plin. epist. 2,17,3; 2,17,5; 2,17,26; Symm. epist. 7,15 Stat. silv. 4,4,12–20 Laurentum Aricia Colum. 10,137–9; Plin. nat. 19,140 Sabina Nomentum Mart. 13,15 Praeneste Stat. silv. 4,4,12–21 Rome Plin. epist. 5,6,4 Sabina Tibur Tusculum Hor. carm. 1,22,9–12 Hor. carm. 1,18,1–2?; Mart. 7,28,1–4?; Plin. nat. 16,237?; Stat. silv. 1,3,17–8; 4,4,12–19; Strab. 5,3,11 Stat. silv. 4,4,12–9 Praeneste Mart. 9,60; Plin. nat. 21,16; 21,20 Tibur Mart. 9,60 Tusculum Mart. 9,60; Plin. nat. 21,27 Leeks Aricia Ostia Colum. 10,137–9 Colum. 10,137–9; Mart. 13,19; Plin. nat. 19,110 Plin. nat. 19,110 Onions Tusculum Plin. nat. 19,105 Radish Mons Algidus Plin. nat. 19,81 Turnips Rome Grapes, vineyards, wine Alba Ardea Plin. nat. 19,77 Ath. Deip. 1,26d; 1,26f; 1,33a; Colum. 3,2,16; Gal. (Kuehn) 6,334; Mart. 13,109; Iuv. 13,213–5; Plin. nat. 14,25; 14,64; see also Tchernia 1986, Appendix II pp. 324–5 Col. 3,9,2 Aricia Plin. nat. 14,12; 17,213 Caere Colum. 3,9,6; Mart. 6,73,3; 13,124 Gabii Gal. (Kuehn) 6,334 Labicum Ath. Deip. 1,26f Laurentum? Nomentum Plin. epist. 2,17,15 Rome Sabina Flowers Ath. Deip. 1,27b; Colum. 3,2,14; 3,3,3; Mart. 1,105; 10,48; 13,119; Plin. nat. 14,23; 14,48–52 Tusculum? Mart. 1,18; 6,92; 10,45; 12,48 Ath. Deip. 1,27b; Colum. 5,8,5; Gal. (Kuehn) 6,334; Hor. carm. 1,9,5–8; 1,20,1–4; Mart. 10,49; Plin. nat. 14,38; Strab. 5,3,1; see also Tchernia 1986, Appendix II pp. 328–9 Ath. Deip. 1,26e; 1,26f; Gal. (Kuehn) 6,334; Hor. carm. 1,18,1–2; Iuv. 11,64–76; Mart. 7,28,1–4; Plin. nat. 14,38; see also Tchernia 1986, Appendix II pp. 328–9 Varro ling. 6,14 Velitrae Ath. Deip. 1,27a; Plin. nat. 14,65 Tibur Mons Algidus 65 chaptEr 4 Table 4.4 Agricultural activities related to speciic villas mentioned in the written sources. Owner Status Area Acilius Sthenelus son of a freedman Nomentum L. Annaeus Seneca knight Nomentum Imp. Colum. 3,3,3; Plin. nat. 14,48–52 wine M. Aquillius Regulus senator Tusculum Imp. Mart. 7,31,9–12 a possible productive farm? Imp. Mart. 7,31,9–13 a possible productive farm? M. Aquillius Regulus senator Via Tiburtina 3rd mile M. Atilius Regulus senator Ager Pupiniensis Emperor/Claudius emperor Bovillae Emperor/Claudius emperor Q. Fabius Maximus Verrucosus Cunctator senator Faustinus ? Period Reference Imp. Plin. nat. 14,48–52 Tibur Mid-Rep. Colum. 1,4,2–3; Val. Max. 4,4,6 CIL XIV 2431; Granino Cecere Imp. 1995, 363 Imp. CIL XIV 3920 Ager Pupiniensis Mid-Rep. Val. Max. 4,8,1 Tibur Imp. dispensator 7 iugera sheep wine, olives coloni, vilicus, productive farm wild boar Fuscus ? Tibur Imp. Mart. 7,28,1–3 knight Digentia Imp. Hor. carm. 1,20; epist. 1,14; 1,16 Q. Hortensius Hortalus senator Laurentum Julia Magnilla senatorial woman Tibur Late Rep. Varro rust. 3,13,2 Imp. CIL XIV 3716 = I.It. IV,1 246 Decimus Junius Juvenalis ? Nomentum/Tibur Imp. Iuv. 11,64–76 L. Licinius Lucullus senator Tusculum Matidia minor imperial family Ficulea C. Plinius Caecilius Secundus senator Laurentum Late Rep. Varro rust. 3,4,3; 3,5,8 Solin 1975, 66–9 Nr. 112; AE 1995, Imp. 372e Imp. Plin. epist. 2,17 M. Pupius Piso Calpurninus (Frugi?) senator Tusculum Late Rep. Varro rust. 3,3,8; 3,13,1 Quinctilii Vari senator Tibur T. Quinctus senator Tusculum Remmius Palaemon ? Nomentum Hor. carm. 1,18,1–2 Mid-Rep. Liv. 7,39,11 Imp. 7 iugera, bailiff dispensator, vilicus Mart. 7,80,11 Q. Horatius Flaccus Imp. Comments wine Plin. nat. 14,48–52 dispensator sheep, chickens, vegetables, fruits aviary procurator summarum (horrearius) kitchen garden, fruits, wine? leporarium (wild boar, roe deer wine? farm wine a possible productive farm? Q. Aurelius Symmachus signo Eusebius senator Tusculum Late Ant. Macrob. Sat. 7,7,14 Q. Aurelius Symmachus signo Eusebius senator Tibur Late Ant. Symm. epist. 6,81; Macr. Sat. 7,16,15 vilicus, coloni; game? M. Seius ? Ostia Late Rep. Varro rust. 3,2,7–13 M. Terentius Varro senator Tusculum Late Rep. Varro rust. 3,3,8; 3,13,1 Late Rep. Cic. fam. 16,18,2; Att. 13,11 wild boar, bees leporarium (wild boar, roe deer) market garden M. Tullius Cicero senator Tusculum Valeria Messallina imperial family Tusculum Imp. CIL XIV 2751 vilica M. Valerius Martialis ? Nomentum Imp. Mart. 13,42; 13,119 wine?, fruits? the vicinity of Rome.288 In conclusion, it can be said that the kind of variation in planting Cato recommends is also visible in the array of passages referring to the Roman region. The two types of intensive cultivation preferred by Cato, vineyards and gardens, are also mentioned more often indicating probably intensive land use in the area. Animal husbandry is the other main type of agricultural production. It is not related directly to soils, but in general to land use. (Table 4.3; Plate III.2.) Sheep and cattle are mentioned as the only traditional animals in connection to the towns of the area. They were raised mainly for obtaining meat, wool and milk, but hides, bone and horns were equally useful by-products. Columella also mentions the proitability of selling young kids, chickens and piglets from suburban farms to the city keeping only a minimum on the farm for reproductive purposes.289 In addition to traditional animal husbandry, pastio villatica or growing animals on the farmstead is mentioned as a lucrative type of production. Pastio villatica covers birds other than chickens, ishponds (particularly in coastal villas) as well as bees, dormice and gamekeeping, e.g., deer and wild boar. It is described as a risky, but potentially inancially very proitable enterprise. Its main aim was to supply the specialty and novelty food market created by major urban feasts, both public and private.290 In addition, pastio villatica could provide aesthetic and recreational enjoyment for the landowner: aviaries, ish ponds and game 288 Spurr 1986, 8; Jongman 1988, 97–154. See Goodchild 2006 for suburbium. Cf. Scheidel 1994 for grain production and villas. 289 Colum. 7,3,13; 7,9,4; 8,5,9. 290 Colum. 8,10,6; Varro rust. 3,2,16. Cf. Kolendo 1995. 66 soils reserves provided beautiful sights and sounds as well as opportunities for physical exercise and hunting. Birds, bees and game are attested for the Roman area, but ish ponds are not mentioned in the written sources. Comparably to plants, no particular animals or animal products can be connected to the Roman area. Versatility was also possibly advisable in connection to animal husbandry. Most animals can transport themselves and all could be slaughtered in the city, so they could be raised even at a distance and transported cheaply and unspoiled to the market.291 Sheep transhumance is perhaps the obvious example of the lexibility of some types of livestock rearing. Moving locks between low-country winter pastures and mountain summer pastures is known in the area from the Bronze Age onwards.292 The market for young animals could mean that vicinity of the city was preferred during reproduction season(s). In terms of land use, animal husbandry would mean pastures, which do not necessarily have to be very close to the farmstead. Roman authors clearly recommend all sorts of agricultural and other economic activity for the area surrounding Rome in general. But what do they write about when their own or, e.g., neighboring villas are considered? Only a few villas in the Roman Campagna are described at length and, in ca. 30 cases, these descriptions include references to agricultural production (Table 4.4). The most famous of these villas is Cicero’s Tusculanum, which he acquired in the early 60’s BC and which probably remained in his possession until his death in 43 BC.293 In the 60’s, Cicero writes a great deal about the process of decorating the villa and, in the 50’s, after his exile, about its destruction and the cost of repairs. Two passages from the 40’s refer to the economic side of the Tusculanum. The irst is a letter to Tiro,294 with Cicero urging his trusted secretary to get a higher rent from the gardens he was letting out. The buildings and water installations had been renovated and the owner wanted more than the 1,000 sesterces that had been paid for it before these renovations took place. The second passage from a letter to Atticus from 45 BC is vaguer, but refers to Cicero’s concerns about getting rents out his tenants, possibly at the Tusculanum.295 Thus, it seems that Cicero’s Tusculanum included agricultural activity, although probably not directly by Cicero himself and mentioned very rarely. When the other cases are considered, the lack of references by Cicero is striking: he is the most important source for the villeggiatura of his times and yet he does not write about the economics of villa life.296 Selling and buying estates as well as what inancial problems the expensive villas caused their owners are topics discussed,297 but other economic activities 291 Cf. Witcher 2005a. E.g., Barker 1981; Bonetto 1999. 293 Some researchers (e.g., Shatzman 1976, Nr. 216) would like to see the fairly vague references concerning the sale of Cicero’s Tusculanum in the 50’s BC as evidence for Cicero selling his villa, which had been robbed and partially destroyed by Gabinius during his exile in 58 BC. In my opinion, the passages do not conirm the sale interpretation, as no temporal gaps in references to the villa can be seen. The speech passages mourning its destruction are from 58 BC (dom. 62 and p. red. in sen. 18), the sale reference was written 57 BC (Att. 4,2,7) and an unsuccessful sale is referred to in 56 BC (ad Q. fr. 2,2,1). The last passage could also be interpreted as referring to the property of Culleo, which Cicero was considering buying. 294 fam. 16,18,2. Mangiatordi 2003, 245–246 gives a slightly different interpretation of the passage: the garden was supposed to be inside the Tusculanum and the products would have been destined for Cicero’s own use. I think that she might have taken conducere to mean “to take care” in this instance (cf. id. 228) and not “to rent” as the whole passage would make one think. 295 Att. 13,11. The area he is talking about could also be Arpinum where the letter was written (cf. Mangiatordi 2003, 241–244). 296 The rather short discussion in Mangiatordi 2003, 241–246 on the production at Cicero’s villas gives a very good picture of how little he writes about productive activity. Despite her bold assertion that the productivity of Cicero’s villas has now been reconsidered, her conclusions remain few and mostly doubtful. 297 Cf. Walcot 1975 and Rawson 1976 on Cicero and private property. See, e.g., Att. 2,1 and 4,2,7 for Cicero’s personal inancial problems. For Cicero’s evaluation of city and country life in general, see Davies 1971. 292 67 chaptEr 4 are almost never mentioned. Cicero’s reluctance to discuss agricultural production means that the topic is treated almost exclusively in specialist literature, i.e., Cato and Varro. Varro mentions four speciic villas in the Roman Campagna:298 two of these feature game and leporaria, a third has both boar and bees and the fourth is the somewhat failed aviary of Lucius Licinius Lucullus (cos. 74 BC). The latter was also used as a triclinium, but it is brought into the economic sphere by mentioning it among the examples of pastio villatica. Three more passages refer to speciic productive farms in the Republican period, but they all belong to 4th or 3rd centuries BC and depict the legendary peasant–soldier–politician of the Middle Republic (Table 4.4). Each had a small amount of land in the vicinity of Rome – 7 iugera or slightly less than 2 hectares is mentioned for two of them – probably cultivated for subsistence purposes. This is clear in the case of Marcus Atilius Regulus (cos. 267 BC), whose vilicus died while he himself was campaigning in Africa and his family would have perished for the lack of cultivation had the state not granted them economic aid. For the Republican period, very little can be said of the type and distribution of agricultural production based on literary evidence on single villas. Slightly more evidence can be presented for the Early Imperial period, both from literature and epigraphy. The poets Horace, Martial and Juvenal all wrote about their own estates in the vicinity of Rome and were happy about their self-suficiency. They also disapproved of and ridiculed estates which did not have agricultural production.299 The most famous villa, apart from that of Horace at Digentia, is the Nomentanum bought by Seneca from Remmius Palaemon, who had made the farm very productive by hard work and a complete renovation of the vineyards. Another famous villa is the Laurentinum of Pliny the Younger which featured an orchard and a good kitchen garden. Part of the food consumed at the villa was produced there, although Pliny also describes the location as convenient because of the proximity of Ostia and Vicus Augustanus, where supplies could be bought. Pliny also writes much more about productive activities at his various villas than Cicero. After the 1st–2nd century AD, the only literary references to productive villas are those to Symmachus’s farms. He laments the problematic maintenance and tenants of the Tiburtine villa in one of his letters. The epigraphic evidence is more ambiguous, but it seems likely that the presence of a dispensator or vilicus/ vilica would mean production.300 What, then, is the picture given by the written sources? The general handbooks of Cato, Varro and Columella recommend cultivation as ingenuous as possible for areas close to the city. Animal husbandry is also recommended. Moreover, the suburban area is the place to which highly productive agriculture is connected.301 Various products from the surroundings of Rome are also often referred to. Descriptions of speciic villas do not seem to exhibit this economic potential at all, but rather accentuate the suburban area as a recreational space. Cicero ignores production almost completely in his writing, but the situation changes slightly in the Imperial period. It could be that it became more acceptable to write about economic matters in a similar manner to luxurious living.302 Consequently, if only written sources are considered, it is easy to understand the concept of the surroundings of Rome as an intellectual space, destined mostly for the civilized otium of the Roman elite in luxurious architectural settings. This would mean that the great economic potential would have been ignored by the very people for whom the agricultural handbooks were meant and who had the capital to invest in property as well as in many types of production which required large initial investments. 298 299 300 301 302 Book 3 of res rusticae on pastio villatica. E.g., Mart. 3,58. Carlsen 1995. Cf. Kolendo 1995. E.g., Corti 1991; Myers 2000. 68 soils This would seem very unlikely and in order to get a more complete picture of the agricultural land use it is necessary to turn to archaeological evidence. 4.5 archaEological EvidEncE for agriculturE Surface survey and excavation sometimes produce clear remains related to agricultural production such as parts of oil and wine presses, various other production spaces and storage facilities. (Appendix III; Plate IV.1.) These have been little studied, and not very recently, which means that the extensive ieldwork carried out in recent decades is excluded from previous studies.303 Approximately 160 certain oil or wine presses have been reported from the surroundings of Rome along with ca. 20 uncertain ones. To these could be added the ca. 80 references to macina or mills in reports, which do not always specify what kind of mill is in question, a lour mill or one for pressing olives. In addition to these, ca. 60 sites have remains or inds that point towards the existence of some unspeciied production. The recent largescale open area excavations in some parts of the Roman region have produced a considerable number of features connected to agricultural activities: ditches for, e.g., vineyards, irrigated crops and drainage as well as pits for planting trees and other plants. These excavations have revealed the very intensive use of almost every square meter of the countryside and support an attempt to reconstruct some of the agricultural environments.304 Pits and ditches have been found in ca. 70 locations, either at the sites or between them. (Plate IV.1.) Plant remains is one further category of evidence for agricultural production, but these have been studied very little. The only site where plant remains have been studied in the research area is Collatia 776 and the results are very interesting. The excavations of the villa revealed carbonized plant remains in the soil layers covering the loors of the production and storage spaces as well as the stables. Most of the remains were olive pits from both edible and oil producing types. Other edible plants found were barley, lentils, broad bean and cherry. In addition, Lolium seeds were found, possibly of Lolium temulentum or Darnel ryegrass, which is grain weed. Carbonized wood pieces were also analyzed and these included wood from probably downy oak (Quercus pubescens), olive, elm, willow and beech. Downy oak is good for irewood, but it has also been used for building purposes. Its acorns are also edible. Some Mediterranean maquis species were also present: Phyllirea latifolia (It. ilatro), Pistacia lentiscus (It. lentischio; mastic tree), Pistacia terebinthus (It. terebinto; terebinth), holm oak and myrtle (Myrtus communis). Of these, the mastic tree, terebinth and myrtle can be used for various purposes, e.g., in cooking and for medicinal purposes.305 The plant remains from two sites outside the research area dating to the Iron Age/Archaic period and to Late Antiquity have been studied recently and it is worthwhile to look at them in this context as comparative material. The Iron Age/Archaic inds have been dated from the 8th through to the 6th century BC and they originate from excavations in central Rome. The remains found contain ive different cereals: einkorn, emmer, spelt, naked wheat as well as barley. Emmer is the dominant grain. Other edible plants include various legumes (horse beans, bitter vetch, common vetch, peas) and some fruits (igs, grapes). Olives are present 303 Rossiter 1978; 1981. E.g., De Seña 2005 lists 50 sites with wine and oil production facilities, but he has not consulted the Latium Vetus publications, which leaves a large gap in the information he provides. The data and distribution maps presented in Corrente 1985 are equally schematic. The data presented in Appendix III is more complete, but it is probably an equally imperfect collection of data, particularly regarding the regions outside the research area. 304 Di Blasi et al. 1999; Di Manzano 2001; Pracchia 2001, 286–308. 305 Musco 1984, 101, nota 35; Campolmi 1993, 157, 179. 69 chaptEr 4 only in small quantities. The species identiied from early Rome match those cultivated also in later times.306 The Late Antique material comes from La Fontanaccia at Allumiere, 50 km northwest of Rome. The site is an abandoned villa, where a small hut was set up in the AD 450’s. The most common seed found there was that of grass pea (Lathyrus sativus), which is currently used as animal fodder or an emergency crop during famines. Barley was the most common cereal and it is of the naked, two-row kind, which is not very common after the Bronze Age. Some wheat grain was found. Remains of fruits were also rare with only a few grape pips and acorns, which might have been also used as fodder. These indings differ greatly from the variability and abundance of the earlier periods and have been interpreted as a sign of the troubled times and degradation of agricultural practices.307 The amount of archaeological evidence for growing plants is not great considering the high number of known settlement sites. In the research area alone, ca. 1,250 villa sites are known and among them 53 partes rusticae, 127 oil or wine presses and 62 grain or olive mills have been found. This means that slightly less than 20% of all sites show evidence of agricultural production. In a collection of one hundred excavated villas published recently, 80% of the villas had evidence of agricultural production.308 In addition, planting pits and ditches have been found in 52 locations – these are naturally located between settlements. Remains related to agricultural production have been found in all kinds of villa buildings, from modest farms to monumental residential complexes. In the research area, the 274 sites with these inds are mostly Class 1 villas, i.e., those with the most elaborate evidence for various building parts, decorations and use. The more modest Class 2 and 3 sites do not feature them very frequently and only one Class 4 site with remains related to production is known. (Table 4.5.) This situation corresponds quite well with that visible in the excavation data: 83 of the one hundred sites had residential and productive parts and 11 sites featured just production parts.309 The most common type of villa in the area is probably a combination of residential and productive units, whose size and quality of decorations vary from modest to very luxurious. Even quite large complexes feature signs of agricultural production. The inds have been dated to all periods ranging from the Archaic to Late Antique and Medieval times (Appendices II–III). The most accurate dates can probably be derived from the excavated sites and 55 of the excavated remains related to agricultural production have been dated.310 The majority of the sites were active in the 1st century BC (39 sites) and/or 1st century AD (21). Only four date to the period before the 2nd century BC, ten can be dated to the 2nd century AD and six are even later. These igures are not representative for the whole habitation of the research area, but they give an impression of the temporal variation in productive activities. In general, they correspond well to the general period of use of the sites, i.e., most sites have been settled during the 1st century BC and 1st century AD.311 How long the installations have been in use cannot be determined. Some new installations were still being built in the 2nd century AD, but their total number seems to diminish starting from the 1st century AD. The distribution of these discoveries in the research area is clearly concentrated in its well-studied and well-published northern and northeastern parts, where it is thus perhaps possible to get an idea of the intensity of the agricultural land use. The distribution is at 306 Constantini and Giorgi 2001; Motta 2002. Sadori and Susanna 2005. 308 De Franceschini 2005, 293–295, 315–320, 349–350. 309 De Franceschini 2005, 349–350. 310 De Franceschini 2005, 369, 381–382. In addition to the actual production remains, I have included here simple dated cocciopesto loors classiied as rustici as well as opus spicatum loors which are often related to production spaces. 311 Cf. Chapter 2.3 and De Franceschini 2005, 297–298. 307 70 soils Table 4.5 Summary of classes, dated sites and remains related to agricultural production. >1 = more than one feature; PR = pars rustica. its lowest in the areas around Tibur and Tusculum and more remains related to agricultural production Site type Number >1 Ditches Presses Mills PR Animals have been found around Class 1 127 11 6 62 13 32 3 Tibur than Tusculum. This Class 2 68 2 6 27 22 10 1 Class 3 36 0 9 16 9 2 0 can relect the residential Class 4 14 0 10 2 2 0 0 character of land use in Early 8 0 7 0 1 0 0 these towns, but could also Rustic–Residential 101 4 8 52 25 6 6 be explained by a lack of Pars rustica 7 1 0 5 0 1 0 recent large-scale open area Other 27 0 20 2 2 0 3 Stray/Stray? 40 0 0 5 35 0 0 excavations. In Tusculum, Total 428 18 66 171 109 51 13 the very intensive modern land use has destroyed many sites or makes studying the remains dificult. The distribution of these production remains is thus biased and it is dificult to draw conclusions for the whole Roman region based on it. Animal husbandry is more dificult to discover archaeologically without excavation than plant growing and processing and even then, determining the function of various spaces is highly uncertain. No certain indications of what a sheep pen or a pig sty should look like can be given312 and they could also be located outside the actual villa buildings like at the Setteinestre Villa in Etruria.313 Finding pastio villatica is equally dificult, e.g., only two aviaries as parts of a villa have been recognized in the research area and even these have to be considered as uncertain interpretations (Appendix III). Fish ponds are also rare, only ten, but the proper identiication of the function of the basins is dificult. The research area is located inland, which also naturally reduces the number of possible ish ponds.314 Two gliraria or pots for raising dormice have also been found at the villa sites Collatia 477315 and Tibur IV 20a. Another source for animal husbandry is bones, but they have been little studied in connection to the Roman countryside: osteological analyses have been reported from only three villa sites and only one has been published.316 The inds from the villa Tusculum 414– 418 included most commonly young pigs, ca. 2 years of age. The material does not allow conclusions to be drawn on the character of the pig breeding. Sheep and goats were less common and generally less than one year old. Cattle could not be identiied with certainty. Chickens were mostly adult females. The inds included also ish and mollusks, such as oysters, grooved carpet shell (Tapes decussatus) and wedgeshell clam (Donax trunculus). In addition, rat bones (Rattus rattus) were found. The results correspond well with, e.g., those from the Setteinestre Villa.317 The villa site Collatia 392 was probably a small pig farm with a herd of ca. 50 animals.318 312 Rossiter 1978, 60–61; Carandini 1985a, 158, 182–188, 195–196. Carandini 1985a, 182–188, 195–196. 314 Higginbotham 1997, 110–167 lists eight inland ish ponds, which are included in Appendix III apart from the two urban ponds found in Rome (Domus Tiberiana and Domus Augustana). All the coastal ish ponds he mentions are from south of Antium: Nettuno, Astura, Circeo, Sperlonga, Formia and Scauri (cf. Lafon 2001a, 164–177 for a very similar distribution). Villas in the stretch between Caere and Antium do not seem to feature ish ponds (Lafon 2001a, catalogue of villas). Cf. also Marzano 2007, 47–63. 315 Carpaneto and Cristaldi 1995. 316 MacKinnon 2004 mentions three sites in the immediate vicinity of Rome: Collatia 392 with Republican and Imperial period deposits, Collatia 386 for Late Antique deposits and the Villa of the Quintilii at Monteporziocatone (= Tusculum site 414–418) for Imperial period deposits. 317 De Grossi Mazzorin 1987; MacKinnon 2004, 153–159. 318 MacKinnon 2004, 153–159. 313 71 chaptEr 4 The rural production/consumption patterns reported are based on a small amount of data, but it seems likely that oxen and cattle were mostly used as draught animals in the central Italian countryside. It has also been suggested that as cattle requires good and large pastures, they would not have been raised much in the intensively used countryside east and north of Rome, but might have been more common to the west and possibly to the south. Sheep and goats were commonly reared for meat, wool and milk and could have been sold in the markets as well. Sheep raising might have been more common after the Second Punic war when peaceful conditions in the countryside made transhumance possible.319 Compared to the total number of sites, relatively few remains related to agricultural production have been found. Deinitive proportional numbers are hard to give – as was mentioned above, only 20% of the survey sites and 90% of the excavated sites feature production remains. The igure for excavation data is probably closer to the truth due to more detailed data. This proportion could be even higher considering how the sites are only partially excavated. The distribution map of the discoveries is biased towards areas that have been recently studied and published. However, despite the problems mentioned above, some conclusions can be drawn. Remains related to agricultural production have been found practically everywhere in the area starting from the outskirts of Rome. Most of the excavated sites are located in the plateau area, which would probably indicate very high production intensity for that landscape unit. The normally large and elaborately decorated villa complexes in the slope areas feature less signs of agricultural production. Chronological patterns are perhaps even harder to detect than those based on geographical distribution, but based on the excavated sites, the period of greatest activity dates from the 2nd century BC to 2nd century AD. Almost all kinds of production is known in the area; olives and grapes are well attested. Grain cultivation is perhaps the most poorly known aspect, as the discovered mills could have ground grain that was produced elsewhere, bought from the market or received from the state. Next, palynological data will be analyzed in order to see if it can reveal more information on the land use in the Roman period, particularly on grain cultivation. 4.6 palynological EvidEncE for agricultural activitiEs Palynology is the study of pollen, spores and other plant related contemporary or fossil particles in sedimentary rocks and sediments.320 They are most commonly retrieved from the sediments deposited at the bottom of lakes or in marshes. Pollen and spores can be recognized fairly accurately and the variations in the amounts of each species or plant group indicate changes in vegetation around the basin where the pollen has been deposited. The occurrence of pollen in the sediments depends mostly on how the plant species spread their pollen, i.e., those plants whose pollen is transported by wind are best represented. As a result, pollen diagrams represent vegetation conditions of a region, but also of areas from some distance of the sampled basin. The most important problem related to determining agricultural activity is that many of the most important cultivated cereals have pollen not dispersed by wind. Thus, they are fairly poorly represented in pollen diagrams.321 Another important Roman crop not visible in pollen analysis is grapes, usually not featured in diagrams at all.322 Despite these 319 320 321 322 Bonetto 1999; MacKinnon 2004, 90–97, 120–133, 153–159, 163–170. Faegri and Iversen 1989. Faegri and Iversen 1989, 127. Cf. Turner and Brown 2004. 72 soils shortcomings it is possible to detect changes, e.g., in the type and intensity of human activity by looking at many indicators such as changes in the amount of arboreal, shrub and grass pollen, or occurrence of anthropogenic weeds and other plants. A number of small crater lakes as well as marshy areas existed in the Roman Campagna until quite recently. Some have been drained for agricultural purposes, e.g., Lago di Castiglione by ancient Gabii, but some still exist as lakes, most notably Lago di Albano and Lago di Nemi in the Alban Hills. All three of these lakes have been subjects of environmental studies during the past decades. Each lake and its sediments have been analyzed for pollen and other indicators for reconstructing the vegetation history and other environmental changes in the area. Lago di Castiglione is located in the plateau area and the two other lakes on the higher slope areas, thus representing vegetation from slightly different landscapes. The longest sequence has been recovered from Lago di Castiglione. Unfortunately for the period of interest here, 500 BC – AD 500, the uppermost part of the sequence is not well preserved and the main interest in the publications has been in the reconstruction of long term climatic and vegetational changes. The uppermost dated zone is roughly Bronze Age (VdC18 3480 uncal. BP) and it reaches almost to the top of the sediment sequence. Mediterranean type vegetation dominates this zone and human activity in the area is visible from 1.15 m upwards as a general decrease of arboreal pollen and spread of chestnut (Castanea) and hazel (Corylus). Nevertheless, nothing can be said of the Roman period vegetation and signs of human impact.323 The research conducted in Lago di Nemi and Lago di Albano has fortunately produced better results with regard to the Roman period as both sediment sequences cover this era. In Lago di Albano, subzone IIId starts at 2950 cal. BP, i.e., during the Late Bronze Age, and the Roman period can be found at a depth of 1.50 m. The most prominent signs of human activity are the cultivated trees, chestnut, walnut (Juglans) and olive (Olea), which all start to increase at the bottom of the subzone and reach high levels in the Roman period. Cereals have also been detected relatively clearly: barley, oat (Avena) and wheat pollen are continuously present for the latter part of the Holocene. Their proportions in the non-arboreal pollen counts do not increase or change radically during the Roman period. Grapes are mentioned as having increased in the Roman period, but they have not been included in the diagrams. The same cultivated trees and cereals are also present in the Lago di Nemi subzone IIId, which marks the Roman period. The greatest difference with the Albano sequence is the rather late increase and continuous occurrence of the cereals. Other indicators of agriculture and human activity in general also follow the same trends in both lakes.324 4.7 intEgrating thE EvidEncE Site type distribution and quality of soils The irst task was to compare the distribution of the site types to the best quality agricultural soils (Fig. 4.4; Table 4.6). The best soils cover 32% of the area featured in the soil survey and they are mostly located in the northwestern and southwestern zones of the research area. In the analysis, the number of sites located in the soil survey area was compared to the distribution of the best soils. A completely random selection of locations would mean that 30% of the sites should be on the best soils. The results show some correlation between certain site types and 323 324 Alessio et al. 1988; Follieri et al. 1988; 1989; 1998. Lowe et al. 1996; Mercuri et al. 2002. 73 chaptEr 4 the best soils (Table 4.6). The number of Class 2 and 3 sites near or directly on the best soils is slightly elevated; whereas the igures for Class 1 and 4 sites are slightly lower, but also higher than the expected random percentage. Class 2 and 3 sites represent the more rustic villa (and farm) types and thus the result of this analysis strengthens their interpretation as villae rusticae. The results in general correspond fairly well with the noted agricultural activity: the sites with either attested or interpreted rustic qualities are situated more often closer to the best agricultural lands than other types of sites. Additional observations can be made, Best soils = 32% though, but mostly concerning smaller Expected = 30% Sites On best On best % regions. As mentioned previously, the Class 1 160 57 36 Sabatine paleosol geological formation Class 2 230 100 43 present in the northwestern area might Class 3 263 111 42 Class 4 412 161 39 have been a preferred type. In the lat Archaic 282 142 50 ridge crests, this paleosol results in very Early/Middle Republic 400 222 56 good agricultural soils and 25% of them New sites 2nd c. BC 82 38 46 are among the best soils (Plate V.1). New sites 1st c. AD 304 110 37 The slopes in its area are among poorer Late Antiquity 89 42 47 soils recommended mostly for shrubs or Fig. 4.4 and Table 4.6 Distribution of best trees (Fig. 4.3c). The settlements of the soils (dark grey) for various crops and northwestern area are mostly Class 2 and 3 number of sites by class and date. villas with high quantities of scatters of tile and pottery, i.e., Class 4 sites (Table 4.7). The best soils were also preferred as 40% of all sites are directly on or very close to them. The Class 1 sites are concentrated in the central zone of the best soils and close to the main roads, the Via Salaria in the west and the Via Nomentana in the east. Most of the remains related to agricultural production have been found in the same zone. Class 2 sites are often located outside the area of the best soils and can be found particularly towards the Aniene. The sites become more rustic and poorer moving away from the city and the main roads. The large villas also have control of the best agricultural soils close to the city with the best connectivity in accordance with the advice given by the agronomical writers.325 When the proposed road lines are added to the picture, the tentative conclusion could be drawn that one major site (Class 1 or 2) with a production part was built in an area bordered on all sides by roads. In addition to the rich villa, two to eight other sites were present in the same area. The owners of the larger villas could have established secondary farms, market gardens or other production facilities for tenants. The sites further away from the city could also belong to independent peasants.326 An example of a larger villa is the site Fidenae 197, which has also been attributed to Phaon, a freedman of the emperor Nero.327 The archaeological remains are poorly preserved, 325 326 327 Cf. Di Gennaro et al. 2005 for a notion of the largest sites concentrating near the city in this area. Cf. Dommelen 1993; Marcone 1997, 129–132. The attribution is based on Suet. Nero 48 describing Nero’s escape to the villa in the area between the fourth 74 soils but the villa was probably of considerable size.328 In addition to the residential part, a wine and/or oil press has been found. Although Suetonius’s text might not pertain exactly to this villa, it is, nevertheless, an interesting description of a villa’s surroundings in this general region. Nero, fearing for his life, leaves the city on a horse by one of the main roads. Then he takes a deverticulum on foot going through a thicket of thorny bushes and other shrubs (fruticetum, vepres), then reaching reeds (harundinetum) near the back wall of the villa. No actual agricultural production is mentioned, but the shrubs and reeds by the villa correspond well with the position of the site near a small river valley with soils best suited for shrub and trees. This is also what Cato329 advised to be grown in river valleys near the borders of the villa. Another interpretation can also be offered: the bushes and shrubs could indicate neglected surroundings, which in Italy quickly become covered by an almost impenetrable and invariably thorny thicket. The image of an unpleasant walk through thorny bushes to the villa emphasizes the unhappy last moments of the emperor. The reeds close to the building could still very well have been part of the agricultural production of the villa as they (and also willows) were needed for supporting grape vines or making baskets.330 The situation in the central part of the research area is different from the northwest. The general fertility and suitability of the Roman region for almost all agricultural activity are evident from both environmental and literary evidence. Most of the area is regarded as good by most ancient authors with the notable exception of the ager Pupiniensis located below Tusculum. When the modern soil suitability classiication is examined, the soils of the plateau on both banks of the Aniene continuing southwards to the modern Via Casilina are among the mediocre or merely good for agricultural purposes. The recent geoarchaeological studies in this area also demonstrate concern for soil conservation and attempts to take care of the agricultural soils. This area is considered good for modern grain cultivation, but the suitability analysis resulted in fairly poor possibilities for growing emmer, other wheat varieties and barley; millet would be fairly suitable. This matches well the ancient descriptions of the area that has traditionally been associated with the ager Pupiniensis. The problems of obtaining a livelihood encountered by the Middle Republican peasant–soldier–politicians mentioned in later Roman literature could be a partial relection of the true conditions. Subsistence farming of grain in the ager Pupiniensis could have been low yielding and problematic. The passages could also be interpreted as a literary topos, accentuating the hardy and truly Roman character of the Republican heroes. But even in this context, the heroic deed of extracting a livelihood from the famously poor soils would have further emphasized the quality of the men. The passages could also be interpreted as a metaphor for how agricultural know-how had diminished in later times: the Late Republican and Imperial landowners could no longer farm as well as their ancestors did.331 As mentioned above, the soils of the central zone are slightly poorer in quality than in most of the research area (Plate V.2). Slightly more small sites can be found, particularly between the Aniene and the modern Via Casilina, the western part of which follows the line of a Roman road, the Via Labicana.332 Twenty-two Class 1 sites are known between the Aniene and the Via Praenestina and most of these are located in the eastern part of that area, away from the city. The eastern area also features the majority of the remains related to agricultural milestones of the Via Salaria and the Via Nomentana. In addition to the accurate distance from the city, a funerary urn CIL VI 34916 of one Claudia Egloge has been found in the area. It has been suggested that the person was Nero’s nurse (Suet. Nero 50). 328 Most detailed description in De Franceschini 2005, 98–101 Nr. 32. 329 agr. 6,3. 330 White 1970, 394. 331 E.g., Colum. 1, praef.; Plin. nat. 18,19–21. Valencia Hernández 1991, 29–64. 332 Quilici 1974a, passim. 75 chaptEr 4 production. The soils get better south of the Via Labicana and the number of Class 1 sites increases. Between the Via Praenestina and the Via Labicana, 45 Class 1 villas have been found and of these, ifteen feature remains related to agricultural production. Even here, the majority of the rich sites concentrate in the eastern zone. Hence, the large landowners do not seem to have been very interested in building their villas in the less fertile area despite its closeness to the city. The importance of the best soils can also be seen in the area south of the Via Labicana: Class 1 sites are found mostly in the area of the best soils. Land in the less fertile zones could have been available as small farms or market gardens, either through tenancies or actual ownership.333 Based on the great number of 1st century AD small sites, it has been suggested that the main type of cultivation would have been vegetable and fruit gardens.334 The interpretation also maintains that gardening as a form of cultivation developed only in the Early Imperial period based on the small amount of attention that Cato and Varro give to gardens and the great interest shown by Columella. Gardens must have existed both for the use of the owner as well as for selling produce for market, but not all small sites can be interpreted as gardens. Few of these sites, particularly of the Class 4 scatters of tile and pottery, have been excavated and published, so it is dificult to determine what these small sites represent. One such site is Collatia 386, which was found in the early 1960’s as a scatter of pottery, tile and some building.335 Excavated, this scatter proved to be a villa, which had been used from the beginning of the 3rd century BC to the 3rd century AD. Four main building phases were recognized and agricultural activities included wine and olive oil production.336 These small independent units with modestly decorated living quarters, production parts as well as often also baths, have been found in many areas around Rome.337 These villae rusticae were probably the basic unit for habitation and production in the area. The small sites are most common in the central part of the research area, where the possibilities for subsistence farming or orchards are not very good, but which are very well suited for wine and olive growing. Small plots could have been under intercultivation with grapes, olives or fruits being the main crop with possibly grain growing between these.338 Some fruits and vegetables could also be grown on the plot. The recently excavated Villa Regina near Pompeii is possibly a good comparison for the sites near Rome and also supplies detailed data on types of cultivation. A small vineyard surrounded the villa with fruit and/or olive trees growing here and there. A small kitchen garden was also found near the entrance to the villa.339 The situation changes moving southeast in the study area. The large villas documented on the slopes west of Tusculum feature only a few production sites (Table 4.8). This is interesting considering that the volcanic soils are suitable for agricultural production despite the steeper slopes and growing elevation. The majority of the sites are large Class 1 and 2 building complexes. A similar area of large complexes can be found on the limestone slopes around Tibur. (Table 4.8.) In Tibur, more remains related to agricultural production have been found scattered fairly evenly in the area. The poorest agricultural soils as well as some of the steepest 333 Recently, something similar has been tentatively suggested for the Via Flaminia region: either small independent farms or tenancies controlled by a larger unit. Here, the sites have been excavated and the buildings include almost invariably living and production quarters as well as a baths and doliaria. Messineo 2003 and 2005 with references. 334 Carandini 1985b. The interpretation is partially based on von Thünen’s model of agricultural production around a city, which has gardens in the innermost areas. 335 Quilici 1974a is a vague description of tiles and pottery, based on which the site was interpreted as a villa rustica. The site description by Kahane and Ward Perkins 1972 is more detailed and even records some site history. 336 See note 115 for full references. 337 E.g., Di Gennaro et al. 2005; Mari 2005. Most of the villas described by De Franceschini 2005 belong to this category as well: on p. 349, she lists 83 cases out of her selection of one hundred sites. 338 Cf. White 1970, 48; Spurr 1986, 6–7. 339 De Caro 1994. 76 soils Table 4.7 Classes of sites at Tibur and Tusculum. slopes of the whole research area can probably be found in Sites Class 1 Class 2 Class 3 Class 4 Production this latter zone, but the poor Tusculum 115 30 50 27 8 5 conditions do not seem to have Tibur 108 48 27 19 14 26 prevented large villas with partes rusticae from being built there. The lack of inds related to agricultural production in the large villas at Tusculum would seem to correspond with the picture transmitted by the ancient authors: they were intended for recreation, not production. But a comparison of the Tiburtine area and the region of Tusculan makes this interpretation problematic: why would there be no signs of agricultural production in the area affording much better opportunities for such activities? One explanation could be the different possibilities for archaeological research: the villas in Tibur are still mostly in the countryside, outside the modern town, whereas the villas are almost all covered by later buildings and surrounded by dense modern habitation at Tusculum. Many ruins were effectively covered by Baroque villas, preventing later land use and opportunities for stray inds. Therefore, the lack of production remains at Tusculum is partially related to site formation. On the other hand, it may relect true conditions. Why the slopes of Tusculum would have been less cultivated than those of Tibur is hard to understand based on the advice given by the Roman agronomists on production in the suburban areas. The last section is the southwestern corner. This area features mostly good or excellent agricultural soils and it is suited to almost every kind of production (Plate V.3). It is also at an equal distance from the city as the northwestern part. These physical conditions make the areas very similar, but the difference is clear when the distribution of sites is compared. The southwestern zone features only a small number of sites compared to the dense habitation of the northwest. This may result from a gap in research and publication, but, on the other hand, it may also relect the Roman period situation. The quality of soil would certainly have not been a hindrance to establishing habitation in this zone. Changes in site selection over time? The chronological distribution of settlement can be studied best in the northern and central parts of the research area for reasons that have been stated above (see Chapter 2.3). The periods examined are the Archaic and Early Republic, 2nd century BC, 1st century AD and then the 4th–5th centuries AD. Each of these shows major changes in habitation. The Archaic period habitation is dense, but starts to diminish in the Early Republic becoming denser again during the Middle Republic. The 2nd century BC is marked by a great number of completely new sites as is also the 1st century AD. The locations of the early sites are analyzed in relation to the best soils, and in later periods, it is interesting to see where the new sites are established. The Late Antique sites can show where the villas longest in use are located. The same type of procedure as in the previous section was conducted with all sites dated to the periods mentioned above, i.e., the site distribution inside the area of the modern soil survey was compared to the distribution of best soils. In the earlier periods, the sites that have not continued to be occupied after the Middle Republic were compared separately to the sites showing signs of occupation in the Late Republic and Imperial times. In the period before the 2nd century BC, a clear connection between site distribution and the best agricultural soils can be observed (Table 4.6; Fig. 4.5a). Almost half of the sites are consistently very near or on the best soils between the Archaic period and Middle Republic. In the Archaic period, the site distribution is even in the northern and central parts; and no difference between the sites that continue to be used later and those that are completely 77 chaptEr 4 Fig. 4.5 Distribution of the best soils (dark grey) and a) sites inhabited only before the 2nd century BC, b) sites established in the 2nd century BC, c) sites established in the 1st century AD and d) sites occupied during the 4th–5th centuries AD. abandoned at the end of the Middle Republic can be po inted out. Half of the Archaic sites were abandoned and this is most clear in the central area where the sites located west of Gabii vanish almost completely. Only a few new sites are established and these are scattered around the old settlements. Continuity is high between the Early Republic and Middle Republic, with the highest numbers of sites connected to the best soils – 60% of those in the soil survey area. The situation changes drastically in the 2nd century BC (Fig. 4.5b). For the irst time since the Archaic period, the total number of sites climbs over 400 and a great number of new sites are established, 184 in total (45% of all sites). The most signiicant areas which seem to have been inhabited for the irst time now are the slopes west of Tibur and Tusculum. Approximately 45% of the new sites are located in the soil survey area and of these, 46% are located in areas with the best soils. Habitation does not increase very much during the 1st century BC. The densest habitation dates to the 1st century AD with 1,167 sites in total (Fig. 4.5c). Of these almost half are new, 501 sites (43%). Most of the new sites are located in the central zone and, consequently, the rate of new sites on the best soils drops to 36%. It is worth noting 78 soils that the new Class 1 and 2 sites are established most often in the Via Labicana region, where the best soils of the central area can be found. From these results, it seems clear that for most of the period between 500 BC and AD 500, locations near the best agricultural soils were preferred for habitation. This is evident for the Archaic and Republican periods, when subsistence farming was the basis of livelihood and grain cultivation was important. The increase in population begins to change this tendency only in the 2nd century BC, when the slope areas become inhabited. The poorer soils in the central zone remain unpopular even in this phase and the new sites still tend to be located near the best soils. Unfortunately, the main expansion area of this period, the slopes near Tibur and Tusculum, remains outside the soil survey. Considering these areas, it is clear that the growing elevation, steeper slopes and, in the Tiburtine area, the calcareous geology make many types of cultivation less productive. The central area becomes densely inhabited only during the next expansion phase in the 1st century AD. The larger sites are then fairly often situated in close proximity to the best soils, which shows that their location could still be selected, perhaps with production potential in mind. The same relative numbers of Late Antique sites are located near the best soils as were in Archaic and Republican eras (Fig. 4.5d). This could point towards an emphasis on grain cultivation and subsistence farming. Changes in types of production? All sources used here generally offer a picture of versatile production. Agronomists recommend planting various crops in the vicinity of the city. Written sources cite many different fruits, vegetables, cereals, wine, olive oil as well as animal husbandry in almost all its forms in connection to the area. Archaeological evidence is clear for wine and olive oil production, but also other types of production, e.g., animal husbandry, are present in the archaeological record. Palynological evidence conirms this picture supplementing it with some cultivated trees as well as hemp and hops. Grain cultivation is veriied by pollen data for the whole period examined. Complex and developed agriculture already began before and continued throughout the Roman period. The good quality of soils allowed for most production and Roman agriculture was already also quite sophisticated and had developed species of plants as well as agricultural practices for tackling most environmental problems. However, it would be foolish to expect that agricultural activities would have remained the same for the whole period considered here. Chronological changes in types of production are mostly hard to detect. In the past, possible changes have been interpreted based on changing numbers of settled sites and literary sources.340 The traditional subsistence farming of free peasants has been suggested to change into commercial agriculture with a slave workforce in the 3rd–2nd century BC.341 This could have meant changes in selecting locations, as the main crops would have been grapes and olives instead of grain. Palynological data offer some help, but the record is not suficiently detailed or well-dated for a very thorough analysis. The pollen for cultivated trees, chestnut, walnut and olive, peaks at 2500 BP and is followed by a drop and another rise soon afterwards. These high levels continue well beyond 2000 BP. The irst peak could perhaps be connected to the dense Archaic settlement, the subsequent drop to the Early/Middle Republic and the new rise to the intensifying in habitation of the 2nd century BC. The later peak in cultivated trees is much lower than that of the mid-1st millennium BC and could indicate diminished cultivation of olives during Roman times. It is unfortunate that grapes are not visible in the pollen record, 340 E.g., Ikeguchi 2000. E.g., Toynbee 1965, 155–189; White 1970, 384–412; Neeve 1984; Pucci 1985; Marcone 1997, 123–150; Volpe 2000. 341 79 chaptEr 4 making comparison of the main crops impossible. Some changes in site distribution can be noted in the 2nd century BC with a considerable number of new sites, particularly in Tibur and Tusculum. The majority (127 of 182 sites, 70%) of the new sites are of Classes 1 and 2. This its well with the idea of a wealthy landowner establishing a farm. As both of the main expansion areas feature few production remains, it is dificult to determine what changes this might have meant for agricultural activities. Higher slopes on the eastern perimeter of the plateau area were selected as sites for building new villas. These areas are not perhaps the best places for agricultural production, but it is not impossible either. The proximity of the best agricultural soils seems to have been appreciated, pointing towards the importance of production potential in site selection on the plateau. The few dated production facilities in the whole Roman area, particularly wine and olive presses, tend to be – at the earliest – from the 2nd century BC. On the other hand, certain vineyards from the Middle Republican period are known based on indings. Thus, the settlement pattern changed as new sites were partially located in different places compared to previous periods and they were also generally of the larger and wealthier type. It is more dificult to connect these sites to the proposed changes in agricultural production.342 For the Late Republican and Early Imperial period, all evidence points towards intensive and variable agricultural production covering arable farming as well as animal husbandry. Some of the written sources, most notably Cicero, offer a very different picture of almost no production at all. Both are probably right. A great need for agricultural production for consumption both in the countryside and in the city existed,343 but villas with little or no production at all were also probably present. Based on archaeological data, these nonproductive villas were probably a minority, but they existed nevertheless. The disapproval of non-productive estates voiced by many different literary sources was probably not a mere literary topos.344 Pliny the Elder’s reference to poorly maintained farms and the resulting low prices of land in the 1st century AD seems real. The conservation of soils was a concern for those working the land, which has been attested archaeologically. New types of production are also attested by mainly written sources – pastio villatica. In addition, hemp iber was produced in the Alban Hills area as shown by pollen data. The greatest density of habitation in the Early Imperial period meant that areas which are perhaps not suited very well for agriculture were also occupied and the whole plateau area seems to have been exploited very intensively. An agricultural crisis concerning commercial crops of grapes and olives has been suggested for Italy for the 2nd century AD based mostly on literary sources as well as some survey and excavation data.345 It has been suggested that cheap imports from the provinces outside Italy caused a decline in Italian agricultural production. This hypothesis has been widely discussed and, based on the growing archaeological record from various regions of Italy, it has been mostly rejected.346 In the Roman region, the 2nd century AD does not seem to be a great period of change where settlement patterns are concerned. Some sites do cease to exist (10%), but, on the other hand, new production facilities are still built. In the pollen record of Lago di Albano, a decline in the amounts of cultivated trees can be seen, but this probably happens later than the 2nd century AD.347 Changes in patterns of landownership have also been suggested based on literary and archaeological evidence. Pliny the Younger 342 For an equally inconclusive discussion concerning the period in southern Etruria, see Patterson et al. 2004, 13–17. Cf. Purcell 1995; De Seña 2005; Witcher 2005a; 2005b; Goodchild 2006. 344 Cf. Purcell 1995. 345 E.g., Pucci 1985; Carandini 1988, 267–285; Sirago 1995, 447–476; 1996, 39–80; Vera 1995a; 1995b; Marcone 1997, 151–172; Volpe 2000; Marzano 2007, 199–222. 346 E.g., Tchernia 1986, 295–299; Patterson 1987; Vera 1995a; 1995b; Majbom Madsen 2003. 347 Mercuri et al. 2002, Fig. 7. 343 80 soils mentions in one of his letters that the price of land was rising because of an edict by Trajan stipulating that senators should own a third of their landed property in Italy.348 This has been connected to the epigraphic evidence for provincial senators present in the area in the 2nd century AD as well as the construction of many large villa complexes on the plateau area close to Rome.349 What changes in land use might be related to these events is less clear. Estates grew in size and the villas on them were intended merely for residential purposes. The archaeological evidence cited comes mostly from outside the surveyed regions, closer to Rome, but as mentioned above, no great changes occur in the areas further away from Rome during this period. A great decline in site numbers can be seen, however, starting from the 3rd century AD and the difference between 3rd and 4th century AD igures is already striking: ca. 980 and ca. 135 respectively. The sites that survive to the 4th–5th centuries AD are relatively few (141, 7% of all sites) and their distribution is mostly concentrated in the vicinity of the best agricultural soils. Some new facilities are still built for wine and/or olive oil production. The most recent of the cultivation pits and ditches reported in the surroundings of Rome are from the Medieval period and they are also those closest to the Aurelian Wall, perhaps indicating a changing pattern of land use in the area closest to the city.350 The pollen record does not feature a great increase in the amounts of weeds (e.g., Chenopodiaceae and Urticaceae) growing in abandoned, disturbed grounds, which seems to indicate continued active maintenance of the land. In general, landscapes do become more forested, which can be seen from the increase in arboreal pollen. However, a small peak in barley, dated roughly to the Late Antique period, can be found and, at the same time, oat and wheat almost vanish. The number of cultivated trees, chestnut, walnut and olive, clearly declines and then increases again, but this latter increase is probably later than the Late Antique period.351 The available sources point towards a continuation of many of the old types of production, but possibly more extensively. Grain cultivation and subsistence farming probably became the main aims of agricultural activities and, for some areas in Italy, Late Antiquity also meant an increase in animal husbandry, raising pigs.352 This is exempliied by one site in the research area, Collatia 392. The Late Republican and Early Imperial villa there was possibly abandoned by the mid-3rd century AD and a large horreum was built instead. In addition, the bone material from the site has revealed a herd of ca. 50 pigs.353 The Gothic Wars in the mid-6th century AD were probably the most problematic period for the countryside around Rome with Totila and his troops present in the area, but this is already beyond the time period of material discussed here. 4.8 conclusions Agriculture was the basis of the Roman economy and its importance can be seen in the selection of sites, ancient literature and archaeological inds. All indicate that good quality soils were sought. Although the surroundings of Rome were often presented in ancient literature 348 epist. 6,19. Coarelli 1986, 35–58. Large complexes, e.g., Sette Bassi (Collatia site 679a), Quintilii (Tellenae site U7) in the research area. Outside the research area, e.g., ad duas lauros and Villa delle Vignacce. 350 Gardens and vineyards on both sides of the Aurelian Wall were very common until quite late in the 19th century; see, e.g., Carandini 1985b; Gross 1990, 152–175 as well as the analysis of an 18th century map of Rome by G. Nolli at The Interactive Nolli Map Website (http://nolli.uoregon.edu/). 351 Cf. Brown 1997, 244. 352 Barnish 1987; King 1999, 190–191. 353 MacKinnon 2004, 153–159. 349 81 chaptEr 4 as an area of relaxation and otium, archaeology demonstrated a more practical picture: the most common building type features production facilities and an abundance of remains of agricultural production have been found. Recent work has also added a true landscape of production excavated from between buildings and consisting of pits and ditches. The continuity of production is also visible in the pollen diagrams. The importance of good soils can even be seen in the distribution of the rich and large sites which are situated in the immediate vicinity of the best soils. The regions of poorer quality soils become densely inhabited only during the Early Imperial period when few other locations were available. Some regional variation can be observed as most of the archaeological inds come from the plateau area. The scarcity of inds related to agricultural production on the slope areas, intensively cultivated today, could be partially explained by the good preservation of platforms and buildings with few scatters of artifacts. The southwestern corner is the problematic area as there the soils are good, but the known sites are few. The Roman Campagna was long perceived as barren and uninhabited in Roman times, as it was in early modern times due to malaria. This image has been slowly banished by the results of the archaeological surveys. Its potential as a producer of food and other perishable goods for the city of Rome has not been understood or studied. The versatility and potential of the area is only now beginning to be understood properly. The general density of habitation and the richness of many sites also make the area a formidable consumer of its own as well as imported products, a new theme that is only now being explored. Many of the sites combine production with comfortable or even luxurious living quarters. Fertile soils were the basis for the production, but other qualities were required to make the location optimal for leisure. 82 5 watEr and roMan villas 5.1 background Water is a fundamental requirement for all living things, for sustaining the life of humans, animals and plants. For Romans, water was not merely a practical requirement for life, but it was also important for recreational purposes as well as for social promotion. Consequently, water was an important factor when selecting a location for a villa.354 The agronomists recommended the presence of a perennial spring on the estate or so near that water can be conducted there. The second choice was a perennial stream which can also be conducted to the villa. A third choice was trying to ind a suitable place for a well with drinkable water and, if all else failed, closed cisterns and/or open reservoirs should be built for collecting rainwater.355 Running water also alleviated the summer heat and added to the pleasantness of the location, but placing the backside of the building towards water was recommended for avoiding smells, unhealthy miasmas or mists as well as dampness in the winter.356 Excessive water in the ground should be avoided for health reasons, i.e., marshes as well as lat areas or sites with depressions where water can gather. In addition, swamp water was not considered suitable for drinking water.357 The obvious importance of water resources has resulted in several studies of Romans and water both in Rome and other cities358 as well as in the Roman countryside.359 The problems caused by excessive water in the Roman region have also been studied, i.e., loods and the occurrence of malaria.360 The aspect of site selection has not been directly examined, although the issue has been treated indirectly while studying the water supply of Roman farms in the surroundings of Rome.361 The results show that normal rainfall in the region would have been suficient for most agricultural purposes, i.e., irrigating ields, vineyards and olive groves. Additional water from other sources would have been required for drinking and other household water as well as for the irrigation of orchards and other types of gardens. These needs were met by a number of methods: collecting rainfall in cisterns and tanks, digging wells, as well as channelling water from aqueducts deriving from springs or streams. In another study concerning southern Etruria, the distance of settlement sites from rivers was calculated and it was noted that most of the sites were located within 500 meters of a minor 354 355 356 357 358 359 360 361 Cato agr. 1,3; Colum. 1,3,3 (also more generally 1,3,4–5.); Varro rust. 1,11,2. Colum. 1,5,1–2; Varro rust. 1,11,2. Colum. 1,5,4; Plin. nat. 18,33; Varro rust. 1,12,1. Colum. 1,5,3; 1,5,6; Pallad. 1,7,4; Plin. nat. 18,33; Varro rust. 1,6,6; 1,12,2–3. E.g., Lanciani 1880; Ashby 1935; Bruun 1991; Taylor 2000; Kleijn 2001; Koloski-Ostrow 2001; Jansen 2002. Evans 1993; Thomas and Wilson 1994; Wilson 1999; 2000; Bannon 2001. Floods: Aldrete 2007. Malaria: Celli 1925; 1927; 1933; Sallares 2002. Thomas and Wilson 1994; see also Dell’Era 2002 for the area north of Rome. chaptEr 5 or major river. Their greatest signiicance was deemed to be the potential for transportation and not water supply.362 Another aspect studied is the use of the public aqueducts crossing the Roman Campagna.363 These were mainly designed to provide water for the city of Rome, but they were also used locally and spurs were built for supplying water to villas both legally and illegally. The importance of aqueduct water along their lines is great, both for practical and representational requirements. The legal aspects of water servitudes between landowners have also been studied in relation to archaeological indings in the Roman Campagna.364 The results suggest that the spread of commercial agriculture might have caused problems in the traditional system of neighborly sharing of water rights. This contributed to the creation of bona ides as a means of evaluating the conduct of the landowners in various disputes as well as a means of resolving some of the problematic situations. The purpose of this chapter is to explore the hydrological resources in relation to site selection and two issues in particular will be explored: sources of water supply and the healthfulness of the site. The structure of this chapter follows mostly that of the previous ones. First, the natural resources available are charted, and then the literary and archaeological evidence for supplies and uses of water will be presented. In the end, all this evidence is compared in order to see how water-related issues might have inluenced the selection of a location for a villa. 5.2 watEr rEsourcEs in thE roMan caMpagna The general climatic factors of temperature and precipitation determine how much water is available both as surface water and groundwater. The region of Rome can be described climatically as Mediterranean with hot and dry summers and humid and mild winters.365 The mean annual temperature in the region is 15º C with a maximum of 20º C and minimum of 11º C (Table 5.1). The warmest period occurs in July–August and the coldest period in January–February. Local variations occur depending mostly on elevation and exposure, as the climate tends to be more severe in mountainous regions, but the greatest changes happen above the highest elevations for the research area, e.g., in Rocca di Papa located at 690 m a.s.l. The data from the Tivoli weather station at 238 m a.s.l does not differ signiicantly from the lower regions. The mean annual rainfall for the whole area is 950 mm with 80 days with rain/year (Table 5.1).366 The greatest amount of rain normally falls October–December and the least in June–August. The distribution of annual rainfall varies in the Roman region: the eastern plateau area receives the least amount of rain, between 700–800 mm, whereas in the area north of Rome as well as on the lower slopes, the rainfall rises to 900–1,000 mm/year. The historical records show regularly occurring longer periods of dry months with less than 30 mm of rain between May and September. These last between 3–5 months. The evapotranspiration, i.e., the loss of water from the soil both by evaporation and by transpiration from plants, is greatest during the summer months. The uneven distribution of rain during the year is evident and the possibility of drought during the summer is great. 362 Goodchild 2007, 160–166. Cf. though Perkins 1999, who did not examine water resources due to the dificulty of reconstructing the ancient situation further north in Etruria. 363 Evans 1993; Wilson 1999. 364 Bannon 2001. 365 Ventriglia 1990a, 25–34; 1990b, 27–34. 366 Ventriglia 1990a, 34–55; 1990b, 35–55; Thomas and Wilson 1994, 140; Mangianti and Leone 2008. 84 watEr Table 5.1 Monthly and annual temperatures and rainfall in Tivoli and Rome 1921–1965. Based on Ventriglia 1990a, Tables 3.8, 3.11 (Rome, 1921–1965); Ventriglia 1990b, Tables 3.2, 3.4 (Tivoli, 1935–1965). Month January February March April May June July August September October November December Annual Tivoli Temperature (°C) 6.5 7.5 10.4 13.6 17.7 22.1 24.8 24.8 21.2 14.2 11.7 8 mean 15.5 Rome/plateau Rainfall (mm) 68 70 69 73 72 46 24 27 70 101 98 95 total 817.8 Days with rain 8.0 8.3 8.4 9.3 8.3 6.0 2.8 3.2 5.9 8.6 10.0 10.0 total 88.8 Temperature (°C) 7.9 8.9 11.3 14.9 18.9 23.4 26.2 25.8 22.7 17.6 12.8 9.0 mean 16.1 Rainfall (mm) 95 93 82 70 64 36 20 24 84 122 133 121 total 941.3 Days with rain 8.6 8.4 8.0 7.3 6.7 3.7 1.9 2.1 5.2 7.9 10.0 9.9 total 79.6 The climate has changed over time, as was already discussed in Chapter 3.3. It has been suggested that the weather zones over Europe and northern Africa move cyclically northwards and southwards causing repeated changes in the climate.367 The summary of suggested weather types for the period studied here (Table 5.2), shows that for most of the period studied here, the modern temperature and rainfall data are valid indicators of the weather conditions. Most of the earlier Republican period would have been similar to the Little Ice Age (from the 15th to mid–19th century) with an oceanic precipitation regime, i.e., with more rain in the summer, and generally lower temperatures with less variation between maxima and minima both monthly and annually. The later period, the Imperial era, featured less rain and the temperatures would have been possibly higher than the current ones. The surface water system in the area consists mainly of small streams draining into the two main river channels: to the Tiber in the northwest and southwest and to the Aniene in rest of the area.368 Thirty-six watersheds have been recognized and 23 of these drain to the Aniene and the rest to the Tiber, apart from the small lake basins of Lago di Castiglione and Lago diAlbano, which have no natural drainage channels. (Fig. 5.1.) The Aniene drains into the Tiber and the Tiber eventually into the Mediterranean Sea. The main channels are perennial, but the smaller branches can dry up seasonally. Most alluvial deposits featured on the geological map are located in the lower reaches of the channels, particularly towards the Aniene (Plate II.1). The drainage pattern follows the natural slopes. In the northern part of the Table 5.2 Climate changes in the Mediterranean area from the Archaic through the early Medieval period. Based on Ortolani and Pagliuca 1994; 1995; 1996; 2003; Caiazza et al. 1999; Molinaro et al. 2001, Fig. 10. 367 368 Period Climate and soil formation until 500 BC Temperature and rainfall similar to today; erosion, formation of soils in the subsoil layers 500–350/300 BC Mean temperature lower than today; oceanic precipitation; moderate formation of soils 350/300 BC – AD 150 Temperature and rainfall similar to today; rapid soil formation AD 150–350 Mean temperature higher than today, rainfall lower; erosion, slow soil formation AD 350–500 Temperature and rainfall similar to today; erosion, formation of soils in the subsoil layers AD 500–750 Mean temperature lower than today; oceanic precipitation; moderate formation of soils Ortolani and Pagliuca 1994; 1996; 2003; Caiazza et al. 1999; Molinaro et al. 2001, 34–37; Bradley et al. 2003. Ventriglia 1990a, 73–156; 1990b, 73–152. 85 chaptEr 5 area, the main drainage direction is from northeast to southwest and in the eastern area, from southeast to northwest. Some of the small channels in the northeastern part drain almost in a north–south direction. The streams draining into the Tiber in the southwestern corner tend to run from east–southeast to west–northwest. The stream banks tend to be steep in most of the southwestern area as well as in the northwest towards the Tiber. The central part features shallower valleys, apart from the area northwest of Lago di Castiglione. The streams in the central northern part also run in deep and steep channels. The amount of possible surface runoff Fig. 5.1 Rivers, streams and lakes in the and the speed of groundwater aquifer research area. River basins outlined with renewal depend on the permeability of the black thick line, watershed with white lines. underlying geological deposits as well as the drainage capacities of the soils.369 The marine clay in the central northern part of the research area and the limestone in the northeast are non-permeable, although the latter can contain issures that allow water to pass through. Most of the volcanic deposits are permeable either by great porosity or by issures, which is a common characteristic of lava. Poorly permeable deposits can be found evenly all over the area and the same applies to normal/good permeability. The most deposits with the best permeability can be found in the area north of the Aniene. (Table 3.1.) The soils in the area drain moderately or well, but coarser soils which drain somewhat excessively are also known. These can be found particularly in the northwestern and southwestern zones. The iner soils of the depressions in the ridge crests as well as in the valley bottoms, e.g., in the Aniene valley drain moderately. (Table 4.1.) Groundwater resources in the area are good and they are mainly renewed by vertical iniltration of rainfall. The volcanic deposits on top of the Pleistocene clays are connected to a continuous base aquifer. The variation of permeability between volcanic and/or other deposits overlying each other can result in a perched aquifer, which is commonly found in the volcanic area. The depth of the modern groundwater level varies according to the depth of the deposits on top of it. The areas closest to the major water basins, such as the Tiber, the Aniene and the crater lakes, feature a high water table. The plateau area has today a groundwater table below 50 m from the current ground level nearest to the major rivers and between 50 and 100 m in the zone closer to the higher slopes. At Tivoli, the water level is at 100 m and in the Frascati– Grottaferrata region at 200 m below ground level. (Plate VI.1.) It is likely that the Roman and later drainage works have caused the groundwater level to sink from the pre-Roman height.370 The aquifer was utilized by digging wells and on hydrogeological maps, the “Roman wells” (It. pozzi romani) mark the features dug manually, 80–100 cm in diameter (Plate VI.1).371 The dates of the well shafts are uncertain, but some of them can be ancient. Most of the 268 wells are located in the northern and southwestern parts of the research area. The majority of the 369 Ventriglia 1990a, 183–185; 1990b, 183–188; Arnoldus-Huyzendveld 2003. Ventriglia 1990a, 86–156, 183–187, carta idrogeologica; 1990b, 93–104, 183–193, carta idrogeologica; Thomas and Wilson 1994, 143–146; Boni et al. 1995. 371 The wells recorded on the topographical maps were not included, as no information of their type, depth, etc., was available. 370 86 watEr wells produce water less than 10 liters of water per second and most of these are located in the northwestern area. Approximately two thirds of the wells are less than 25 m deep, probably corresponding to the 25 to 50 m groundwater level in most of the area where they have been found. The majority of the shallow wells are found in the northern zone and deeper ones can be found in all parts of the research area.372 The aquifer becomes naturally available in valleys, where the water table reaches the ground surface. The most important features are small streams, already discussed above, and springs. Another typical phenomenon for this region is stream seep, which means that groundwater seeps through to the surface, but does not necessarily form clear concentrations, such as springs or ponds. Stream seep has been studied in the area south of the Aniene. The springs recorded in the topographical and hydrogeological maps as well as the spring lines are plotted in Plate VI.2.373 A total of 129 springs are known in the research area and they are evenly spread excepting the central part which is almost devoid of springs. Most of the springs (109) discharge less than 10 liters of water per second, being thus small/medium in size. There are relatively few large springs: 11 springs have a low of over 10 l/s and 9 a low of over 100 l/s. The large springs are found in the northeast in the vicinity of the travertine deposits, in the central part just south of the Aniene and in the southeastern part, in the zone north of Lago di Albano. The largest sweet water springs are those of Salone (900 l/s, 4 springs) just south of the Aniene and Acquoria (2800 l/s) below Tivoli. The largest spring is actually the Laghi Colonnelle–Regina, i.e., ancient Aquae Albulae, discharging sulphurous water at a volume of 3500 l/s. In general, it can be assumed that the opportunities for acquiring water were good in the Roman Campagna during the Roman period. Rainfall was probably suficient – or sometimes even excessive – for most of the year. The volcanic geology of the region created also further possibilities for water supply, featuring stream seep and a perched aquifer. The qualities of the soils and geology also offer an opportunity to gather relatively clean water iltering through them. The Roman supply and use of water is recorded in an extensive infrastructure related to public and private water management in written as well as archaeological sources. The next task is to look at these sources and to see what can be deduced about the relationship of villas and water based on that evidence. 5.3 writtEn sourcEs on thE watEr rEsourcEs and watEr usE The literary and epigraphic references related to water in the research area include information on natural bodies of water as well as on man-made structures such as aqueducts. Many of the issues have been dealt with in detail in previous research,374 so the following discussion repeats some of the same issues, but the focus will be on the key subjects of healthfulness, 372 Ventriglia 1990a, 199–262; 1990b, 207–240 and the list of springs and wells (= Parte IV) in Ventriglia 1989; 1990a; 1990b. 373 Ventriglia 1990a, 199–206, carta idrogeologica, carta acque minerali; 1990b, 207–223, carta idrogeologica, carta acque minerali and the list of springs and wells (= Parte IV) in Ventriglia 1989; 1990a; 1990b; Boni et al. 1995; Heiken et al. 2005. Only one intermittent spring on the topographical maps was included as others are the same as on the hydrogeological maps. The spring data presented by Thomas and Wilson 1994 for the Collatia area is derived from Boni et al. 1987, which I have been unable to consult. Thomas and Wilson 1994, Fig. 8 shows a great number of springs in the central area, but few modern fountains have been marked on the IGM topographical map. Ventriglia 1990b lists hardly any springs there, and the same applies to the new hydrogeological map of Rome (Capelli et al. 2008). Consequently, these were not included in this study, as the data available was insuficiently detailed. 374 See above for references, particularly Thomas and Wilson 1994. 87 chaptEr 5 water supply and use of water. Salubritas of the location chosen for a villa is emphasized in the advice given by the agronomical writers. What it means exactly is more dificult to understand, but in relation to water, this would seem to mean avoiding stagnant water and marshy areas as well as placing the buildings with their backs towards lowing water.375 Another important aspect is a temperate climate without excess heat or cold.376 Choosing an elevated location can be partially helpful, as in such locations seasonal weather differences are more moderate. Favorable winds are also important and their type depends on the region and in this regard, the problem of a high site is windiness: every breeze in every season affects the inhabitants.377 The research area was considered mostly healthy, particularly the elevated zones, such as Tibur and Tusculum.378 Tibur’s healthfulness can also be derived indirectly from Martial,379 who describes Curiatius’s sudden death in the town as surprising considering its reputation. In addition, the town’s cool and temperate climate is mentioned often.380 The salubriousness of the plateau area is discussed on some occasions. Strabo381 mentions the plains towards the sea as being unhealthy, but that the other plateau zones were pleasant to dwell in and accordingly dotted with wealthy villas. The only region regarded as unhealthy is the ager Pupiniensis which is located on the plateau below Tusculum, corresponding more or less with the central part of the research area. Marcus Atilius Regulus (cos. 267 BC), had a small farm in the ager Pupiniensis and his advice, based on his own experience was to avoid even fertile land if it was in an unwholesome area. Pupinia was very bad as it was considered both lean and pestilential.382 The central Italian area around the city of Rome is regarded as a well-watered region in literary sources.383 Despite this, references to water in the Roman area are relatively scarce. Some passages describe rains and these concern both excessive rain as well as lack of it.384 It is obviously impossible to draw more general conclusions on the climatic conditions from these references, but indirect evidence for rains can be found in the frequent passages concerning the looding of Tiber, and on one occasion, in AD 105, of the river Aniene.385 Pliny the Younger’s lively description makes it clear that the reason for the lood was excessive rainfall. The period between the 2nd century BC and the 2nd century AD was marked by a great intensity of loods and this has been taken as evidence for a positive hydrological balance, or, in other words, a wet period, in the central Italian area.386 Rivers and lakes were major features of the landscape and were connected to good aquatic resources; such is the case of Tibur and the Aniene in Early Imperial poetry.387 The river furnished water for irrigation, as well as pleasant shade with trees growing on its banks.388 375 See above for references. Colum. 1,4,10; Varro rust. 1,4,4. 377 Varro rust. 1,12,1 on wind varying by region; Colum. 1,4,10 on the problems concerning windiness. 378 Tusculum: Cicero rep. 1,1; Strab. 5,3,12. 379 4,60. 380 Fronto 2,6,3; Mart. 1,12,1–2; 4,57; 4,64,31–33; 5,71; Stat. silv. 1,3,1–8. 381 5,3,12. 382 Colum. 1,4,2–3. Plin. nat. 18,27 gives the same advice, but does not mention the ager Pupiniensis. Other references to Atilius Regulus’s farm do not mention the unhealthfulness of the area and also not even the area; Dio Cass. 11 frg. 43,20; Front. strat. 4,3,3; Val. Max. 4,4,6; Varro rust. 1,9,5. Cf. Sallares 2002, 236, where the rest of the Roman region is labelled unhealthy based on Cicero and Livy praising the healthfulness of Rome’s hills. 383 Colum. 11,2,61; Strab. 5,3,11–12. 384 Excessive rains: Colum 11,2,61 and Plin. epist. 8,17. Draught in 428 BC: Liv. 4,30,7–8. 385 Tiber: Aldrete 2007. Aniene: Plin. epist. 8,17. 386 Giraudi 2005. 387 E.g., Hor. carm. 1,7,12–14; 3,29,6–8; 4,2,29–32; 4,3,10–11; Ov. am. 3,6,45–46; fast. 4,71–72. 388 Irrigation: Hor. carm. 1,7,12–14; Prop. 4,7,81–86. Shade: Stat. silv. 1,3. 376 88 watEr Table 5.3 Public aqueducts running through the research area. Aqueduct Built Source References Aqua Appia 312 BC Between seventh and eighth milestones of the Via Prenestina? Front. aq. 1,5 Anio Vetus 272 BC Aniene Valley Front. aq. 1,6 Aqua Marcia 144 BC Aniene Valley Front. aq. 1,7; 1,12; Plin. nat. 31,41; Strab. 5,3,13 Aqua Tepula 125 BC Aqua Iulia 33 BC 2 miles south of the tenth milestone of the Via Latina, Sorgente Front. aq. 1,8; Valenti 2003, 292 site 611 Preziosa? 2 miles south of the twelfth milestone of the Via Latina; Ponte Front aq. 1,9; LTURS s.v. Iulia Aqua (Mari) degli Squarciarelli? Front. aq. 1,10; 2,70; Plin. nat. 31,42; Quilici 1974a, 123–38 Salone springs? site 45. Front. aq. 1,5; 1,12; LTURS s.v. Augusta Aqua (Mari); s.v. Appia North of the sixth milestone of the Via Prenestina Aqua (Mari) Aqua Virgo Aqua Appia Augusta 19 BC AD 11–14? Anio Novus AD 38 Aniene Valley Front. aq. 1,15 Aqua Claudia AD 38 Aniene Valley Front. aq. 1,14 spring(s) on the eastern edge of the Pantano Borghese LTURS s.v. Alexandrina Aqua (Mari) Aqua Alexandrina early 3rd century AD In addition, water reduces heat and the waters of the Aniene were often described as cold or cool.389 Apart from major rivers, a few other streams are mentioned by name: Almo,390 lumen Tutia and rivus Herculaneus are three streams located in or near the research area. The lumen Tutia is mentioned in connection to Hannibal’s military campaign in 211 BC as one camp site in the Roman region.391 The third stream, rivus Herculaneus, is mentioned in connection to the springs of Aqua Virgo.392 References to the crater lakes in the area are equally rare. Lago di Albano is mentioned in connection to unexpected and unexplainable loods occurring in the Early Republican period.393 Lacus Regillus is mentioned in connection to the war between Rome and the Latin peoples in the Early Republic, but it is not referred to in later times.394 Of the many springs in the area, the ones known as Aquae Albulae (It. Acque Albule) are mentioned most often. Their sulphurous waters were considered good for drinking and bathing as well as having healing qualities.395 These springs also had a religious signiicance for their healing qualities with regard to both humans and animals.396 Other springs are mentioned in connection to the origins of the Roman aqueducts. The aqueducts are the water feature most often treated in ancient literature. The longest aqueducts – Anio Vetus, Aqua Marcia, Anio Novus and Aqua Claudia – originated from springs located outside the research area, east of Tibur running along the Aniene and the Via Valeria/Tiburtina. The remaining aqueducts began in the area between the Aniene and the Via Appia and all the aqueducts ran through the research area in the same zone (Table 5.3; Fig. 5.2). Their main purpose was to service the city of Rome, but a fairly large proportion of the water delivery, almost 30%397 was directed to the countryside outside the city, probably beyond the seventh milestone (Table 5.4).398 In addition, a branch from the Anio Vetus was 389 Lucan. 1,581–582; Sil. 10,363–364; 12,539–540; Stat. silv. 4,4,17; Verg. Aen. 7,681–685; cf. Colum. 1,5,4. Mart. 3,47; LTURS s.v. Almo (Pisani Sartorio). Probably Fosso dello Statuario between the Via Appia and the Via Latina. 391 Liv. 26,10–11. LTURS s.v. Tutia luvius (Mari). Probably Fosso di Settebagni located on the northwestern edge of the research area (Quilici and Quilici Gigli 1980, 207–210). 392 Plin. nat. 31,42. Probably Fosso di Ponte di Nona (Quilici 1974a, 120 under site 43). 393 E.g., Cic. div. 2,69; Diod. Sic. 7,5,11; Liv. 5,15,2; Val. Max. 1,6,3. 394 E.g., Dion. Hal. Ant. Rom. 5,50–6,13; 6,4–13; Liv. 2,18–20. The site of the lake has been widely discussed, and two locations have been suggested: the small crater lakes of Pantano Secco and Prataporci situated north of Tusculum. Pantano Secco is considered the most likely one (Quilici 1974a, 879–881 site 804). 395 E.g., Paus. 4,35,10; Plin. nat. 31,10; Strab. 5,3,11; cf. CIL XIV 3911. 396 Frizell 2004; Edlund-Berry 2006; cf. inscriptions CIL XIV 3908–3912, I.It. IV,1 591–592, 606. 397 Front. aq. 78–86; cf. Wilson 2000, 315. 398 Front. aq. 1,19; 2,70; 2,72; Bruun 1991, 148. 390 89 chaptEr 5 Fig. 5.2 Public aqueducts in the Roman region. destined for Tibur and the archaeologically known loop-line of the Anio Vetus has been suggested as having served the same purpose.399 Of the total amount reported by Frontinus, 60% of the water went to private parties and 40% for imperial use. The only aqueduct coming to the city from the west in Frontinus’s time, the Aqua Alsietina, delivered all of its water to the countryside. The deliveries of the individual aqueducts to the countryside east of Rome vary from less than 1% of the Aqua Appia to 50% of the Aqua Virgo. Most of the countryside deliveries went to the emperor (70%) with the Aqua Marcia and the Anio Novus serving only imperial needs.400 The users of the water delivered by the Aqua Appia and the Aqua Virgo are not speciied. This is unfortunate in the case of the Aqua Virgo, as so large a proportion of its water was destined for private use. However, considering the total amount of water destined for private use, it seems likely that most of the water from the Aqua Virgo went to private users and not to the emperor. The private deliveries probably went to legal owners of water rights, who had paid for the privilege granted by the emperor.401 The aqueducts were also tapped without permission, sometimes even by dishonest aquarii for inancial gain.402 Illegal taps were punished with ines or with the threat of the coniscation of the lands where the water was used.403 The aqueduct water was used by private parties for both practical purposes as well as for pleasure.404 However, little is said of how the water was used in the countryside. The irrigation 399 Front. aq. 1,6; 2,66; cf. Evans 1993. Plin. nat. 31,42 contradicts this concerning the Aqua Marcia as he states that its deliveries did not reach Rome due to private use outside the city (in villas ac suburbana). 401 Front. aq. 2,88; 2,103–106; 2,118; cf. Bruun 1991, 66–71. The proits were partially used for covering the maintenance costs of the aqueducts (Front. aq. 118). 402 Front. aq. 1,7; 1,9; 2,69; 2,72. 403 Front. aq. 2,97; 2,129. 404 Front. aq. 1,11; 1,13; 1,23; 2,75; 2,95; Plin. nat. 31,42. 400 90 watEr Table 5.4 Consumption of aqueduct water in quinaria according to Frontinus (aq. 76–88). No attempt has been made to correct the data which added up do not match the total reported by Frontinus. * = The amount of water from the Aqua Claudia and the Anio Novus arriving in Rome is presented as a total, hence the countryside igures are also totals. ** = Aqua Alsietina is on the right bank of the Tiber. Rome Countryside Total Town/ Country Private Emperor Not speciied Private/ Emperor 9,955 4,063 14,018 70:30 2,345 1,718 ? 60:40 699 5 704 99:1 ? ? 5 ? Anio Vetus 1,509 573 2,082 70:30 169 404 ? 30:70 Marcia Aqueduct All Appia 1,472 263 1,735 85:15 ? 263 ? 0:100 Iulia 548 114 662 85:15 56 58 ? 50:50 Virgo 2,304 2,000 4,304 50:50 ? ? 2,000 ? Claudia 3,498 656 4,154 70:30* 439 217 ? 70:30 ? Anio Novus Total Alsietina** ? 728 10,030 4,339 ? 392 392 ? ? 728 ? 0:100 ? 664 1,670 2,005 30:70 0:100 138 254 ? 35:65 of gardens and ields is often mentioned, including the pleasure gardens or horti surrounding the center of the city. Cato mentions irrigated gardens as one of the most lucrative agricultural activities, second only to vineyards.405 In the same passage, he also mentions another type of cultivation requiring plenty of water: osier-beds. However, these were probably mostly grown in places that were naturally rich in moisture. Irrigated meadows were also considered important for a productive and self-suficient estate.406 A water source could be easily exploited for purposes of agriculture and inancial gain. The uses of water considered luxurious are not speciied by Frontinus, but even the irrigation of gardens could be regarded as such, if it prevented regular supplies to the city.407 In addition to the great public aqueducts, there were also smaller ones, provided by small towns for their inhabitants. Part of the water supplies of Tibur were provided by the Anio Vetus as well as by one local aqueduct and channels taking in water directly from the Aniene, which are known from archaeological evidence.408 There were also three Tiburtine magistrates involved in water management.409 Frontinus does not mention arrangements for deliveries from public aqueducts to the region of Tusculum. Instead, both Frontinus and Cicero mention a local aqueduct called the Aqua Crabra. It lowed past the intake of the Aqua Iulia and was left untouched by Agrippa for the beneit of landowners in Tusculum. Despite this decision, the Roman aquarii had tapped the Aqua Crabra supplementing the low of the Aqua Iulia mostly for their own beneit by arranging illegal taps from the main line. Frontinus maintains that he shut off the Aqua Crabra taps and restored its waters for Tusculum’s use at the emperor’s command and probably to great surprise of the landowners.410 The Aqua Crabra provided abundant water for Cicero’s Tusculanum in the 1st century BC. There seems to be even too much water for Cicero’s needs and he questions Tiro as to what to do with it all.411 A lead water pipe bearing the name of a village, ad Decimum, has been found near the tenth 405 Front. aq. 2,75–76; 2,92; 2,97; Cato agr. 1,7. Cf. Wilson 2000, 315. Cato agr. 1,9. 407 Front. aq. 2,75. 408 Front. aq. 2,66; Evans 1993; Tibur I sites 74, 82, 111, 147, 149b, 152. 409 Praefectus rivis supernatis (CIL XIV 3682 = I.It. IV,1 199; 2nd century AD); tribunus aquarum (CIL XIV 3674 = I.It. IV,1 197); curator aquae Tiburtinae (CIL X 6427; from Circeii). 410 Front. aq. 1,9. 411 Cic. fam. 16,18,3, leg. agr. 3,9. A third reference to water rights and Cicero’s Tusculanum is Cic. Balb. 45, but the Aqua Crabra is not mentioned there explicitly. As the passage is about what legal references to use in questions of water rights, it would seem likely that the Aqua Crabra matter dealt with in leg. agr. might also be under discussion here. 406 91 chaptEr 5 milestone of the Via Latina, but it is not known where the water came from.412 Two other local aqueducts situated outside the research area are known from inscriptions. One supplied water to Gabii in the Hadrianic period.413 A series of cippi for a line called the Aqua Augusta in the area east of Tusculum has been found, but little else is known of this line.414 A burial inscription found near modern Marino also records the existence of an Aqua Albana which could be a third local aqueduct.415 The private channelling of water was another means of water supply. Relationships between landowners with regard to water rights and conducting water through estates have been recorded in legal texts and these have been treated in research literature at length.416 This practice is probably recorded in a fragmentary inscription from Tibur giving indication of times of day and quantities of deliveries of aqueduct water to estates.417 The last issue to be discussed is speciic villas, their relationships to bodies of water and water resources. Relatively little is said of how villas were located in relation to natural bodies of water. The practical reasons for locating a villa near water are outlined by the Roman agronomists, but the proximity of water was also appreciated for purposes of pleasure as can be seen from a passage by Seneca418 condemning the practice of building luxurious villas and resorts on all possible shores. The most common location mentioned in relation to water is the sea, the villa maritima, but the maritime villa will not be discussed here at length.419 Pliny the Younger’s Laurentinum can be mentioned as a prime example of such a villa near Rome.420 The rims and shores of the crater lakes feature archaeologically attested villas, but these are not known from literary sources.421 Emperor Domitian’s Albanum on Lago di Albano is called arx albana, but the lake is rarely mentioned.422 Several brief passages record great numbers of villas on the banks of the Tiber, but these are located upriver, outside the research area.423 The banks of the Aniene were also popular dwelling places and one of the most elaborate descriptions of a villa, that of Statius on the villa of Manilius Vopiscus in Tibur, concerns a building complex stretching over the river on both of its banks.424 The aquatic resources of the villa also included a connection to the Aqua Marcia, which contradicts Frontinus’s information that no water was delivered from the Aqua Marcia to private parties. Statius describes the pipe as a daring arrangement and this could mean that is was exceptional. It has also been suggested that the way the pipe crossed the river – probably under the river and not over it on a bridge or other structure as was more Bovillae site 121; CIL XIV 4229 = CIL XV 7811. CIL XIV 2797. The same springs that supplied the Aqua Alexandrina on the eastern edge of the Pantano Borghese basin have been suggested as the sources of this aqueduct (LTURS s.v. Alexandrina Aqua (Mari). 414 CIL XIV 2567a–e = ILS 5748; Wilson 2000, 317; LTURS s.v. Augusta Aqua (Mari). The name “Aqua Augusta” is used for several aqueduct lines in the Roman region: two were located in the Aniene valley, one is the branch of the Aqua Appia mentioned above and the Alban Hills line is the fourth. The Aqua Virgo and the Aqua Alsietina were also often called Augusta and it seems to have been the unoficial name for the Anio Novus as well. 415 CIL XIV 2466. The Aqua Albana could also possibly refer to Lago di Albano, but as the latter is quite consistently called lacus, it would seem more likely that a river, stream or an aqueduct is indicated. 416 Thomas and Wilson 1994, 147–149; Bannon 2001. 417 CIL XIV 3676 = I.It. IV,1 239add.; Lanciani 1880, 324–326; Mari 1991, 43; Bianco 2007, 152–153. A similar inscription, CIL VI 1261, depicting water lines, distribution times, quantities and landowners has also been connected to either Tibur or Tusculum (Bianco 2007, 153–157), but its provenance is unclear. 418 epist. 89,21. 419 For discussions of the villa maritima, see Lafon 2001a; Marzano 2007, 13–81. 420 epist. 2,17. 421 Plin. epist. 9,7 can be mentioned as a comparison: he had two buildings on Lago di Como: one on the cliffs above the lake and the other right by the lakeshore. 422 One exception is Plin. paneg. 82,1; cf. Darwall-Smith 1994; Mayer 2005, 94–96. 423 Cic. Catilin. 3,5; Mil. 64; S.Rosc. 20; Plin. nat. 3,54. 424 silv. 1,3 passim. 412 413 92 watEr common – was the daring part.425 Living by a river was not always pleasant as serious damages to habitations on river banks during loods of the Tiber and the Aniene are also mentioned. In the late 5th century BC, a lood of the Tiber destroyed Veientine farms and in 192 BC, another lood destroyed the countryside along the river, including villas. In the lood of AD 105, the waters of the Aniene broke out of the river’s usual channel due to blockages caused by landslides.426 Even people living on higher ground were in trouble because of the excessive rainfall and the looding river. The major river banks were not shunned despite the danger of loods or other water-related problems, perhaps for reasons of economics, as a navigable river and irrigated meadows were considered precious assets for a villa.427 Lucius Licinius Lucullus’s (cos. 74 BC) estate at Tusculum is mentioned twice in connection to public aqueducts. Both the Aqua Tepula and the Aqua Iulia had their origin on Lucullus’s property.428 It is not known whether Lucullus’s property received water from the public aqueducts, but the situation probably illustrates a quite common relationship between the public aqueducts and private villas. The aqueducts started and ran mostly on private property and this made the maintenance of the lines sometimes dificult. Although some of the lands of the lines were bought by the state, the parts closest to the city, i.e., beyond the seventh milestones, were, according to Frontinus, almost all on private property. The proprietors were in theory required to keep ifteen feet around springs and ive feet of ground on both sides of the aqueduct clear of vegetation and buildings, but this rule was often neglected. Damages and causing problems for the maintenance of the aqueducts could lead to ines and paying the cost of repairs. This situation can also have encouraged illegal tapping of the aqueducts.429 The public aqueducts were thus not necessarily only a beneit for the landowner, but also a possible problem. References to water installations in villas are also relatively few, which is surprising considering how common they are in archaeologically known buildings. Pliny’s villa at Laurentinum did not have running water, but instead, wells and springs provided drinking water, as groundwater was close to the surface near the sea. Cicero briely mentions how the artiicial streams in villas pompously called Niles and Euripi pale when compared to real rivers such as the Fibrenus near Arpinum.430 In connection to the market garden he was renting, Cicero also mentions an emissarium that was built in the garden to improve the plot and to make it more proitable. The character of this water channel as well as its relationship to the water supply from the Aqua Crabra remains unknown.431 The most common water-related feature mentioned in connection to villas is the bath. From the descriptions of how various Romans spent their days in their villas, it is clear that bathing was an important event in the daily schedule.432 The Junii Bruti had a villa in Tibur in the 2nd century BC and this featured probably the earliest known bath in the Roman region. Along with another very early bath, that of Scipio Africanus’s Liternum, this shows that bathing was already part of villa life very early.433 Later on, Cicero’s Tusculanum had a bath, as he writes to Terentia to prepare everything for him and his friends’ visit in 47 BC including 425 Stat. silv. 1,3,67. See Evans 1993, 452 for the alternative explanation of the water pipe. Plin. epist. 8,17. 427 Plin. epist. 8,8,6 mentions also how the banks of the river Clitumnus in Umbria were built up with villas wherever the area was beautiful. 428 Front. aq. 1,8 and 1,10 respectively. 429 Front. aq. 2,124–2,129. 430 Cic. leg. 2,1,2. 431 Cic. fam. 16,18,2. Mangiatordi 2003, 221 nota 37 regards the emissarium as an aqueduct. 432 E.g., Cic. de orat. 1,27; Plin. epist. 3,1; 3,5; 9,36; 9,40. 433 Junii Bruti: Cic. de orat. 2,55. Scipio Africanus: Sen. epist. 86. 426 93 chaptEr 5 taking care of the basin in the bath. The villa of Crassus where the discussions of Cicero’s de oratore took place was also furnished with a bath which demonstrated the urban pleasures of his estate. Manilius Vopiscus’s villa at Tibur had a bath located slightly higher on the banks of the Aniene.434 A bath of one’s own was not always necessary, as villages and towns nearby probably featured public baths. Pliny mentions in connection to his Laurentinum, how the vicus nearby conveniently had three baths that could be used if the visit to his own villa happened suddenly or if it was very short.435 Several public baths are known in the Roman countryside. The ones established at the sulphurous springs of Aquae Albulae were well known. Gabii featured baths for hire in the late 1st – early 2nd century AD.436 Some individuals also donated baths to their communities, as, e.g., those given by a soldier, Marcus Helvius Rufus, east of Tibur, or by Aurelia Faustiniana near Ficulea. An anonymous donor left a bath next to his house at Tibur for public use for ten months of the year.437 Baths still feature in Late Antiquity as one establishment at Tibur was re-dedicated in AD 376–377 by Furius Maecius Gracchus (praef. urb. AD 376). Gaius C(a)eionius Ruius Volusianus (signo Lampadius; praef. urb. AD 365– 366) built a public bath in the area in the early 4th century AD.438 No public baths are known from Tusculum or the southern parts of the research area. 5.4 archaEological EvidEncE for watEr rEsourcEs and watEr usE The archaeological evidence concerning water resources in the research area consists of various structures interpreted as related to water. Approximately 670 of the 1,940 settlement sites (35%) have provided some structures or inds related to water and, in addition, approximately 300 other sites featuring water-related installations are known, sometimes more than one installation per site. (Appendix IV; Table 5.5.) In the following, the various structures are introduced with basic statistics and distribution maps in the framework of the main functional groups of supply, movement, storage and use. The archaeological evidence for water supply from natural sources, rainfall and groundwater, is not very common in the research area. Nevertheless, some signs of collecting rainwater, of development of springs, as well as of digging wells and underground channels for extracting drinking/household water have been found. Most of the evidence comes from settlement sites; 179 as opposed to 59 non-settlement sites. The settlement features include atria/impluvia, open cisterns, springs and wells, of which the irst two are not present at nonsettlement sites.439 Rainfall has traditionally been an important source of household water in the Mediterranean. In ancient times, rainwater was collected inside a house as roof runoff by a compluvium – impluvium roof arrangement in the atrium.440 The runoff from the roofs of colonnades in peristyle gardens or other courtyards could also have been collected. Rainwater was probably collected mostly for domestic consumption inside the buildings, as perhaps 434 435 436 437 438 439 440 Cicero: Cic. de orat. 1,27. Crassus: Cic. de orat. 1,27. Vopiscus: Stat. silv. 1,43–44. Plin. epist. 2,17,26. Aquae Albulae: Strab. 5,3,11; possibly also Suet. Aug. 82,2. Gabii: Iuv. 7,3–5. Rufus: CIL XIV 3472. Faustiniana: Ficulea site 192 or 228, CIL XIV 4015. Anonymous: Scaev. dig. 32,35,3. Maecius: CIL XIV 3594 = I. It. IV,1 151add. Lampadius: AE 1986, 109; Moretti 1986. Mansio and statio sites could possibly include architecture similar to houses. Vitr. 6,3. 94 watEr Table 5.5 Summary of data presented in Appendix IV with comparison to excavated sites. Uncertain cases in brackets. * = only private aqueducts. Feature Class 1 Class 2 Class 3 Class 4 Other Total All sites 278 239 146 5 260 928 Excavated 90 More than one feature 201 79 22 0 63 365 61 Spring 9 (1) 6 (1) 2 3 43 (4) 63 (6) ? Well 44 (6) 24 (4) 23 (4) 0 19 (2) 110 (16) ? Impluvium 29 (12) 5 (1) 0 0 0 34 (13) 30 Aqueduct* 23 (3) 8 (2) 6 (1) 0 15 (2) 52 (8) 6 184 (22) 153 (27) 78 (12) 1 92 (23) 508 (84) 67 Cistern Cuniculi 69 (3) 47 (3) 26 (2) 1 81 (5) 224 (13) ? Masonry channel 54 (1) 23 (2) 13 (2) 0 15 (6) 105 (11) ? Lead pipe 48 (12) 25 (6) 2 (1) 0 27 (2) 102 (21) ? Clay pipe 42 (4) 18 (2) 17 0 9 86 (6) ? Basin 86 16 (1) 3 0 22 (3) 127 (4) 27 Nymphaeum 31 2 4 (1) 0 9 (3) 46 (4) 13 Bath 119 (39) 30 (18) 4 (3) 0 6 (2) 159 (62) 54 Other 0 0 0 0 5 5 ? already indicated by the ways it was done.441 Survey material rarely allows such detailed information as recognition of speciic architectural forms, so the number of inds remains low. Most of the sites, where an atrium and/or an impluvium have been documented, have been excavated or feature well-preserved building remains. All the sites with identiied atria and/or impluvia are from Classes 1 and 2, and they are spread evenly in the research area (Appendix IV). Survey evidence for atria and impluvia can be compared to the excavated sites: in the sites catalogued, atria and impluvia were found in 31 buildings and connections between impluvia and cisterns were mentioned in six cases.442 Other features interpreted as indicating collection of rainwater inside a building are underground cisterns, wellheads/shafts and rooless aboveground cisterns. The irst two have been assumed to be almost universally present in the central part of the research area.443 This interpretation is not always correct in the Roman Campagna, where underground channels and spaces were used to collect water iltering through porous volcanic soils and stones. Well shafts can lead to such cisterns or also be “true” wells tapping groundwater. The rooless aboveground cisterns are commonly interpreted as having been designed to collect rainwater, but the evidence of roolessness often remains uncertain. A large rooless cistern is at the same time a basin or a tank, which could also have had other functions such as a pool, ishpond, etc. These features are located lower than the villas and the usual interpretation of their function is irrigation of the lands near/below the buildings.444 The relative amounts of the underground and aboveground cisterns from the excavated sites are quite different from those in the survey data: 1:1 compared to 1:3. Cisterns in general, and underground ones in particular, must have been more common, but their connection to roof runoff collection is less evident. Despite the relative lack of evidence, it would seem plausible to assume that rainwater collection must have been a common phenomenon. Rainwater was collected in towns, such as Pompeii,445 where water could also be acquired from public fountains and aqueducts. Public water supply was not generally available in the countryside, so it is likely that all possible means of water collection were exploited. 441 442 443 444 445 Thomas and Wilson 1994, 140, 161–162. De Franceschini 2005, 376–379, 383. Thomas and Wilson 1994, 140, 186. E.g., Tibur IV site 154; Mari 1991, 39; 2005, 17. E.g., Ohlig 2001; Jansen 2002. 95 chaptEr 5 Springs located directly on archaeological sites or near them have been reported in connection to only ca. 65 sites (Appendix IV). Almost half of the springs are connected to settlement sites and almost all of them are located in the northern part of the research area (Plate VI.2). Many of the sites in the northwestern zone have been inhabited from the Archaic period onwards. The inhabitants of the area could have been more dependent on natural water sources in earlier times and hence chosen locations near springs. The ways of developing the springs vary: they have been used via fountains, basins, nymphaea and wells. The local and Roman aqueducts took their water from some of the largest springs in the area (Table 5.3). Small underground tunnels, cuniculi, also often functioned as ways of using and directing spring water.446 The sulphurous springs of Aquae Albulae were used from the 3rd century BC onwards for various purposes with a large bath by the main spring starting from the 1st century AD.447 One site has been interpreted as a iglina in connection to a spring and a pond.448 Wellheads or shafts have been mentioned at 91 settlement sites, sometimes in connection to underground cisterns. The shafts could have been used to tap groundwater or they could have been accesses to underground cisterns. These features are similar to the pozzi romani: shafts of varying depth and shape, not exceeding a diameter of 1 meter and commonly equipped with footholds. The distribution of wells concentrates in the areas below 90 m a.s.l., although some sites at higher altitudes are known. This corresponds roughly with the modern groundwater depth of below 50 m, enabling relatively easy access. Nine of the Roman public aqueducts run in the research area in two main zones: in the northeast through the Aniene valley and on the slopes south of Tibur and in the central part between the Aniene and Tusculum (Fig. 5.2). The lines of the aqueducts descending from the Apennines are fairly well known, as long tracts of them are built on arches above ground and these can be connected to each other with relative certainty. The aqueducts starting from the volcanic area have been more dificult to trace as most of the lines have run in underground channels, which make their discovery and recognition quite dificult. The exception to this is the Aqua Alexandrina, which again runs on arches for great stretches of its line. Water from the great public aqueducts was available only in relatively limited areas: in the Tiburtine region as well as below the main lines towards the lowest slopes of the Alban Hills. Later, the Aqua Alexandrina probably provided water for sites in the central area. The public aqueducts were tapped by private users and some of the branches are quite well known, e.g., the ones leading to the Villa of the Quintilii (Tellenae site U7) and the Villa Sette Bassi (Collatia site 679a). These both derive from the main lines and are built on impressive stretches of arches crossing the countryside, visible from the roads (Via Appia and Via Latina) passing the villas.449 The aqueduct lines run by a great number of large villas near Tibur, but very few branches are known. The aqueducts run underground and are thus dificult to trace. The clearest taps lead to the Aqua Alexandrina, particularly to its western part and two can be securely connected to the aqueduct and lead to villas.450 On Tusculum sites 109, 112, 127, a spring has been connected to a series of cuniculi and a cistern, but the inal destination of the water remains unknown. 447 Tibur III site 340. 448 It. polla sorgiva. Tibur III site 99. 449 Some channels run along the ridges immediately below the main aqueducts east of Sette Bassi and between the Via Labicana and the Via Latina (Collatia sites 711, 757; Bovillae sites 16, 56) and they could be branches based on their topographical location. Collatia site 711 is a series of arches forming a crossing of the Fosso di S. Maura leading to the large villa complex Collatia site 713. 450 The branch leading to the cistern and probable villa Collatia site 448 as well as the branch Collatia site 439 leading to villa Collatia site 441. Two other possible connections can also be mentioned: the channels found running towards the crossroads village at Collatia site 194 or towards villa Collatia site 405. In the irst case, it has been suggested that the tunnels form an independent system and are not connected to the Aqua Alexandrina due to the greater elevation of the site (Thomas and Wilson 1994, 184). In the latter case, the line is not mentioned in the description of the site, but is marked on the map as a clear branch. 446 96 watEr The known taps from the aqueducts to villas lead to some of the largest building complexes known, but the public aqueducts seem not to have attracted much habitation to their surroundings. Considering the risks of illegal and the cost of legal tapping, it is likely that the Class 1 and 2 sites would have been established closer to the aqueducts. The distribution of Class 1 sites does show some correlation with the lines, particularly in the Tiburtine area where many of the largest sites are found just below the aqueducts (Plate VII.1). The only natural water resource available on the slopes is rainwater.451 It has been suggested that the villas would have been built in the area because of the aqueduct and that the small and rare cisterns would be an indication of a constantly lowing water source.452 The double line of the Anio Novus might also have been constructed in order to provide water for Tibur.453 The second area of large Class 1 villas close to the public aqueducts can be found between the Aqua Alexandrina and the aqueducts south of it. The main villa habitation at Tusculum is located high above the aqueducts and could not have access to the public aqueducts. The Aqua Alexandrina region features a concentration of Class 1 sites at elevations lower than the aqueduct. However, the aqueduct is a late addition and the villas were already established before its construction; thus the aqueduct was not signiicant for the selection of their locations. The relative absence of sites directly in the aqueduct areas is also interesting; Class 1 sites are rare. This could indicate observation of the regulations mentioned by Frontinus: the aqueduct lines had to be left unbuilt upon for maintenance purposes. Rich landowners might have been well aware of the punishments and avoided building on top of the lines. In addition to the public aqueducts, local aqueduct systems existed (Plate VII.1). They are most commonly simple underground tunnels dug directly into stone. Some lines known from the Alban Hills area supplied water to Bovillae and the area north and northwest of it.454 The springs are situated on the hills above Bovillae and the channels run downwards along the main ridges dug in peperino or tuff. Another well-known channel system is found in Tibur which takes water from the Aniene and distributes it to the town.455 In addition to the public and local aqueducts, a great number of small water channels are known (Plate VII.1). The best known case is the cuniculus or small underground channel dug in subsoil layers.456 The phenomenon of digging drainage cuniculi has been studied best in Etruria, but the features found in Latium are very similar.457 According to these studies, the cuniculi were used to drain excessive moisture from large tracts of land in periods of wet weather conditions. The channels are found in areas of lithic tuff or with layers of pozzolana that tend to harden from bottom to top, i.e., in geological areas where permeability tends to decrease from top to bottom and ground near the surface can become waterlogged. The soils topping the sequence could also contribute to the need, as poorly permeable soils are often common in the areas where the channels are found. The channelling activity is mostly connected to the 7th to 4th centuries BC, which coincided with wet and cold conditions. The reason behind the activity would have been mainly to create better agricultural lands, 451 The piscina Tibur IV site 71 is a very large open basin, which could have served several villas in its vicinity (Tibur IV sites 70, 72, 73 have been suggested). 452 Mari 1991, 40. 453 Evans 1993. 454 Bovillae sites 168, 185, 188, 304.9, 304.16, 304.18, 304.24, 325, 437. Fountains have been found in the Bovillae area as well as possible connections to villas (Bovillae sites 173, 183). 455 Tibur I sites 187, 188, 189, 190. A local aqueduct has been suggested based on aqueduct remains on the right bank of the Aniene (they have been connected to the Anio Vetus although it runs on the left bank) and epigraphic evidence (CIL X 6247 mentions aqua Tiburtina); Panimolle 1968, 53–55. 456 Confusingly, the term is also used to describe a variety of structures and it is sometimes dificult to know exactly what kind of installation is meant. 457 Ward Perkins 1961; Judson and Kahane 1963; cf. Ravelli and Howarth 1989; Thomas and Wilson 1994, 143. 97 chaptEr 5 but the prevention of malaria has also been suggested.458 Compared to the later, much more sophisticated and unsuccessful attempts to rid the Roman Campagna and the Pontine Marshes of malaria, it seems likely that the cuniculi were not successful either.459 The tunnels were also used for draining small lakes and ponds, collecting and moving water from springs as well as sewage disposal. In connection to settlement sites, the cuniculi are interpreted as drains, sewage channels as well as possible parts of a cunicoli cisterns. Approximately 250 cuniculi have been found in the research area (Appendix IV). Most of the tunnels were dug into the harder subsoil layers: tuff, peperino or hardened pozzolana.460 Sometimes the tunnel was provided with masonry parts or waterproof plaster. The most frequently mentioned function in connection to the villas was drainage. Tunnels were also often found in connection to cisterns, fountains, basins and other water-related installations and they were interpreted as water pipes taking water from its source to its inal destination. In some cases, the collection of water iltering through soils and stone was mentioned in the survey reports. Another important function was directing and controlling water in rivers and streams.461 Some of the tunnels are quite long and have clearly functioned as aqueducts.462 Some villas also had small aqueduct systems of their own based on cuniculi, starting from springs and leading water to single sites.463 Sometimes tunnels hewn in rock were used as corridors in the building on top of them.464 In one case, the cuniculi had nothing to do with water, but were the result of pozzolana extraction.465 The distribution of the cuniculi is concentrated in the central zone and Tusculum (Plate VII.1).466 They have been found more frequently on higher and sloping areas. No great differences between volcanic and limestone geologies can be seen, as cuniculi have also been found on the predominantly limestone slopes around Tibur. Most of the long stretches run along the ridge crests, possibly indicating the use of gravity low for transporting water from a source on a higher elevation to a inal destination in a lower place.467 In most cases, the source of water remains unknown since the conduit cannot be followed to its beginning. The cuniculi as well as the buildings using the water are found on crests of the ridges; hence, water from the streams between the ridges could not be used. Connecting the various bits of cuniculi found on ridge crests to a system is unfortunately impossible due to the fragmentary evidence. A few channels run across ridges, but this could indicate function as drainage channels, rather than as water supply lines. Roman water reservoirs are perhaps one of the most durable and most common architectural features in the Roman Campagna. They also indicate areas where water was available or needed in large quantities and are in such a way relevant to the selection of location. Over 600 cisterns have been recognized in the research area and over 500 of these 458 Ravelli and Howarth 1989. Sallares 2002, 76–79. 460 Digging into rock was mentioned almost 85 times. In most cases, the channel was mentioned only in passing. 461 E.g., Tibur III sites 35, 118, 292; Tibur IV site 79. 462 E.g., Tellenae site 80; possibly also Ficulea site 86; Fidenae site 113; cf. Dell’Era 2002. 463 Conduits at Tibur I site 206 and Tibur IV sites 25–26 led water to Tibur I site 209. Springs near the Villa Adriana were exploited by several conduits (Tibur IV sites 84, 150, 152). Tusculum site 67 for villa Tusculum 61 and sites Tusculum 431–432 for villa Tusculum 190–194. Villa Bovillae site 101 probably has three cisterns located on the slopes above it (sites 102, 110, 118) connected to the villa by an aqueduct (site 110). Bovillae site 173a had an aqueduct of its own (site 168) and a cistern along the line (site 174). 464 E.g., Tibur IV site 68. 465 Bovillae site 101. 466 A neat line of cuniculi sites seems to be located between the Aqua Alexandrina and the aqueduct lines in the central area. The line corresponds to the modern Via Casilina and the railway line running along it, both of which have cut the ridges and revealed the underground channels. 467 Cf. Thomas and Wilson 1994, 185. 459 98 watEr are found on settlement sites. Approximately 20% of the settlement sites feature more than one cistern and the maximum number is seven.468 Cisterns connected to settlement sites are found everywhere in the research area, but those on non-settlement sites are clearly concentrated in Tusculum (Plate VII.2). The cisterns can be divided into two basic categories: underground cisterns dug into rock and aboveground cisterns in masonry spaces of varying forms. Within these categories, there is great variation based on, e.g., ways of arranging spaces. The most common underground cisterns are formed of single or multiple spaces in various shapes. Cisterns a cunicoli consist of tunnels sometimes connected to larger spaces. The aboveground cisterns are rectangular, single spaces built of opus caementicium and covered with waterproof plaster, but multiple spaces in rows or parallels lines have also been built.469 Cisterns are also found as rooms in the buildings, but this tends to be a relatively rare phenomenon. The underground types were accessed through well shafts and/or openings sometimes equipped with wellheads. The aboveground cisterns are generally located outside the main villa building, on the slope above it to allow use of gravity low for distributing water. The sizes of the spaces vary greatly: the ground areas have been reported for half of the cisterns and they vary between 1 m² and 1,400 m². Capacity and water volume are harder to assess, as information on the height of the building as well as the level of the water table in the cistern is very rarely available.470 Underground and aboveground cisterns are distributed evenly over the research area. The most common type of underground cistern is the a cunicoli cistern, i.e., one or more underground channels sometimes connected to larger spaces for collecting and holding water seeping through soil and rock.471 Many of the underground cisterns feature waterproof plaster which is interesting considering the assumed means of water supply: the plaster layer is on the inner surface effectively stopping water from iniltrating from the porous stone. The water could therefore have been collected somewhere further away and just moved by and stored in the cuniculi.472 The sizes of the cisterns indicate possible need and availability of water. The cisterns were divided into three categories according to the size of their ground area: smaller than 80 m², between 81 and 160 m² and larger than 161 m² (Plate VII.2; Table 5.6). Approximately 70–80% of the cisterns are found in Class 1 or 2 villas in all sizes. The distribution of the two smaller groups is similar to the distribution of cisterns in general, but the largest reservoirs are clearly concentrated in region of Tusculum. The storage capacity could also be divided between several cisterns and when the distribution of sites with more than one reservoir is added, the position of Tusculum is further accentuated. Water was clearly needed in the villas of the area and it was available for storage in great quantities. Water was used in many, many ways in the villas: for basic household necessities, for irrigation of both pleasure and productive gardens, for animals, for other productive activities, and for baths. Water was also otherwise an important luxury: lowing water created both coolness on a hot summer day as well as a pleasant auditory backdrop. The archaeological evidence for uses of water are scarce as are other water-related materials: baths, basins and/or nymphaea have been found in ca. 270 sites (Appendix IV). Evidence for water consumption 468 Bovillae site 101. The number of cisterns in the Villa Adriana has proven to be dificult to ind out and it remains unclear (De Franceschini 1991 and online; Ricotti 2001). Considering the great number and huge sizes of spaces where water was used as well as the lack of central cisterns (cf. Villa Iovis on Capri; Krause 1998), the number could be higher than seven. 469 Cf. Riera 1994, 311–373; De Franceschini 2005, 305–311. 470 The ground area is not necessarily a reliable indicator of the cistern’s capacity as some are tower-like, high and narrow, and in these cases, the full height of the structure remains unknown. 471 Cf. Dell’Era 2002, Fig. 8, 11; De Franceschini 2005, Tavv. 4–7. 472 The various cuniculi systems have not been studied systematically and many unveriied assumptions exist. 99 chaptEr 5 related to productive purposes is clearly missing from that list, although basins can be used in many different ways and their functions are dificult to determine. Basins were also used in various productions not connected to water, e.g., for processing grapes and olives.473 Subsistence farming probably required no irrigation as it has been determined that the annual rainfall would have been suficient for grain Site type < 80 m² 80–160 m² > 160 m² No size Total cultivation.474 Some of the rooless cisterns/ Class 1 76 28 32 130 266 pools could have been used for irrigation Class 2 46 11 19 106 177 of gardens and orchards. Baths and basins Class 3 16 11 7 52 86 can be recognized with a certain security Class 4 0 0 0 1 1 even among fragmentary survey inds, Table 5.6 Sizes of cisterns found but inding a nymphaea requires quite in classes of sites. well-preserved architecture or excavation results. Most baths, basins and nymphaea are found at Class 1 sites. The distribution of baths and basins covers most of the research area. Nymphaea, on the other hand, are concentrated almost exclusively in the areas close to the villa centers of Tusculum and Tibur. (Plate VIII.1.) The evidence for use of water is strongly connected to the Class 1 sites and it is spread evenly in the research area. What is perhaps noteworthy is the lack of evidence fpr use of water from Tusculum compared to the very strong presence of cisterns indicating a great need for storing water. It should be remembered that the sites in the region of Tusculum are very commonly incorporated into modern buildings, allowing access only to the platforms. Therefore, few details of the actual buildings on the platforms have survived or are accessible. More than ten baths have been found outside settlement sites, although it is dificult to distinguish “independent” baths and those related to villas. The two cases from the Tiburtine area seem fairly clearly public baths: the baths of Aquae Albulae exploited the sulphurous spring waters and another establishment has been found inside the town walls.475 More than half of the possible public baths are located by the main roads, which could be considered additional evidence for interpreting them as public. The combination of public spaces and functions in private buildings could have been more common than what can be seen based on the archaeological evidence. 5.5 intEgrating thE EvidEncE Healthy site equals high and dry site? Water was not only important to the Romans, but possibly also slightly problematic. A water source near the site selected for habitation was a basic requirement, but water could also cause The basins and pools listed in Appendix IV are inds that are probably not impluvia, nymphaea, parts of baths or productive installations not using water. Two sites have been identiied as related to pottery production: Ficulea site 201 and Tibur III site 156. The cistern of Collatia site 646 has been connected to dying of cloth. 474 Thomas and Wilson 1994, 157–172. 475 Aquae Albulae: Tibur III site 340. Tibur: Tibur I site 169. 473 100 watEr problems by emanating unpleasant smells or unhealthy miasmas. It has been suggested that the most serious detriment to human and animal health caused by stagnant water and poorly drained soils already in ancient times would have been malaria, which has ravaged the Roman Campagna in more recent periods.476 The ancient solution to the problem was to build the houses at high elevations facing away from open water as well as to avoid marshy zones as much as possible. Therefore, the elevations of the settlement sites were studied in relation to their surroundings. The orientation of buildings to bodies of water was also analyzed. Then, the areas of possible waterlogging and marshiness were identiied and compared with the site distribution. The variable topography of the research area is discussed here briely in order to evaluate the connection of the locations to elevation and water – the main discussion on the topic follows in Chapter 6. Most of the area is covered with volcanic ridges and in the northwest, they form wide plateaus incised by small rivers. South of the Aniene, the ridges are long and narrow with generally smooth and gentle slopes descending into ravines. The southwestern zone also has long and narrow ridges, but the slopes and ridge crests are more irregular than elsewhere in the research area. The watersheds run along the ridge crests where most Roman roads were also built. The ravine bottoms were not used for habitation; instead, roads or agricultural ditches and pits are found there. A southwest–northeast-oriented hill system separates the valley of the Fosso di Pratolungo/Marco Simone/Santa Lucia and the travertine plateau. The steepest slopes can be found in the pre-Apennine mountain chain around Tibur. The region of Tusculum also consists of steep slopes, varying between wide, even hillsides, narrow ridges and spurs. The settlement sites are found on the highest parts of the ridges, but tend not to be right on the ridge crest. They are generally slightly removed from the road, closer to the ridge edges. This means that part of the site always faces the ravine and, most commonly, the body of water below. The architecture of the buildings is not usually known in such detail that the entrance, the various parts of the villa or their orientation can be identiied with certainty.477 However, based on the locations of the roads and the ridge edges, it seems more likely that the front of the house faced the road and its back the ridge edge. This would also afford panoramic views from possible representative rooms located at the back of the house, as is the case, e.g., at the Setteinestre Villa in Cosa where the location of the villa’s entrance and the functions of its spaces are known.478 The Villa of the Quintilii (Tellenae site U7) on the Via Appia is structured in a similar manner: the huge circus-shaped garden is built between the main complex and the road. The main complex is on the steep edge of the lava ridge with magniicent views towards the countryside. One part of the house thus faced a body of water, but was clearly situated above it. What spaces were included in the section facing water varies, but they tend to be parts of the residence, of the peristyle or of both. The difference in elevation between the lat or very gently sloping ridge crests and the ravine bottoms is between 10 and 20 meters. Therefore, most of the locations chosen for building in the plateau area had open space on all sides. Even on the steep slopes, the buildings had open areas on both sides and in front of it. The difference in elevation and the general open character of the terrain around the building afforded free movement of air. This might have been enough to keep the smells and miasmas rising from the water below at bay. When the distribution of settlement is examined in relation to elevation chronologically, some trends can be noted. Most of the sites with signs of pre-Late Republican phases have 476 E.g., Celli 1925; 1927; 1933; Sallares 2002, 235–261. The ground plans are also published in such a way that the relationship between the topography and the buildings is hard to understand (e.g., Romizzi 2001; De Franceschini 2005). The frequently strong, three-dimensional character of the locations is ignored (exceptions being Tessaro Pinamonti 1984 and Zarmakoupi 2005; 2006). 478 Carandini 1985a, passim. 477 101 chaptEr 5 Fig. 5.3 Early sites (white) and those established in the 2nd century BC (black) with 100 m a.s.l. contour. been found in the northern and eastern parts of the research area and most of these are located on the higher slopes of the northern zone. The central plateau area and slopes, on the other hand, show a few signs of settlement. Change occurs when new sites are established in the 2nd century BC: these sites are quite commonly placed on the higher slopes above the 100 m a.s.l. line (Fig. 5.3). The higher ground covers 30% of the research area and over 40% of the new sites are located in those areas. The ridge crests are also the highest point in the local terrain and selecting them for settlement sites indicates a preference for elevated positions. A tendency to place the buildings on more elevated points in the topography existed and it became popular during the 2nd century BC. Large bodies of open water are rare in the research area. The two large rivers, the Tiber and the Aniene, were probably the most important features. Only one site has been reported on the Tiber loodplain in the northwest.479 On the other hand, the Aniene valley features a dense net of settlement sites from Classes 2 to 4. The largest and richest Class 1 sites are found at a slight distance from the immediate vicinity of the river valley. (Fig. 2.1a.) The streams are often seasonal with dry or nearly dry channels during the summer months and 479 Fidenae site 53. If the loodplain was inhabited, later alluvium has covered all sites. 102 watEr this might have been considered unhealthy because of stagnant pools of water. Five small lakes probably existed in the area during ancient times and their water level was controlled or drained by channels.480 The distribution of the archaeological remains on the upper rims of the lake basins could indicate that the lakes held water during ancient times. The upper rim sites also correspond with the general preference for elevated locations with a steep slope to one side. The buildings were probably positioned in a similar manner as the ridge crest sites: the front and the entrance were turned away from the water and the back was towards the basin. In addition, water from the lake could have been used for irrigating lands around it. A dried up lake could have been exploited as an irrigated meadow, described by Cato as a lucrative agricultural enterprise. Several zones in the research area have been marshy in recent times (Plate VI.2). The largest marshy area in the whole region, Pantano Borghese (“Borghese swamp”), lies northeast of the Prataporci Basin. The region between Tusculum and Gabii is dotted by smaller and larger depressions which have held water and have been poorly drained. The Aniene valley is another environment where pools of stagnant water formed easily: the valley is fairly lat and the river and its tributaries had their greatest amount of water in that area. The stream seep areas are also located in the Aniene valley and these further increase the amount of water there. The bottoms of the seasonal river channels and small lakes were at risk of turning marshy for part of the year. In addition, the combination of impermeable geology and welldrained soil might have caused waterlogging – the ridge crests are sometimes affected by this problem (Plate VI.2). The travertine plateau with abundant sulphurous springs and very lat terrain was another swampy area. All types of sites are found, even in the potentially marshy zones, but most of the sites belong to Class 4. The Class 1 sites tend to be built further away from the Aniene valley and the Pantano Borghese area. The sites established in the 1st century AD are found mostly in the Aniene valley compared to the sites inhabited for the irst time in the 2nd century BC (Fig. 2.5d–e). Open bodies of water were not actively avoided when selecting a location for a large villa. This corresponds well with the literary evidence describing the great desire to build on all shores. The danger posed by loods was not great enough to scare the landowners away from choosing sites near the rivers. The somber advice from agronomists to have the building’s back towards the water was maintained. Open water could also be an economic advantage as a mode of transportation and the Aniene is described as a navigable river in connection to the stone quarries supplying Rome with building materials. A well-watered area near the villa also offered other opportunities for economic gain, such as good meadows or a place for an easily irrigated market garden. A high location was not always automatically a dry site as indicated by the possibility of waterlogging on the ridge crests where the villas were most often built. Selecting sites near the ridge edge could have solved this problem: the road ran along the highest and lattest part, while the habitation lay closer to the slope, being thus better drained. This kind of location also meant that the drainage channels probably required to keep the building dry did not need to be long and could drain directly into the valley. The locations of the largest sites show that they tended to avoid the marshy areas, whereas the smallest and poorest sites of Class 4 are also found in these zones. The sites established in the 1st century AD are also commonly found in marshy areas, showing that in the period of greatest need of land, even these were regarded suitable for exploitation and probably also for habitation. Those who could afford to select the best sites continued to favor the slightly elevated and drier areas. 480 Certain lakes: Lago di Castiglione, Prataporci and Pantano Secco (i.e., “dry swamp”). There was possibly also a lake-like formation in the valley of the Fosso dell’Acqua Marciana slightly northwest of Grottaferrata (cf. Valenti 2003, 27) and a small lake in the northern part of the area, Laghetto near Marco Simone. Channels controlling water level: sites Ficulea 352a = Tibur III 20a, Collatia 804a and Collatia 820a. 103 chaptEr 5 Malaria? The main problems caused by open and/or stagnant water were unhealthy miasmas and animacula considered to be connected to illnesses. Recently, it has been argued that the Roman Campagna was already infested with malaria possibly from the Early Republican period onwards.481 Unfortunately, no direct evidence can be offered and the claim is based on the rather meager amount of ancient literary evidence as well as on the later and well-documented problems with malaria in the region. Many of the human activities related to agricultural and infrastructural practices changed the vegetation and topography of the area and turned it into a more suitable habitat for mosquitoes. Animal husbandry as the dominant form of agriculture and transhumance might be regarded harmful since it removes the animals from the region for long periods of time and leaves humans as the main targets of the mosquitoes. Market gardening with irrigated plantations could also create breeding sites. In the modern period, the dominant form of landownership in the malaria-ridden, low-lying areas has been an extensively cultivated large estate, a latifundium in ancient terms, with the majority of the population living in towns and villages located in higher places. Another feature of later agricultural practices has been the coincidence of the lower limits of vineyards and the upper limits of the malaria-infected zones. Changes in climate might also have contributed to the process, although rainy and cold weather as well as warm weather may have favored the spread of the disease. The Romans did not understand the connection between malaria and mosquitoes, but the insect’s typical breeding grounds were perceived as unhealthy. Selecting higher sites on the edges of the ridges with plenty of open space around them would have been a good mosquito prevention method: the insects are not strong liers. The distribution of settlement sites offers three arguments in support of the proposed malaria problem. First, the area has been densely inhabited from the Archaic period onwards with a slight decline in the number of sites in the Early and Middle Republican periods in the central area. The decline coincides with poor climate and possibly problems with malaria. However, general social disturbance in the whole central Italian area could also have discouraged population in the countryside. Second, the central area, the ager Pupiniensis, is also described as unhealthy and unsuitable for agricultural purposes in the 4th and 3rd centuries BC and it was more sparsely inhabited until the 1st century AD. Third, the 2nd century BC new sites were often located on higher slopes and the richest Class 1 sites tended to be located at a distance from the marshy zones. The other evidence for malaria is harder to detect. The main forms of agriculture were probably grapes and olive growing and these occurred in the plateau area with less evidence from the higher hill slopes. Market gardening was possibly important, but it is hard to assess its signiicance because of lack of clear evidence. Animal husbandry probably did include transhumance of sheep in the earlier periods, but, later, pig farming might have been of greater importance. The dense habitation would also have excluded large pastures needed for sheep. The intensive agriculture would have required draught animals which would be targets for the mosquitoes all year round. The main type of building is clearly intended for both habitation and production, not just animal shelters or temporary huts for commuting town dwellers. Even the most unpopular locations were inhabited during the peak settlement in the 1st and 2nd centuries AD. Archaeological evidence can tell little about who owned the farms and who did 481 The following discussion is based on Sallares 2002, especially pp. 235–261. He (2002, 246–247, 256–258) disregards the archaeological evidence without thorough consideration. Sallares’s idea of the Roman Campagna in the Roman period is based very heavily on its barren image in the early modern period. This bias has been already noted in Quilici 1974b and 1979 and it is rather unfortunate that such opinions can be expressed with such great conviction as late as 2002. For a discussion of the re-evaluated Etrurian material by the Tiber Valley Project, see Patterson et al. 2004. 104 watEr the farming, but the overwhelming evidence for intensive agriculture, continuing habitation and complete saturation of the area with settlement can be regarded as solid evidence against major health problems in the area during the Late Republican and Early Imperial period. Sparse settlement accompanied by possible extensive agriculture with an emphasis on animal husbandry occurs only in the 4th century AD or later. This coincides with social disturbance as well as a possible poorer climate, starting in the early 6th century AD, both similar to the Early and Middle Republican periods. Malaria may have been present in the Roman Campagna in ancient times, but it does not seem to have affected life there as much as it did in later times.482 Water supply for the villas? The literary evidence suggests that perennial springs or streams on the estate were the preferred sources of water, a well with drinkable water being the third and a cistern or open piscina the last resort. Accordingly, areas with many springs or other good natural water sources should be more popular for habitation than others. Most of the springs known today are located in the northern part of the research area as well as in the zone of Tusculum. Most of the archaeological sites connected to springs, both habitation and non-habitation ones, are also found in these areas. Only a few springs are known in the central area and the sites with reported connections to springs could be associated with the stream seep phenomenon.483 On the other hand, springs can dry out and inding them, e.g., at some distance from an archaeological site, can be quite dificult. The distribution of habitations of different periods or classes does not correspond very clearly with springs. The settlement is dense in the northern, water-rich zone in the early periods and sparse in the water-poor central zone until the maximum settlement density. On the other hand, the dense Archaic settlement in the central area was most likely not dependent on spring water which probably did not exist. The Middle Republican and 2nd century BC new sites in the central area are often located near the spring lines and stream seep areas which could be interpreted as seeking out locations near natural water sources (Plate VI.2; Fig. 5.3). Many of the sites are located on small spurs of the ridge edges. This could be regarded as a way of seeking water, as the spur slopes cut the soil and stone layers and could have made it possible to harvest surface water iltering through layers or stream seep water. Some ravines in the northern area feature small springs in these types of topographical locations and have a settelement site very near.484 The settlement sites are located above the possible water sources, so gravity low-dependent water systems inside the buildings could not have been used, but the water could have been exploited for other purposes. The methods of supplying water to villas varied greatly according to the possibilities afforded by each individual location. If a preferred source was not available, others could be resorted to and, consequently, water supply does not seem to have been a very signiicant factor in site selection. The archaeological evidence suggests that water did not necessarily have to be available on the site, in spite of the advice of the agronomical writers. Water could easily be transported from springs and other sources to its inal destination, probably often crossing estate boundaries on the way. The legal sources concentrate on water servitudes and this could be regarded as evidence for this kind of communal practice.485 Securing a steady supply of water required effort on the part of many landowners in creating and maintaining 482 Malaria does leave signs on the human skeleton, but as few anthropological studies of the inhumation burials found in the Roman region have been conducted, this source of information cannot be used. Cf. Soren and Soren 1999. 483 E.g., Collatia sites 148a, 187c, 196a, 200c, 624b. 484 Particularly Fossi di Cinquina and Cesarina. 485 Bannon 2001. 105 chaptEr 5 the systems and problems were likely to arise. This makes the agronomists’ advice to check out the quality of the neighborhood before buying land and for maintaining good relations with neighbors signiicant. Water for irrigation and display? Water was used for many different purposes and most of the archaeological evidence is related to luxurious consumption, such as baths, fountains and nymphaea. Evidence for use of water, e.g., for productive purposes, is harder to ind. The amounts of water needed for irrigation have been calculated previously,486 but the areas where this might have been needed have not been considered. In order to ind the areas in need of irrigation, the soils draining excessively and permeable geology were combined to see where water probably passed too quickly. Such combinations exist in relatively limited areas in the north and southwest (Plate VI.2). The possibly dry areas also occur close to the zones of possible water accumulation, especially in the southwest where most of the area is possibly too wet or too dry. This may be relected in the relatively sparse settlement evidence from that area. The areas of possible subsoil water accumulation also feature plenty of small and medium-sized Class 3 and 4 sites interpreted as market gardens or smallholdings where self-suficiency and gardening would have been important. The soils are not very suitable for orchards, but the aquatic resources support gardening or other types of irrigated agriculture. Irrigation needs might also have varied with climatic conditions. If the weather did turn drier and hotter starting from the mid/late 2nd century AD, the need for irrigation even for cultivation that earlier were left to be watered by rainfall may have increased. More water would have been needed. It would be tempting to connect the building of the Aqua Alexandrina in the beginning of the 3rd century AD to this phenomenon, but as the purpose of the aqueduct was probably mostly to service the restored Baths of Nero in Rome,487 this interpretation does not seem very likely. Taps from the Alexandrina to the villas are known, but their number is so small that it is not possible to draw any conclusions. Most of the cisterns remain undated, but in the excavated sites, most installations have been dated to the 1st century BC and 2nd century AD.488 The main building phases of the villas in the Imperial period are in the 1st and 2nd centuries AD and these often include building a bath in a separate pavilion-like wing.489 Many cisterns were built in the 2nd century AD, but their purpose might have been to service the new baths and not irrigation.490 The sizes of the cisterns are also interesting when the use of water is considered. The largest cisterns are located in the area of Tusculum and, in addition, the density of cisterns is at its highest in the same area. Considering the size of the building complexes, density of habitation and the relatively poor agricultural possibilities, it would seem likely that irrigation or other productive purposes were not the main reason for storing water. In addition, the position of the cisterns higher than the buildings would 3point towards use in the villas. The 486 Thomas and Wilson 1994, 158–163. LTURS s.v. Aqua Alexandrina (Mari). 488 De Franceschini 2005, 378: 35 dated aboveground cisterns with one from the 2nd–1st centuries BC, ten from the 1st century BC, four from the 1st century BC – 1st century AD, four from the 1st century AD, one from the 1st–2nd centuries AD,sixteen from the 2nd century AD, three from the 2nd–3rd centuries AD and three from the 3rd century AD. 489 De Franceschini 2005, 380: 35 dated sites with baths with one from the 1st century BC – 1st century AD, ten from the 1st century AD, six from the 1st–2nd centuries AD, 18 from the 2nd century AD, six from the 2nd–3rd centuries AD and one from the 3rd century AD. 490 Cf. Thomas and Wilson 1994, 157–172 where cisterns have been discussed only in connection to irrigation and other productive purposes. 487 106 watEr large cisterns could relect possible problems in the water supply and a need to store water. The water could have been used for ornamental pools, fountains, nymphaea, baths, etc. Pastio villatica and market gardening have also been mentioned in connection to Tusculum and these could have required part of the stored water. However, most of the evidence points to use of water for luxury and display. Water as a luxury is conspicuous in Pompeii where aqueduct water is led to the fountains in the atria and peristyles of the large houses. It could also have been used as household water, but the remains of piping seem to indicate display as the main purpose of bringing water into the house.491 This kind of display may be relected in the distribution of the sizes and number of cisterns in the Roman Campagna: they belong to the greater complexes with signs of lavish use of ornamental water. In addition, the aqueducts, both public and local, were connected to some of the largest villas in the area and the buildings further away from the main lines received water from private conduits. The known villa taps into the public aqueducts run partially on monumental arches and it has been suggested that the arches were larger than actually needed. The tap lines were also visible from the main roads and could have accentuated the importance of the building and its owner by their size and visibility.492 5.6 conclusions Water is fundamentally important for life, but in the Roman Campagna, inding water seems not to have been a major issue. Suficient rainfall, springs, aquifer easily accessible, etc., served most of the needs. Water could also be moved via channels and stored for later use. No special effort was really needed to ind locations with good water resources. Water could also be a problem, particularly when the perceived healthfulness of an area or a speciic site was considered. Only one part of the research area was regarded as unhealthy in antiquity, the ager Pupiniensis, and it was less populated until the 1st century AD. The low settlement density might also be related to the poor soils of the area; another factor the ancient inhabitants were well aware of. Although the Roman region might already have been troubled by malaria in ancient times, the Late Republican and Early Imperial settlement density testiies that the disease could have not been as devastating as it was in the early modern period. What emerges from the evidence related to water supply, storage and use is the use of this resource for display and pleasure. Most of the water-related installations are baths, pools and nymphaea. Even the aqueduct tap lines could be perceived as display, showing the villa owner’s wealth and social status. Building and maintaining the local aqueduct lines also required cooperation between the landowners and participation in some communal activities. 491 Jansen 2002. Quilici 1974a, site 679 with regard to Sette Bassi. In the case of the Villa of the Quintilii on the Via Appia, the arches were needed to cross a valley, but could have served the same purpose of display. 492 107 6 tErrain and roMan villas 6.1 background Terrain is intimately connected to many weather phenomena, e.g., catching the warmth of sunshine, avoiding frosts in the winter and torrential surface runoff after heavy rains, avoiding noxious winds and catching good ones. Many of the characteristics of weather and terrain had also to do with the perceived healthfulness of the area. Ancient authors give much attention to selecting proper terrain for a country estate.493 Different types of landscapes were recognized and the best places were considered according to these types – in lowlands one would select a site differently from a mountainous region.494 Lowlands tend to have hotter climates and thus an estate there would suffer from excessive summer heat. Too high a location in the mountains would similarly suffer from cold winters.495 Temperateness was sought and a location somewhere on a medium elevation was the best choice.496 The characteristics discussed most often are slope and aspect, i.e., whether to select a lat or a sloping site, how steep the slope should be and what direction the buildings should be facing. The authors generally agree that sloping ground is preferable to lat ground, as water was liable to collect in lat areas. The worst choice was lat and uneven ground, where pools of water could form easily.497 Hollows should also be avoided for the same reason as well as due to the danger of frost in the winter. Depressions were also susceptible to poor ventilation and would allow the breeding of insects and smelly miasmas would not be blown away in the summer.498 A location halfway up a hill should be chosen, if possible, and a slightly elevated location compared to its surroundings. This way, the estate would avoid most climatic excesses, since there would be some breezes in the summer, but not too much rain and wind in the winter. Possible torrential surface runoff would low past an eminence and thus the foundations of the buildings would not be harmed.499 An elevated position was also safe from possible robber bands roving around the countryside: visibility in a depression was poor and the estate could easily be surprised by plundering gangs.500 The function of the estate could also be a signiicant factor when selecting the terrain. A good farm should consist of varying terrain for exploiting various resources and for producing 493 494 495 496 497 498 499 500 Varro rust. 1,4,4. Colum. 1,2,4; Varro rust. 1,6,2. Varro rust. 1,6,3. Colum. 1,5,8; Vitr. 1,4,1 (cf. Vitr. 6,6,1). Varro rust. 1,6,6. Colum. 1,4,10; Varro rust. 1,12,3–4. Colum. 1,4,10; Varro rust. 1,13,7. Cf. Varro rust. 1,12,4. tErrain different crops. An estate with partly level, partly hilly terrain would provide enough variation for growing grain, having pastures and woods. Slopes above the estate could also provide irrigation for the meadows below them and even water for household purposes.501 Cato’s often repeated advice of placing a farmstead at the foot of hill, mountain or a mountain range is probably for the purpose of acquiring diverse terrain and the opportunities for production it provided.502 The agronomists naturally provide no advice for choosing a location for a residential villa that would have little productive signiicance. Pliny the Younger’s Tuscan villa has one of the better descriptions. It was placed on higher ground, on the healthy slopes of the Apennines. These were so cold in the winter that many of the common crops, such as olives, could not be grown there, but the villa did provide a cool and pleasant retreat in the summer. In the descriptions of both of Pliny’s villas, the scenery is of great importance and a high location yielded spectacular views.503 Some of the villas built on elevations were compared to military camps for their commanding positions and styles.504 Locations related to bodies of water were also often favored for luxurious residences. Aspect of the site was important for natural light and warmth. Recommendations varied according to the general climate: in hot climates, the house should face north, south in cold areas and east in temperate ones.505 Easterly or southerly orientations were favored since they received cooling breezes in the summer, but less stormy winds in the winter. The morning sun would also help warm the house in the winter. A northerly direction was considered noxious. In the summer, the east/south direction is described as giving shade as well as keeping the estate dry to prevent insects breeding and the creation of miasmas.506 The orientation of the various buildings of the estate, for production, storage and habitation, is discussed at even greater length than the selection of the site itself.507 Despite this wealth of material in ancient sources, the terrain and landscape of villas have not been studied in great detail. Survey reports may briely discuss site selection; more commonly when it corresponds to the ancient descriptions. Sometimes the similarity of sites to each other is mentioned, but not described or analyzed in greater detail.508 In a study concerning agricultural production in Southern Etruria, altitude, slope and aspect were analyzed in respect to Roman sites and their resources. It was noted that the sites were located on lower altitudes on latter but not completely lat areas oriented east–southeast–east. All were deemed favorable features for arable cultivation.509 In a study concerning Etruscan and Roman settlement in the Albegna valley in Tuscany, some differences were noted for the two settlement phases concerning elevation, slope steepness and aspect. The Romans preferred the mid-altitudes (50–100 m a.s.l.), slight slopes and s southern aspect whereas the Etruscans preferred east–south and west–north aspects possibly related to general weather systems as well as level ground.510 Recent studies concerning villa architecture have also little to say about the relationship of the villa and the terrain it was built on, at best a general 501 Colum. 1,2,3–4; Varro rust. 1,6,5. Cf. also Varro rust. 1,12,1. Cato agr. 1,3. Repeated in Plin. nat. 17,36 and Varro rust. 1,7,1. 503 Plin. epist. 2,17; 5,6. Visibility and viewability are discussed in Chapter 7. 504 Sen. epist. 51,11 for the Baian villas of Marius, Pompey and Caesar and Val. Max. 8,1 for the story of a man who was punished for building his villa on a too high a location. 505 Pallad. 1,7,3; Plin. nat. 18,33. 506 Colum. 1,5,5; 1,5,8. Cf. Cato agr. 1,3; Colum. 1,2,3; Varro rust. 1,12,1; 1,12,3. 507 Colum. 1,6; Vitr. 6,6,1–7. 508 For the research area, see Quilici 1974a, 27, 29, 40, 49; De Rossi 1979, 15, 18; Quilici and Quilici Gigli 1986, 407, 413; 1993, 483, 491, 492; Mari 1983a, 12, 33; 1991, 28, 32, 38, 40, 41; Valenti 2003, 55, 56, 57, 61, 62. 509 Goodchild 2007, 123–140. 510 Perkins 1999, 44–48. 502 109 chaptEr 6 description of the type “on a hill,” “on a slope,” “on top of a hill,” or “in a valley.”511 The most comprehensive study discusses a series of spectacular luxury villas in the Bay of Naples area and their relationship to the terrain. Despite their beauty, these sites unfortunately are exceptional cases and represent only a very small portion of villas in general.512 The aim of this chapter is to explore the terrain where the building was as well as its surroundings. The basic attributes used in the process are elevation, slope and aspect – these will be examined with the help of a digital elevation or terrain model (DEM/DTM)513 as well as by using printed topographical maps. The landscape and terrain types are introduced irst and this is followed by a survey of literary and archaeological evidence. The chapter inishes with an analysis and discussion correlating all the evidence 6.2 landscapE and tErrain in thE roMan caMpagna Geology and climate regulate how the terrain will be at any given location. The research area can be divided into a dozen major landscape units based on geological divisions (Plate VIII.2). Most of the area is covered by ignimbrites and lava forming a gently undulating plateau crossed by small rivers. Closer to the crater lakes in the Alban Hills, the plateau rises into steeper slopes. In the northeast, the limestone geology crops up with two types of landscape: the travertine plateau and the steep slopes above it. The hills along the Via Nomentana are also part of the pre-volcanic landscape. The research area is crossed by one major river valley, that of the Aniene, cutting both limestone and volcanic landscapes. The Tiber valley lanks the area in the west.514 The uniformity of the geological landscape offers a great deal of local variation when it is examined in greater detail. The elevations vary between 15 and 925 m a.s.l. and the lowest areas (15–20 m a.s.l.) can be found in the major river valleys (Table 6.1; Plate IX.1). The highest slopes can be found in the northeastern and southeastern parts of the research area. The slopes around Tibur reach 590 m a.s.l. in the north and 525 m a.s.l. in the south. In the region of Tusculum, the ridge where the town is located reaches 620 m a.s.l. The highest hill tops can be found on the slopes of Monte Cavo or the ancient Mons Latinus. The elevations rise from southwest to northeast in the northern area and from the northwest to southeast south of the Aniene. The central northern edge features some higher elevations, rising up to 140 m a.s.l. In the southern zone, the rise tends to be regular apart from some lava ridges rising above their surroundings. The volcanic plateau consists of ridges of locally varying shapes. In the northwest, they are longer, wider and latter than south of the Aniene. The edges of the ridges are also slightly steeper as well as more irregular. The ridges run from north to south near the Tiber and northeast to southwest elsewhere. South of the Aniene, the ridges are very long and smooth511 De Franceschini 2005, each site in the catalogue, 295–296, 351–352; Romizzi 2001 discusses only architecture, but mentions terrain types in the catalogue. 512 Tessaro Pinamonti 1984. For an analysis of the villa at Anguillara Sabazia and its surroundings, see Zarmakoupi 2005. The orientation of the villa buildings was studied in an unpublished MA thesis at the Department of History at the University of Helsinki, see Kolbe 1982. 513 The DEM was created by digitizing the contours in the IGM topographical maps, sheets Fº 150 I N.O., I S.O., II N.O., III N.O., III S.O., III N.E., IV N.O., IV N.E., IV S.E., IV S.O. either completely or partially. The point maps were then interpolated, converted to raster, combined and iltered to create the inal DEM. The terrain displayed on the old printed maps is based on the situation of the 1930’s or 1940’s (sometimes even earlier), which was ideal for this study. The great growth of suburban habitation starting after WWII has changed many areas completely and new maps relect this situation. Thus, using the old maps gives perhaps the closest reconstruction of the old terrain. 514 Description and map based on Arnoldus-Huyzendveld et al. 1997. 110 tErrain Table 6.1 Elevations in the research area and number of sites on each height zone by class and date. Elevation m a.s.l. Area ha Area % Sites Class 1 Class 2 Class 3 Class 4 Early Pre 2nd BC 2nd–1st BC 15–20 21–30 31–40 41–50 51–60 61–70 71–80 81–90 91–100 101–110 111–120 121–130 131–140 141–150 151–200 201–250 251–300 301–350 351–400 401–500 501–600 > 601 992 2,709 4,136 5,333 5,227 4,761 5,346 3,127 2,494 2,122 1,525 1,134 933 739 2,438 1,312 855 687 527 842 338 24 2 6 9 11 11 10 11 7 5 5 3 2 2 2 5 3 2 1 1 2 <1 <1 Area = 8% 74/4% 12/4% 12/3% 16/4% 26/4% 5/2% 11/3% 13/5% Area = 52% 1,052/54% 145/44% 232/51% 256/56% 419/60% 207/74% 281/69% 99/38% Area = 26% 587/3 % 113/34% 134/29% 129/29% 211/30% 63/23% 112/27% 57/22% Area = 15% 262/14% 70/21% 84/18% 53/12% 55/8% 9/3% 14/3% 73/28% edged and they start from the Alban Hills, gradually running downwards towards the Aniene. Their orientation is almost perfectly from the southeast to northwest. Ridges with steeper slopes can be found in the lava outcrops north and northwest of Lago di Castiglione. In the southwestern part, the ridges run in a south–southeast to west–northwest direction. The slopes tend to be steeper and more irregular and the crests of the ridges are not as lat as elsewhere in the research area. Instead, they are most often formed of sequences of small hillocks. The volcanic ridge landscape is interrupted in some areas by hills and lat plateaus. In the central northern part, the marine clay form short ridges with fairly steep slopes and narrow crests oriented northwest to southeast. Just above the travertine plateau, the terrain is irregular with steep slopes and many changes in ridge directions. The travertine area and the zone around the modern town of Ciampino are the lattest terrains in the research area. The Ciampino area is located on the lower reaches of the Alban Hills’ slopes by the Via Appia and recent studies have revealed that the lat terrain of the area was formed by lahar activity of the Alban Lake crater.515 The steepest slopes can be found around Tibur and Tusculum. The hills around Tibur are long ridges with very narrow crests. The steepest slopes with more than 20º steepness form long, smooth and continuous surfaces. The Alban Hills high terrain is more uneven and varying with ridges running mostly from southeast to northwest. The aspect of the various areas is naturally closely linked to the terrain shapes and their orientation (Table 6.2; Plate IX.2). The ridges in the northern part tend to run from northwest to southeast and consequently, the southern part below the ridge crest is exposed to an east– southeast–southerly direction and the northern part correspondingly to a west–northwest– northerly direction. The travertine plateau is mostly open to the east–south–southeast. The limestone slopes around Tibur are mostly exposed to the southwest–west–northwest–north except for some areas north of the Aniene valley. The ridges south of the Aniene run from southeast to northwest and so their slopes are exposed mostly to the northeast and southwest. In the southwestern section, the ridges have plenty of east–southeast–south exposure. Most of the Ciampino plain and the Alban Hills’ slopes are exposed to the west–northwest–north. 515 A mudlow from the crater gushed over the area below and smoothed the terrain. Funiciello et al. 2002; 2003; Freda et al. 2006; De Benedetti et al. 2008. 111 chaptEr 6 Table 6.2 a) Aspects in the research area. b) Aspects of the settlement sites by class and date. Note that most sites are open to both main directions. a Aspect Degrees Area ha Area % lat N NE E SE S SW W NW 0 = lat 1 = 338–23 2 = 23–68 3 = 68–113 4 = 113–158 5 = 158–203 6 = 203–248 7 = 248–293 8 = 293–338 1,936 6,273 5,492 4,142 3,471 4,750 6,657 8,266 6,688 4 13 12 9 7 10 14 17 14 b W–NW–N–NE = 56% E–SE–S = 26% E–SE–S–SW = 40% W–NW–N = 45% Aspect Sites Class 1 Class 2 Class 3 Class 4 Early Pre2nd BC 2nd BC 1st AD or later E–SE–S 1,132 58% 1,740 90% 945 49% 1,941 227 68% 303 92% 200 60% 333 285 62% 414 90% 242 53% 459 258 58% 402 90% 212 47% 449 362 52% 618 88% 291 42% 700 153 55% 236 84% 112 40% 280 311 76% 405 99% 270 66% 410 155 59% 235 90% 129 49% 262 297 58% 450 87% 235 46% 517 W–NW–N Both In general, it can be said that the northern area has more exposure to the east–southeast– south and the southeastern area is mostly towards the west–northwest–north. The ridges on the ignimbrite plateau are exposed towards the northeast and southwest with more east– southeast–eastern exposure in the southwestern corner. 6.3 writtEn sourcEs for landscapE and tErrain Many of the major landscape features of the Roman Campagna have been described or mentioned by ancient authors. These include most notably hills and mountains in the region, but the plateau area is also mentioned occasionally. The northwestern section of the research area was important during the Archaic and Early Republican periods due to many wars and skirmishes between Rome, Veii and Fidenae. Despite many descriptions of events taking place in the area in that period, the terrain is rarely described or even mentioned. The hill where the city of Fidenae was probably located is very impressive in height even today, but it is mentioned only once.516 The hills between Fidenae and the Aniene are mentioned once in the same period.517 The Mons Sacer near the third milestone from Rome by the Aniene is mentioned as the place where the secession of the plebs led them to live in 494 BC.518 The impressive hilltops around Tibur are mentioned often, as the town was one of the preferred holiday resorts with a cool climate due to its elevation, but no real description is ever provided. Strabo mentions the water falls of the Aniene and how both Tibur and Praeneste are located on the same mountain range. One of the hilltops was possibly called Mons Catillus after one story of how the town was founded.519 Aefula or Mons Aelanus is another arx or 516 Liv. 4,21,6–7 (435 BC); possibly also Liv. 3,42,3 (449 BC). Liv. 4,17,11 (437 BC). 518 dio BCiv. 1,1; Ascon. Cic. Corn. 67–68; Cic. Brut. 14,54; Dion. Hal. Ant. Rom. 6,45,2; 6,90,1; Fest. 372; Flor. 1,23,1; Liv. 2,32,2–5; 2,33,3; Ov. fast. 3,663–666; Val. Max. 8,9,1; one further reference is from 449 BC, Liv. 3,52,3. The Cornicolani hills to the north of the research area are also possibly mentioned in Dion. Hal. Ant. Rom. 1,16,5, stating that Ficulea is close to the Cornicolani mountains. 519 Hor. carm. 2,6,21–22; 3,4,21–24; Iuv. 3,190–192; 14,87–90; Mart. 1,12,1–2; Serv. Aen. 7,83; 7,672 (Mons Catillus); Strab. 5,3,11. The Praenestine mountains are also mentioned in Liv. 3,8,6 and Verg. Aen. 7,681–685. 517 112 tErrain hilltop near Tibur mentioned by name. A citadel existed there during the Second Punic War (211 BC) and later, a temple of Bona Dea was located nearby.520 Moving slightly southwards, the old town of Collatia was mentioned in connection to a hill.521 Tusculum was also a popular resort and its high hills and cool climate are described mostly in the same vague way,522 apart from Strabo (5,3,12), whose beautiful description is worth quoting in full: But still closer to Rome than the mountainous country where these cities lie, there is another ridge, which leaves a valley (the valley near Algidum) between them and is high as far as Mt. Albanus. It is on this chain that Tusculum is situated, a city with no mean equipment of buildings; and it is adorned by the plantings and villas encircling it, and particularly by those that extend below the city in the general direction of the city of Rome; for here Tusculum is a fertile and well-watered hill, which in many places rises gently into crests and admits of magniicently devised royal palaces. Adjoining this hill are also the foothills of Mt. Albanus, with the same fertility and the same kind of palace. (Trans. H.L. Jones, Loeb Classical Library.) Strabo comments on the height, shape and exposure of the slopes, as well as the density and wealth of habitation. His “viewing point” can be imagined to be from the west, from the direction of Rome, as the description concentrates on that view. Pliny the Elder523 mentions a Collis Corne in the vicinity of Tusculum, where a grove dedicated to Diana lay. Another famous hill in Tusculum is Mons Algidus located outside the research area.524 Mons Albanus is also mentioned sometimes as the previous citation shows. Slightly further southwest, Bovillae is mentioned as being on a hill.525 The other dominate element of the Roman countryside, the plateau with its gently undulating hills is rarely described and the passages are mostly brief mentions of its existence.526 Strabo527 explains slightly more: some of the plains connect Rome and its suburbs and some connect to the sea. The latter were considered generally unhealthy, but the suburban plains were regarded as good for settlement and the houses were as wealthy as at Tusculum. Campi Tiberiani is possibly the name for the area north of the Aniene and the use of the word campus refers to the latness of the area.528 Speciic descriptions of villas reveal very little of the terrain and landscape where they were built. The best known villa in the area is, of course, Cicero’s Tusculanum, but he never Liv. 26,9,9 (citadel, 211 BC); CIL XIV 3530 = I.It. IV,1 611 (temple of Bona Dea). Verg. Aen. 6,773–774. The location of the town has recently been discussed again after Quilici 1974a, site 100 placed the town on the hill of Lunghezza. The many Iron Age and Archaic inds near La Rustica closer to Rome have been suggested as a possible town site; Musco 2001, 191. 522 Hor. carm. 3,29,6–8 and Schol. Hor. carm. 3,29,8; Liv. 3,7,1–3; 3,8,6; Sil. 7,691–693; Dion. Hal. Ant. Rom. 6,3,3 mentions hills above Lacus Regillus. 523 nat. 16,242. Passienus Crispus, cos. AD 27 and 44 as well as Agrippina the Younger’s husband, used to lie under one of the old trees nearby as well as hugging it and irrigating it with wine. 524 The hill is mentioned often in connection to the war against the Aequi in the mid-5th century BC, Liv. 3,23,5–6; 3,38; 3,60,1; Diod. Sic. 12,24,5; Hannibal approached Tusculum from the direction of Mons Algidus – Liv. 26,9,12. Strab. 5,3,9 gives the locations of a town called Algidum and the road station ad Pictas on the 15th mile nearby. 525 Dion. Hal. Ant. Rom. 8,20,1. 526 Dion. Hal. Ant. Rom. 5,37,4; Mart. 3,47; Strab. 5,3,7; 5,3,11–12. 527 5,3,12. 528 Lib.col. (Lachmann) II p.254,5–9; p.255,28–29; p.255,28–29. The irst and third references place the Campi Tiberiani between Rome and Tibur, but the expression videntur in the irst should be noted – the localization seems uncertain. The irst passage is also part of a catalogue of regions exclusively on the left bank of the Tiber. The second passage is on Fidenae and this would also connect the Campi Tiberiani to the Tiber. This would also make sense considering the use of the word Tiber in the place name. Cato fr. 57 (Peter) refers to a campus Tiburtinus which could be the same if the area was indeed between Rome and Tibur. 520 521 113 chaptEr 6 describes its surroundings or views from it. The references to moving up and down inside the villa indicate buildings parts on different levels. Cicero also mentions a great number of other estates, but equally little information can be gleaned from those passages. Clodius’s villa above Bovillae is described as being high thanks to its high platform or foundations.529 Lucullus’s famous villa at Tusculum lay on the typical sloping environment of the hills: he had neighbors above and below his house.530 Cicero’s quip that Quintus Metellus’s (Tiburtine?) villa was visible from the Esquiline Gate could also be interpreted to mean that the building was on high ground and thus stood out.531 The aediicator Cretonius built his various country houses on the heights of Tibur and in other popular resorts.532 Pliny the Younger’s description of the lood of the Aniene in AD 105 mentions villas at varying elevations.533 The villa of Julius Martialis is outside the research area, but the location is described with greater detail than most of the others:534 A high retreat rises from the hills; the lat summit, a moderate swelling, enjoys serener sky, shining with a light all its own while mist covers the winding valleys. (Trans. W.C.A. Ker, Loeb Classical Library.) The descriptions of the Roman Campagna are not very detailed and most of them concentrate on heights, which also often have place names. The villas described are often luxurious residences and the implication is that they were almost invariably located on high positions in the landscape. Some passages also refer to lower locations, e.g., by the main rivers of the plateau area. 6.4 archaEological EvidEncE for sitE sElEction and tErrain typEs Literary evidence does not offer many clues for studying terrain selection. Archaeological evidence, on the contrary, does: almost every reported villa site can be placed on a map and connected to the terrain. Roads tend to run on the ridge crests and the settlement sites tend to be found on the ridge shoulders, often on small spurs jutting out from the ridge shoulder. In areas of long, continuous slopes, such as the areas around Tibur, the buildings often follow the contours, being long and narrow. In order to examine and quantify the types of terrain villas were built on, the sites were plotted in the DEM and the contour curves derived from the DEM. The DEM provided elevation information as well as general lay of the land and the contours helped in recognizing the terrain types. The data was checked against original topographic maps and the clearest choice was selected in ambiguous situations. (Table 6.3.) The basic data was the elevation of the site (Table 6.1; Plate IX.1). Most of the sites were encountered in areas below 100 m a.s.l. and most commonly they were located at elevations between 30 and 80 m a.s.l. The difference between the lowest and highest elevation encountered in the site area gives an idea of the steepness of the site. For almost half of the sites, the difference remains between 0 and 5 meters, which can indicate a slope from 0% up 529 530 531 532 533 534 Cic. Mil. 53. Cic. leg. 3,13,30–31. Cic. de orat. 2,68. Iuv. 14,86–89. epist. 8,17. Mart. 4,64. 114 tErrain Table 6.3 Attributes used to describe terrain for settlement sites. to 20%, depending on the site size.535 Most of the sites were thus found on moderate elevations. The differences in elevation Elevation inside the sites indicate site locations close m a.s.l. as average of highest and lowest point of the site to edges of slopes, not entirely on level Main axis of site if could be determined, according to cardinal direction; what the axis data was ground. based on (including the type of symbol used on the original map) The orientation of the site was the Direction of open area in front of/around the site according to cardinal directions next type of data collected (Table 6.4; Fig Position on slope 6.1).536 The long axis of the site was used ridge top, shoulder, backslope, valley, lat and its direction was compared to those Direction of slope of the main topographical features around according to cardinal directions from the bottom up Terrain around the site the site: contours, direction of the spur, behind, in front of and to either side; gentle/steep slope up/down, lat and direction of road(s). Most sites had Landform an axis from northwest to southeast and spur point, shoulder, etc.; can also be two different landform types the opposite direction, from northeast to Other possible data southwest, was the second most common whether the DEM or IGM printed map was used, etc. orientation. The direction mostly followed those of contours or the general orientation of the spur. Road lines commonly follow the contours and/or ridges and many sites followed the direction of both the roads and ridges.537 The sites were also commonly oriented at 90º angle against one or two of the main features.538 Most of the sites followed the natural orientation of the locations they were built upon, either along or against the main contours or following the direction of a spur. Road lines were probably also an important factor in deciding the orientation of the buildings. The next attribute is the direction of the open area in front of the site (Fig. 6.2). Most of the sites have an ascending slope on one side and a descending slope on the other. The descent is determined as the front of the site and the ascent the area behind the site. The assumption is that the building’s entrance would have been towards the road, possibly with a productive part attached to that part of the house, either Direction Sites Sites % Contour Spur Road Hill Other in front of or next to the living quarters. NE–SW 654 34 322 256 48 22 6 The residence would have been placed NW–SE 946 49 447 423 48 20 8 towards the valley with the possibility of N–S 142 7 67 56 12 4 3 views. This interpretation is based on the W–E 199 10 96 67 27 3 6 few cases where the villa buildings and its Fig. 6.1 and Table 6.4 Direction of the surroundings have been excavated, such as main axis of sites and the features the Setteinestre, Auditorium, Via Gabina the orientation follows 10 and Villa Regina villas.539 A steep slope 535 The exact slope on each site axis was not calculated as this would have to have been done manually and would have required more time than the information would be worth. 536 If the site axis was unclear, the general orientation of the location was used. These include 30% of the sites for each orientation category. 537 The most roads are indicated in the Collatia, Fidenae and Ficulea publications and the sites, whether indicated with raster/lines/symbol, are often drawn following the road lines. The authors also mention the orientation along or against the road lines. 538 At 161 sites (8%) the orientation followed one main feature and was at a 90º angle to another. 539 For Setteinestre, see Carandini 1985a. For the Auditorium Villa, see Carandini et al. 1997; 2006. For the Via Gabina sites, see note 115. For the Villa Regina, see De Caro 1994. 115 chaptEr 6 Fig. 6.2 Number and direction of slopes around the settlement sites. 116 tErrain or even a body of water in front of the villa does not prevent it also having an entrance in this direction, as has been assumed in the cases of the Loc. Santa Maria Villa at Nemi and the Villa di Arianna at Castellammare di Stabia.540 Both have entrances from below the main building in addition to an entrance from a road running behind the villa. The direction and the extent of the open area in front of the site were determined by checking the elevation information on the site and its surroundings (Fig. 6.3; Table 6.5). The areas below the Width of view Sites Sites % Direction Sites Sites % site as well as those on both sides lower or 45° 1 0 NW–N 572 30 roughly at the same level were determined 90° 45 2 N–NE 418 22 to be visible from it. Most sites had a 135° 308 16 SW–W 240 13 wide open view in front of them, being E–SE 170 8 180° 561 29 S–SW 164 8 225° 699 36 most often between 135º and 225º. A Others 346 18 270° 193 10 small number of sites, located on top of a 315° 43 2 small hill rising higher than its immediate 360° 71 4 surroundings, have a 360º view. The center Other 20 1 point of all views was most often oriented Fig. 6.3 and Table 6.5 Most common view towards west–northwest–north–northeast. directions and view widths. The opposite direction of east–southeast– south–southwest occurs in roughly half the number of sites compared to this. The most common extent and orientation of a single view is from the southwest to northeast covering 225º. From the analysis of rise and fall of slope as well as the extent and direction of the open area in front of the site, it seems that the location preferred was one with a slight upward slope on one side and a descent on the other. The general view to the surroundings of the site was broad and directed towards west–northwest–north–northeast. The attributes describing the nature of the site were also observed. In addition to the rise behind the site and the fall in front of the site, the situation on both sides was examined. Level ground on both sides was the most common, i.e., the terrain around the site was similar to the site itself. Another common type was a gentle descent on both sides of the site. The ive most popular site types can be described by combining all the terrain data and they cover 70% of the sites (Fig. 6.2). The most commonly selected site type was on the edge of a slope or on a spur. This is veriied by analysis of the direction of the slope starting from below and ending on top of the hill. The most common situation is the one where the slope rises from one direction only and the most common direction is from northwest to southeast. The second most common location features slopes in three directions and the most common direction is SW–NE/NW–SE/NE–SW. The directions also show the narrowness of the spurs as the most common types have parallel edges with opposing ascents and the point is 90º against these directions. From this rather complex quantitative analysis emerge two typical locations for a Roman villa in the Roman Campagna. The irst one is the upper edge of a ridge with the terrain continuing lat on both sides of the site. The site is open towards the northwest/northeast, but 540 For Nemi, see Guldager Bilde 2004; 2005. For the Villa di Arianna, see Romizzi 2001, 189–190 sito 50. 117 chaptEr 6 some also look towards the southwest/southeast. The second type is a small spur, usually a fairly narrow, lattish area lanked on three sides by slopes, most commonly oriented towards the northwest–north–northeast. In addition, more gentle landforms resembling spurs, oriented westwards, were sometimes chosen. An abundant open space towards the descent forming wide open views ranging between 135º and 225º is also a common characteristic of the terrain. The most common viewing direction is northwest–north–northeast. 6.5 intEgrating thE EvidEncE More regional variation is connected to the terrain than with the any of the other attributes analyzed in previous chapters. Terrain and landforms change from one zone to another – what is a high elevation in one part of the research area is low in another. In this last part of the chapter, all evidence is integrated to see whether the site selection in the different parts of the Roman Campagna was done according to the general rules set out by the Roman agronomists and how the choice of terrain might have changed over time. The characteristics examined are elevation, landform type, slope gradient and aspect. Mid-slope on an eminence? In general, it can be concluded that the terrain of the research area rises towards its edges. The ridges of the plateau area rise smoothly or in small steps towards the higher slopes. According to the Roman authors, an ideal location for a farm is at the foot of a hill or on a mid-slope, but not on top of a hill. A small eminence would be preferable for catching the breezes of summer, yet avoiding bad weather in winter. This kind of location would also be safe from erosion by torrential surface runoff after heavy rains as well as from water collecting in depressions of lat terrain. Based on the archaeological evidence, approximately half of the sites are located in the area between 30 and 75 m a.s.l. This corresponds to the elevation distribution in the research area: the elevations between 30 and 75 m a.s.l. cover roughly half of the research area (Table 6.2). The proprotions of site classes are similar in elevations below 150 m a.s.l., but above that, more Class 1 and 2 sites can be found. A chronological comparison of site distributions shows that most of the sites used before the 2nd century BC are located below 75 m a.s.l., whereas the sites established in the 2nd and 1st century BC are situated in the higher areas, above 150 m a.s.l. Thus, large and luxurious residential villas tend to be located at higher elevations than modest farms. This coincides with what the ancient sources say in general about locations of productive farms and elite country houses. The next issue is the position of the site on the slope and how to recognize and quantify the important landforms in the research area. Dividing a continuous terrain surface into landforms is not easy, but a cognitive division can be made using the data derived from the DEM. First, the ridge crests were plotted by using a low map derived from the DEM and checked against the DEM. Secondly, the valley bottoms were plotted using the river lines. Thirdly, the ridge shoulders were isolated by using a gradient map: areas where the slope steepness was strong, i.e., 12–20%, were plotted against the ridge and valley data. The same was done with lat areas (0–6%) and very steep slopes (> 20%). In the resulting map (Plate X.1), the lattish ridge crests541 are lanked by the strong (or very steep) slopes forming the shoulder and backslope. 541 In some areas, the lat valley bottoms and ridge crests unfortunately still merge. 118 tErrain Toe slopes and valley bottoms are gently sloping or lat. The elevation difference between the ridge crest and valley bottom is dificult to quantify because of the steady rise of the terrain. A comparison was made manually and the difference was between 10 and 30 meters for the northern and eastern plateau area and slightly lower for the southwestern part. In the hilly areas, the difference could be as much as 60 meters. From this division of landforms, it can be concluded that only a few gently sloping backslopes recommended by the Roman agronomists exist in the research area. Instead of the backslope, the ridge crests – the highest points in the landscape – were selected for settlement. This can be veriied by comparing the distribution of sites to the ridge crests (Plate X.1). A 25 m buffer zone was plotted along the ridge crests and these buffers zones cover 3% of the research area. A very large proportion of the sites, 34% (666 sites), lay directly on the buffer zone or close enough to touch it. The highest numbers (40%) come from the plateau which consists mostly of ridges and valleys. Distributions by class and date show similar igures: over 30% of the sites are close to ridge crests. Class 1 sites and sites established before the 2nd century BC are even more frequently located near ridge crests (40%). The highest points in the plateau topography were thus strongly favored. The narrow ridge crests end in a shoulder signiied by a slope gradient of 6–20% and these ridge shoulders cover 34% of the research area. In the central area, the ridge slopes tend to be gentler. In the region of Tusculum, the 6–20% gradient is not necessarily a ridge shoulder, but more often represents a generally sloping area, commonly with a gradient of over 20%. Sixty-three percent of sites (1224 sites) plotted against the 6–20% gradient have this kind of slope in their areas. The gradient is most common in the sites in the northwestern and southern areas which could be expected due to the landforms. Class 1 and 2 sites are most often near or on the shoulder (70%). A study of chronological distribution shows that the sites established in the 2nd century BC favored the shoulder zones or in general (in the region of Tusculum) slightly steeper slopes. Putting the ridge crest and shoulder data together shows that 23% of all the sites are on both terrain types and of these, 30% are Class 1 sites. The sites inhabited only in earlier periods were not very often situated on ridge crest/shoulder locations. The ones that continued to be inhabited later or established in the 2nd century BC favor these terrains. The areas best matching the site description of the Roman authors are probably the ridge shoulders, the area between the lat crest and the steep backslope. This is also evident from the description of a typical site derived from the archaeological analysis: the edge of a ridge or a small spur on the ridge edge with a wide open area in front. The protruding element is caused by gully erosion and these could match the ancient descriptions of an eminence on the slope. The spur would function as natural erosion protection for the buildings. A site on the ridge shoulder would also mean varying terrain for different resources: lat ridge crest for arable cultivation, slopes could be used for grapes, olives or orchards, the valley bottom offered the possibility of irrigated meadows as well as that of growing reeds. The surface runoff as well as cut geological layers offered the possibility of collecting water and having spring(s) near the site. Another favorable feature of a mid-slope location mentioned by the ancient authors is the protection against the cold winter winds, which is again dificult to model or test, but possibly held true.542 The only area where the sites are literally mid-slope is the Tiburtine zone: the large Class 1 sites on both sides of the town are located near the bottom of the steepest slopes, before the gentler toe slope section begins. Recognizing speciic features such as a spur on the digital maps is quite dificult, if not impossible, so controlling their frequency in the research area against site distributions is 542 Personal experience from the ield trips in the northwestern area also conirms this: the sites located slightly below the ridge crests were protected from the winds that could be felt while moving along the ridge crest. 119 chaptEr 6 not possible. The ridge and shoulder sites identiied above were classiied as spur point, spur top, shoulder and ridge crest locations (Table 6.6; Fig. 6.4). The data collected based on archaeological evidence testiies to the popularity of the spur-related locations for habitation: 45% of all and 42% of early sites. The proportion is the same for almost the whole research area apart from the northeastern zone where the travertine plateau and steep limestone Back Front Sides Topography Sites Sites % slopes create different landforms compared ascent descent lat shoulder 465 24 to the volcanic ones. Approximately half ascent descent descent spur 315 15 ascent descent descent/lat shoulder of spur 250 13 of the Class 1 sites are on spurs and this lat descent descent/lat spur 151 8 is the highest proportion of all the classes. lat descent lat shoulder 149 8 The early sites are slightly more commonly Pre 2nd 2nd/ 1st Class Class Class Class found on ridge shoulders than other Spur All BC 1 2 3 4 Early BC chronological groups – the sites inhabited Near point 653 131 159 157 206 88 161 101 On top 213 35 40 43 95 30 54 28 later also are related to spurs. Total sites 866 166 199 200 301 118 215 129 One further observation based on Sites % 45 50 43 45 43 42 52 49 the topographical analysis is the slight differences found in the southwestern area. Fig. 6.4 and Table 6.6 Names for parts of hill, While the area is good for agriculture and typical site locations and numbers of sites on also features plenty of southerly slopes, the these. Spur sites by class and date. habitation there is relatively scarce.543 The terrain of the area consists mainly of ridges as in, e.g., in the central area, but these feature much rougher crests and steeper backslopes. As a result, fewer spur or ridge shoulder sites preferred for Roman habitation exist. This could be a partial explanation for the relative lack of sites in that area. The kind of site promoted by the agronomists’ advice was not easily found in the Roman Campagna, but locations offering the same beneits could be found. There seems to be little variation in site selection in different site classes, but the large and luxurious ones tend to be located on higher areas and on spurs. Chronologically, the variation is also small, but the later sites tend to be again higher and spur-related. This may signify that the preferred types of locations were deined only in the 2nd century BC and that these ideals were kept thereafter. Aspect and orientation The recommendation for the orientation of the villa was to the east or south for the temperate climate present in the Roman Campagna. The reality of the area was different from the ideal. Slightly over half of the slopes (56%) are facing from west to northeast and of the remaining area, most (45%) faces southwest–west–northwest (Plate IX.1). Some regional variation exists, as in the northern zone, the easterly–southerly directions dominate (51–54%). In the northeast, the southerly slopes are concentrated in the western region and the area around Tibur faces mostly southwest–west–northwest. In the zone of Tusculum, the northwest–north direction predominates with almost 80% of all the slopes. Relatively few east–south slopes can be found and these are mostly located in the northern and southwestern areas. Of these, the northern area has larger, homogeneously south–east oriented surfaces while the slopes 543 Possibly due to gaps in surveys and lack of plowed ields in the area. See Chapter 2.3. 120 tErrain Table 6.7 Sun altitude and azimuth during solstices and equinoxes. in the southwestern area are discontinuous. Plotting together the east–southeast– south and the west–northwest–north March 21 6:30 am 12:00 pm 6:15 pm directions with the sites gives the following sun elevation angle/altitude 2.4° 47.9° 0.4° results. The southerly orientation covers azimuth 92.1° 173.5° 269.6° 26% of the research area and almost 60% June 21 4:45 am 12:00 pm 7:30 pm sun elevation angle/altitude 0.8° 71.4° 2.1° of the sites have a part of their ground azimuth 58.5° 171.6° 300.1° area on such orientation. The northerly September 21 6:15 am 12:00 pm 6:00 pm direction covers 45% and 90% of the sites sun elevation angle/altitude 2.8° 49.0° 1.2° have part of their ground area towards that azimuth 91.3° 178.8° 270.1° December 21 7:45 am 12:00 pm 4:30 pm direction. Approximately 50% of the sites sun elevation angle/altitude 0.9° 24.7° 1.0° feature both directions. The main land azimuth 123.2° 178.0° 236.6° form of fairly narrow ridges offers a partial explanation, as most preferred locations near the ridge crests would feature areas to both main directions. The fairly wide ridge crests in the northwestern zone offer fairly uniform southerly slopes, but the narrow ridges afford locations open to all directions. The ridge side towards south was chosen if it was possible. There does not seem to be much difference between different parts of the research area, except for the obvious high proportions in the northern part with the most south slopes and low igures in the central and southern section with few south slopes. The settlements in the southwestern area differ from the rest of the southern region as almost half of them are on southerly spots. The sites that continued to be inhabited after the early periods also feature more southerly sections than those inhabited only in the early periods or established later. Having part of the site with a southern exposure was thus important for all settlement sites in all periods. The varying exposures of the site could have served the different needs of warmth and light recommended for the many spaces of the villa. Class 1 sites are most often located on southerly spots (70%/227 sites) which is interesting considering that many of them are located in the regions where few such slopes can be found, i.e., near Tibur and Tusculum. For the large Class 1 villas established on elevated locations in the 2nd century BC, a good southerly exposure cannot have been an imperative, as so many of the sites are located in Tibur and Tusculum. Based on literary evidence, many of these villas were destined for short visits probably during most seasons of the year. The possibly least frequented period would be summer, when the destination would most likely be at a greater distance from Rome. The cool climate and fresh air of the most popular resorts were appreciated and the villa buildings oriented away from the sun would have contributed to the pleasantness of the villa in the hotter seasons. On the other hand, if the house was used more frequently during the colder periods, a southerly exposure would have helped in keeping the house warm and dry. One further analysis of orientation was conducted and that was to check the amount of sunlight during different seasons for each location. The sun elevation angle and azimuth for sunrise, midday and sundown were plotted as hillshade maps. These were then reclassiied and added together to see which parts of the research area had the most sun on March 21, June 21, September 21 and December 21 (Table 6.7; Fig 6.5).544 The results for spring and fall are practically identical: the most sun was received by the southerly spots in the whole area. The June sun reached almost every part of the research area most of the day. Tibur and Tusculum 544 The angles and azimuths as well as times used were calculated at the National Research Council Canada Sunset/ Sunrise Calculator (http://www.nrc-cnrc.gc.ca/eng/services/hia/sunrise-sunset.html) using the longitude and latitude of Rome. Taking account of changes in sun movements in the past 2,000 years was deemed unnecessary because of the large scale of the analysis. 121 chaptEr 6 Fig. 6.5 Areas receiving most sunlight during the day on a) March 21, b) June 21, c) September 21 and d) December 21. Darker areas receive more light. also got their fair share of the summer heat. Exposure to sun all day in the summer may have been perceived as a healthful factor, as that was thought to prevent insects from breeding.545 The importance of the southerly exposure becomes most clearly underlined by the December sunlight distribution: the southern spots were in the sun for most of the day during the coldest period of the year. The analysis also shows that very few points can be found in the research area where some part of the higher slopes would shadow the area adjacent to it. This happens mostly only in the mornings in the regions of Tibur and Tusculum. The last issues to be discussed in connection to orientation are the direction of the view, i.e., the open area in front of the site, and exposure to winds. Based on the archaeological material, the main center point direction for 58% the site views is between west–northwest– north–northeast and 29% of sites have a view in the opposite direction, i.e., east–southeast– south–southwest. This relects the general topography of the research area and its regional variation. In the northwestern zone, plenty of southerly areas can be found and, consequently, most sites tend to be on ridge shoulders or spurs (195 sites/40%) facing southeast–south– 545 Varro rust. 1,12,3. 122 tErrain southwest (217/44%). The same igures also apply to the northeastern zone, but there the most common site type is the ridge shoulder facing southeast–south–southwest. In the southwestern part, the proportions for northerly and southerly directions are equal. In the central zone, the northerly direction is prevalent and the main site types are either ridge shoulders or spurs facing north. In the southeastern area, a clear minority of sites – and of the terrain – faces south. The directions which are almost constantly avoided are northeast and east. The eastern slopes are relatively rare in the whole area and the largest proportion of northeastern slopes can be found in the central zone. According to the recommendations, an easterly direction should have been favored, but this did not happen in the Roman Campagna. The choice of location was clear: if a southerly site was impossible, the westerly–northerly ones were preferred. In the autumn and winter, the current dominant wind directions in the area are from the north or northeast. In the spring and summer, the opposite directions prevail: south or southwest. The autumn and winter winds come from the mountains (Tramontana and Grecale) and are cold and dry, whereas the southerly winds (Mezzogiorno and Scirocco) tend to be humid and warm. The southwesterly–westerly (Libeccio and Ponentino) wind direction is from the sea and in the summer, it is an important factor in generating air movement and relieving heat in the Roman Campagna.546 The site orientation is well adapted for these prevailing conditions: avoiding direct north and northeast meant that the sites could catch the summer winds and avoided the worst winter gales, precisely according to the recommendations of the Roman authors. In addition, the very hot and humid southeasterly wind, the Scirocco, could be avoided by selecting these sites. The high locations also function well in catching breezes in the summer periods, but in the winter they were probably bitterly cold. The arrangement of the rooms so that the entertainment spaces were towards the open landscape and more likely to catch winds could also mean that the smells of the productive parts did not reach them.547 6.6 conclusions In addition to soils, the terrain where the buildings of a country estate were to be placed is of great importance to the ancient authors. The placement of the villa in general and then the orientation of its various parts also had signiicance for their function. The favored types of location, spurs and ridge shoulders, were already used for the earliest settlement sites, but become almost a norm in the 2nd century BC. For the rich and large sites, locations on the high slopes were selected. For most sites, it was also important to have open space on one side of the building. The orientation was not always the one recommended, but the east–south slopes cannot be found in most of the area. The eastern aspect might even have been avoided in order to catch heat-reducing winds in the summer and to evade the cold winter winds blowing from the east. Light and natural heat was available from all directions on most sites on the plateau, but the southern spots would have beneitted most from the winter sun. Terrain, including altitude, aspect, slope and geomorphology, is probably the most repeated piece of advice given by the Roman authors. A certain terrain was sought time after time over very long periods. All kinds of buildings were located in the same terrain, showing the strength and the success of the tradition. 546 E.g., Naval Intelligence Division 1944, 517, Figs. 84–89; Colacino and Dell’osso 1978; Mastrantonio et al. 1994; Giuliacci et al. 2001; Schipani 2008, Tav. 1.7. 547 E.g., Colum. 1,6,11; 9,5,1. For a brief discussion of smells in general, see Hobson 2009, 105–115. Cf. Langley 2006, 322 for a suggestion that the buildings were set in such a way that the smells were swept away from the living quarters by the winds. 123 7 visibility, viEwability and roMan villas 7.1 background A beautiful view is not mentioned as a basic requirement for site selection by the Roman agronomists, and indeed, what working farm would have a real need for a beautiful view? A view does not beneit the production aspect of the farm directly, but it could inluence the relationship of the owner with the estate. The Roman agronomists placed great emphasis on the fact that the estate and its the buildings should provide the owner with comfort and pleasure – that way it would attract the owner to stay there for longer periods of time and, at the same time, to supervise and manage the production himself.548 Thus, selecting a site with beautiful views and placing the estate’s buildings so that they could be enjoyed would potentially enhance its production. It is also clear that different estates existed: some were destined for production and some for pleasure. In the latter kind, the view – or rather views – it could provide were of great importance for the enjoyment of the villa.549 Being able to see the area around the dwelling could also be perceived as a useful feature. The villa should not be placed in a hollow because poor visibility of its surroundings made controlling activities around the villa dificult.550 In addition, the owner might want to be able to see his game parks or woodlands serving the same purpose in the vicinity of his dwelling. This can be seen as a desire to control the estate, but also as the opportunity for enjoying the sight of woods and wild animals.551 The need for light and prospect was recognized in Roman law. A landowner could build his house as high as he wished, but could not block the light from his neighbor completely. A builder had the possibility to obtain servitude to block the light and a neighbor could get servitude to limit the height of the building next door. Light and having it is also connected to views. Some of the views or prospects could also be regarded as so valuable that they were protected even without servitude, and if not, such servitude could be obtained by the owner. In the context of the servitude, prospect meant a beautiful, a delightful view, i.e., it had to be worth protecting for the owner. The view was regarded as part of the estate, a feature that probably raised its value. Mountain and sea views were mentioned as such possibly valuable prospects. The legislation concerning views and the legal concept of prospectus dates to the late Republic, which its well with what is thought of the development of the residential 548 Cato agr. 4,1; Colum. 1,4,8. E.g., Plin. epist. 2,17 and 5,6 are classic passages. Cf. also Auson. Mos. 318–326; Cic. ac. 2,80; ad Q. fr. 3,1,1; Att. 12,9; Att. 14,13,1; fam. 7,1,1; Mart. 4,64; Plin. epist. 9,7; Sen. epist. 51; 55,6; Sidon. epist. 2,9,1; Stat. silv. 2,2,72–82. 550 Varro rust. 1,12,4. 551 Colum. 9, praef.; 9,1,1. 549 visibility and viEwability villa.552 Views from villas are described relatively rarely, the most well-known cases being the two buildings described by Pliny the Younger. His Laurentine villa was located on the coast and had different views of the sea, but part of the view consisted of the neighboring estates and their buildings as well as of the nearby woods. The view was also directed inwards, to the rooms and spaces inside the building complex. His Tuscan villa was located inland and the views there consisted of a wide panorama of mountains and agricultural landscapes. The building itself formed part of the view there as well.553 In the other references, bodies of water are often the central theme of the view: the villas were very often located by a lake or on the sea,554 in a high position affording wide panoramas of the surrounding areas. On a visit to Hortensius’s villa at Bauli on the Bay of Naples, Cicero describes how the view from the xystus stretched to nearby Cumae and Puteoli, as well as all the way to Pompeii had Cicero’s eyesight not been so poor that he was not able to see Catulus’s villa there.555 Variety in views was obviously preferred: the main themes were a body of water or an agricultural landscape with maybe mountains and woods. The surrounding villas or urban centers could also be part of the view, and, lastly, parts of the building itself were visible. Wild nature, mountains and woods tended to function as frames for the views, being somewhere at a distance from the viewing position.556 The descriptions of the views are most often given from inside the building, and through a window. Pliny the Younger describes what could be seen from the windows of his villas. At Laurentinum, the sea view was exploited to maximum effect as it could be enjoyed from a number of spaces. Many of these rooms were used for entertainment and reception by the villa owner, and thus a splendid view would enhance the status and pleasure of the room.557 The description of the Tuscan villa is different as the views from the rooms are not described in great detail. The general view is described as a magniicent panorama with mountains in the distance and an agricultural landscape in the foreground. The views from the windows were directed towards the building complex, not outwards to the sweeping panorama. This turning inwards might emphasize the more private nature of the Tuscan villa: its function was not as formal as of the Laurentinum near Rome. Pliny also compares the view from his Tuscan villa to a landscape painting, and the description brings to mind the references to landscape paintings on walls.558 In a villa, a real landscape could be used, framed by a window, but in a town house, that would not be possible. Instead, depictions of nature in wall paintings were used to bring nature into the house and to combine it cleverly with art.559 Cicero writes about villa life at length, but rarely mentions such things as views. His interests lie in what happens inside the building and these events rarely extend beyond the walls. The passage on Bauli mentioned above and another one on an unspeciied villa use the views to explain and discuss theories of eyesight and seeing. In the second passage, the view is discussed in relation to the width of windows. Cicero’s architect Cyrus recommended narrow windows based on his theory of seeing, but the narrow windows were criticized by others 552 Rodger 1972, 38–89, 124–140. For the Laurentine villa, see epist. 2,17,5, 10–12, 21. For the Tuscan villa, see epist. 5,6,7, 13–14, 18–19, 30. Agricultural landscape from a villa is also mentioned in Verg. georg. 2,284–287. 554 Villas with views over bodies of water: Auson. Mos. 318–326; Cic. ad Q. fr. 3,1,1; Att. 12,9; Att. 14,13,1; Plin. epist. 9,7; Plin. nat. 3,54; Sen. epist. 89,21; Sidon. epist. 2,2,11; Stat. silv. 1,3,39–40; 2,2,72–82. 555 Cic. ac. 2,80. For a wide panorama, see also Auson. Mos. 318–326; Cic. fam. 7,20,1; Mart. 4,64; Plin. epist. 5,6,13–14; Sen. epist. 51; 89,21; Sidon. epist. 2,2,11. 556 Cic. Att. 12,9; 14,13,1; Colum. 9,1,1; Plin. epist. 2,17,5; 2,17,21; 5,6,7; 5,6,13–14; 5,6,18–19; Sidon. epist. 2,9,1; Stat. silv. 1,3,39–40. 557 Cf. Leach 1997; Allison 2004, 131–132, 168, 170–171. 558 Plin. epist. 5,6,13–14. Other references, e.g., Plin. nat. 35,116–117 and Vitr. 7,5,2. 559 Bergmann 2002; Hackworth Petersen 2006, 136–162. 553 125 chaptEr 7 for constraining the view. The passage shows that these theories might also have inluenced architectural solutions. In another instance, Cicero mentions the magniicent scenery of Stabiae and how his friend Marius enjoyed it from his cubiculum during the morning hours.560 Having a view and seeing around the buildings is one part of visibility, the other is that of being seen, viewability. Villas were an integral part of the common agricultural landscape as well as that of the coastal areas. Literary descriptions of such landscapes from the outside, looking towards the villa, are relatively rare. This might be regarded as surprising considering how common the villa landscape was. The large country houses on high locations would have been visible to people moving in the area on land or approaching a coast by boat. They could be visible from afar – the villas of Catulus at Cumae and Pompeii could be seen from Bauli.561 Villa landscapes are sometimes described in a manner that suggests a view from a distance.562 The coastal villas could be approached by sea or by land, and depending on this, the impact of the building would be very different.563 Viewing inland sites might be made dificult or hindered altogether by walls around estates or by vegetation, either natural or planted.564 Plantations for protecting the buildings from viewers were not explicitly recommended, but hedges and walls could have served such a purpose. Visibility and viewability have not been discussed very much in research literature. The magniicent views are habitually mentioned in connection to studies of architecture and the social prestige of the buildings, but the issue has never been handled in depth.565 The purpose of this chapter is to look at irst at the evidence for visibility and viewability in the research area and then try to draw conclusions on how they might have inluenced the choice of locations for villas. The main points studied are the visibility from the villa and the viewability of the building from the outside. 7.2 visibility and viEwability in thE roMan caMpagna The possibilities for seeing and being seen mostly depend on three things: types of terrain, vegetation and man-made viewing hindrances, such as buildings, fences and walls. The topography of the Roman Campagna consists of a gently undulating plateau lanked by high slopes in the northeast and southeast. Moving away from Rome, the relief tends to rise in almost all directions apart from southwest, i.e., towards the coast. Only a few topographical hindrances for views exist in the area between Rome, Tibur and Tusculum and these are high spurs of lava encountered in the Alban Hills area. The Alban Hills are perhaps the clearest landmark in all of the Roman Campagna: they are visible from almost everywhere in the area. The more distant and smaller Cornicolani Hills to the north of the research area are also a landmark, but they are not as clearly discernible and as commonly visible as the Alban Hills. The terrain is such that locations with expansive panoramic views are quite common and easy to ind. In clear weather, it is possible to see for very long distances from most elevated Unspeciied villa: Att. 2,3,2. Stabiae: fam. 7,1,1. In addition, fam. 7,20,1 is on a house or possibly a villa in Velia and how the building would have had an extensive view, if a lotus tree in front of it could have been cut down. See also Davies 1971 for an analysis of Cicero’s perception of natural beauty. 561 Cic. ac. 2,80. 562 E.g., Istria in Cass. Var. 12,22,5 and the coast of Laurentum in Plin. epist. 2,17, 21; 2,17,27. See also Sen. dial. 5,21,5; epist. 55,6; Verg. ecl. 1,82–83. 563 Stat. silv. 2,2,6–12. 564 Cato agr. 6,3; Liv. 33,6,7; Varro rust. 1,14–15. 565 E.g., Tessaro Pinamonti 1984, 50–55; Mielsch 1987, 137–140; Purcell 1995; Schneider 1995, 76–93; Hales 2003, 44–46. For a brief analysis of Portuguese villas, see Langley 2006. 560 126 visibility and viEwability points, e.g., from Frascati, roughly 20 km from Rome as the crow lies, it is possible to discern large buildings in Rome. Vice versa, from the slopes of the Gianicolo hill in Rome, it is easy to see the small towns on the slopes of the surrounding hills, and even single buildings, such as the villas around Frascati. The same viewing is not possible from Tibur located at a slightly lower elevation and at a much greater distance from Rome. The best viewing points on the plateau are the ridge crests and this can be seen in the cumulative viewshed calculated using the DEM with ridge lines as viewing points (Plate X.2a).566 The cumulative viewshed means that the visibility is calculated from each point of the viewing zones – or in this case, lines – and added together to show which areas are visible from most points. The resulting maps show not only what areas were visible, but also which could be seen from a multitude of points. The image for the ridge crests shows how most of the research area is visible from them. The poorest viewing possibilities occur in the river valleys large and small, as is shown by the cumulative viewshed calculated using the channel lines as the viewing point (Plate X.2b). Only the largest river valleys with wide lower reaches afford better visibility. Even in these valleys, the best visibility is limited to the valley bottoms and does not necessarily reach even the ridge crests. The shoulder areas, on the other hand, are visible from both the ridge crests and valley bottoms. The ridge shoulders afford a good opportunity for viewing their immediate vicinity and might even have a more extended view, but not as wide and farreaching as from the ridge crests.567 What is striking for both viewsheds is that the slopes of Tusculum are so well visible from both ridge crests and valley bottoms. Another feature worth noting is the poor visibility in the southwestern corner of the research area for both viewing locations. The DEM does not take into account the possible effects of vegetation, as the vegetation cover of the area in Roman times is dificult to evaluate and to reconstruct. Considering the density of settlement, it would seem safe to assume that no great forests could be found except perhaps on the hill areas. This is also shown by the pollen analyses from the lakes in the Alban Hills, where the amount of arboreal or tree pollen diminishes during the Roman period.568 It is even more dificult to evaluate the vegetation created by human activity, but the results of the recent excavations in the eastern parts of the Roman Campagna and the environmental reconstructions based on them indicate that the landscape was intensely cultivated.569 Tree crops, i.e. fruit and olive trees, were probably cultivated widely. Individual olive trees are planted relatively far apart from each other, which means that they would not necessarily have created dense vegetation covers hindering visibility. The same probably applies to fruit trees and vineyards. The hill slopes were also exploited for cultivation. The roads were also often constructed by digging until a harder surface was found and this process had to be renewed when the road surface was worn down by trafic. The process sometimes resulted in very deep road cuts.570 The main roads, however, were paved with stone and probably maintained their original levels. The natural topography thus afforded good opportunities for having good views and being seen from most of its points despite the problems of evaluating the effects vegetation cover might have had. More problems might have occurred due to human activity, mostly by 566 A viewing height of 1.70 m and 20 km radius. The radius was selected based on the attested line of sight between Rome and Frascati. These two parameters have been used in calculating all the viewsheds. A resolution of 25 m was used on the raster maps. Calculating cumulative viewsheds from multiple viewing areas or points was selected partly for reducing work time since calculating a viewshed for each site is a slow process. For viewsheds in general, see Wheatley and Gillings 2000. 567 The results of the viewshed analysis are strengthened by my own experience of the Roman Campagna during long walks in the ield and photographing views from various types of archaeological sites. 568 Lowe et al. 1996. 569 Pracchia 2001, 286–308. 570 Musco 2001, passim; Pracchia 2001, 261–270; Kuusisto and Tuppi 2009. 127 chaptEr 7 buildings and by possibly creating deliberate barriers for visibility. 7.3 writtEn sourcEs for visibility and viEwability The landscapes of the Roman Campagna are described relatively frequently in Roman literature, probably due to the signiicance of the area to Roman Italy. The writers mostly lived in Rome or its surroundings and often used what they saw in their works. They do not normally discuss visibility directly, but some conclusions can be drawn from the texts indirectly. In addition, descriptions of smaller regions and even of some villas can be found. Strabo571 describes the area as part of his geographical treatise, starting from the northeast with Tibur and Praeneste and then proceeding southwards to Tusculum. He mentions that these towns were visible from Rome. The villas and plantings around Tusculum and in the western Alban Hills areas are admired in the following passages. Martial’s poem describing the Janiculan villa of Julius Martialis572 lists the sites visible from the building starting with the hills of Rome and then continuing with the classic Latin cities of Alba and Tusculum. The sites north of Rome – Fidenae, Rubra and the grove of Anna Perenna – are mentioned next, and at the end, Tibur and Praeneste as the current fashionable watering places. The last two also lie furthest away from Rome. Statius writes a similar list while describing the beginning of the hot summer season and holidays: Praeneste, Diana’s grove in the Alban Hills, Algidus, Tusculum and Tibur.573 The cities and other locations in the Roman area are also compared to various other villa resorts, such as the coast of Latium, Baiae and the Bay of Naples in general.574 The same features are repeated in the descriptions: the viewing point is from west to east, i.e., from Rome towards the countryside and the hills at its borders. The towns and other places mentioned are the same, especially Tibur, Praeneste and Tusculum. The emphasis is on the coolness and shadiness of these places, probably referring to woods, and in the case of Tibur, also to the cool waters of the Aniene. The evocation of the villa resorts in the surroundings of Rome is based on viewing the area from Rome and many of the readers would probably have been able to picture the panorama with the towns and villas clearly in their minds. Sometimes more limited landscapes are described. If the passages mentioned above emphasized the great panorama, these descriptions are more local and mostly give information on where buildings might have been located and whether there were woods or cultivation in the area. Pliny the Younger’s description of his Laurentinum also includes a short passage on the landscapes on the way to the villa. The diversity of the landscape is emphasized: woods limiting the views and wide stretches of meadows. The seashore with the buildings adorning it is described slightly later as part of the view from the villa.575 The townscape and surroundings of Tusculum are described in connection to an incident with Camillus in the early 4th century BC.576 Fields and pastures with locks surround the walled and gated town. These descriptions are not necessarily accurate as they were written centuries after the incident took place, but perhaps give an idea of what was perceived as a proper town. Gabii is 571 5,3,11–12. 4,64. 573 Stat. silv. 4,4. Cf. also Flor. 1,5,6–7; Hor. carm. 3,29,6–8; Iuv. 14,87–90. 574 E.g., Fronto 2,6,3 (Naples, Puteoli); Hor. carm. 3,4,21–24 (Baiae); Mart. 4,57 (Lucrine Lake, Baiae); 4,60 (Ardea); 5,1 (Nemus, Antium, Terracina, Caieta, Circeii); 5,71 (Trebula); 6,43 (Baiae); 10,30 (Antium, Caieta, Circeii, Minturnae, Lucrine Lake); Stat. silv. 1,3,83–89 (Ardea, Antium, Formiae, Circeii, Caieta, Anxur). 575 On the way to the Laurentinum, epist. 2,17,3. Neighbors seen from Pliny’s villa, epist. 2,17,27. 576 Liv. 6,25–26, Plut. Vit. Cam. 28,1–3. 572 128 visibility and viEwability also described as a populous town gravitating towards the main road and possibly surviving on the income from the inns lanking the road. The ruins of the old city are still visible in its surroundings.577 Pliny’s other description of the area is related to the AD 105 lood of the Aniene and the destruction it caused.578 The river banks before the lood were covered with villas and had also woods or groves (nemus). Public buildings are mentioned in connection to higher areas also affected by the lood. The few references to villas include the Janiculan villa of Julius Martialis mentioned above. It was not very large, but located high on a ridge crest and its rooftops were visible even at a distance. Pliny’s Laurentinum and Cicero’s Tusculanum are not described from the outside at all. Horace offers a long description of his Sabinum situated in Digentia (modern Licenza) outside the research area.579 The villa is located in a mountainous region with hills limiting the view and the estate included cultivations, tree crops as well as a spring. The villa of Emperor Nero’s freedman Phaon near the fourth milestones of the Via Salaria and Via Nomentana is described by Suetonius.580 A small bypath led to the villa from the main road and its surroundings were riddled with reeds and brambles. In addition, a pit for extracting sand, a pool possibly used as a water source and some muddy ground are mentioned. Again, the views are not described, nor is the building’s viewability, but considering that it was removed from the main road and surrounded partially by scrub, it seems likely not to have been clearly visible. A short quip by Cicero referring to a villa of Q. Caecilius Metellus Pius (cos. 80 BC) outside the Esquiline Gate could be interpreted to mean that his villa, possibly in Tibur, could be seen all the way from Rome.581 The jest indicates that the villa was so large that even the myopic Cicero was able to see it from Rome. Another reference to the viewability of a building is an inscription found from Aquae Albulae below Tibur, a votive text erected at the temple located in the area. The offering is witnessed by the “painted facades of the Aelian villa”, i.e., the Villa Adriana.582 This could also be merely a reference to the imperial villa and not mean that it was actually visible from Aquae Albulae or vice versa. Today, the western part of the villa affords an unhindered view towards the Roman Campagna and also towards Aquae Albulae. The few references do not allow general conclusions, but connected to the descriptions of landscape, they refer to the visibility and viewability of the villas in the Roman region, at least of those large enough and located on the elevated points. Some evidence also refers to possible viewing hindrances created by vegetation, e.g., concerning the villa of Phaon. Cato recommends planting trees, such as elms and poplars, on the borders of the estate and along the roads as well as having poplars and reed thickets along river banks and wet ground. Similar advice is given by Varro, who describes various types of walls and fences to be built around the villa to protect it. Along the public roads and rivers, the landowner should dig a ditch and raise a bank to protect the estate from both intruders and water. Estate borders should be secured with planted trees. In the region of Tusculum, masonry walls were a common fence type. Columella mentions that fruit orchards and gardens should be fenced as well as sheep, cattle and game enclosures.583 Protection from 577 Dion. Hal. Ant. Rom. 4,53,1. epist. 8,17. 579 epist. 1,16,1–16. 580 Nero 48. 581 de orat. 2,68. The location outside the Esquiline Gate could refer to Tibur or nearby since the Via Tiburtina started from that gate. The family had a Tiburtine estate during the life of Pius’s father, Q. Caecilius Metellus Numidicus (cos. 109 BC; Cic. de orat. 2,65). 582 CIL XIV 3911 = I.It. IV,1 596; translation by Kitchell in Frizell 2004. 583 Cato agr. 6,3; Varro rust. 1,14–15; Colum. 1,6,24; 9,1,1–4; 11,3,2. Cf. also the writings of Roman land surveyors, Campbell 2000, passim. 578 129 chaptEr 7 viewers is never mentioned explicitly, but it could have been implied. Walls, banks, trees and shrubs, e.g., by the roadsides could all function as effective barriers to viewing. 7.4 archaEological EvidEncE for sitE sElEction, visibility and viEwability The villa locations discussed in the previous chapter can be simpliied even further when visibility is concerned. One type is the villa situated on the plateau, on ridge crests or shoulders and on small spurs oriented towards westerly or northerly directions. The second main type is the villa located on the higher slopes of the mountain ranges lanking the plateau. Almost all the locations selected for villas afford very wide views ranging between 135 degrees to a nearly complete circular view. Both location types afford fairly wide views in one direction – the terrain tends to rise behind the buildings forming thus an effective viewing barrier. The locations on the plateau area can have a wide panoramic view if the site is slightly higher than its surroundings, but the sites on the high slopes have almost certainly a very wide and farreaching landscape in front.584 Many of the sites are thus naturally suited for having good views, but knowing how these possibilities were exploited is more dificult. Most villas were slightly removed from the roads and probably built so that the main entrance was towards the road and the more private spaces were further away from the roadside, towards the slopes and the panoramic view. More detailed studies of the relationship between the landscape and the buildings are hampered by various problems in the material. Villa remains are often very fragmentary, only partially excavated, and the walls have commonly survived only slightly above foundation level. Consequently, it is very dificult to recognize spaces and their possible uses. The building parts suited for rooms with views were often built on artiicial platforms that are now mostly destroyed. Windows or other openings have thus survived only rarely. Some better preserved villas can be found on the plateau, but they are almost totally lacking on the slopes. More plateau sites have also been excavated which gives a better idea of the use of individual spaces and the division of various functions inside the complex. The composition of the buildings varies very much in the area as few regular or common ground plans exist.585 Despite the great variation, some common features exist. Rooms are grouped around courtyards, such as atria with pools in the middle and peristyles with the classic four porticoes or other variations.586 In addition, simple, undecorated courtyards have been found connected to the production and service parts of the building. These are often found in the entrance area, thus generally facing the road. On small farms, the living quarters and production spaces are located around the same courtyard, whereas in larger complexes they can be more clearly separated. One of the most easily recognizable building parts is the bath, usually located close to the edges of the building or built in a separate wing.587 The general functions of 584 Viewsheds from single sites were calculated to check the results of the cumulative viewsheds from the topographical points presented above. The results from single sites corresponded with the results from cumulative viewsheds and it was deemed suficient to calculate only the cumulative viewsheds to see what was visible from most sites representing, e.g., different classes of sites. It could also be added that the viewsheds from the high locations continue outside the DEM of the research area. 585 Cf. the ground plans in Romizzi 2001 and De Franceschini 2005. 586 De Franceschini 2005, 383–384 lists 31 atria and 23 peristyles for the hundred villas she has studied. Of these, six had separate residential and production courtyards and thirteen had both atrium and peristyle. 587 A bath in a separate wing is called a padiglione by De Franceschini 2005, 313–315. She lists 54 sites with baths (p. 380), of which 21 are of the a padiglione type. In the survey material, ca. 150 baths have been identiied. 130 visibility and viEwability the buildings can thus be identiied, but drawing conclusions on how the location and the surrounding landscape were used is more dificult. This depends either on missing data on topography or the structures themselves. On the plateau, the higher parts of the building would have had good possibilities for panoramic views of the surrounding countryside. Most of the villas in this region were probably destined for both production and habitation,588 and would not be necessarily built with good views in mind. In the slopes, the buildings are often on artiicial platforms and the slope rises steeply behind. Here also the higher parts of the building, e.g., on the upper platform, would have wider views. The upper platforms were generally used as living quarters and the lower platforms were gardens, peristyles, etc. In addition, the edges of substructures could in theory have been used for creating lookout points or areas, but in reality, e.g., the cryptoporticoes tend to have only small, high windows that let in light and air, but did not allow looking out. Most of the sites on steeper slopes are from Classes 1 and 2, i.e., the larger and more elaborately decorated ones. Three of the villas on the plateau area have survived suficiently well to show positions of windows: Quintilii (Tellenae site U7), Sette Bassi (Collatia site 679a) and the Villa Adriana (Tibur IV site U1). These villas are, of course, quite exceptional even in the Roman Campagna. They are some of the largest complexes known and in general built for residential and entertainment purposes. They could have functioned as models for the architecture of more modest buildings. Based on the evidence they offer, few reliable trends can be delineated for all villas, but as the only surviving evidence, they should not be ignored. The Villa of the Quintilii is built on the northeastern shoulder of a lava ridge with a relatively lat top and steep slopes (Plates XI.1; XII.1b). The ridge rises high above the surrounding countryside towards northeast. The closest main road is the Via Appia running on the ridge crest southwest of the villa. The villa features a large hippodrome-shaped garden (R) towards the road with a nymphaeum (H) at the end. The ridge crest features few buildings but a large garden (Stadium, S) has previously been envisaged there. The main building complex (A–B) is located on the ridge shoulder, facing in an easterly direction. Living quarters on two levels (B) have been built on the slope, but most of the buildings are on top of the ridge shoulder. Relatively few high walls and windows survive apart from the massive bath complex in two separate buildings (D–E). A viewing platform can be found in the main complex (A11), facing south/southeast. The view would have consisted of the small bath (M) and possibly the Alban Hills behind them.589 The bath complex (D–E), on the other hand, features windows on two loors. The frigidarium (D) is on higher ground, possibly offering a 360º view. On the irst loor, the huge windows on the main axis opened in two directions: to the southwest, the view was internal, to the hippodrome garden. The view to the northeast was towards open landscape, assuming that no other buildings or high walls blocked it. The upper loor windows would have afforded a magniicent view all around. The caldarium (E) is located on a lower level and closer to the ridge shoulder. It had large windows in three directions: in the southeast towards the main building (A), in the northwest towards the ridge crest (Plate XI.2) and in the northeast towards the open landscape. The caldarium and the oval building (F) were placed so that the latter does not block the former’s views, merely framing the view towards southeast. The upper loor views would have reached very far towards the southeast and northwest.590 588 E.g., of the excavated sites 80% are such farms (De Franceschini 2005, 349–350). The heights of the platform and the surrounding buildings have not been indicated in publications, so this aspect remains uncertain. 590 If an upper loor existed, the main rooms could have been very high and no windows on two levels towards northeast, where the largest windows opened, can be seen. The analysis is based on ground plans and descriptions of the villa, especially in De Franceschini 2005, site 81, as well as on my own visits, photographs and photos found at various internet sites. 589 131 chaptEr 7 The Villa Sette Bassi is located on a gentle spur near the Via Latina running roughly south of the villa (Plate XII.1). The main building (A–B) is on the highest part of the spur with a massive hippodrome-shaped garden (D) below it to the west and south. The northwest end of the hippodrome also features rooms on a lower level (C). The main building offered good possibilities for views in all directions, as few natural obstacles around the site exist. The western side opened onto the hippodrome garden and the countryside around it. The compact central block in the southeast featured a courtyard towards the north with internal views from the irst loor windows. The upper loor windows would have afforded panoramic views in all directions. The rooms at the end of the hippodrome garden were possibly a bath and they would have opened internally towards the garden and possibly outwards to the small valley below the main complex – the building probably continued to northwest. Little survives of the hippodrome walls, but they feature one square and several circular tower-like structures which could have been used as viewing points.591 The Villa Adriana is the third villa in the Roman Campagna with high walls and good preservation (Plate XII.2). The complex is vast and built on various levels of a ridge with fairly steep river valleys to the northeast and southwest. From the higher points of the terrain, it was possible to have a 360º view, but, in reality, visibility depended on the location inside the complex, the height of the building as well as the height of the buildings around it. Many of the central entertainment spaces, such as the Canopus (A), the nymphaeum/stadium (B) or the central complex with its peristyles are located in such a way that extensive views were probably not likely from them. Internal views were common, but not external ones. The external views can be veriied for the western part, towards Rome and the countryside. The northwestern edge of the hippodrome-shaped Pecile (C) could have had a great panorama towards the countryside, but most of its walls have not survived above the substructures. The walls of the porticoes on both sides are windowless, but the gable would have afforded a view to the north–northwest.592 The Great Bath (D) had a series of fairly high windows opening westwards. The building with the ish pond (Peschiera, E) could also have had upper loor windows to the west. In addition, the small tower (Torre di Roccabruna, F) is located on the western edge of the villa. On the east side, the view opened towards Tibur and the mountains. The area of the Greek Theater (G) and Palaestra (H) as well as Piazza d’Oro (I) could have opened to the east. The Piazza d’Oro seems to have not exploited the views, as the surviving high wall has few openings. The peristyles and courtyards in the higher main building offered internal views of the buildings and towards the east/northeast they were framed by the high mountains rising above.593 In addition, surviving windows on many levels in the buildings on the northern part, the Bibliothecae (K), can be found. The terrain next to these buildings slopes downwards and thus the views would have been unobstructed. These examples show that in an optimal case, views from the villas could have been spectacular and that external views were exploited. The other side of the equation is looking at the buildings from the outside – could the villas be seen from a distance? Today, the archaeological remains – even the huge artiicial platforms – tend to blend with vegetation and the rest of the landscape; they are not easily noticeable except to the trained eye (Plate XIII.1). Many of the buildings were huge in size compared to most modern structures, covering several hectares of land. The platforms were originally probably decorated with colorful plaster and 591 The analysis is based mainly on my own visits to the site as well as on the ground plans and descriptions in De Franceschini 2005, site 75. Old photographs in Bloch 1958 were also useful. 592 In the miniature model of the villa, the northwest gable walls have windows, but it is dificult to know how accurate the reconstruction is. 593 Map 4 of gardens in De Franceschini online indicates “closed” and “open” gardens in the complex. The irst are integral parts of the building, inside peristyles and open courtyards and the latter more ambiguous terraced spaces with few remains of buildings around them. 132 visibility and viEwability the actual buildings rose maybe even two loors above them. No large physical obstacles were found in front of the elevated locations and this would ensure that the villa was visible from many points in its surroundings. Certainly the largest buildings on high locations could have been easily seen from a distance and from many places in the Roman Campagna. Even on the plateau, an elevated location would ensure high viewability in the immediate surroundings of the villa. The buildings were viewed from various positions in the landscape. A passer-by would normally look at the landscape from the roads. The main roads were built in long, more or less straight lines crossing the terrain in various ways (Plate XIII.2a).594 The Via Salaria followed the Tiber valley edges, lanking the high ridges east of it. Visibility from its line was not very extensive outside the valley. Seeing the buildings on the ridge shoulders was dificult, and only the ones at a distance would have been visible. The Via Nomentana enters the research area through the Aniene valley and runs on top of irregular ridges, turning sharply northwards at 10 miles from Rome (Plate XIII.2b). Visibility from the road line is best to its immediate surroundings, particularly to northwest, but also crosses the eastern plateau all the way to the Alban Hills. The lower parts of the northern zone, e.g., the travertine area and the river valleys remain out of sight. The Via Tiburtina runs through most of the research area (Plate XIII.2c). Visibility from the road line is again best in its immediate vicinity, but moving along the road from Rome to Tibur, it is possible to see most of the research area. The travertine area, the higher slopes of the Alban Hills and the Tiburtine region are very well covered. The irst main road south of the Aniene is the Via Praenestina running west to east in an almost straight line (Plate XIII.2d). The visibility from the road line is best in its immediate vicinity as on all the other roads, but the slopes of the Alban Hills and the Tiburtine region are also well covered. Many parts of the northwestern zone are also visible. The Via Labicana is located at a slightly higher elevation and it passes through the area diagonally towards the southeast irst crossing ridges and then, towards the Alban Hills, following their direction (Plate XIV.1a). The Via Labicana’s zone of good visibility is very large, covering most of the northern as well as the eastern region. The Via Latina is often regarded as one of the oldest roads in the Roman Campagna, as it connected the areas of the old Roman and Latin tribes (Plate XIV.1b). The road leads southeast from Rome following the general ridge direction and climbing the Alban Hills towards their central caldera. The Via Latina has perhaps the best overall visibility of the whole area, but the hills west of Tusculum remain mostly out of sight. Even the southwestern part of the research area is visible from this road. The Via Appia is built on a high lava ridge all the way up to the Alban Hills area (Plate XIV.1c). It is slightly higher than most of the surrounding countryside and thus the zone of best visibility from its line is very wide, but it does not cover the whole region as thoroughly as the Via Latina. The lines of the main roads running through the southwestern region are not as certain as the others. The Via Ardeatina was possibly one of them and it mostly crosses the ridges and valleys in the area instead of following them (Plate XIV.1d). Its visibility pattern is very different compared to the other roads since the eastern area remains almost completely out of sight. Even the immediately adjacent areas are poorly visible. The best visibility covers only the roadsides and the southwestern slopes of the Alban Hills. The secondary and tertiary road network in the research area is also dense, but poorly known compared to the main roads. The main lines of the known secondary roads tend to run on the ridge crests in the northern and eastern parts, but crossing the ridges in the southwest. The visibility they offer can be partially deduced from the visibility from the ridge lines (cf. Plate X.2a), but a separate viewshed was also calculated for the secondary road network 594 The road lines used in the analyses are based on the information provided by the survey reports. The accuracy and certainty of the lines varies, but considering the scale of the maps used and research objectives, small changes brought about by later research probably would not change the results very much. 133 chaptEr 7 (Plate XIV.2b). Compared to the viewshed from the ridges and the main roads (Plate XIV.2a), the secondary road viewshed differs mostly in range – the roads do not run on every ridge and in the southwestern area, they run across the ridges. The eastern zone is well covered by secondary and tertiary roads, and the northwest is second best. The roads naturally cover the northeastern travertine area better than the ridges. The viewsheds presented above are, of course, theoretical possibilities for seeing. They do not take into account the impact of vegetation, walls, tombs and other buildings lanking the roadsides. In addition, the roads were often cut deeper into the landscape, running below the general ground surface595 which would have made them useless as points for viewing the landscape. Very little is known of stone or masonry walls surrounding estates, but they could have existed. Hedges may have been common enough, but inding evidence of them is even more dificult than for walls.596 Tombs of all kinds, high and low, lanked the main roads, particularly in the stretches closer to Rome (Plate XV.1).In places where larger burial grounds were located, tombs could have totally blocked the view. Very few other buildings are found built directly by the roads. Gaps in the obstacles affording glimpses of the surrounding countryside could probably always have been found. This is also obvious when walking on the modern roads: most of the time, the view is partially blocked, but every now and then the landscape opens up and it is possible to see far and wide. 7.5 intEgrating thE EvidEncE Literary descriptions imply that Romans enjoyed beautiful scenery and good views; they were even regarded valuable enough to be safeguarded by means of servitude. Mountain and sea views seem to have been valued based on court cases as well as on other texts. Views also included those inside the building – a great variety was appreciated. The buildings were designed in such a way that views could be enjoyed from various rooms, but the architectural descriptions are vague and it remains unclear how the rooms were connected to each other and what the overall loor plan looked like. The same uncertainty applies to the archaeological remains: only parts of the building are visible or excavated, little remains of the structures and the function of spaces cannot be determined. The sites most often afford the possibility of panoramic views and, the higher the site, the more sweeping the panorama. How this possibility was acted upon – and if it was acted upon – is more dificult to determine. The tendency to select the high locations in the Tusculum and Tibur areas starts from the beginning of the Late Republic and could indicate that views became more important in this period. The appearance of prospectus as a legal concept at about the same time could also be regarded as a possible indication of the development. The “open” villa? The irst task is to look at villa architecture and determine if it enabled views from the building as well as to examine what the building looked like to people viewing it from the outside. It has been suggested that villas were open in loor plan and open to their surroundings.597 A recent comparison of the amount of open spaces as part of entertainment space in town houses 595 Cf. Musco 2001, 149–236 passim; Pracchia 2001, 259–270 with more detail. Pracchia 2001, 241–246, 251–259. 597 E.g., Schneider 1995, 76. In addition, the basic villa types presented in Romizzi 2001 are based on open spaces, the atrium and the peristyle. 596 134 visibility and viEwability of Pompeii and Herculaneum and villas in the Bay of Naples region has tentatively shown that the igure was relatively stable for houses at 15–25%, whereas the same igure varied a great deal among the villas, from 7% to 75%.598 The majority of villas feature less than 30% of open space, but one third has over 40% of open space. The amount of open space does not seem to correlate very clearly with the total ground area of the villa. Thus, some of the villas had plenty of open space inside the building, but not necessarily all of them. Of interest here is the connection between the open space inside and the landscape outside, and this is explored by studying villa architecture and how it evolved. The chronological development of villa architecture is still mostly unknown. At the moment, it seems that the earlier buildings were most often compact rectangular or square masses with few wings or other protruding elements.599 The ground plans include often two of the central Roman architectural elements, the atrium and the colonnaded courtyard. Both of them can be described as internal features. The open space was surrounded by rooms that opened onto the courtyard, but the courtyard itself had no openings except, of course, upwards. It is possible that the rooms opening onto a peristyle/atrium could have had windows on the outside, but this is dificult to verify. In the rectangular loor plan with a courtyard in the middle, only the rooms on the outside could open outwards and have an external view. The best examples of a fairly complete building of this type are the Villa dei Misteri just outside Pompeii and possibly the Setteinestre Villa near Cosa. In the earlier phase of the Villa dei Misteri, the side on the sea is lanked by a portico and the rooms have few windows. In the later building, the porticoes are limited to the sides and the main façade has an exedra lanked by corridors, both with large windows towards the sea. Few of the spaces have windows – the room with the famous wall paintings is one of the exceptions. This room had a sea view through its door and the window opened towards Pompeii and the coastline through the southern portico.600 The Setteinestre building also had a portico on the outside. None of the reception rooms could have had windows opening directly outwards.601 Single porticoes lanking the building could open outwards and have extensive views, but the actual rooms have few windows. In the tiny sample, the irst loors feature few windows outwards, even in open countryside. The views tend to be internal, towards gardens, peristyles, or porticoes. The possible upper loor(s) probably featured windows, but nothing is known of these. Later ground plans show plenty of variation: pavilions and wings are built. The buildings stretch out in all directions from a central core. This can also be seen very clearly in the Roman region. In addition to fashion, the architecture also had to follow the topography of the locations selected. A long and narrow ground plan suited building on a steep slope and space could be increased by adding step-like platforms.602 A square or rectangular ground plan suited a spur,603 but if the building was large enough, it could spread along the ridge beyond the spur. On the plateau, it was possible to use the large spaces afforded by the more gentle relief. The Villa dei Quintilii has its main core near the ridge shoulder. Its garden wings stretch outwards from the core along the shoulder and towards the center of the ridge crest. At Sette Bassi, the core is built on a low spur and the protruding hippodrome below 598 Calculations are based on data provided by Adams 2006, Tables 3–13. It should be noted that Adams used all kinds of villas (of the total 49, 39 are included here), but only some of the very large town houses in Pompeii (12) and Herculaneum (6). This probably makes the igures skewed and the comparison only tentative. Two of the town houses had only 10% or less open space and one had more than 50%. 599 E.g., Terrenato 2001; Becker 2005; 2006. 600 Ground plans in Romizzi 2001, site 29, Tav. 29a–b; see also Kirsch 1993. 601 Ground plans in Carandini 1988. 602 E.g., as at most sites on the slopes around Tibur. 603 E.g., Tibur I site 209 (Villa of Quinctilii Vari) or Bovillae site 96 (Villa Centroni). 135 chaptEr 7 it on a lat area.604 The wings, pavilions and long, narrow buildings could mean that more possibilities for exploiting the views around them existed. They had more external walls and often included porticoes. Sometimes the extensions end in small towers and other types of buildings that seem to be designed around the views from them.605 Single rooms could also have a large opening towards the view.606 It is dificult to know how much the purpose of the building located in the countryside inluenced its architecture. If the villa was intended to be a functional farm with little or no amenities, would the landscape and views be needed? Open spaces – courtyards, atria, peristyles – are found in all villas. They were the basic elements of any Roman domestic building, but not commonly used for exploiting the landscape. The small, compact houses rarely feature the porticoed spaces opening outwards that can be seen in the large buildings. In addition, the distribution of the Class 3 sites in the research area points towards no desire, need or, perhaps, possibility for views. The Class 3 sites tend to be located at low elevations (the majority below 100 m a.s.l.) and as a result, have a more limited visibility of their surroundings compared to the Class 1 sites (Plate XV.2). The agricultural activities concentrated on the plateau and the buildings featured both productive and residential elements. The locations where the Class 3 houses were built little differ from those of Class 1 sites as both can be found on spurs and ridge shoulders with open space in front and on the sides. The greatest difference is perhaps the more limited visibility caused by the low elevation of the Class 3 sites. From the point of view of exploiting the landscape in the villa, it is possible to suggest that the spaces could be divided into three categories. The traditional central elements, atrium and peristyle tended to be internal features on the irst loor. The rooms opened into the courtyards, but not necessarily outwards, offering mostly internal views of the villa. The visual axes through the entire building present in, e.g., Pompeian town houses, were rarely seen in villa architecture. It was not necessary to reveal the inside of the house in the countryside,607 but the lines of sight were not commonly used to exploit the external views either. The upper loors probably had more windows, based on the few surviving samples, and any opening from an upper loor would probably have afforded wide external as well as internal views. The outside of the building was often used for porticoes and other open spaces that clearly opened outwards. In addition to the platforms, a portico probably also formed the most visible part of the villa from the outside and it has been suggested that, e.g., in wall paintings, the villa as a building could be symbolized by a portico.608 The openness of the building outwards was created by building porticoes on the façades of the villa, but it did not necessarily mean that the building really was open outwards. Experiencing the view? What was then visible from the buildings, from the windows and porticoes? The answer is relatively simple: surrounding countryside, Rome in the distance, and mountains (particularly 604 Cf. also the Villa cd. ad duas lauros found at Centocelle near Rome, Gioia and Volpe 2004. E.g., the Villa c.d. ad duas lauros at Centocelle near Rome (Pellegrini 1997; Gioia and Volpe 2004) or the Torre di Roccabruna at the Villa Adriana. 606 E.g., the nymphaea Bergantino and Dorico of the Villa of Domitian at Castelgandolfo (Romizzi 2001, site 11, Tav. 11). 607 Cf. Leen 1991; Wallace-Hadrill 1998; Hales 2003, 32–39; Myers 2005. For the signiicance of the villa for selfpromotion, see Bodel 1997. 608 Mantha Zarmakoupi in a paper given at the Archaeological Institute of America’s 2006 Annual Meeting in Montreal, Canada, January 2006. The abstract can be found at the AIA website under 2006 Abstracts, Session 5A: Roman Houses and Villas. See also Thagaard Loft 2004 for not inding evidence of a connection between actual villa buildings and those depicted in wall paintings. 605 136 visibility and viEwability the Alban Hills). Bodies of water could also be seen in many locations, e.g., the small crater lakes and in the major river valleys. The majority of sites would have offered a view of agricultural landscapes lanked by mountains in the distance. The variation desired by the Roman authors could have been created by using internal views of the villa. In some cases, the villas were either purposefully built in – or became part of – a quarry landscape. Some of the villas near the huge quarries of the Aniene tuff were in such locations that the extraction activity could have been one of the central themes of their view. This probably was not always intentional, as in some cases the quarries actually caused the abandonment of the villa.609 Extraction of stone and pozzolana were probably very common practices on the estates and in many cases, the straight cuts of the quarries might resemble the natural erosion of river banks. The quarries were a natural part of the landscape of production in the area, and it can be surmised that overlooking them was not actively avoided when selecting locations for villas. The locations of the sites on ridge shoulders and spurs towards northerly–westerly directions would have meant that the view did not include villas in the immediate vicinity. The distances between individual villas are so great that the question of blocking someone else’s view dealt with in legal cases would not have been very important in the countryside. The differences of elevation in the terrain would also have made that problem unimportant. In the lower regions, the most common sight from a spur or ridge shoulder would have been the adjacent river valley, possibly exploited for intensive agriculture or gardening (Plate XVI.1a). If the villa was built on a steeper slope at a higher elevation, the view would have been almost automatically panoramic and far-reaching (Plate XVI.1b). The sites are most commonly oriented from west to north, which gives a view over the Roman Campagna with Rome in the background. The mountains in that direction are mostly far away and barely visible as a boundary to the landscape. The signiicance of Rome as part of the view can be debated. It has been suggested, e.g., that the early modern villas of the Frascati region were built deliberately in such a way that their facades were towards Rome.610 The reason would have been creating a visual link between the city and the villas, between the center of power and the holders of power or those who were aspiring to grab it. These later villas can be seen from Rome and even recognized based on their locations and appearance. They have been almost invariably (and, again, deliberately) built on top of ancient Roman villas,611 which could lead to the conclusion that the Roman villas were oriented in the same way, and perhaps for the same reason. However, little direct evidence for this interpretation for the Roman period can be found. The literary descriptions of the Roman Campagna are almost always written looking out from Rome, thus offering little help. The platforms in the Frascati region are often built aligned in the general direction of Rome, but then the terrain is such that any building in the area would almost automatically point towards Rome. In addition, so little remains of the buildings on platforms that it is impossible to say what they looked like. What is perhaps more interesting is the elevation of the buildings. The large Class 1 villas in the region of Tusculum, ca. 30 of them, are found above 250 m a.s.l. and also mostly above modern Frascati. A clear gap in the dense distribution of Class 1 villas below that elevation can be seen. The elevation gave the villas a very good visibility of the surrounding area and also towards Rome (Plate XVI.2a). That the greater elevation was selected as a building site, could also point towards the desire for a view of Rome. The situation is very different for the Class 1 villas on the slopes of Tibur: the main visibility is limited to nearby areas (Plate XVI.2b). The distance is too far for seeing Rome properly and placing the villas at higher elevations probably logistically impossible – the very 609 610 611 E.g., Collatia sites 4c and 4d. Franck 1956. Cf. Ehrlich 2002, 55–67; Valenti 2003, passim. 137 chaptEr 7 steep slopes begin right above the ancient buildings. Thus, of the Class 1 site areas, region of modern Frascati is the best suited for viewing Rome. The visual connection between the villa owners and Rome as a center of power could have been desired, but this cannot be wholly veriied. The view was most often enjoyed not directly, but through the portico, limiting the extent of the view both horizontally and vertically. It also controlled the amount of light and protected the room/viewers from the most immediate and extreme effects of the weather. The literary descriptions almost invariably contemplate the view from inside a building, from some room in it. The structures perhaps created a picture frame effect; the landscape could have functioned like a large central image in a wall painting. However, the visible landscape had very little in common with the detailed landscape paintings. Depending on where the villa was located and where the opening was in the villa, the visible landscape would have been quite different. The facade of the building was towards the open landscape, whereas the sides gave on the immediate surroundings. On a high spur or at a tip of ridge, the sides could also have been towards wide open space. At a fairly low elevation, the facade view would have ended relatively soon, showing mostly the surrounding countryside, an agricultural landscape. At higher locations, the view extended dramatically and looking out from a window reclining on a couch would probably have shown a very distant agricultural landscape with maybe some mountains. From certain locations, the view could have been Rome and, from the highest points, probably mostly just sky. The side view could have contained parts of the building itself as well as the surrounding countryside in the immediate vicinity. If the outside of the building was lanked by a portico, walking in the portico would have given a wide panorama of the surroundings. Depending on the height of the location, one would also have been able to see what was immediately in front of the building. If the slope was steep and terraces high, very little would be visible, and vice versa. The main focus of the view was probably in the middle- and long-distance, not in the short-distance.612 The effect of the high location, a long-distance view from the facade area, could be that of isolation. The viewer could perceive him(her)self as being in the only building in the area. The same can also apply to a sea view: the open sea offers little other details to watch apart from the ever-changing sea and sky. The long-distance view of a landscape (as opposed to a seascape) with sky could be perceived as more monotonous, but, nevertheless, similarly isolating. Other parts of the building were centered mostly on itself, on the internal views of courtyards. In the very densely inhabited popular resorts, the long-distance view combined with the internal views created a sense of privacy and isolation that did not necessarily correspond with reality.613 This isolation was, of course, broken by views of other villas and other human occupation of the landscape. If the villa owner so wanted, he/she could also perceive him/herself as part of a community. The viewable villa? The last section of this chapter is about the viewability of the villa in its surroundings, about being visible. A high location means that the building is in a commanding position: the villa controls its surroundings, and it is itself also clearly viewable. Some of the villas built on the Bay of Naples by great generals and war heroes, Marius, Pompey and Caesar, are likened to 612 Cf. Higuchi 1983 for dividing the landscape into ranges. In an analysis of Portuguese villas, a similar feature was noted: the view was extensive, but the intervisibility between the villas was low (Langley 2006, 322). 613 Schneider 1995, 103–104 suggests that the view along with the general villa architecture created a feeling of control and power that was otherwise dificult to achieve during the imperial period. 138 visibility and viEwability fortresses for their locations and layouts.614 The elevated location and large size evoke awe and admiration in the viewers. The buildings might have also reminded one of the town houses in Rome built on the slopes of the Palatine, viewable from below in the Forum Romanum. Cicero makes a point of buying such a house to enhance his political status; his house was part of his public agenda.615 Even though the villa was characteristically less public than the town house, it inluenced viewers in a similar manner: the location and splendor of the building added to the glory of the owner.616 For maximizing the effect of good visibility, inding a visible location in any given area was important. The viewsheds presented above are strikingly similar: the best overall intervisibility is connected to the area roughly between the Via Nomentana and the Via Appia; one area is visible from another and vice versa.617 Entering the area from almost any direction, it was possible to see many parts of it, recognize the Alban Hills and to understand one’s location in the landscape easily. In the northwest, the Via Salaria is the westernmost point, but there the Tiber valley is also a formidable gap in the landscape and if the right bank is viewed from the left bank, little is visible apart from the edge of the landscape north and south of Fidenae. In the northeast and southeast, the high slopes of the Apennines and the Alban Hills continuing outside the edges of the research area form a natural limit to the intervisibility. They also form a background for the villas on the lower slopes. In the southwest, very few obvious elements that would limit the intervisibility exist, but despite this, it almost cannot be seen from the other parts of the research area. The Capo di Bove lava ridge is a natural barrier to viewing between the eastern and southwestern areas, but no barriers can be found towards, e.g., the Alban Hills area and yet the southwestern zone is not included in what can be seen from the sites along the Via Appia. In addition, the intervisibility of the southwestern area is poor, which can be seen from the viewsheds of the roads and the settlement sites (Plates XIV.1d; XV.2).618 The area of the dense settlement starting from the Via Appia and continuing all the way to the Tiber also forms a zone with fairly good intervisibility. The dense settlement seems to be concentrated in an area that created a visual entity – the intervisibility could enhance a feeling of community. The signiicance of a good neighborhood in site selection is mentioned many times in the advice given by the Roman agronomists.619 Seeing the prosperous landscape surrounding the site at a glance would give a good idea of the value of the land. In this zone of good intervisibility, the slopes around Frascati are the most commonly visible area, which can be seen from everywhere else, except from the southern edge of the research area. This area was also visible all the way to Rome (Plate XVII.1). Considering this, it is not at all surprising that Tusculum was one of the most popular villa resorts in the vicinity of Rome. A villa in that region ensured maximum viewability from the whole region of Rome as well as from Rome itself. Cicero as a homo novus was consciously creating a political career as well as seeking social status and his choices of the locations of his house in Rome and his villa in the Roman region are telling: a house on the Palatine and villa in Tusculum. 614 Plin. nat. 18,32; Sen. epist. 51. Cic. fam. 5,6,2; dom. 37,100; Att. 1,13,6; off. 1,39,138–140; Gell. 12,12. See also Vell. Pat. 2,14,3 for M. Livius Drusus (tr. pl. 91 BC) and his (possibly the same?) house on the Palatine. Cf. Berg 1997; Hales 2003, 41–44. 616 See Llobera 2001 for an introduction to the concept of topographical prominence, i.e., how the vertical position of the viewer and the area viewed contribute to the viewing experience. 617 This naturally depends partly on the extent of the research area, the DEM and the locations of the viewing points included. 618 It should be recalled that the southwestern area is quite possibly very poorly surveyed. However, the viewsheds discussed are calculated from roads, not just from settlement sites, so the basic visibility does not change even if there were more sites. What would change in that case is the interpretation of the distribution of settlement. 619 Cato agr. 1,2; 1,4; Colum. 1,3,3; Varro rust. 1,16,1; 1,16,6. 615 139 chaptEr 7 The second popular resort, Tibur, was further away from the city and only the highest slopes of the Apennine range are visible from the research area as well as from Rome. The villas were built below this zone of best viewability. This makes the effect of a villa there local and less powerful universally. The Via Tiburtina was an important route to the northeast and to the Apennines, and it served as a transhumance route.620 The temple of Hercules Victor was another site often visited in Tibur. A villa in Tibur was visible, but it probably did not offer as high a social prestige as one in Tusculum. The political and social prestige that once could be acquired by a Tusculanum was perhaps not as necessary in the new imperial political climate as it had been before. This could explain why Tibur is mentioned in literary sources in connection to villa habitation until Late Antiquity and Tusculum becomes less important starting from the 2nd century AD. This is supported by archaeological evidence in the villas in Tusculum, which show relatively few signs of late construction.621 The areas with the best viewability were fairly limited and these feature dense Class 1 occupation. The way these areas were seen when, e.g., approaching them on the main roads, would create a formidable sight of a community of the rich and powerful in Roman society. The owners of the villas and sometimes even their ownership histories were well known, as shown by some discussions, especially of Cicero’s Tusculanum. He was disapproved for owning a villa that had once been the property of the Lutatii Catuli; a family deemed more honorable than the new man Cicero. A similar story is told of Lucius Cornelius Balbus (cos. suff. 40 BC) and his Tusculanum.622 Cicero is reported to have learned the locations of his colleagues’ houses and villas as part of his skills as a professional politician.623 Cicero’s sentences on the society of Tusculum are also interesting considering what persons he mentions. Dozens of important senators of mostly consular rank are referred to as villa owners, but only one knight and one freedman and they remain anonymous.624 For Cicero, Tusculum signiied Roman aristocratic society extending outside the capital and his own aspirations as part of that society. The other landowners knew who their neighbors were and travelers passing by were also vaguely aware of who were part of this glorious community when they saw the “magniicently devised royal palaces.”625 The area with the best viewability is limited to the higher slopes, but another zone of many Class 1 sites existed below on the plateau. Comparing the distribution to the viewsheds shows that the sites are located in some of the best spots for viewability on the plateau. In addition, a strong intervisibility between this area and the slopes around Tusculum can be found (Plate XV.2a) – the plateau sites fall roughly into the middle-distance view from the higher villas. A strong visual connection between the higher and lower sites exists, perhaps adding to the prestige of the lower sites. The hills and the villas in the distance functioned as an impressive background for the lower buildings, forming a strong association between the two. The situation is not repeated in the Tiburtine region because of the very different topography. Instead of gently undulating ridges, a lat travertine area is found below Tibur. This area does not feature the terrain preferred for villas, although a number of sites have been built there. Most of the Class 1 sites are, however, located on the elevated regions, many of them also on the western parts of the ridges and hills and not necessarily in sight of the villas on the Tiburtine slopes. This could point towards the different roles these two villa resorts 620 E.g., Bonetto 1999. Valenti 2003, 61–64. 622 Cic. Att. 4,5,2; cf. Balb. 56. The villa could have also belonged to Lucius Cornelius Sulla (cos. 88 BC) based on Plin. nat. 22,6,12, but Cicero himself never mentions Sulla in connection to his villa. Balbus in Cic. Balb. 56. 623 Plut. Vit. Cic. 7. Cf. also Rhet.Her. 4,50,63 of a man pretending to be richer and more famous than he is and expecting “everyone” to know where his house is. 624 Cf. catalogue in Shatzman 1976. The knight and freedman in Cic. leg. 3,13,30. 625 Strab. 5,3,12; translation H. L. Jones (Loeb Classical Library). 621 140 visibility and viEwability had: Tusculum deinitely had a more public character, whereas Tibur was more private and secluded. The owners of the archaeologically known villas can be rarely identiied with certainty, but the imperial buildings are the notable exception to this. In the Roman region, three sites are known where the owner (and the builder) most certainly was the emperor: Hadrian’s Villa Adriana in Tibur, Nero’s Tusculanum in Frascati and Domitian’s villa on Lago di Albano in the Alban Hills. Their locations are also all interesting when visibility and viewability are concerned. The earliest site is located in Tusculum, which had its greatest loruit in the Late Republican and Early Imperial period. The imperial ownership of Agrippina Minor, Nero’s mother, and later Nero himself is attested in literature as well as by stamps on lead water tubes.626 Various members of the imperial family also owned other sites in the vicinity, e.g., the Emperor Sulpicius Galba and Matidia Augusta, Trajan’s niece or her daughter.627 The Frascati site was located in the lower part of the best intervisibility zone and most of the Class 1 sites were above it (Plate XVI.2a). When the area is viewed from Rome or from the countryside below, Frascati lies in the foreground, below all the other buildings. It was not visible from the villas above and around, but it was a focal point of the whole landscape. The Villa Adriana is located outside and below the greatest concentration of Tiburtine Class 1 sites. The huge building plan required a site in the plateau area – the steep slopes above would not have allowed such an enormous complex. But the location was again well selected with viewability in mind: all the sites above it would have had the imperial villa in their middledistance view. In addition, the site was also visible on the plateau area. Domitian’s villa on the rim of Lago di Albano is exceptional compared to the other two sites, as it is not surrounded by a dense villa concentration. The Alban Hills were popular, but no such focal points as Tibur or Tusculum could be found there. The modern town of Castelgandolfo and the papal villa on top of the imperial villa ruin are located on the rim of the crater so that they are visible from the Roman Campagna and Rome. It is possible that Domitian’s villa was also visible in a similar manner. The imperial villas in the Bay of Naples area can be mentioned here as a comparison. The known sites were on the islands of Capri and Ischia outside the main habitation, isolated but in plain sight of the villas located on the mainland. In each case, the imperial presence in major villa resorts in accentuated by the viewability of its location and it is hard to believe that they were selected by chance. The last question to be discussed is what the villas looked like. Today, most of the sites are reduced to either artifact/building debris scatters in the ields or undecorated terraces. The few reconstructions of the villas give an idea of the imposing character of the buildings.628 They were built on elevated locations on high terraces placed on top of each other and the terraces were also often decorated with painted wall plaster and architectural features such as niches.629 The outside of the building and the façade was often formed by a portico. The comparison to royal palaces offered by Strabo is apt, but the comparison also suits the Hellenistic temple sites known from the Roman region, especially those of Hercules Victor in Tibur and Fortuna Primigenia in Praeneste.630 A poem by Juvenal describing the achievements of one Cretonius/ Centronius in villa construction compares the villas to the sacred buildings of Fortuna and Tac. ann. 14,3; CIL XIV 2659 = XV 7853; CIL XV 7817. Galba: Tusculum site 93–99; Suet. Galba 4,3 and 18,2; CIL XIV 2737; Granino Cecere 2000. Matidia Augusta: Tusculum site 397–409; CIL XV 7822. 628 E.g., Förtsch 1993, passim; Lafon 2001a, passim. 629 E.g., Tusculum site 397–409. 630 See Coarelli 1987 for the temples. The façade portico of the villa at Loc. Santa Maria on the western side of Lake Nemi was modelled after the temple of Diana on the northern shore of the lake, forming a strong visual bond between the temple and the villa; Guldager Bilde 2004. 626 627 141 chaptEr 7 Hercules.631 Another comparison offered for the villas of Marius, Caesar and Pompey in Baiae is to a military camp. They were built near mountain tops and even the style of the buildings resembled camps.632 The commanding position offered the visibility and viewability required for a military camp intended to intimidate the enemy or of a temple intended to impress the worshippers as well as to glorify the deity. The visible villa and its architecture created an allusion to worldly or divine power that was probably well understood by viewers. The visibility mostly concerned one part of the building, the façade. The villas were mainly located away from the main roads, connected to these by secondary roads. The entrance sides were often utilitarian courtyards, and no direct connection to the private spaces of the villa was offered. The view of the magniicent façade was the most public one, accentuating the owner’s wealth and social status. It could also be perceived as protecting the privacy of the building. The architecture allowed few glimpses inside the building and the location could even create confusion as to how to enter the villa. The entrance side could be closed and unfriendly, offering no clues to the opulence of the building. For the owner, a good view of the surrounding area provided the possibility to control it and to see those approaching; the actual, more secluded entrance secured the privacy (and safety) of the owner. 7.6 conclusions Visibility and viewability were important issues when sites for large, probably residential buildings were chosen. The emphasis was on gaining a middle- and/or long-distance view, mostly towards the Roman Campagna and Rome. The smaller and poorer sites could enjoy beautiful scenery, but it was then limited to short- and middle-distances. Views and visibility also worked in the context of the ambiguity of the Roman villa. It was mostly considered a private building, a place for isolation and peace. However, most of the popular locations for villas resembled more Rome in the density of settlement and the social competition of the inhabitants. A villa visibly part of a high society resort served the owner’s social and political aspirations. Being in the villa and views both internal and external created the sense of isolation and privacy. 631 Iuv. 14,86–95. Sen. epist. 51. This can, of course, be merely part of Seneca’s moral rhetoric concerning those individuals and their position in Roman history, but it could also be an accurate description. 632 142 8 roads, towns, villagEs and roMan villas 8.1 background The last aspect to be studied is the relationship between villas, transportation routes, most importantly roads, and habitation centers. In addition to these, another issue to be discussed is neighborhood or community. A loosely related subject is also land division or centuriation, which is discussed in the last section. The objects of study differ from previous ones in the respect that they are all man-made and no environmental material is used; although, e.g., the DEM will be used in the last section. Roman agronomists advised prospective landowners to look at the physical aspects of a plot of land, e.g., fertile soil and good water sources, but other characteristics needed to be thought about as well. Agricultural and other production were intended to provide selfsuficiency for the owner, but selling potential surplus was equally important. It was also necessary to bring in supplies, as no unit could be completely self-reliant. Transporting the products to and from the farm needed to be easy and preferably cheap. For this purpose, having a farm near a road or a navigable stream was preferable, and transportation by sea is also mentioned.633 Good connectivity also made it easy for the landowner to travel to the farm often and manage its activities personally.634 On the other hand, a major road could cause disruption to daily life and agricultural activities, as well as attract more visitors looking for lodgings and entertainment. Some passages suggest villas should be some distance from a major roadway, especially if this was a major military route.635 The road network also offered good visibility for villas and tombs erected along the main roads or at important crossroads and could be used for promoting the family of the landowner or the deceased.636 A town or other market for selling and buying products in the vicinity was also deemed important when placing a villa.637 The farm should be located near the owner’s town house for frequent and easy visits. A town or even a small village provided not only a market, but also other services. The population of the town could be hired as free laborers for the harvest and other times of urgent need for extra hands.638 Pliny the Younger describes the vicus near his 633 For transportation needs and costs, see Laurence 1998, 130–136. Cato agr. 1,3; Colum. 1,3,3; Varro rust. 1,16,1–2; 1,16,6. Transportation by sea: Colum. 1,2,3. 634 Colum. 1,2,1; 1,3,3. 635 Colum. 1,5,6–7. 636 One of the most important instances of this discourse in the ancient literature is Cicero’s search for a burial site for his daughter, Tullia. For discussions on the topic, see, e.g., Hesberg 1992, 5–18; Bodel 1997, 18–26; Griesbach 2005. 637 Cato agr. 1,3. 638 Cato agr. 1,3. Ikeguchi 2000 offers a brief analysis of the use of labor, slave and free, and how it could be seen in archaeological data. chaptEr 8 Laurentine villa as capable of illing anyone’s modest needs.639 Horace writes to Vala about visiting Velia and the region of Salernum, asking questions on the condition of the roads as well as on how well the towns could provide decent foodstuff.640 The habitation center could also provide other entertainment.641 The landowner could participate in the local political and religious life to enhance his political and economic connections, even if his main political career was conducted in Rome. The local inhabitants could also be bothersome if relaxation and peace were wanted.642 The most popular watering places, such as the Bay of Naples or the Roman Campagna, were crowded with Roman aristocrats and social life could be very similar to that of Rome.643 Few of the Romans had kinship or other ties to the towns and territories, perhaps to ensure a peaceful stay. Like a major road, a town could be useful, but also a troublesome distraction. The neighborhood of the villa was also considered important when assessing a plot. A prosperous and well-managed area was preferred,644 as it boded well for one’s own farm. Bad maintenance of neighboring lands could harm the adjacent properties.645 The safety of the area was also considered important.646 A good neighbor could be relied to lend a hand or supplies or whatever necessary in times of need. Reciprocity was also recommended: it was important to be a good neighbor.647 Visits to neighbors formed a signiicant part of country life and thus acquiring a place near friends was probably regarded as important, even if not directly advised by the agronomists.648 The connections between the transportation routes, habitation centers and villas have been studied as part of the economy of Roman agriculture. Land transport has often been regarded as too expensive compared to using sea and other water routes, but despite this roads, had to be used for transporting people and goods to reach most parts of Italy.649 Land and water routes probably complemented each other rather than competed with each other. The chronological association of road building and spread of the villa as a center of productive activities is also interesting. The road network built across Italy from the 3rd–2nd century BC onwards enhanced the opportunities for successful commercial agriculture even in fairly remote areas. It has also been observed that the villas were placed at a distance from the roads as well as from the towns or villages. This way, the villa remained independent in appearance, separated from habitation centers as well as from major arteries of land transportation. Despite the separation, it was fully integrated into the landscape and the economy of the countryside, the town and Roman Italy.650 639 Plin. epist. 2,17,26. Hor. epist. 1,15. 641 E.g., the library at the temple of Hercules Victor at Tibur is mentioned in Gell. 9,14,3 and 19,5,4. Baths were also common in towns and villages and served their surroundings as well, see Chapter 5. See also, e.g., D’Arms 1970, 55–61, 142–152 for the cultural life on the Bay of Naples. 642 Plin. epist. 9,15 on trouble with local farmers. 643 E.g., Cic. Att. 5,2; Plin. epist. 5,6,45. Cf. D’Arms 1970; Champlin 1982. 644 Cato agr. 1,2; 1,4; Colum. 1,3,5–7; Varro rust. 1,16,1. 645 Varro rust. 1,16,6. 646 Varro rust. 1,16,1. 647 Cato agr. 4,1. 648 The descriptions of villa life recorded in the letters of Cicero, Pliny the Younger and Symmachus testify to the importance and longevity of this aspect of the country life. 649 The importance of roads is also accentuated by a recent analysis on various factors affecting river transport on the Tiber; Graham 2005. The results also suggest limitations to river transport that could have made using roads more cost-effective even for moving such bulky materials as bricks and tiles. 650 Laurence 1998, 138–143; 1999, 95–108; Goodchild 2007, 166–174. Some survey publications discuss distributions of settlement in the countryside in relation to towns, e.g., Yntema 1993. 640 144 roads, towns and villagEs The relationship between villas and neighborhoods has not been studied in great detail.651 Landowners of senatorial rank have been studied and the data collected is also related to the question of neighborhoods, although this issue has not been addressed directly.652 Archaeological data has rarely been used on these occasions, perhaps due to the problems of connecting written sources, even epigraphic ones, to archaeological remains. 8.2 writtEn sourcEs on habitation cEntErs and transportation The Roman Campagna is crossed by most of the main roads in central Italy: from the north clockwise Salaria, Nomentana, Tiburtina, Collatina, Praenestina(/Gabina), Labicana, Latina, Appia, Ardeatina. The history and development of the road network follows fairly closely the expansion of Rome in the Apennine peninsula.653 Most of the early roads were named after their destination: Gabina lead to Gabii, Nomentana to Nomentum, etc. The Via Salaria, which derives its name from the Latin for “salt,” is often presumed to be one of the oldest routes in the area, following the river valley and forming an important trade route from the mouth of the Tiber to the Sabine country and other inland areas. The Via Latina is presumably another old road following a natural route southeast to the Alban Hills and beyond to the lands of the Latin League. The most famous road is the Via Appia, which was built in 312 BC. It was the irst road purposefully built using public funding and the irst to be named after its builder, censor Appius Claudius Caecus. Some minor roads have also been mentioned, like the road connecting the Via Salaria and the Via Nomentana at the fourth milestone, on which the villa of Emperor Nero’s freedman Phaon was located.654 The importance of the roads for military efforts is clear, but the sources say very little about their signiicance for commercial purposes. In fact, dio mentions the Aniene as a major route for transporting building stone to Rome.655 Roads could also offer other services than just a means of transport. Road stations, stationes, mansiones and mutationes featured at regular intervals in the countryside outside Rome.656 The road stations varied quite a lot in what was included in the services provided, but they were often located at crossroads, required a source for water, included buildings for storage, animals, housing, religious activities and entertainment. Villas could also be used as road stations.657 The most important town in the Roman Campagna was, of course, Rome, but a number of smaller habitation centers in and around the research area also existed: Apiolae, Bovillae, Castrimoenium, Collatia, Ficulea, Fidenae, Gabii, Nomentum, Pedum, Tellenae, Tibur, and Tusculum. These are all referred to as towns in ancient literature, but a number of villages, vici and pagi are known based on inscriptions. They probably functioned as secondary centers 651 D’Arms 1970 discussing the development and character of the Roman settlement on the Bay of Naples could be regarded as an exception. 652 Shatzman 1976; Andermahr 1998. 653 See, e.g., RE SXIII s.v. viae publicae romanae (Radke); Wiseman 1970; Coarelli 1988; Laurence 1999, 11–57. As with the towns, the following is not intended as a complete history. More information on the history of roads (including archaeological data) can be found in the relevant survey publications. 654 Suet. Nero 48. 655 5,3,11. 656 Fidenae on the Via Salaria (town?). Aquae Albulae on the Via Tiburtina. Ad Quintanas on the ifteenth milestone of the Via Labicana. Ad Decimum on the tenth milestone of the Via Latina (village?). Aricia (modern Ariccia) on the sixteenth milestone of the Via Appia. List based on Geogr. Rav., Guidonis Geographica, Itin. Ant. and Tabula Peutengiarina. 657 Corsi 2000, 20–78. 145 chaptEr 8 in larger town territories.658 The towns and villages have varying histories, and some are known very well, Tibur and Tusculum particularly.659 Most have a mythological story of foundation660 and had an independent existence in the early periods as members of the Latin peoples with their own politics. Rome’s expansion in its surroundings changed the status of many towns or even destroyed them.661 Some of the old towns reduced in size and power, even taken over by private citizens’ estates.662Rome’s interests concentrated irst in the north and extended then to the area east of it. Tusculum became a municipium after the Latin Wars of 381 BC being thus one of the irst towns ever to receive the grant.663 In comparison, Tibur retained an independent position until the 1st century BC, only becoming a municipium after the Social War in 90 BC.664 During the Imperial period, epigraphic sources mention civil servants in many towns, indicating the existence of a local government. Literary sources reveal relatively little about the towns and villages apart from their histories, some of their institutions and administrative bodies. Most of the information comes from inscriptions.665 Most towns featured temples or sacred places where various deities were worshiped. Hercules was perhaps one of the more popular of the gods with a major temple at Tibur and many smaller ones in most parts of the research area.666 Other major temples in Vicus Licinianus on the Via Tiburtina, CIL VI 9871 = (?) I.It. IV,1 168. Vicus near the tenth milestone and between the Via Latina and the Via Labicana, EE IX 685. Vicus on the seventh milestone of the Via Latina, CIL VI 1324. Vicus Angusculanus in Tusculum, AE 1906, 79. Vicus Sulpicius at the eighth milestone of the Via Appia, Chiofi 1999, 56–60. Vicus on the Laurentine coast, Plin. epist. 2,17,26. Pagus Mandelae near Digentia, Hor. epist. 1,18,104–106 and Schol. Hor. epist. 1,18,105. Pagus Ulmanus and pagus Transulmanus Pelectanus in the territory of Ficulea, CIL XIV 4012. Pagus Amentinus minor at the Via Appia near the fourth or ifth milestone, Chiofi 1999, 56–60. Pagus Amentinus maior?, at the eighth or ninth milestone of the Via Appia, Chiofi 1999, 56–60. Res Publica Decimiensis or the village of Ad Decimum at the tenth milestone of the Via Latina, Itin. Ant. 305, CIL XIV 4229 = CIL XV 7811. Tarpin 2002; Todisco 2004; Hernández Martínez 2006; 2007, 138–177. 659 The following is not intended as a thorough history of the towns in the region, but only a very brief overview. Each town’s history is presented in the relevant survey publication, in entries in RE as well as in Mayer 2005. See also Cornell 1995, particularly 293–326 and 369–398. 660 E.g., Tibur, Praeneste, Gabii, Tusculum, Tellenae and possibly Bovillae: Diod. Sic. 7,5,9; Origo Rom. chron. 17,6. 661 E.g., Apiolae, Tellenae and Mugilla are not mentioned after the 5th century BC. 662 Bovillae and Gabii: Cic. Planc. 23; Schol. Bob. Cic. Planc. 23. Fidenae and Gabii: Hor. epist. 1,11,7–8; Iuv. 3,190–192; 10,99–102; Lucan. 7,391–399; Prop. 4,1,33–36 (also Bovillae). Fidenae: Strab. 5,3,2; Plin. nat. 3,69. Labicum: Strab. 5,3,9. 663 Liv. 6,26,8. 338 BC: Liv. 8,14. 664 App. BCiv. 1,65. 665 See also Jouffroy 1986 for a study of public building in Italy. 666 Hercules Victor at Tibur: e.g., Gell. 19,5,4; Strab. 5,3,11; Suet. Aug. 72,2. A large number of inscriptions mentioning Hercules Victor are known and the relevant entries in CIL XIV and I.It. IV,1 should be consulted; see also Coarelli 1987, 85–112. Other occurrences of Hercules: at Bovillae (CIL XIV 2426), Castromoenium (CIL XIV 2455–2456), in the Collatia territory (CIL VI 325, CIL VI 341 = CIL XIV 278, 3905), Ficulea (Vicario 1976, 84 Nr. 8), Fidenae (CIL XIV 4056), Gabii (CIL XIV 2788, 2789). The following are some of the deities, probably from public shrines in the area. Bona Dea: at Aefula (CIL XIV 3530 = I.It. IV,1 611; also Granino Cecere 1992, 132–140) and Tusculum (EE IX 698). Sol: possibly at Ficulea (CIL XIV 3568 = I.It. IV,1 68, could also be from Rome) and at Tibur (CIL XIV 3567 = I.It. IV,1 67; CIL XIV 3568 = I.It. IV,1 68). Fortuna: at Ficulea (CIL XIV 4002 with Victoria ), Tibur (CIL XIV 3539 = I.It. IV,1 41) and Tusculum (CIL XIV 2577). Isis: at Ficulea (AE 1995, 37), Tibur (CIL XIV 3633 = I.It. IV,1 169) and Tusculum (CIL XIV 2589). Diana/Minerva: at Tibur (CIL XIV 3536 = I.It. IV,1 3; CIL XIV 3537 = I.It. IV,1 7; CIL XIV 3570 = I.It. IV,1 65) and Tusculum (Valenti 2003, 117 Nr. 54; CIL XIV 2495–2495a, 2633). Juno: at Tibur (CIL XIV 3693 = I.It. IV,1 232; CIL XIV 3556 = I.It. IV,1 61). Juppiter: at Tibur (CIL XIV 3555 = I.It. IV,1 60; CIL XIV 3557 = I.It. IV,1 58; CIL XIV 3559 = I.It. IV,1 59; CIL XIV 3586 = I.It. IV,1 99) and Tusculum (Liv. 27,4,11; CIL XIV 2562, 2579). Venus: at Tibur (CIL XIV 3569 = I.It. IV,1 71; I.It. IV,1 70) and Tusculum (CIL XIV 2584). Augustales: Bovillae (EE IX 679, CIL XIV 2388–2404, 2405, 2406, 2412), Tibur (again numerous sources, CIL and I.It. IV,1 should be consulted), Tusculum (CIL XIV 2620, 2637). Imperial cult: Bovillae (Paribeni 1926, 206), Ficulea (CIL VI 764), Gabii (CIL XIV 2794–2800), Tibur (CIL XIV 3575 = I.It. IV,1 75; CIL XIV 3576 = I.It. IV,1 76; CIL XIV 3577 = I.It. IV,1 79; CIL XIV 3578 = I.It. IV,1 80; CIL XIV 3579a = I.It. IV,1 78; CIL XIV 3580 = I.It. IV,1 81; I.It. IV,1 32, 74), Tusculum (AE 1914, 54; CIL XIV 2496a, 2497; EE IX 970 = CIL VI 31563c; CIL XIV 2591–2597). See also Lega 1995 for deities protecting agricultural work. 658 146 roads, towns and villagEs the area were those of Juno at Gabii and Fortuna Primigenia at Praeneste.667 In addition to the temples and shrines, little else is known of public buildings. Walls and gates are mentioned in some sources.668 Theaters, amphitheaters and circuses are also sometimes referred to.669 The major villa resorts, Tibur and Tusculum, feature only a few records of possible entertainment activities. Walls, temples and shrines were most commonly built during the Republican and early Imperial era, whereas in the later periods, public building concentrated more on such entertainment buildings as baths, theaters and amphitheaters.670 The concept of neighborhood is the last issue to be discussed in this chapter and it is probably the most dificult to grasp. No direct references to what was thought of the Roman Campagna as a neighborhood can be found.671 Some of the aspects of a good neighborhood, such as afluence and well-tended farms and plantations, may be seen in more general descriptions of the area.672 Another way to approach the problem is to look at the people who have been active in the area, e.g., as landowners, at different times and try to see what they perceived “neighborhood” to be.673 Cicero describes decorating his own villa, visits to neighboring villas and discussions on philosophy in their rooms and gardens. His friends in Tusculum were the same as in Rome, the society Cicero participated in did not change when he moved out of the capital. No locals are mentioned and apart from a few exceptions, all persons mentioned by name are Romans of senatorial rank. Pliny the Younger gives a hint that the situation has continued in a similar way since he complains how stays in Tibur or Tusculum require more formal dress and manners than living at his more remote Tuscan villa.674 The Late Antique references to Tibur in the letters of Symmachus also give a relatively similar picture of villa life in the Roman Campagna, of mostly Romans sharing their leisure time with each other even in the countryside. Moreover, the poets offer similar views of their activities in the area.675 It has been suggested that the suburbium of Rome would have been mostly free of obligations and client relationships to the landowners coming from the capital.676 Very few landowners had their origins in the area and very few held any ofices in the towns. Families from Tusculum, such as the Porcii Catones or the Fulvii, already participated in Roman politics in the 3rd century BC, and were thus perhaps considered more Romans than originating in Tusculum. Tiburtine families became more important in state politics only in the late 1st century BC and their loruit lasted only until the end of 1st century AD. One exceptional ofice in this respect is the curator fani of the temple of Hercules at Tibur, an ofice held also by some Roman senators. Some town patroni are known during the Imperial era, but even added together, these two do not form a very high percentage of all known senators or 667 Gabii: Verg. Aen. 7,681–685 and, e.g., Jiménes Salvador 1981 or Coarelli 1987, 11–21. Praeneste: see, e.g., Coarelli 1987, 35–84. The deity of the large temple at Tusculum has been debated for a long time, but currently the most probable suggestion is Castores, who are featured both in literary and epigraphic sources (Quilici and Quilici Gigli 1995, 533–534). 668 Castrimoenium: CIL XIV 2466. Bovillae: AE 1991, 389. Aefula: Liv. 26,9,9 (211 BC). Gabii: Liv. 24,10,9 (214 BC). 669 Bovillae: circus, Tac. ann. 2,41; 15,23; theater, CIL XIV 2408. Fidenae: Oros. hist. 7,4,11; Suet. Cal. 31; Suet. Tib. 40; Tac. ann. 4,62–63. Amphitheater at Tibur: CIL XIV 4259 = I.It. IV,1 202. 670 Jouffroy 1986; Lomas 1997. 671 Dyson 1992 is an attempt to comprehend what community (and neighborhood?) could have meant in different parts of Roman Italy. 672 E.g., Strab. 5,3,12 concerning the region of Tusculum gives an idea of a very prosperous and well-to-do region. 673 Published lists of landowners in, e.g., Grossi Gondi 1901 (Tusculum); Shatzman 1976; Mari 1991 (Tibur); Andermahr 1998; Valenti 2003 (Tusculum). 674 Plin. epist. 5,6,45. 675 Mayer 2005, 149–219. 676 Champlin 1982; also Mayer 2005. 147 chaptEr 8 knights. The Roman elite probably had various subgroups based on connections by descent, marriage, political views, ambitions, etc. It has been suggested that the occurrence of many senators of Spanish origin in the Tiburtine area during the 1st and 2nd centuries AD was not a coincidence, but a conscious choice on their part.677 The local elite must also have possessed some land and wealth in order to be able to be magistrates in their towns. It could be that these two landowning groups did not mingle very much, at least not in ways that left any records. There were many different groups of people living and being active in the area and they created their own communities and neighborhoods, which were geographically located in the same region, but did not necessarily have very much to do with each other. 8.3 archaEological EvidEncE for roads, villagEs and towns The archaeologically known road network in the Roman Campagna is very dense, consisting of the main roads as well as of a great number of secondary and tertiary routes (Fig. 8.1).678 The routes of the main roads have also been plotted based on archaeological remains and aerial photographs as well as burials built along them. Most of the main roads have been located with fairly accurately and many of the secondary routes are also relatively well known. Excavations have also given some indication of the ages of the roads and the regularization of the network has been dated to the 4th century BC in the eastern part of the area, although its inception is of greater age. It has been suggested that until the 3rd century BC, the routes followed natural passages between points of interest, e.g., between habitation centers, to the main rivers, to the sea, but later military needs required straighter, faster and easier routes. The general perception of a Roman road in Italy is that of a straight and wide consular road paved with large stone blocks, but new excavations have revealed many unpaved roads cut into the subsoil. These needed to be periodically renewed to get rid of the ruts in the road surface and this has resulted sometimes in very deep road cuts. The paved roads tend to remain closer to the modern ground surface.679 The archaeology of the towns in the Roman Campagna offers some supplements to the written sources.680 Most of the towns known by name have been located and recognized with certainty based Fig. 8.1 Road networks in the research area on epigraphic evidence.681 (Plate XVII.2.) with bridges and crossroads. 677 Syme 1983. The network presented in the survey publications has been used here. Some problematic points in the border zones of some survey areas can be pointed out, particularly between Collatia, Tusculum and Bovillae, where the continuation of some roads remains uncertain. I assigned the classiication of secondary and tertiary routes and based it on the length of each road and its perceived importance in its surrounding. 679 Pracchia 2001, 259–270. 680 This section is based on the survey publications of Bovillae, Collatia, Fidenae, Ficulea, Mugilla, Tellenae, and Tibur. 681 Certain locations: Bovillae, Gabii, Tibur and Tusculum. Fairly certain: Fidenae and Tellenae (Moltesen 1978; 678 148 roads, towns and villagEs Few of the towns have been properly excavated and most are under dense modern settlement. Thus, possible public buildings are sually relatively poorly known. Some of the towns feature only remains of town walls and other fortiications682 and possible votive deposits of shrines or temples used mostly during the Archaic, Early and Middle Republican periods.683 Some of the old town sites feature later villas, perhaps in correspondence with written sources stating that some of the old, abandoned towns were taken over as private properties.684 The towns with the most epigraphic evidence from the Imperial period tend to be also archaeologically well known. Tibur, Tusculum, Bovillae and Gabii685 are all in this category, although Gabii is poorly preserved compared to the other three. Tibur and Tusculum are located on high hills with control over a main road: the Via Tiburtina (and the Aniene) and the Via Latina, respectively. Both also have a possible arx, a small castle-like part of the town. Bovillae is located on the lower slopes of the Alban Hills, not on a high hill of its own, but at the crossroads of the Via Appia and a road leading to the sea. Gabii is located on the lattish eastern shore of Lago di Castiglione, on the Via Praenestina. Remains of walls and/ or fortiications have been found in Tibur, Tusculum and Bovillae. Three of the towns feature a large temple whose deities have been identiied based on written sources: Hercules Victor in Tibur, Juno Gabina in Gabii and possibly Castores in Tusculum. Other temples are also known in all three towns. No certain archaeological remains of a temple have been found at Bovillae. Remains of a forum with other public administrative buildings have been found from all except Bovillae. Tibur and Tusculum also have an amphitheater on the outskirts of the town center, whereas Bovillae features a large circus. The temples in Gabii and Tibur included theaters, which was found in connection to the forum in Tusculum. The theater in Bovillae was possibly associated with the circus. Various structures connected to water supply (cisterns, aqueducts, nymphaea, wells) have also been found in all the towns excepting Gabii. Building stone was also quarried in the town areas of Gabii and Tusculum.686 In addition, a number of smaller villages have also been identiied archaeologically. Very few of the villages have been excavated and what constitutes a village in the Roman Campagna is not very well known. The early, Archaic and pre-Late Republican, villages tend to be large sites featuring signs of settlement and a necropolis in the vicinity. Sometimes the presence of a village is suggested based on great density of settlement around some area.687 1980; 1988; Moltesen and Brandt 1994). Uncertain: Castrimoenium, thought to be the modern town of Marino, but no archaeological remains are known (Daicovici 1930). Ficulea has been located in the hills around the Casale Marco Simone Vecchio, but this has been recently challenged and a location a few kilometers westwards, between Casale Coazzo and Casale Capobianco has been suggested based on dense Archaic settlement and written sources (Pantano 2001). Collatia has been located at the castle of Lunghezza, but a new site has been proposed in the zone of La Rustica, closer to Rome. This is based on results of new excavations both at Lunghezza and at La Rustica. (Musco 2001, 191–192.) Mugilla has also been located in the research area, but so little is known of it from all possible sources, that no real certainty exists. 682 Fidenae, Ficulea, maybe also Collatia. Aefula has been connected to the remains of walls and terraces running on the hilltops south of Tibur. 683 Fidenae, Collatia. Votive deposits: Bouma 1996, 37–38 Nr. 67 (Fidenae), Bouma 1996, 42 Nr. 77 (Tellenae). Votive deposits at the sites suggested to have been Apiolae and Politorium have also been found: Bouma 1996, 60 Nr. 97 (Castel Savello = A) and 33 Nr. 55 (Castel di Decima = P). At Aefula, a votive deposit dating from the Archaic period to the 4th century BC was found at Monte S. Angelo in Arcese at the southern end of the complex (Bouma 1996, 102 Nr. 120). 684 Politorium, Tellenae; Strab. 5,3,2. 685 Tusculum: Quilici and Quilici Gigli 1991; Coarelli 1993; Bouma 1996, 103–104 Nr. 124; Duprè 2000. Gabii: Coarelli 1993; Melis and Vardaro 1993; Bouma 1996, 40–41 Nr. 73. 686 Fidenae also needs to be mentioned in this connection. 687 In the Fidenae and Ficulea areas, three such regions have been suggested where a village might have existed: in Fidenae at the crossroads of the modern Via delle Vigne Nuove and the road leading from Fidenae towards the southeast; in Ficulea near the modern Casale Coazzo and Casale Capobianco and at the crossroads of the modern Via di Tor Giovanni. In the Collatia area, the following sites could be centers for a number of smaller settlements around them: 26 (La Rustica), 73 (Salone), 189 and 194 (Tor Angela), 333 (Osteria dell’Osa), 529 (Torraccio S. Antonio), 149 chaptEr 8 In some cases, some structures, e.g., remains of fortiications, have been found.688 These sites vanish after the Early Republican period and the later settlement is established nearby, but not directly on the old site. Based on the distribution of the early settlements, some of the routes can be determined to have been used in early times in the northern and eastern areas. The main emphasis of settlement in the latter region seems to be on the banks of the Aniene.The surroundings of bodies of water and springs were popular locations for early villages, as there is dense settlement around Lago di Castiglione and the Pantano Secco basin.689 The later villages are established in new locations, very commonly on the main roads. Some of them have also functioned as shrines as Middle/Late Republican votive deposits have been found.690 The most well-known sites are the Res Publica Decimiensis or Ad Decimum on the Via Latina and Ad Nonum on the Via Praenestina. Both sites were in use for a very long period of time starting from the Middle Republican period through Late Antiquity. Both are also multifunctional sites: they were habitation centers, road stations and shrines. In addition, large cemeteries have been found in the vicinity and Ad Decimum features Christian catacombs from the 4th century AD. A somewhat similar center could be Aquae Albulae developed on the Via Tiburtina around sulphurous springs. In addition to shrines and baths, the site probably functioned as a road station.691 Apart from shrines, tombs and cemeteries, few other structures have been identiied. More ambiguous remains of villages are found by the roads, most commonly near major crossroads. The road stations possibly attracted settlement and small villages developed around them. The irst stations tended to be four to six miles from Rome and after that, the next stops were at nine to eleven miles distance. (Plate XVII.2.) The Via Salaria and the Via Nomentana have no stations and this could point to some of the roads having been better serviced or then to gaps in archaeological knowledge. In some cases, the station was built on a natural stopping place on the journey, e.g., by the bridges. Some sites were also located on crossroads of important routes, but in some cases, the distance from Rome might have been the decisive factor for the placement. Towns and centers developed around other services probably also functioned as stations, e.g., Fidenae and Aquae Albulae. Two villas in and near the research area were also used as road stations: the one attributed to Gallienus near the ninth milestone of the Via Appia and the recently excavated complex at Torre Fiscale near the fourth milestone of the Via Latina.692 It is also possible that many other structures along the roads were used as oficial, semi-oficial or private stops, such as small shrines or the semi-public baths sometimes found in the countryside. These could also attract settlement around them even when located further away from the routes. The last issue in this context, the neighborhood or community, could be perhaps considered to be beyond archaeological evidence. A community, although connected often to a physical place, is very much a mental construct, not necessarily linked to structures, but rather to the people inhabiting them. Some of the elements of a good neighborhood, such as afluence, can also be gauged archaeologically. The distribution of different settlement sites can give an indication of where different communities/neighborhoods existed: large and luxurious villas in areas used more for seasonal relaxation and small units with agricultural 553 (Finocchio). 688 Ponte Mammolo, Collatia site 1 = Ficulea site 593; Osteria delle Molette, Ficulea site 289 = Tibur III site 178. Possibly also Fidenae site 177. 689 Villages have also been identiied outside the research area on the Solfotara sulphurous lake (Apiolae site 409) and by the small crater at Castel Savello (Apiolae site 85). 690 Bovillae site 121 (Ad Decimum on the Via Latina): Bouma 1996, 120 Nr. 146. Collatia site 224 (Ad Nonum on the Via Praenestina): Bouma 1996, 120–121 Nr. 147. 691 Tibur III sites 339–340. For use as a road station, see Corsi 2000, 87. 692 Corsi 2000; Rea 2003, 205–214; Spera and Mineo 2004; Donda 2006. 150 roads, towns and villagEs remains in more rural communities destined for permanent settlement. Afluence can also be seen in the amount of precious materials used. Almost half of the Class 1–3 sites feature marble on walls, loors, architectural elements or other decorations. Opus sectile has been found in almost 10% and sculpture in almost 20% of these sites. Mosaics are more common as 30% of the sites have remains of them. Excavated data matches these proportions from surveys, although mosaics and opus sectile are even more common (70% and 30% respectively).693 Excavations outside the buildings have also given an idea of how intensely the area was used agriculturally. Another way the people connected to the area were burials and inscriptions found with them. Tombs were commonly built on visible sites near the roads and their density can tell something about the signiicance of the road. The main roads, all but the Via Ardeatina, were lanked by a great number of separate burials and cemeteries of varying sizes. (Plate XVIII.1.) Plotting the single tombs and cemeteries on the same map with the roads shows that in addition to the main roads, two routes were important in their areas: that of the modern Via delle Vigne Nuove and the one from Torraccio dell’Inviolata to modern Guidonia, both in the northern part of the research area. Little research has been conducted on the distribution of burials of individuals, but studying the distribution of inscriptions could be used to trace where individuals of differing social status buried their dead and whether changes can be seen in selection of places over time. The relationship between villas and burials, however, has been studied and some chronological differences were noted. In the Late Republican and Early Imperial periods, the monumental tombs were placed along main roads, probably separately from the deceased’s possible estates. Viewability helped to commemorate the dead, but also promoted the deceased and his/her family. From the 2nd century AD onwards, tombs began to be built on the estates, even in direct connection to the villas. A new spiritual mentality emphasizing private grief and commemoration could have been one reason for this, but changes in politics had also occurred when the emperor seized political power and personal/ familial promotion became unnecessary or even dangerous. It has also been suggested that space along the roads could merely have run out in the densely populated Roman Campagna and tombs needed to be built elsewhere, i.e., on the actual estates.694 8.4 intEgrating thE EvidEncE Near a good road, but not directly on one? The main roads would probably have attracted the most trafic and would have been the kind of a road whose proximity should have been sought out and at the same time avoided by not building directly by the road. The nine main roads in the research area all originate from Rome and thus create the familiar fan-like distribution – all roads lead to (or from) Rome. The secondary and tertiary road network is dense, almost every ridge has its own road, and thus all villas lie within just a few minutes’ walking distance of a good secondary route that led to one of the main roads. In addition to the roads, only one water route could be used: the Aniene 693 De Franceschini 2005, 363–375. Griesbach 2000; 2005; Di Gennaro and Griesbach 2003; Di Gennaro et al. 2004. The last comment was made by Dott.ssa Rita Volpe, Sovraintendenza ai Beni Culturali Roma, on a paper by Claudio Borgognoni titled “Sepoltura in villa vel in horto. L’associazione tra architettura residenziale e architettura funeraria a Roma e nel suburbio. Problemi, metodologie, linee di ricerca.” at the Koninklijk Nederlands Instituut Rome in Rome, June 19th, 2006. Borgognoni’s results were mostly the same as those presented by Griesbach on earlier occasions. 694 151 chaptEr 8 Fig. 8.2 Distribution of settlement sites compared to a 500 m wide buffer zone around the main roads. a) Classes 1 (white) and 2 (black). b) Classes 3 (white) and 4 (black). which, according to literary evidence, was navigable during ancient times. It was certainly used for transporting building stone from the tuff quarries on the river and even further away. Only one villa with a connection to the river has been reported,695 but more such villas could have existed. The most luxurious sites tend to be slightly further away from the river which could mean that the river was not important to them. The research area is an inland zone and road transport was more important than water transport. The entry points of the main roads into the research area are 2–4 km apart and the distance between the roads grows the further away they extend from Rome. Almost 80% of the ground surface in the research area was within 2 km distance of one main road, i.e., within a 20–30 minute walk depending on the terrain and the need to go around estates. As most sites are consequently located very close to at least one main road, I did not consider necessary to study the distance from the road lines. Instead, the areas closest to the main roads were deemed more interesting to see whether the roads attracted settlement to its immediate vicinity. A 500 m buffer zone was created around the main roads, and these buffer zones cover 21% of the research area (Fig. 8.2a–b). The distribution of Class 1 to 3 sites was plotted against the 500 m buffer zones and 28% of the settlement occurred in them, i.e., the areas closest to the main roads could be considered to attract settlement to some extent. Of the Class 4 sites, on the other hand, only 20% were near the main roads, showing thus no clear connection to them. As the main roads often run perpendicular to ridges and the sites are most commonly along the ridge shoulders, the main routes did not offer that many suitable locations for villas and could not attract very dense settlement along them. The gaps between the 2 km wide buffer zones around the main roads were checked separately (Fig. 8.3). The largest gaps occur in the northern area between the Via Salaria and the Via Nomentana, in the northeast between the Via Nomentana and the Via Tiburtina, in the eastern area between the Aniene and the Via Praenestina and in the southwestern area west of the Via Ardeatina. These zones feature few Class 1 sites apart from the area north of the Via Tiburtina, where the general lack of a main road seems not to have prevented the building of large villas. Two secondary routes in the northern gap zones, the Via delle Vigne Nuove and the Inviolata–Guidonia road, were of great importance as shown by the tombs built along 695 Loc. S. Eusebio, probably Collatia site 57; cf. Calci and Mari 2003, 184. 152 roads, towns and villagEs Fig. 8.3 Distribution of settlement sites compared to a 2 km wide buffer zone around the main roads. Secondary roads also indicated. a) Classes 1 (white) and 2 (black).b) Classes 3 (white) and 4 (black). them. They were the main arteries for the settlement of the area and little settlement can be found beyond their immediate surroundings. The densest Class 1 and 2 site distribution occurs further south, between the Via Praenestina and the Via Labicana. In the southwestern region, the Class 1 and 2 sites are located almost exclusively between the Via Ardeatina and the Via Appia. The large number of major and minor roads made almost the whole area easily accessible by foot or by any vehicle. The gap zones between the main roads cover only a small section of the Roman Campagna and they were not necessarily selected for building the largest villas. In some areas, the lack of a main road seems to be of some importance (northwest, southwest), but in others, secondary routes were suficient to attract even the richest settlement (north–northeast, the Aniene valley). Crossroads between main and secondary routes as well as bridges were plotted on a map (Plate XVIII.2), and a 500 m wide buffer zone was drawn around crossroads. The crossroad zones cover 20% of the research area and the greatest number of them occur in the northwestern and eastern areas where many west–east and north–south routes meet. When Class 1–3 sites were plotted against the crossroads zones, 30% of them were found to be located totally or partially in these. The vicinity of a major crossroad was somewhat attractive to settlement, but not all crossroads were equally popular. Those in the southwestern area as well as many between the Via Salaria and the Via Nomentana feature no settlement sites in or around them. The most densely settled crossroad areas can be found on the Via Tiburtina, the Via Praenestina and the Via Labicana matching some of the densest settled areas in general. At some of these crossroads are also known villages or road stations.696 Centuriations? Centuriation was used for allocating land and it could inluence site selection: depending on how the land division was conducted, the resulting plots might or might not include sites 696 Ponte Mammolo and Settecamini on the Via Tiburtina; Muraccio dell’Olmo, Tor Angela and Ponte di Nona on the Via Praenestina; Giardinetti on the Via Labicana; Villa Senni on the Via Latina; and Casale il Palombaro on the Via Appia. 153 chaptEr 8 regarded as suitable for building. Three land divisions in the Late Republican and Early Imperial periods known from ancient literature occurred in the research area and the town territories affected were Bovillae, Castrimoenium and Tusculum, all centuriated during the Sullan period.697 The Campi Tiberiani, possibly located somewhere between Rome and Tibur, were centuriated during the reign of Emperor Tiberius.698 These are all very late considering the age of the dispersed settlement in the Roman region and very little is known of possible earlier land divisions. Rome conquered its surroundings and some changes in the landholding patterns might have occurred, old inhabitants moved out and new ones moved in, but no unambiguous references to land divisions are made in the Roman history books before the 1st century BC.699 Road lines in addition to other archaeological remains have been used to reconstruct the centuriations (Plate XIX.1). Land division results in similarly oriented structures in the countryside, such as roads, water channels/ditches or even buildings themselves. They have been recognized from aerial photographs in northern and central Italy and often the centuriation can also be connected to archaeological remains.700 In many centuriated areas, the land division represents a fairly easily recognizable event horizon in the archaeological record, the beginning of the Roman occupation in a new area. In this respect, the Roman region is quite different. By the time of the centuriations, the area had been inhabited by Romans for centuries and the last division also coincides with the greatest settlement density. The centuriated regions were the ones which had remained sparsely inhabited until the 1st century BC, i.e., the central area, and the centuriation could be regarded as an active measure to occupy the last open spaces in the area. However, despite the sparse settlement, noting a new, different habitation phase might be dificult – the old features in the landscape might have been used as a basis for the centuriation.701 If this happened, it remains open how the old structures could be differentiated from the new ones. The reconstructions suggested for the research area are based on modern topographical maps and archaelogical inds. The dates of the sites have also been taken into consideration when evaluating the suitability of the orientations, but similar care has not perhaps been taken in the initial interpretation.702 Most of the base lines for the divisions (Plate XIX.1) are roads running on ridge crests, the natural lines of movement in the area. Many of the roads might have been used from very early periods onwards and thus they probably preceded the centuriations in many cases. The sites built after the centuriation period and cited as following the divisions also follow the general orientations of the terrain. Considering the evidence for the whole area on the dependence of the site orientation on its surroundings, it is dificult to know whether the centuriation or the general topography of the area was the reason even 697 Lib.col. (Lachmann) II p.254,5–9; p.255,28–29; p.255,28–29. Archaeological evidence presented in Chouquer et al. 1987, 92–98, 285–288, Figs. 2–4. Centuriation in general is a huge topic and the following discussion is necessarily too brief to cover all aspects. Some recent discussions of the centuriation in the Roman area (particularly in Hernández Martínez 2007) inspired these comments. 698 See note 528 on the location of the Campi Tiberiani. 699 Some of the towns were possibly destroyed, indicating that their inhabitants were either killed or transferred somewhere else. The land was found for the gens Claudia in the 6th century BC without problems (e.g., Liv. 2,16,3– 5) and the plebs of Rome occupied the sacer mons in the same region equally easily slightly later (e.g., Liv. 2,32,2–5). Some further suggestions can be found in Hernández Martínez 2007, 177–220. 700 E.g., Chouquer et al. 1987; Campbell 1996; 2000; Schubert 1996. 701 Campbell 1996, 83–84. Cf. also Caravello and Giacomin 1993 for an explanation of centuriation divisions by using ecological sustainability as a main argument. 702 The Campi Tiberiani reconstruction has been criticized: Quilici 1994; LTURS s.v. Tiberiani campi (Mari); Campbell in his commentary on Liber Coloniarum (Campbell 2000). Chouquer and Favory 1999 is a reply to Quilici 1994. Using the Aqua Alexandrina built in the 2nd or 3rd century AD as evidence is suspicious because it could not have followed a centuriation but the terrain in order to maintain the gravity low. Cf. Hernández Martínez 2007, 321–362 for accepting the divisions and presenting additional evidence. 154 roads, towns and villagEs at these sites. The suggested orientations of the centuriations also match neatly the general topographical orientation of the area, particularly on the Collatia–Gabii region.703 The number of sites matching the orientation diminishes towards the west and the east where the direction of the ridges changes slightly. It should also be noted that only a little evidence perpendicular to the division orientations can be presented. The land divisions noted in ancient sources probably did happen, but inding evidence for them is more dificult to discover than what has been suggested. Towns, villages and road stations? The Roman region has an almost regular distribution of small towns and villages (Plate XVII.2). Rome was naturally the center, but around it the irst towns were located within 7 to 15 km in the north, but in all other directions, only villages are known up to 10 to 15 km from Rome. Most of the towns are beyond the 15 km radius. Rome dominates its immediate surroundings and the other towns probably had fairly small territories compared to the capital. Even the furthest of the towns in the research area, Tibur, could be reached within a half a day’s travel.704 An estate would have been quite near a town or a village almost everywhere in the area. Dispersed settlement has been the main settlement pattern in the Roman Campagna from the 7th–6th century BC onwards.705 Changes in geographical and chronological distribution patterns occur, but the small, separate settlement sites exist in large numbers up to Late Antiquity. The relationship between the dispersed settlement and the towns on the left bank of the Tiber is currently interpreted in the following manner. Before the 7th century BC, all settlement concentrated in different centers: pre/proto-urban centers (such as Rome), smaller centers and villages. The onset of dispersed settlement is explained by the emerging aristocracy settling its lands by allocating farms to family members, friends and supporters. In the Archaic and Early Republican periods, the farms sites are assumed to gravitate around the habitation centers. The political independence of towns before Roman rule required economic independence, a town territory with agricultural production. The fast spread of dispersed settlement stops in the 5th century BC and is followed by a slower increase in numbers involving abandonment of old sites and establishment of new ones in slightly different places. This is connected to changes in the pattern of property holding. The aristocracy possibly acquired most of the lands and this led to the civil unrest between patricians and plebeians known from literary sources in the Early and Middle Republican periods. This period sees also few habitation centers apart from the main towns in the area. With the advancement of Rome’s dominion in its surroundings, the signiicance of the towns diminishes drastically and some even vanish completely. Rome becomes the most important town and the dispersed settlement gravitates around the capital. Small town territories did exist, but they probably had more administrative rather than political or economic signiicance. The period of greatest unrest and wars in the area already ends in the late 4th century BC, although the Second Punic War and Hannibal’s campaign in the area might have caused 703 Chouquer et al. 1987, Fig. 104. E.g., Laurence 1999, 81–82. 705 The following paragraph, particularly for the earlier periods, is based mainly on Carafa 2000 and 2004, which use the most complete material available for the region north of Rome. The references to earlier studies are based on syntheses presented in the survey publications, particularly those on Fidenae and Ficulea. See also Patterson et al. 2004 and Di Giuseppe 2005 for a similar analysis of the situation on the right bank of the Tiber. The pattern is similar to the left bank: dense settlement in the Archaic period, then a reduction until the Late Republican/Early Imperial period, when the settlement reaches its peak. More luctuation can be noted in the Etrurian material in the Early/Middle Republican settlement than in the Latin area, where the tendency is a slight, steady rise in site numbers. 704 155 chaptEr 8 Fig. 8.4 Distribution of settlement sites compared to habitation centers (towns in dark grey, villages in light grey). a) Archaic, b) Early Republican, c) Middle Republican and d) 2nd century BC. disruption in the late 3rd century BC. The events of the 1st century BC might have also had an effect on the life in the Roman Campagna. The next great increase in site numbers occurs in the 1st century AD, the period of greatest settlement density. Little is said of the relationship between farms/villas and habitation centers apart from the connection between the luxurious villas and the main resort towns, Tibur and Tusculum. The long peace of the Imperial period ceased in the 4th century AD, when Rome was also losing its position as the only center of power. The dispersed settlement seems to disappear during the following centuries and most habitation is assumed to concentrate in the towns. Some problems occur regarding the analysis of settlement patterns of the Archaic and Early Republican periods. The locations of all the early towns in the Roman region are not known and the survey publications of the southern part lack information on possible pre-2nd century BC sites. In the research area, the densest Archaic settlement can be found in the northern part between Fidenae and Ficulea. In the eastern region, most of the Archaic sites are 156 roads, towns and villagEs Fig. 8.5 Distribution of votive deposits (large grey), habitation centers (black) and crossroads (white). located south of the Via Praenestina near Gabii.706 (Fig. 8.4a.) The Early Republican period site distribution features relatively little changes in the northern part, but the habitation around the Via Praenestina more or less disappears (Fig. 8.4b). The Middle Republican period sees settlement return to the eastern area, but this time a little bit further south, on the lower slopes of the Alban Hills (Fig. 8.4c). In the 2nd century BC, the slopes above 110–120 m a.s.l. get settled, but the plateau and the southwestern zone are more sparsely inhabited (Fig. 8.4d). Apart from the main towns, only a few villages are known to have existed in the area from the Archaic to the Late Republican era. During the same period, rustic shrines and votive deposits were common in central Italy and these could indicate points of congregation for people in the area. Approximately twenty votive deposits and twenty other sanctuaries are known in the research area and they are distributed relatively evenly, apart from the northern and southwestern regions, where few or none are known (Fig. 8.5).707 In general, it cannot be 706 Neither of the proposed sites for Collatia, La Rustica or Lunghezza, seems to attract much habitation around them – here it should be remembered that the new, only partially published results of excavations around La Rustica change this distribution pattern. 707 The list of votive deposits is based on Bouma 1996 and supplemented with other sites found in the survey publications. 157 chaptEr 8 said that farm sites concentrate around the shrines. The northern zone is also an exception as plenty of sites have been found, but no shrines or votive deposits.708 Some of the sites with votive deposits are later known as villages and/or towns, e.g., Bovillae, Ad Nonum on the Via Appia, Ad Decimum on the Via Latina and Ad Nonum on the Via Praenestina. The shrines connected to main roads perhaps developed in places where stopping was natural, e.g., at crossroads and bridges, and then villages formed around them with services for travelers. Some of the votive deposits have also been found in later villas, which could be interpreted in two ways. On one hand, the shrine and the villa could be separate entities and the later buildings covered an old shrine site by chance. On the other hand, in light of the recent discoveries in the Auditorium Villa in Rome, the villa could have had a private shrine, which could also have functioned as a shrine for the community around, it in its early phases.709 During the Late Republican and Imperial periods, the towns still functioning were Bovillae, Castrimoenium, Fidenae, Gabii, Tibur, Tusculum and possibly Ficulea (Plate XVII.2). The paciication of the relationships between the powerful hill towns, Tusculum and Tibur, and Rome opened their areas for villa habitation from the 2nd century BC onwards based on both archaeological and written sources. The majority of sites around these two towns are of Class 1 or 2, itting their descriptions as villa resorts. The known villages, ca. 20 of them, are located by the main roads and commonly also functioned as road stations. Although some of the villages are uncertain, a fairly regular network of small centers existed in the plateau with fewer sites in the northeastern and nothing in the southwestern zones. The villages or road stations do not seem to attract much settlement around them, but the dispersed settlement of the Late Republican and Imperial era is extremely dense. Minor concentrations around Fidenae and Ad Nonum on the Via Appia can be seen, but the proximity of towns or villages could hardly have played a major part in selecting building sites. Commercial agriculture and any surplus products could easily be sold either locally710 or in Rome due to the short distances and good transport routes. Neighborhoods and communities? A neighborhood consists of people in the same geographically deined space. Although people, their actions and perceptions may be the deining factors in creating a neighborhood, the concrete space and its character inluence the outcome.711 Space creates physical borders with river valleys or mountain ranges, it offers a varying range of resources for agriculture and other economic activity, it inluences design and construction of buildings. Many of the aspects of the environment are not necessarily ever contemplated consciously by people experiencing and using them every day, but the environment does inluence their behavior in many ways.712 The different environments found in the research area together with archaeological and written sources can be used to hypothesize how the Roman region could have been perceived by its inhabitants and visitors. The two main topographical components are the plateau and the slopes lanking it in the northeast and southeast. The areas also differ climatically as the higher areas are cooler. The plateau is in general better suited for agriculture than the slope areas, partially due to good 708 This is also interesting when the location of Ficulea is discussed. Most of the other early centers feature votive inds or deposits, but there seems to be none from either of the suggested locations for Ficulea. 709 D’Alessio and Di Giuseppe 2005. 710 Cf. Morley 1996 for markets in the surroundings of Rome and Witcher 2005a for the demography of the Roman hinterland and its economic and political possibilities. 711 E.g., Relph 1976, 33–36; Norberg-Schulz 1980; Stedman 2003. For the concept of space and place in general, see, e.g., Cresswell 2005, particularly pp. 1–14. 712 E.g., Appleton 1975 for landscape. See also references in note 711. 158 roads, towns and villagEs geology and water resources. The majority of the plateau villas are probably combinations of residences and productive parts displaying a certain amount of wealth as indicated by frequent use of marble, wall paintings, as well as baths. It is not possible to determine whether the inhabitants owned their lands or whether they were tenants since neither written sources nor archaeological material can answer that question deinitively. Some dependency, tenancy particularly, has been suggested based mostly on density of settlement and possibly topography, which would render large landholdings dificult or even impossible to manage.713 The slope areas belong mostly to the territories of Tibur and Tusculum and the villas in their areas are of Classes 1 and 2, being often large and luxurious. Most of the written sources are connected to these two areas, citing a formidable array of powerful Romans as landowners and visitors. These two areas, the plateau and the slopes, could be interpreted as two different neighborhoods: the slopes reserved for luxurious relaxation and the lower zones for agricultural production, particularly in the area north of the Aniene. Tibur and Tusculum combine many of the important aspects for selecting a site for a residential villa. They offer a slightly cooler climate than the plateau, spectacular views, space for large buildings as well as the unique possibility of being viewable in a very central area of the empire. They were also close to the capital thus making it possible to withdraw from the negotium of the city to the otium of the villa quite rapidly.714 Compared to the towns within a 30 km radius from Rome, these two had also survived the conquest of Rome well and were active towns with good infrastructure during the Late Republican and Imperial times. The fame and popularity of Tibur and Tusculum were celebrated widely in Imperial poetry; Tibur especially seems to have been the favorite of the poets as Catullus, Horace, Juvenal, Martial and Statius all write about it. What impact the villas and their inhabitants had on the towns is more dificult to judge as the written sources record only the visitor’s point of view. Few of the persons connected to the area functioned in any oficial capacity or left any visible signs of their involvement in town life. The same applies to the Bay of Naples, where the local politics were also irmly in the hands of the locals.715 The plateau can be divided into different subzones, perhaps indicating variation in the neighborhoods. The northern part between the Tiber and the travertine area is marked by a fairly dense settlement with most Class 1 and 2 sites gravitating towards the main roads. Single burials and monumental tombs tend to be located near the road lines, but a great number of cemeteries of various sizes can also be found (Plate XVIII.1). Cemeteries are mostly lacking in the rest of the research area.716 Few villages have been recorded in the area, but Fidenae, the two pagi by the Via Nomentana and possibly Ficulea served as centers. Two baths built by private individuals could be used by the public. The frequent road stations on the Via Tiburtina as well as by the main bridges on the Aniene probably also served the local settlement. The many inds connected to agricultural activity combined with excellent soils further accentuate the impression of a mainly agricultural neighborhood. The relatively poor connectivity with Rome, the frequent small cemeteries as well as poor viewability all point towards an area away from the main resort areas concentrating on productive activities and local life. The eastern part of the research area below the slopes of Tusculum also features plenty of agricultural remains, particularly in the zones further away from Rome (Plate IV.1). The soils are only mediocre, but the proximity to Rome probably made this area worth exploiting 713 Messineo 2005; Witcher 2006a, 115–118. Cf. Champlin 1982; Mayer 2005, 25–30, 158–161. 715 E.g., D’Arms 1970; 1984. 716 This could be the result of less excavation in the eastern and southern zone, as small cemeteries can remain invisible in surface surveys (discussions with Dr. Jochen Griesbach in September 2005 and December 2007). 714 159 chaptEr 8 in any case. The general character of the area is similar to the northern zone, but with a slightly different tone. The Class 1 and 2 sites in this area tend to be large and are most abundant between the Via Praenestina and the Via Latina, in the area of the best intervisibility with the villas on the slopes of Tusculum. Three main roads, the Via Praenestina, the Via Labicana and the Via Latina made this a very heavily traficked zone. Tombs are located by the main roads exploiting the viewability to many passers-by. They are mostly single tombs and monumental ones with only a few cemeteries. The known villages are also by the roads and the only town is Gabii. The connections to Rome are excellent, and there is a strong visual connection between the sites here and the villa resort above. The eastern plateau was connected to Rome, as well as possibly to Tusculum, and served perhaps as a second best choice for those who could not afford or otherwise acquire a villa in Tusculum. The eastern zone coincided roughly with the ancient ager Pupiniensis, which had a bad reputation as poor agricultural land as well as being unhealthy.717 The distribution of settlement, its density and richness give an indication that during the Early Imperial period, this reputation did not bother the landowners. It has to be pointed out that this density is relatively late in the eastern area, only coinciding with the greatest saturation of settlement in all of the Roman Campagna. The area south of the Via Latina is very different compared to the northern and eastern plateau zones. The terrain is partially lat which was unappealing for building villas. The settlement concentrates on the areas above 100 m a.s.l. as well as on the high ridges. Bovillae and Castrimoenium were both functioning towns, although little is known of the latter. Despite its obvious importance, judging by the burials lanking it, the Via Appia did not attract much settlement. It is possible that the lava ridge on which it was built offered poorer possibilities for agricultural activities and the Via Appia might have been too close for comfort on the narrow ridge. The area southwest of the Via Appia is the most sparsely inhabited in the entire research area. Although the low density might be partially due to lack of ieldwork, some aspects make it different from the rest of the area. Agriculturally, the southwestern area is as good as the northern zone, but little evidence of activity has been found. The only main road, the Via Ardeatina, did not attract many burials indicating its relative unimportance as a transport route. In addition, no villages, towns nor road stations are known except closer to the Tiber, west of the research area.718 Topographically this is also possibly not a very suitable area for villas: the ridges are formed of small hillocks and the commonly used spurs are relatively rare. These ridges with no lat tops also functioned poorly as routes of transport towards the Tiber. Most habitation, particularly Class 1 and 2 sites, concentrated in the region between the Via Ardeatina and the Via Appia, where most burials are also located. The most striking difference, however, is the lack of intervisibility in the southwestern area (Plates X.2, XIII.2, XIV.1, XV.1). The cumulative viewshed calculated from the main roads (Plate XIV.2a) shows how the plateau and the slopes west of Tusculum are part of an area of good intervisibility starting from the Tiber and ending at the Via Appia. The lava ridge on which the Via Appia is built forms a partial viewing barrier, but the best visibility from the Via Appia is towards its northeastern and eastern side, not towards the west or southwest. The intervisibility inside the southwestern area is also relatively poor compared to the rest of the research area: the zone of best visibility from the Via Ardeatina is limited compared to the other roads. In many ways, the southwestern corner did not offer the same, almost ideal conditions for villa construction as the rest of the research area and it will be interesting to see whether new ieldwork can raise its settlement density to the same levels as in the north and east. 717 718 E.g., Cic. leg. agr. 2,96; Colum. 1,4,2–3; Val. Max. 4,4,6. E.g., Il Torrino and Magliana Vecchia; Bedini 1984; Corsi 2000, 121–122. 160 roads, towns and villagEs The area of good intervisibility between the Tiber and the Via Appia coincides with the densest settlement in the research area. It is also the same area that can be seen from Rome: Martial’s description of the view from Julius Martialis’s villa on the Janiculum719 roughly covers the same area. It is also worth noting that the Gianicolo (80 m a.s.l.) and Monte Mario (135 m a.s.l.) hills effectively cut the view from east to west as the hills of Rome remain generally below the height of 60 m a.s.l. This forms a strong visual bond between Rome and the area east of it. The intervisibility might have been part of the creation of a neighborhood: whenever a panoramic vista opens up in the eastern Roman Campagna, one can see roughly the same landmark slopes lanking the area as well as the city of Rome. The dense settlement and apparently intense agricultural activity are connected to the same region, the heavily traficked roads enhancing the impression of a busy and afluent area. Variation inside the area can be found, but the overall impression is similar for most of the zone of best intervisibility. 8.5 conclusions Roads for easy movement and transportation of goods were an important aspect when selecting a location for a country estate. The Roman Campagna was very well connected to the capital with a dense network of roads. The area was also very near to the city as it could be reached within a half a day’s journey even from the furthest points of the Roman Campagna. Easy access and markets were at anyone’s reach in the area. Only the southwestern corner of the research area remains somewhat isolated from the main roads and the habitation centers. Analysis of the distribution of different types of burials along the main roads and elsewhere in the research area combined with the visibility analysis indicates that the human activity varied in the area. The strong intervisibility between the Via Salaria and the Via Appia corresponds with the most settlement and the most burials – particularly those by the main roads intended to be seen by travelers. This area could be perceived as the main neighborhood in the countryside east of Rome with the southwestern corner remaining outside. The archaeology of the plateau indicates its rustic nature; an intensively cultivated countryside with dense settlement. The slopes above the plateau, on the other hand, were spotted with large and luxurious villas viewable from most of the area. Even if the inhabitants of either area did not intend to create communities, their surroundings and how they were used gave obvious clues to the passers-by of who the inhabitants were and what their inancial status might have been. 719 4,64. The description in the text starts, however, from Saxa Rubra on the right bank of the Tiber and concentrates on the northern section of the landscape, but it mentions the whole of Rome and it surroundings to the Alban Hills. 161 9 finding thE idEal location The ancient Roman Campagna was perceived as an uninhabited and malaria-ridden region until the late 20th century when archaeological surveys had “repopulated” the landscape. The work conducted on the right bank of the Tiber provides an opportunity to analyze and discuss a fairly large area and its settlement patterns. The archaeological material is mostly lawed and lacking data in many respects, but placing it into a wider context, archaeological, geographical and environmental, can make the results more reliable. Studying site selection over a long period of time gives insights into the ways Romans perceived their environment as well as, e.g., into economic development. The various aspects related to selection of location have so far been discussed as separate issues, but they are connected to each other often in many ways, e.g., geology is the basis of soil formation which also depends on terrain and water. The aim of this inal chapter is to combine the various recommended qualities for an ideal location and to discuss the results with respect to settlement as in the previous chapters. Modelling the ideal location Six aspects of the site environment were discussed in the previous chapters and the model of the ideal location is based on these. Geology was deemed relatively insigniicant for site selection, but good soils were actively sought. Water was not a major problem; if the site did not feature a spring or ground water for digging a well, water could be conducted there. However, dense and rich settlement sites are often located in the areas where the most known springs can be found in the northern zone and in the Alban Hills area. The terrain type selected is also clear: a spur, ridge shoulder or slope opening towards the west–northwest–north. Most of the area was also well connected to the main roads crossing the Roman region as well as near towns and villages. Visibility, both views from as well as viewability of the building itself in its surroundings, was a signiicant factor for some of the largest and richest sites. All in all, nine qualities were used in the model for the ideal location and they are listed and described in Table 9.1. In addition to these, others could also have been used, Table 9.1 Qualities used for modelling e.g., altitude, but this was not speciically the ideal location. mentioned by ancient authors. Geology in Quality Modelled as Image the form of materials suitable for building Geology Formations providing suitable building material Fig. 9.1a is more clearly referred to, although not Soils Best soils for agricultural purposes Plate III.2 included in the most desirable qualities Water Areas at least 100 m away from open water Fig. 9.1b listed for the ideal villa location (Fig. 9.1a). Water Springs with a 1 km wide buffer zone Fig. 9.1c Terrain Areas opening towards east–southeast–south Plate IX.2 It was a useful resource for the estate itself Terrain Almost level to moderately sloping (1–12%) Fig. 9.1d as well as commercially and as such, it was Terrain Ridge shoulder Plate X.1 included in the model. The others have Roads Areas within 2 km of main roads Fig. 8.3 been discussed in the appropriate chapters Centers Areas within 3 km of habitation centers Fig. 9.1e idEal location Fig. 9.1 a) Geological formations used as building materials. b) Areas at a distance greater than 100 m from open water. c) Areas within 1 km of a known spring. d) Almost level to moderately sloping (1–12%) terrain. e) Areas within 3 km of habitation centers. 163 chaptEr 9 and are deined in the same way as in those analyses. The instruction concerning avoiding open water was modelled by creating a 100 m wide buffer zone around all bodies of open water as well as potential marshy areas. The area used in this analysis is what remains outside the buffer zones (Fig. 9.1b). The other quality related to water is springs and a buffer zone of 1 km was created around each known spring (Fig. 9.1c). The distance was chosen arbitrarily, but the resulting map outlines well the areas where springs occur. Moderately sloping was considered to mean a gradient of 1–12% (Fig. 9.1d). Only the habitation centers with a known location were included and they were surrounded with a buffer zone of three km or two Roman miles (Fig. 9.1e). The distance was again a fairly arbitrary choice and an equal buffer zone was made for both towns and villages. The map for best quality soils (Fig. 4.4) is somewhat problematic as it covers only part of the research area, but with the eight other maps, it does not affect the result too much – the slopes on the Alban Hills as well as in the northern part might be better represented if the soil map would cover the entire area. Two of the aspects discussed in the previous chapters are not included at all, i.e., visibility and viewability. They are not listed as desired qualities in the ancient recommendations, except for residential villas. Visibility could have been included in two ways: views from the sites or how the sites themselves were viewable. In both, the map would have covered most of the research area apart from the southwestern corner. In the end, I decided to exclude visibility and viewability as they were of importance only to part of the building projects. The model was created simply by overlaying each thematic map and adding the values in each pixel in each map. Thus, the maximum number of qualities one pixel could have is nine. The resulting map was reclassiied into ive classes from very poor to excellent and the end result can be seen in Plate XIX.2. Almost half of the area features four or ive of the qualities for an ideal location and almost a quarter is either good or excellent. The best areas are distributed fairly evenly in the whole region with a slight emphasis on the zones north of the Aniene and west of Tusculum. The central and southwestern areas remain mainly mediocre or poor with small patches of more favorable zones scattered here and there. Villas in ideal locations The distribution of the settlement sites was then compared to the areas deemed good and excellent. The best quality areas cover 23% of the research area and of all the sites, over 50% are located partially in them (Table 9.2; Plate XX.1). When classes of sites are concerned, only the poorest Class 4 sites are less frequently located in the best areas, but even in their case, the relative amount rises to almost 50%. The same trend can be seen in the chronological break down (Table 9.2; Plate XXI.1): until the 1st century AD, over 60% of the sites are located in the best areas. The sites established during the 1st century AD, the period of maximum density, needed to be located in mediocre or poor quality zones more often than before, because the best spots were already taken. It is also worth noting that the highest percentage of sites in the best areas comes from the sparse Late Antique settlement – only the estates in the best locations were viable. The best quality locations were obviously also sought starting from the irst appearance of dispersed settlement in the area. The distribution of the Archaic, Early and Middle Republican sites is clearly concentrated in the best areas (Plate XXI.1a–c). Although the concentration of early sites in the northern part of the research area is partially due to the intensive recent survey activity, it could also be a relection of its good quality for settlement. It has been suggested that the Early Republican period would have brought about a change in site selection,720 but in 720 Carafa 2004. See also Patterson et al. 2004 for Southern Etruria where the early sites are abandoned and later ones established in new locations. 164 idEal location Table 9.2 Number of sites located in good and excellent areas by class and date. this analysis, no great changes can be seen; on the contrary, almost 70% of the Early Republican sites can be found in the best Site Type Sites Sites % areas. The early periods are characterized All sites 1036 53 by many old sites being abandoned and new Class 1 192 58 ones being established. The reasons for this Class 2 276 60 cannot be determined with certainty, but Class 3 228 51 Class 4 331 47 the general selection criteria for locations Archaic 285 63 were the same all the time. The northern Early Republic 212 69 area features prominently in Roman history Middle Republic 217 60 and many wars were needed before the area 2nd century BC 260 63 came securely under Roman rule. Land 2nd century BC, new sites 116 63 1st century AD 660 56 division related to the Roman occupation 1st century AD, new sites 245 48 could mean that the new settlement sites 4th and 5th century sites 105 73 needed to be built in different locations, but no unambiguous evidence for that kind of activity can be found. Locating the buildings of a new estate in a slightly different location could also be related to trying to avoid the remains of the old one, particularly if these also included burials. In the later periods, the building tends to remain in the same location and the subsequent building phases add to the earlier structures. The difference compared to the earlier sites could also be related to the types of building or building techniques. It might have been easier to start a new house from the beginning than using the old structures. Another period of possible changes in settlement pattern starts from the late 1st century AD and continues through the 2nd century AD. It has been suggested that imperial orders requiring senators to have part of their landed property in Italy would have resulted in new interest in the Roman region and the creation of large estates.721 Imperial rule would have diminished the opportunities for making political careers for the senators and knights and this would have led to moving to the countryside and concentrating on civilized leisure. In the Roman Campagna, this development would have resulted in the large building complexes and estates on the plateau. The villa architecture also diversiies at this point: the building begins to have wings and pavilions sprawling outwards from the main core.722 This kind of building requires more space and more fairly level ground than the traditional, compact house. Construction of these buildings on the steep slopes traditionally favored for the large villas would have been dificult. The 1st century AD is the period of maximum settlement density in the research area and even the most unsuitable locations were used. The known large building complexes on the plateau tend to be dated to the 2nd century AD or later with a possible beginning in the 1st century AD or earlier,723 but the locations were most often already inhabited prior to the expansion of the buildings. Most of the new sites of the 1st century AD were established in the central part of the research area (320 of 502 sites) which had been relatively sparsely inhabited. In addition, the new sites are small and poor Class 4 sites (244 of 502). The 1st and 2nd centuries AD signiied an intense period in the Roman Campagna with plenty of building activity. This could be the result of senatorial interest in the area renewed by legislation, but few signs of the formation of large estates can be seen in the dense settlement. The new settlement in the central area could result from the centuriations known to have been made, but it could be also explained by the general over-population of 721 722 723 Coarelli 1986. Romizzi 2001, 102–119. E.g., Villa of the Quintilii, Sette Bassi, Villa Adriana in the research area. Cf. De Franceschini 2005. 165 chaptEr 9 the area. The distribution map for the best quality areas could also be used as a predictive model for future surveys in the Roman region, particularly in the poorly known southwestern area. The best areas were sought so consistently that using them as a guideline for starting surveys could be an easy way to start.724 The southwestern area still features relatively large amount of open landscape and thus a survey there would be of great importance before the ever expanding modern settlement destroys it for good. 724 The Carta storica archeologica monumentale e paesistica del suburbio e dell’agro romano (1988) features a dozen or slightly more sites interpreted as villas and an equal amount of aree di frammenti ittili in the southwestern zone. These are mostly located directly on or in the vicinity of the best areas. 166 10 conclusion The villa was the basic unit for rural settlement and economy, but it was never one-dimensional in character. The appearance of the villa changed over time, but modest and luxurious country houses probably co-existed from very early periods onwards. The deinition and practical content of “luxurious” also changed over time. The concept of the ideal location for a villa, on the other hand, did not vary very much, but remained the same for centuries. Many of the required qualities are essential for agricultural production and/or for life in general which explains the stability of the ideal. However, some changes were observed in the analysis of some aspects, particularly altitude, visibility and viewability. The 2nd century BC marked the beginning of a certain stability in settlement: many new sites were established, but compared to the earlier periods, few were abandoned in the next centuries. In the 2nd century BC, the settlement also spreads to the higher slopes of the Alban Hills and the pre-Apennines. The sites chosen on the higher slopes were less than ideal, more often mediocre (cf. Plate XX.1a–b).725 The visibility analysis showed that the high locations occupied at this time were, however, the best, taking into consideration both views from the buildings and their viewability. In addition, most of the new sites established in high locations develop into the rich Class 1 and 2 buildings (Plate XX.2). A slightly different concept of a villa was developed in the 2nd century BC and it required a different ideal location, i.e., high, with panoramic views from and offering great viewability for the buildings from the surroundings. The high and visible locations were desirable for the villas whose purpose was mostly to advance the owner’s social status. It has been suggested that the ideal for the high location was adopted along with the architecture from Hellenistic palaces of the eastern Mediterranean, but the Romans did have examples of their own. The houses of the elite in Rome were often located on top of the central hills, particularly the Palatine overlooking the Forum Romanum. The model for the impressive country house could have been a domestic development rather than borrowed from elsewhere. What changes in the 2nd century BC is perhaps the way the villa was exploited for social and political promotion. In ancient literature, the villa is often described as private compared to the public town house, but despite this, it was often morally judged similarly to the domus. A too large and too luxurious villa is as bad as similarly equipped town house. A villa in the right place added to the status of its owner and in Tusculum and Tibur, the villa owners represented the highest political elite. The socially appropriate place to own a villa in the vicinity of Rome was on the slopes of these two towns, as relected in the 1st century BC texts and based on archaeology, the development already started in the previous century. That the country house might have already been large and luxurious before that is shown by the Auditorium Villa, but it started to get used for social and political purposes more commonly only in the 2nd century BC. 725 The soil map does not reach Tusculum or Tibur and in the case of Tusculum might have added positively to the quality of the area. chaptEr 10 The “magniicently devised royal palaces” described by Strabo in the 1st century BC possibly represented the culmination of this development. The Imperial period diminished the political power of the elite and showing off one’s wealth and social position could be potentially dangerous. The villa served as a possible refuge from the dangers of Rome. A high location with long distance views gave an illusion of isolation and safety from the surrounding world although neighbors were situated right beside, above or below. The instructions for inding an ideal location for a Roman estate and its buildings were irst voiced by Cato in the 2nd century BC. The analysis of the aspects and their combination compared to the site distributions starting from the Archaic period shows that what Cato said was already known long before he wrote his book on agriculture. The ideal was probably already created by experience of the environment in the times preceding the Archaic period, from hundreds of years of practicing agriculture in central Italy. Cato and his successors mostly just collected and wrote down what was probably commonly known among the people who lived in the area. The rational arguments behind each recommendation were supplemented by the literary tradition. The longevity of the concept of the ideal location shows that it functioned well in this particular area – the agronomists were also well aware that different environments meant that a different location should be chosen. Many of the basic requirements – need for water and good soils – are universal to any society based on subsistence agriculture. They could be generally applied to almost any environment, but the speciics were always derived from the area. The Romans are sometimes perceived as urban people with little connection to the nature around them. This might have applied to a certain portion of the population in the later periods, but many depended on their natural environment for their livelihood. What was needed environmentally for successful cultivation and animal husbandry could be relatively easily rationalized and explained even to those who did not have personal experience. The use of certain locations to promote one’s social status might have not been equally easily explained or consciously understood, but it was also derived from the centuries of inhabiting the environment typical of the Roman region. 168 appEndiX i sitE cataloguE The names for each site indicate the survey area and the site number in its catalogue. “U” before the number indicates a site not included in the original survey. Each entry comprises a simpliied list of inds from the site except for the water installations and remains of agricultural production which are listed in Appendices III and IV. References are to literature published after the survey report’s publication date. AFF = area dei frammenti ittili; AFC = area dei frammenti ceramici; AME = area di material edilizia; AF = area di frammenti. For translations of the Italian terms used as site headings, see Chapter 2.3. class 1 sitEs Fidenae site 1. Villa?: owner?, platform, cryptoporticus, brick, inscription. Fidenae site 8a. Villa?: owner?, foundations, reticulatum, listatum, marble, tuff, lava, brick, tile, inscription. Pergola et al. 2003, Via Salaria Nr. 48. Fidenae site 31a. Villa: owner?, architectural elements, sculpture, wall plaster, mosaic, spicatum, quadratum, reticulatum, cocciopesto, marble, tuff, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware). Fidenae site 71a. Villa: owner?, living quarters?, architectural elements, mosaic, signinum, spicatum, incertum, reticulatum, marble, travertine, tuff, brick, tile, pottery (African Red Slip), inscription, connected to possible burial at site 71b? Di Gennaro et al. 2002. Fidenae site 78. Villa rustica: foundations, living quarters, garden, architectural elements, wall plaster, mosaic, spicatum, signinum, reticulatum, cocciopesto, marble, travertine, tuff, lint, brick, tile, pottery (African Red Slip, olla perforata, dolium), inscription. Ammannato and Belelli Marchesini 1988, 465–7; De Franceschini 2005, 56–7. Fidenae site 83. Casa rustica/Villa?: 100 x 80 m, platform, living quarters, service areas, quadratum, incertum, reticulatum, tuff, pottery (dolium, amphora), burials?, inscription? Pergola et al. 2003, Via Salaria Nr. 60; De Franceschini 2005, 57–9. Fidenae site 114. Villa: 180 x 70 m, living quarters, architectural elements, sculpture, wall plaster, stucco, mosaic, spicatum, sectile, reticulatum, cocciopesto, marble, tuff, lava, peperino, brick, tile, pottery (black gloss, terra sigillata, African Red Slip C, dolium), glass, inscription. Dell’Era 2002, 254–7; Pergola et al. 2003, Via Nomentana Nr. 111. Fidenae site 121. Villa: platform?, architectural elements, wall plaster, mosaic, spicatum, marble, tuff, brick, pottery (terra sigillata, African Red Slip, amphora), coins, inscription. Di Gennaro et al. 2004, 109; Di Gennaro 2006. Fidenae site 139a. Villa: 180 x 180 m, platform?, living quarters, wall plaster, mosaic, spicatum, incertum, reticulatum, marble, limestone, tuff, lava, brick, tile, pottery (terra sigillata, African Red Slip, plain/cookware, dolium, amphora), connected to cemetery at site 139c and road at site 140, same as Class 2 villa at site 141? De Franceschini 2005, 77–80. Fidenae site 182a. Villa: living quarters, wall plaster, spicatum, reticulatum, listatum, cocciopesto, brick, tile, pottery (terra sigillata, African Red Slip D, dolium, amphora, lamp), inscription. Messineo and Sorella 1990b; Pergola et al. 2003, Via Nomentana Nr. 126; De Franceschini 2005, 81–3. Fidenae site 197. Villa: owner?, 300 m long, enclosed spaces, architectural elements, sculpture, sectile, reticulatum, cocciopesto, marble, travertine, tuff, lava, brick, tile, pottery (terra sigillata, African Red Slip), inscription. Messineo and Sorella 1990a; Pergola et al. 2003, Via Nomentana Nr. 105; De Franceschini 2005, 98–101; De Seña 2005, Nr. 6. Fidenae site 250. Villa: reticulatum, cocciopesto, tuff, peperino, brick, tile, pottery (black gloss, terra sigillata, African Red Slip D, dolium, amphora), sarcophagus, inscription. De Franceschini 2005, 66–7. Ficulea site 75. Villa: owner?, 5,000 m², 3 areas, architectural elements, sculpture, wall plaster, mosaic, sectile, spicatum, signinum, mortar, cocciopesto, marble, travertine, tuff, lava, brick, tile, pottery (terra sigillata, African Red Slip A & C, African cookware, plain/cookware, dolium, amphora), glass, inscription. Pergola et al. 2003, Via Nomentana Nr. 139. Ficulea site 127. Villa: 5,800 m², platform?, architectural elements, wall plaster, mosaic, mortar, cocciopesto, marble, travertine, tuff, tile, pottery (African Red Slip A & D, amphora). Barbina 1998, 314; Pergola et al. 2003, Via Nomentana Nr. 130. Ficulea site 129a. Villa: architectural elements, wall plaster, mosaic, spicatum, quadratum, reticulatum, marble, limestone, tuff, lava, brick, tile, pottery (black gloss, terra sigillata, thin-walled ware, African Red Slip, plain/cookware, dolium, amphora), coins, glass. De Franceschini 2005, 91–3. Ficulea site 138a. Villa: owner?, foundations, enclosed spaces, quadratum, reticulatum, travertine, limestone, lava, brick, tile, pottery, sarcophagus, connected to cemetery at site 138b. Ficulea site 144. Villa: living quarters, service areas, architectural elements, wall plaster, mosaic, sectile, spicatum, quadratum, incertum, quasireticulatum, reticulatum, mixtum, listatum?, vittatum?, cocciopesto, marble, travertine, tuff, brick, tile, pottery (African Red Slip D, plain/cookware, dolium?, amphora, lamp), glass, inscription. Carbonara and Messineo 1995; Sorella 1998, 158–9; Pergola et al. 2003, Via Tiburtina appEndiX i Nr. 228; De Franceschini 2005, 103–7. La Porta 2004, 168. Ficulea site 148a. Villa?/Mansio?: enclosed spaces, architectural elements, sculpture, reticulatum, listatum?, vittatum?, marble, limestone, tuff, brick, tile, pottery (African Red Slip A & D, African cookware, plain/cookware, dolium, amphora), glass. Pergola et al. 2003, Via Nomentana Nr. 110. Ficulea site 292 = Tibur III site 176. Villa: platform, walls?, sculpture, mosaic, signinum, mortar, marble, travertine, limestone, brick, pottery (black gloss, terra sigillata, African Red Slip). Moscetti and La Porta 2004, 171. Ficulea site 152a. Villa: living quarters, architectural elements, wall plaster, mosaic, reticulatum, marble, travertine, lava, brick, tile, pottery (black gloss, terra sigillata, African Red Slip D, dolium). Sorella 1998, 160. Ficulea site 159b. Villa rustica: platform, walls, mosaic, spicatum, quadratum, reticulatum, travertine, limestone, tuff, lava, brick, tile, pottery (impasto, black gloss, terra sigillata, African Red Slip D, African cookware, plain/cookware, dolium). Ficulea site 163a. Villa rustica: foundations, walls, reticulatum, travertine, tuff, tile, pottery (terra sigillata), burial. Pergola et al. 2003, Via Nomentana Nr. 152. Ficulea site 302. Villa?: platform, wall, reticulatum, travertine, limestone, tuff, tile, pottery (terra sigillata, dolium, amphora). Ficulea site 310b. Villa: 30 x 7–8 m, platform, foundations, walls, architectural elements, wall plaster, mosaic, spicatum, reticulatum, marble, travertine, limestone, tuff, peperino, brick/ tile, pottery (black gloss, terra sigillata, African Red Slip). Moscetti and La Porta 2004, 172. Ficulea site 327. Villa: 200 x 100 m, architectural elements, wall plaster, sectile, spicatum, mortar, cocciopesto, marble, limestone, tuff, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium). Ficulea site 172. Villa: 3 areas, living quarters, storage spaces, signinum, quadratum, reticulatum, cocciopesto, tuff, tile, pottery (terra sigillata, African Red Slip, dolium, amphora). Sorella 1998, 170–3; De Franceschini 2005, 93–4. Ficulea site 329a. Villa: foundations, enclosed spaces, architectural elements, wall plaster, mosaic, sectile, spicatum, mixtum, marble, limestone, tuff, brick, tile, pottery (black gloss, thin-walled ware, terra sigillata, African Red Slip, plain/ cookware, dolium). Ficulea site 180b. Villa: owner?, 38,500 m², 2 areas, architectural elements, sculpture, mosaic, marble, limestone, tuff, brick, tile, pottery (impasto, African Red Slip C & D, African cookware, amphora), inscription. Pergola et al. 2003, Via Nomentana Nr. 149. Ficulea site 342b = Tibur III site 47a. Villa: owner?, 80 m long, platform, foundations, architectural elements, sculpture, mosaic, sectile, spicatum, signinum, listatum, marble, travertine, tuff, brick, tile, pottery (black gloss, thin-walled ware, terra sigillata, African Red Slip), votive artifacts, inscription. Ficulea site 183. Villa: 50 x 35 m, platform, foundations, walls, mortar, marble, tuff, brick, tile, pottery (terra sigillata). Pergola et al. 2003, Via Nomentana Nr. 156. Ficulea site 350 = Tibur III site 21. Villa: 35 m long, platforms, architectural elements, wall plaster, spicatum, quadratum, reticulatum, cocciopesto, limestone, tuff, brick, pottery (plain/ cookware, dolium, amphora). Ficulea site 185a. Villa rustica: 45 x 23 m, living quarters, storage spaces, sculpture, sectile, spicatum, quadratum, marble, travertine, peperino, tile, pottery (dolium), inscription. Pergola et al. 2003, Via Nomentana Nr. 157; De Franceschini 2005, 83–5. Ficulea site 187a. Villa: enclosed spaces, sculpture, wall plaster, incertum, reticulatum, cocciopesto, tuff, tile, pottery (terra sigillata, African Red Slip D). Ficulea site 192. Villa: owner?, 55,000 m², living quarters, storage spaces, sculpture, wall plaster, mosaic, spicatum, signinum, quadratum, reticulatum, mixtum, vittatum, cocciopesto, marble, tuff, lava, peperino, brick, tile, pottery (black gloss, terra sigillata, African Red Slip A & C), inscription. Pantano 1998; Pergola et al. 2003, Via Nomentana Nr. 155; De Franceschini 2005, 85–6. Ficulea site 201a. Villa: owner?, living quarters, metal workshop?, service areas, garden, architectural elements, sculpture, wall plaster, stucco, mosaic, sectile, signinum, spicatum, quadratum, reticulatum, mixtum, cocciopesto, marble, travertine, tuff, lava, brick, tile, pottery (olla perforata, dolium, amphora), coins, lead artifacts, iron bars, inscription. Sorella 1998, 170–3; Pergola et al. 2003, Via Nomentana Nr. 231; De Franceschini 2005, 94–8; Bruto and Vigna 2006. Ficulea site 218. Villa: mosaic, tile, pottery (African Red Slip, amphora), inscription. Calci and Sorella 1995, 125, nota 41; Moscetti 1997, 145. Ficulea site 219b. Villa: owner?, architectural elements, sculpture, mortar, marble, travertine, limestone, tuff, tile, pottery (terra sigillata), inscription. Ficulea site 237. Villa: wall plaster, quadratum, mortar, cocciopesto, marble, travertine, brick, tile, pottery (terra sigillata). Ficulea site 254a. AFF/Villa?: enclosed spaces, spicatum, mortar, marble, tile, pottery (dolium, amphora). Moscetti and La Porta 2004, 167. Ficulea site 258. Villa: architectural elements, mortar, marble, lava, brick, pottery (terra sigillata, dolium). Ficulea site 261a. Villa rustica: 100 m diameter, mortar, tile, pottery (terra sigillata, African Red Slip, dolium). Moscetti and 170 Ficulea site 356a = Tibur III site 6a. Villa: wall plaster, stucco, mosaic, sectile, spicatum, signinum, mortar, cocciopesto, marble, travertine, limestone, tuff, tile, pottery (black gloss, terra sigillata, African Red Slip), cemetery, inscription. Quilici Gigli 1987; Curti and Moscetti 1996, Nr. 2. Ficulea site 360 = Tibur III site 2a. Villa: platforms, foundations, living quarters, architectural elements, wall plaster, mosaic, sectile, spicatum, signinum, reticulatum, marble, inscription, connected to road at site Tibur III 2d and cemetery at site Tibur III 3. Curti and Moscetti 1996, Nr. 1. Ficulea site 364a. Villa: architectural elements, mosaic, spicatum, quadratum, listatum, cocciopesto, marble, travertine, tuff, lava, brick, tile, pottery (terra sigillata, African Red Slip), inscription. Ficulea site 371b. Villa: architectural elements, sculpture, mosaic, spicatum, quadratum, reticulatum, cocciopesto, marble, travertine, limestone, tuff, lava, peperino, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware), inscription. Moscetti 2003, 145–6. Ficulea site 374a. Villa: 1,970 m², 2 areas, cryptoporticus, living quarters, service areas, architectural elements, mosaic, quadratum, reticulatum, mixtum, cocciopesto, marble, travertine, limestone, tuff, peperino, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium, amphora), coins, inscription. Calci and Mari 2003, 196, 199; Calci and Sorella 1995, 124–5, nota 38; Mari et al. 1995, 215– 32. Ficulea site 380a. Villa rustica: architectural elements, mosaic, reticulatum, cocciopesto, marble, travertine, limestone, tuff, brick, tile, pottery (black gloss, plain/cookware). Mari et al. 1995, 216. Ficulea site 398b–c. Villa: platform, cryptoporticus, architectural elements, mosaic, quadratum, incertum, cocciopesto, marble, travertine, limestone, tuff, lava, tile, pottery (terra sigillata, African Red Slip), burial, inscription. Carbonara and Messineo 1992; Calci and Sorella 1995, 124, nota 37; Mari et al. 1995, 215–30; De Franceschini 2005, 286–9. Ficulea site 405a. Villa rustica: platform?, architectural elements, mosaic, spicatum, reticulatum, cocciopesto, marble, sitE cataloguE travertine, tuff, brick, pottery (African Red Slip). Ficulea site 415a. Villa: architectural elements, wall plaster, mosaic, spicatum, reticulatum, cocciopesto, marble, travertine, tuff, lava, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium), inscription. Carbonara and Messineo 1992; Curti and Moscetti 1996, Nr. 18; Calci 1998, 111. Ficulea site 415c. Villa rustica?: workshops?, architectural elements, sculpture, wall plaster, mosaic, sectile, signinum, quadratum, reticulatum, listatum, cocciopesto, marble, travertine, tuff, lava, brick, tile, pottery (dolium, amphora, lamp), coin, inscription. Carbonara and Messineo 1992; Calci and Sorella 1995, 121–2, notes 24, 32; Calci 1998, 95–103. Ficulea site 435. Villa: architectural elements, quadratum, mortar, travertine, tuff, tile, pottery (terra sigillata, African Red Slip, dolium). Ficulea site 455a. Villa rustica: platform, architectural elements, wall plaster, mosaic, spicatum, reticulatum, mixtum, cocciopesto, tuff, brick, tile, pottery (African Red Slip A), later cemetery, inscription. Sorella 1998, 160–1; Pergola et al. 2003, Via Tiburtina Nr. 229; De Franceschini 2005, 101–3; Livi 2006. Ficulea site 460a. Villa?: walls, quadratum, cocciopesto, travertine, tuff, peperino. Ficulea site 473a. Villa rustica: owner?, foundations, living quarters, garden, architectural elements, wall plaster, mosaic, spicatum, signinum, quadratum, quasireticulatum, reticulatum, cocciopesto, marble, travertine, limestone, tuff, lava, peperino, tile, pottery (terra sigillata, dolium), inscription. Sorella 1998, 161–4, De Franceschini 2005, 107–11. Ficulea site 479. Villa: architectural elements, mosaic, sectile, spicatum, reticulatum, marble, tile, pottery (African Red Slip, dolium). Di Nicola and Tantari 1998, 149; De Franceschini 2005, 121–2. Ficulea site 487a. Villa: owner?, platform, living quarters, lime furnace, architectural elements, sculpture, spicatum, quadratum, incertum?, quasireticulatum?, reticulatum, cocciopesto, marble, tuff, brick, tile, pottery (terra sigillata, African Red Slip, plain/ cookware, amphora, lamp). Di Nicola and Tantari 1998, 153; Pergola et al. 2003, Via Tiburtina Nr. 211; De Franceschini 2005, 120–1. Ficulea site 489a. Villa: 2 platforms, foundations, living quarters, storage spaces, harbor?, architectural elements, sculpture, wall plaster, mosaic, spicatum, quadratum, reticulatum, mixtum, marble, travertine, limestone, tuff, brick, tile, pottery (black gloss, terra sigillata, African Red Slip A & C & D, dolium, amphora), later burials. Di Nicola and Tantari 1998, 152; De Franceschini 2005, 129–30. Ficulea site 503a/c = Collatia site 1b–c. Villa: platforms?, foundations, enclosed spaces, signinum, quadratum, reticulatum, vittatum, tuff, tile, pottery (terra sigillata). Ficulea site 507b = Collatia site 2–4. Villa: owner?, enclosed spaces, wall plaster, mosaic, reticulatum, tuff, brick, tile, pottery. Ficulea site 517c. Villa: enclosed spaces, wall plaster, mosaic, reticulatum, marble, tuff, pottery (terra sigillata), inscription. Calci 1998, 82. Ficulea site 536a. Villa: foundations, living quarters, service areas, architectural elements, wall plaster, mosaic, sectile, signinum, quadratum, incertum, reticulatum, cocciopesto, travertine, limestone, tuff, brick. Carbonara and Messineo 1992; Calci 1998, 84; De Franceschini 2005, 112–4. Ficulea site 539d. Villa: owner?, platform, foundations, spicatum, reticulatum, marble, travertine, tuff, lava, pottery (terra sigillata, African Red Slip). Carbonara and Messineo 1992. Ficulea site 540a. Villa: wall plaster, mosaic, signinum, quadratum, mortar, travertine, tuff, tile, pottery (African Red Slip). Calci and Sorella 1995, 117–8; Calci 1998, 89; Pergola et al. 2003, Via Tiburtina Nr. 214. Ficulea site 573a = Tibur III Site 249a. Villa: 50 x 50 m, architectural elements, sculpture, mosaic, sectile, signinum, reticulatum, marble, tuff, inscription. Ficulea site 581b. Villa: architectural elements, sculpture, wall plaster, mosaic, quadratum, reticulatum, marble, travertine, limestone, tuff, tile, pottery (terra sigillata, African Red Slip, plain/cookware). Ficulea site 604a. Villa: owner?, platforms, foundations, living quarters, garden, storage spaces, harbor?, sculpture, mosaic, sectile, signinum, wall plaster, stucco, reticulatum, mixtum, listatum, cocciopesto, tuff, tile, pottery (olla perforata), inscription. Calci 1998, 78–82; Calci and Sorella 1998, 192–4; Pergola et al. 2003, Via Tiburtina Nr. 219; Calci and Mari 2003, 184–6; De Franceschini 2005, 122–6; Filippini 2006. Tibur III site 22. Villa: 40 x 30 m, service areas, architectural elements, wall plaster, mosaic, sectile, spicatum, reticulatum, mixtum, cocciopesto, marble, travertine, limestone, tuff, brick, pottery (black gloss, terra sigillata, plain/cookware, lamp), inscription. Mari and Moscetti 1992, 100; Moscetti 1995, 17; 2001, 114–20. Tibur III site 35. Villa: 230 x 100 m, platform, polygonal, quadratum, reticulatum, travertine, tuff, lava, brick, tile, inscription. Moscetti 1991, Nr. 9. Tibur III site 42a. Villa: 80 x 70 m and 45 x 30 m, 2 platforms, architectural elements, mosaic, spicatum, quadratum, reticulatum, marble, travertine, brick/tile, pottery (black gloss, terra sigillata, African Red Slip), inscription. Moscetti 1991, Nr. 12. Tibur III site 45a. Villa: owner?, 10,000 m², natural platform?, cryptoporticus, living quarters, sculpture, wall plaster, mosaic, sectile, spicatum, signinum, quadratum, quasireticulatum, reticulatum, marble, travertine, limestone, tuff, brick, pottery (black gloss, terra sigillata, African Red Slip, amphora, lamp), sarcophagus, inscription, cemetery at site 45b. Gatti 1991; Moscetti 1991, Nr. 14; 1994, 189–92; 1999b; 2003, 149–52; Calci and Mari 2003, 204–6; Curti and Moscetti 1996, Nr. 3. Tibur III site 55. Villa: mosaic, spicatum, signinum, reticulatum?, travertine, limestone, tile, pottery (black gloss, terra sigillata, African Red Slip). Moscetti 1991, Nr. 19. Tibur III site 80. Villa: 30 x 30 m, architectural elements, wall plaster, spicatum, signinum, mortar, marble, travertine, tuff, brick/tile, pottery (black gloss, African Red Slip, plain/ cookware), inscription. Tibur III site 83a. Villa: 30 x 15 m, platform, living quarters, architectural elements, sectile, reticulatum, marble, travertine, tuff, brick/tile, cemetery at site 83b. Tibur III site 108a. Villa: 110 x 40 m, architectural elements, sculpture, wall plaster, mosaic, signinum, mortar, marble, travertine, brick, tile, pottery (black gloss, terra sigillata, African Red Slip), inscription, cemetery at site 108b. Moscetti 1991, Nr. 87. Tibur III site 109a. Villa: natural platform, living quarters, architectural elements, sculpture, mosaic, mortar, cocciopesto, marble, travertine, limestone, tuff, brick, pottery (black gloss). Mari and Moscetti 1992, 100–15. Tibur III site 110. Villa: walls, architectural elements, sculpture, wall plaster, reticulatum, cocciopesto, marble, travertine, tuff, lava, brick, tile, pottery (dolium), inscription. Tibur III site 118. Villa: 80 x 30 m, platforms, signinum, quadratum, reticulatum, cocciopesto, marble, travertine, tuff, brick, tile, inscription. Moscetti 1991, Nr. 23. Tibur III site 138. Villa: foundation, walls, architectural elements, spicatum, reticulatum, marble, travertine, brick, inscription. Tibur III site 144a. Villa: mosaic, mortar, cocciopesto, marble, travertine, brick, tile, pottery (black gloss, terra sigillata, African Red Slip). Tibur III site 162a. Villa: enclosed spaces, architectural elements, sculpture, wall plaster, mosaic, sectile, mortar, marble, travertine, brick/tile, pottery (black gloss, terra sigillata, African 171 appEndiX i Red Slip). Moscetti 1991, Nr. 82. plain/cookware), inscription? Mari and Moscetti 1992, 164–7. Tibur III site 195a. Villa: 70 m long, architectural elements, wall plaster, mosaic, signinum, reticulatum, cocciopesto, marble, travertine, tuff, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, lamp), glass, sarcophagus, inscription, burial at site 195b. Tibur III site 345a. AME/Villa?: 3 areas?, foundations?, architectural elements, sculpture, mosaic, reticulatum, cocciopesto, marble, travertine, brick, pottery (black gloss, terra sigillata, African Red Slip), inscription. Curti and Moscetti 1996, Nr. 10. Tibur III site 197a. Villa: platforms, foundations, mosaic, sectile, spicatum, reticulatum, travertine, tuff, lava, brick/ tile, pottery (black gloss, terra sigillata, African Red Slip), inscription, cemetery at site 197b. Tibur III site 346. Villa: architectural elements, cocciopesto, marble, travertine, tuff, brick, pottery (plain/cookware, terra sigillata, African Red Slip, amphora). Tibur III site 202a. Villa: architectural elements, mosaic, sectile, spicatum, signinum, quadratum, mortar, marble, travertine, tuff, brick, tile, pottery (black gloss, terra sigillata, African Red Slip), inscription. Moscetti 1991, Nr. 57. Tibur III site 217a. Villa: 150 x 70 m?, platform, storage spaces, architectural elements, mosaic, sectile, spicatum, signinum, mortar, marble, limestone, tuff, brick, tile, pottery, inscription. Moscetti 1991, Nr. 59. Tibur III site 254a. Villa: 80 x 70 m?, 2 platforms, signinum, reticulatum, cocciopesto, travertine, tuff, brick, tile, sarcophagus. Moscetti 1991, Nr. 53. Tibur III site 263a. Villa: walls, mosaic, signinum, reticulatum, travertine, brick/tile, pottery (black gloss, African Red Slip), inscription. Tibur III site 265a. Villa: 50 x 50 m, natural platform?, wall plaster, mosaic, quadratum, reticulatum, marble, travertine, tuff, brick/tile, pottery (black gloss, African Red Slip, dolium), inscription, cemetery at site 266. Tibur III site 270a. Villa: 45 m long, foundations, signinum, quadratum, reticulatum, cocciopesto, travertine, limestone, tuff. Tibur III site 277a. Villa: 100 x 50 m?, architectural elements, mosaic, mortar, cocciopesto, marble, travertine, tuff, brick, tile, pottery (black gloss, African Red Slip D, plain/cookware, dolium, amphora, lamp), inscription, cemetery at site 276b. Tibur III site 287a. Villa: 40 m long, foundations, architectural elements, mosaic, marble, travertine, tuff, brick, tile, pottery (plain/cookware). Tibur III site 369. Villa/Statio?: 75 x 35 m, platforms, architectural elements, wall plaster, mosaic, signinum, mortar, marble, travertine, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, lamp), glass, inscription, connected to burial at site 368a? Mari and Moscetti 1992, Nr. 5. Tibur III site 375. Villa: living quarters, architectural elements, mosaic, incertum, cocciopesto, travertine, limestone, tuff. Tibur III site 387a. Villa: platform, service areas, architectural elements, wall plaster, mosaic, spicatum, incertum, mixtum, cocciopesto, marble, travertine, limestone, tuff, lava, brick, pottery (black gloss, plain/cookware, dolium, amphora), coins. Moscetti 1991, Nr. 89; Mari and Moscetti 1992, Nr. 4. Tibur III site 390a. Villa: mosaic, sectile, spicatum, signinum, travertine, limestone, tuff, brick, tile, inscription. Tibur III site 393. Villa: owner?, platform, enclosed spaces, architectural elements, reticulatum, cocciopesto, travertine, brick, pottery, inscription. Tibur III site 396. Villa: walls, architectural elements, sculpture, cocciopesto, marble, travertine, pottery (black gloss, African Red Slip, amphora), inscription. Tibur III site 402a. Villa: architectural elements, sculpture, wall plaster, mosaic, spicatum, quasiresticulatum, cocciopesto, marble, travertine, brick. Tibur III site 414. Villa: 50 x 10 m, 3 areas, architectural elements, mosaic, signinum, reticulatum, vittatum, cocciopesto, marble, travertine, limestone, tuff, brick, inscription. Moscetti 1991, Nr. 94; Curti and Moscetti 1996, Nr. 12. Tibur III site 294. Villa: platform?, architectural elements, wall plaster, spicatum, reticulatum, marble, travertine, tuff, brick/tile, pottery (dolium). Tibur III site 421. Villa: platform?, architectural elements, mosaic, spicatum, mortar, marble, travertine, brick, pottery (black gloss, terra sigillata, African Red Slip, amphora), inscription, cemetery at sites 422–4. Tibur III site 298. Villa: 3 areas?, mosaic, sectile, cocciopesto, travertine, lava, pottery (black gloss, terra sigillata, African Red Slip, dolium, lamp). Tibur III site 429a. Villa: architectural elements, mosaic, mortar, cocciopesto, marble, travertine, brick, pottery (terra sigillata), inscription. Moscetti 1991, Nr. 96. Tibur III site 302a. Villa: living quarters, architectural elements, wall plaster, mosaic, sectile, spicatum, signinum, mortar, cocciopesto, marble, travertine, limestone, tuff, brick, pottery (black gloss, African Red Slip), inscription. Moscetti 1991, Nr. 48; Curti and Moscetti 1996, Nr. 7. Tibur III site 432. Villa: living quarters, architectural elements, wall plaster, mosaic, signinum, mortar, marble, travertine, limestone, tuff, brick, tile, pottery, inscription. Tibur III site 303a. Villa: 100 x 50 m, enclosed spaces, architectural elements, wall plaster, mosaic, sectile, signinum, marble, travertine, limestone, tuff, brick, tile, pottery (black gloss, African Red Slip A & B & D, lamp), inscription. Curti and Moscetti 1996, Nr. 8. Tibur III site 434. Villa: 50 m long, living quarters, architectural elements, mosaic, sectile, spicatum, mortar, marble, travertine, limestone, tuff, brick, pottery (black gloss). Moscetti 1991, Nr. 96; Curti and Moscetti 1996, Nr. 13. Tibur III site 448a. Villa: 30 x 15 m, wall plaster, mosaic, reticulatum, cocciopesto, travertine, limestone, tuff, brick/tile. Tibur III site 304a. Villa: enclosed spaces, architectural elements, sculpture, mosaic, spicatum, signinum, marble, travertine, coin, inscription. Tibur III site 449. Villa: 100 x 30 m, architectural elements, mosaic, reticulatum, cocciopesto, marble, travertine, tuff, brick, pottery (black gloss, African Red Slip, plain/cookware, dolium, amphora). Tibur III site 308a. Villa: owner?, living quarters, architectural elements, wall plaster, mosaic, spicatum, marble, travertine, pottery (plain/cookware, amphora), inscription. Tibur III site 454. Villa: platforms, architectural elements, quadratum, reticulatum, marble, travertine, tuff, brick, pottery (dolium). Tibur III site 313a = Collatia site 91. Villa: natural platform, foundations, spicatum, quadratum, mortar, cocciopesto, travertine, tuff, lava, peperino, tile. Tibur III site 318a = Tibur IV site 216a = Collatia site 296. Villa: architectural elements, spicatum, quadratum, mortar, cocciopesto, marble, travertine, tuff, peperino, brick, pottery (terra sigillata, African Red Slip, dolium), inscription. Tibur III site 325a. Villa: architectural elements, mortar, cocciopesto?, travertine?, lava, brick, tile, pottery (black gloss, 172 Tibur III site U2. Villa rustica: enclosed spaces, spicatum, mixtum, brick. Moscetti 1991, Nr. 44. Tibur IV site 1a. Villa: owner?, 2 platforms, foundations, living quarters, architectural elements, quadratum, reticulatum, marble, travertine, limestone, tuff, brick/tile, pottery (black gloss). Curti and Moscetti 1996, Nr. 15. Tibur IV site 8. Villa: 2 platforms, living quarters, architectural elements, mosaic, spicatum, polygonal, incertum?, reticulatum?, sitE cataloguE cocciopesto, travertine, limestone, tuff, brick/tile, pottery, inscription. Tibur IV site 11a. Villa: owner?, 2 foundations, architectural elements, wall plaster, mosaic, signinum, quadratum, incertum, reticulatum, cocciopesto, marble, travertine, limestone, tuff, brick/tile, lead?, inscription. Curti and Moscetti 1996, Nr. 11. Tibur IV site 16. Villa: foundations, cryptoporticus?, polygonal, reticulatum, travertine, limestone, tuff. Tibur IV site 20a. Villa: 200 m long, platforms, foundations, living quarters, architectural elements, sculpture, mosaic, sectile, reticulatum, listatum, cocciopesto, travertine, limestone, tuff, pottery (black gloss), burials. Tibur IV site 22a. Villa: 50 x 40 m, platform, foundations, living quarters, wall plaster, mosaic, sectile, reticulatum, limestone, tuff, pottery. Tibur IV site 27. Villa: owner?, cryptoporticus, architectural elements, sculpture, incertum, reticulatum, marble, travertine, limestone, inscription. Tibur IV site 37a. Villa: owner?, 70 x 27 m, 2 platforms, architectural elements, sculpture, polygonal, quadratum, incertum?, reticulatum?, cocciopesto, marble, limestone, brick, tile, inscription. Tibur IV site 40. Villa: owner?, 130 x 75 m, 2 platforms?, living quarters, polygonal, quadratum, incertum, reticulatum, cocciopesto, travertine, limestone, brick, inscription. Tibur IV site 41a. Villa: owner?, platform, foundation, cryptoporticus, architectural elements, wall plaster, mosaic, polygonal, incertum, quasireticulatum, reticulatum, travertine, limestone, tuff, tile, pottery (black gloss). Tibur IV site 44a. Villa: enclosed spaces, architectural elements, mosaic, sectile, reticulatum, cocciopesto, marble, travertine, limestone, tuff, brick, tile, inscription. Tibur IV site 60. Villa: platforms, foundations, cryptoporticus, architectural elements, sculpture, wall plaster, incertum, quasireticulatum, reticulatum, marble, limestone, tuff, inscription. Tibur IV site 67. Villa: owner?, 110 x 80 m, 2 platforms, cryptoporticus, garden, wall plaster, incertum, quasireticulatum, reticulatum, cocciopesto, limestone, tuff, brick, pottery, road. Tibur IV site 68. Villa: owner?, 37 m long, 2 platforms, cryptoporticus, garden, wall plaster, mosaic, incertum, reticulatum, limestone, inscription. gloss, terra sigillata), inscription. Tibur IV site 141a/c. Ruderi/Villa?: platform?, wall plaster, mosaic, reticulatum, tuff, brick, tile, inscription. Tibur IV site 143. Villa: 100 m long, platforms, wall plaster, mosaic, spicatum, incertum, reticulatum, cocciopesto, marble, tuff, brick/tile, pottery (amphora). Tibur IV site 148. Villa: 80 m long, foundations, architectural elements, wall plaster, mosaic, reticulatum, cocciopesto, limestone, tuff, inscription. Tibur IV site 149. Villa: 8,000 m², living quarters, quadratum, reticulatum, mixtum, cocciopesto, marble, travertine, tuff, brick/ tile. Tibur IV site 153. Villa: owner?, 230 x 100 m, 2 platforms, foundations, living quarters, garden, architectural elements, wall plaster, mosaic, reticulatum, mixtum, cocciopesto, marble, travertine, tuff, brick/tile, inscription. Tibur IV site 157. Villa: owner?, 400 m long, platforms, cryptoporticus, living quarters, architectural elements, sculpture, wall plaster, mosaic, incertum, reticulatum, mixtum, cocciopesto, marble, brick, inscription. Tibur IV site 202a. Villa: foundations, enclosed spaces, wall plaster, mosaic, spicatum, polygonal?, quadratum, incertum, reticulatum, cocciopesto, marble, travertine, tuff, lava, peperino, brick/tile, inscription. Tibur IV site 224. Villa: owner?, platforms, foundations, cryptoporticus, architectural elements, sculpture, incertum, reticulatum, marble, limestone, tuff. Tibur IV site 225. Villa: owner?, 20,000 m², platforms, cryptoporticus, living quarters, polygonal, quadratum, incertum, quasireticulatum, reticulatum, listatum, cocciopesto, limestone, tuff, pottery, inscription. Boanelli 1992. Tibur IV site U1. Villa Adriana: owner?, many platforms, living quarters, service areas, gardens, almost all possible inds. Ricotti 2001. Tibur I site 95. Villa: foundations, walls, reticulatum, mixtum, lava, brick/tile, connected to Class 1 villa at site 96? Tibur I site 96. Villa: 50 x 30 m, platform, foundation, cryptoporticus, wall plaster, quadratum, incertum, reticulatum, mixtum, travertine, limestone, tuff, connected to Class 1 villa at site 95 and Class 2 villa at site 112? Tibur I site 97. Villa: over 30 m long, 2 platforms, cryptoporticus, reticulatum, travertine, limestone, connected to Class 1 villa at site 99? Tibur IV site 70. Villa: owner?, 2 platforms, foundations, cryptoporticus, sculpture, wall plaster, incertum, reticulatum, cocciopesto, travertine, limestone, tuff, brick/tile, inscription. Tibur I site 99. Ruderi/Villa?: foundations?, inscription, connected to Class 1 villa at site 97? Tibur IV site 72. Villa: owner?, 2 platforms, cryptoporticus, sculpture, wall plaster, mosaic, polygonal, incertum, quasireticulatum, reticulatum, cocciopesto, travertine, limestone, brick, tile, inscription. Tibur I site 106–107a. Villa?: 80 m long, foundation?, architectural elements, mosaic, incertum, quasireticulatum, reticulatum, mixtum?, vittatum?, cocciopesto, marble, travertine, limestone, brick/tile, inscription. Rasch 1998, 51–106. Tibur IV site 73. Villa: owner?, 200 m long, 2 platforms, foundations, cryptoporticus, living quarters, garden, architectural elements, sculpture, wall plaster, mosaic, incertum, reticulatum, travertine, limestone, tuff, pottery (black gloss), inscription. Tibur I site 131. Villa: enclosed spaces, wall plaster, incertum, quasireticulatum, travertine, limestone, connected to cryptoporticus at site 120? Troccoli 1984. Tibur IV site 86. Villa: owner?, 100 x 80 m, platform, foundations, cryptoporticus, living quarters, wall plaster, incertum, limestone, tuff, brick, tile. Tibur IV site 87. Villa: owner?, 20,000 m², 2 platforms, foundations, cryptoporticus, living quarters, architectural elements, wall plaster, mosaic, quadratum, incertum, reticulatum, cocciopesto, marble, travertine, limestone, tuff, brick, tile, inscription. Tibur IV site 100. Villa: platform, living quarters, incertum, mixtum, limestone, tuff, brick. Tibur IV site 104. Villa: owner?, 23,000 m², 2 platforms, cryptoporticus, living quarters, garden, architectural elements, wall plaster, mosaic, spicatum, incertum, reticulatum, cocciopesto, marble, limestone, tuff, brick, tile, pottery (black Tibur I site 136. Villa: 2–3 platforms, cryptoporticus, architectural elements, wall plaster, signinum, incertum, marble, limestone, brick, pottery. Tibur I site 178–179 = Tibur II site 82. Villa: owner?, foundations, enclosed space, architectural elements, mosaic, incertum, reticulatum, marble, limestone, inscription. Venetucci 1992a, 168. Tibur I site 198–200 = Tibur II site 28. Villa: owner?, 130 m long, foundations, grottoes, signinum, polygonal, quasireticulatum, incertum, cocciopesto, limestone, tuff, brick/ tile, inscription?, connected to rock-cut at site 201 and grotto at site 202. Venetucci 1992a, 146–9; Giuliani 2005. Tibur I site 205. Villa: owner?, foundation, architectural elements, sculpture, wall plaster, mosaic, incertum, reticulatum, marble, limestone, brick/tile. 173 appEndiX i Tibur I site 207–208. Villa: owner?, 150 x 65 m, 3 platforms, foundations, architectural elements, stucco, mosaic, polygonal, incertum, quasireticulatum, reticulatum, marble, travertine, limestone, brick. Venetucci 1992a, 154–7. Tibur I site 209. Villa: owner?, 450 x 174 m and 270 x 152 m, 3 platforms, cryptoporticus, architectural elements, sculpture, wall plaster, stucco, mosaic, sectile, signinum, polygonal, incertum, quasireticulatum, reticulatum, mixtum, marble, travertine, limestone, tuff, peperino, brick, tile, inscription. Neudecker 1988, 234–5; Mari and Boanelli 1991; Venetucci 1992a, 159–62. Tibur I site 212. Villa: owner?, 120 x 75 m, platform, foundation, cryptoporticus, living quarters, garden, architectural elements, sculpture, wall plaster, mosaic, incertum, reticulatum, marble, travertine, limestone, tuff. Neudecker 1988, 235–7; Mari 1994, Nr. 1. Tibur II site 85. Villa: owner?, 400 m², 3 areas, signinum, incertum, reticulatum, limestone, pottery (black gloss). Mari 1988. Tibur II site 88. Villa: owner?, 2 platforms?, foundation, sculpture, wall plaster, sectile, reticulatum, marble, tuff, brick, inscription. Collatia site 195a. Villa: 100 x 20–50 m, sculpture, wall plaster, mosaic, reticulatum, listatum, travertine, tuff, brick, tile, inscription. Neudecker 1988, 213–4. Collatia site 202a. Aggregato rustico/Villa?: architectural elements, wall plaster, mortar, marble, tuff, lava, brick, tile, pottery, inscription. Collatia site 204b. Sito antico/Villa?: 260 x 60 m, platform?, architectural elements, sculpture, wall plaster, mosaic, spicatum, signinum, reticulatum, marble, tuff, lava, peperino, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium, amphora), inscription. Collatia site 223. Villa rustica: enclosed space?, architectural elements, spicatum, marble, tuff, tile, pottery (terra sigillata). Bellini 1985, 124. Collatia site 224p. Villa: platform, wall plaster, mosaic, spicatum, signinum, mortar, marble, travertine, tuff, lava, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/ cookware, dolium, amphora), glass. Collatia site 229c. Villa: platform, enclosed spaces, architectural elements, sculpture, marble, travertine, tile, pottery (terra sigillata). Collatia site 241. Villa rustica: platform?, architectural elements, mortar, travertine, tuff, lava, peperino, brick, tile. Musco 2001, 201. Tibur II site 214. Villa: owner?, 200 x 100 m, 3 platforms, cryptoporticus, living quarters, service areas, architectural elements, sculpture, wall plaster, mosaic, spicatum, signinum, polygonal, incertum, reticulatum, cocciopesto, marble, limestone, brick/tile, inscription. Mari 1983b, Nr. 35; 1984; Neudecker 1988, 229–34; Venetucci 1992a, 87–140; 1992b, 223–35. Collatia site 242. Villa rustica: 31 x 24 m, platform, wall plaster, mosaic, spicatum, quadratum, reticulatum, cocciopesto. Musco and Zaccagni 1985, 101–2; De Franceschini 2005, 135–7. Tibur II site 217. Villa: owner?, 90 m long, platform, foundations, garden, architectural elements, wall plaster, stucco, polygonal, incertum, reticulatum, cocciopesto, limestone, brick/ tile, inscription. Neudecker 1988, 228–9; Mari and Moscetti 1993, 130–42. Collatia site 258b. Villa rustica: 120 x 85 m, platforms, architectural elements, spicatum, mortar, travertine, limestone, tuff, peperino, tile. Collatia site 4d. Villa rustica: platforms, wall plaster, mosaic, spicatum, quadratum, reticulatum, cocciopesto, marble, tuff, tile, pottery (amphora). Rea 1985b, 119–20; De Franceschini 2005, 131–3. Collatia site 245. Villa rustica: 90 x 60 m, 2 platforms, walls, architectural elements, reticulatum, marble, tuff, lava, brick, tile. Collatia site 282. Villa rustica: foundations, wall plaster, mortar, marble, travertine, tuff, lava, brick, tile. Collatia site 14b. Villa: 100 x 80 m, architectural elements, sculpture, mosaic, quadratum, incertum, marble, tuff, peperino, tile, pottery. Collatia site 284. Villa rustica: platforms, living quarters, wall plaster, mosaic, polygonal?, quadratum?, reticulatum, cocciopesto, marble, tuff, lava, peperino, tile, pottery. Montalcini De Angelis d’Ossat 1983; Musco and Zaccagni 1985, 101; De Franceschini 2005, 138–9. Collatia site 43a–c. Villa: 80 m long, architectural elements, sculpture, wall plaster, reticulatum, listatum, cocciopesto, marble, travertine, tuff, brick, tile, pottery (black gloss, terra sigillata). Collatia site 321b. Villa: 1,100 m², platforms, foundations, architectural elements, sculpture, mortar, marble, travertine, tuff, peperino, brick/tile, pottery, inscription, connected to Class 4 rural cluster at site 321c? Musco 2001, 205. Collatia site 45c. Villa: architectural elements, sculpture, mosaic, marble, travertine, brick, inscription. Collatia site 355a. Villa rustica: 200 x 200 m, platforms, enclosed spaces, wall plaster, mosaic, spicatum, signinum, reticulatum, marble, travertine, tuff, lava, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium, amphora), glass, burial, inscription. Collatia site 50a. Villa rustica: architectural elements, sculpture, quasireticulatum, reticulatum, cocciopesto, marble, travertine, tuff, peperino, brick, tile, pottery (terra sigillata, dolium, amphora), sarcophagus, burial. Musco 2001, 193–7. Collatia site 106a. Villa: owner?, architectural elements, mosaic, sectile, reticulatum, mixtum, marble, tuff, peperino, tile, pottery (dolium), inscription. Musco 2001, 201. Collatia site 123a. Villa rustica: 130 x 60 m, enclosed spaces, sculpture, wall plaster, spicatum, reticulatum, cocciopesto, travertine, brick/tile, pottery (dolium). Collatia site 147. Villa rustica: platform, wall plaster, reticulatum, pottery (dolium, amphora), inscription. Collatia site 156a. Villa rustica: platform, walls, wall plaster, mosaic, reticulatum, travertine, tuff, tile, pottery (terra sigillata). Collatia site 177S. Villa: 50 x 20 m, sculpture, wall plaster, mosaic, mortar, cocciopesto, marble, lava, brick, cippus, sarcophagus?, inscription. Collatia site 194l. AFF/Villa: walls, sculpture, wall plaster, mosaic, sectile, spicatum, signinum, mortar, marble, tuff, tile, pottery (terra sigillata, African Red Slip, plain/cookware, dolium, amphora). 174 Collatia site 360a. Villa rustica: 70 x 30 m, foundations, enclosed spaces, wall plaster, mosaic, signinum, quadratum, reticulatum, tuff, lava, peperino, tile, pottery (terra sigillata, plain/cookware, amphora). Musco et al. 2002. Collatia site 362b. Villa: 160 x 140 m, platforms, architectural elements, wall plaster, mosaic, sectile, signinum, listatum, cocciopesto, marble, travertine, tuff, lava, peperino, brick, tile, pottery (impasto, black gloss, African Red Slip, plain/cookware, dolium, amphora), glass, inscription. Collatia site 365a. Villa rustica: 400 x 150 m, platforms, enclosed spaces, sculpture, mosaic, mortar, cocciopesto, marble, tuff, lava, peperino, brick, tile, pottery (terra sigillata, African Red Slip, plain/cookware, dolium, amphora), inscription. Collatia site 372a–c. Villa: 70 x 70 m, architectural elements, sculpture, wall plaster, mosaic, signinum, mortar, cocciopesto, marble, travertine, tuff, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium, amphora), burials?, inscription. Collatia site 373. Villa rustica: walls, architectural elements, sitE cataloguE wall plaster, mosaic, spicatum, quadratum, reticulatum, marble, tuff, lava, peperino, brick, tile, pottery (terra sigillata, African Red Slip). Collatia site 377a. Villa rustica: 200 x 120 m, mosaic, signinum, quadratum, reticulatum, marble, tuff, lava, tile, pottery (terra sigillata, African Red Slip, plain/cookware, dolium, amphora, lamp), inscription. Collatia site 386. Villa rustica: platform, living quarters, garden, wall plaster, mosaic, sectile, spicatum, signinum, cocciopesto, quadratum, reticulatum, marble, tuff, peperino, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/ cookware, dolium, amphora), inscription. Cotton 1979; OliverSmith and Widrig 1981; Widrig 1980; 1981; 1983a; 1983b; 1987; Musco and Zaccagni 1985, 99; De Franceschini 2005, 167–70. Collatia site 388a. Villa: 85 m long, 2 areas, platform?, architectural elements, sculpture, wall plaster, mosaic, spicatum, signinum, reticulatum, cocciopesto, marble, travertine, limestone, tuff, lava, brick, tile, pottery (terra sigillata, African Red Slip, plain/cookware, dolium, amphora), glass, inscription. Collatia site 392a. Villa rustica: 200 x 70 m, platform, living quarters, architectural elements, mosaic, sectile, signinum, reticulatum, cocciopesto, marble, travertine, tuff, lava, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/ cookware, dolium, amphora), glass, inscription. Cotton 1979; Oliver-Smith and Widrig 1981; Widrig 1980; 1981; 1983a; 1983b; 1987. Collatia site 394a. Villa: 120 x 180 m, architectural elements, sculpture, wall plaster, mosaic, spicatum, signinum, reticulatum, cocciopesto, marble, travertine, tuff, peperino, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium, amphora), glass, cippus, sarcophagus, inscription. Montalcini de Angelis d’Ossat 1983, 30; De Franceschini 2005, 166–7. Collatia site 405b. Villa urbana: architectural elements, sculpture, wall plaster, mosaic, spicatum, quadratum, reticulatum, cocciopesto, marble, tuff, brick/tile, pottery (dolium, amphora). Musco and Zaccagni 1985, 106? Collatia site 431. Villa: 65 x 35 m, 2 areas, platforms, architectural elements, wall plaster, mosaic, sectile?, reticulatum, listatum, tuff, peperino, tile, pottery. Collatia site 485a. Villa: wall plaster, mosaic, mortar, marble, tuff, lava, brick, tile, inscription. Marcelli 2002. Collatia site 489b. Villa: 160 x 80 m, enclosed spaces, architectural elements, marble, travertine, tuff, peperino, tile, pottery, cippus, sarcophagus, inscription? Collatia site 495a. Villa: owner?, 280 x 100 m, platforms, enclosed spaces, architectural elements, wall plaster, mortar, marble, travertine, tuff, lava, peperino, brick, tile, pottery (African Red Slip), inscription. Collatia site 516a. Villa rustica: 40 x 40 m, platforms, architectural elements, mortar, tuff, lava, peperino, tile, pottery (terra sigillata). Collatia site 517b. Villa: 250 x 300 m, platforms, enclosed spaces, architectural elements, wall plaster, mosaic, sectile, signinum, quadratum?, reticulatum, cocciopesto, marble, travertine, limestone, tuff, lava, peperino, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium, amphora), glass. Musco et al. 2002. Collatia site 521b. Villa: 100 x 50 m, platforms, architectural elements, mortar, cocciopesto, marble, tuff, peperino, tile, pottery (African Red Slip). Collatia site 540. Villa rustica: 150 x 150 m, platform, architectural elements, mortar, tuff, peperino, brick, tile, pottery. Collatia site 552a. Villa: 70 x 60 m, platform?, enclosed spaces, mortar, cocciopesto, tuff, brick, pottery. Collatia site 560b. Villa: 230 x 100 m, architectural elements, mortar, cocciopesto, marble, limestone, tuff, lava, peperino, tile, pottery (African Red Slip, dolium). Devoti 1978, Nr. 83. Collatia site 567a. Villa: architectural elements, quadratum, mortar, marble, travertine, tuff, lava, peperino, brick, tile, pottery (black gloss, dolium, amphora), inscription. Collatia site 582a. Villa: 150 m long, platform?, enclosed spaces, architectural elements, sculpture, wall plaster, reticulatum, listatum, marble, travertine, tuff, brick, tile, pottery (African Red Slip), inscription. Collatia site 583a. Villa: 240 x 110 m, platforms, storage spaces, architectural elements, wall plaster, reticulatum, cocciopesto, marble, tuff, tile, pottery, cippus, inscription. Collatia site 584. Villa rustica: walls, architectural elements, signinum, mortar, travertine, tuff, lava, brick, tile, pottery. Collatia site 432. Villa: owner?, architectural elements, sculpture, wall plaster, mosaic, spicatum, signinum, quadratum, incertum, reticulatum, mixtum, cocciopesto, marble, travertine, tuff, brick, tile, pottery (African Red Slip), inscription. Collatia site 586. Villa: 140 x 170 m, platforms, enclosed spaces, mosaic, reticulatum, mixtum, cocciopesto, marble, tuff, brick, tile, pottery (terra sigillata), inscription. Collatia site 440a. Villa rustica: foundation, walls, reticulatum, travertine, inscription. Collatia site 590e. Villa: platform, service areas, architectural elements, sculpture, mosaic, spicatum, reticulatum, cocciopesto, marble, tuff, lava, tile, pottery, sarcophagus, burials?, inscription. Collatia site 441. Villa: enclosed spaces, sculpture, wall plaster, mosaic, signinum, mortar, cocciopesto, marble, travertine, tuff, tile, pottery, inscription. Collatia site 444. Villa rustica: 150 x 130 m, 2 areas, architectural elements, sculpture, mosaic, spicatum, reticulatum, cocciopesto, marble, travertine, tuff, peperino, tile, pottery (dolium), inscription. Collatia site 454b. Villa urbana: owner?, platform?, living quarters, architectural elements, wall plaster, mosaic, sectile, quadratum, quasireticulatum, reticulatum, mixtum, cocciopesto, marble, travertine, tuff, lava, peperino, tile, pottery (black gloss), inscription. Tartara 1988; De Franceschini 2005, 182–5. Collatia site 460. Villa: enclosed spaces, architectural elements, wall plaster, mosaic, spicatum, quadratum, reticulatum, cocciopesto, tuff, pottery (terra sigillata). Collatia site 472. Villa: walls, mosaic, marble, travertine, limestone, tuff, lava, peperino, tile, glass, inscription. Collatia site 477. Villa rustica: foundations, walls, wall plaster, mosaic, mortar, marble, tuff, lava, pottery (dolium, amphora). Collatia site 479b. Villa urbana: 270 x 125 m, platforms, foundations, architectural elements, wall plaster, reticulatum, listatum, cocciopesto, marble, travertine, tuff, brick. Collatia site 591f/j. Ruderi/Villa?: architectural elements, sculpture, wall plaster, mosaic, reticulatum, marble, lava, tile, pottery (amphora), inscription. Devoti 1978, Nr. 59? Collatia site 593a. Villa: owner?, platforms, architectural elements, sculpture, reticulatum, listatum, cocciopesto, marble, travertine, tuff, lava, tile, pottery (terra sigillata, amphora). Devoti 1978, Nos. 60–61. Collatia site 594. Villa: 200 x 70 m, wall plaster, mortar, tuff, lava, tile, pottery (African Red Slip), inscription. Collatia site 596a. Villa rustica: enclosed space, wall plaster, spicatum, cocciopesto, lava, tile, pottery, inscription. Collatia site 598a. Villa: 60 m long, wall plaster, mosaic, reticulatum, cocciopesto, tuff. Collatia site 601a. Villa: 300 x 150 m, foundations, storage spaces?, architectural elements, sculpture, mosaic, quadratum, reticulatum, marble, travertine, tuff, lava, peperino, brick, tile, pottery (African Red Slip, amphora). Collatia site 604a. Villa: architectural elements, sculpture, wall plaster, mosaic, reticulatum, cocciopesto, marble, tuff, brick, tile, pottery (plain/cookware, dolium), sarcophagus, inscription. Collatia site 607a. Villa: architectural elements, mortar, 175 appEndiX i cocciopesto, marble, tuff, lava, peperino, tile, pottery (plain/ cookware). mortar, marble, travertine, tuff, lava, peperino, brick, tile, pottery (amphora). Collatia site 616b–c. Villa: 250 x 200 m, architectural elements, mortar, cocciopesto, marble, travertine, tuff, peperino, brick, tile, pottery (amphora), inscription. Collatia site 812. Villa: 180 x 60 m, foundations, architectural elements, wall plaster, reticulatum, marble, travertine, tuff, lava, peperino, brick, tile, pottery (African Red Slip), inscription. Collatia site 618. Villa rustica: mortar, cocciopesto, tuff, peperino, tile, pottery (terra sigillata). Zaccagni 1984. Collatia site 828. Villa: 260 x 160 m, 3 platforms, walls, architectural elements, mortar, marble, travertine, tuff, peperino, tile, pottery, cippus, inscription. Collatia site 628a. Aggregato/Villa?: foundations, spicatum, reticulatum, cocciopesto, marble, tuff, lava, tile, pottery (African Red Slip). Collatia site 634a. Villa rustica: enclosed spaces, architectural elements, reticulatum, cocciopesto, travertine, tuff, brick, pottery, inscription. Collatia site 646. Villa rustica: 27 x 28 m, living quarters, textile dyeing workshop?, service areas, wall plaster, mosaic, spicatum, signinum, incertum, reticulatum, cocciopesto, marble, travertine, limestone, tuff, lava, peperino, tile, pottery (terra sigillata). Morelli 1984; Musco and Zaccagni 1985, 103–5; De Franceschini 2005, 205–7. Collatia site 647. Villa: 100 x 60 m, platforms, walls, wall plaster, spicatum, reticulatum, cocciopesto, marble, tuff, tile, inscription. Collatia site 649. Villa: 200 x 100 m, platform, living quarters, service areas, architectural elements, sculpture, wall plaster, sectile, listatum, vittatum, marble, travertine, tuff, lava, peperino, brick, tile, pottery (terra sigillata, African Red Slip, dolium, amphora), inscription. Corrente 1984a; De Franceschini 2005, 208–9. Collatia site 651a. Villa rustica: 150–160 x 130 m, platforms, sculpture, wall plaster, sectile, reticulatum, cocciopesto, marble, brick, tile. Collatia site 669. Villa: enclosed space, architectural elements, reticulatum, cocciopesto, marble, tuff, lava, brick, tile. Collatia site 679a. Villa urbana: owner?, 2 platforms, cryptoporticus, living quarters, service areas, gardens, temple, mosaic, signinum, reticulatum, mixtum, marble, tuff, peperino, brick, tile, inscription. Corrente et al. 1988; Neudecker 1988, 207–9; De Francesco 1990, 58–9; Coarelli 1993, 148–54; Pergola et al. 2003, Via Latina Nr. 295; De Franceschini 2005, 209–14. Collatia site 695a. Villa?: owner?, garden?, architectural elements, reticulatum, cocciopesto, marble, travertine, tuff, peperino, tile, pottery (terra sigillata, plain/cookware), inscription? Collatia site 713a. Villa: living quarters?, garden?, architectural elements, spicatum, mortar, cocciopesto, marble, travertine, tuff, brick, tile, pottery (terra sigillata, plain/cookware, dolium). Collatia site 728. Villa rustica: 160 x 100 m, platform, mortar, marble, travertine, tuff, lava, peperino, brick, tile, pottery (dolium). Collatia site 761a. Villa rustica: enclosed spaces, listatum, cocciopesto, tuff, lava, peperino, brick. Collatia site 771a. Villa rustica: walls, wall plaster, mortar, cocciopesto, marble, tuff, lava, brick, tile, pottery (terra sigillata, plain/cookware, amphora). Musco 1984; Rea 1985b, 121. Collatia site 776. Villa rustica: 10,000 m², living quarters, storage spaces, architectural elements, wall plaster, sectile, quadratum, reticulatum, listatum, vittatum, marble, limestone, tuff, peperino, tile, pottery (terra sigillata, dolium). Musco 1984; Rea 1985b, 121; De Franceschini 2005, 214–5. Collatia site 787a. Villa rustica: enclosed spaces, wall plaster, mosaic, reticulatum, lava, tile. Devoti 1978, Nr. 53. Collatia site 795. Villa: 220 x 160 m, platform, cryptoporticus, wall plaster, mortar, marble, tuff, lava, peperino, tile, pottery (African Red Slip), inscription. Collatia site 803. Villa: 400 x 110 m, platforms, enclosed spaces, garden?, architectural elements, sculpture, quadratum, 176 Collatia site 832a. Villa?: owner?, 250 x 100 m, sculpture, travertine, tuff, lava, brick/tile, pottery (African Red Slip). Neudecker 1988, 158. Collatia site 839c Villa: platform, architectural elements, mosaic, spicatum, mortar, cocciopesto, marble, travertine, tuff, lava, peperino, tile, pottery (African Red Slip). Collatia site 842. Villa rustica: 90–100 x 90–100 m, platform, spicatum, mortar, cocciopesto, marble, tuff, peperino, tile. Collatia site 855. Villa: 17 x 37 m, platforms, enclosed spaces, architectural elements, wall plaster, quadratum, mortar, marble, travertine, tuff, lava, peperino, tile, pottery (terra sigillata). Devoti 1978, Nr. 75. Tellenae site 5b. Villa: owner?, living quarters?, architectural elements, sculpture, wall plaster, mosaic, reticulatum, tuff, peperino, brick/tile, pottery (lamp), inscription. Quilici 1969, Nr. 1583. Tellenae site 29a. Villa: 50 m long, platform, enclosed spaces, wall plaster, mosaic, spicatum, reticulatum, mixtum, marble, tuff, brick, tile, inscription. Quilici 1969, Nr. 1086; Cecchini et al. 1990, 120. Tellenae site 37b. Villa: wall plaster, mosaic, reticulatum, tuff, lava, inscription. Quilici 1969, Nr. 1986; Moltesen 1980; 1988; Spera and Mineo 2004, 176–7. Tellenae site 40. Villa: foundations, peperino, brick, tile, pottery (black gloss, terra sigillata, dolium, amphora), inscription. Quilici 1969, Nr. 1977. Tellenae site 62. Villa: enclosed spaces?, architectural elements, reticulatum, marble, peperino, brick, tile, pottery (plain/ cookware, dolium, amphora), inscription. Tellenae site 79. Villa: platform, architectural elements, reticulatum, listatum, peperino, brick, tile, pottery (black gloss, terra sigillata, dolium, amphora), inscription. Quilici 1969, Nr. 2059 (also 1080?). Tellenae site 88. Villa: mosaic, reticulatum, mixtum, listatum, cocciopesto, brick, tile, pottery (black gloss, terra sigillata, dolium, amphora), inscription. Quilici 1969, Nr. 1935. Tellenae site 97a. Villa: 100 x 115 m, 2 platforms, living quarters, architectural elements, sculpture, wall plaster, mosaic, spicatum, quadratum, incertum, reticulatum, latericium, marble, travertine, tuff, peperino, brick, tile, pottery (dolium), inscription? Quilici 1969, Nr. 1064; Neudecker 1988, 197–8; Liverani 1992; De Franceschini 2005, 241–4. Tellenae site 117a. Villa: wall, mortar, peperino, brick, tile, pottery (terra sigillata, amphora), inscription. Quilici 1969, Nr. 1903. Tellenae site U7. Villa: owner?, 450 x 250 m, living quarters, garden, service areas, architectural elements, sculpture, wall plaster, mosaic, sectile, signinum, quadratum, reticulatum, mixtum, listatum, cocciopesto, marble, tuff, lava, peperino, brick, tile, inscription. Castagnoli et al. 1972, 143–7; Ricci 1986; 1998; Quilici 1989, 49–51; Coarelli 1993, 55–8; De Franceschini 2005, 222–36; Spera and Mineo 2004, 144–53. Bovillae site 11. Villa: 2 platforms, foundations, enclosed spaces, mixtum, marble. Bovillae site 19. Villa: 35 x 20 m, platform, foundations, enclosed spaces, architectural elements, sculpture, reticulatum, cocciopesto, marble, travertine, peperino, brick, tile. Bovillae site 48. Villa: owner?, architectural elements, sculpture, mosaic, sectile, mixtum, listatum, cocciopesto, marble, travertine, tuff, peperino, brick, pottery (impasto), sitE cataloguE inscription. Devoti 1978, Nr. 180; Corrente 1984b; 1988a, 400, nota 6; Neudecker 1988, 209–11; LTURS s.v. Latina via, VII miglio; De Franceschini 2005, 239–41. Bovillae site 273. Villa: platform, cryptoporticus, living quarters, architectural elements, wall plaster, marble, peperino, inscription. Bovillae site 96. Villa: owner?, foundations, architectural elements, sculpture, spicatum, quadratum, cocciopesto, marble, travertine, tuff, peperino, brick, tile, pottery (dolium), inscription, connected to road at site 68. Bovillae site 285b. Villa: architectural elements, wall plaster, mosaic, sectile, spicatum, signinum, quadratum, reticulatum, cocciopesto, marble, pottery (terra sigillata), inscription?, connected to Class 1 villa at site 286? Bovillae site 101. Villa: owner?, 300 x 120 m?, 2 platforms, cryptoporticus, living quarters, mithraeum?, wall plaster, mosaic, signinum, incertum, vittatum, cocciopesto, limestone, lava, peperino, brick, tile, inscription. Di Matteo 2002; 2003b; De Franceschini 2005, 163–6. Bovillae site 286. Villa: living quarters, architectural elements, sculpture, sectile, spicatum, reticulatum, mixtum, peperino, pottery (amphora), inscription?, connected to Class 1 villa at site 285b? Bovillae site 103. Villa: owner?, 120 x 30 m, foundations, cryptoporticus, living quarters?, garden, sculpture, wall plaster, mosaic, spicatum, signinum, incertum, reticulatum, mixtum, cocciopesto, marble, travertine, tuff, lava, peperino, inscription. Corrente 1985, 116–7; Neudecker 1988, 206–7; De Francesco 1990, 51; De Franceschini 2005, 244–9. Bovillae site 113. AFF/Villa?: enclosed spaces, spicatum, reticulatum, cocciopesto, tuff, peperino, pottery (dolium). Bovillae site 121.1. Villa: owner?, 110 x 120 m, 2 platforms, enclosed spaces, garden, spicatum, incertum, reticulatum, listatum, cocciopesto, lava, peperino, inscription. Bovillae site 123a. Villa: living quarters, architectural elements, sculpture, wall plaster, mosaic, quadratum, reticulatum, mixtum, cocciopesto, marble, limestone, peperino, pottery (amphora), burial. Corrente 1988b, 400–1; Neudecker 1988, 206; De Franceschini 2005, 237–9. Bovillae site 131. Villa?: architectural elements, wall plaster, mosaic, marble, brick. Corrente 1988a, 400. Bovillae site 141. Villa: enclosed spaces, architectural elements, mosaic, quadratum, cocciopesto, travertine, peperino, brick, ustrinum? Bovillae site 160. Villa: foundations, walls, architectural elements, wall plaster, quadratum, reticulatum, marble, peperino, brick, tile, pottery (plain/cookware, dolium, amphora), inscription. Bovillae site 165. Villa: foundations, signinum, reticulatum, cocciopesto, peperino, brick, inscription. Bovillae site 169. Villa: living quarters, architectural elements, wall plaster, mosaic, spicatum, reticulatum, mixtum, marble, peperino, brick, inscription. Bovillae site 173a. Villa: owner?, 103 x 80 m, 2 platforms, cryptoporticus, living quarters, service areas, garden, architectural elements, sculpture, wall plaster, mosaic, sectile, spicatum, reticulatum, marble, travertine, peperino, pottery (dolium), bronze, sheets of talcum, inscription. Neudecker 1988, 169–70; Pensabene 1999, 104, 159–60, 182, 195, 197, 206–7. Bovillae site 179.–180. Villa: owner?, 240 x 110 m, platforms, architectural elements, sculpture, wall plaster, mosaic, signinum, reticulatum, mixtum, vittatum, cocciopesto, marble, tuff, lava, peperino, brick, pottery (amphora), inscription. Bovillae site 200. Villa: owner?, architectural elements, sculpture, mosaic, mortar, cocciopesto, marble, travertine, lava, peperino, tile, burials, road, inscription. Neudecker 1988, 184. Bovillae site 209. Villa: foundations, architectural elements, sculpture, mosaic, mortar, marble, lava, brick, tile, pottery (plain/cookware), road?, inscription. Spera and Mineo 2004, 178. Bovillae site 222a. Villa: living quarters, architectural elements, sculpture, mosaic, sectile, mortar, marble, travertine, lava, peperino, pottery (dolium), bronze, lead, inscription. Neudecker 1988, 159. Bovillae site 249a. Villa: enclosed spaces, architectural elements, sculpture, wall plaster, mosaic, marble, coin, inscription. Bovillae site 254. Villa: mosaic, spicatum, mortar, cocciopesto, peperino. Bovillae site 337. Villa: architectural elements, sculpture, reticulatum, marble, peperino, brick, inscription. Bovillae site 387. Villa: enclosed spaces?, architectural elements, sculpture, wall plaster, incertum, cocciopesto, marble, peperino, pottery (dolium, amphora, lamp), inscription. Bovillae site 391. Villa: foundation, enclosed spaces, architectural elements, sculpture, sectile, quadratum, reticulatum, mixtum, marble, peperino, brick, pottery (black gloss, terra sigillata, dolium, amphora), inscription. Bovillae site 404. Villa: 55 x 30 m, 3 platforms, cryptoporticus?, garden, architectural elements, wall plaster, reticulatum, peperino, brick/tile. Bovillae site 417. Villa: 60 x 60 m, platforms, foundations, cryptoporticus, architectural elements, sculpture, mosaic, sectile, reticulatum, marble, peperino, brick/tile, pottery (black gloss, terra sigillata, amphora, lamp), glass, bronze, ivory, inscription. Bovillae site 423. Villa: walls, architectural elements, sculpture, wall plaster, spicatum, reticulatum, cocciopesto, marble, lava, peperino, brick/tile, inscription. Bovillae site 432. Villa: owner?, 40 x 20 m, 2 platforms, living quarters, sculpture, mosaic, wall plaster, quadratum, reticulatum, cocciopesto, marble, travertine, lava, peperino, tile, inscription. Granino Cecere 1995, 381–6. Bovillae site 436a. Villa: enclosed spaces, architectural elements, mosaic, reticulatum, marble, peperino, brick. Bovillae site U1. Villa: owner?, living quarters, sculpture, reticulatum, listatum, marble, cocciopesto, brick, coin, inscription. Corrente 1984b; LTURS s.v. Latina via, VII miglio. Tusculum site 35–41. Villa: 135 m long, 2 platforms, living quarters, garden, incertum, reticulatum, cocciopesto, lava, pottery (black gloss, African Red Slip A & D, African cookware, plain ware, dolium, amphora), coins, sarcophagus. Tusculum site 93–99. Villa: 200 x 180 m, 2 platforms?, cryptoporticus, enclosed spaces, architectural elements, sculpture, stucco, incertum, reticulatum, lava, brick/tile. Tusculum site 135–136. Villa: 2 areas, foundations, enclosed spaces, architectural elements, sculpture, mosaic?, reticulatum, cocciopesto?, marble, lava. Tusculum site 154. Villa: owner?, cryptoporticus, living quarters?, architectural elements, wall plaster, incertum, reticulatum, mixtum, lava, peperino, brick, Class 2 villa at site 156–158 part of this? Tusculum site 190–194. Villa: foundations, cryptoporticus, living quarters, architectural elements, sculpture, mosaic, sectile, incertum, reticulatum, cocciopesto, lava, brick/tile, inscription. Tusculum site 208–219. Villa: owner?, 2 platforms, foundations, architectural elements, sculpture, sectile, spicatum, incertum, quasireticulatum, reticulatum, mixtum, latericium, cocciopesto, marble, lava, tile, coin, cemetery, inscription, Tusculum site 235–242. Villa: 105 x 135 m, 2 platforms, foundations, living quarters, architectural elements, sculpture, incertum, latericium, mixtum, cocciopesto, lava, brick, tile. Tusculum site 314–316. Villa: platform, cryptoporticus?, mosaic, signinum, incertum, reticulatum?, mixtum, cocciopesto, lava, brick, pottery (terra sigillata, African Red Slip C & D, plain/cookware), inscription? 177 appEndiX i Tusculum site 318–323. Villa: platform, living quarters, garden?, architectural elements, mosaic, incertum, quasireticulatum?, marble, lava, brick. Tusculum site 362–379. Villa: owner?, 400 m long, foundations, cryptoporticus?, architectural elements, sculpture, mosaic, sectile, reticulatum, mixtum, cocciopesto, marble, lava, peperino, tile, pottery (amphora), inscription. Tusculum site 380–392. Villa: 105 x 250 m, 2 platforms, foundations, living quarters, architectural elements, sculpture, mosaic, sectile, incertum, reticulatum, cocciopesto, marble, lava, peperino, brick, pottery (black gloss, terra sigillata, African Red Slip A & C, African cookware), sarcophagus, urn, burial?, inscription. Tusculum site 397–409. Villa: owner?, 3 platforms?, foundations, cryptoporticus?, sculpture, wall plaster, spicatum, polygonal?, incertum, quadratum, reticulatum, mixtum?, tuff, lava, peperino, brick, inscription. Tusculum site 426. Villa: owner?, 250 x 220 m, platform, foundations, architectural elements, wall plaster, spicatum, incertum, quasireticulatum, reticulatum, mixtum, latericium, cocciopesto, marble, lava, brick, tile, inscription. Tusculum site 437–487. Villa: owner?, 5–6 platforms, foundations, cryptoporticus?, architectural elements, sculpture, mosaic, sectile, wall plaster, stucco, quadratum?, incertum, quasireticulatum, reticulatum, mixtum, latericium, cocciopesto, marble, travertine, lava, peperino, brick, tile, sarcophagus, road, inscription. Tusculum site 500–508. Villa: owner?, 220 m long, 3 platforms, foundations, living quarters, garden?, architectural elements, sculpture, mosaic, incertum, reticulatum, marble, lava, inscription. Tusculum site 511–517. Villa: 180 x 50 m, 3 platforms, sculpture, mosaic, sectile, wall plaster, stucco, incertum, reticulatum, marble, tuff, lava, inscription. Tusculum site 525–535. Villa: owner?, 240 m long, platforms, living quarters, garden, architectural elements, sculpture, reticulatum, latericium, cocciopesto, marble, peperino, tile, pottery (plain/cookware?), lead artifact, sarcophagus, cemetery?, inscription. Tusculum site 546–552. Villa: owner?, 150 x 130 m, platform, architectural elements, sculpture, sectile, mosaic, wall plaster, polygonal, quadratum, incertum, reticulatum, marble, lava, peperino, burials. brick/tile, connected to Class 1 villa at site 650–655? Tusculum site 687–695. Villa: owner?, 89 x 300 m, 2 platforms?, enclosed spaces, architectural elements, sculpture, mosaic, quasireticulatum, latericium, marble, peperino, lava, brick/tile, inscription. Tusculum site 720–736. Villa: owner?, 105 x 90 m and 60 x 60 m, platform?, foundations, cryptoporticus, living quarters, garden?, architectural elements, sculpture, mosaic, quadratum, incertum, quasireticulatum, reticulatum, latericium, cocciopesto, marble, lava, peperino, bronze, burials, road, inscription. Tusculum site 764–768. Villa: owner?, platform, foundations, architectural elements, sculpture, mosaic, spicatum, wall plaster, stucco, quadratum, reticulatum, latericium, cocciopesto, marble, lava, brick/tile, pottery (black gloss, terra sigillata, lamp), coin, inscription. Tusculum site 789–794. Villa: owner?, 120 x 50 m and 140 x 140 m, 2 platforms, cryptoporticus, living quarters, architectural elements, sculpture, sectile, mosaic, signinum, quasireticulatum, cocciopesto, marble, lava, peperino, pottery (terra sigillata, African Red Slip A), burial?, inscription. Tusculum site 804–806. Villa: 32 x 37 m, 2 platforms, cryptoporticus, architectural elements, sculpture, mosaic, sectile, wall plaster, polygonal, incertum, quasireticulatum, reticulatum, mixtum?, cocciopesto, marble, tuff, lava, brick/tile, inscription. Tusculum site 832.–836. Villa: platform, cryptoporticus, living quarters?, grotto, architectural elements, sculpture, stucco, mosaic, spicatum, signinum, quadratum, quasireticulatum, reticulatum, mixtum, latericium, cocciopesto, marble, tuff, lava, peperino, brick, pottery (terra sigillata, dolium, amphora), cippus, cemetery, inscription. class 2 sitEs Fidenae site 10. AFF?: foundation, mosaic, spicatum, mortar, tuff, tile, pottery (terra sigillata, African Red Slip). Fidenae site 11a. Insediamento?: architectural elements, quadratum, wall plaster?, marble, travertine, tuff, brick/tile/ pottery. Fidenae site 21. Villa: architectural elements, mosaic, reticulatum, marble, tuff, brick/tile, pottery (bucchero, black gloss), inscription. Tusculum site 561–569. Villa: owner?, 150 x 150 m, platform, foundations, enclosed spaces, garden?, mosaic, spicatum, reticulatum, mixtum, marble, travertine, tuff, lava, brick, cippus, burial, inscription. Fidenae site 25a. Villa: platform, walls, mosaic, quadratum, mortar, tuff, sarcophagus. Tusculum site 574. Villa: 300 x 180 m, platforms, walls, architectural elements, mosaic, wall plaster, mortar, marble, glass. Fidenae site 73. Villa: platform, wall plaster, spicatum, quadratum, incertum, reticulatum, marble, tuff. Tusculum site 592–599. Villa: owner?, 115 x 210 m, 2 platforms, foundations, living quarters, gardens, architectural elements, spicatum, incertum, quasireticulatum, reticulatum, lava, peperino, tile, burials, inscription. Tusculum site 634–642. Villa: 200 x 130 m and 100 x 130 m, 2 platforms, foundations, architectural elements, sculpture, mosaic, sectile, reticulatum, cocciopesto, marble, travertine, lava, peperino, brick, pottery (lamp), burial, road, inscription, connected to Class 2 villa at site 632–633? Tusculum site 650–655. Villa: owner?, 121 x 90 m, 3,600 m², 200 m² and 4,900 m², 2 platforms, foundations, living quarters?, architectural elements, sculpture, mosaic, sectile, spicatum, polygonal?, reticulatum, marble, lava, peperino, brick, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium, amphora), burial, inscription, connected to Class 1 villa at site 656–660? Tusculum site 656–660. Villa: owner?, 2 platforms, foundations, cryptoporticus?, garden?, architectural elements, wall plaster, incertum, quasireticulatum, reticulatum, marble, lava, peperino, 178 Fidenae site 48. Villa?: architectural elements, mortar, tuff, tile, pottery (black gloss, African Red Slip). Fidenae site 86. Villa: marble, tuff, tile, pottery (black gloss, African Red Slip, plain/cookware, amphora). Fidenae site 88b. Villa rustica: 80 x 45 m, living quarters, mortar, marble, tuff, tile, pottery (terra sigillata, plain/ cookware). Pergola et al. 2003, Via Salaria Nr. 61; Di Gennaro et al. 2004, Nr. 3. Fidenae site 91. Villa: platform, architectural elements, wall plaster, spicatum, reticulatum, marble, travertine, tuff, brick, tile, pottery (black gloss). Perego 1990; Di Gennaro et al. 2004, 111? Fidenae site 117a. Villa rustica: architectural elements, travertine, tuff, lava. Fidenae site 124. Villa rustica: 180 m x 210–220 m, platform, mortar, tuff, tile, pottery. Fidenae site 132. Villa: wall plaster, mosaic, reticulatum, cocciopesto, tuff, tile, pottery (terra sigillata, plain/cookware, dolium). Fidenae site 141. Villa: mortar, tile, pottery (terra sigillata), same as Class 1 villa at site 139a? De Franceschini 2005, 79–80. sitE cataloguE Fidenae site 144. Villa: reticulatum, cocciopesto, tuff, brick, tile. Fidenae site 146a. Villa rustica: wall plaster, mosaic, incertum, reticulatum, cocciopesto, limestone, tuff, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium), glass. Fidenae site 148b. Villa: platform, quadratum, reticulatum, marble, tile. Fidenae site 149. Villa: platforms? Fidenae site 153. Villa: mosaic, sectile, quadratum, reticulatum, marble, travertine, tuff, brick/tile. Fidenae site 157a. Villa rustica: wall plaster, incertum, reticulatum?, travertine, tuff, connected to Class 3 villa at site 156a? Fidenae site 158a. Villa: owner?, 350 x 80 m, platforms, spicatum, reticulatum, tuff, brick/tile. Fidenae site 160a. Villa: walls, wall plaster, mosaic, reticulatum, cocciopesto, tuff, tile, pottery (terra sigillata, plain/cookware). Fidenae site 163. Villa rustica: enclosed spaces, architectural elements, vittatum, cocciopesto, tuff, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, amphora), connected to cemetery at site 162? Pergola et al. 2003, Via Nomentana Nr. 113; De Franceschini 2005, 80–1. Fidenae site 172a–b. Villa rustica: 10,500 m², foundation, architectural elements, wall plaster, mosaic, sectile, spicatum, reticulatum, cocciopesto, marble, travertine, tuff, lava, tile, pottery (black gloss, terra sigillata, African Red Slip A & D, African cookware, plain/cookware, dolium, lamp), glass, later burials, inscription? Dell’Era 1998, 299; Fraioli 2000, 233–4; Pergola et al. 2003, Via Nomentana Nr. 120; Di Gennaro et al. 2004, Nr. 86, 89, 104, 113. Fidenae site 186. Rustico impianto/Villa rustica?: foundations, enclosed spaces, signinum, mortar, tuff, lava, tile, pottery (terra sigillata, dolium), coin. Fidenae site 206. Villa: walls, wall plaster, mosaic, mortar, cocciopesto, marble, tuff, tile, pottery (terra sigillata, plain/ cookware), coin, inscription. Fidenae site 212a. Ruderi/Villa?: foundation, reticulatum, cocciopesto, tuff. Fidenae site 212b. Villa?: foundations, wall plaster, stucco, signinum, quadratum, incertum?, reticulatum?, mixtum, marble, lava, peperino, same as Class 2 villa at site 213a? African Red Slip D, dolium), inscription. Ficulea site 7b. Villa rustica: living quarters, mosaic, spicatum, signinum, quadratum, reticulatum, cocciopesto, travertine, brick, tile, pottery (black gloss, African Red Slip A & D, dolium). Pergola et al. 2003, Via Nomentana Nr. 119; De Franceschini 2005, 54–6. Ficulea site 52. Villa: owner?, architectural elements, sculpture, mosaic, quadratum, marble, travertine, tuff, brick, tile, pottery (black gloss, terra sigillata), sarcophagus, inscription. Pergola et al. 2003, Via Nomentana Nr. 198. Ficulea site 71b. Villa: architectural elements, spicatum, mortar, marble, travertine, tuff, lava, brick, tile, pottery (terra sigillata, dolium, amphora). Ficulea site 73. Villa: wall plaster, mosaic, spicatum, mortar, marble, travertine, tile, pottery (black gloss, terra sigillata, African Red Slip, amphora). Pergola et al. 2003, Via Nomentana Nr. 175. Ficulea site 80. AFF/Villa?: 300 m², architectural elements, mosaic, mortar, cocciopesto, marble, travertine, limestone, tuff, lava, brick, tile, pottery (black gloss, terra sigillata, thin-walled ware, African Red Slip A & C & E, plain/cookware, dolium, amphora), inscription. Pergola et al. 2003, Via Nomentana Nr. 199. Ficulea site 81. AFF/Villa?: 60 x 100 m, walls, wall plaster, mosaic, quadratum, marble, tuff, brick, tile, pottery (black gloss, terra sigillata). Ficulea site 82. Villa: mortar, cocciopesto, travertine, lava, tile, pottery (black gloss, terra sigillata, dolium). Ficulea site 86. Villa rustica?: pottery (black gloss), connected to possible villa at site 87–88 (not in the catalogue since it was only marked on the distribution map)? Ficulea site 119. Casa rustica: tile, pottery. Ficulea site 124b. Villa: 2,850 m², architectural elements, mosaic, reticulatum, cocciopesto, marble, travertine, tuff, lava, pottery (black gloss, terra sigillata, African Red Slip D, African cookware, amphora), burial. Pergola et al. 2003, Via Nomentana Nr. 145. Ficulea site 133. Villa: platform, walls, mortar, tuff, tile, pottery. Ficulea site 137. Villa: architectural elements, marble, travertine, tuff, tile, connected to cemetery at site 136? Ficulea site 155a. AFF/Villa?: 40 m diameter, foundations, limestone, tile, pottery (impasto, black gloss). Fidenae site 213a. Villa: enclosed spaces, architectural elements, sculpture, mosaic, reticulatum, marble, tuff, pottery (lamp), inscription, same as Class 2 villa at site 212b? Ficulea site 155b Villa rustica: mosaic, reticulatum, cocciopesto, limestone, tuff, brick, tile, pottery (black gloss, terra sigillata). Fidenae site 215a. Villa: owner?, architectural elements, sculpture, mosaic, reticulatum, marble, travertine, tuff, lava, peperino, brick/tile, pottery (terra sigillata), inscription. Ficulea site 156b. Villa?: platform, reticulatum, vittatum, travertine, limestone, tuff, tile, pottery (terra sigillata, African Red Slip A & C). Fidenae site 230. Villa rustica: architectural elements, sculpture, reticulatum, marble, peperino, tile, pottery (dolium), sarcophagus, burials, inscription. Ficulea site 164b. Villa: platform?, building debris. Fidenae site 234b. Villa rustica: platform, foundations, mortar, tile, pottery. Ficulea site 166b. Villa: quadratum, mortar, marble, travertine, tuff, peperino, tile, pottery (black gloss, terra sigillata, dolium, amphora). Fidenae site 239. Villa rustica: 200 m², platforms, walls, mosaic, spicatum, cocciopesto, marble, travertine, tuff, tile. Ficulea site 169a. Villa: owner?, 150 x 200 m, several areas, architectural elements, sculpture, mosaic, marble, tile, pottery (terra sigillata, African Red Slip, dolium), inscription. Fidenae site 267. Villa: platforms, enclosed spaces, reticulatum, tile, pottery (black gloss). Ficulea site 173a. Villa?: walls, quasireticulatum, tuff, tile, pottery (African Red Slip). Ficulea site M. Villa?: platforms, mortar, tuff, lava, tile, pottery (terra sigillata, dolium). Ficulea site 205a. Villa?: architectural elements, quadratum, mortar, travertine, limestone, tuff, peperino, pottery (terra sigillata, African Red Slip). Ficulea site R. Villa?: 200 x 100 m, spicatum, mortar, marble, tuff, brick, tile, pottery (terra sigillata, African Red Slip, amphora). Ficulea site 214. Villa: sculpture, travertine, inscription. Ficulea site S. AFF/Villa?: 120 x 90 m, platform, tile, pottery (black gloss, terra sigillata, African Red Slip), inscription. Ficulea site 228b. AFF/Villa?: owner?, 44,000 m², platform, living quarters, mosaic, reticulatum, cocciopesto, brick, inscription? Pantano 1998, 312. Ficulea site 1. AFF/Villa?: architectural elements, reticulatum, cocciopesto, travertine, tuff, brick, tile, pottery (terra sigillata, Ficulea site 228c. AFF/Villa?: architectural elements, sculpture, spicatum, quadratum, marble, travertine, peperino, tile, pottery 179 appEndiX i (African Red Slip). Ficulea site 228f. AFF/Villa?: mosaic, reticulatum, cocciopesto, tile, pottery (terra sigillata). Ficulea site 234. Villa: 40 m diameter, platform, walls, sectile, marble, tile, pottery (black gloss, dolium). Ficulea site 236a. Villa: mortar, marble, travertine, tile, pottery, inscription, connected to cemetery at site 236b. Ficulea site 491. Villa: platform, architectural elements, marble, tile, pottery. Ficulea site 501a. Villa: enclosed spaces, reticulatum, cocciopesto, tuff, brick/tile, pottery (dolium), inscription. Ficulea site 502. Villa: foundations, enclosed spaces, reticulatum, tuff. Ficulea site 504a. Villa: reticulatum, tile, pottery. Ficulea site 243. Villa rustica?: architectural elements, marble, tuff, tile, pottery (dolium, amphora), inscription, Ficulea site 522. Villa: walls, wall plaster, tuff, lava, tile, pottery (black gloss, African Red Slip). Ficulea site 296b. Villa: architectural elements, mortar, marble, travertine, limestone, tuff, brick, tile, pottery. Moscetti and La Porta 2004, 171. Ficulea site 525. Villa: signinum, incertum, reticulatum, mixtum, listatum, cocciopesto, tuff, brick/tile, pottery (terra sigillata). Ficulea site 315b. Villa?: spicatum, cocciopesto, marble, limestone, lava, brick, tile, pottery (African Red Slip, dolium, amphora), inscription. Moscetti and La Porta 2004, 172. Ficulea site 527. Ruderi: enclosed spaces, architectural elements, travertine, brick/tile/pottery. Ficulea site 316a. Villa?: owner?, wall plaster, mosaic, spicatum, mortar, cocciopesto, marble, limestone, tuff, lava, brick, tile, pottery (black gloss, thin-walled ware, terra sigillata, African Red Slip, plain/cookware, amphora), inscription. Granino Cecere and Magioncalda 2003; Moscetti and La Porta 2004, 173. Ficulea site 318b Villa?: architectural elements, mortar, tuff, peperino, brick, tile, inscription? Ficulea site 345b = Tibur III site 185. Villa: mosaic, spicatum, mortar, cocciopesto, marble, tuff, tile, pottery (black gloss, terra sigillata, African Red Slip, amphora), inscription. Ficulea site 346 = Tibur III site 186. AFF/Villa: wall, architectural elements, mortar, marble, brick, tile. Ficulea site 351a. Villa rustica: platform, wall plaster, tile, pottery (dolium). Ficulea site 358a = Tibur III site 4. Ruderi/AFF: wall, mortar, brick, tile/pottery. Ficulea site 363a = Tibur III site 1. Villa: walls, quadratum, mortar, marble, tuff, tile, pottery (African Red Slip, plain/ cookware). Ficulea site 377. Villa?: spicatum, reticulatum, cocciopesto, tuff, peperino, tile, pottery (dolium, amphora). Ficulea site 378b. Sporadic inds/Villa?: owner?, architectural elements, mosaic, mortar, marble, travertine, brick, tile, pottery, sarcophagus, inscription. Carbonara and Messineo 1992. Ficulea site 389. AFF/Villa?: 2 areas, enclosed spaces, mortar, cocciopesto, marble, tuff, brick, tile, pottery (African Red Slip). Ficulea site 404a. Villa?: mosaic, mortar, marble, tuff, brick, tile, pottery. Curti and Moscetti 1996, Nr. 17; Calci 1998, 108–9. Ficulea site 408a. Villa?: enclosed spaces, mosaic, mortar, marble, tuff, brick, tile. Ficulea site 410. Villa: wall plaster, mortar, marble, limestone, tuff, brick, tile, pottery (terra sigillata, amphora, lamp). Ficulea site 423. Villa?: owner?, 90 x 120 m, enclosed spaces, wall plaster, mosaic, reticulatum, inscription. Di Matteo 2003a. Ficulea site 427c. AFF/Villa?: foundations, quadratum, mortar, marble, travertine, tuff, lava, peperino, tile. Ficulea site 432b. AFF/Villa?: walls, mosaic, quadratum, incertum, tuff, pottery (black gloss, terra sigillata, African Red Slip). Ficulea site 442. AFF/Villa?: architectural elements, quadratum, travertine, tuff, peperino, pottery (African Red Slip), votive artifacts? Calci and Sorella 1995, 126, nota 49. Ficulea site 466a. Villa rustica?: spicatum, mortar, tuff, lava, tile, pottery. Ficulea site 469. Villa: architectural elements, quadratum, tuff, tile, pottery. Ficulea site 482. Ruderi/Villa?: platform?, enclosed spaces? Ficulea site 488b. Villa?: mortar, tuff. 180 Ficulea site 529. Villa rustica: walls, mortar, travertine, tuff, tile, pottery. Ficulea site 531. Villa: architectural elements, sculpture, wall plaster, mosaic, marble, tile, pottery (terra sigillata, African Red Slip). Ficulea site 532. Villa?: architectural elements, mosaic, mortar, peperino, tile, pottery (amphora). Ficulea site 534d. Villa: architectural elements, quadratum, mortar, marble, travertine, tuff, brick/tile, pottery (terra sigillata), cippus. Ficulea site 537a. AME/Villa?: architectural elements, mortar, marble, travertine, tile, pottery (terra sigillata). Ficulea site 539e. Villa?: walls, mortar, tuff, brick/tile. Ficulea site 539f. Villa: walls, sculpture, wall plaster, reticulatum, vittatum, cocciopesto, marble. Ficulea site 539i. Villa: architectural elements, brick/tile. Ficulea site 547c. Villa/Mansio?: owner?, enclosed spaces, reticulatum, listatum. Calci and Sorella 1995, 122–3, nota 33. Ficulea site 552. Casa rustica: mosaic, mortar, cocciopesto, tuff, brick, tile, pottery (amphora). Calci and Sorella 1995, 124, nota 36. Ficulea site 553c. AFF: 20 x 10 m and 10 x 2 m, 2 areas, wall plaster, mosaic, travertine, lava, brick, tile, pottery (black gloss, terra sigillata). Ficulea site 557b. Insediamento: enclosed spaces, reticulatum, tuff. Ficulea site 559 = Tibur III site 243. Villa: owner?, 40 x 30 m, architectural elements, sculpture, mosaic, spicatum, cocciopesto, marble, travertine, limestone, tuff, tile, pottery (black gloss, African Red Slip), inscription. Ficulea site 570a = Tibur III site 242a. Villa: sculpture, quadratum, mortar, tuff, peperino, tile. Ficulea site 577d = Tibur III site 246. AFF: enclosed spaces, architectural elements, reticulatum, marble, travertine, tuff, brick, tile, pottery. Ficulea site 580 = Tibur III site 244. Villa: 30 x 15 m, mosaic, signinum, quadratum, mortar, cocciopesto, tuff, lava, brick, tile, pottery (African Red Slip). Ficulea site 599 = Collatia site 58. Ruderi/Villa rustica?: mosaic, tile, pottery. Calci and Sorella 1998, 195 nota 26. Ficulea site 601 = Collatia site 30. Villa: owner?, storage spaces, architectural elements, sculpture, mosaic, sectile, signinum, reticulatum, marble, tuff, brick/tile, votive artifacts, inscription. De Franceschini 2005, 126–7. Ficulea site U1. Villa: 30 x 25 m, service areas?, sectile, reticulatum, listatum, cocciopesto, marble, peperino, brick. Calci and Sorella 1995, 125 nota 47; Calci 1998, 82. Tibur III site 12a. Villa rustica: architectural elements, wall plaster, mosaic, spicatum, cocciopesto, marble, tuff, pottery (black gloss, African Red Slip), inscription, cemetery at site 11b? sitE cataloguE Tibur III site 29. Villa: sculpture, mosaic, spicatum, reticulatum, marble, brick, pottery (black gloss, terra sigillata, African Red Slip), glass, inscription. Moscetti 1991, Nr. 5. Tibur III site 31a. Villa: architectural elements, sectile, spicatum, mortar, marble, travertine, limestone, tuff, brick, cemetery at site 31b. Moscetti 1991, Nr. 6. Tibur III site 64. Villa: 150 m², architectural elements, wall plaster, mosaic, sectile, mortar, travertine, tuff, brick, tile, pottery (terra sigillata, African Red Slip). Moscetti 1991, Nr. 22. Tibur III site 69. Villa: spicatum, reticulatum, marble, travertine, tuff, brick/tile, pottery (dolium). Tibur III site 70. Villa: mosaic, mortar, travertine, tuff, lava, brick/tile, pottery (black gloss, African Red Slip). Tibur III site 74. Villa: storage spaces, wall plaster, mosaic, reticulatum, mixtum, listatum, vittatum, cocciopesto, marble, tuff, brick. Moscetti 1991, Nr. 45. Tibur III site 229a. Villa: architectural elements, mosaic, signinum, spicatum, marble, brick, pottery (black gloss). Tibur III site 231. Villa: owner?, mosaic, marble, travertine, brick, pottery (black gloss, African Red Slip, amphora), inscription. Tibur III site 257. Villa: mosaic, mortar, cocciopesto, travertine, tuff, brick/tile, pottery (black gloss, dolium, amphora). Tibur III site 260. Villa?: architectural elements, mosaic, spicatum, tuff, brick/tile, pottery (black gloss, African Red Slip). Tibur III site 278a. Villa: 30 m long, natural platform, architectural elements, mosaic, reticulatum, cocciopesto, limestone, tuff, brick, pottery (black gloss), cemetery at site 278b. Tibur III site 283a. AFF: architectural elements, marble, travertine, brick, tile, pottery (plain/cookware). Tibur III site 285a. AFF/Villa?: mortar, marble, limestone. Tibur III site 85. Villa: architectural elements, spicatum, mortar, marble, travertine, brick/tile, pottery (dolium). Tibur III site 290a. Villa: owner?, architectural elements, marble, travertine, tuff, pottery, inscription, connected to Class 3 AFF at site 291? Moscetti 1991, Nr. 46. Tibur III site 90. Villa: sculpture, wall plaster, marble, travertine, pottery (black gloss, terra sigillata, African Red Slip), inscription. Tibur III site 292. Villa: architectural elements, mosaic, spicatum, signinum, marble, travertine, tuff, brick, pottery (black gloss, terra sigillata, African Red Slip), sarcophagus? Tibur III site 92a. Villa: architectural elements, mosaic, quadratum, mortar, cocciopesto, marble, travertine, tuff, brick/ tile, pottery (black gloss, terra sigillata), cemetery at site 92b (and 94?). Tibur III site 305a. Villa: 80 x 80 m, architectural elements, wall plaster, stucco, sectile, spicatum, mortar, cocciopesto, marble, travertine, limestone, tuff, brick, tile, pottery, cemetery, inscription. Di Sante and Presen 2002; Pirro 2002. Tibur III site 93a. Villa: mosaic, cocciopesto, marble, travertine, brick, tile, pottery (dolium), cemetery at site 93b. Tibur III site 319. AME/villa rustica?: architectural elements, limestone, tuff, lava, brick, tile, pottery (dolium, amphora). Tibur III site 112. Villa: architectural elements, sculpture, mosaic, cocciopesto, marble, travertine, brick/tile, inscription. Tibur III site 321. Villa: architectural elements, sectile, marble, travertine, brick, tile, pottery (black gloss, amphora). Tibur III site 116a. AME/Villa?: mosaic, brick, pottery. Moscetti 1991, Nr. 26. Tibur III site 123. Villa: architectural elements, mortar, travertine. Tibur III site 124. Villa: wall plaster, mosaic, sectile, spicatum, reticulatum, tuff, tile, pottery (black gloss, African Red Slip). Tibur III site 131. Villa: architectural elements, mosaic, sectile, spicatum, mortar, marble, limestone, tuff, brick/tile, pottery (black gloss, terra sigillata, African Red Slip), inscription. Curti and Moscetti 1996, Nr. 4. Tibur III site 135a. Villa?: architectural elements, marble, travertine, lava, brick/tile, connected to early site 134a? Tibur III site 136. AFF/Villa?: sculpture, brick/tile, pottery, sarcophagus. Tibur III site 139. Villa: mosaic, marble, travertine, lava, brick, pottery (terra sigillata, African Red Slip), coin. Moscetti 1991, Nr. 84. Tibur III site 143. AFF/Villa?: architectural elements, marble, brick/tile, pottery (black gloss, dolium), inscription. Tibur III site 149a. Villa: sculpture, wall plaster, mortar, travertine, brick/tile, inscription. Tibur III site 153. Villa: quasireticulatum?, reticulatum?, Moscetti 2002, 66 Nr. 6. architectural elements, travertine, inscription. Tibur III site 182. Villa: mosaic, mortar, marble, limestone, tuff, brick, pottery (terra sigillata, African Red Slip D). Tibur III site 207a. Villa: architectural elements, mosaic, spicatum, marble, travertine, brick, pottery (black gloss, terra sigillata, African Red Slip, dolium, lamp), glass. Tibur III site 324. Villa: architectural elements, mosaic, spicatum, marble, travertine, brick, pottery (black gloss, African Red Slip), inscription. Tibur III site 332. Villa?: walls, wall plaster, mosaic, spicatum, travertine, tuff, lava, brick. Tibur III site 344a. Villa: sectile, spicatum, reticulatum, travertine, brick/tile, pottery (black gloss, terra sigillata, African Red Slip), inscription. Tibur III site 349. Villa: mosaic, marble, travertine, brick/tile, pottery (African Red Slip A). Tibur III site 352a. AME/Villula?: 100 x 50 m, architectural elements, mosaic, marble, brick, pottery (black gloss, African Red Slip), cemetery at site 352b? Tibur III site 356. Villa: 50 x 30 m, mosaic, mortar, marble, limestone, tuff, brick, tile, pottery (black gloss, terra sigillata). Tibur III site 357a. Villa: wall plaster, reticulatum, travertine, limestone, tuff, brick, pottery. Tibur III site 381a. Villa: architectural elements, sculpture, mortar, marble, travertine, limestone, brick, tile, pottery (African Red Slip, amphora). Tibur III site 391. Ruderi/Villa?: walls, reticulatum, marble, travertine, limestone, brick, tile, pottery (dolium). Tibur III site 394. Villa: architectural elements, sectile, travertine, pottery (black gloss, African Red Slip A & B, amphora), inscription. Moscetti 1991, Nr. 90. Tibur III site 404. Villa: mosaic, mortar, limestone, brick, inscription. Moscetti 1991, Nr. 93; 2002, 67. Tibur III site 406. Villa: architectural elements, reticulatum, travertine, brick/tile, pottery (dolium, amphora). Tibur III site 211a. Villa: architectural elements, mosaic, sectile, mortar, marble, travertine, tuff, brick, tile, pottery (black gloss, terra sigillata, African Red Slip, dolium), inscription. Tibur III site 408. Villa: architectural elements, sculpture, mosaic, sectile, marble, travertine, brick, tile, pottery (black gloss, African Red Slip), inscription. Tibur III site 218. Villa?: wall plaster, mortar, marble, brick/ tile, pottery (black gloss, terra sigillata, African Red Slip), inscription. Tibur III site 412. Villa: mosaic, spicatum, mortar, marble, travertine, limestone, tuff, brick, pottery (dolium, amphora), inscription. Moscetti 1991, Nr. 91; 2002, 76; Curti and Moscetti 181 appEndiX i 1996, Nr. 11. Tibur III site 413. Villa: enclosed spaces, wall plaster, mosaic, sectile, latericium, mortar, cocciopesto, marble, travertine, limestone, tuff, brick, tile, pottery (black gloss, plain/cookware, dolium), cemetery. Moscetti 1991, Nr. 95; 2004, 177–82. Tibur III site 417. Villa: architectural elements, mosaic?, marble, travertine, brick/tile, pottery (dolium). Tibur III site 426. Villa: enclosed spaces, architectural elements, wall plaster, mosaic, marble, limestone, brick/tile, pottery (black gloss), inscription. Tibur III site 440. Villa: architectural elements, mosaic, marble, travertine, limestone, tuff, brick, pottery (dolium), coins. Tibur III site U1. Villa: cocciopesto, tuff. Moscetti 1991, Nr. 25. Tibur IV site 19a. Villa: foundations, wall plaster, sectile, polygonal, incertum, reticulatum, marble, limestone, tuff. Tibur IV site 23. Villa?: platform, polygonal, incertum, reticulatum, limestone. Tibur IV site 36. Villa: 85 x 65 m, 2 platforms, mosaic, wall plaster, incertum, reticulatum, tuff, limestone, brick/tile, inscription. Tibur I site 112. Villa?: foundations, enclosed spaces, incertum?, quasireticulatum?, reticulatum, limestone, connected to Class 1 villa at site 96? Tibur I site 119. Villa: owner?, architectural elements, sculpture, mosaic, marble. Tibur I site 137. Ruderi/Villa?: foundations?, mosaic. Tibur I site 163. Ruderi: enclosed space, wall plaster, mosaic, sectile, reticulatum, marble, tuff. Mari and Moscetti 1993, Nr. 5. Tibur I site 197 = Tibur II site 4. Villa: owner?, platforms, foundations, wall plaster, mosaic, marble, travertine. Collatia site 4c. Villa: enclosed spaces, architectural elements, sculpture, wall plaster, mosaic, mortar, marble, tuff, pottery (amphora). Collatia site 5b. Villa: architectural elements, sculpture, mosaic, marble, travertine, limestone, tuff, peperino, pottery (amphora). Pergola et al. 2003, Via Tiburtina Nr. 226. Collatia site 38a. Villa rustica: architectural elements, reticulatum, cocciopesto, marble, travertine, tuff, lava, tile, pottery. Collatia site 49b. Sito rustico: architectural elements, marble, travertine, limestone, tuff, peperino. Tibur IV site 45a. Villa: owner?, enclosed spaces, architectural elements, sculpture, incertum, reticulatum, mixtum, travertine, brick, inscription. Collatia site 64. Villa rustica: walls?, tile, pottery, inscription. Tibur IV site 49a. Villa: owner?, walls, mortar, marble, limestone, brick/tile, pottery. Collatia site 87b. Villa: architectural elements, sculpture, wall plaster, mosaic, marble, travertine, tuff, tile, pottery (dolium), inscription. Tibur IV site 65. Villa: owner?, 60 x 20 m, platform, quadratum, incertum?, quasireticulatum?, travertine. Tibur IV site 66. Villa: owner?, 55 x 20 m, platform, mosaic, polygonal, incertum?, limestone. Tibur IV site 69. Villa: platform, enclosed spaces, wall plaster, reticulatum, cocciopesto, limestone, brick, tile. Tibur IV site 76a. Villa: 250 x 200 m, platform, architectural elements, wall plaster, reticulatum, cocciopesto, marble, limestone, tuff, brick/tile, pottery (black gloss). Collatia site 79. Villa rustica: wall plaster, mortar, marble, tuff, brick, tile, pottery (plain/cookware, dolium). Collatia site 93. Villa rustica: tile, pottery. Collatia site 95a. Villa: architectural elements, wall plaster, marble, travertine, limestone, tuff, peperino, brick/tile, pottery (black gloss, lamp), inscription. Collatia site 98. Villa rustica: architectural elements, sculpture, marble, tuff, peperino, tile, pottery. Collatia site 104a. Villa rustica: walls?, wall plaster, reticulatum, tuff, tile, pottery (terra sigillata). Tibur IV site 77. Villa: foundation, reticulatum, cocciopesto, marble, travertine, tuff, brick/tile, pottery (black gloss, plain/ cookware). Collatia site 111a. Villa rustica: 100 x 55 m, platform?, enclosed spaces, tuff, lava, tile, pottery. Tibur IV site 82b. Villa: platforms? Collatia site 116. Villa rustica: platform, enclosed spaces?, mortar, tuff, tile, pottery. Tibur IV site 85. Villa: natural platform, mosaic, reticulatum, cocciopesto, marble, tuff, brick/tile, pottery (black gloss, African Red Slip), inscription. Tibur IV site 101a. Villa: 100 x 45 m, platform, enclosed spaces, incertum, reticulatum, cocciopesto, tuff. Tibur IV site 137a. Villa: enclosed spaces, architectural elements, mosaic, sectile, incertum, reticulatum, cocciopesto, marble, travertine, limestone, tuff, brick, tile, inscription. Tibur IV site 145. Villa: 4 areas, architectural elements, reticulatum, travertine, tuff, brick, inscription. Mari 1996, Nr. 2; Moscetti 2002, 67. Tibur IV site 164. Villa: natural platform, platform, walls, wall plaster, reticulatum, tuff. Tibur IV site 167. Villa: wall plaster, reticulatum, travertine, tuff, pottery (black gloss), inscription. Tibur IV site 168. Villa: wall plaster, incertum, cocciopesto, marble, travertine, tuff, pottery, inscription. Tibur IV site 196. Villa: owner?, architectural elements, sculpture, mosaic, sectile, spicatum, quadratum, marble, travertine, tuff, pottery (terra sigillata), inscription. Tibur IV site 200. Villa: living quarters, sculpture, reticulatum, cocciopesto, travertine, tuff, pottery (terra sigillata, African Red Slip, plain/cookware). Tibur IV site 214. Villa: platform, architectural elements, spicatum, mortar, cocciopesto, marble, travertine, lava. Tibur IV site 219a. Villa: platform, travertine, brick/tile. 182 Collatia site 113. Villa rustica: 90 x 40 m, platform. Collatia site 120. Villa: 170 x 80 m, platforms, enclosed spaces, travertine, tile, pottery. Collatia site 129. Villa rustica: architectural elements, mortar, cocciopesto, travertine, tuff, lava, brick, tile, pottery (African Red Slip). Collatia site 131. Villa: 100 x 120 m, platform?, walls, travertine, tile, pottery (African Red Slip). Collatia site 133. Villa: platform, quadratum, marble, travertine, tuff, brick/tile, pottery (African Red Slip). Collatia site 134a. Villa: 70 x 60 m, enclosed spaces, wall plaster, quadratum, marble, travertine, limestone, tuff, lava, brick, tile, pottery. Collatia site 136. Villa: 100 x 300 m?, enclosed spaces, mortar, marble, travertine, tuff, tile, pottery (African Red Slip, plain/ cookware, amphora). Collatia site 141a. Villa rustica: wall plaster, brick/tile, pottery (amphora). Collatia site 142. Villa: architectural elements, reticulatum, cocciopesto, marble, tuff, brick, tile, pottery (terra sigillata). Collatia site 155. Sito antico/Villa?: architectural elements, marble, travertine, tuff, tile, pottery (black gloss). Collatia site 157. Villa rustica: cocciopesto, tile, pottery. Collatia site 170b. Villa rustica: architectural elements, sculpture, marble, tile, pottery, inscription. sitE cataloguE Collatia site 174a. Casa rustica: enclosed space?, brick/tile/ pottery. Collatia site 349a. Villa: 110 x 80 m, enclosed space, sculpture, mosaic, quadratum, tuff, tile, pottery (amphora). Collatia site 177T.a Villa: walls, mosaic. Collatia site 351. Villa rustica: space?, spicatum, mortar, tuff, lava, peperino, tile, pottery (terra sigillata). Collatia site 180a. Villa: architectural elements, wall plaster, marble, tuff. Collatia site 180h.1 Villa: 120 x 100 m, mosaic, reticulatum, marble, pottery (dolium), inscription. Collatia site 180o.1/4 Villa/Burial?: foundations, walls?, reticulatum, tuff, lava, brick, tile, pottery. Qulici Gigli 1987. Collatia site 186a. Villa rustica: spicatum, cocciopesto, marble, travertine, tuff, tile, pottery. Collatia site 187a. Villa rustica: 120 x 80 m, walls, architectural elements, reticulatum, marble, travertine, tuff, tile, pottery (African Red Slip). Collatia site 195b. Villa:100 x 80 m, platform, tile, pottery. Collatia site 198a. Villa: 110 x 40 m, enclosed spaces, architectural elements, sculpture, marble, travertine, tuff, lava, tile. Collatia site 201a. Aggregato rustico/Villa?: walls, sculpture, spicatum, signinum, mortar, tile. Collatia site 227. Casa rustica: platform?, tile, pottery (terra sigillata). Collatia site 230a. Villa rustica: architectural elements, mortar, cocciopesto, marble, tuff, lava, peperino, tile, pottery (terra sigillata). Collatia site 236. Villa rustica: platform?, enclosed space, tile, pottery (amphora). Collatia site 248. Villa rustica: architectural elements?, sectile?, marble, travertine, tuff, peperino, brick, tile, pottery (amphora). Rea 1985b, 121. Collatia site 254a. Villa: 50 x 20 m, platform, enclosed spaces. Collatia site 256a. Villa rustica: walls, travertine, tile, pottery (plain/cookware, amphora), cippus. Collatia site 260. Villa rustica: 110 x 90 m, enclosed spaces, reticulatum, tuff, lava, tile, pottery. Collatia site 261d. Villa rustica: wall plaster, mortar, marble, tuff, lava, peperino, brick, tile, pottery (impasto, terra sigillata, African Red Slip). Collatia site 267. Villa rustica: architectural elements, marble, tuff, tile, pottery (terra sigillata, plain/cookware). Collatia site 269. Villa rustica: sectile?, spicatum, mortar, marble, tuff, lava, tile, pottery (African Red Slip, dolium), cippus. Collatia site 280a. Villa rustica: reticulatum, tuff, lava, tile, pottery. Collatia site 285b. Villa rustica: mosaic, mortar, cocciopesto, tuff, lava, brick/tile/pottery. Collatia site 301. Villa rustica: living quarters, architectural elements, sculpture, wall plaster, mosaic, signinum, mortar, marble, travertine, tuff, lava, peperino, tile, pottery (amphora), inscription. Musco 2001, 208; Musco and Cima 2006. Collatia site 305. Villa rustica: wall plaster, quadratum, mortar, cocciopesto, marble, travertine, tuff, lava, peperino, brick, tile, pottery (amphora), lead artifact. Collatia site 326a. Villa rustica: architectural elements, mortar, tuff, lava, peperino, tile, pottery. Collatia site 363. Villa rustica: foundations, walls, reticulatum, tuff, tile, pottery. Collatia site 368. Sito rustico/Villa?: architectural elements, mosaic, spicatum, signinum, marble, travertine, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, dolium, amphora). Collatia site 369a. Villa rustica: wall plaster, mosaic, sectile, spicatum, signinum, quadratum, mortar, cocciopesto, marble, tuff, lava, peperino, tile, pottery (terra sigillata, African Red Slip, plain/cookware, amphora), glass. Collatia site 376. Casa rustica: mosaic, sectile, signinum, reticulatum, marble, tuff, tile, pottery (black gloss, African Red Slip, plain/cookware, dolium, amphora). Collatia site 379a–b. AFF/Villa?: wall plaster, marble, lava, tile, brick/tile, pottery (impasto, terra sigillata, African Red Slip, plain/cookware, amphora), glass, inscription. Collatia site 380. Villa rustica: 160 x 50 m, 2 areas, platforms, enclosed spaces, mortar, tuff, brick/tile, inscription? Collatia site 383. Villa rustica: architectural elements, spicatum, reticulatum, marble, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware). Collatia site 385. Villa rustica: enclosed spaces, cocciopesto, tile, pottery (terra sigillata, plain/cookware). Collatia site 387a–b. Sito rustico/Villa?: sculpture, wall plaster, mosaic, sectile, spicatum, signinum, reticulatum, marble, lava, tile, pottery (terra sigillata, African Red Slip, plain/cookware, dolium, amphora, lamp), glass, cippus, cemetery?, inscription. Collatia site 396. Villa rustica: architectural elements, mosaic, spicatum, signinum, mortar, cocciopesto, marble, travertine, tuff, lava, peperino, brick, tile, pottery (terra sigillata, African Red Slip, plain/cookware, dolium, amphora), inscription. Collatia site 403. Villa rustica: architectural elements, mosaic, spicatum, mortar, marble, tuff, lava, peperino, tile, pottery (black gloss, terra sigillata, African Red Slip, plain/cookware, amphora), inscription. Collatia site 407. Villa rustica: architectural elements, wall plaster, mosaic, sectile, spicatum, signinum, mortar, marble, travertine, limestone, tuff, lava, tile, pottery (terra sigillata, African Red Slip, amphora, lamp). Collatia site 409a. Villa: 80 x 40 m, enclosed spaces, architectural elements, reticulatum, tuff, lava, peperino, tile, pottery. Neudecker 1988, 214–5. Collatia site 409b. Villa: sculpture, mosaic, marble, mortar, tuff, tile, pottery, inscription. Neudecker 1988 214–5. Collatia site 410. Villa rustica: reticulatum, cocciopesto, tile, pottery (African Red Slip, dolium), inscription? Collatia site 422b. Aggregato rustico/Villa?: marble, tuff, lava, tile, pottery (terra sigillata), inscription. Collatia site 437. Villa: reticulatum, marble, travertine, tuff, peperino, tile, pottery (amphora). Collatia site 445c. Villa: 80 m long, foundation, walls, mortar, travertine, tuff, brick, tile, pottery (terra sigillata, amphora). Collatia site 331. Villa rustica: mosaic, tuff, peperino, tile. Collatia site 456. Villa rustica: architectural elements, mortar, cocciopesto, travertine, tuff, peperino, tile, pottery (dolium, amphora). Collatia site 343a. Villa rustica: architectural elements, mosaic, spicatum, cocciopesto, marble, travertine, tuff, lava, peperino, tile, pottery, road? Collatia site 465. Villa rustica: 100 x 100 m, enclosed spaces, architectural elements, wall plaster, spicatum, quadratum, mortar, travertine, tuff, peperino, tile, pottery (amphora). Collatia site 345. Villa rustica: enclosed spaces, architectural elements?, quadratum, reticulatum, cocciopesto, tuff, lava, peperino, brick/tile, pottery (amphora), burial. Musco and Delino 2002. Collatia site 476b. Villa rustica: enclosed space, marble, brick/ tile/pottery. Collatia site 488a. Villa: 100 m long, quadratum, reticulatum, cocciopesto, marble, travertine, tuff, lava, brick, tile, pottery 183 appEndiX i (amphora). Collatia site 507b. Villa: 3 areas, architectural elements, sculpture, mosaic, mortar, marble, travertine, limestone, tuff, lava, peperino, tile, pottery (terra sigillata, plain/cookware), sarcophagus, cippus, inscription. Collatia site 528a. Villa rustica: architectural elements, spicatum, quadratum, mortar, cocciopesto, tuff, lava, peperino, tile, pottery (African Red Slip). Collatia site 551a. Villa: owner?, architectural elements, marble, tuff, peperino, tile, pottery, inscription. Collatia site 554. Villa rustica: 180 x 60 m, platform. Collatia site 556a. Villa rustica: living quarters, grotto, marble, tile, pottery (lamp). Collatia site 559a–b. Villa rustica/Casa rustica: walls?, architectural elements, marble, brick/tile/pottery. Collatia site 562. Villa: architectural elements, wall plaster, spicatum, reticulatum, cocciopesto, marble, travertine, tuff, lava, peperino, brick, tile, pottery (dolium), inscription. Devoti 1978, Nr. 82. Collatia site 568b. AFF/Villa?: architectural elements, marble, brick/tile/pottery. Collatia site 569e. Villa?: enclosed spaces, architectural elements, mortar, marble, travertine, tuff, lava, peperino, brick, tile, pottery (terra sigillata). Collatia site 569g. AFF/Villa?: wall, grotto, architectural elements, wall plaster, mortar, marble, travertine, lava, brick, tile, pottery (African Red Slip). Collatia site 655d. Villa: foundations, mortar, travertine, lava, brick/tile, burial? Collatia site 659a. Villa: platform?, mortar, lava, brick/tile. Collatia site 660b. Villa: enclosed space, reticulatum, marble, tuff, lava. Rea 1985a; Pergola et al. 2003, Via Labicana Nr. 266. Collatia site 662c. Villa?: 60 x 32 m, platform, architectural elements, reticulatum, tuff, lava, peperino, tile, pottery (black gloss). Rea 1985a?; Caruso et al. 1998, 288; Pergola et al. 2003, Via Labicana Nr. 267. Collatia site 663a. Villa: architectural elements, sculpture, wall plaster, mosaic, mixtum, marble, road, colombarium. Rea 1985a. Collatia site 663b. Villa: living quarters, architectural elements, wall plaster, mosaic, reticulatum, vittatum, latericium, cocciopesto, brick, tile, inscription. Rea 1985a?; De Franceschini 2005, 188–90? Collatia site 688. Villa rustica: mosaic, spicatum, tile, pottery, inscription. Collatia site 692. Villa rustica: platforms?, tile. Collatia site 704. Villa: enclosed spaces, architectural elements, mixtum, marble, peperino, brick, pottery. Collatia site 706. Villa rustica: 2 platforms, tile, pottery. Collatia site 731. Villa rustica: 220 x 100 m, sculpture, tile, pottery. Collatia site 739. Villa: architectural elements, wall plaster, mosaic, mortar, cocciopesto, marble, tuff, tile, pottery (terra sigillata). Collatia site 570c. Villa rustica: architectural elements, marble, travertine, tuff, brick, tile, pottery (amphora). Collatia site 740. Villa: architectural elements, sculpture, marble, tuff, peperino, brick/tile, pottery (olla perforata), sarcophagus, inscription. Collatia site 578. Villa: architectural elements, sculpture, marble, travertine, tuff, peperino, tile, pottery (terra sigillata). Collatia site 748. Villa rustica: architectural elements, spicatum, mortar, marble, tuff, peperino, tile, pottery (terra sigillata, amphora), cippus. Collatia site 581. Villa: walls, wall plaster, mosaic, reticulatum, tuff, brick, tile, pottery, inscription. Collatia site 590c. Villa: platform, marble, brick, pottery, inscription. Collatia site 592. Villa rustica: wall plaster, mosaic, tile, pottery, inscription. Collatia site 599a. Villa rustica: enclosed spaces, architectural elements, marble, tuff, tile, pottery. Collatia site 600c. Villa: platform, walls? Collatia site 750. Villa rustica: 90 x 60 m, architectural elements, wall plaster, spicatum, mortar, cocciopesto, marble, tuff, peperino, brick, tile, pottery (terra sigillata, dolium). Collatia site 756a–b. Villa rustica: architectural elements, quadratum, mortar, cocciopesto, marble, travertine, tuff, lava, peperino, brick, tile. Collatia site 756c. Villa rustica: architectural elements, quadratum, mortar, marble, travertine, tuff, lava, peperino, brick/tile, pottery (terra sigillata), cippus, inscription. Collatia site 603a. Villa: architectural elements, sculpture, wall plaster, mosaic, mixtum, cocciopesto, marble, tuff, lava, peperino, tile, pottery (plain/cookware, lamp), coins. Morelli and Musco 1984. Collatia site 771b. Aggregato rustico/Villa?: 300 x 30 m, walls, mortar, tuff, lava, brick, tile, pottery (terra sigillata, plain/ cookware, amphora), inscription. Musco 1984. Collatia site 613b. Villa rustica: platform?, foundations, living quarters, wall plaster, sectile?, spicatum, quadratum, reticulatum, cocciopesto, marble, travertine, tuff, lava, brick, tile, pottery (terra sigillata), cippus. Morelli and Musco 1984. Collatia site 784. Villa: owner?, architectural elements, sculpture, wall plaster, mosaic, signinum, mortar, marble, travertine, tuff, lava, peperino, brick, tile, pottery (African Red Slip, plain/cookware), coins, inscription. Devoti 1978, 103. Collatia site 614. Villa rustica: architectural elements, mosaic, marble, travertine, tile, pottery (amphora), bones, shells. Collatia site 788a. Villa: architectural elements, wall plaster, reticulatum, marble, travertine, lava, tile, pottery (black gloss, African Red Slip). Collatia site 625b. Villa?: uncertain site. Collatia site 627a. Sito rustico/Villa?: walls, reticulatum, tile, pottery. Collatia site 796. Villa rustica: platform, mortar, tile, pottery. Collatia site 629. Villa: enclosed space, sculpture, listatum, tuff, lava, brick, tile, pottery (African Red Slip). Collatia site 798a. Villa: 210 x 150 m, grotto, architectural elements, wall plaster, mosaic, mortar, marble, tuff, lava, peperino, brick, tile, pottery (African Red Slip), inscription. Devoti 1978, Nr. 85. Collatia site 630. Villa: 10–20 x 10–20 m, enclosed spaces, spicatum, reticulatum, listatum, tile, burial? Collatia site 799. Villa: platforms?, cocciopesto, marble, tile, pottery (terra sigillata). Collatia site 641a. Villa rustica: tuff, lava, tile, pottery (dolium). Collatia site 642b. Villa: architectural elements, sculpture, marble, travertine, tile, pottery. Collatia site 802. Villa: architectural elements, sculpture, marble, brick, tile, pottery, sarcophagus, inscription. Devoti 1978, Nr. 84. Collatia site 650. Villa rustica: mortar, marble, brick, tile. Collatia site 809a. Villa: owner?, architectural elements, peperino, tile, pottery, inscription. Collatia site 653b. Villa rustica: marble, travertine, lava, peperino, brick/tile/pottery. Collatia site 823a. Villa rustica: architectural elements, wall plaster, cocciopesto, marble, brick, tile, pottery, inscription. 184 sitE cataloguE Collatia site 823b. Villa rustica: architectural elements, spicatum, mortar, marble, tile, pottery (black gloss, terra sigillata, dolium), coins, inscription. Tellenae site 133. Villa: spicatum, reticulatum, cocciopesto, brick, tile, pottery (black gloss, terra sigillata, dolium, amphora). Quilici 1969, Nr. 2220. Collatia site 825b. Villa rustica: architectural elements, wall plaster, marble, travertine, tuff, lava, peperino, tile, pottery (amphora). Tellenae site 135b. Villa: mortar. Quilici 1969, Nr. 1058. Collatia site 829b. Villa: architectural elements, spicatum, marble, travertine, lava, peperino, tile, pottery. Collatia site 832b. Villa: 250 x 100 m, architectural elements, marble, travertine, tuff, lava, brick/tile, pottery (terra sigillata, plain/cookware). Collatia site 835a. Villa rustica: architectural elements, reticulatum, cocciopesto, tuff, peperino, tile, pottery, coins, inscription. Devoti 1978, Nr. 81. Collatia site 838. Villa: architectural elements, mosaic, mortar, marble, travertine, tuff, lava, peperino, tile, pottery, inscription. Collatia site 848a. Villa: spicatum, mortar, marble, tuff, brick, tile, pottery (terra sigillata, amphora). Devoti 1978, Nr. 78. Collatia site 848b. Villa: mortar, cocciopesto, tuff, lava, brick, tile, pottery (African Red Slip). Collatia site 851. Villa: architectural elements, quadratum?, mortar, marble, travertine, tuff, lava, peperino, tile, pottery (terra sigillata). Collatia site 854. Villa: 70 x 75 m, platforms, mortar, lava, peperino, brick/tile/pottery. Devoti 1978, Nr. 76. Collatia site U1. Villa: enclosed spaces, wall plaster, mosaic, reticulatum, tuff, brick. Messineo and Sorella 1991; Messineo 1992; Calci and Mari 2003, 182. Tellenae site 14b. Villa?: foundations, brick/tile. Quilici 1969, Nr. 1115. Tellenae site 24b–c. Villa: wall plaster, mosaic, spicatum, mortar, cocciopesto, marble, peperino, brick, tile, pottery (black gloss, terra sigillata, dolium, amphora), inscription. Quilici 1969, Nr. 1101; Cecchini et al. 1990, 120. Tellenae site 30. Villa: mosaic, wall plaster, peperino, brick, tile, pottery (black gloss, terra sigillata, dolium, amphora), inscription. Quilici 1969, Nr. 1598; Cecchini et al. 1990, 120. Tellenae site 32. Villa: wall plaster, mosaic, spicatum, reticulatum, tuff, peperino, brick, tile, pottery (terra sigillata, plain/cookware, dolium). Tellenae site 42. Villa: mosaic, spicatum, signinum, tuff, brick, tile, pottery (black gloss, terra sigillata, dolium), bronze, sarcophagus, inscription. Tellenae site 46b. Villa? Quilici 1969, Nr. 1994; Bellini 1985, 127. Tellenae site 55. Villa: mosaic, reticulatum, listatum, brick, tile, pottery (black gloss, terra sigillata, dolium), inscription. Quilici 1969, Nr. 1085. Tellenae site 64a. Villa: architectural elements, mosaic, mortar, marble, tuff, peperino, brick, tile, pottery (plain/cookware, dolium, amphora), inscription. Tellenae site 66. Villa: architectural elements, wall plaster, mosaic, spicatum, mortar, brick, tile, pottery (terra sigillata, dolium, amphora), inscription. Quilici 1969, Nr. 1980. Tellenae site 80. Villa: wall plaster, mosaic, spicatum, mortar, brick, tile, pottery (terra sigillata, dolium, amphora), inscription. Tellenae site 81a. Villas?: inscription? Quilici 1969, Nr. 1918. Tellenae site 85. Villa: mosaic, spicatum, mortar, brick, tile. Tellenae site 102. Villa: foundations, tuff. Quilici 1969, Nr. 1067. Tellenae site 103. Villa: wall plaster, spicatum, reticulatum, peperino, brick, tile, pottery (black gloss, terra sigillata, dolium, amphora). Quilici 1969, Nr. 1066. Tellenae site 126. Villa: foundations, quadratum, tuff, peperino, tile, pottery. Tellenae site 157. Villa: mosaic, peperino, brick, tile, pottery (black gloss, terra sigillata, dolium, amphora), inscription. Quilici 1969, Nr. 1897. Tellenae site 166. Villa: mosaic, spicatum, reticulatum, tuff, peperino, brick, tile, pottery (black gloss, terra sigillata, dolium, amphora), inscription. Quilici 1969, Nr. 1941. Tellenae site 171. Villa: architectural elements, mortar, marble, brick, tile, pottery (amphora). Quilici 1969, Nr. 1936. Tellenae site U5. Villa: pottery. Cecchini et al. 1990, 120. Bovillae site 29. Villa: sculpture, spicatum, mortar, brick, tile, pottery (African Red Slip). Bovillae site 32. Fattoria: spicatum, mortar, cocciopesto, marble, travertine, tuff, lava, peperino, brick, tile, pottery (terra sigillata, plain/cookware), inscription. Bovillae site 41. Villa: 10,000 m², walls, mosaic, signinum, quadratum, incertum, reticulatum, cocciopesto, lava, brick, tile, pottery (plain/cookware). Corrente 1988a, 400, nota 5; LTURS s.v. Latina via, VII miglio. Bovillae site 46. Villa: 4,650 m², mosaic, reticulatum, mixtum, marble, brick, tile, pottery (plain/cookware), inscription. Corrente 1988a, 400, nota 6; LTURS s.v. Latina via, VII miglio. Bovillae site 70. Villa?: architectural elements, quadratum, brick/tile. Bovillae site 74–75. Ruderi/Villa: architectural elements, wall plaster, reticulatum, cocciopesto, marble, tuff, peperino, brick/ tile. Bovillae site 79. Villa/Fattoria: architectural elements, mortar, marble, tuff, peperino, brick/tile, inscription. Bovillae site 83. Villa: living quarters, peperino. Bovillae site 145. Villa: architectural elements, sculpture, wall plaster, mosaic, spicatum, reticulatum, marble, travertine, peperino, brick, tile, pottery (terra sigillata, dolium, amphora). Bovillae site 152a. Fattoria: walls, reticulatum, cocciopesto, marble, peperino, brick, tile, pottery (dolium, amphora). Bovillae site 157. AFF/Fattoria?: architectural elements, reticulatum, marble, peperino, brick, tile, pottery (dolium). Bovillae site 166. Villa: enclosed spaces, architectural elements, reticulatum, mixtum, marble, peperino, brick, tile. Bovillae site 171. Villa: wall plaster, architectural elements, sculpture, marble, peperino, tile, inscription. Bovillae site 172. Villa: architectural elements, sculpture, marble, peperino. Bovillae site 201. Villa: architectural elements, sculpture, mosaic, quadratum, reticulatum, marble, lava, peperino, brick, tile, inscription. Bovillae site 220. Villa: owner?, architectural elements, sculpture, wall plaster, stucco, mosaic, reticulatum, mixtum, marble, travertine, peperino, brick, tile, pottery (dolium), votive artifacts? Spera and Mineo 2004, 179. Bovillae site 245. Villa: architectural elements, wall plaster, reticulatum, marble, lava, peperino, brick, tile, pottery (terra sigillata, dolium). Bovillae site 250. Villa: platform, mixtum, peperino, brick/tile. Bovillae site 264a. Villa? Bovillae site 275a–276. Villa: owner?, foundations, walls, inscription. Bovillae site 296. Villa: architectural elements, quadratum, reticulatum, peperino, brick, tile, pottery (dolium). Bovillae site 302b–303. Villa: foundation, walls, quadratum, mortar, marble, lava. 185 appEndiX i Bovillae site 304.1. Villa: walls, architectural elements? Bovillae site 304.4. Villa?: walls, wall plaster, mosaic. Bovillae site 304.7b. Villa: wall, reticulatum. Bovillae site 304.19. Villa: living quarters?, sculpture, mosaic, marble, pottery (dolium), inscription. Bovillae site 311. Villa: enclosed spaces, architectural elements, sculpture, quadratum, reticulatum, marble, peperino, votive artifacts, inscription. Bovillae site 313. Villa: 100 x 45 m, living quarters, architectural elements, reticulatum, listatum, marble, peperino, pottery (dolium). Bovillae site 317. Villa: 200 x 150 m, platforms, brick/tile/ pottery. Bovillae site 319a. Villa: owner?, cryptoporticus, marble, peperino. Bovillae site 332a. Villa: cryptoporticus, architectural elements, mortar, marble, lava, peperino. Bovillae site 333b. Villa: architectural elements, peperino, brick, tile, pottery (plain/cookware). Bovillae site 335b. Villa: enclosed spaces, reticulatum, mixtum, cocciopesto, limestone, lava, brick/tile. Bovillae site 339. Villa: cryptoporticus?, architectural elements, wall plaster, marble, peperino, brick, tile, pottery (terra sigillata, dolium). Bovillae site 356. Villa: reticulatum, cocciopesto, marble, tuff, peperino, brick, tile, pottery (dolium), burial? Bovillae site 364a. Villa?: enclosed spaces. Bovillae site 371. Villa: walls, wall plaster, reticulatum, listatum, cocciopesto, marble, tuff, peperino, brick, tile, pottery (black gloss, terra sigillata, dolium). cryptoporticus?, living quarters?, architectural elements, mortar, marble, travertine, peperino, brick/tile, connected to Class 2 villa at site 60–62? Tusculum site 72. Villa/Fattoria?: platform?, mosaic?, polygonal. Tusculum site 74–75. Villa: enclosed spaces, architectural elements, sculpture, incertum?, quasireticulatum?, marble, travertine, lava, peperino. Tusculum site 79–82. Villa: 120 x 100 m, platform, foundation, cryptoporticus, architectural elements, wall plaster, mosaic, polygonal, incertum, marble, tuff, lava, brick, pottery (terra sigillata, African cookware, plainware). Tusculum site 101–105. Villa?: platform, enclosed space, architectural elements, mosaic, reticulatum. Tusculum site 116–117. Villa: 100 m long?, platform, foundations, reticulatum, cocciopesto. Tusculum site 130–132. Villa: enclosed spaces, architectural elements, sculpture, sectile, wall plaster, reticulatum, marble, brick, pottery (lamp), inscription? Tusculum site 133. Villa: platform, foundations, enclosed spaces, mosaic, wall plaster, mortar, cocciopesto, limestone, lava, tile, pottery (black gloss, African Red Slip C, plain/ cookware). Tusculum site 143–148. Villa?: 150 x 150 m?, platform?, foundations, enclosed spaces, incertum, mixtum, latericium, lava, brick. Tusculum site 149. Villa?: platform? Tusculum site 151. Villa: owner?, 180 x 130 m, platform, foundations, reticulatum, cocciopesto, lava. Tusculum site 156–158. Villa?: platform, walls, incertum, reticulatum, lava, connected to Class 1 villa at site 154? Bovillae site 374. Villa: architectural elements, sculpture, mosaic, reticulatum, listatum, marble, lava, peperino, brick, tile, pottery (dolium), inscription. Tusculum site 167–171. Villa: 180 x 90 m, platform, sculpture, polygonal, incertum, reticulatum, cocciopesto, marble, lava, brick/tile, inscription. Bovillae site 382. Fattoria: foundation, quadratum, peperino, brick. Tusculum site 174. Villa?: platform, reticulatum. Bovillae site 383. Villa: architectural elements, incertum, peperino, pottery (black gloss), burials, inscription. Tusculum site 176. Villa: 130 m long, platform, mosaic, mortar, lava. Bovillae site 411. Villa: peperino. Tusculum site 177–178. Villa: platform?, foundations, architectural elements, spicatum, peperino. Bovillae site 414. Villa: foundation, quadratum, peperino, brick, tile, pottery. Tusculum site 181–182. Villa?: platform?, enclosed spaces, quadratum, incertum, lava. Bovillae site 416. Villa: sculpture, quadratum, reticulatum, marble, tuff, burial?, inscription. Tusculum site 184. Villa: 200 x 150 m, platform? Bovillae site 427. Villa: walls, architectural elements, sculpture, mosaic, sectile, spicatum, marble. Tusculum site 196–201. Villa?: walls, architectural elements, mosaic, spicatum, signinum, reticulatum, marble, lava, peperino, brick, pottery (amphora), inscription. Bovillae site U2. Villa Rustica: living quarters?, mosaic, reticulatum, vittatum, cocciopesto, cemetery? LTURS s.v. Latina via, VII miglio. Tusculum site 256–258. Villa: foundations, architectural elements, sculpture, reticulatum, marble, travertine, limestone, tuff, peperino, burial? Apiolae site 342a. Villa: sculpture, spicatum, mortar, marble, brick, tile, pottery (black gloss, terra sigillata, dolium, lamp), inscription. Tusculum site 260–264. Villa: owner?, 200 x 100 m, platform, architectural elements, sculpture, incertum, quasireticulatum, reticulatum, cocciopesto, lava, burial?, inscription. Apiolae site 343a. Villa: architectural elements, wall plaster, mosaic, spicatum, reticulatum, cocciopesto, travertine, peperino, brick, tile, pottery (black gloss, terra sigillata, dolium). Tusculum site 266–267. Villa?: enclosed space, architectural elements, sculpture, marble. Tusculum site 5–8. Villa: owner?, cryptoporticus?, enclosed spaces?, architectural elements, mosaic, reticulatum, mixtum, cocciopesto, marble, tuff, peperino. Tusculum site 46–48. Villa: 2 platforms, enclosed space, architectural elements, sculpture?, polygonal, incertum, marble, lava, peperino. Tusculum site 60–62. Villa: 150 x 150 m, enclosed space, quasireticulatum, reticulatum, cocciopesto, lava, brick, tile, pottery (black gloss, terra sigillata, plain/cookware), inscription, connected to Class 2 villa at site 63–65? Tusculum 186 site 63–65. Villa: platform, foundation, Tusculum site 394. Villa: foundations, enclosed spaces, reticulatum, lava. Tusculum site 395–396. Villa: 110 m long, platform, foundation, quasireticulatum, reticulatum, lava. Tusculum site 414–418. Villa: owner?, 60 m long?, foundations, architectural elements, sculpture, mosaic, reticulatum, marble, lava, pottery, inscription. Tusculum site 423–425. Villa?: walls, mosaic, inscription. Tusculum site 540–545. Villa: owner?, foundations, incertum, reticulatum, latericium, cocciopesto, lava, burial? Tusculum site 581–583. Villa: owner?, platforms?, wall plaster, mosaic, spicatum, quadratum?, reticulatum, latericium, sitE cataloguE cocciopesto, tuff, lava, brick, burials. Tusculum site 603–605. Villa: 150 x 100 m, architectural elements, sculpture, incertum, cocciopesto, marble, limestone, lava, peperino, sarcophagus. Tusculum site 607. AFF: architectural elements, sectile, marble, peperino. Tusculum site 616–622. Villa: owner?, 150 x 200 m, enclosed spaces, architectural elements, spicatum, reticulatum, mixtum, marble, peperino, brick/tile, burial, inscription, connected to Class 2 villa at site 623–624? Tusculum site 623–624. Villa: foundations, enclosed spaces, reticulatum, mixtum, latericium, cocciopesto, peperino, brick, connected to Class 2 villa at site 616–622? Tusculum site 631. Platform/Villa? Tusculum site 632–633. Villa: cryptoporticus, sculpture, mosaic, inscription, burial, connected to Class 1 villa at site 634–642? Tusculum site 699–701. reticulatum, lava, peperino. Villa: owner?, foundations, Tusculum site 712–715. Villa?: platform, foundations, architectural elements, mosaic, reticulatum, lava, sarcophagus? Tusculum site 743–747. Villa: 50 x 150 m, 3 platforms?, foundations, cryptoporticus?, quasireticulatum, reticulatum, mixtum, listatum, lava, brick/tile, inscription. Tusculum site 752. Villa?: foundations, incertum, mixtum. Tusculum site 774–775. cocciopesto, marble, lava. Villa?: foundations, mortar, Tusculum site 786–787. Villa: mosaic, incertum, marble, lava, brick/tile, pottery, inscription. Tusculum site 808–814. Villa: 150 x 120 m, 3 platforms, foundations, wall plaster, stucco, reticulatum, cocciopesto, lava, brick/tile, inscription. Tusculum site 820–821. Villa: foundations, enclosed spaces, reticulatum, tuff, lava. Tusculum site 823–825. Villa: platform, enclosed space, stucco, reticulatum. Tusculum site 903–905. Villa?: owner?, architectural elements, sculpture, wall plaster, reticulatum, marble, travertine, inscription. Tusculum site 907–908. Villa: platform, architectural elements, reticulatum. Tusculum site 915–917. Villa: foundations, reticulatum, marble, lava, brick/tile. Tusculum site 921. Villa: 100 x 100 m, platform, incertum, quasireticulatum, lava, peperino. Fidenae site 61b. Villa?: enclosed spaces, wall plaster, quadratum, reticulatum, brick/tile. Fidenae site 77. Villa: walls, architectural elements, mixtum, brick, tile, pottery (black gloss, dolium). Fidenae site 81a. Villa: cocciopesto, tuff, tile, inscription. Fidenae site 94. Villa rustica: enclosed spaces, architectural elements, mortar, travertine, tuff, tile, pottery (black gloss, African Red Slip). Di Gennaro et al. 2004, 111. Fidenae site 95. Villa rustica: tile, pottery (black gloss), burials. Di Gennaro et al. 2004, Nr. 1–2. Fidenae site 101b. Fabbricato: enclosed spaces, reticulatum, vittatum, cocciopesto, tuff, tile, burials. Dell’Era 2002, 250–3; Pergola et al. 2003, Via Nomentana Nr. 117. Fidenae site 113b. Villa rustica: reticulatum, cocciopesto, tuff, tile, pottery. Dell’Era 2002, 254–7; Di Gennaro et al. 2004, 111. Fidenae site 120. Villa rustica: spicatum, mixtum, marble, tile, pottery (plain/cookware, amphora), cemetery? Pergola et al. 2003, Via Nomentana Nr. 112. Fidenae site 127. Villa: reticulatum. Fidenae site 138. Ruderi: 50 m long, wall, mortar, tuff. Fidenae site 156a. Villa urbana: 300 m long, walls, reticulatum, connected to Class 2 villa at site 157a? Fidenae site 174. Villa rustica: 90 m diameter, foundation, mortar, marble, tuff, tile, pottery (black gloss, African Red Slip D), burials. Fraioli 2000, 226, 232–5; Di Gennaro et al. 2004, Nr. 72. Fidenae site 178. Villa rustica: 2,600 m², walls, tuff, tile, pottery (impasto, terra sigillata, African Red Slip D, African cookware, amphora). Fraioli 2000, 226, 232; Pergola et al. 2003, Via Nomentana Nr. 122; Di Gennaro et al. 2004, 110. Fidenae site 189. Villa rustica: enclosed spaces, mosaic, tile, pottery (African Red Slip). Di Gennaro 1990, 225–6; Pergola et al. 2003, Via Nomentana Nr. 124. Fidenae site 196. AFF: architectural elements, marble, tile, pottery. Fidenae site 199. Ruderi/Villa?: foundations, sculpture. Fidenae site 202. Villa rustica?/Burial?: architectural elements, spicatum, mortar, cocciopesto, marble, travertine, tuff, tile, pottery. Fidenae site 204. Villa: walls, reticulatum. Fidenae site 212e. Villa?: 2 areas, walls, architectural elements, sculpture, reticulatum, marble, travertine, peperino, inscription, same as burial at site 214? Pergola et al. 2003, Via Nomentana Nr. 101. Fidenae site 226. Villa rustica?: reticulatum, tuff, tile, pottery (dolium). class 3 sitEs Fidenae site 231. Villa: owner?, architectural elements, marble, lava, tile, pottery, sarcophagus, inscription? Fidenae site 20b. AFF: enclosed spaces, wall plaster, mortar, tuff, brick/tile, pottery. Fidenae site 236. Ruderi/Villa: walls, architectural elements, sculpture, marble, coins, inscription. Pergola et al. 2003, Via Nomentana Nr. 109. Fidenae site 22b. AFF: mosaic, building debris. Fidenae site 24c. Casa rustica: architectural elements, wall plaster, mortar, travertine, tuff, brick/tile, pottery. Fidenae site 26a. Villa?: enclosed spaces, quadratum, reticulatum, mixtum, tuff, brick, tile, pottery, bones. Fidenae site 32. Ruderi: owner?, architectural elements, marble, brick/tile, pottery, inscription. Fidenae site 45. AFF: architectural elements, marble, brick/tile, pottery. Fidenae site 51c. Villa?/Burial?: service areas?, architectural elements, cocciopesto, marble, inscription, connected to burials at site 51a–b? Fidenae site 53a. Ruderi/Villa?: architectural elements, marble. Fidenae site 243. Villa rustica: 100 m diameter, tile, pottery (amphora, lamp), coin, cemetery. Di Gennaro and De Filippis 1995; Pergola et al. 2003, Via Nomentana Nr. 108. Fidenae site 262. Villa rustica: architectural elements, mortar, tile, pottery. Ficulea site O.2 Villa rustica: spicatum, mortar, cocciopesto, brick, tile, pottery (African Red Slip, amphora). Ficulea site 12. Villa: 10,000 m², architectural elements, spicatum, quadratum, mortar, travertine, tuff, peperino, brick, tile, pottery (terra sigillata, amphora). Pergola et al. 2003, Via Nomentana Nr. 195. Ficulea site 43. Villa: mosaic, mortar, tile, pottery. Ficulea site 45. Villa: mosaic, mortar, marble, brick, tile, pottery (black gloss, terra sigillata). 187 appEndiX i Ficulea site 51. Villa: spicatum, marble, brick, tile, pottery (plain/cookware, amphora). Ficulea site 53. AFF: mosaic, marble, brick, tile, pottery (terra sigillata, amphora). Ficulea site 54. Villa: walls, architectural elements, wall plaster, reticulatum, marble, tuff, pottery. brick. Tibur III site 190. AFF: tile, pottery (bucchero, terra sigillata, African Red Slip), inscription. Tibur III site 291. AFF: architectural elements, travertine, pottery, connected to Class 2 villa at site 290a? Tibur III site 343a. Villa: owner?, sculpture, mortar, brick. Ficulea site 66b. Villa: 80 m diameter, mosaic, cocciopesto, marble, tuff, lava, tile, pottery (black gloss, terra sigillata, African Red Slip, amphora), same as Class 4 AFF at site 66a?, connected to burials at site 68a. Tibur III site 354. AME/Villa?: natural platform, architectural elements, travertine, brick/tile, pottery (lamp). Ficulea site 83. Villa: mosaic, mortar, marble, tile, pottery (terra sigillata, African Red Slip). Tibur III site 395. AFF: tile, pottery (terra sigillata, dolium, amphora), inscription. Ficulea site 97b. Villa rustica: architectural elements, travertine, tile, pottery. Tibur III site 398a. Villa: mosaic, inscription. Ficulea site 115. Villa rustica: spicatum, mortar, lava, tile, pottery (terra sigillata). Tibur III site 382d. Villa: owner?, incertum? Tibur III site 399. AME/Villa?: architectural elements, marble, travertine, brick, pottery (black gloss). Ficulea site 175. Villa: walls, reticulatum. Tibur III site 416. AME: architectural elements, travertine, limestone, tuff, brick/tile, pottery (black gloss). Ficulea site 179a. AFF: 43,750 m², walls, tile, pottery (impasto, terra sigillata, African Red Slip D, African cookware, amphora), connected to cemetery at site Tibur III 175b. Pergola et al. 2003, Via Nomentana Nr. 150. Tibur III site 431. Villa: architectural elements, travertine, limestone, tuff, brick/tile, pottery. Ficulea site 233. Villa?: platforms, enclosed spaces, mortar, tuff. Tibur III site 447. AME/Villula?: architectural elements, travertine, brick/tile, pottery (dolium). Ficulea site 252. Villa rustica: mortar, tile, pottery (terra sigillata), inscription. Tibur IV site 46. AFF: wall plaster, mosaic, signinum, marble, lava, brick, tile. Ficulea site 294 = Tibur III site 175a. Villa rustica: platform, tuff, peperino, connected to cemetery at site Tibur III 175b. Moscetti 1996, 60; Moscetti and La Porta 2004, 171. Tibur IV site 48. Villa: walls, reticulatum, travertine, pottery (black gloss). Ficulea site 306b. Villa: quadratum, mortar, travertine, limestone, tuff, tile, pottery. Moscetti and La Porta 2004, 172. Ficulea site 367 = Tibur III site 17a. Villa: 2 areas, mortar, travertine, limestone, brick, tile, pottery (black gloss, African Red Slip, dolium), connected to road at site Tibur III 17b? Ficulea site 420. AFF: quadratum, travertine, tuff, tile, pottery (black gloss, terra sigillata, amphora). Ficulea site 454b. Villa: quadratum, marble, tuff, tile, pottery (terra sigillata, African Red Slip, plain/cookware, amphora). Ficulea site 456a. Villa: spicatum, tile, pottery. Ficulea site 467a. Villa: spicatum, tile, pottery. Ficulea site 477. Villa: enclosed spaces, reticulatum. Ficulea site 546a. Casa rustica?: reticulatum, cocciopesto, tile, pottery. Calci and Sorella 1995, 124, nota 35. Ficulea site 560a = Tibur III site 232b. Villa: walls, reticulatum, travertine, tuff, lava. Ficulea site 584d = Collatia site 75b. Villa rustica: architectural elements, travertine, tuff, peperino, lava, tile, pottery, cippus, inscription. Ficulea site 605. Villa?: mosaic?, tile, pottery. Tibur III site 430. Sporadic inds/Villa?/Burial?: owner?, architectural elements, marble, jewelry? Tibur IV site 52a. Villa?: owner?, mortar, marble, limestone, brick, tile, inscription. Tibur IV site 99. Villa?: small areas, architectural elements, marble, pottery (terra sigillata), inscription, brick/tile. Tibur IV site 107. Villa?: walls, mortar, marble, brick/tile. Tibur IV site 129. Villa: walls, polygonal?, quadratum, mortar, inscription. Tibur IV site 166. Villa?: mortar, cocciopesto, inscription. Tibur IV site 199. Villa: travertine, tuff, brick/tile, pottery (black gloss). Tibur IV site 218. Villa: architectural elements, spicatum, travertine, tuff, brick/tile. Tibur I site 20. Villa: owner?, sculpture, mosaic. Tibur II site 7. Villa: owner?, sculpture, marble, inscription. Tibur II site 86. Villa: wall plaster, mosaic, signinum, mortar. Tibur II site 220. Villa: owner?, 45 x 30 m, incertum, reticulatum. Neudecker 1988, 228–9. Collatia site 2. Villa rustica: architectural elements, mortar, travertine, tuff, tile. Collatia site 11a. Villa rustica: spicatum, tuff, tile, pottery. Tibur III site 51. AME: mortar, tuff, brick, inscription. Collatia site 11b. Villa rustica: spicatum, tile, pottery. Tibur III site 76. Villa: reticulatum, mortar, cocciopesto, marble, travertine, tuff, lava, pottery (dolium). Collatia site 28. Villa rustica: architectural elements, travertine, lava, tile. Tibur III site 78a. Villa: mortar, marble, tuff, lava, brick/tile, pottery, inscription, cemetery at site 78b. Collatia site 31b. Villa rustica?: architectural elements, travertine, tuff, tile. Tibur III site 88. Villa: walls, cocciopesto, marble, travertine, brick/tile, pottery (dolium), inscription. Moscetti 1997, 145. Collatia site 52. Villa rustica: enclosed space. Collatia site 62b. Casa rustica: architectural elements, marble, tile, pottery. Tibur III site 102. Villa: 40 x 30 m, cocciopesto, tuff, tile, pottery (black gloss, African Red Slip, dolium), lead, jewelry, connected to burials at site 103? Collatia site 82e. Ruderi: enclosed space, mortar, cocciopesto, tuff. Tibur III site 128. Villa: spicatum, travertine, brick/tile, pottery. Collatia site 99. Villa rustica: tile, pottery, cippus. Tibur III site 141. Villa: 150 m², mosaic, marble, brick, pottery (black gloss, terra sigillata, African Red Slip). Collatia site 109a. Villa rustica: reticulatum, travertine, tile, pottery. Tibur III site 145a. AME: mosaic, limestone, brick/tile, pottery. Collatia site 122. Villa rustica: walls, reticulatum, travertine, peperino. Tibur III site 155a. Villa: walls, mortar, marble, travertine, 188 sitE cataloguE Collatia site 139. Aggregato rustico: 300 m long, walls, mortar, marble, travertine, tuff, lava, brick, tile, pottery (terra sigillata, dolium). Musco 2001, 193. Collatia site 447c. Villa rustica: tile, inscription? Collatia site 140a. Villa rustica: 90 x 50 m, enclosed spaces, tile, pottery. Collatia site 449b. Sito antico: owner?, marble, pottery (terra sigillata), inscription. Collatia site 165. Villa rustica: mortar, cocciopesto, travertine, tuff, lava, peperino, brick, tile, pottery, cippus. Collatia site 178. Villa rustica: quadratum, lava, peperino, tile, pottery (black gloss, African Red Slip). Collatia site 180b. Villa: enclosed spaces, mortar, tile, pottery. Collatia site 182. Villa rustica: 160 m long, enclosed spaces?, mortar, tile, pottery (terra sigillata). Collatia site 185a. Villa rustica: architectural elements, travertine, brick/tile/pottery. Collatia site 188. Villa rustica: 160 x 160 m?, enclosed spaces, tile, pottery. Collatia site 196b. AME: architectural elements, mortar, marble, travertine, tuff, tile. Collatia site 207. Villa rustica: 40 m long, enclosed spaces, brick/tile/pottery. Collatia site 208. Villa rustica: 60 x 25 m and 70 x 120 m, enclosed spaces, tuff, lava, tile, pottery. Collatia site 210a. Villa rustica: 90 x 60 m, walls. Collatia site 221a. Aggregato rustico: reticulatum, tuff, lava, tile, pottery (dolium). Collatia site 233. Villa rustica: 60 x 40 m, walls, tile, pottery. Collatia site 244. Casa rustica: enclosed spaces, tile. Collatia site 274. Villa rustica: walls, mortar, marble. Montalcini De Angelis d’Ossat 1983, 30–1; Musco and Zaccagni 1985, 105–6. Collatia site 294. Casa rustica: architectural elements, marble, travertine, tile, pottery. Collatia site 299a. Casa rustica: architectural elements, travertine, peperino, tile. Musco 2001, 208. Collatia site 303. Villa rustica: spicatum, reticulatum, travertine, tuff, peperino, tile. Collatia site 306a. Villa rustica: walls, quadratum, mortar, tuff, peperino, lava, tile. Collatia site 317. Villa rustica: quadratum, tuff, lava, peperino, brick, tile, pottery (amphora). Collatia site 340a. Villa rustica: 200 x 60 m, enclosed spaces, tuff, peperino, tile, lava, pottery (terra sigillata). Collatia site 362c. AFF: pottery, inscription. Collatia site 367. Villa rustica: walls, mortar, tuff, lava, peperino, tile, pottery (African Red Slip). Collatia site 371. Villa rustica: spicatum, quadratum, mortar, tuff, lava, tile, pottery (terra sigillata). Collatia site 382a. Villa rustica: quadratum, mortar, cocciopesto, marble, travertine, tuff, lava, peperino, tile, pottery (African Red Slip), inscription. Collatia site 382b. AFF: quadratum, mortar, cocciopesto, marble, travertine, tuff, peperino, tile. Collatia site 397. Aggregato rustico/Villa: mosaic, mortar, marble, tuff, tile, pottery (black gloss, plain/cookware, terra sigillata, African Red Slip, amphora). Collatia site 413a. Villa rustica: architectural elements, mortar, tuff, tile, pottery. Collatia site 416a. Sito rustico: mortar, tuff, tile, pottery (terra sigillata, amphora), inscription. Collatia site 428. Villa: platforms, mortar, travertine, tuff, tile, pottery (African Red Slip). Collatia site 429. Villa rustica: spicatum, travertine, tile, pottery (amphora). Collatia site 448c. Villa rustica: mortar, tuff, brick/tile, same as Class 4 AFF at site 448b? Collatia site 452. Villa rustica: architectural elements, marble, brick/tile/pottery. Collatia site 466. Sito rustico: tile, pottery, cippus, inscription. Collatia site 470. Villa rustica: architectural elements, limestone, tile, pottery (amphora). Collatia site 490. Villa rustica: brick, tile, pottery, inscription? Collatia site 499. Villa rustica: architectural elements, marble, travertine, peperino, tile, pottery (amphora), sarcophagus. Collatia site 509. Casa rustica: spicatum, mortar, marble, tuff, tile, pottery. Collatia site 523. Villa: architectural elements, mortar, marble, travertine, tuff, peperino, tile. Collatia site 530. Aggregato rustico: 65 x 70 m, enclosed spaces?, tile, pottery. Collatia site 546a. Villa rustica: 100 x 150 m, enclosed spaces, tile. Collatia site 558b. Villa: 70 x 30 m, enclosed spaces, quadratum, signinum, tile. Collatia site 558c. Villa: spicatum, lava, brick/tile, pottery. Collatia site 571. Villa: architectural elements, mortar, marble, tile, pottery (terra sigillata). Collatia site 589a. Villa rustica: sculpture, tile, pottery (African Red Slip, dolium, amphora). Collatia site 606. Villa rustica: architectural elements, marble, tile, pottery. Collatia site 608. Villa rustica: 220 x 60–80 m, tile, pottery (amphora), sarcophagus. Collatia site 652. Sito rustico: architectural elements, peperino, tile, pottery (amphora). Collatia site 653a. Aggregato rustico: architectural elements, peperino, tile, pottery. Collatia site 659d. AME/Villa?: marble, inscription. Collatia site 682I. AFF: mortar, brick, tile, pottery, burial? Collatia site 690. Villa rustica: architectural elements, marble, brick/tile/pottery. Collatia site 694. Casa rustica: architectural elements, marble. Collatia site 696. Villa rustica: architectural elements, mortar, marble, travertine, tuff, peperino, tile, pottery. Collatia site 698. Villa rustica: enclosed spaces, mosaic, tile, pottery (amphora). Collatia site 708. Villa rustica: sculpture, marble, tile, pottery (amphora). Collatia site 710. Aggregato rustico: architectural elements, marble, tuff, peperino, tile, pottery (amphora). Collatia site 737. Villa rustica: walls, mortar, tuff, peperino, tile. Collatia site 753a. Villa rustica: architectural elements, mortar, marble, travertine, tuff, peperino, brick, tile, pottery. Collatia site 755. Villa rustica: mortar, marble, tuff, lava, brick, tile/pottery, inscription. Collatia site 758a. Villa rustica: spicatum, mortar, tuff, brick/ tile, pottery. Collatia site 762 = Tusculum site 2. AFF/Villa rustica?: spicatum, mortar, brick, tile, pottery. Collatia site 767. Aggregato rustico: 120 x 80 m, wall plaster, mortar, marble, travertine, tuff, lava, tile, pottery (terra sigillata, dolium, amphora), glass, iron. 189 appEndiX i Collatia site 769b. Sito rustico: architectural elements, mortar, marble, tuff, lava, peperino, brick, tile, pottery (terra sigillata, amphora). Collatia site 774a–b. Villa rustica/Casa rustica: 300 m diameter, walls, travertine, tuff, peperino, lava, brick, tile, pottery (amphora). Collatia site 775a. Villa rustica: architectural elements, marble, tile, pottery. Collatia site 780b. Villa rustica: architectural elements, spicatum, marble, travertine, peperino, tile, pottery. 1898. Tellenae site 160. Villa: mosaic, mortar, peperino, brick, tile, pottery (plain/cookware, dolium, amphora). Quilici 1969, Nr. 1900. Tellenae site U4. Villa: enclosed spaces, mortar, lava. Cecchini et al. 1990, 119. Bovillae site 2. Villa rustica?: spicatum, reticulatum, listatum, mortar, travertine, peperino, brick, tile, pottery (African Red Slip, dolium). Bovillae site 4. Villa?: listatum, marble, tuff, brick/tile. Collatia site 793a. Villa rustica: architectural elements, marble, tile, pottery (black gloss). Bovillae site 5. Villa?: wall plaster, listatum, tuff, brick. Collatia site 801b. Villa: architectural elements, mortar, cocciopesto, marble, tuff, lava, peperino, tile, pottery (amphora). Bovillae site 15. AFF: enclosed spaces?, brick/tile/pottery. Collatia site 806a. Casa rustica: enclosed space, peperino, tile, pottery. Collatia site 819a. Villa rustica: spicatum, reticulatum, tuff, peperino, tile, pottery. Bovillae site 8a. AFF: enclosed spaces?, brick/tile/pottery. Bovillae site 53/U3. Ruderi/Villa?: walls, reticulatum, mixtum, latericium, burials, inscription. LTURS s.v. Latina via, VII miglio. Bovillae site 86. Villa: walls. Collatia site 819b. Villa rustica: 140 x 40 m, enclosed spaces, tile, pottery. Bovillae site 88. Fattoria: spicatum, quadratum, reticulatum, brick, tile, pottery (dolium), inscription. Collatia site 822. Villa rustica: signinum, mortar, peperino, tile, pottery (dolium). Bovillae site 93. Villa: brick/tile, pottery, inscription. Collatia site 829c. Villa rustica: architectural elements, lava, peperino, tile. Collatia site 829d. Villa: architectural elements, tuff, lava, tile, pottery. Collatia site 835b. Villa rustica: architectural elements, mortar, cocciopesto, marble, travertine, tuff, peperino, lava, tile, pottery (African Red Slip). Tellenae site 9. Sporadic material/Villa?: architectural elements, peperino. Quilici 1969, Nr. 1671. Tellenae site 11. Villa: owner?, brick, tile, pottery (terra sigillata, plain/cookware, amphora), inscription. Quilici 1969, Nr. 1119–23; Cecchini et al. 1990, 119. Tellenae site 26. Villa: spicatum, reticulatum, lava, brick, tile, pottery (plain/cookware). Quilici 1969, Nr. 2214; Cecchini et al. 1990, 120. Tellenae site 27. Villa: architectural elements, marble, pottery (black gloss, terra sigillata), inscription. Quilici 1969, Nr. 2213; Cecchini et al. 1990, 120. Tellenae site 31. Villa: reticulatum, brick, tile, pottery, inscription. Quilici 1969, Nr. 1597; Cecchini et al. 1990, 120. Tellenae site 53. Villa: architectural elements, reticulatum, listatum, tuff, brick, tile, pottery (plain/cookware, dolium). Bovillae site 114a. AFF: wall plaster, reticulatum, cocciopesto, peperino, brick. Bovillae site 120a. Villa: mosaic, mortar, inscription. Bovillae site 125. Villa: brick, inscription. Corrente 1988a, 400. Bovillae site 146. Fattoria: signinum. Bovillae site 147a. AFF: architectural elements, peperino, brick/tile/pottery. Bovillae site 183. Villa?: enclosed space, reticulatum, connected to Class 3 villa at site 186? Bovillae site 186. Villa: walls, reticulatum, connected to Class 3 villa at site 183? Bovillae site 194. Villa?: foundations. Bovillae site 211. Villa: walls. Bovillae site 248b. Villa rustica?: reticulatum, peperino. Bovillae site 274b. Villa?: reticulatum. Bovillae site 289. Villa?: reticulatum, lava. Bovillae site 295a. Villa: marble, peperino, brick, tile, pottery (black gloss), inscription. Bovillae site 301a. AFF: brick/tile/pottery, inscription. Bovillae site 304.20. Ruderi/Burial?/Villa?: inscription. Bovillae site 305. Villa: sarcophagus, inscription. Tellenae site 82. Villa: mosaic, mortar, brick, tile, pottery (plain/ cookware, dolium, amphora). Quilici 1969, Nr. 1917. Bovillae site 340. Villa: spicatum, reticulatum, cocciopesto, marble, peperino, brick, tile, pottery (terra sigillata, dolium). Tellenae site 90. Villa: mortar, peperino, brick, tile, pottery (black gloss, terra sigillata, dolium), inscription. Quilici 1969, Nr. 1937. Bovillae site 379a. Villa?: architectural elements, listatum, marble, peperino. Tellenae site 91. Villa: spicatum, mortar, tile, pottery (black gloss, terra sigillata), inscription. Tellenae site 108. Villa: marble, peperino, brick, tile, pottery, inscription. Quilici 1969, Nr. 1092. Tellenae site 115a. Villa: spicatum, peperino, brick, tile, pottery (plain/cookware, amphora), inscription. Tellenae site 130. Villa: architectural elements, peperino, brick, tile, pottery. Quilici 1969, Nr. 1923–4. Tellenae site 140a. Villa: reticulatum, tuff, lava, brick, tile, pottery (black gloss). Quilici 1969, Nr. 1902. Tellenae site 143a. Villa: mortar, marble, brick, tile, pottery (dolium, amphora), inscription. Tellenae site 145b. Villa: walls. Tellenae site 158a. Villa: mosaic, brick, tile, pottery (black gloss, plain/cookware, dolium, amphora). Quilici 1969, Nr. 190 Bovillae site 380a. Villa?: reticulatum, peperino, brick, tile, pottery (terra sigillata, plain/cookware). Bovillae site 412. Villa rustica/Casa rustica?: architectural elements, mortar, cocciopesto, lava, peperino. Bovillae site 434. AFF: architectural elements, reticulatum, marble, peperino, brick. Bovillae site 435. Villa: architectural elements, mixtum, lava, peperino. Bovillae site 438. Villa: owner?, quadratum, marble, peperino, inscription. Tusculum site 21. Villa?: owner?, enclosed spaces? Tusculum site 30–31. Villa?: walls, inscription. Tusculum site 68. Villa?: sculpture?, peperino. Tusculum site 69–70. Villa?: enclosed spaces?, architectural elements, reticulatum, marble, peperino. sitE cataloguE Tusculum site 71. Villa: foundations, sculpture?, peperino, brick/tile, inscription. Tusculum site 106.–108. Villa: platform, incertum, lava, tile, pottery (black gloss). Tellenae: 51. Villa; 70b. Villa (Quilici 1969, Nr. 1603–4; Cecchini et al. 1990, 119; Spera and Mineo 2004, 157); 83. Villa (Quilici 1969, Nr. 1919); 169a. Villa. Tusculum site 114. Villa?: walls, reticulatum. Bovillae: 112b. Fattoria/Villa rustica; 132. Villa; 155. Villa; 338. Villa?; 351. Villa?; 425. AFF; 430. Villa. Tusculum site 189. Villa?: enclosed space, incertum, tuff. Tusculum: 25. Villa?; 58–59. Villa?; 910. Villa. Tusculum site 243. Villa?: natural platform. Tusculum site 246–250. Villa: foundations, enclosed spaces, incertum, reticulatum, cocciopesto, lava, iron pipe, pottery (amphora). Tusculum site 521. Villa: owner?, walls, reticulatum, latericium. Tusculum site 627–628. Villa?: owner?, walls, architectural elements, incertum, marble, peperino, inscription, burials. Tusculum site 643. Villa?: wall, enclosed spaces, mosaic, incertum. Brick, tile, pottery Fidenae: 221. Villa (Quilici Gigli 1987). Ficulea: 140c. Villa; 151. AFF; 222a. AFF; 244a. AFF; 322b. AFF; 485a. Villa; 543c. Villa?; 584c = Collatia site 75a. AFF; 589b = Collatia site 66a. Villa; 600. AFF. Tibur III: 101a. AFF; 199b. AFF; 269. Villa. Tibur IV: 217. AFF. Tusculum site 828–829. Villa: enclosed spaces, mosaic, mixtum, brick/tile, inscription. Collatia: 22a. Villa rustica; 24. Casa rustica; 44a. Casa rustica; 71f. Casa rustica; 80d. Aggregato rustico; 105a. Casa rustica; 151a. Villa rustica; 214. Casa rustica; 219b. Casa rustica; 220b. Casa rustica; 296a. Villa rustica; 327c. AFF; 336b. Aggregato rustico; 366a. Sito rustico; 408. Sito rustico; 420b. Villa rustica; 455. Villa rustica?; 459d. Villa; 469a. Aggregato rustico; 474. Villa rustica; 524b. Casa rustica; 537. Villa rustica; 569h. AFF; 577a. Villa rustica; 657f. AFF; 662b. Aggregato rustico; 675a. Villa rustica; 683. Casa rustica; 792a. Casa rustica; 797b. AFF; 813. Villa rustica; 826a. Casa rustica; 841a. Villa rustica (connected to Class 4 villa at site 841b?); 843e. Villa rustica. Tusculum site 840. Villa?: quadratum, reticulatum. Tellenae: 109b. Villa (Quilici 1969, Nr. 1096). Tusculum site 893. Villa: walls, mosaic, reticulatum. Bovillae: 45. AFF; 365b. AFF. Tusculum site 661–662. Villa?/Burial?: owner?, sculpture, wall plaster, mosaic, latericium, marble, peperino, tile, burials. Tusculum site 667. Villa?: walls, reticulatum. Tusculum site 750–751. Villa?: platform, architectural elements, sculpture, marble, peperino, inscription. Tusculum site 753–754. Villa?: enclosed spaces, spicatum?, signinum?, reticulatum?, brick, pottery (terra sigillata). Tusculum site 901. Villa?: walls, reticulatum, mortar, lava. Tusculum site 914. Villa: sculpture, reticulatum, brick, burials. Tusculum site 923. AF: walls? Building debris (wall plaster, mortar, cocciopesto, stone, brick, tile), pottery Old observations or only water installations or remains of agricultural production Ficulea: 208b. Villa; 465d. Villa; 500d. Villas?; 573b = Tibur III site 249b?. Insediamento? Fidenae: 16b. AFF; 80. Casa rustica; 99a. Villa rustica; 106a. Villa rustica; 133b. Villa; 137. AFF; 151b. Villa?; 165. Villa; 166. Villa rustica; 183b. Villa rustica; 190a. Villa; 207a. Villa; 218b. Villa; 235a. Fabbricato; 245b. Villa (Pergola et al. 2003, Via Nomentana Nr. 129); 274. Villa rustica (Pergola et al. 2003, Via Nomentana Nr. 116). Tibur III: 341. Villa? Ficulea: E.3 Villa?; 20a. Villa rustica; 70. AFF; 77. Villa; 103. AFF; 107. Villa; 109. Villa; 120. AFF; 130. Villa rustica; 132. Villa rustica; 146a. Casa rustica; 210. Villa?; 228e. AFF/Villa?; 239. AFF; 324. Villa; 412. Villa rustica (Calci 1998, 103); 424b. Casa rustica?; 463. Villa rustica; 464. Villa rustica?; 515a. Villa; 523. Villa; 548c. Casa rustica?; 550. Villa; 553a. Villa; U2. Villa. Bovillae: 80. Villa; 164. Villa; 304.16b. Villa. Tibur III: 14. AFF; 81. AME; 122b. Villa; 276a. AFF; 311. Villa; 312. Villa; 334c. AME; 427. Villa; 437. Villa. Tibur IV: 189a. Villa?; 210a. Villa?; 212. Villa? (conneced to Class 3 AFF at site 213?); 215. Villa? Collatia: 23a. Villa rustica; 35. Villa rustica; 53a. Aggregato rustico; 97a. Villa rustica; 125. Villa rustica; 128a. Casa rustica; 138a. Villa rustica; 164a. Villa rustica; 180q.1/3 Villa; 199b. Villa rustica; 205a. Villa rustica; 211. Villa rustica; 263c. Villa rustica?; 273a. Villa rustica; 319. Villa rustica; 333c. AFF; 336c–e. Aggregato rustico; 390. Villa rustica; 419a. Sito rustico; 424a. Villa rustica; 468a. Villa rustica; 498a. Sito rustico; 502a. Villa rustica; 504. Aggregato rustico; 564a. Villa; 610. Villa rustica; 638. Aggregato rustico; 655i. Villa; 667c. Villa; 673. Villa rustica; 676a. Villa rustica; 702. Villa rustica; 707b. Casa rustica; 732a. Villa rustica; 735a. Villa rustica; 741. Casa rustica; 744a. Villa rustica; 761b. Casa rustica?; 768a. Villa rustica; 779. Villa rustica; 783. Villa rustica; 790a. Villa rustica; 791b. Villa; 811a. Villa rustica; 833. Sito rustico. Tibur IV: 51b. Villa. Collatia: 55c. Villa rustica?; 108c. Villa?; 480b. Villa rustica; 555a. Villa rustica; 821. Villa rustica. Tellenae: 4b. Villa (Quilici 1969, Nr. 1576); 137. Villa. Tusculum: 896. Villa? class 4 sitEs Fidenae site 50d–e. Casa rustica/AFF: pottery (bucchero, dolium), connected to burials at site 50a–c? Fidenae site 76. Villa: quadratum, travertine, lava, brick/tile/ pottery. Fidenae site 98. Villa rustica: building debris, brick, tile, pottery (black gloss, dolium), burials. Di Gennaro et al. 2004, Nr. 26–7. Fidenae site 142. Villa rustica: reticulatum, tile. Fidenae site 193. Villa rustica: reticulatum. Fidenae site 269. Villa?: architectural elements, sculpture, mosaic, mortar, marble, travertine, tuff, brick, tile, pottery (terra sigillata, African Red Slip), votive artifacts, iron, coins, sarcophagus, inscription. Ficulea site 4. Villa rustica: reticulatum, tuff, tile, pottery. Ficulea site 291. Villa: tile, pottery, inscription. Ficulea site 382a. AFF: reticulatum, tuff, tile, pottery. Carbonara and Messineo 1992. 191 appEndiX i Ficulea site 400. Casa rustica: reticulatum, tuff, lava, pottery. et al. 1990, 119–20. Ficulea site 430. AFF: quadratum, tuff, tile, pottery (African Red Slip). Tellenae site 20. Villa: enclosed spaces, brick. Quilici 1989, 55; Cecchini et al. 1990, 119; Coarelli 1993, 62; Spera and Mineo 2004, 172. Ficulea site 518a. Villa: quadratum, mortar, travertine, pottery. Ficulea site 556. AFF: quadratum, reticulatum, tuff, tile, pottery. Ficulea site 577c. AFF: reticulatum, tuff, brick, tile, pottery. Ficulea site 595. AFF: architectural elements, quadratum, tuff, tile, pottery. Tibur III site 8. AFF: brick/tile, pottery (black gloss), connected to burials at site 7? Tibur III site 45b. AFF: mortar, cemetery for Class 1 villa at site 45a? Tibur III site 326a. Villa: travertine, lava, inscription. Tibur IV site 47a. Villa: reticulatum, travertine, brick/tile, pottery (black gloss). Tibur IV site 54. AFF: pottery (impasto, black gloss), connected to burial at site 53. Tibur IV site 194a. Villa: travertine, tuff, brick/tile, pottery (African Red Slip), inscription. Moscetti 2002, 67. Tibur IV site 213. AFF: marble, brick/tile, pottery (plain/ cookware), connected to Class 3 villa at site 212? Tibur I site 117. Ruderi: reticulatum, limestone. Collatia site 55b. Villa rustica?: incertum, tuff. Collatia site 186b.Villa rustica: quadratum, peperino, tile, pottery (black gloss, amphora). Collatia site 191b. Villa rustica: peperino, tile, pottery (amphora), inscription. Collatia site 193b. Villa: reticulatum. Collatia site 300. Villa rustica: quadratum, travertine, tuff, tile. Collatia site 302. Villa rustica: quadratum, travertine, tuff, peperino. Musco 2001, 209. Collatia site 307. Villa rustica: 1,100 m², architectural elements, quadratum, cocciopesto, tuff, tile, pottery (plain/ cookware, amphora), coins. Musco 2001, 205–6. Collatia site 321c. Aggregato rustico: tile, connected to Class 1 villa at site 321b? Collatia site 364a. Casa rustica: foundation, reticulatum, marble, tuff, tile, pottery (black gloss, African Red Slip, plain/ cookware, amphora). Musco et al. 2002. Collatia site 448b. AFF: mortar, tile, pottery, same as Class 3 villa at site 448c? Collatia site 524c. Casa rustica: foundation, mortar, tuff, lava, tile, pottery. Musco et al. 2002. Collatia site 549. Villa: quadratum, mortar, marble, peperino, tile, pottery. Collatia site 587. Villa rustica: quadratum, marble, tuff, lava, tile, pottery (African Red Slip). Collatia site 640a. Casa rustica: quadratum, tile, pottery (black gloss). Zaccagni 1984; Musco and Zaccagni 1985, 99. Collatia site 766. Villa rustica: quadratum, tuff, lava, peperino, tile, pottery. Collatia site 836b. Villa: architectural elements, marble, brick, tile, pottery. Collatia site 841b. Villa rustica: cocciopesto, marble, travertine, tile, pottery, connected to Class 3 villa at site 841a? Tellenae site 15. Villa: listatum, tuff, brick, tile, pottery (dolium). Quilici 1969, Nr. 1116; Bellini 1985, 127. Tellenae site 17. Villa: architectural elements, brick, tile, pottery (black gloss, terra sigillata, dolium), inscription. Quilici 1969, Nr. 2215; Cecchini et al. 1990, 120. Tellenae site 19. Villa: architectural elements, reticulatum, marble, peperino, brick, tile. Quilici 1969, Nr. 1601–2; Cecchini 192 Tellenae site 50b. Villa?: incertum. Quilici 1969, Nr. 1098. Tellenae site 52. Villa: reticulatum, marble, brick, tile, pottery (plain/cookware, dolium). Tellenae site 74. Villa: reticulatum, mortar, brick, tile, pottery (terra sigillata, plain/cookware, amphora). Quilici 1969, Nr. 1594. Tellenae site 96. Villa: walls? Quilici 1969, Nr. 1061. Tellenae site 105. Villa: reticulatum, brick, pottery (plain/ cookware). Quilici 1969, Nr. 1078. Tellenae site U1a. Villa/Statio?: reticulatum, brick, inscription. Cecchini et al. 1990, 119; Spera and Mineo 2004, 157. Tellenae site U6. Villa: reticulatum, brick/tile/pottery. Quilici 1969, Nr. 1062; Cecchini et al. 1985, 247; Santangeli Valenziani and Volpe 1988, 544–53. Bovillae site 116. Villa: reticulatum, mortar, marble, tuff, peperino, brick. Bovillae site 227. AFF: reticulatum, peperino, brick, tile. Bovillae site 281. AFF/Villa rustica: quadratum, mortar, lava, peperino, brick, tile, pottery (plain/cookware). Bovillae site 298. AFF: reticulatum, peperino, brick/tile/pottery. Bovillae site 307. Fattoria: quadratum, reticulatum, peperino, brick/tile, pottery. Bovillae site 327. AFF: reticulatum, listatum, marble, lava, peperino, brick/tile, pottery (black gloss, terra sigillata, dolium). Bovillae site 366a. AFF: architectural elements, marble, peperino, inscription. Bovillae site 381. Villa: sectile, reticulatum, cocciopesto, marble, peperino, brick, tile. Bovillae site 397. Villa: reticulatum, cocciopesto, marble, peperino, brick, tile. Building debris (wall plaster, mortar, cocciopesto, stone, brick, tile), pottery Fidenae: site 49. AFF; 82. Casa rustica?; 93. Villa rustica; 109. Villa (Di Gennaro et al. 2004, Nr. 23?); 126. Casa rustica; 128. Villa; 131. Villa rustica; 134. AFF; 201. Ruderi; 219. Villa rustica; 223. Villa rustica; 232. Fabbricato rustico (Di Gennaro et al. 2005); 251. Villa rustica; 253. Villa rustica. Ficulea: 3. Villa rustica (Cali 1998, 310?); 9. Villa; 11. Casa rustica?; 15. Villa rustica; 17. Villa; 79b. AFF; 85b. Villa rustica/ Casa rustica; 101. AFF; 123. Villa rustica (Pergola et al. 2003, Via Nomentana Nr. 146); 128. AFF; 149. AFF; 179b. AFF (Pergola et al. 2003, Via Nomentana Nr. 150); 184. Villa?; 195d. AFF; 202. Villa rustica; 213. Villa rustica; 226. Villa rustica?; 240. AFF; 241. Villa; 245. Villa rustica?; 257. Villa; 299. Villa rustica (Moscetti and La Porta 2004, 171); 337. Villa; 397. Casa rustica; 403. Casa rustica?; 440. AFF; 457. Villa; 483. Villa rustica; 506. Villa?; 520. Villa; 542. Villa rustica?; 591. AFF. Tibur III: 72. AME/Casa rustica?; 73. AFF;98. AFF/Villa?; 106. AFF; 116b. AME; 119. AFF; 132. AME; 254d. AFF; 275. AME; 289. AME; 313b = Collatia site 92. AFF; 317. AFF; 327. AME/ Villa?; 428. AFF. Tibur IV: 138a. Villa; 140a. Villa?; 191a. Villa; 201. AFF. Tibur II: 219. Villa. Collatia: 9. Casa rustica; 17b. Villa rustica; 41a. Villa rustica; 51. Villa rustica; 53c. Aggregato rustico; 83. Sito rustico; 85. Casa rustica; 102a–b. Villa rustica; 111b. Casa rustica; 115. Villa rustica; 117. Casa rustica; 124b. Casa rustica; 127a–b. Casa rustica; 135. Casa rustica; 159. Villa rustica;162. Villa rustica; 168b. Villa rustica; 171a. Villa rustica; 180q.4 Ruderi?/ sitE cataloguE Villa?; 194g. AFF; 212. Casa rustica; 215. Casa rustica; 217. Villa rustica; 234. Villa rustica; 235. Casa rustica; 243. Casa rustica; 268. Casa rustica; 275. Villa rustica; 293. Casa rustica; 304a. Villa rustica; 313a. Villa rustica; 318. Villa rustica; 347a. Villa rustica; 348. Villa rustica; 350a. Casa rustica; 350b. Casa rustica; 356. Casa rustica; 359a. Casa rustica; 381. Casa rustica; 391. Sito rustico; 393a. Sito rustico; 398a. Villa rustica; 401b. Casa rustica; 411. Villa rustica; 412. Villa rustica; 413b. Villa rustica; 423. Villa rustica; 424b. Villa rustica; 433. Casa rustica; 435. Villa rustica; 458. Villa rustica; 462. Sito rustico; 475. Villa rustica; 478. Villa rustica; 481. Casa rustica; 483a. Casa rustica; 484. Villa rustica; 485b. Villa rustica; 491a. Villa rustica; 503. Casa rustica; 513. Casa rustica; 515a. Villa rustica; 525. Aggregato rustico; 527. Casa rustica; 545. Aggregati rustici; 561. Villa rustica; 575a. Sito rustico; 589b. AME?; 590b. AFF; 595. Villa rustica; 613a. Villa rustica; 615b. Sito rustico; 619a. Casa rustica; 621b. Casa rustica; 623a–b. Villa rustica; 635a. Villa rustica; 654a. Casa rustica; 666. Villa rustica; 689a. Casa rustica; 717. Villa rustica; 721. Casa antica; 726a. Sito antico; 730. Villa rustica; 746. Villa rustica; 749. Aggregato rustico; 758b. Villa rustica; 760b. Casa rustica; 763. Sito rustico; 765a– b. Villa/Casa rustica; 768b. Villa rustica; 770. Villa rustica; 778. Casa rustica; 792d. Casa rustica; 792e. Villa rustica; 800a–b. Villa/Casa rustica; 816. Villa; 829a. Villa; 832c. Villa rustica; U3. AFF. Tellenae: 39. Villa; 47. Villa; 72. Villa (Quilici 1969, Nr. 1599); 89a. Villa (Quilici 1969, Nr. 2061); 119. Villa (Quilici 1969, Nr. 2222); 141. Villa (Quilici 1969, Nr. 1901); 164. Villa (Quilici 1969, Nr. 1929); 172. Villa (Quilici 1969, Nr. 2218); U2. Villa (Cecchini et al. 1990, 119). Bovillae: 3. AFF; 10. AFF; 23. AFF/Villa rustica?; 31. AFF; 43. AFF; 84. Fattoria/Villa rustica; 153. AFF; 154. AFF; 156. AFF; 159. AFF; 178. Ruderi; 210. AFF; 260. Villa; 294b. AFF; 309. AFF; 349. Villa; 353. AFF; 363. AFF; 370. AFF/Casa/Villa rustica; 385. AFF; 386. AFF; 389a. AFF; 395. AFF/Villa/Casa rustica?; 396a. Insediamento?; 400. AFF; 401. AFF; 402. AFF; 408. AFF/Villa/Casa rustica?; 428. AFF. Apiolae: 178. Villa; 185. AFF; 281. Villa. Tusculum: 4. AF. Brick, tile, pottery Fidenae: 12. AFF?; 14b. AFF; 17b. AFF; 75a–b. AFF/Villa; 89. AFF; 90. AFF; 116. Villa rustica; 119. AFF; 123. Fabbricati; 145. Villa rustica?; 154a. Villa rustica; 164a. Villa/Burial?; 181. Villa rustica; 192. AFF (Pergola et al. 2003, Via Nomentana Nr. 128); 227. Villa rustica?; 229. Villa rustica; 237a. Villa; 240. Villa rustica; 244. Casa rustica; 248. Fabbricato rustico; 254. Villa rustica; 259. AFF; 266. AFF; 276. Villa rustica? Ficulea: E.2 AFF; P. AFF; Q. AFF; 5. Villa rustica; 56a. AFF; 60. Villa rustica?; 66a. AFF (same as Class 3 villa at site 66b?); 67a. AFF; 69b. Villa rustica (Pergola et al. 2003, Via Nomentana Nr. 173); 94. AFF; 99. AFF; 100. AFF; 108. AFF; 112. AFF; 117. AFF; 118. AFF (Pergola et al. 2003, Via Nomentana Nr. 141); 153. Villa; 160a. Villa rustica; 171. Villa rustica?; 189. Casa rustica; 206a. AFF; 211. Villa rustica?; 232. AFF (Pantano 1998, 310); 235. AFF; 248b. Villa rustica; 301. AFF; 319. AFF; 322a. AFF; 330. AFF; 331b. AFF; 332. AFF; 336. Villa rustica?; 348a. AFF; 349. Villa rustica?; 352b = Tibur III site 20b. Casa/ Villa rustica; 383. Villa rustica (Carbonara and Messineo 1992); 384. AFF; 385c. Villa rustica; 425b. Villa rustica; 447. Casa rustica?; 450. Casa rustica?; 458b. Villa; 470b. Villa rustica; 471. Villa rustica; 478. Casa rustica?; 490. Casa rustica?; 496b. Casa rustica; 545b. Villa rustica? (Calci and Sorella 1995, 124, nota 35); 555. Villa rustica; 575. Villa; 585b = Collatia site 74. AFF; 597. Villa?; 598 = Collatia site 60. Villa rustica. Tibur III: 10. AFF; 13. AFF; 49. AFF; 52. AFF;68. AFC; 86. AFF; 96. AFF/Casa rustica/Burials?; 104. AFF; 107. AFC; 113. AFF; 115a. AFF (Moscetti 1991, Nr. 26); 129. AFF; 150. AFC (Moscetti 1991, Nr. 85); 188. AFC; 193. AFF; 194. AFF; 201. AFF; 255. AFF; 256. AFF; 271. AFF; 274. AFF; 296c. AFF; 315. AFF; 333. AFF;335. AFF; 410. AFF; 415. AFF; 439. AFF; 445. AFF/Villa? Tibur IV: 29. Insediamento pastorale; 47b. AFF. Collatia: 3. Villa rustica; 6. Casa rustica; 7. Casa rustica; 8. Villa rustica; 10a. Villa rustica; 10b. Villa rustica; 15a. Villa rustica (Guaitoli and Zaccagni 1985); 15b. Villa rustica (Guaitoli and Zaccagni 1985); 16. Villa rustica; 17a. Villa rustica; 19b. Casa rustica; 20. Villa rustica; 21. Casa rustica; 25. Casa rustica; 27. Casa rustica; 29b. Villa rustica?; 31c. Villa rustica?; 33. Villa rustica?; 36a. Casa rustica; 37. Casa rustica; 38b. Casa rustica; 39b. Casa rustica; 40. Villa rustica; 41b. Casa rustica; 42. Casa rustica; 45d. AFF?; 46. Sito rustico; 56. Casa rustica; 59a Villa rustica; 59b. Villa rustica; 62a. Casa rustica; 63. Villa rustica; 65. Casa rustica; 67. Sito rustico; 68. Sito rustico; 76b. Casa rustica; 77. Casa rustica; 78. Villa rustica; 86. Sito rustico; 89a. AFF; 101a. Villa rustica; 110. Villa rustica; 112. Casa rustica; 114. Casa rustica; 119. Sito rustico; 130. Casa rustica; 132. Casa rustica; 134b. Casa rustica; 137a. Casa rustica; 143. Villa rustica; 145. Villa rustica; 146. Villa rustica; 148b. AFF; 148c. AFF; 150. Villa rustica; 152. Casa rustica; 153. Villa rustica; 154. Casa rustica; 158. Villa rustica; 161. Casa rustica; 167a. Villa rustica; 167b. Casa rustica; 171b. Casa rustica; 173c. Aggregato rustico; 175. Villa rustica; 176a.Villa rustica; 176b.Villa rustica; 179c. AFF; 181. Casa rustica; 184. Casa rustica; 185b. Villa rustica; 189a. Aggregato rustico; 190. Casa rustica; 192b. Villa rustica; 203. Villa rustica; 206. Casa rustica; 209. Casa rustica; 213. Villa rustica; 216. Casa rustica; 218. Casa rustica; 222b–c. Casa rustica; 226e. Sito arcaico/Villa?; 231. Casa rustica; 237. Casa rustica; 238. Casa rustica; 249a. Casa rustica; 250. Casa rustica; 252. Casa rustica; 271. Villa rustica; 277. Casa rustica; 278a. Casa rustica; 279. Casa rustica; 286a. Villa rustica; 287. Casa rustica; 288a. Sito rustico; 288b. Sito rustico; 289. Villa; 291. Villa rustica; 295. Villa rustica; 297a. Casa rustica; 298. Villa rustica; 306b. Casa rustica; 310. Casa rustica; 311. Villa rustica; 312. Villa rustica; 314. Casa rustica; 315. Casa rustica; 316. Casa rustica; 322. Villa rustica; 323. Casa rustica; 324. Casa rustica; 325b. Casa rustica; 328. Villa rustica; 329. Villa rustica; 335a. Villa rustica; 337a. Casa rustica; 337c. Casa rustica; 338. Casa rustica; 339. Aggregato rustico; 344. Villa rustica; 354c. Casa rustica; 357c. Insediamento rustico; 358. Villa rustica; 361. Casa rustica; 374b. AFF; 389. Casa rustica; 400. Casa rustica; 401a. Casa rustica; 420a. Villa rustica; 421. Casa rustica; 425a. Villa rustica; 425b. Casa rustica; 427. Villa rustica; 430. Casa rustica; 434. Villa rustica; 446. Villa rustica; 450. Villa rustica; 451a–b. Villa rustica; 453a. Villa rustica; 457. Casa rustica; 461a. Villa rustica; 463. Villa rustica; 464. Casa rustica; 467. Casa rustica; 471. Villa rustica; 473. Villa rustica; 480a. Villa rustica; 486. Villa rustica; 493. Villa rustica; 494. Casa rustica; 496a. Villa rustica; 500. Villa rustica; 501a. Villa rustica; 506. Casa rustica; 510. Casa rustica; 511. Casa rustica; 514. Casa rustica; 518. Casa rustica; 520. Casa rustica; 522. Casa rustica; 526. Casa rustica; 532. Casa rustica; 533a. Casa rustica; 533b. Casa rustica; 534a. Casa rustica; 534b. Casa rustica; 535a. Casa rustica; 535b. Casa rustica; 536. Villa rustica; 538. Aggregato rustico; 539a. Casa rustica; 539b. Casa rustica; 542. Aggregato rustico; 543. Villa rustica; 544a. Aggregato rustico; 544b. Villa rustica; 546b. Villa rustica; 547. Casa rustica; 548. Villa rustica; 557a. Villa rustica; 557b. Casa rustica; 566. Aggregato rustico; 572. Villa rustica; 573. Casa rustica; 574. Villa rustica; 576. Sito rustico; 579. Villa rustica; 580. Villa rustica; 583b. AFF; 585a. Villa rustica; 597. Villa rustica; 605. Aggregato; 612. Casa rustica; 615a. Sito rustico; 621a. Casa rustica; 632. Villa rustica; 635b. Casa rustica; 636. Villa rustica; 637a. Villa rustica; 639a. Casa rustica; 643. Villa rustica; 644. Casa rustica; 648. Villa rustica; 654c. Casa rustica; 661b. Sito rustico; 664. Villa rustica; 665. Casa rustica; 674. Casa rustica; 677. Casa rustica; 680a. Villa rustica; 684. Villa rustica; 686. Villa rustica; 687. Villa rustica; 689b. Casa rustica; 691. Casa rustica; 699. Casa rustica; 700. Casa rustica; 705. Villa rustica; 707a. Villa rustica; 709. Villa rustica; 714a. Villa rustica; 714b. Casa rustica; 714c. Casa rustica; 715. Villa rustica; 716. Casa rustica; 718a. Casa rustica; 718b. Casa rustica; 719. Casa rustica; 720. Casa rustica; 722. Casa rustica; 723. Casa rustica; 724. Villa rustica; 725. Casa rustica; 729a. Casa rustica; 729b. Casa rustica; 732b. 193 appEndiX i Casa rustica; 733. Casa rustica; 734. Casa rustica; 736a. Casa rustica; 736b. Villa rustica; 738. Villa rustica; 742. Villa rustica; 743. Casa rustica; 751. Casa rustica; 752a. Casa rustica; 752b. Casa rustica; 760a. Villa rustica; 760c. Casa rustica; 764a. Casa rustica; 764b. Casa rustica; 773a. Casa rustica; 773b. Casa rustica; 780a. Casa rustica; 781a. Casa rustica; 781b. Casa rustica; 782a. Casa rustica; 785. Casa rustica; 786. Villa rustica; 789. Casa rustica; 792c. Casa rustica; 804b. Casa rustica; 805a. Casa rustica; 805b. Casa rustica; 807b. Casa rustica; 811b. Villa rustica; 815. Villa rustica; 818. Casa rustica; 824a. Villa rustica; 824b. Villa rustica; 825a. Villa rustica; 827. Villa rustica; 830a. Casa rustica; 830b. Casa rustica; 834. Casa rustica; 844a. Casa rustica; 844b. Casa rustica; 846. Sito rustico; 847. Casa rustica; 852. Casa rustica; 856a. Villa rustica; 856b. Casa rustica. Tellenae: 58. Villa (Quilici 1969, Nr. 1082); 69. Villa; 101. Villa (Quilici 1969, Nr. 1071); 142. Villa; 175a. Villa (Quilici 1969, Nr. 2217); U3. Villa (Cecchini et al. 1990, 119; Spera and Mineo 2004, 156). Bovillae: 17. AFF; 18. AFF; 22. AFF; 52. AFF; 90. AFF; 119. Villa; 126b. Villa (Corrente 1988a, 400; LTURS s.v. Latina via, VII miglio); 136. Villa; 162. AFF; 163. AFF; 229. AFF; 237. AFF; 326. AFF; 330. AFF; 347. AFF; 348. AFF; 357. AFF; 358. AFF; 365a. AFF; 367. AFF; 368. AFF; 369. AFF; 372. AFF; 388. AFF; 394a. AFF; 424. AFF. Tusculum: 73. AFF; 779. AFF. Old observations or only water installations or remains of agricultural production Fidenae: 9a. AFF?; 62. Villa?; 125. Villa rustica; 208. Ruderi/ Villa?; 209. Ruderi/Villa?; 238. Villa. Ficulea: 206b. Villa rustica; 375. AME; 387a. Villa rustica. Tibur III: 378. Villa; 397. Villa. Tibur IV: 130. Villa?; site 192. Villa? Tellenae: 3. Villa (Quilici 1969, Nr. 1581); Tellenae site 21b. Villa? (Castagnoli et al. 1972, 156; Cecchini et al. 1990, 119; Spera and Mineo 2004, 172); 129b. Villa? (Quilici 1969, Nr. 2060). Bovillae: 77. AFF; 140. Villa?; 142. Villa rustica/Casa rustica?; 148. Villa?; 217. Villa. Tusculum: 202. Villa?; 422. Villa?; 614. AF; 649. AFF; 759. Villa? sitEs inhabitEd only bEforE 2nd cEntury bc Fidenae site 27. AFF: wall, quadratum, tuff, tile, pottery. Fidenae site 35. Foundations: quadratum, tuff. Fidenae site 36a. Walls: tuff. Fidenae site 169. AFF: walls, tuff, tile, pottery. Di Gennaro et al. 2004, 111. Fidenae site 195. AFF: walls, cocciopesto, tuff, tile. Ficulea site A.1 AME?: architectural elements, marble, travertine, brick/tile, pottery. Collatia site 406. Casa rustica: platform?, enclosed spaces?, tuff, pottery. Tile, pottery Fidenae: 8b. Insediamento preistorico; 15. AFF?; 23. AFF; 25b. Insediamento preistorico; 52b. AFF; 74c. Casa rustica; 84. AFF; 85. AFF; 104. AFF; 105. AFF; 110. AFF; 112a. AFF; 115. AFF; 117b. AFF; 118. AFF; 122. AFF; 133a. AFF; 147. AFF; 170. 194 AFF; 173. AFF; 175. AFF (Fraioli 2000, 226; Di Gennaro et al. 2004, 110–1); 177. AFF (Fraioli 2000, 226; Di Gennaro et al. 2004, 111, 119, 132?); 179. AFF (Di Gennaro et al. 2004, 110?); 183a. AFF; 187. AFF; 194. AFF; 200. AFF; 210b. AFF; 233. AFF; 246. AFF; 247. AFF; 249. AFF; 252. AFF; 255. AFF; 256. AFF; 261. AFF; 263. AFF; 264. AFF; 265. AFF; 268. Enclosed space; 271b. AFF; 278. AFF. Ficulea: I. AFF; L. AFF; N. AFF; T. AFF; 6. AFF; 8. AFF; 18. AFF; 58. AFF; 61. AFF; 64. AFF; 65. AFF; 69a. AFF; 71a. AFF; 76. AFF; 78. AFF; 79a. AFF; 90. AFF; 95. AFF; 97a. AFF; 98. AFF; 102. AFF; 110. AFF; 111a. AFF; 113. AFF; 114. AFF; 122. AFF; 125. AFF; 140b. AFF; 145. AFF; 150. AFF; 156a. AFF; 159a. AFF; 162. AFF; 166a. AFF; 174b. AFF (Slaska 2002); 178a. AFF; 182. AFF; 186b. AFF; 187b. AFF; 188. AFF; 190a. AFF; 194. AFF; 195c. AFF; 198. AFF; 199. AFF; 203. AFF; 207. AFF; 209. AFF; 212. AFF; 215. AFF; 219a. AFF; 220. AFF; 224. AFF; 227. AFF; 228a. AFF; 242. AFF; 247. AFF; 248a. AFF; 256. AFF; 296a. AFF; 298a. AFF; 300. AFF; 304a. AFF; 306a. AFF; 308. AFF; 310a. AFF; 314. AFF; 315a. AFF; 317. AFF; 318a. AFF; 320a. AFF; 321. AFF; 323. AFF; 325. AFF; 326. AFF; 328. AFF; 331a. AFF; 334. AFF; 338. AFF; 339. AFF; 340. AFF; 342a = Tibur III site 48. AFF/AFC; 344 = Tibur III site 183a. AFF; 345a = Tibur III site 185. AFF; 353 = Tibur III site 19. AFF; 361. AFF; 366 = Tibur III sites 14* and 19*. AFF?; 368 = Tibur III sites 14* and 19*. AFF; 371a. AFF; 373. AFF; 376. AFF; 381. AFF; 385a. AFF; 385b. AFF; 386a. AFF; 388. AFF; 390. Casa rustica; 398a. AFF; 399. AFF; 401a. AFF; 402. AFF; 407. AFF; 418. AFF; 419. AFF; 421a. AFF; 422. AFF; 424a. AFF; 425a. AFF; 427b. AFF; 428. AFF; 429. AFF; 431. AFF; 432a. AFF; 433. AFF; 436. AFF; 437. AFF; 438a. AFF; 443. AFF; 445a. AFF; 448. AFF; 449. AFF; 451. AFF; 452. Villa?; 453. AFF; 454a. AFF; 458a. AFF; 459. AFF; 461. AFF; 462. AFF; 470a. AFF; 474. AFF; 476. AFF; 480. AFF; 484a. AFF; 504e. AFF; 507a. AFF; 530a. AFF; 538. AFF; 539c. AFF; 545a. AFF; 558a. AFF; 577a. AFF; 581a. AFF; 583a. AFF; 584a. AFF; 585a = Collatia site 74. AFF; 587. AFF; 590. AFF; 594. AFF/Aggregato rustico; 603a. AFF. Tibur III: 11a. Insediamento; 75. AFC; 84. AFC; 100a. AFF; 111. AFF; 127. AFF; 134a. AME; 140a. AFF; 142. AFC; 144b. AFC; 154b. AFC; 192a. AFF; 206. AFF; 230b. AFC; 351. AFF; 425a. AFF; 441. Sito arcaico? Tibur IV: 198a. AFC. Collatia: 11c. Insediamento; 15c. AFF; 18. Villa rustica; 19a. Casa rustica; 39a. Aggregato rustico; 48. Casa rustica; 57a. Casa rustica; 57b. Casa rustica; 69a. Casa rustica; 73a. Insediamento rustico; 90. Casa rustica; 121. Casa rustica; 123b. Sito arcaico; 124a. Casa rustica; 126. Casa rustica; 163. Aggregato rustico; 173b. Aggregato rustico; 191a. Casa rustica; 222a. Casa rustica; 259. Sito arcaico; 266. Casa rustica; 270. Sito arcaico; 272. Sito arcaico; 280b. Sito arcaico; 281. Casa rustica; 290. Casa rustica; 325c. Casa rustica; 342. Aggregato rustico; 352. Casa rustica; 353. Casa rustica; 357b. Insediamento arcaico; 364b. Casa rustica; 375. Casa rustica; 393b. Sito rustico; 395. Sito rustico; 426. Casa rustica; 461b. Casa rustica; 483b. Casa rustica; 487. Sito arcaico; 508. Sito arcaico; 512. Casa rustica; 539c. Casa rustica; 551b. AFF; 553a. Aggregato rustico; 575b. Sito rustico; 611. Casa rustica; 617. Casa rustica; 637b. Casa rustica; 654b. Casa rustica; 661a. Sito rustico; 772. Villa rustica; 782b = Tusculum site 1. Casa rustica; 806b. Casa rustica; 809b. Sito arcaico. Tusculum: 28. Sito preistorico; 609. AFF; 610. Insediamento?; 611. AFF; 612. AFF; 702–703. AFF; 741. AFF; 778. AFF; 895. Insediamento? appEndiX ii tablE of datEd sitEs PH = Prehistoric, Arch = Archaic, ER = Early Republican, MR = Middle Republican, I–V BC/ AD = 1st, 2nd, 3rd, 4th, 5th century BC/AD and ? = uncertain date. Name Fidenae 1 Fidenae 8a Fidenae 31a Fidenae 71a Fidenae 78 Fidenae 83 Fidenae 114 Fidenae 121 Fidenae 139a Fidenae 182a Fidenae 197 Fidenae 250 Ficulea 74–75 Ficulea 127 Ficulea 129a Ficulea 138a Ficulea 144 Ficulea 148a Ficulea 152a Ficulea 159b Ficulea 163a Ficulea 172 Ficulea 180b Ficulea 183 Ficulea 185a Ficulea 187a Ficulea 192 Ficulea 201a Ficulea 218 Ficulea 219b Ficulea 237 Ficulea 254a Ficulea 258 Ficulea 261a Ficulea 292 = Tibur III 176 Ficulea 302 Ficulea 310b Ficulea 327 Ficulea 329a Ficulea 342b = Tibur III 47a Ficulea 350 = Tibur III 21 Ficulea 356a = Tibur III 6a Ficulea 360 = Tibur III 2a Ficulea 364a Ficulea 371b Ficulea 374a Ficulea 380a Ficulea 398b–c Ficulea 415a Ficulea 415c Ficulea 435 Ficulea 455a Ficulea 473a Ficulea 479 Ficulea 487a Ficulea 489a Ficulea 503a–c = Collatia 1b–c Ficulea 507b Ficulea 517c Ficulea 536a Ficulea 539d Class 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 PH Arch ER MR II BC I BC x ? x x ? x x x x x ? x x x x ? x x x x x ? x x x x x x x x x x x x x x x x x ? x x x x x x x x x ? x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x ? x x x x x x x x x x ? x ? x x x x x x x x x x ? x x x x x x x x x I AD II AD III AD IV AD x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x V AD x x x x x x x appEndiX ii Name Ficulea 540a Ficulea 573a = Tibur III 249a Ficulea 581b Ficulea 604a Tibur III 22 Tibur III 35 Tibur III 42a Tibur III 45a Tibur III 55 Tibur III 80 Tibur III 108a Tibur III 109a Tibur III 118 Tibur III 138 Tibur III 144a Tibur III 162a Tibur III 195a Tibur III 197a Tibur III 202a Tibur III 217a Tibur III 254a Tibur III 263a Tibur III 265a Tibur III 270a Tibur III 277a Tibur III 298 Tibur III 302a Tibur III 303a Tibur III 304a Tibur III 308a Tibur III 318a = Collatia 296d = Tibur IV 216a Tibur III 345a Tibur III 346 Tibur III 369 Tibur III 387a Tibur III 390a Tibur III 396 Tibur III 402a Tibur III 414 Tibur III 421 Tibur III 429a Tibur III 432 Tibur III 434 Tibur III 449 Tibur IV 1a Tibur IV 8 Tibur IV 11a Tibur IV 16 Tibur IV 20a Tibur IV 22a Tibur IV 27 Tibur IV 37a Tibur IV 40 Tibur IV 41a Tibur IV 44a Tibur IV 60 Tibur IV 67 Tibur IV 68 Tibur IV 70 Tibur IV 72 Tibur IV 73 Tibur IV 86 Tibur IV 87 Tibur IV 100 Tibur IV 104 Tibur IV 143 Tibur IV 148 Tibur IV 149 Tibur IV 153 Tibur IV 157 Tibur IV 202a Tibur IV 224 Tibur IV 225 Tibur IV U1 Tibur I 96 Tibur I 99 Tibur I 106 Tibur I 198–200 Tibur I 205 Tibur I 207–208 Tibur I 209 Tibur I 212 196 Class 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 PH Arch x ER MR II BC I BC I AD II AD III AD IV AD x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x ? x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x ? ? ? ? ? x ? ? x x ? x x x ? x ? x x x x x x ? x x ? x x x x ? x ? x ? x ? ? ? ? x x x ? x x x x ? x ? x ? x x x x ? ? ? ? ? ? ? x x x x ? ? x x x x x x x x x ? ? x x ? ? x x x x x x x x x x x ? ? x x x x ? ? ? ? ? ? ? x x x x ? x x x x x x x x x x ? ? x x x x x x ? x ? ? ? ? x x x x x x x x x x x x x x x ? ? x x x x x x ? x x x x x x x x ? x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x V AD x x x x x x x ? x x x x x x x x x x x x x ? x ? x x x x x x x x x x x x x x x x x x x x x x x x ? x ? x x x x x x x x ? x ? x x x x x x x x x x x x x x x x x x x x x x x ? x x ? x x x x x x x x x x x ? x x x x x x x datEd sitEs Name Tibur II 85 Tibur II 214 Tibur II 217 Collatia 4d Collatia 43a–c Collatia 50a Collatia 106a Collatia 123a Collatia 147 Collatia 194l Collatia 195a Collatia 202a Collatia 204b Collatia 223 Collatia 224p Collatia 229c Collatia 241 Collatia 242 Collatia 284 Collatia 321b Collatia 355a Collatia 360a Collatia 362b Collatia 365a Collatia 372a–c Collatia 373 Collatia 377a Collatia 386 Collatia 388a Collatia 392a Collatia 394a Collatia 405b Collatia 431 Collatia 432 Collatia 444 Collatia 454b Collatia 472 Collatia 477 Collatia 479b Collatia 485a Collatia 495a Collatia 516a Collatia 517b Collatia 540 Collatia 552a Collatia 560b Collatia 567a Collatia 582a Collatia 583a Collatia 586 Collatia 590e Collatia 593a Collatia 594 Collatia 596a Collatia 601a Collatia 604a Collatia 607a Collatia 616b–c Collatia 618 Collatia 634a Collatia 646 Collatia 649 Collatia 651a Collatia 679a Collatia 695a Collatia 713a Collatia 728 Collatia 771a Collatia 776 Collatia 795 Collatia 803 Collatia 812 Collatia 828 Collatia 832a Collatia 839c Collatia 855 Tellenae 5b Tellenae 29a Tellenae 37b Tellenae 40 Tellenae 97a Tellenae U7 Class 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 PH Arch ER x MR ? x II BC I BC x x x x x x x x x x ? x x x x x ? ? ? ? ? ? x x x x ? ? ? x x x x ? x ? ? ? ? ? x x x ? x x x ? ? ? x x ? ? x x x x x ? x x x x x x x x x x ? ? ? x ? ? ? x x ? ? ? ? x x x x x x ? x ? x x x x x x x x ? x x x x x ? x ? x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x I AD x ? x ? ? ? x x x x x x x ? ? x ? x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x ? x x II AD ? x x III AD IV AD V AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x ? x x x x ? x x x x x x ? x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x ? x x x 197 appEndiX ii Name Bovillae 19 Bovillae 48 Bovillae 96 Bovillae 101 Bovillae 103 Bovillae 121.1 Bovillae 123a Bovillae 141 Bovillae 160 Bovillae 169 Bovillae 173a Bovillae 179–180 Bovillae 200 Bovillae 209 Bovillae 222a Bovillae 249a Bovillae 254 Bovillae 273 Bovillae 285b Bovillae 337 Bovillae 387 Bovillae 391 Bovillae 404 Bovillae 417 Bovillae 423 Bovillae 432 Bovillae 436a Bovillae U1 Tusculum 93–99 Tusculum 135–136 Tusculum 152–154 Tusculum 190–194 Tusculum 208–219 Tusculum 235–242 Tusculum 314–316 Tusculum 318–323 Tusculum 362–379 Tusculum 380–392 Tusculum 397–409 Tusculum 426 Tusculum 437–487 Tusculum 500–508 Tusculum 511–517 Tusculum 525–535 Tusculum 546–552 Tusculum 561–569 Tusculum 592–599 Tusculum 634–642 Tusculum 650–655 Tusculum 656–660 Tusculum 687–695 Tusculum 720–736 Tusculum 764–768 Tusculum 789–794 Tusculum 804–806 Tusculum 832–836 Fidenae 10 Fidenae 11a Fidenae 21 Fidenae 25a Fidenae 48 Fidenae 73 Fidenae 86 Fidenae 88b Fidenae 91 Fidenae 124 Fidenae 132 Fidenae 141 Fidenae 144 Fidenae 146a Fidenae 148b Fidenae 149 Fidenae 153 Fidenae 157a Fidenae 158a Fidenae 160a Fidenae 163 Fidenae 172a–b Fidenae 186 Fidenae 206 Fidenae 212a Fidenae 212b 198 Class 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 PH Arch ER x x MR ? ? ? ? II BC I BC I AD II AD III AD IV AD x x x x x x x x x x x x x ? x x x x x x x x x x x ? x ? x x x x x x ? x x x x x ? x x x x x x x x x x x x x x x x x x x x ? x ? x x x x x x ? x x x x ? x x x x ? x x x x ? x x x x x x x x x x x x x x x x x x ? ? x ? ? ? x x ? ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x ? ? x ? x x ? ? x x x ? x ? x x ? x x x x x x x x ? x x x x x x x x x x ? x ? x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x ? ? x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x V AD x x x x x x x x x x x x x datEd sitEs Name Fidenae 213a Fidenae 215a Fidenae 230 Fidenae 234b Fidenae 239 Fidenae 267 Fidenae 269 Ficulea M Ficulea R Ficulea S Ficulea 1 Ficulea 7b Ficulea 52 Ficulea 71b Ficulea 73 Ficulea 80 Ficulea 81 Ficulea 82 Ficulea 86 Ficulea 119 Ficulea 124b Ficulea 133 Ficulea 137 Ficulea 155a Ficulea 155b Ficulea 156b Ficulea 164b Ficulea 166b Ficulea 169a Ficulea 173a Ficulea 205a Ficulea 228b Ficulea 228c Ficulea 228f Ficulea 234 Ficulea 236a Ficulea 243 Ficulea 296b Ficulea 315b Ficulea 316a Ficulea 318b Ficulea 345b = Tibur III 185 Ficulea 346 = Tibur III 186 Ficulea 351a Ficulea 363a = Tibur III 1 Ficulea 377 Ficulea 378b Ficulea 389 Ficulea 404a Ficulea 423 Ficulea 427c Ficulea 432b Ficulea 442 Ficulea 466a Ficulea 469 Ficulea 491 Ficulea 504a Ficulea 522 Ficulea 525 Ficulea 527 Ficulea 529 Ficulea 531 Ficulea 532 Ficulea 534d Ficulea 537a Ficulea 539f Ficulea 539i Ficulea 552 Ficulea 559 = Tibur III 243 Ficulea 570a = Tibur III 242a Ficulea 580 = Tibur III 244 Ficulea 599 = Collatia 58 Ficulea 601 = Collatia 30 Ficulea U1 Tibur III 12a Tibur III 29 Tibur III 64 Tibur III 69 Tibur III 70 Tibur III 74 Tibur III 90 Tibur III 92a Class 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 PH Arch ER x x x x x x x MR x x x x II BC x x x x x x x x x x x x ? x x x x x I AD II AD III AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x ? x x I BC x x x x x x x x x x x ? ? x x x x x x x ? ? x x x x ? x ? x x x x x x x x x x x x x x x x ? x x x ? x x x ? ? x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x ? x x IV AD V AD x x ? x x x x x x x x x x ? x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? ? x ? x x x x x x x ? ? x x x x x x x x x x x x x x x x x 199 appEndiX ii Name Tibur III 116a Tibur III 124 Tibur III 131 Tibur III 136 Tibur III 139 Tibur III 143 Tibur III 182 Tibur III 207a Tibur III 211a Tibur III 218 Tibur III 229a Tibur III 231 Tibur III 257 Tibur III 260 Tibur III 278a Tibur III 292 Tibur III 305a Tibur III 324 Tibur III 344a Tibur III 349 Tibur III 352a Tibur III 356 Tibur III 381a Tibur III 394 Tibur III 408 Tibur III 413 Tibur III 426 Tibur III 440 Tibur III U1 Tibur IV 19a Tibur IV 23 Tibur IV 36 Tibur IV 65 Tibur IV 66 Tibur IV 69 Tibur IV 76a Tibur IV 85 Tibur IV 101a Tibur IV 137a Tibur IV 167 Tibur IV 168 Tibur IV 196 Tibur IV 200 Tibur IV 214 Tibur I 119 Tibur I 163 Collatia 5b Collatia 38a Collatia 79 Collatia 93 Collatia 95a Collatia 98 Collatia 104a Collatia 111a Collatia 120 Collatia 129 Collatia 134a Collatia 136 Collatia 142 Collatia 155 Collatia 157 Collatia 170b Collatia 174a Collatia 177Ta Collatia 186a Collatia 195b Collatia 198a Collatia 227 Collatia 230a Collatia 236 Collatia 248 Collatia 256a Collatia 260 Collatia 261d Collatia 269 Collatia 280a Collatia 301 Collatia 326a Collatia 331 Collatia 343a Collatia 345 Collatia 349a 200 Class PH 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 x x Arch ER MR x x ? x ? x ? x ? x ? x x x ? x ? ? ? x x x ? II BC I BC I AD II AD III AD IV AD x x x x x ? x x x x x ? x x x x x x x x x x x x x x x x x x x x x ? x x ? x x x x x x x x x x x x x x x x ? x ? x ? x ? x x x x x x x ? ? ? ? x ? ? ? x ? x x x x x x x x x x x x ? ? x x x x x x x x x x x x x x ? ? x ? x x x x x x x ? x x x ? x x x x x x x ? x x ? ? ? ? x x x x x x x x x x x x x x x x ? ? ? ? x ? x x x x x x x x x x ? ? ? ? x x x x x ? x x x x x x x x x x x ? x x x x x x x x ? x x x ? x ? x x x x x x V AD x x x x x x x x x x x x x x x x x ? x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x ? x x x ? x x ? x ? x x x ? x x ? x ? x x x ? x x x x x x x x x x x ? x x x ? ? x ? x ? x x datEd sitEs Name Collatia 363 Collatia 368 Collatia 369a Collatia 376 Collatia 379a–b Collatia 383 Collatia 385 Collatia 387a–b Collatia 396 Collatia 403 Collatia 407 Collatia 409a Collatia 409b Collatia 410 Collatia 422b Collatia 437 Collatia 445c Collatia 456 Collatia 507b Collatia 528a Collatia 556a Collatia 559a–b Collatia 562 Collatia 569e Collatia 569g Collatia 578 Collatia 581 Collatia 592 Collatia 599a Collatia 603a Collatia 613b Collatia 614 Collatia 625b Collatia 642b Collatia 650 Collatia 653b Collatia 660b Collatia 662c Collatia 663b Collatia 688 Collatia 704 Collatia 706 Collatia 731 Collatia 739 Collatia 740 Collatia 748 Collatia 756a–b Collatia 771b Collatia 784 Collatia 788a Collatia 796 Collatia 798a Collatia 799 Collatia 802 Collatia 809a Collatia 825b Collatia 829b Collatia 832b Collatia 835a Collatia 838 Collatia 848a Collatia 848b Collatia 851 Collatia 854 Tellenae 30 Tellenae 32 Tellenae 42 Tellenae 55 Tellenae 81a Tellenae 102 Tellenae 103 Tellenae 157 Tellenae 166 Tellenae 171 Bovillae 29 Bovillae 32 Bovillae 41 Bovillae 46 Bovillae 74–75 Bovillae 145 Bovillae 166 Bovillae 201 Class 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 PH Arch ER MR x ? ? x x II BC I BC ? ? ? x x x x x x x x ? ? x x x ? ? ? x x ? ? x x x x x ? ? ? ? ? x ? ? ? ? ? x ? ? ? I AD II AD III AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x ? ? x ? x ? x x ? ? x ? ? ? ? ? ? ? ? ? ? x x x x x x x x x x x x x x x x x ? ? x ? x x x x x x x x x x x x x x x x x x x x x ? ? ? ? x ? x x ? x x ? IV AD V AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? ? x x x x x x x ? x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x ? ? ? 201 appEndiX ii Name Bovillae 220 Bovillae 250 Bovillae 264a Bovillae 296 Bovillae 311 Bovillae 313 Bovillae 333b Bovillae 335b Bovillae 339 Bovillae 356 Bovillae 371 Bovillae 374 Bovillae 383 Bovillae 414 Bovillae 416 Bovillae U2 Tusculum 5–8 Tusculum 46–48 Tusculum 63–65 Tusculum 74–75 Tusculum 79–82 Tusculum 116–117 Tusculum 130–132 Tusculum 133 Tusculum 143–148 Tusculum 151 Tusculum 156–158 Tusculum 167–171 Tusculum 174 Tusculum 181–182 Tusculum 196–201 Tusculum 256–258 Tusculum 260–264 Tusculum 393–394 Tusculum 395–396 Tusculum 414–418 Tusculum 423–425 Tusculum 540–545 Tusculum 581–583 Tusculum 603–605 Tusculum 616–622 Tusculum 623–624 Tusculum 632–633 Tusculum 699–701 Tusculum 752 Tusculum 786–787 Tusculum 808–814 Tusculum 820–821 Tusculum 903–905 Tusculum 907–908 Tusculum 915–917 Tusculum 921 Fidenae 16b Fidenae 20b Fidenae 22b Fidenae 24c Fidenae 26a Fidenae 32 Fidenae 45 Fidenae 61b Fidenae 77 Fidenae 80 Fidenae 94 Fidenae 95 Fidenae 99a Fidenae 101b Fidenae 106a Fidenae 113b Fidenae 117a Fidenae 120 Fidenae 127 Fidenae 133b Fidenae 137 Fidenae 156a Fidenae 165 Fidenae 166 Fidenae 174 Fidenae 178 Fidenae 183b Fidenae 189 Fidenae 190a Fidenae 196 202 Class 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 PH Arch ER ? ? MR II BC I BC I AD II AD III AD IV AD V AD x x x x x x ? x x x x x x x x x x x x x x x ? x x x x x x x x ? x x ? x x ? x ? x x x x x x x x x x x x ? ? x ? ? ? ? x ? x ? ? ? ? ? x x x x x x x x x x x ? ? x ? x ? x x x x ? x x x x x x ? x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x ? ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x datEd sitEs Name Fidenae 202 Fidenae 204 Fidenae 207a Fidenae 212e Fidenae 218b Fidenae 221 Fidenae 226 Fidenae 231 Fidenae 236 Fidenae 243 Fidenae 245b Fidenae 262 Fidenae 274 Ficulea E.3 Ficulea O.2 Ficulea 12 Ficulea 20a Ficulea 43 Ficulea 45 Ficulea 51 Ficulea 54 Ficulea 66b Ficulea 70 Ficulea 77 Ficulea 83 Ficulea 97b Ficulea 103 Ficulea 107 Ficulea 109 Ficulea 115 Ficulea 120 Ficulea 130 Ficulea 132 Ficulea 146a Ficulea 151 Ficulea 175 Ficulea 179a Ficulea 210 Ficulea 222a Ficulea 228e Ficulea 239 Ficulea 244a Ficulea 252 Ficulea 306b Ficulea 322b Ficulea 324 Ficulea 367 = Tibur III 17a Ficulea 412 Ficulea 420 Ficulea 424b Ficulea 454b Ficulea 456a Ficulea 463 Ficulea 464 Ficulea 467a Ficulea 485a Ficulea 515a Ficulea 543c Ficulea 546a Ficulea 550 Ficulea 560a = Tibur III 232b Ficulea 573b = Tibur III 249b? Ficulea 584c = Collatia 75a Ficulea 584d = Collatia 75b Ficulea 589b = Collatia 66a Ficulea 600 Ficulea 605 Tibur III 14 Tibur III 78a Tibur III 101a Tibur III 102 Tibur III 128 Tibur III 141 Tibur III 145a Tibur III 190 Tibur III 199b Tibur III 269 Tibur III 276a Tibur III 311 Tibur III 312 Tibur III 343a Tibur III 354 Class 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 PH Arch ER MR II BC I BC I AD II AD III AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? ? ? ? x x x x x x x x x ? x x x x x ? ? x ? x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x V AD x x x x x x x x x x x x x x x x x IV AD x x x x x x ? x x ? ? x x x x x x x x x x x x x ? x x x x x ? x x x x x x x x x x x x x x x x x x 203 appEndiX ii Name Tibur III 416 Tibur III 447 Tibur IV 48 Tibur IV 199 Tibur IV 212 Collatia 2 Collatia 11a Collatia 11b Collatia 22a Collatia 23a Collatia 24 Collatia 31b Collatia 35 Collatia 44a Collatia 53a Collatia 62b Collatia 71f Collatia 80d Collatia 82e Collatia 97a Collatia 99 Collatia 105a Collatia 125 Collatia 128a Collatia 138a Collatia 151a Collatia 178 Collatia 180b Collatia 180q1/3 Collatia 185a Collatia 188 Collatia 191b Collatia 194g Collatia 199b Collatia 205a Collatia 207 Collatia 208 Collatia 214 Collatia 219b Collatia 220b Collatia 221a Collatia 244 Collatia 263c Collatia 274 Collatia 294 Collatia 296a Collatia 303 Collatia 306a Collatia 317 Collatia 319 Collatia 333c Collatia 336b Collatia 336c–e Collatia 340a Collatia 362c Collatia 366a Collatia 367 Collatia 371 Collatia 382a Collatia 390 Collatia 397 Collatia 408 Collatia 413a Collatia 416a Collatia 419a Collatia 420b Collatia 424a Collatia 428 Collatia 429 Collatia 447c Collatia 448c Collatia 449b Collatia 459d Collatia 466 Collatia 468a Collatia 469a Collatia 470 Collatia 474 Collatia 490 Collatia 502a Collatia 504 Collatia 509 204 Class 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 PH Arch x x ER ? ? MR ? ? ? x ? ? x x II BC ? x ? x x x I BC I AD II AD III AD x x ? x x x x x ? x x x x x x x x x x x ? x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x ? x x x x x x x x x x x x x x x ? x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x ? x x x x x x x x ? x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x ? x x ? ? x x ? x ? x x x x ? ? x x x x ? x x x x x x ? ? x x x x x x x ? ? x x ? x x ? x x ? x ? x x ? x ? ? x ? x x x x x x x x ? x x x x x x x IV AD V AD x x x x x x x x x x x datEd sitEs Name Collatia 524b Collatia 530 Collatia 537 Collatia 546a Collatia 571 Collatia 577a Collatia 589a Collatia 610 Collatia 638 Collatia 652 Collatia 657e Collatia 662b Collatia 675a Collatia 676a Collatia 682I Collatia 683 Collatia 690 Collatia 694 Collatia 702 Collatia 707b Collatia 708 Collatia 710 Collatia 732a Collatia 735a Collatia 741 Collatia 762 = Tusculum 2 Collatia 767 Collatia 768a Collatia 769b Collatia 774a–b Collatia 779 Collatia 780b Collatia 790a Collatia 793a Collatia 792a Collatia 791b Collatia 797b Collatia 801b Collatia 806a Collatia 811a Collatia 813 Collatia 819b Collatia 822 Collatia 826a Collatia 829d Collatia 829c Collatia 833 Collatia 835b Collatia 836b Collatia 841a Collatia 843e Tellenae 11 Tellenae 20 Tellenae 26 Tellenae 27 Tellenae 51 Tellenae 129b Tellenae 137 Tellenae 140a Bovillae 2 Bovillae 4 Bovillae 5 Bovillae 53 = U3 Bovillae 88 Bovillae 120a Bovillae 248b Bovillae 340 Bovillae 430 Tusculum 106–108 Tusculum 114 Tusculum 189 Tusculum 246–250 Tusculum 627–628 Tusculum 643 Tusculum 661–662 Tusculum 828–829 Tusculum 840 Tusculum 901 Tusculum 914 Fidenae 9a Fidenae 12 Fidenae 14b Class 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 PH Arch ER MR x ? x ? II BC I BC ? x ? x x x x I AD II AD III AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x IV AD V AD x x x x ? x x x x x ? ? x ? ? x x ? x ? x x x x ? ? ? ? x x x x ? x ? ? ? ? ? x ? ? x x x x x x x x x x x x x x x x x x x ? x ? x x x x x x x x x x x ? x ? x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x 205 appEndiX ii Name Fidenae 17b Fidenae 49 Fidenae 50d–e Fidenae 75a–b Fidenae 76 Fidenae 82 Fidenae 89 Fidenae 90 Fidenae 93 Fidenae 98 Fidenae 109 Fidenae 116 Fidenae 119 Fidenae 123 Fidenae 125 Fidenae 126 Fidenae 128 Fidenae 131 Fidenae 134 Fidenae 142 Fidenae 145 Fidenae 154a Fidenae 164a Fidenae 181 Fidenae 192 Fidenae 193 Fidenae 208 Fidenae 209 Fidenae 219 Fidenae 223 Fidenae 227 Fidenae 229 Fidenae 232 Fidenae 237a Fidenae 240 Fidenae 244 Fidenae 248 Fidenae 251 Fidenae 253 Fidenae 254 Fidenae 259 Fidenae 266 Fidenae 276 Ficulea E.2 Ficulea P Ficulea Q Ficulea 3 Ficulea 4 Ficulea 5 Ficulea 9 Ficulea 11 Ficulea 15 Ficulea 17 Ficulea 56a Ficulea 60 Ficulea 66a Ficulea 67a Ficulea 69b Ficulea 79b Ficulea 85b Ficulea 94 Ficulea 99 Ficulea 100 Ficulea 101 Ficulea 108 Ficulea 112 Ficulea 117 Ficulea 118 Ficulea 123 Ficulea 128 Ficulea 149 Ficulea 153 Ficulea 160a Ficulea 171 Ficulea 184 Ficulea 189 Ficulea 195d Ficulea 202 Ficulea 206a Ficulea 206b Ficulea 211 Ficulea 213 206 Class 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 PH Arch ER MR x x x x x ? x ? ? ? x x II BC ? ? I BC ? ? ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x ? ? ? ? x ? x x x ? x x x x x x x x ? x x x x x x ? x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x I AD II AD III AD x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x IV AD V AD ? ? x x x x x x datEd sitEs Name Ficulea 226 Ficulea 232 Ficulea 235 Ficulea 240 Ficulea 241 Ficulea 245 Ficulea 248b Ficulea 257 Ficulea 291 Ficulea 299 Ficulea 301 Ficulea 319 Ficulea 322a Ficulea 330 Ficulea 331b Ficulea 332 Ficulea 336 Ficulea 337 Ficulea 349 Ficulea 352b = Tibur III 20b Ficulea 382a Ficulea 383 Ficulea 384 Ficulea 385c Ficulea 387a Ficulea 397 Ficulea 400 Ficulea 425b Ficulea 430 Ficulea 440 Ficulea 447 Ficulea 450 Ficulea 457 Ficulea 458b Ficulea 470b Ficulea 471 Ficulea 478 Ficulea 483 Ficulea 490 Ficulea 496b Ficulea 506 Ficulea 520 Ficulea 542 Ficulea 545b Ficulea 555 Ficulea 575 Ficulea 577c Ficulea 585b = Collatia 74 Ficulea 591 Ficulea 595 Ficulea 598 = Collatia 60 Tibur III 8 Tibur III 10 Tibur III 13 Tibur III 49 Tibur III 52 Tibur III 68 Tibur III 73 Tibur III 86 Tibur III 104 Tibur III 106 Tibur III 107 Tibur III 116b Tibur III 119 Tibur III 129 Tibur III 150 Tibur III 188 Tibur III 193 Tibur III 194 Tibur III 201 Tibur III 254d Tibur III 255 Tibur III 274 Tibur III 313b = Collatia 92 Tibur III 327 Tibur III 410 Tibur III 415 Tibur III 428 Tibur IV 54 Tibur IV 194a Tibur IV 213 Collatia 3 Class 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 PH Arch ER MR II BC I BC x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x I AD II AD III AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x ? x x ? x x x ? x ? x x ? x x x ? x ? ? ? x x ? x x x IV AD V AD x x x x x x x x x x x x x x x x x x x x x ? x ? x x ? ? x ? x x x ? x ? ? x ? ? x x x x ? ? x ? x ? x x ? x x x ? x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x ? ? x x x x ? ? x x ? ? x x x x x ? ? x x x x x x x x x x x x x x x x x x ? x x x x 207 appEndiX ii Name Collatia 6 Collatia 7 Collatia 8 Collatia 9 Collatia 10a Collatia 10b Collatia 15a Collatia 16 Collatia 17a Collatia 17b Collatia 21 Collatia 25 Collatia 29b Collatia 31c Collatia 33 Collatia 36a Collatia 37 Collatia 38b Collatia 40 Collatia 42 Collatia 46 Collatia 53c Collatia 56 Collatia 59a Collatia 59b Collatia 62a Collatia 63 Collatia 65 Collatia 67 Collatia 68 Collatia 76b Collatia 78 Collatia 85 Collatia 86 Collatia 101a Collatia 102a–b Collatia 111b Collatia 112 Collatia 114 Collatia 115 Collatia 117 Collatia 119 Collatia 124b Collatia 127a–b Collatia 130 Collatia 132 Collatia 135 Collatia 137a Collatia 143 Collatia 145 Collatia 148b Collatia 148c Collatia 150 Collatia 153 Collatia 154 Collatia 158 Collatia 159 Collatia 162 Collatia 167b Collatia 168b Collatia 171a Collatia 171b Collatia 173c Collatia 175 Collatia 176b Collatia 179c Collatia 181 Collatia 184 Collatia 186b Collatia 189a Collatia 190 Collatia 203 Collatia 206 Collatia 209 Collatia 212 Collatia 213 Collatia 215 Collatia 216 Collatia 217 Collatia 218 Collatia 222b–c Collatia 226e 208 Class 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 PH Arch ER MR II BC I BC x x x x x x x I AD II AD III AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x ? x x x x x x x ? x x x x x x x x x x x x x x IV AD V AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? ? x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x datEd sitEs Name Collatia 231 Collatia 234 Collatia 235 Collatia 237 Collatia 238 Collatia 243 Collatia 249a Collatia 268 Collatia 271 Collatia 277 Collatia 278a Collatia 286a Collatia 287 Collatia 288a Collatia 289 Collatia 291 Collatia 295 Collatia 297a Collatia 298 Collatia 300 Collatia 302 Collatia 304a Collatia 306b Collatia 307 Collatia 312 Collatia 313a Collatia 314 Collatia 315 Collatia 316 Collatia 318 Collatia 321c Collatia 322 Collatia 323 Collatia 324 Collatia 325b Ficulea 348a Collatia 328 Collatia 329 Collatia 335a Collatia 337c Collatia 339 Collatia 344 Collatia 347a Collatia 348 Collatia 350a Collatia 350b Collatia 356 Collatia 357c Collatia 358 Collatia 359a Collatia 361 Collatia 364a Collatia 374b Collatia 381 Collatia 389 Collatia 391 Collatia 393a Collatia 398a Collatia 400 Collatia 401a Collatia 401b Collatia 411 Collatia 413b Collatia 420a Collatia 421 Collatia 423 Collatia 424b Collatia 425a Collatia 425b Collatia 427 Collatia 430 Collatia 433 Collatia 434 Collatia 435 Collatia 446 Collatia 448b Collatia 450 Collatia 451a–b Collatia 453a Collatia 457 Collatia 461a Collatia 462 Class 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 PH Arch ER x x ? MR II BC I BC x ? ? x ? ? x ? ? x x x ? x x x x x x x x x x x x x x x x x ? ? ? ? ? ? ? ? ? x x x x x x ? x x x ? II AD III AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x ? x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x IV AD V AD x x x x x I AD ? x x ? ? x x x x x x x x x 209 appEndiX ii Name Collatia 464 Collatia 467 Collatia 471 Collatia 473 Collatia 475 Collatia 478 Collatia 480a Collatia 481 Collatia 483a Collatia 484 Collatia 485b Collatia 486 Collatia 491a Collatia 493 Collatia 494 Collatia 496a Collatia 500 Collatia 501a Collatia 503 Collatia 510 Collatia 511 Collatia 513 Collatia 514 Collatia 515a Collatia 518 Collatia 520 Collatia 522 Collatia 524c Collatia 525 Collatia 526 Collatia 527 Collatia 532 Collatia 533a Collatia 533b Collatia 534a Collatia 534b Collatia 536 Collatia 538 Collatia 539a Collatia 542 Collatia 543 Collatia 544b Collatia 545 Collatia 546b Collatia 547 Collatia 548 Collatia 557a Collatia 557b Collatia 561 Collatia 566 Collatia 572 Collatia 573 Collatia 574 Collatia 575a Collatia 576 Collatia 579 Collatia 580 Collatia 583b Collatia 585a Collatia 587 Collatia 595 Collatia 597 Collatia 605 Collatia 612 Collatia 615a Collatia 615b Collatia 621a Collatia 621b Collatia 623a–b Collatia 632 Collatia 635a Collatia 635b Collatia 636 Collatia 637a Collatia 639a Collatia 640a Collatia 643 Collatia 644 Collatia 654a Collatia 664 Collatia 665 Collatia 666 210 Class 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 PH Arch ER x MR II BC I BC x x x x I AD II AD III AD x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? ? ? ? ? x ? ? ? x x x x ? x ? ? ? x ? ? ? ? ? x x ? ? x ? ? ? x x x ? x x ? x x ? ? x x x ? x x x x x x x x ? ? ? ? x x x x x x x x x x x x x x x x IV AD V AD datEd sitEs Name Collatia 674 Collatia 677 Collatia 680a Collatia 684 Collatia 687 Collatia 689b Collatia 699 Collatia 705 Collatia 707a Collatia 709 Collatia 714a Collatia 714b Collatia 714c Collatia 715 Collatia 716 Collatia 717 Collatia 718a Collatia 718b Collatia 719 Collatia 720 Collatia 721 Collatia 722 Collatia 723 Collatia 724 Collatia 725 Collatia 726a Collatia 729a Collatia 729b Collatia 730 Collatia 732b Collatia 733 Collatia 734 Collatia 736a Collatia 736b Collatia 738 Collatia 742 Collatia 743 Collatia 746 Collatia 749 Collatia 752b Collatia 760a Collatia 760b Collatia 760c Collatia 763 Collatia 764a Collatia 764b Collatia 765ab Collatia 766 Collatia 768b Collatia 770 Collatia 773a Collatia 778 Collatia 781a Collatia 781b Collatia 782a Collatia 785 Collatia 786 Collatia 789 Collatia 792c Collatia 792d Collatia 792e Collatia 800a–b Collatia 804b Collatia 805a Collatia 811b Collatia 816 Collatia 818 Collatia 824a Collatia 824b Collatia 827 Collatia 825a Collatia 832c Collatia 834 Collatia 841b Collatia 844a Collatia 844b Collatia 846 Collatia 847 Collatia 852 Collatia 856a Collatia 856b Tellenae 39 Class 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 PH Arch ER MR II BC I BC I AD II AD III AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x IV AD V AD x x x x x x x x x x x x x x x x ? ? ? x x x ? x x ? x x x x x ? ? ? x x x x x x x x x x x x x ? ? ? ? x 211 appEndiX ii Name Class Tellenae 69 Tellenae 101 Tellenae 105 Tellenae U3 Bovillae 23 Bovillae 84 Bovillae 126b Bovillae 281 Bovillae 370 Tusculum 73 Tusculum 779 Fidenae 8b Fidenae 15 Fidenae 23 Fidenae 25b Fidenae 27 Fidenae 35 Fidenae 52b Fidenae 74c Fidenae 84 Fidenae 85 Fidenae 104 Fidenae 105 Fidenae 110 Fidenae 112a Fidenae 115 Fidenae 117b Fidenae 118 Fidenae 122 Fidenae 133a Fidenae 147 Fidenae 169 Fidenae 170 Fidenae 173 Fidenae 175 Fidenae 177 Fidenae 179 Fidenae 183a Fidenae 187 Fidenae 194 Fidenae 195 Fidenae 200 Fidenae 210b Fidenae 233 Fidenae 246 Fidenae 247 Fidenae 249 Fidenae 252 Fidenae 255 Fidenae 256 Fidenae 261 Fidenae 263 Fidenae 264 Fidenae 265 Fidenae 271b Fidenae 278 Ficulea A.1 Ficulea I Ficulea L Ficulea N Ficulea T Ficulea 6 Ficulea 8 Ficulea 18 Ficulea 58 Ficulea 61 Ficulea 64 Ficulea 65 Ficulea 69a Ficulea 71a Ficulea 76 Ficulea 78 Ficulea 79a Ficulea 90 Ficulea 95 Ficulea 97a Ficulea 98 Ficulea 102 Ficulea 110 Ficulea 111a Ficulea 113 Ficulea 114 4 4 4 4 4 4 4 4 4 4 4 Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early 212 PH Arch ER MR ? II BC ? I BC ? x x x x ? x x x x x x x x x x x ? ? x x x x x x x ? ? x x x x x x x x x x ? x x x ? x x x x x x x x x x x x x ? x x x ? x x x x x x x x x x x x x x x x x x x x ? ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x II AD III AD x x x x x x x x x x x x x x x x x ? ? ? ? x x x x x x x I AD x x x x x x x x x x x x x x IV AD V AD datEd sitEs Name Class Ficulea 122 Ficulea 125 Ficulea 140b Ficulea 145 Ficulea 150 Ficulea 156a Ficulea 159a Ficulea 162 Ficulea 164a Ficulea 166a Ficulea 174b Ficulea 178a Ficulea 182 Ficulea 186b Ficulea 187b Ficulea 188 Ficulea 190a Ficulea 194 Ficulea 195c Ficulea 198 Ficulea 199 Ficulea 203 Ficulea 207 Ficulea 209 Ficulea 212 Ficulea 215 Ficulea 219a Ficulea 220 Ficulea 224 Ficulea 227 Ficulea 228a Ficulea 242 Ficulea 247 Ficulea 248a Ficulea 256 Ficulea 296a Ficulea 298a Ficulea 300 Ficulea 304a Ficulea 306a Ficulea 308 Ficulea 310a Ficulea 314 Ficulea 315a Ficulea 317 Ficulea 318a Ficulea 320a Ficulea 321 Ficulea 323 Ficulea 325 Ficulea 326 Ficulea 328 Ficulea 331a Ficulea 334 Ficulea 338 Ficulea 339 Ficulea 340 Ficulea 342a = Tibur III 48 Ficulea 344 = Tibur III 183a Ficulea 345a = Tibur III 185 Ficulea 353 = Tibur III 19 Ficulea 361 Ficulea 366 = Tibur III 14*/19* Ficulea 368 = Tibur III 14*/19* Ficulea 371a Ficulea 373 Ficulea 376 Ficulea 381 Ficulea 385a/b Ficulea 386a Ficulea 388 Ficulea 398a Ficulea 399 Ficulea 401a Ficulea 402 Ficulea 407 Ficulea 418 Ficulea 419 Ficulea 421a Ficulea 422 Ficulea 424a Ficulea 425a Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early PH Arch ER MR x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x II BC I BC I AD II AD III AD IV AD V AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x 213 appEndiX ii Name Ficulea 427b Ficulea 428 Ficulea 429 Ficulea 431 Ficulea 432a Ficulea 433 Ficulea 436 Ficulea 437 Ficulea 438a Ficulea 445 Ficulea 448 Ficulea 449 Ficulea 451 Ficulea 452 Ficulea 453 Ficulea 454a Ficulea 458a Ficulea 459 Ficulea 461 Ficulea 462 Ficulea 470a Ficulea 474 Ficulea 476 Ficulea 480 Ficulea 484a Ficulea 504e Ficulea 507a Ficulea 530a Ficulea 538 Ficulea 539c Ficulea 545a Ficulea 558a Ficulea 577a Ficulea 581a Ficulea 583a Ficulea 584a Ficulea 585a = Collatia 74 Ficulea 587 Ficulea 590 Ficulea 594 = Collatia 61 Ficulea 603a Tibur III 11a Tibur III 75 Tibur III 84 Tibur III 100a Tibur III 127 Tibur III 134a Tibur III 140a Tibur III 142 Tibur III 144b Tibur III 154b Tibur III 192a Tibur III 206 Tibur III 351 Tibur III 425a Tibur IV 198a Collatia 11c Collatia 15c Collatia 18 Collatia 19a Collatia 39a Collatia 48 Collatia 69a Collatia 73a Collatia 90 Collatia 121 Collatia 123b Collatia 124a Collatia 126 Collatia 163 Collatia 173b Collatia 191a Collatia 222a Collatia 259 Collatia 266 Collatia 270 Collatia 272 Collatia 280b Collatia 281 Collatia 290 Collatia 325c Collatia 342 214 Class Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early PH Arch ER MR x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x x x x x x x x x x x x x x x x x x ? II BC I BC I AD II AD III AD IV AD V AD datEd sitEs Name Collatia 352 Collatia 353 Collatia 357b Collatia 364b Collatia 375 Collatia 393b Collatia 395 Collatia 406 Collatia 426 Collatia 461b Collatia 483b Collatia 487 Collatia 508 Collatia 512 Collatia 539c Collatia 551b Collatia 553a Collatia 575b Collatia 611 Collatia 617 Collatia 637b Collatia 654b Collatia 661a Collatia 772 Collatia 782b = Tusculum 1 Collatia 806b Collatia 809b Tusculum 28 Tusculum 609 Tusculum 610 Tusculum 611 Tusculum 612 Tusculum 702–703 Tusculum 778 Tusculum 895 Class Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early Early PH Arch ER MR II BC I BC I AD II AD III AD IV AD V AD x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ? x x x x 215 appEndiX iii rEMains rElatEd to agricultural production “Early” refers to settlement sites inhabited only before the 2nd century BC. “Near” means that theremains have been found near some previously known site. “Other” means sites probably not used for habitation. “Stray” is a stray ind. Site Fidenae 78 Fidenae 83 Fidenae 114 Fidenae 121 Fidenae 182a Fidenae 197 Fidenae 250 Ficulea 129a Ficulea 144 Ficulea 152a Ficulea 172 Ficulea 185a Ficulea 192 Ficulea 201a Ficulea 218 Ficulea 261a Ficulea 310b Ficulea 342b = Tibur III 47a Ficulea 350 = Tibur III 21 Ficulea 371b Ficulea 374a Ficulea 398b–c Ficulea 405a Ficulea 455a Ficulea 460a Ficulea 473a Ficulea 479 Ficulea 487a Ficulea 489a Ficulea 536a Ficulea 573a = Tibur III 249a Ficulea 581b Ficulea 604a Tibur III 55 Tibur III 80 Tibur III 108a Tibur III 109a Tibur III 118 Tibur III 138 Tibur III 195a Tibur III 217a Tibur III 263a Tibur III 298 Tibur III 304a Tibur III 308a Tibur III 313a = Collatia 92 Tibur III 387a Tibur III 390a Tibur III 396 Tibur III 432 Tibur III 434 Tibur III 449 Tibur IV 8 Tibur IV 20a Tibur IV 22a Tibur IV 27 Tibur IV 73 Class Type 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 trapetum torcularium; bones pars rustica pars rustica; ditches/furrows/pits torcularium torcularium torcularium trapetum? torcularium pars rustica torcularium? trapetum torcularium torcularium pars rustica pars rustica torcularium millstone torcularium (wine) torcularium (oil) trapetum pars rustica torcularium pars rustica pars rustica; ditches/furrows/pits torcularium meta of a mill pars rustica; bones torcularium; bones torcularium pars rustica; 3 millstones; ishponds?; ditches/furrows/pits millstone; bones pars rustica torcularium pars rustica torcularium trapetum torcularium (oil) torcularium torcularium pars rustica torcularium torcularium torcularium pars rustica torcularium (wine) pars rustica torcularium torcularium (wine) torcularium (wine) pars rustica torcularium (wine) pars rustica trapetum; glirarium pars rustica torcularium pars rustica Source rEMains rElatEd to agricultural production Site Tibur IV 100 Tibur IV 143 Tibur IV 149 Tibur IV 153 Tibur IV 202a Tibur IV 225 Tibur IV U1 Tibur I 198 Tibur I 209 Tibur II 85 Tibur II 214 Tibur II 217 Collatia 4d Collatia 43a–c Collatia 50a Collatia 106a Collatia 223 Collatia 229c Collatia 241 Collatia 242 Collatia 245 Collatia 258b Collatia 284 Collatia 321b Collatia 360a Collatia 362b Collatia 373 Collatia 377a Collatia 386 Collatia 392a Collatia 444 Collatia 454b Collatia 477 Collatia 489b Collatia 516a Collatia 517b Collatia 582a Collatia 583a Collatia 590e Collatia 593a–c Collatia 616b–c Collatia 618 Collatia 646 Collatia 649 Collatia 679a Collatia 695 Collatia 761a Collatia 771a Collatia 776 Collatia 787a Collatia 803 Tellenae 5b Tellenae 62 Tellenae 97a Tellenae 117a Tellenae U7 Bovillae 101 Bovillae 103 Bovillae 123 Bovillae 173a Bovillae 179–180 Bovillae 209 Bovillae 285b Bovillae 286 Fidenae 139a Tibur III 360a Tusculum 832–836 Fidenae 86 Fidenae 88b Fidenae 141 Fidenae 144 Fidenae 146a Fidenae 163 Fidenae 230 Ficulea S Ficulea 1 Ficulea 7b Ficulea 82 Ficulea 119 Ficulea 169a Ficulea 228 Ficulea 243 Class Type 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 Source pars rustica pars rustica torcularium pars rustica torcularium pars rustica; bones ishponds ishponds ishponds; aviary? torcularium (wine) pars rustica; ishponds? pars rustica torcularium meta of a mill pars rustica; ditches/furrows/pits torcularium (oil); ditches/furrows/pits meta of a mill torcularium torcularium torcularium? torcularium trapetum torcularium (oil) torcularium torcularium torcularium trapetum? torcularium torcularium; bones; ditches/furrows/pits pars rustica; bones torcularium pars rustica pars rustica; glirarium catillus & meta of a mill torcularium trapetum meta of a mill pars rustica pars rustica trapetum? torcularium torcularium torcularium pars rustica separate pars rustica? ditches/furrows/pits trapetum torcularium pars rustica; plant remains pars rustica meta of a mill millstones millstone pars rustica millstone separate pars rustica?; torcularium? separate pars rustica? pars rustica; ishponds? trapetum pars rustica trapetum millstone torcularium millstone pars rustica; bones pars rustica pars rustica torcularium pars rustica pars rustica torcularium (wine) torcularium (wine) pars rustica millstone trapetum torcularium torcularium torcularium pars rustica catillus of a mill? horrea?; ditches/furrows/pits; more than one site?; Late Rep.–Imp. catillus & meta of a mill 217 appEndiX iii Site Ficulea 378b Ficulea 525 Ficulea 552 Ficulea 601 = Collatia 30 Ficulea U1 Tibur III 29 Tibur III 69 Tibur III 90 Tibur III 93a Tibur III 112 Tibur III 136 Tibur III 292 Tibur III 324 Tibur III 344a Tibur IV 19a Tibur IV 45a Tibur IV 49a Tibur IV 167 Tibur IV 200 Collatia 49b Collatia 87b Collatia 131 Collatia 201a Collatia 230a Collatia 248 Collatia 331 Collatia 345 Collatia 409a Collatia 456 Collatia 507b Collatia 556a Collatia 642b Collatia 663b Collatia 739 Collatia 740 Collatia 756c Collatia 825b Tellenae 30 Tellenae 46b Tellenae 64a Bovillae 32 Bovillae 172 Bovillae U2 Tusculum 46–48 Tusculum 196–201 Tusculum 256–258 Tusculum 414–418 Tusculum 603–605 Ficulea 315b Ficulea 534d Collatia 614 Fidenae 172a–b Collatia 301 Fidenae 174 Fidenae 178 Fidenae 226 Fidenae 243 Ficulea 210 Ficulea 294 = Tibur III 175a Ficulea 367 = Tibur III 17a Ficulea 424b Ficulea 463 Ficulea 485 Ficulea 584d = Collatia 75b Tibur IV 51b Tibur IV 199 Tibur IV 215 Tibur IV 218 Collatia 55c Collatia 221a Collatia 317 Collatia 319 Collatia 333 Collatia 382a Collatia 390 Collatia 408 Collatia 419a Collatia 509 Collatia 555a Collatia 610 Collatia 683 Collatia 779 218 Class Type 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 catillus & meta of a mill meta of a mill ditches/furrows/pits pars rustica pars rustica meta of a mill pars rustica torcularium trapetum trapetum catillus of a mill meta of a mill millstone torcularium torcularium torcularium (oil) parts of a mill torcularium pars rustica trapetum trapetum meta of a mill torcularium catillus & meta of a mill torcularium (oil) meta of a mill torcularium (wine); ditches/furrows/pits; bones meta of a mill torcularium (oil) meta of a mill pars rustica meta of a mill torcularium (wine) meta of a mill meta of a mill torcularium meta of a mill millstone torcularium millstone torcularium trapetum torcularium pars rustica meta of a mill torcularium bones torcularium bones torcularium? bones, shells ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits; Imp. ditches/furrows/pits; Rep. catillus of a mill pars rustica; ditches/furrows/pits trapetum torcularium (wine) catillus of a mill torcularium torcularium ditches/furrows/pits meta of a mill pars rustica torcularium torcularium (oil) torcularium torcularium torcularium torcularium trapetum ditches/furrows/pits torcularium meta of a mill meta of a mill torcularium meta of a mill meta of a mill torcularium meta of a mill torcularium Source rEMains rElatEd to agricultural production Site Class Type Collatia 797b Bovillae 88 Fidenae 75a–b Ficulea 171 Collatia 307 Fidenae 95 Fidenae 98 Collatia 302 Ficulea A Collatia 57 Fidenae 183 Fidenae 221 Fidenae 232 Ficulea 119 Ficulea 174 Ficulea 356a = Tibur III 6a Ficulea 378 Ficulea 382 Ficulea 383 Ficulea 398 Ficulea 404a Ficulea 415 Ficulea 422 Ficulea 433 Ficulea 530 Ficulea 536 Ficulea 539 Ficulea 545 Collatia 15 Collatia 15 Collatia 22 Collatia 50 Collatia 90 Collatia 221 Collatia 230 Collatia 240 Collatia 245 Collatia 270 Collatia 299 Collatia 310 Collatia 318 Collatia 580 Collatia 679 Tellenae 150 Fidenae 140g Ficulea 217 Ficulea 414b Ficulea 528c Tibur III 50b Tibur IV 51a Collatia 100 Collatia 180n Collatia 624d Bovillae 138 Tibur III 87a Tibur III 121a Tibur III 151 Tibur III 180a Tibur III 384 Tibur IV 39b Collatia 177Ub Tellenae 44 Tellenae 86 Tellenae 173 3 3 4 4 4 4 4 4 Early Early near near near near near near near near near near near near near near near near near near near near near near near near near near near near near near near near near near Other Other Other Other Other Other Other Other Other Other Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray meta of a mill? pars rustica (oil?) mill parts 2 metae of a mill torcularium; ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits catillus of a mill ditches/furrows/pits; Rep. ditches/furrows/pits; Mid-Rep.–Late Rep. ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits; Mid-Rep. trapetum; ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits; Imp. ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits torcularium ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits; more than one site? ditches/furrows/pits; more than one site? ditches/furrows/pits ditches/furrows/pits ditches/furrows/pits torcularium ditches/furrows/pits agricultural terrace? agricultural terrace? ditches/furrows/pits pars rustica ditches/furrows/pits ishponds? torcularium ditches/furrows/pits millstone? millstone? torcularium? meta of a mill? parts of a mill? meta of a mill? meta of a mill? meta of a mill? millstone? millstone? millstone? millstone? near near Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other ditches/furrows/pits ditches/furrows/pits; Mid-Rep.–Late Rep. ishponds? agricultural terraces ditches/furrows/pits torcularium ditches/furrows/pits; Late Rep. ditches/furrows/pits ditches/furrows/pits; Late Ant./Med. ditches/furrows/pits; more than one site; Roman ditches/furrows/pits; more than one site; Roman ditches/furrows/pits ishponds ditches/furrows/pits ditches/furrows/pits; Med. ditches/furrows/pits ditches/furrows/pits; Med. Source OUTSIDE THE RESEARCH AREA Collatia 296 Collatia 328 Tibur IV 119a/b Tibur II 159b Via Flaminia km 23 Nomentum 1,102 Crustumerium, town Monte Mario S. Agata, Rome Porta Pia, Rome TAV km 17.609–18.490 Near Fosso di San Giuliano Acquafredda, Via Aurelia km 9.600 Monteverde, Rome Corcolle, Via Lunano Via Appia UT115 Via dei Granai di Nerva Via Grotta Perfetta Quilici Gigli 1987 Pala 1976; De Seña 2005, Nr. 12 Amoroso 2000 Quilici Gigli 1987 Gilkes et al. 1994 Musco 2001 Musco 2001 Quilici Gigli 1987 Higginbotham 1997 Musco et al. 2002 Spera 1999 Santangeli Valenziani and Volpe 1988 Ricciardi 2002; 2005 219 appEndiX iii Site Class Type Torrino Other ditches/furrows/pits; more than one site?; Mid-Rep. Acqua Acetosa–Laurentina Other ditches/furrows/pits; Arch.–Rep. Villa di Quarto Cappello del Prete Villa pars rustica; ishponds? Villa di Casal Bertone Villa torcularium? Villa di Torre Spaccata, Sito 1, Via Prenestina/ Villa torcularium Via M. Lizzani Villa della Via Appia Nuova, Via Appia Villa torcularium Nuova km 7 Villa di Via della Magliana Villa torcularium Villa del Torrino, Sito 10 Villa torcularium Tibur III 170 Villa torcularium Tibur III 223a Villa torcularium Tibur III 456a Villa torcularium Tibur IV 14a Villa torcularium Tibur IV 18a Villa torcularium Tibur IV 120a Villa torcularium Tibur IV 127 Villa torcularium Tibur IV 209b Villa trapetum Tibur II 14 Villa ishponds? Tibur II 49 Villa torcularium Tibur II 70 Villa torcularium Tibur II 121b Villa aviary? Tibur II 207 Villa torcularium Tibur II 229a Villa torcularium Tusculum 295–299 Villa meta of a mill Villa di Fiano Romano “Della Standa” Villa torcularium Villa dei Volusii a Lucus Feroniae, Fiano Villa torcularium Romano Villa di Procoio Nuovo, Via Tiberina km 7 Villa torcularium Nomentum 300 Villa torcularium Nomentum 216 Villa millstone Nomentum 94 Villa millstone Nomentum 257 Villa millstone Villa di Campetti a Veio Villa torcularium? Villa della Via Tiberina, Via Tiberina km 3.500 Villa torcularium Villa di Quarto di Montebello Villa millstone Nomentum 70 Villa millstone Nomentum 37 Villa millstone Crustumerium 109 Villa meta of a mill Villa del Cimitero Flaminio Villa torcularium Crustumerium 65 Villa mill parts Villa della Via Tiberina, Via Tiberina km 0.850 Villa torcularium Villa di Livia a Prima Porta Villa torcularium? Crustumerium 74 Villa torcularium Villa di Via Barbarano Romano Villa millstone Villa di Casale Ghella Villa torcularium Villa della Borgata Ottavia Villa torcularium? Via del Fosso della Crescenza Villa ditches/furrows/pits; Late Rep.–Imp. Villa della Muracciola Villa torcularium Villa di Grottarossa Villa torcularium?; ishponds Villa dell’Auditorium Villa torcularium Via Romagnosi, Rome Villa torcularium Galleria d’arte moderna, Rome Villa ishponds Villa di Via dei Casalotti Villa torcularium Via Manfredi, Rome Villa torcularium Via Carpineto Sinello Nr. 11 Villa ditches/furrows/pits Villa di Casal Bruciato Villa torcularium?; millstone Villa di Tor de’ Schiavi “Dei Gordiani” Villa torcularium Praeneste 130 Villa torcularium Villa di Castel di Guido Villa torcularium Via Appia UT39 Villa pars rustica Villa di Via Togliatti Villa torcularium Centocelle, Via Casilina Villa ditches/furrows/pits; more than one site Villa di Centocelle “Della Piscina” Villa ishponds? Villa di Fregene, Loc. Campo delle Corse Villa torcularium? Praeneste 79 Villa torcularium Villa di Cinecittà, Loc. Subaugusta Villa torcularium Villa di Numisia Procula a Tor Marancia Villa torcularium? Villa delle Vignacce Villa separate pars rustica? Villa di Via del Quadraro Villa pars rustica? Villa di Tor Carbone Villa separate pars rustica; torcularium? Villa della Via Ardeatina, Via Ardeatina km 5 Villa pars rustica? Villa di Dragoncello, Sito A Villa torcularium? Villa di Acilia, Loc. Fralana Villa torcularium Villa di Dragoncello, Sito F Villa pars rustica Apiolae 20 Villa torcularium Apiolae 7 Villa ditches/furrows/pits Apiolae 372 Villa millstone Apiolae 84 Villa millstone Albano Laziale, Loc. S. Maria della Stella Villa torcularium Villa di Castelfusano, “Di Plinio” Villa ishponds? Apiolae 225 Villa torcularium; ditches/furrows/pits Tibur II 46 Villa? meta of a mill 220 Source Bedini 1984; 1997 Bedini 1984; 1997 De Franceschini 2005, Nr. 55 De Franceschini 2005, Nr. 52 De Franceschini 2005, Nr. 65 De Franceschini 2005, Nr. 77 De Franceschini 2005, Nr. 86 De Franceschini 2005, Nr. 87 De Franceschini 2005, Nr. 98 De Franceschini 2005, Nr. 99 De Franceschini 2005, Nr. 1 Pala 1976 Pala 1976 Pala 1976 Pala 1976 De Franceschini 2005, Nr. 2 De Franceschini 2005, Nr. 3 De Franceschini 2005, Nr. 4 Pala 1976 Pala 1976 Quilici and Quilici Gigli 1980 De Franceschini 2005, Nr. 5 Quilici and Quilici Gigli 1980 De Franceschini 2005, Nr. 6 De Franceschini 2005, Nr. 7 De Franceschini 2005, Nr. 9 De Franceschini 2005, Nr. 14 De Franceschini 2005, Nr. 16 De Franceschini 2005, Nr. 18 Messineo 2005 De Franceschini 2005, Nr. 21 De Franceschini 2005, Nr. 28 De Franceschini 2005, Nr. 39 Pavolini et al. 2003 Pavolini et al. 2003 De Franceschini 2005, Nr. 47 Pavolini et al. 2003 Musco et al. 2002 De Franceschini 2005, Nr. 51 De Franceschini 2005, Nr. 53 Muzzioli 1970 De Franceschini 2005, Nr. 54 Spera 1999 De Franceschini 2005, Nr. 63 De Franceschini 2005, Nr. 60–2 De Franceschini 2005, Nr. 60 De Franceschini 2005, Nr. 68 Muzzioli 1970 De Franceschini 2005, Nr. 67 De Franceschini 2005, Nr. 72 De Franceschini 2005, Nr. 71 De Franceschini 2005, Nr. 70 De Franceschini 2005, Nr. 78 De Franceschini 2005, Nr. 80 De Franceschini 2005, Nr. 90 De Franceschini 2005, Nr. 92 De Franceschini 2005, Nr. 91 De Rossi 1970 De Rossi 1970 De Rossi 1970 De Rossi 1970 Caserta 2006 De Franceschini 2005, Nr. 93 De Rossi 1970; De Franceschini 2005, Nr. 97 rEMains rElatEd to agricultural production Site Class Type Source Tibur II 186 Tibur II 202 Tibur II 205 Tusculum 327–328 Nomentum 89 Torrimpietra 368 Torrimpietra 478 Torrimpietra 31D Torrimpietra 172 Torrimpietra 57 Torrimpietra 461 Torrimpietra 203 Via Fleming, Rome Torrimpietra 661 Torrimpietra 274 Corcolle, Colle Sant’Angeletto Praeneste 82 Villa di Dragoncello, Sito G Apiolae 36 Apiolae 444 Apiolae 295a Apiolae 453 Apiolae 79 Apiolae 101 Apiolae 109 Apiolae 206 Apiolae 161 Apiolae 245 Nomentum 151 Nomentum 1,38 Nomentum 1,41 Nomentum 354 Crustumerium 9 Torrimpietra 340 Torrimpietra 502 Torrimpietra 551 Torrimpietra 732 Torrimpietra 773 Torrimpietra 802f Torrimpietra 231 Torrimpietra 261 Torrimpietra 681 Torrimpietra 279 Praeneste 140 Apiolae 269 Apiolae 218 Apiolae 148 Apiolae 425a Veii 536 Veii 76 Veii 127 Veii 521 Veii 438 Veii 190 Veii 313 Veii 239 Veii 368 Veii 363 Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Villa? Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray Stray? Stray? Stray? Stray? Stray? Stray? Stray? Stray? Stray? Stray? torcularium agricultural terraces ishponds? meta of a mill torcularium millstone millstone torcularium millstone millstone millstone millstone ditches/furrows/pits millstone millstone ditches/furrows/pits; Late Rep. torcularium pars rustica torcularium millstones torcularium torcularium millstone millstone torcularium torcularium; ditches/furrows/pits torcularium torcularium millstone? millstone? millstone? millstone? catillus & meta of a mill millstone millstone? millstone? millstone millstone? millstone? millstone millstone millstone? trapetum? torcularium? torcularium? torcularium? millstone? millstone? millstone millstone millstone millstone millstone catillus of a mill quern 2 rotary mills millstone quern Pala 1976 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Quilici Gigli 1987 Tartara 1999 Tartara 1999 Musco et al. 1995 Muzzioli 1970 De Franceschini 2005, Nr. 89 De Rossi 1970 De Rossi 1970 De Rossi 1970 De Rossi 1970 De Rossi 1970 De Rossi 1970 De Rossi 1970 De Rossi 1970; Quilici Gigli 1987 De Rossi 1970 De Rossi 1970 Pala 1976 Pala 1976 Pala 1976 Pala 1976 Quilici and Quilici Gigli 1980 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Tartara 1999 Muzzioli 1970 De Rossi 1970 De Rossi 1970 De Rossi 1970 De Rossi 1970 Kahane et al. 1968 Kahane et al. 1968 Kahane et al. 1968 Kahane et al. 1968 Kahane et al. 1968 Kahane et al. 1968 Kahane et al. 1968 Kahane et al. 1968 Kahane et al. 1968 Kahane et al. 1968 torcularium torcularium torcularium ishponds De Seña 2005, Nr. 8 De Seña 2005, Nr. 4 De Seña 2005, Nr. 5 Higginbotham 1997 OUTSIDE THE MAP AREA Ager Capenas, Monte Canino Sutri, Poggiolo Suligano Sutri, fattoria Contea Flacchi Villa of Horace, Licenza 221 appEndiX iv watEr installations “Other” means sites probably not used for habitation. The sizes of all cisterns could not be calculated. See, e.g., Fidenae 139a where the area refers to only one cistern. Site Fidenae 1 Fidenae 8a Fidenae 31a Fidenae 71a Fidenae 78 Fidenae 83 Fidenae 114 Fidenae 121 Fidenae 139a Fidenae 182a Fidenae 197 Ficulea 75 Ficulea 129a Ficulea 144 Ficulea 148a Ficulea 152a Ficulea 163a Ficulea 172 Ficulea 180b Ficulea 187a Ficulea 192 Ficulea 201a Ficulea 218 Ficulea 219b Ficulea 237 Ficulea 258 Ficulea 261a Ficulea 302 Ficulea 327 Ficulea 329a Ficulea 342b = Tibur III 47a Ficulea 350 = Tibur III 21 Ficulea 356a = Tibur III 6a Ficulea 360 = Tibur III 2a Ficulea 364a Ficulea 371b Ficulea 374a Ficulea 380a Ficulea 398b–c Ficulea 405a Ficulea 415a Ficulea 415c Ficulea 435 Ficulea 455a Ficulea 460a Ficulea 473a Ficulea 479 Ficulea 487a Ficulea 489a Ficulea 503a/c = Collatia 1b–c Ficulea 507b = Collatia 2–4 Ficulea 536a Ficulea 540a Ficulea 573a = Tibur III 249a Ficulea 581b Ficulea 604a Tibur III 22 Tibur III 35 Tibur III 42a Tibur III 45a Tibur III 55 Class Water Installations 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 bath well; cistern (246 m²); lead pipe? basin; bath? bath cistern; masonry channel; basin; bath cistern (155 m²); masonry channel aqueduct cistern; masonry channel; bath? well; impluvium; 2 cisterns (10 & 19 m²); masonry channel; nymphaeum; bath well; impluvium; 2 cisterns (72 m², a cunicoli); masonry channel; basin; bath 2 cisterns (371 m², a cunicoli); basin cuniculi; lead pipe; basin; bath (site 74) cistern (4 m²); masonry channel; lead pipe?; basin; bath well; 3 cisterns (72 & 81 m²); masonry channel; lead pipe?; basin; bath water channels well; clay pipe masonry channel; lead pipe well; impluvium; masonry channel; basin lead pipe?; bath? well; cistern; masonry channel; basin cuniculi; bath well; cistern; lead pipe; basin; nymphaeum; bath bath lead pipe; bath? clay pipe; bath bath? spring masonry channel cistern; cuniculi cistern (115 m²); cuniculi cistern (9 m²); cuniculi; bath? 2 cisterns; bath 2 cisterns (187 m²); clay pipe; lead pipe?; bath? cistern; cuniculi cistern (198 m²); cuniculi; bath? cistern (48 m²) impluvium?; cistern?; cuniculi; bath cistern? well; 3 cisterns (14 & 265 m²); basin; bath cistern (108 m²) cistern (89 m²); bath basin; bath; lead pipe cistern spring; well; cuniculi; lead pipe?; basin; bath cuniculi impluvium; cistern (27 m²); bath 2 cisterns (a cunicoli?); bath well; impluvium; cistern (a cunicoli); basin cistern (25 m²); masonry channel basin well; cuniculi impluvium; 3 cisterns; bath cistern (31 m²); basin cistern (71 m²); cuniculi; basin; bath? masonry channel; clay pipe; bath 2 cisterns? (90 m², a cunicoli?); nymphaeum; bath? well; cistern (120 m²); cuniculi; basin; bath cistern?; cuniculi; masonry channel; clay pipe 2 cisterns (14 & 27 m², a cunicoli?) cistern (360 m²); bath bath? watEr installations Site Tibur III 80 Tibur III 83a Tibur III 108a Tibur III 109a Tibur III 110 Tibur III 118 Tibur III 138 Tibur III 144a Tibur III 195a Tibur III 197a Tibur III 202a Tibur III 217a Tibur III 254a Tibur III 265a Tibur III 270a Tibur III 277a Tibur III 287a Tibur III 298 Tibur III 302a Tibur III 303a Tibur III 304a Tibur III 308a Tibur III 318a = Collatia 296d = Tibur IV 216a Tibur III 325a Tibur III 345a Tibur III 346 Tibur III 369 Tibur III 375 Tibur III 387a Tibur III 390a Tibur III 393 Tibur III 396 Tibur III 402a Tibur III 414 Tibur III 429a Tibur III 432 Tibur III 448a Tibur III 449 Tibur IV 1a Tibur IV 8 Tibur IV 11a Tibur IV 16 Tibur IV 20a Tibur IV 22a Tibur IV 27 Tibur IV 37a Tibur IV 40 Tibur IV 41a Tibur IV 44a Tibur IV 60 Tibur IV 67 Tibur IV 68 Tibur IV 70 Tibur IV 72 Tibur IV 73 Tibur IV 86 Tibur IV 87 Tibur IV 100 Tibur IV 104 Tibur IV 141a/c Tibur IV 143 Tibur IV 148 Tibur IV 149 Tibur IV 153 Tibur IV 157 Tibur IV 202a Tibur IV 224 Tibur IV 225 Tibur IV U1 Tibur I 96 Tibur I 99 Tibur I 106 Tibur I 131 Tibur I 136 Tibur I 198–200 Tibur I 205 Tibur I 207–208 Tibur I 209 Tibur I 212 Tibur II 85 Tibur II 88 Tibur II 214 Class Water Installations 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 cistern cistern cistern (66 m²); cuniculi; bath cistern (21 m²); masonry channel; lead pipe; basin; bath? lead pipe spring; cuniculi; clay pipe; lead pipe; basin cistern (38 m²) bath? cuniculi; lead pipe; basin; bath cuniculi cistern (75 m²); cuniculi; bath? cistern cistern (104 m²); cuniculi; lead pipe clay pipe well; cistern? (a cunicoli?); cuniculi cistern cistern cistern?; basin; bath? impluvium?; 2 cisterns (both a cunicoli?) spring; lead pipe lead pipe impluvium? cuniculi; masonry channel; clay pipe?; bath? cistern (58 m²); bath? cistern (131 m²); basin cistern cistern (5 m²); masonry channel; bath impluvium?; cistern (45 m²); masonry channel bath impluvium; cistern; basin; bath spring; masonry channel? well; basin well; impluvium?; cistern; basin? cistern (128 m²); basin; bath? cistern (144 m²); bath impluvium?; basin cistern; clay pipe bath? cistern?; bath basin well; aqueduct; cistern?; cuniculi; clay pipe; basin; bath cistern (191 m²) 2 cisterns (9 m²); lead pipe cistern (40 m²) cuniculi? 2 cisterns (84 m²) cistern (348 m²); clay pipe; bath? well; cuniculi; masonry channel well; clay pipe; basin spring; cistern? (a cunicoli?); cuniculi; lead pipe? cistern?; masonry channel; basin cistern (26 m²) cistern (15 m²); cuniculi; clay pipe; basin; bath well?; 2 cisterns (23 m²); cuniculi; basin; bath cistern; cuniculi; masonry channel; basin; nymphaeum cistern? cistern?; cuniculi?; basin; nymphaeum well?; impluvium?; cistern (100 m²); basin aqueduct?; cistern (225 m²); cuniculi; nymphaeum; bath lead pipe; bath? aqueduct?; basin; bath? 2 cisterns (9 & 20 m²); clay pipe; bath? cistern spring?; well; aqueduct; cistern? (147 m²); cuniculi; masonry channel; clay pipe; lead pipe; basin; bath well; aqueduct; 4 cisterns (30 & 80 m²); masonry channel; basin; nymphaeum; bath well?; masonry channel; basin 2 cisterns; basin; nymphaeum 3 cisterns (72 m²); masonry channel; nymphaeum spring; well; aqueduct; masonry channel; clay pipe; lead pipe; basin; nymphaeum; bath cistern (a cunicoli?); cuniculi nymphaeum impluvium?; cistern? (64 m²); nymphaeum? cistern 2 cisterns aqueduct; lead pipe?; basin; nymphaeum clay pipe?; basin?; nymphaeum 2 cisterns (31 & 36 m²); nymphaeum aqueduct; 3 cisterns (41, 45 & 1346 m²); cuniculi; basin; bath? well?; aqueduct; 2 cisterns (100 m²); nymphaeum impluvium?; cistern (323 m²); basin lead pipe?; nymphaeum cistern (73 m²); cuniculi?; lead pipe?; basin; nymphaeum 223 appEndiX iv Site Tibur II 217 Collatia 4d Collatia 14b Collatia 43a–c Collatia 45c Collatia 50a Collatia 106a Collatia 123a Collatia 147 Collatia 156a Collatia 177S Collatia 195a Collatia 202a Collatia 224p Collatia 242 Collatia 245 Collatia 282 Collatia 284 Collatia 321b Collatia 360a Collatia 362b Collatia 365a Collatia 372a–c Collatia 377a Collatia 386 Collatia 392a Collatia 394a Collatia 405b Collatia 431 Collatia 432 Collatia 441 Collatia 444 Collatia 454b Collatia 460 Collatia 472 Collatia 477 Collatia 479b Collatia 485a Collatia 489b Collatia 495a Collatia 517b Collatia 540 Collatia 552a Collatia 560b Collatia 567a Collatia 583a Collatia 584 Collatia 586 Collatia 590e Collatia 591f/j Collatia 593a Collatia 594 Collatia 598a Collatia 601a Collatia 604a Collatia 616b–c Collatia 618 Collatia 628a Collatia 634a Collatia 646 Collatia 649 Collatia 651a Collatia 679a Collatia 695a Collatia 713a Collatia 761a Collatia 771a Collatia 776 Collatia 787a Collatia 795 Collatia 803 Collatia 812 Collatia 828 Collatia 832a Collatia 842 Collatia 855 Tellenae 5b Tellenae 29a Tellenae 37b Tellenae 88 Tellenae 97a Tellenae U7 224 Class Water Installations 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 aqueduct; cistern (33 m²); basin; nymphaeum cistern (70 m²); basin; bath? aqueduct; cistern (a cunicoli) cuniculi; bath? bath well; cistern (a cunicoli); cuniculi; clay pipe well; cistern?; masonry channel; bath cistern; cuniculi cistern (a cunicoli) well; clay pipe cistern; masonry channel; bath clay pipe; bath cistern (42 m²) cistern? (a cunicoli?); cuniculi well; impluvium; clay pipe; basin cistern? (a cunicoli?); basin 2 cisterns (24 & 39 m²); masonry channel cistern (68 m²); masonry channel cistern cistern (a cunicoli) cistern (70 m²); bath? clay pipe bath cistern? (128 m²); bath? well; impluvium; cistern? (a cunicoli?); cuniculi; basin; bath well; bath well; cistern; cuniculi; clay pipe; lead pipe; bath aqueduct; bath masonry channel; basin; bath basin; bath aqueduct; cistern; lead pipe?; basin well; cistern; basin impluvium?; 2 cisterns (a cunicoli?); cuniculi; masonry channel; lead pipe; basin cistern (29 m²); cuniculi; masonry channel cistern? (a cunicoli?) cuniculi cistern (82 m²); cuniculi; basin; bath masonry channel; basin; bath cistern 2 cisterns (59 & 79 m²); cuniculi 2 cisterns (a cunicoli?); cuniculi cistern (1400 m²) cuniculi cistern (44 m²); basin cistern (4 m²); bath? bath? well cistern (44 m²) well; basin; bath? cistern (660 m²); clay pipe; bath 2 cisterns (59 & 72 m²) cistern (28 m²); clay pipe masonry channel; bath aqueduct?; cuniculi 2 cisterns? (a cunicoli?); masonry channel; basin; bath? bath bath 2 cisterns (23 & 27 m²); masonry channel; clay pipe cistern (44 m²); basin well; impluvium; 2 cisterns (385 m²); basin; bath aqueduct; cistern; masonry channel; basin; bath bath aqueduct; 4 cisterns (143 m²); cuniculi; masonry channel; basin; nymphaeum; bath well?; cistern (29 m²); lead pipe well?; 2 cisterns (a cunicoli?); cuniculi basin basin well; cistern; basin cistern; cuniculi cistern cistern? (669 m²); cuniculi cistern (45 m²); masonry channel; clay pipe? cuniculi lead pipe?; bath? cistern? (a cunicoli?) well; cistern? cistern (55 m²); lead pipe; bath clay pipe cistern (254 m²); cuniculi; basin; nymphaeum? clay pipe; bath? impluvium; 2 cisterns (58 m²; a cunicoli); clay pipe; bath? aqueduct; 5 cisterns (124, 124, 144 & 365 m²); masonry channel; clay pipe; lead pipe; basin; nymphaeum; bath watEr installations Site Class Water Installations Bovillae 11 Bovillae 19 Bovillae 48 Bovillae 96 1 1 1 1 Bovillae 101 1 Bovillae 103 Bovillae 121.1 Bovillae 123a Bovillae 131 Bovillae 141 Bovillae 165 Bovillae 169 Bovillae 173a Bovillae 179–180 Bovillae 200 Bovillae 209 Bovillae 222a Bovillae 254 Bovillae 273 Bovillae 285b Bovillae 286 Bovillae 337 Bovillae 387 Bovillae 391 Bovillae 404 Bovillae 423 Bovillae 432 Bovillae U1 Tusculum 35–41 Tusculum 135–136 Tusculum 154 Tusculum 190–194 Tusculum 208–219 Tusculum 235–242 Tusculum 314–316 Tusculum 318–323 Tusculum 362–379 Tusculum 380–392 Tusculum 397–409 Tusculum 426 Tusculum 437–487 Tusculum 500–508 Tusculum 511–517 Tusculum 525–535 Tusculum 546–552 Tusculum 561–569 Tusculum 592–599 Tusculum 634–642 Tusculum 656–660 Tusculum 687–695 Tusculum 720–736 Tusculum 764–768 Tusculum 789–794 Tusculum 804–806 Tusculum 832–836 Fidenae 10 Fidenae 21 Fidenae 48 Fidenae 73 Fidenae 86 Fidenae 91 Fidenae 132 Fidenae 141 Fidenae 144 Fidenae 146a Fidenae 157a Fidenae 160a Fidenae 163 Fidenae 172a–b Fidenae 186 Fidenae 206 Fidenae 212a Fidenae 212b Ficulea R Ficulea 7b Ficulea 73 Ficulea 86 Ficulea 155a Ficulea 166b Ficulea 173a Ficulea 205a Ficulea 228b 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 cistern cistern well; 2 cisterns (8 & 46 m²); bath lead pipe impluvium; aqueduct; 7 cisterns (55, 55, 60, 310 & 340 m²); masonry channel; clay pipe; lead pipe; basin; nymphaeum; bath aqueduct; 3 cisterns (79 m²); clay pipe; lead pipe; basin; bath cuniculi well; cistern; nymphaeum cistern; bath lead pipe; basin 2 cisterns (6 & 24 m²); bath? impluvium; lead pipe; basin impluvium; aqueduct; cistern (775 m²); masonry channel; clay pipe; lead pipe; basin; nymphaeum; bath cistern (76 m²); bath cistern (a cunicoli); masonry channel; clay pipe?; lead pipe; nymphaeum cistern?; bath impluvium; basin aqueduct; clay pipe; bath? impluvium? clay pipe; bath? impluvium?; bath well; bath cistern; cuniculi lead pipe cistern? masonry channel; clay pipe 2 cisterns (507 m²); cuniculi; masonry channel; lead pipe lead pipe; bath 3 cisterns (7, 60 & 187 m²) cuniculi 4 cisterns (84 & 900 m²); cuniculi 2 cisterns (533 & 1245 m²); cuniculi spring; cistern; cuniculi; lead pipe; basin; bath cistern (157 m²); cuniculi; basin; bath 3 cisterns (115 & 450 m², a cunicoli?); bath? lead pipe; basin 2 cisterns (27 & 697 m²); cuniculi; basin; bath 2 cisterns (233 m²); cuniculi aqueduct; cistern; cuniculi; lead pipe; bath 2 cisterns (214 & 682 m²) 2 cisterns; cuniculi; masonry channel; clay pipe; lead pipe; nymphaeum; bath? 2 cisterns 2 cisterns (67 m²); masonry channel; clay pipe?; nymphaeum 2 cisterns (27 m², a cunicoli?); cuniculi; lead pipe cistern; cuniculi cistern (207 m²); masonry channel cistern (414 m²); masonry channel 2 cisterns (48 m²); basin; nymphaeum; bath 2 cisterns (a cunicoli?); masonry channel; clay pipe; nymphaeum 2 cisterns?; basin masonry channel clay pipe; bath? impluvium; 4 cisterns (10, 120 & 126 m²); masonry channel; lead pipe; basin; nymphaeum; bath 2 cisterns (63 m²); cuniculi; bath impluvium; 2 cisterns (145 m², a cunicoli?); basin clay pipe? well well cistern (a cunicoli) bath? lead pipe cistern? (a cunicoli?); clay pipe masonry channel cistern (100 m²); bath? well; cistern cistern (a cunicoli) cuniculi well; cistern? (a cunicoli?) cistern (> 1m²?) cuniculi masonry channel aqueduct cistern (99 m², a cunicoli?) bath? impluvium; 2 cisterns (10 m², a cunicoli?); cuniculi?; masonry channel; basin clay pipe well; cuniculi cistern; basin masonry channel well; cuniculi cistern well?; masonry channel?; bath? 225 appEndiX iv Site Ficulea 228c Ficulea 228f Ficulea 315b Ficulea 316a Ficulea 377 Ficulea 378b Ficulea 389 Ficulea 404a Ficulea 408a Ficulea 423 Ficulea 466a Ficulea 488b Ficulea 501a Ficulea 504a Ficulea 525 Ficulea 527 Ficulea 531 Ficulea 534d Ficulea 537a Ficulea 552 Ficulea 557b Ficulea 570a = Tibur III 242a Ficulea 580 = Tibur III 244 Ficulea 599 = Collatia 58 Ficulea 601 = Collatia 30 Ficulea U1 Tibur III 12a Tibur III 31a Tibur III 74 Tibur III 90 Tibur III 112 Tibur III 123 Tibur III 139 Tibur III 143 Tibur III 149a Tibur III 153 Tibur III 211a Tibur III 218 Tibur III 229a Tibur III 257 Tibur III 290a Tibur III 292 Tibur III 305a Tibur III 319 Tibur III 321 Tibur III 357a Tibur III 381a Tibur III 391 Tibur III 404 Tibur III 408 Tibur III 413 Tibur III U1 Tibur IV 19a Tibur IV 23 Tibur IV 36 Tibur IV 45a Tibur IV 65 Tibur IV 69 Tibur IV 76a Tibur IV 77 Tibur IV 82b Tibur IV 85 Tibur IV 101a Tibur IV 137a Tibur IV 145 Tibur IV 164 Tibur IV 167 Tibur IV 168 Tibur IV 196 Tibur IV 200 Tibur IV 214 Tibur I 137 Tibur I 197 = Tibur II 4 Collatia 4c Collatia 5b Collatia 93 Collatia 95a Collatia 116 Collatia 129 Collatia 136 Collatia 141a Collatia 155 226 Class Water Installations 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 well?; masonry channel?; bath? well?; cistern; masonry channel; bath? cistern (41 m²) cistern; cuniculi cistern lead pipe cuniculi cistern; cuniculi cistern lead pipe well well?; cistern? (a cunicoli?); cuniculi well; cistern (a cunicoli); cuniculi cistern 2 cisterns (2 m², a cunicoli) cistern?; cuniculi bath cistern cuniculi clay pipe well; cistern? cistern (a cunicoli?); cuniculi cistern; cuniculi cistern well; cistern (109 m²); lead pipe; bath? well; cistern? (a cunicoli?); cuniculi; basin cistern cistern (35 m²) spring; cistern? (11 m²) aqueduct lead pipe cistern (51 m²); lead pipe? lead pipe; bath? spring cistern (54 m²) cistern (19 m²) cistern bath? cistern; masonry channel cistern? cistern cistern; cuniculi aqueduct; cistern; cuniculi; lead pipe; basin spring basin cistern; cuniculi cistern (71 m²) well cistern (18 m²) aqueduct; lead pipe cuniculi; masonry channel cistern? (22 m²); cuniculi; bath cistern? cistern (20 m²); masonry channel cistern (44 m²); cuniculi cistern (113 m²); lead pipe? cistern; masonry channel masonry channel cistern cistern? (a cunicoli?) cistern (95 m²) cistern 2 cisterns (2 & 13 m²) cistern? (a cunicoli?) cistern; bath? spring? cistern? (a cunicoli?); cuniculi cistern (36 m²); cuniculi; bath? spring; lead pipe; bath? impluvium; aqueduct?; 2 cisterns (18 & 47 m²); cuniculi; masonry channel; basin spring; 3 cisterns (68 m²) cistern? cistern well basin bath? aqueduct? cistern? (a cunicoli?); masonry channel cistern cuniculi cistern? (a cunicoli?) masonry channel watEr installations Site Collatia 157 Collatia 174a Collatia 180a Collatia 180h1 Collatia 180o1/4 Collatia 186a Collatia 198a Collatia 230a Collatia 256a Collatia 269 Collatia 301 Collatia 305 Collatia 331 Collatia 345 Collatia 379a–b Collatia 380 Collatia 387a–b Collatia 396 Collatia 407 Collatia 409a Collatia 410 Collatia 437 Collatia 456 Collatia 476b Collatia 488a Collatia 507b Collatia 551a Collatia 554 Collatia 556a Collatia 562 Collatia 570c Collatia 581 Collatia 603a Collatia 613b Collatia 614 Collatia 625b Collatia 627a Collatia 629 Collatia 630 Collatia 641a Collatia 650 Collatia 653b Collatia 659a Collatia 660b Collatia 662c Collatia 663a Collatia 663b Collatia 739 Collatia 740 Collatia 748 Collatia 750 Collatia 784 Collatia 788a Collatia 798a Collatia 802 Collatia 809a Collatia 823a Collatia 823b Collatia 835a Collatia 848a Collatia 848b Collatia 851 Collatia 854 Collatia U1 Tellenae 14b Tellenae 32 Tellenae 42 Tellenae 55 Tellenae 80 Tellenae 85 Tellenae 135b Tellenae 171 Tellenae U5 Bovillae 29 Bovillae 41 Bovillae 70 Bovillae 74–75 Bovillae 79 Bovillae 83 Bovillae 145 Bovillae 166 Bovillae 171 Class Water Installations 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 bath? cistern? cuniculi; bath? cistern? (a cunicoli?); clay pipe; lead pipe; bath masonry channel cuniculi well cistern (7 m²) well cistern impluvium; cistern? (22 m², a cunicoli?); basin; bath cistern; lead pipe; basin masonry channel well; 2 cisterns; cuniculi? lead pipe? cistern?; lead pipe? well cistern; masonry channel cistern (16 m²); masonry channel cistern cistern aqueduct; cistern (59 m²) cuniculi cistern cistern cistern (28 m²) cistern (92 m²) cuniculi cuniculi cistern (66 m²) cistern? (a cunicoli?); cuniculi clay pipe cistern (28 m²) bath cuniculi cistern; cuniculi clay pipe cistern? (a cunicoli?); cuniculi; basin cuniculi? bath bath bath well?; cistern; bath? cistern (65 m²); basin cistern; basin aqueduct; cistern; cuniculi impluvium; lead pipe?; bath clay pipe; basin cuniculi cistern (a cunicoli) cistern (10 m², a cunicoli); clay pipe cistern; clay pipe; lead pipe; nymphaeum cistern (45 m²) cistern 2 cisterns cistern cistern; clay pipe? cistern (180 m²); cuniculi; clay pipe; lead pipe 2 cisterns (21 m²) cistern (40 m²) bath? cistern (7 m²) cistern (191 m²) well cistern (200 m²) cistern cistern? (a cunicoli?) clay pipe cuniculi cistern cistern cistern (320 m²); cuniculi bath cuniculi cistern (a cunicoli?) well; cistern clay pipe cistern (100 m²) cistern (a cunicoli) cistern? (a cunicoli?) masonry channel clay pipe; lead pipe? 227 appEndiX iv Site Bovillae 201 Bovillae 220 Bovillae 245 Bovillae 264a Bovillae 296 Bovillae 304.7b Bovillae 313 Bovillae 333b Bovillae 335b Bovillae 356 Bovillae 364a Bovillae 374 Bovillae 383 Bovillae 411 Bovillae 414 Bovillae 416 Bovillae 427 Tusculum 5–8 Tusculum 60–62 Tusculum 79–82 Tusculum 101–105 Tusculum 116–117 Tusculum 130–132 Tusculum 143–148 Tusculum 151 Tusculum 167–171 Tusculum 176 Tusculum 177–178 Tusculum 184 Tusculum 196–201 Tusculum 256–258 Tusculum 260–264 Tusculum 394 Tusculum 414–418 Tusculum 540–545 Tusculum 581–583 Tusculum 603–605 Tusculum 616–622 Tusculum 699–701 Tusculum 743–747 Tusculum 774–775 Tusculum 786–787 Tusculum 808–814 Tusculum 820–821 Tusculum 823–825 Tusculum 903–905 Tusculum 907–908 Tusculum 915–917 Fidenae 61b Fidenae 81a Fidenae 95 Fidenae 99a Fidenae 101b Fidenae 113b Fidenae 120 Fidenae 127 Fidenae 137 Fidenae 151b Fidenae 156a Fidenae 174 Fidenae 183b Fidenae 189 Fidenae 202 Fidenae 221 Fidenae 235a Fidenae 236 Fidenae 243 Fidenae 245b Ficulea E.3 Ficulea 70 Ficulea 120 Ficulea 132 Ficulea 228e Ficulea 294 = Tibur III 175a Ficulea 367 = Tibur III 17a Ficulea 412 Ficulea 464 Ficulea 467a Ficulea 515a Ficulea 523 Ficulea 550 Ficulea 553a 228 Class Water Installations 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 cistern; lead pipe 2 cisterns (a cunicoli); clay pipe cistern? bath cuniculi cistern? impluvium? cistern 2 cisterns (56 m²) cistern (21 m²) basin? clay pipe cuniculi nymphaeum cistern? (70 m²); basin cistern (325 m²) clay pipe; lead pipe 2 cisterns (42 & 136 m²) 2 cisterns (a cunicoli?); lead pipe cistern cistern (821 m²); lead pipe 2 cisterns (54 m²) 2 cisterns (a cunicoli?) 2 cisterns (30 m²); masonry channel; basin 2 cisterns (29 & 218 m²) 2 cisterns (104 & 183 m²); cuniculi cistern (322 m²) cistern cistern cuniculi bath? cistern (56 m²) cistern (384 m²) 4 cisterns (300 & 636 m²); lead pipe 2 cisterns (a cunicoli?); cuniculi; masonry channel cistern (12 m²); masonry channel cistern (238 m²) cistern cistern? cistern? (360 m²) 2 cisterns (857 m²) cistern (600 m²); bath 2 cisterns (243 & 299 m²) cistern (84 m²) cistern (788 m²) lead pipe cistern cistern masonry channel 2 cisterns? (a cunicoli?) cuniculi; bath? cistern well?; cistern (a cunicoli); masonry channel; basin aqueduct cistern cistern? (a cunicoli?) well masonry channel well 3 cisterns (120 & 120 m²?, a cunicoli?) cistern cuniculi well cuniculi cuniculi lead pipe? well; cistern; masonry channel; basin cistern (a cunicoli) clay pipe; bath well cistern (20 m², site 121a) cuniculi well?; masonry channel?; bath? well; cuniculi 2 cisterns (28 & 106 m²) cistern? cuniculi well; cuniculi cuniculi cistern cistern (a cunicoli) 2 cisterns (232 m², a cunicoli?) watEr installations Site Ficulea 600 Ficulea U2 Tibur III 122b Tibur III 311 Tibur III 312 Tibur III 437 Tibur IV 107 Tibur IV 129 Tibur IV 166 Tibur IV 212 Tibur IV 217 Tibur II 7 Tibur II 86 Tibur II 220 Collatia 22a Collatia 55c Collatia 182 Collatia 192b Collatia 205a Collatia 214 Collatia 273a Collatia 317 Collatia 366a Collatia 367 Collatia 382b Collatia 390 Collatia 420b Collatia 448c Collatia 449b Collatia 455 Collatia 459d Collatia 474 Collatia 480b Collatia 523 Collatia 537 Collatia 564a Collatia 577a Collatia 589a Collatia 608 Collatia 610 Collatia 638 Collatia 667c–d Collatia 682I Collatia 698 Collatia 702 Collatia 707b Collatia 708 Collatia 710 Collatia 732a Collatia 735a Collatia 741 Collatia 744a Collatia 758a Collatia 761b Collatia 768a Collatia 775a Collatia 783 Collatia 811a Collatia 813 Collatia 819a Collatia 821 Collatia 822 Collatia 833 Collatia 841a Tellenae 4b Tellenae 21b Tellenae 70b Tellenae 83 Tellenae 90 Tellenae 109b Tellenae 129b Tellenae 137 Tellenae 145b Tellenae 169a Tellenae U4 Bovillae 4 Bovillae 5 Bovillae 45 Bovillae 80 Bovillae 120a Bovillae 125 Bovillae 132 Class Water Installations 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 cuniculi cistern cistern (95 m²) cistern (95 m²) cistern (118 m²) cistern cistern? well; basin; nymphaeum? cistern cistern cistern?; cuniculi aqueduct?; masonry channel?; nymphaeum? cistern (227 m²) cistern (500 m²) cistern (a cunicoli?) well clay pipe cuniculi well cistern? (a cunicoli?) well well well clay pipe clay pipe spring cuniculi aqueduct; cistern (400 m²) cistern; bath? cuniculi well?; cuniculi? cistern? (a cunicoli?) clay pipe masonry channel cuniculi cistern (29 m²) cuniculi masonry channel; clay pipe masonry channel; clay pipe cistern (a cunicoli) cuniculi cistern? (a cunicoli?); clay pipe cistern (64 m²) cuniculi?; masonry channel well well clay pipe masonry channel clay pipe cistern (81 m²) cuniculi well clay pipe clay pipe clay pipe masonry channel cistern (238 m²) cistern; masonry channel; clay pipe well cuniculi cistern cistern spring; cistern (31 m²); cuniculi well cistern (33 m²) cistern (173 m²) cistern (74 m²) cuniculi cistern (21 m²) cistern (12 m²) cistern (7 m²) cistern (256 m²) cistern (77 m²) clay pipe cistern nymphaeum nymphaeum well cistern cistern; cuniculi cistern cistern 229 appEndiX iv Site Class Water Installations Bovillae 146 Bovillae 155 Bovillae 164 Bovillae 183 Bovillae 194 Bovillae 248b Bovillae 289 Bovillae 295a Bovillae 304.16b Bovillae 304.20 Bovillae 338 Bovillae 351 Bovillae 379a Bovillae 380a Bovillae 412 Bovillae 430 Tusculum 25 Tusculum 58–59 Tusculum 68 Tusculum 69–70 Tusculum 71 Tusculum 106–108 Tusculum 114 Tusculum 189 Tusculum 246–250 Tusculum 643 Tusculum 750–751 Tusculum 828–829 Tusculum 910 Tusculum 914 Fidenae 75a–b Ficulea 101 Ficulea 118 Collatia 307 Collatia 364a Fidenae 11b Fidenae 16a Fidenae 68 Fidenae 74ac Fidenae 154b Fidenae 161 Fidenae 177 Fidenae 180 Fidenae 185a Fidenae 210a Fidenae 212c Fidenae 212d Fidenae 222a Fidenae 261 Fidenae 271a Fidenae 275 Fidenae U4 Ficulea B Ficulea F.1 Ficulea F.2 Ficulea O.1 Ficulea 20b Ficulea 87 Ficulea 140d Ficulea 146 d Ficulea 187b Ficulea 219ac Ficulea 223b Ficulea 250 Ficulea 256 Ficulea 339 Ficulea 344 = Tibur III 183b Ficulea 352a = Tibur III 20a Ficulea 358b Ficulea 413 Ficulea 415c.4 Ficulea 444 Ficulea 455c Ficulea 484b Ficulea 492d Ficulea 500c Ficulea 519c Ficulea 521b Ficulea 528b Ficulea 528c Ficulea 533a Ficulea 539h 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other 230 cistern? (a cunicoli?) cistern aqueduct aqueduct cistern cistern? cistern (77 m²) clay pipe well? lead pipe aqueduct; cistern (27 m²); clay pipe cistern (64 m²) cistern cistern cistern 2 cisterns (133 m²) cistern (36 m²) cistern? cistern cistern cistern cistern (159 m²); cuniculi cuniculi cistern? 2 cisterns (35 m²) cistern? cistern cistern (155 m²) cistern cistern (120 m²) spring spring spring cistern (a cunicoli) cuniculi well masonry channel well?; cuniculi well spring; nymphaeum? spring; cuniculi; basin spring spring cuniculi; basin well; cuniculi? cistern (9 m²) cistern (a cunicoli) cuniculi spring? spring lead pipe (stray) cistern (a cunicoli) spring well cistern? (376 m²) spring spring lead pipe (stray) lead pipe basin spring spring; lead pipe? cistern (a cunicoli?) cuniculi spring cuniculi cuniculi cuniculi spring spring; cuniculi lead pipe; masonry channel; nymphaeum well spring; well; masonry channel; basin cistern (24 m²) cistern (a cunicoli?) cistern (a cunicoli?) water installation? cuniculi bath? cuniculi well; basin cistern (45 m²); basin watEr installations Site Class Water Installations Ficulea 554 Ficulea 557a Tibur III 27c Tibur III 34 Tibur III 39a Tibur III 59 Tibur III 60b Tibur III 63a Tibur III 82 Tibur III 99 Tibur III 126a Tibur III 156 Tibur III 323a Tibur III 339 Tibur III 340 Tibur III 355b Tibur III 379 Tibur III 457 Tibur IV 3b Tibur IV 25–26 Tibur IV 35 Tibur IV 51a Tibur IV 55b Tibur IV 71 Tibur IV 79 Tibur IV 80a Tibur IV 84a Tibur IV 92a Tibur IV 136 Tibur IV 144b Tibur IV 152a Tibur IV 152b Tibur IV 154 Tibur IV 163 Tibur IV 210b Tibur I 15 Tibur I 17 Tibur I 31 Tibur I 81 Tibur I 103 Tibur I 115 Tibur I 127 Tibur I 138 Tibur I 143 Tibur I 144 Tibur I 145 Tibur I 148 Tibur I 156 Tibur I 169 Tibur I 171 Tibur I 183 Tibur I 213 Tibur I 144 Tibur I 148 Tibur II 2 Tibur II 82–90 Tibur II 209 Tibur II 210a Collatia 22b Collatia 43d Collatia 44b Collatia 44c Collatia 45a Collatia 71g Collatia 80f = U2 Collatia 81 Collatia 97d Collatia 100 Collatia 107a Collatia 148a Collatia 177U.b Collatia 187c Collatia 194h Collatia 194m Collatia 194n Collatia 194o Collatia 196a Collatia 200c Collatia 219a Collatia 220a Collatia 224a Collatia 224b Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other basin cuniculi; basin water installation? cuniculi?; masonry channel? cuniculi well head (stray) cuniculi spring?; cuniculi spring; cuniculi spring? cuniculi basin spring spring lead pipe; bath lead pipe (stray) spring; masonry channel?; basin lead pipe (stray) cistern? (a cunicoli?) spring; aqueduct; basin? spring aqueduct; cistern?; basin? masonry channel cistern?; masonry channel; basin? cistern?; cuniculi cistern? spring; aqueduct cistern (24 m²) cistern? (13 m²) cistern? spring; aqueduct; cuniculi cistern (30 m²) basin cuniculi water installation? cistern (a cunicoli?) cistern cuniculi masonry channel cistern cistern well; cuniculi lead pipe (stray) cuniculi? cistern?; cuniculi cuniculi cistern; cuniculi masonry channel; basin; nymphaeum lead pipe; bath cistern well cistern cistern? cistern cuniculi lead pipe (stray) cistern? cistern cuniculi well well well spring; aqueduct cuniculi cistern spring; cistern; bath well lead pipe cistern (239 m²) spring; aqueduct cistern spring; aqueduct cuniculi cuniculi cuniculi? cuniculi? spring; cuniculi; nymphaeum spring; cuniculi; nymphaeum spring; cuniculi spring; cuniculi spring; masonry channel?; nymphaeum well; bath 231 appEndiX iv Site Class Water Installations Collatia 224e Collatia 224n Collatia 224o Collatia 224q Collatia 224r Collatia 296c Collatia 296e Collatia 327d Collatia 332d Collatia 349b Collatia 591g Collatia 624b Collatia 625d Collatia 628b Collatia 657b Collatia 658b Collatia 662d Collatia 682u Collatia 726b Collatia 757b Collatia 759 Collatia 804a Collatia 807a Collatia 820a Collatia 849 Tellenae 12 Tellenae 37a Tellenae 44 Tellenae 113a Tellenae 113b Tellenae 1146 Bovillae 1 Bovillae 16 Bovillae 36 Bovillae 55 Bovillae 67 Bovillae 72b Bovillae 73 Bovillae 85 Bovillae 89 Bovillae 91 Bovillae 111 Bovillae 115 Bovillae 117 Bovillae 121 Bovillae 130 Bovillae 143 Bovillae 144a Bovillae 150 Bovillae 168a Bovillae 168b Bovillae 170a–c Bovillae 185 Bovillae 291 Bovillae 300 Bovillae 304.3 Bovillae 304.9 Bovillae 304.13 Bovillae 304.15 Bovillae 304.16d–e Bovillae 304.18 Bovillae 304.24 Bovillae 342 Bovillae 359 Bovillae 384 Bovillae U3 Tusculum 9 Tusculum 12 Tusculum 19–20 Tusculum 26ab Tusculum 27 Tusculum 67 Tusculum 89–90 Tusculum 109/112/127 Tusculum 118–119 Tusculum 125 Tusculum 137 Tusculum 155 Tusculum 159 Tusculum 166 Tusculum 195 Tusculum 206 Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other 232 cistern cistern (a cunicoli?) cistern (a cunicoli?) cuniculi cistern?; cuniculi? cuniculi? cuniculi cuniculi cuniculi lead pipe (stray) cistern (a cunicoli?) spring lead pipe; basin clay pipe (stray) cuniculi; basin cuniculi; clay pipe clay pipe (stray) cistern? aqueduct?; cuniculi aqueduct; cuniculi spring cuniculi spring cuniculi aqueduct; lead pipe cistern (73 m²) water installation? clay pipe (stray) cistern (28 m²) cistern cistern? cistern (146 m²) cuniculi cistern? cistern (14 m²) cistern? (a cunicoli?) cistern (a cunicoli?) cistern cistern basin; bath? cistern; cuniculi lead pipe cistern? cuniculi lead pipe cuniculi cuniculi cuniculi cuniculi cuniculi spring; aqueduct; basin water installation? spring; aqueduct cistern (130 m²) cistern cistern (101 m²) spring; aqueduct cistern? cistern? well?; aqueduct well; cuniculi; basin; nymphaeum cistern (15 m²); clay pipe cistern cistern spring?; well lead pipe cistern? cistern cistern cistern; lead pipe cistern cuniculi cistern (8 m²); clay pipe spring; cistern (900 m²); cuniculi cistern (35 m²); lead pipe lead pipe (stray) masonry channel? basin cistern (692 m²); masonry channel; lead pipe? cistern (50 m²); lead pipe cuniculi cuniculi watEr installations Site Class Water Installations Tusculum 226 Tusculum 231 Tusculum 232 Tusculum 244 Tusculum 251 Tusculum 252 Tusculum 254 Tusculum 255 Tusculum 317 Tusculum 326 Tusculum 333 Tusculum 334 Tusculum 359 Tusculum 427 Tusculum 431 Tusculum 432 Tusculum 518 Tusculum 519 Tusculum 522 Tusculum 539 Tusculum 554 Tusculum 556 Tusculum 558 Tusculum 579 Tusculum 584 Tusculum 625 Tusculum 626 Tusculum 680 Tusculum 681 Tusculum 739 Tusculum 749 Tusculum 777 Tusculum 783 Tusculum 784 Tusculum 785 Tusculum 788 Tusculum 796 Tusculum 797 Tusculum 799 Tusculum 817 Tusculum 831 Tusculum 876 Tusculum 878 Tusculum 898 Tusculum 891 Tusculum 909 Tusculum 913 Tusculum 918 Tusculum 922 Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other Other cuniculi cuniculi masonry channel? cuniculi cistern (100 m²) cuniculi cuniculi cuniculi cuniculi basin? cistern? masonry channel? cistern cistern? cuniculi cuniculi cistern lead pipe cistern cuniculi lead pipe (stray); clay pipe (stray) cistern (594 m²) cistern (90 m²) cuniculi cuniculi clay pipe cistern (34 m²) nymphaeum? cistern (541 m²) cistern? (87 m²) aqueduct?; cuniculi cistern lead pipe cuniculi lead pipe cuniculi cistern cistern cistern? cuniculi cistern cistern (53 m²) cuniculi spring; nymphaeum? cistern (465 m²) cistern masonry channel; clay pipe cistern cistern? 233 bibliography abbrEviations Names of Latin authors and their works abbreviated according to Thesaurus Linguae Latinae; Greek authors and their works according to Oxford Classical Dictionary. AE = L’année épigraphique. CIL = Corpus inscriptionum Latinarum. EE = Ephemeris epigraphica: corporis inscriptionum latinarum supplementum. I. It. = Inscriptiones Italiae. LTURS = Fiocchi Nicolai et al. 2001; 2004; 2005; 2006; 2008. OLD = Oxford Latin Dictionary. Ed. P. G. W. Glare. Oxford: Clarendon Press, 1996. RE = Real-Encyclopädie der classischen Alterthums-wissenschaft. Adams, G. W. 2006. The Suburban Villas of Campania and their Social Function. BAR International Series S1542. Adams, G. W. 2008. Rome and the Social Role of Élite Villas in its Suburbs. BAR International Series S1760. Agusta-Boularot, S. 1998. Banlieue et faubourgs de Rome: approche linguistique et déinition spatiale. In Suburbia. Les faubourgs en Gaule Romaine et dans les regions voisines, ed. R. Bedon, 35–62. Caesarodunum XXXII. Aldrete, G. S. 2007. Floods of the Tiber in Ancient Rome. Baltimore: Johns Hopkins University Press. Alessio, M., L. Allegri, F. Bella, G. Calderoni, C. Cortesi, G. Dai Pra, D. De Rita, D. Esu, M. Follieri, S. Improta, D. Magri, B. Narcisi, V. Pedrone, and L. Sadori. 1986. 14C Dating, Geochemical Features, Faunistic and Pollen Analyses of the Uppermost 10 M Core from Valle di Castiglione (Rome, Italy). Geologia Romana XXV, 287–308. Allison, P. M. 2004. Pompeian Households: An Analysis of the Material Culture. Cotsen Institute of Archaeology Monograph 42. Amanti, M., M. Bencivenga, F. Bramerini, P. Cara, A. Caserta, S. Castenetto, R. Colozza, S. Coppari, A. Corazza, R. Crescenzi, E. Di Loreto, M. Dolce, C. Faccenna, R. Funiciello, G. Gisotti, E. Guidoboni, L. Liperi, L. Lombardi, L. Malagnini, F. Marra, D. Molin, M. Moscato, B. Narcisi, A. Paciello, M. Pecci, M. Piro, M. Prosperi, M. Rebuffat, F. Riguzzi, C. Rosa, A. Rossi, A. Rovelli, F. Sabetta, S. Salvi, A. Tertulliani, G. Traina, and R. Vallesi, eds. 1995. La geologia di Roma: il centro storico. Memorie descrittive della Carta Geologica d’Italia L. Ammannato, F., and B. Belelli Marchesini. 1988. Via Nomentana/Via Salaria. Castel Giubileo (circ. IV). Bullettino della Commissione Archeologica Comunale di Roma, 465–467. Amoroso, A. 2000. Crustumerium, da città arcaica a suburbium di Roma. Bullettino della Commissione Archeologica Comunale di Roma, 263–282. Amouretti, M.-C., and J.-P. Brun, eds. 1993. La production du vin et de l’huile en Méditerranée. Oil and Wine Production in the Mediterranean Area. Actes du symposium international organisé par le Centre Camille Jullian et le Centre archéologique du Var, Aix-en-Provence et Toulon, 20–22 Novembre 1991. Bulletin de correspondance hellénique Supplément XXVI. Andermahr, A. M. 1998. Totus in praediis. Senatorischer Grundbesitz in Italien in der Frühen und Hohen Kaiserzeit. Antiquitas 3:37. Anschuetz, K. F., R. H. Wilshusen, and C. L. Scheick. 2001. An Archaeology of Landscapes: Perspectives and Directions. Journal of Archaeological Research 9:2, 157–211. Appleton, J. 1975. The Experience of Landscape. Chichester: J. Wiley. Appleton, J. 1990. The Symbolism of Habitat. An Interpretation of Landscape in the Arts. Seattle: University of Washington Press. Arnoldus-Huyzendveld, A. 1995. Via Maresciallo Pilsudski – Via P. De Coubertin (circ. II). Prime Considerazioni paleoambientali. Bullettino della Commissione Archaeologica Comunale di Roma, 280–281. Arnoldus-Huyzendveld, A. 2003. I suoli di Roma: due passi sulle terre della città. Carta dei suoli del Comune di Roma in scala 1:50,000. Roma: Comune di Roma, Dipartimento X – U. O. Sviluppo Sostenibile. Arnoldus-Huyzendveld, A. 2004. Lo sfruttamento agricolo e le costruzioni di età repubblicana. In Centocelle I. Roma S.D.O. Le indagini archeologiche, eds. P. Gioia and R. Volpe, 462. Soveria Mannelli: Rubbettino. Arnoldus-Huyzendveld, A. 2008. Valorizzazione ambientale dei suoli non-urbanizzati. In La geologia di Roma dal centro storico alla periferia II, eds. R. Funiciello, A. Praturlon, and G. Giordano, 31–50. Memorie descritteve della Carta Geologia d’Italia LXXX. bibliography Arnoldus-Huyzendveld, A., A. Corazza, D. De Rita, and F. Zarlenga. 1997. Il paesaggio geologico ed i geotipi della Campagna Romana. Quaderni dell’Ambiente 5. romano. Città, agricoltura, commercio: materiali di Roma e del Suburbio, eds. R. Bussi and V. Vandelli, 122–129. Modena: Panini. Arthur, P. 1991. Romans in Northern Campania. Settlement and Land-use around the Massico and the Garigliano Basin. Archaeological Monographs of the British School at Rome 1. Belvedere, O. 1994. La ricognizione sul terreno. Journal of Ancient Topography IV, 69–84. Ashby, T. 1902. Classical Topography of the Roman Campagna. Papers of the British School at Rome I, 125–281. Ashby, T. 1906. Classical Topography of the Roman Campagna. Papers of the British School at Rome III, 1–212. Ashby, T. 1907. Classical Topography of the Roman Campagna. (The Via Latina). Section I. Papers of the British School at Rome IV, 1–159. Berg, B. 1997. Cicero’s Palatine Home and Clodius’ Shrine of Liberty: Alternative Emblems of the Republic in Cicero’s De domo sua. Studies in Latin Literature and Roman History 8, 122–143. Collection Latomus 239. Berg, R. 2010: Il mundus muliebris nelle fonti latine e nei contesti Pompeiani. PhD diss., University of Helsinki. Ashby, T. 1910. Classical Topography of the Roman Campagna. (The Via Latina) Section II. Papers of the British School at Rome V, 213–432. Bergmann, B. 2002. Art and Nature in the Villa at Oplontis. In Pompeian Brothels, Pompeii’s Ancient History, Mirrors and Mysteries, Art and Nature at Oplontis, & The Herculaneum “Basilica,” 87–120. Journal of Roman Archaeology Supplement 47. Ashby, T. 1914. La Campagna Romana al tempo di Paolo III. Mappa della Campagna Romana del 1547 di Eufrosino della Volpaia. Roma: Danesi. Bianco, A. D. 2007. Aqua ducta, aqua distributa. La gestione delle risorse Idriche in età romana. Roma: Silvio Zamorani Editore. Ashby, T. 1927. The Roman Campagna in Classical Times. London: E. Benn. Blake, M. E. 1947. Ancient Roman Construction in Italy from the Prehistoric Period to Augustus: A Chronological Study Based in Part upon the Material Accumulated by the Late Dr. Esther Boise van Deman. Washington D.C.: Carnegie Institution. Ashby, T. 1935. The Aqueducts of Ancient Rome. Oxford: Clarendon Press. Bannon, C. J. 2001. Servitudes for Water Use in the Roman Suburbium. Historia 50, 34–52. Bloch, H. 1958: Sette Bassi Revisited. Harvard Studies in Classical Philology LXIII, 401–414. Ashmore, W., and A. B. Knapp, eds. 1999. Archaeologies of Landscapes. Contemporary Perspectives. Oxford: Blackwell. Boanelli, F. 1992. La villa romana di San Martino fra tardoantico e basso-medioevo. Atti e Memorie della Società Tiburtina di Storia e d’Arte LXV, 37–83. Barbanera, M. 1998. Archeologia degli italiani. Storia, metodi e orientamenti dell’archeologia classica in Italia. Roma: Editori Riuniti. Bodel, J. 1997. Monumental Villas and Villa Monuments. Journal of Roman Archaeology 10, 5–35. Barbina, P. 1998. Tenuta Boccone Borghese. Bullettino della Commissione Archeologica Comunale di Roma, 313–314. Barker, G. 1981. Landscape and Society, Prehistoric Central Italy. London: Academic Press. Bonetto, J. 1999. Ercole e le vie della transumanza. Il santuario di Tivoli. Ostraka 81, 291–307. Boni, C., P. Bono, and G. Capelli. 1987. Hydrogeological Scheme of Central Italy: Map and Tables. Barker, G. 1995. A Mediterranean Valley: Landscape Archaeology and Annales History in the Biferno Valley. London: Leicester University Press. Boni, C., P. Bono, S. Lombardi, L. Mastrorillo, and C. Percopo. 1995. Hydrogeology, Fluid Geochemistry and Thermalism. In The Volcano of the Alban Hills, ed. R. Trigila, 221–242. Roma: Tipograia. Barker, G., J. Lloyd, and D. P. Webley. 1978. A Classical Landscape in Molise. Papers of the British School at Rome XLVI, 35–51. Borda, M. 1943. Monumenti archeologici tuscolani nel Castello di Agliè. Roma: La Libreria dello Stato. Barnish, S. J. B. 1987. Pigs, Plebeians and Potentes. Rome’s Economic Hinterland c. AD 350–600. Papers of the British School at Rome LV, 157–185. Becker, J. A. 2005. Investigating Early Villas: The Case of Grottarossa. In Papers in Italian Archaeology VI: Communities and Settlements from the Neolithic to the Early Medieval Period. Proceedings of the 6th Conference of Italian Archaeology held at the University of Groningen, Groningen Institute of Archaeology, The Netherlands, April 15–17, 2003, eds. P. Attema, A. Nijboer, and A. Zifferero, 813–821. BAR International Series S1452. Becker, J. A. 2006. The Villa delle Grotte at Grottarossa and the Prehistory of Roman Villas. Journal of Roman Archaeology 19, 213–220. Bedini, A. 1979. Abitato protostorico in località Acqua Acetosa – Laurentina. Quaderni del Centro di studio per l’archeologia etrusco-italica 3. Archeologia Laziale II, 21–28. Bedini, A. 1984. Scavi al Torrino. Quaderni del Centro di studio per l’archeologia etrusco-italica 8. Archeologia Laziale VI, 84–90. Bedini, A. 1997. Modi di insediamento e boniica agraria nel suburbio di Roma. In Uomo, acqua e paesaggio. Atti dell’Incontro di studio sul tema: Irregimentazione delle acque e trasformazione del paesaggio antico (S. Maria Capua Vetere, 22–23 novembre 1996), ed. S. Quilici Gigli, 165–184. Atlante tematico di topograia antica Supplementum II. Bellini, G. R. 1985. Gli impianti dei settori Nord, Ovest, Sud. In Misurare la terra: centuriazione e coloni nel mondo Borda, M. 1958. Tuscolo. Itinerari dei musei, gallerie e monumenti d’Italia 98. Bouma, J. W. 1996. Religio votiva. The Archaeology of Latial Votive Religion. The 5th–3rd c. BC Votive Deposit South West of the Main Temple at “Satricum” Borgo Le Ferriere. Part III Corpus of Latial Cult Places: Setting, Nature, Contents and Chronology. Groningen: University of Groningen. Bradley, R. S., M. K. Hughes, and H. F. Diaz. 2003. Climate in Medieval Time. Science 302, 404–405. Brown, A. G. 1997. Alluvial Geoarchaeology: Floodplain Archaeology and Environmental Change. Cambridge, New York: Cambridge University Press. Brown, A. G, and C. Ellis. 1995. People, Climate and Alluviation: Theory, Research Design and New Sedimentological and Stratigraphic Data from Etruria. Papers of the British School at Rome LXIII, 45–73. Bruto, M. L., and L. M. Vigna. 2006. Via Nomentana. Sant’Alessandro (Municipio V). Villa Romana. In Roma. Memorie dal sottosuolo. Ritrovamenti archeologici 1980/2006, ed. M. A. Tomei, 257–260. Milano: Electa. Bruun, C. 1991. The Water Supply of Ancient Rome. A Study of Roman Imperial Administration. Commentationes Humanarum Litterarum 93. Bruun, C. 2000. Senatorial Owners of What? (Review of Andermahr 1998.) Journal of Roman Archaeology 13, 498–506. Bruun, C. 2003. Le istule acquarie e i proprietari terrieri nel suburbium di Roma. In Suburbium: Il suburbio di Roma dalla 235 bibliography crisi del sistema delle ville a Gregorio Magno, eds. P. Pergola, R. Santangeli Valenzani, and R. Volpe, 485–501. Collection de l’École française de Rome 311. Bruun, C., ed. 2005. Interpretare i bolli laterizi di Roma e della valle del Tevere: produzione, storia economica e topograica. Acta Instituti Romani Finlandiae 32. Buck, R. J., and A. M. Small. 1980. The Topography of Roman Villas in Basilicata. In Attività archeologica in Basilicata, 1964– 1977: Scritti in onore di Dinu Adamesteanu, ed. M. Padula, 561–570. Matera: Meta. Bulgarini, F. 1848. Notizie storiche antiquarie, statistiche ed agronomiche intorno all’antichissima città di Tivoli e suo territorio. Roma: G. B. Zampi. Buonocore, M., ed. 1997–2002. Appunti di topograia romana nei Codici Lanciani della Biblioteca apostolica vaticana. Vol. I–V. Bussi, R., and V. Vandelli, eds. 1985. Misurare la terra: centuriazione e coloni nel mondo romano. Città, agricoltura, commercio: materiali di Roma e del Suburbio. Modena: Panini. Cabral, S., and F. Del Re. 1779. Delle ville e de’ più notabili Monumenti Antichi della città, e del territorio di Tivoli. Nuove ricerche. Roma: Stamperia del Puccinelli al Governo Vecchio. Caiazza, D., F. Ortolani, and S. Pagliuca. 1999. Modiicazioni ambientali cicliche e rilessi sulle attività antropiche in Campania durante il periodo storico. In Archeologia e ambiente. Atti del convegno internazionale Ferrara Fiere 3–4 aprile 1998, ed. F. Lenzi, 267–273. Documenti 30. Cali, A. 1998. Tenuta della Cesarina, settore meridionale. Bullettino della Commissione Archeologica Comunale di Roma, 309–310. Calci, C. 1998. La via Tiburtina antica dal Fosso della Marranella a Settecamini. In Roma oltre le mura. Lineamenti storico topograici del territorio della V circoscrizione, ed. C. Calci, 51–111. Roma: Associazione Culturale “Roma oltre le mura.” Calci, C., and Z. Mari. 2003. Via Tiburtina. In Suburbium: Il suburbio di Roma dalla crisi del sistema delle ville a Gregorio Magno, eds. P. Pergola, R. Santangeli Valenzani, and R. Volpe, 175–209. Collection de l’École française de Rome 311. Calci, C., and R. Sorella. 1995. Forme di paesaggio agrario nell’ager Ficulensis. Atlante tematico di topograia antica IV, 117–127. Calci, C., and R. Sorella. 1998. Dal Fosso di Pratolungo al Monte dello Spavento. In Roma oltre le mura. Lineamenti storico topograici del territorio della V circoscrizione, ed. C. Calci, 177–202. Roma: Associazione Culturale “Roma oltre le mura.” Calzolari, M. 1994. Toponimi fondiari romani. Una prima raccolta per l’Italia. Annali dell’Università di Ferrara Sezione VI Lettere VII:3. Cambi, F. 2003. Archeologia dei paesaggi antichi: fonti e diagnostica. Le bussole 79. Cambi, F., and N. Terrenato. 1994. Introduzione all’archeologia dei paesaggi. Roma: Nuova Italia Scientiica. Campbell, B. 1996. Shaping the Rural Environment: Surveyors in Ancient Rome. Journal of Roman Studies LXXVI, 22–39. Campbell, B., ed. 2000. The Writings of the Roman Land Surveyors. Introduction, Text, Translation and Commentary. Journal of Roman Studies Monograph 9. Campolmi, B. 1993. Piante selvatiche in cucina. Firenze: Olimpia. Capelli, G., R. Mazza, and S. Taviani. 2008. Carta idrogeologica dell’area di Roma. In La geologia di Roma dal centro storico alla periferia II, eds. R. Funiciello, A. Praturlon, and G. Giordano, CD rom. Memorie descritteve della Carta Geologia d’Italia LXXX. Carafa, P. 2000. Una nuova analisi archeologica per il 236 settore settentrionale del Suburbio di Roma. Bullettino della Commissione Archeologica Comunale di Roma, 185–196. Carafa, P. 2004. Il paesaggio etruco-italico. In Bridging the Tiber. Approaches to Regional Archaeology in the Middle Tiber Valley, ed. H. Patterson, 45–59. Archaeological Monographs of the British School at Rome 13. Carandini, A., ed. 1985a. Setteinestre. Una villa schiavistica nell’Etruria romana. Vol. I–III. Modena: Panini. Carandini, A. 1985b. Hortensia – Orti e frutteti intorno a Roma. In Misurare la terra: centuriazione e coloni nel mondo romano. Città, agricoltura, commercio: materiali di Roma e del Suburbio, eds. R. Bussi and V. Vandelli, 66–75. Modena: Panini. Carandini, A. 1988. Schiavi in Italia. Gli strumenti pensanti dei Romani fra tarda Repubblica e medio Impero. Studi NIS Archeologia 8. Carandini, A., M. T. D’Alessio, and H. Di Giuseppe, eds. 2006. La fattoria e la villa dell’Auditorium nel quartiere Flaminio di Roma. Roma: “L’Erma” di Bretschneider. Carandini, A., G. Ricci, M. T. D’Alessio, C. De Davide, and N. Terrenato. 1997. La villa dell’Auditorium dall’età arcaica all’età imperiale. Mitteilungen des Deutschen Archäologischen Instituts, Römische Abteilung 104, 117–148. Caravello, G. U., and F. Giacomin. 1993. Landscape Ecology Aspects in a Territory Centuriated in Roman Times. Landscape and Urban Planning 24:1–4, 77–85. Carbonara, A., and G. Messineo. 1992, Via Tiburtina. Settecamini (circ. V). III. Complesso idraulico in via Casal Bianco. Bullettino della Commissione Archeologica Comunale di Roma, 98–100. Carbonara, A., and G. Messineo. 1995. Via Nomentana. Via Nomentana, km 9,700. Località La Cecchina (circ. V). Bullettino della Commissione Archeologica Comunale di Roma, 239–260. Carlsen, J. 1995. Vilici and Roman Estate Managers until AD 284. Analecta Romani Instituti Danici Supplementum 24. Carocci, S. 1988. Tivoli nel basso medioevo. Società cittadina ed economia agraria. Istituto Storico Italiano per il Medio Evo Nuovi Studi Storici 2. Carpaneto, G. M., and M. Cristaldi. 1995. Dormice and Man: A Review of Past and Present Relations. Hystrix 6:1–2: 303–330. Carrington, R. C. 1931. Studies in the Campanian villae rusticae. Journal of Roman Studies XXI, 110–130. Carta storica archeologica monumentale e paesistica del suburbio e dell’agro romano. Comune di Roma, 1981–1988. Caruso, G., P. Gioia, and R. Volpe. 1998. Indagini archeologiche preliminari alla realizzazzione del Sistema Direzionale Orientale. Bullettino della Commissione Archeologica Comunale di Roma, 280–291. Caserta, E. 2006. Albano Laziale (Roma). Loc. S. Maria della Stella. Vita quotidiana e attività produttive. Lazio & Sabina 3, 169–176. Castagnoli, F., A. M. Colini, and G. Macchia, eds. 1972. La via Appia. Roma: Edizioni Scientiiche Italiane – Istituto di Studi Romani. Cecchini, M. G., N. Pagliardi, and L. Petrassi. 1985. Via Ardeatina. Via di Grottaperfetta (circ. XI). Bullettino della Commissione Archeologica Comunale di Roma, 245–247. Cecchini, M. G., N. Pagliardi, and L. Petrassi. 1990. Via Appia. Località Torricola. (circ. XI). Bullettino della Commissione Archeologica Comunale di Roma, 114–121. Celli, A. 1925. Storia della Malaria nell’Agro Romano: opera postuma con illustrazioni del P. Ambrogetti. Città di Castella: Società anonima tip. “Leonardo da Vinci.” Celli, A. 1927. Malaria e colonizzazione nell’Agro Romano dai più antichi tempi ai nostri giorni. Firenze: Vallecchi. Celli, A. 1933. The History of Malaria in the Roman Campagna. London: Bale & Danielsson. bibliography Champlin, E. 1982. The Suburbium of Rome. American Journal of Ancient History 7, 97–117. Circulation in the Rome Area: Surface Observations. Boundary Layer Meteorology 14:2, 133–151. Champlin, E. 1993. Aeternumque tenet per saecula nomen: Property, Place-names and Prosopography. In Prosopographie und Sozialgeschichte. Studien zur Methodik und Erkenntnismöglichkeit der kaiserzeitlichen Prosopographie. Kolloquium Köln 24.–26. November 1991, ed. W. Eck, 51–59. Köln: Böhlau. Conolly, J., and M. Lake. 2006. Geographical Information Systems in Archaeology. Cambridge, New York: Cambridge University Press. Chiofi, L. 1999. Epigraia ed insediamenti: il caso del suburbio di Roma. In La forma della città e del territorio. Esperienze metodologiche e risultati a confronto, Atti dell’incontro di studio – S. Maria Capua Vetere 27–28 novembre 1998, ed. S. Quilici Gigli, 51–60. Atlante tematico di topograia antica, Supplementum V. Chiofi, L. 2003. I nomi dei proprietari dall’analisi epigraica. In Suburbium: Il suburbio di Roma dalla crisi del sistema delle ville a Gregorio Magno, eds. P. Pergola, R. Santangeli Valenzani, and R. Volpe, 437–484. Collection de l’École française de Rome 311. Chiofi, L. 2005. “Sepulchra in extremis i nibus…etiam in mediis possessionibus sepulchra faciunt.” In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B Santillo Frizell and A. Klynne, 125–134. The Swedish Institute in Rome Projects and Seminars 2. Chiofi, L. 2006. Suburbana e sepulchra. Nomi di proprietari nel suburbio di Roma dall’iscrizioni su monumenti funebri. Bullettino della Commissione Archeologica Comunale di Roma, 215–239. Constantini, L., and J. Giorgi. 2001. Charred Plant Remains of the Archaic Period from the Forum and Palatine. Journal of Roman Archaeology 14, 239–248. Cornell, T. 1995. The Beginnings of Rome: Italy and Rome from the Bronze Age to the Punic Wars (c.1000–264 BC). London: Routledge. Corrente, M. 1984a. Via Latina. Cinecittà Est (circ. IX). Bullettino della Commissione Archeologica Comunale di Roma, 87–89. Corrente, M. 1984b. Via Latina. Fosso di Gregna (circ. IX). Bullettino della Commissione Archeologica Comunale di Roma, 89. Corrente, M. 1985. Alcuni esempi di forme economiche nel settore Est del suburbio romano. In Misurare la terra: centuriazione e coloni nel mondo romano. Città, agricoltura, commercio: materiali di Roma e del Suburbio, eds. R. Bussi and V. Vandelli, 112–118. Modena: Panini. Corrente, M. 1988a. Via Latina. Ville al VII miglio (circ. X). Bullettino della Commissione Archeologica Comunale di Roma, 398–401. Corrente, M. 1988b. Via Latina. Località Morena. Tracciato sotterraneo dell’Anio Novus (circ. X). Bullettino della Commissione Archeologica Comunale di Roma, 401–405. Chisholm, M. 1962. Rural Settlement and Land Use. An Essay in Location. London: Hutchinson University Library. Corrente, M., L. Buonamico, and A. Vodret. 1988. Via Latina. Villa dei Sette Bassi (circ. X). Bullettino della Commissione Archeologica Comunale di Roma, 394–398. Chouquer, G., M. Clavel Lévêque, F. Favory, and J.-P. Vallat. 1987. Structures agraires en Italie centro-méridionale. Cadastres et paysages ruraux. Collection de l’École française de Rome 100. Corsi, C. 2000. Le strutture di servizio del Cursus Publicus in Italia. Ricerche topograiche ed Evidenze Archeologice. BAR International Series S875. Chouquer, G., and F. Favory. 1999. Réponse à Lorenzo Quilici à propos des limitations de l’Italie centrale. Analecta Romana Instituti Danici XXVI, 47–55. Claridge, A. 1998. Rome. An Oxford Archaeological Guide. Oxford: Oxford University Press. Clarke, W. 2003. Production and Distribution of Decorative Building Stone. Tiber Valley Workshop 2003. (http://www.bsr. ac.uk/BSR/sub_arch/Attach/Decorative_building_stone.pdf; accessed August 2008.) Clüver, P. 1624. Italia Antiqua. Lugduni Batavorum: Oficina Elseviriana. Coarelli, F. 1986. L’urbs e il suburbium. In Società romana e impero tardoantico. Vol. 1. Istituzioni, ceti, economie, ed. A. Giardina, 1–58, 395–412. Coarelli, F. 1987. I santuari del Lazio in età repubblicana. Studi NIS archeologia 7. Coarelli, F. 1988. Colonizzazione romana e viabilità. Dialoghi di archeologia III 6:2, 35–48. Coarelli, F. 1993. Dintorni di Roma. Guide archeologiche Laterza 7. 2. edizione. Coarelli, F., and F. Pesando, eds. 2006. Rileggere Pompei. Studi della Soprintendenza archeologica di Pompei 12. Coccia, S., and D. J. Mattingly, eds. 1992. Settlement History, Environment and Human Exploitation of an Intermontane Basin in the Central Apennines. The Rieti Survey 1988–1991. Part 1. Papers of the British School at Rome LX, 213–289. Coccia, S., and D. J. Mattingly, eds. 1995. Settlement History, Environment and Human Exploitation of an Intermontane Basin in the Central Apennines. The Rieti Survey 1988–1991. Part 2. Land-use Patterns and Gazetteer. Papers of the British School at Rome LXIII, 105–158. Colacino, M., and L. Dell’osso. 1978. The Local Atmospheric Corti, R. 1991. La tematica dell’otium nelle Silvae di Stazio. In Continuità e trasformazioni fra repubblica e principato. Istituzioni, politica, società. Atti dell’incontro di studi organizzato dall’Università di Bari (Dipartimento di Scienze dell’Antichità), École Française de Rome, in collaborazione con Università di Firenze, Dottorato in storia (Storia politica e culturale dell’antichità classica), Bari, 27–28 gennaio 1989, ed. M. Pani, 189–224. Documenti e studi 8. Cosentino, D., M. Parotto, and A. Praturlon, eds. 1993. Lazio. Guide Geologiche Regionali. Cosgrove, D. E. 1984. Social Formation and Symbolic Landscape. Madison: University of Wisconsin Press, 2nd ed. 1998. Costantini, E. A. C., F. Urbano, and G. L’Abate. 2004. Soil Regions of Italy. (http://www.soilmaps.it/download/csiBrochureSR_a4.pdf; accessed September 2006.) Cotton, M. A. 1979. Una villa ed un grande ediicio romani lungo la via Gabina. Quaderni di archeologia etrusco-italica 3. Archeologia Laziale II, 82–85. Cresswell, T. 2005. Place: A Short Introduction. Malden: Blackwell Publishing. Curti, E., and E. Moscetti. 1996. “Marmi” colorati in alcune ville romane tra le vie Nomentana e Tiburtina. Annali dell’Associazione Nomentana di Storia e Archeologia 2, 23–35. Daicovici, C. 1930. Castrimoenium e la così detta via Castrimeniense. Ephemeris Dacoromana IV, 29–71. Dal vulcano all’uomo. Caratteristiche e impiego della pietra sperone e del peperino di Marino. Quaderni delle Scuderie Aldobrandini 2. 2003. D’Agostino, B. 1991. The Italian Perspective on Theoretical Archaeology. In Archaeological Theory in Europe: The Last Three Decades, ed. I. Hodder, 52–64. London, New York: Routledge. 237 bibliography D’Alessio, M. T., and H. Di Giuseppe. 2005. La villa dell’Auditorium a Roma tra sacro e profano. In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B. Santillo Frizell and A. Klynne, 177–196. The Swedish Institute in Rome Projects and Seminars 2. Projects in Imperial Rome. Journal of Roman Archaeology Supplement 25. D’Arms, J. H. 1970. Romans on the Bay of Naples. A Social and Cultural Study of the Villas and Their Owners from 150 BC to AD 400. Loeb Classical Monograph 846. Dell’Era, F. 1998. Tenuta Radicicoli del Bene. Bullettino della Commissione Archeologica Comunale di Roma, 297–297. D’Arms, J. H. 1984. Ville rustiche e ville di otium. In Pompei 79: raccolta di studi per il decimonono centenario dell’eruzione vesuviana, ed. F. Zevi, 65–86. Napoli: G. Macchiaroli. Darwall-Smith, R. 1994. Albanum and the villas of Domitian. In Les années Domitien. Colloque organisé à l’université de Toulouse-Le Mirail par J.-M. Pailler et R. Sablayrolles les 12, 13, 14 octobre 1992, 145–165. Pallas 40. Davies, J. C. 1971. Was Cicero Aware of Natural Beauty? Greece & Rome 18:2, 152–165. Day, J. 1932. Agriculture in the Life of Pompeii. Yale Classical Studies 3, 166–208. De Benedetti A. A., R. Funiciello, G. Giordano, G. Diano, E. Caprilli, and M. Paterne. 2008. Volcanology, History and Myths of the Lake Albano Maar (Colli Albani Volcano, Italy). Journal of Volcanology and Geothermal Research 176, 387–406. De Caro, S. 1994. La villa rustica in località Villa Regina a Boscoreale. Roma: G. Bretschneider. De Franceschini, M. online. La Villa Adriana di Tivoli. (http:// www.villa-adriana.net/; accessed March 28, 2007.) De Franceschini, M. 1991. Villa Adriana – Mosaici, pavimenti, ediici. Roma: “L’Erma” di Bretschneider. De Franceschini, M. 2005. Ville dell’Agro Romano. Monograie della Carta dell’Agro Romano 2. De Francesco, D. 1990. Le donazioni constantiniane nell’Agro Romano. Vetera Christianorum 27, 47–75. De Grossi Mazorin, J. 1987. La fauna. Quaderni di archeologia etrusco-italica 14. Archeologia Laziale VIII, 234–235. De Rita, D., and C. Giampaolo. 2005. Local Vocanic Building Stones Used in the Construction of Ancient Rome. In Cultural Responses to the Volcanic Landscape. The Mediterranean and Beyond, eds. M. S. Balmuth, D. K. Chester, and P. A. Johnston, 165–184. Archaeological Institute of America Colloquia and Conference Papers 8. De Rossi, G. M. 1967. Tellenae. Forma Italiae Regio I:IV. De Rossi, G. M. 1970. Apiolae. Forma Italiae Regio I:IX. De Rossi, G. M. 1979. Bovillae. Forma Italiae Regio I:XV. De Seña, E. C. 2003. Seeing the Trees and the Forest: Toward a More Reined Understanding of Socio-cultural Systems in Classical Antiquity. The Case of Olive Oil in Ancient Latium. Archaeologiae 1:1, 11–32. De Seña, E. C. 2005. An Assessment of Wine and Oil Production in Rome’s Hinterland: Ceramic, Literary, Art Historical and Modern Evidence. In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B. Santillo Frizell and A. Klynne, 135–149. The Swedish Institute in Rome Projects and Seminars 2. Del Lungo, S., ed. 1996. La toponomastica archeologica di Provincia di Roma. Vol. I–II. Roma: Regione Lazio. Del Re, A. 1611. Dell’Antichità Tiburtine. Roma: Giacomo Mascardi. DeLaine, J. 1995. The Supply of Building Materials to the City of Rome. In Settlement and Economy in Italy 1500 BC – AD 1500. Papers of the Fifth Conference of Italian Archaeology, ed. N. Christie, 555–562. Oxbow Monograph 41. DeLaine, J. 1997. The Baths of Caracalla: A Study in the Design, Construction, and Economics of Large-scale Building 238 DeLaine, J. 2001. Bricks and Mortar. Exploring the Economics of Building Techniques at Rome and Ostia. In Economies Beyond Agriculture in the Classical World, eds. D. Mattingly and J. Salmon, 230–268. Leicester–Nottingham Studies in Ancient Societies 9. Dell’Era, F. 2002. Ville e paesaggio: gli impianti idraulici. Bullettino della Commissione Archeologica Comunale di Roma, 249–262. Devoti, L. 1978. Cisterne del periodo romano nel Tuscolano. Itinerari tuscolani. Frascati: Poligraico Laziale. Devoti, L. 1981. L’eremo tuscolano e la villa detta dei Furii. Frascati: Ass. Tuscolana “Amici di Frascati.” Di Blasi, L., L. De Marco, M. Fellak, and E. Foddai. 1999. Elementi e linee ricostruttive di un paesaggio agrario del suburbio di Roma. In Campagna e paesaggio nell’Italia antica, eds.L. Quilici and S. Quilici Gigli, 95–114. Atlante tematico di topograia antica 8 Di Gennaro, F. 1990. Via Nomentana/Via Salaria. Via G. Pasquati (circ. IV). Bullettino della Commissione Archeologica Comunale di Roma, 225–226. Di Gennaro, F. 2006. Tenuta Radicicoli Maffei. Villa presso Via Sandro Pertini. In Roma. Memorie dal sottosuolo. Ritrovamenti archeologici 1980/2006, ed. M. A. Tomei, 235–236. Milano: Electa. Di Gennaro, F., P. Barbina, M. De Filippis, F. Dell’Era, G. Fratianni, and P. Togninelli. 2005. Il liberto Faonte, il notabile Marco Claudio Ponzio Ponziano Marcello e i loro vicini. In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B. Santillo Frizell and A. Klynne, 27–48. The Swedish Institute in Rome Projects and Seminars 2. Di Gennaro, F., and M. De Filippis. 1995. Un sepolcreto d’età imperiale nella tenuta Boccone D’Aste. Quaderni del Centro di studio per l’archeologia etrusco-italica 23. Archeologia Laziale XII:2, 267–274. Di Gennaro, F., F. Dell’Era, F. Fraioli, J. Griesbach, and P. Barbina. 2004. Strutture insediative e tracce di uso agrario del territorio idenate in età romana. In Atti del IV Congresso di Topograia Antica. Insediamenti e strutture rurali nell’Italia Romana. (Roma 7–8 Marzo 2001), 83–148. Journal of Ancient Topography XIV. Di Gennaro, F., and J. Griesbach. 2003. Le sepolture all’interno delle ville con particolare riferimento al territorio di Roma. In Suburbium: Il suburbio di Roma dalla crisi del sistema delle ville a Gregorio Magno, eds. P. Pergola, R. Santangeli Valenzani, and R. Volpe, 123–166. Collection de l’École française de Rome 311. Di Gennaro, F., M. Vitti, F. Fraioli, S. Panciera, S. Biagini, A. Malizia, P. Catalano, S. Di Giannantonio, and A. Starace. 2002. Roma. Via Salaria. La villa “di Marco Claudio Ponzio Ponziano Marcello” e la basilica di San Michele Arcangelo sulla collina di Castel Giubileo. Notizie degli Scavi di Antichità, 465–541. Di Giuseppe, H. 2005. Realtà micro-regionali a confronto sulle due sponde del Tevere tra l’età arcaica e repubblicana. In Papers in Italian Archaeology VI: Communities and Settlements from the Neolithic to the Early Medieval Period. Proceedings of the 6th Conference of Italian Archaeology held at the University of Groningen, Groningen Institute of Archaeology, The Netherlands, April 15–17, 2003, eds. P. Attema, A. Nijboer, and A. Zifferero, 1056–1066. BAR International Series S1452. Di Manzano, P., ed. 2001. Ad deverticulum. Scavi archeologici lungo la bretella Nomentana–GRA. Roma: Provincia di Roma, Generalvie spa. Di Matteo, F. 2002. La villa dei Centroni. Rendiconti bibliography dell’Accademia Nazionale dei Lincei. Classe di Scienze morali, storiche e ilologiche IX:XIII, 243–297. Di Matteo, F. 2003a. Domus suburbana Claudii. Rendiconti dell’Accademia Nazionale dei Lincei. Classe di Scienze morali, storiche e ilologiche IX XIV:1, 79–87. Di Matteo, F. 2003b. Roma. Via Anagnina, vocabolo “Centroni Grotte.” Saggi di scavo nella Villa dei Centroni. Notizie degli Scavi di Antichità, 257–330. Di Nicola, M., and A. Tantari. 1998. Da Ponte Mammolo al Torraccio della Cecchina. In Roma oltre le mura. Lineamenti storico topograici del territorio della V circoscrizione, ed. C. Calci, 147–155. Roma: Associazione Culturale “Roma oltre le mura.” Di Sante, S., and G. Presen. 2002. Guidonia: note di scavo in località Martellona. Annali dell’Associazione Nomentana di Storia e Archeologia 8, 88–101. Dommelen, P. van. 1993. Roman Peasants and Rural Organisation in Central Italy: An Archaeological Perspective. In Theoretical Roman Archaeology: First Conference Proceedings, ed. E. Scott, 167–186. Worldwide Archaeology Series 4. Donda, L. 2006. Tenuta Radicicoli Maffei. Villa–Mansio. In Roma. Memorie dal sottosuolo. Ritrovamenti archeologici 1980/2006, ed. M. A. Tomei, 237. Milano: Electa. Drerup, H. 1959. Die römische Villa. Marburger WinckelmannProgramm 1959, 1–24. Fiocchi Nicolai, V., M. G. Granino Cecere, and Z. Mari, eds. 2001. Lexicon Topographicum Urbis Romae – Suburbium. Vol. I A–B. Roma: Edizioni Quasar. Fiocchi Nicolai, V., M. G. Granino Cecere, and Z. Mari, eds. 2004. Lexicon Topographicum Urbis Romae – Suburbium. Vol. II C–F. Roma: Edizioni Quasar. Fiocchi Nicolai, V., M. G. Granino Cecere, and Z. Mari, eds. 2005. Lexicon Topographicum Urbis Romae – Suburbium. Vol. III G–L. Roma: Edizioni Quasar. Fiocchi Nicolai, V., M. G. Granino Cecere, and Z. Mari, eds. 2006. Lexicon Topographicum Urbis Romae – Suburbium. Vol. IV M–Q. Roma: Edizioni Quasar. Fiocchi Nicolai, V., M. G. Granino Cecere, and Z. Mari, eds. 2008. Lexicon Topographicum Urbis Romae – Suburbium. Vol. V R–Z. Roma: Edizioni Quasar. FitzPatrick, E. A. 1983. Soils. Their Formation, Classiication and Distribution. London: Longman. Flach, D. 1990. Römische Agrargeschichte. Handbuch der Altertumswissenschaft 3:9. Fleming, A. 2006. Post-processual Landscape Archaeology: A Critique. Cambridge Archaeological Journal 16:3, 267–280. Follieri, M., M. Giardini, D. Magri, and L. Sadori. 1998. Palynostratigraphy of the Last Glacial Period in the Volcanic Region of Central Italy. Quaternary International 47–48, 3–20. Duncan Jones, R. 1974. The Economy of the Roman Empire. Quantitative Studies. Cambridge: Cambridge University Press. Follieri, M., D. Magri, and L. Sadori. 1988. 250,000-year Pollen Record from the Valle di Castiglione (Roma). Pollen et spores 30:3–4, 329–356. Duprè, X., ed. 2000. Scavi archeologici di Tusculum. Rapporti preliminari delle campagne 1994–1999. Roma: XI Comunità montana del Lazio “Castelli romani e prenestini,” Escuela Española de Historia y Arqueología, CSIC. Follieri, M., D. Magri, and L. Sadori. 1989. Pollen Stratigraphical Synthesis from Valle di Castiglione (Roma). Quaternary International 3–4, 81–84. Dyson, S. L. 1978. Settlement Patterns in the Ager Cosanus: The Wesleyan University Survey, 1974–1976. Journal of Field Archaeology 5:3, 251–268. Foss, J. E. 1988. Paleosols of Pompeii and Oplontis. In Studia Pompeiana & Classica in Honor of Wilhelmina F. Jashemski. Vol. I: Pompeiana, ed. R. I. Curtis, 127–144. New Rochelle: A. D. Caratzas. Dyson, S. L. 1992. Community and Society in Roman Italy. Baltimore, London: Johns Hopkins University Press. Fraioli, F. 2000. Recenti scavi nella Tenuta Radicicoli. Bullettino della Commissione Archeologica Comunale di Roma, 225–236. Dyson, S. L. 2003. The Roman Countryside. London: Duckworth. Franck, C. 1956. Die Barockvillen in Frascati. Ihre Gestaltung aus den landschaftlichen Gegebenheiten. München: Deutschen Kunstverlag. Edlund-Berry, I. 2006. Hot, Cold, or Smelly: the Power of Sacred Water in Roman Religion, 400–100 BCE. In Religion in Republican Italy, eds. C. E. Schultz and P. B. Harvey, Jr., 162–180. Yale Classical Studies 33. Ehrlich, T. L. 2002. Landscape and Identity in Early Modern Rome. Villa Culture at Frascati in the Borghese Era. Cambridge, New York: Cambridge University Press. Eriksdotter, G. 2005. Bakom fasaderna. Byggnadsarkeologiska sätt att fånga tid, rum och bruk. Lund Studies in Medieval Archaeology 36. Evans, H. B. 1993. In Tiburtium usum. Special Arrangements in the Roman Water System (Frontinus Aq. 6,5). American Journal of Archaeology 97, 447–455. Frank, T. 1924. Roman Buildings of the Republic. An Attempt to Date Them from Their Materials. Papers and Monographs of the American Academy in Rome III. Frank, T. 1940: An Economic Survey of Ancient Rome. Vol. IV. Rome and Italy of the empire. Baltimore: Johns Hopkins Press. Freda, C., M. Gaeta, D. B. Karner, F. Marra, P. R. Renne, J. Taddeucci, P. Scarlato, J. N. Christensen, and L. Dallai. 2006. Eruptive History and Petrologic Evolution of the Albano Multiple Maar (Alban Hills, Central Italy). Bulletin of Volcanology 68, 567–591. FAO. 1993. World Soil Resources. An Explanatory Note on the FAO World Soil Resources Map at 1:25,000,000 scale. World Soil Resources Reports 66 Rev. 1. Frizell, B. Santillo. 2004. Curing the Flock. The Use of Healing Waters in Roman Pastoral Economy. (With an Appendix by K. F. Kitchell, Inscription CIL XIV, 3911 from Aquae Albulae, Tivoli.) In Pecus. Man and Animal in Antiquity. Proceedings of the Conference at the Swedish Institute in Rome, September 9–12, 2002, ed. B. Santillo Frizell, 80–93. The Swedish Institute in Rome Projects and Seminars 1. FAO. 1998. World Reference Base for Soil Resources. World Soil Resources Reports 84. Frutaz, A. P. 1972. Le carte del Lazio. Vol. I–III. Roma: Istituto di Studi Romani. Favro, D. 1996. The Urban Image of Augustan Rome. Cambridge: Cambridge University Press. Funiciello, R., G. Giordano, and D. De Rita. 2003. The Albano Maar Lake (Colli Albani Volcano, Italy): Recent Volcanic Activity and Evidence of Pre-Roman Age Catastrophic Lahar Events. Journal of Volcanology and Geothermal Research 123, 43–61. Faegri, K., and J. Iversen. 1989. Textbook of Pollen Analysis. Chichester: Wiley. Fentress, E., D. Kennet, and I. Valenti. 1990. A Sicilian villa and its landscape (Contrada Mirabile, Mazara del Vallo, 1988). Opus 5, 75–90. Filippini, P. 2006. Località Sant’Eusebio (Municipio V). Via Carciano. Villa residenziale. In Roma. Memorie dal sottosuolo. Ritrovamenti archeologici 1980/2006, ed. M. A. Tomei, 263– 264. Milano: Electa. Funiciello, R., G. Giordano, D. De Rita, M. L. Carapezza, and F. Barberi. 2002. L’attività recente del cratere del Lago Albano di Castelgandolfo. Rendiconti Lincei Science Fisiche e Naturali 9:13, 113–143. 239 bibliography Funiciello, R., and M. Parotto 2001: General Geological Features of the Campagna Romana. The World of the Elephants – International Congress, Rome 2001, eds. G. Cavarretta, P. Gioia, M. Mussi, and M. R. Palombo, 48–52. Roma: Istituto Salesiano Pio XI. Funiciello, R., A. Praturlon, and G. Giordano, eds. 2008. La geologia di Roma dal centro storico alla periferia. Vol. I–II. Memorie descritteve della Carta Geologia d’Italia LXXX. Förtsch, R. 1993. Archäologischer Kommentar zu den Villenbriefen des jüngeren Plinius. Beiträge zur Erschließung hellenistischer und kaiserzeitlicher Skulptur und Architektur 13. Gaffney, V., and Z. Stančič. 1991. GIS Approaches to Regional Analysis: A Case Study of the Island of Hvar. Ljubljana: Znanstveni inštitut Filozofske fakultete. Gatti, S. 1991. Guidonia (Roma). Località Inviolata. Indagini preliminari nell’area della villa romana. Bollettino di archeologia 11–12, 218–221. Garnsey, P. 1979. Where Did Roman Peasants Live? Proceedings of the Cambridge Philological Society XXV, 1–25. Giampaolo, C, G. Lombardi, and M. Mariottini. 2008. Pietre e costruito della città di Roma: dall’antichità ai giorni nostri. In La geologia di Roma dal centro storico alla periferia II, eds. R. Funiciello, A. Praturlon, and G. Giordano, 273–406. Memorie descritteve della Carta Geologia d’Italia LXXX. Grahame, M. 2000. Reading Space: Social Interaction and Identity in the Houses of Roman Pompeii: A Syntactical Approach to the Analysis and Interpretation of Built Space. BAR International Series S886. Granino Cecere, M. G. 1992. Epigraia dei santuari rurali del Latium Vetus. Mélanges d l’École française de Rome. Antiquité 104:1, 125–143. Granino Cecere, M. G. 1995. Villa Mamurrana. Rendiconti. Atti della Accademia nazionale dei Lincei Serie IX, VI:2, 361–386. Granino Cecere, M. G. 2000. I Sulpicii e il Tuscolano. Rendiconti. Atti della Pontiicia Accademia Romana di Archeologia LXIX, 233–251. Granino Cecere, M. G., and A. Magioncalda. 2003. L’ara di C. Vibullius Fidus e i procuratori della Syria. Mélanges d l’École française de Rome. Antiquité 115:2, 615–638. Griesbach, J. 2000. Le aree funerarie del contado suburbano: alcune rilessioni su rinvenimenti recenti nel territorio ad est di Fidenae. Bullettino della Commissione Archeologica Comunale di Roma, 213–224. Gibson, J. J. 1979. The Ecological Approach to Visual Perception. Boston: Houghton Miflin. Griesbach, J. 2005. Villa e mausoleo: trasformazioni nel concetto della memoria nel suburbio romano. In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B. Santillo Frizell and A. Klynne, 113–124. The Swedish Institute in Rome Projects and Seminars 2. Gilkes, O. J., S. Passigli, and R. Schinke. 1994. Porta Pia: Excavation and Survey in an Area of Suburban Rome. Part 2. Papers of the British School at Rome 62, 101–137. Griesbach, J. 2007. Villen und Gräber. Siedlungs- und Bestattungsplätze der römischen Kaiserzeit im Suburbium von Rom. Internationale Archäologie 103. Gioia, P., and R. Volpe, eds. 2004. Centocelle I. Roma S.D.O. Le indagini archeologiche. Soveria Mannelli: Rubbettino. Gros, P. 2001. L’Architecture Romaine du début du IIIe siécle av. J.-C. à la in du Haut-Empire 2. Maisons, palais, villas et tombeaux. Paris: Picard. Giraudi, C. 2005. Late-Holocene Alluvial Events in the Central Apennines, Italy. The Holocene 15:5, 768–773. Giuliacci, M., S. Abelli, and G. Dipierro. 2001. Il clima dell’Italia nell’ultimo ventennio. Milano: Alpha Test. Gross, H. 1990. Rome in the Age of Enlightenment: The PostTridentine Syndrome and the Ancien Régime. Cambridge: Cambridge University Press. Giuliani, C. F. 1966. Tibur. Pars altera. Forma Italiae Regio I:VII. Grossi Gondi, F. 1901. Le ville Tusculane nell’epoca classica e dopo il Rinascimento. La villa dei Quintilii e la villa di Mondragone. Roma: Unione cooperativa. Giuliani, C. F. 1970. Tibur. Pars prima. Forma Italiae Regio I:XX. Grossi Gondi, F. 1908. Il Tuscolano nell’età classica. Escursioni archeologiche. Roma: Loescher. Giuliani, C. F. 1990. L’edilizia nell’antichità. Roma: La nuova Italia scientiica. Guaitoli, M. and P. Zaccagni. 1985. Via Collatina. Località La Rustica. Interventi di scavo collegati ad opere di urbanizzazione (circ. VII). I. Scavo 1975. Bullettino della Commissione Archeologica Comunale di Roma, 119–124. Giuliani, C. F. 2005. La Villa Gregoriana a Tivoli. Le testimonianze archeologiche e gli interventi sul corso dell’Aniene. Tivoli: Tiburis Artistica ed. Gliozzo, E., and G. Filippi. 2005. Archeologia e archeometria della doliare bollata “urbana:” ulteriori dati e rilessioni. In Interpretare i bolli laterizi di Roma e della valle del Tevere: produzione, storia economica e topograica, ed. C. Bruun, 229– 247. Acta Instituti Romani Finlandiae 32. Goodchild, H. 2003. Modelling Agriculture in the Middle Tiber Valley. Tiber Valley Workshop 2003, II. Towns and Hinterland. (http://www.bsr.ac.uk/BSR/sub_arch/Attach/Modelling_ Agriculture.pdf; accessed June 2009) Goodchild, H. 2006. Modelling Roman Demography and Urban Dependency in Central Italy. In TRAC 2005: Proceedings of the Fifteenth Annual Theoretical Roman Archaeology Conference Which Took Place at the University of Birmingham, 31st March – 1st April, eds. B. Croxford, H. Goodchild, J. Lucas, and N. Ray, 42–56. Oxford: Oxbow Books. Goodchild, H. 2007. Modelling Roman Agricultural Production in the Middle Tiber Valley, Central Italy. PhD diss., University of Birmingham. (http://etheses.bham.ac.uk/175/1/ Goodchild07PhD.pdf; accessed June 2009) Graham, S. 2005. Of Lumberjacks and Brick Stamps: Working with the Tiber as Infrastructure. In Roman Working Lives and Urban Living, eds. A. MacMahon and J. Price, 106–124. Oxford: Oxbow Books. 240 Guldager Bilde, P. 2004. Caesar’s Villa? Nordic Excavations of a Roman Villa by Lake Nemi, loc. S. Maria (1998–2001). Analecta Romana Instituti Danici XXX, 7–42. Guldager Bilde, P. 2005. The Roman Villa by Lake Nemi: from Nature to Culture – Between Private and Public. In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B. Santillo Frizell and A. Klynne, 211–219. The Swedish Institute in Rome Projects and Seminars 2. Hackworth Petersen, L. 2006. The Freedman in Roman Art and Art History. Cambridge: Cambridge University Press. Haggett, Peter, A. D. Cliff, and A. Frey. 1977. Locational Analysis in Human Geography. Vol. I–II. London: Edward Arnold. Hales, S. 2003. The Roman House and Social Identity. New York: Cambridge University Press. Hayes, J. W., and I. P. Martini, eds. 1994. Archaeological Survey in the Lower Liri Valley, Central Italy. BAR International Series S595. Heiken, G., R. Funiciello, and D. De Rita. 2005. The Seven Hills of Rome. Princeton: Princeton University Press. Hernández Martínez, M. 2006. Una aproximación a la bibliography identiicación de los uici del ager Tusculanus. Latomus 65:1, 17–33. Hernández Martínez, M. 2007. Redes de comunicación y vialidad en el Lacio meridional. El ager Tusculanus. PhD diss., Universidad autónoma de Madrid. Hesberg, H. von. 1992. Römische Grabbauten. (Monumenta. I sepolcri romani e la loro architettura). Biblioteca di Archeologia 22. Hietala, H. J., and P. A. Larson, eds. 1984. Intrasite Spatial Analysis in Archaeology. Cambridge: Cambridge University Press. Higginbotham, J. A. 1997. Piscinae: Artiicial Fishponds in Roman Italy. Chapel Hill, London: The University of North Carolina Press. Higuchi, T. 1983. The Visual and Spatial Structure of Landscapes. Cambridge: MIT Press. Hobson, B. 2009. Latrinae et foricae. Toilets in the Roman World. London: Duckworth. Hodder, I., and C. Orton. 1976. Spatial Analysis in Archaeology. New Studies in Archaeology 1. Karner, D. B., F. Marra, and P. R. Renne. 2001a. The History of the Monti Sabatini and Alban Hills Volcanoes: Groundwork for Assessing Volcanic-tectonic Hazards for Rome. Journal of Volcanology and Geothermal Research 107, 185–219. Karner, D. B., L. Lombardi, F. Marra, P. Fortini, and P. Renne. 2001b. Age of Ancient Monuments by Means of Building Stone Provenance: A Case Study of the Tullianum, Rome, Italy. Journal of Archaeological Science 28, 387–393. Kehoe, D. P. 1993. Investment in Estates by Upper-Class Landowners in Early Imperial Italy: The Case of Pliny the Younger. In De agricultura. In memoriam Pieter Willem de Neeve (1945–1990), eds. H. Sancisi-Weerdenburg et al., 214– 237. Dutch Monographs on Ancient History and Archaeology X. Kehoe, D. P. 1997. Investment, Proit, and Tenancy: The Jurists and the Roman Agrarian Economy. Ann Arbor: University of Michigan Press. King, A. 1999. Diet in the Roman World: A Regional Intersite Comparison of the Mammal Bones. Journal of Roman Archaeology 12, 168–202. Holstenius, L. 1666. Annotationes in Geographiam sacram Caroli a S. Paulo, Italiam antiquam Cluuerii, et Thesaurum georaphicum Ortelii: quibus accedit Dissertatio duplex de sacramento conirmationis apud græcos. Roma: Typis Iacobi Dragondelli. Kircher, A. 1671. Latium: Id est, Nova & parallela Latii tum veteris tum novi descriptio. Qua quaecumque vel natura, vel veterum Romanorum ingenium admiranda effecit, geographico-historico-physico ratiocinio, juxta rerum gestarum, temporumque seriem exponitur & enucleatur. Amstelodami: Joannem Janssonium à Waesberge & haeredes Elizei Weyerstraet. Ikeguchi, M. 2000. A Comparative Study of Settlement Patterns and Agricultural Structures in Ancient Italy. A Methodology for Interpreting Field Survey Evidence. Kodai 10, 1–59. Kirsch, J. H. I. 2003. Villa dei Misteri. Bauaufnahme, Bautechnik, Baugeschichte. (http://www.freidok.uni-freiburg. de/volltexte/20/; accessed April 2008.) Ingold, T. 2000. The Perception of the Environment: Essays in Livelihood, Dwelling and Skill. London: Routledge. Kleijn, G. de. 2001. The Water Supply of Ancient Rome. City Area, Water, and Population. Dutch Monographs on Ancient History and Archaeology XXII. Jackson, M. D., and F. Marra. 2006. Roman Stone Masonry: Volcanic Foundations of the Ancient City. American Journal of Archaeology 110, 403–436. Jackson, M. D., F. Marra, R. L. Hay, C. Cawood, and E. M. Winkler. 2005. The Judicious Selection and Preservation of Tuff and Travertine Building Stone in Ancient Rome. Archaeometry 47:3, 485–510. Jackson, M. D., F. Marra, D. Deocampo, A. Vella, C. Kosso, and R. L. Hay. 2007. Geological Observations of Excavated Sand (Harenae Fossicae) Used as Fine Aggregate in Roman Pozzolanic Mortars. Journal of Roman Archaeology 20, 25–53. Jansen, G. 2002. Water in de Romeinse stad: Pompeji – Herculaneum – Ostia. Leuven: Peeters. Jiménes Salvador, J. L. 1981. El santuario de Gabii: estudio arquitectónico del templo de Juno Gabina. Zaragoza: Facultad de Filosofía y Letras, Universidad de Zaragoza. Jones, G. D. B. 1963. Capena and the Ager Capenas. Part II. Papers of the British School at Rome XXXI, 100–158. Joolen, E. van. 2003. Archaeological Land Evaluation. A Reconstruction of the Suitability of Ancient Landscapes for Various Land Uses in Italy Focused on the First Millennium BC. PhD diss., University of Groeningen. (http://irs.ub.rug.nl/ ppn/248069004; accessed June 2006) Knapp, A. B., and W. Ashmore. 1999. Archaeological Landscapes: Constructed, Conceptualized, Ideational. In Archaeologies of Landscapes. Contemporary Perspectives, eds. W. Ashmore and A. B. Knapp, 1–30. Oxford: Blackwell. Kolbe, L. 1982. Campanian villa-arkkitehtuuri antiikin orientoitumisteorioiden esimerkkinä. MA thesis, University of Helsinki. Kolendo, J. 1994. Praedia suburbana e loro redditività. In Landuse in the Roman Empire, eds. J. Carlsen, P. Ørsted, and J. E. Skydsgaard, 59–72. Analecta Romani Instituti Danici, Supplementum 22. Koloski-Ostrow, A. O., ed. 2001. Water Use and Hydraulics in the Roman City. Archaeological Institute of America Colloquia and Conference Papers 3. Krause, C. 1998. L’ediicio residenziale di Villa Jovis. Capri: La Conchiglia. Kutbay, B. L. 1998. Palaces and Large Residences of the Hellenistic Age. Studies in Classics 8. Kuusisto, A., and J. Tuppi. 2009. Research on the Crustumerium Road Trench. Fasti Online Documents & Research Folder 143. (http://www.fastionline.org/docs/FOLDER-it-2009-143.pdf; accessed August 2009.) Jongman, W. 1988. The Economy and Society in Pompeii. Dutch Monographs on Ancient History and Archaeology 4. Kuziščin, V. I. 1984. La grande proprietà agraria nell’Italia romana, II sec. a.C. – I sec. d.C. Roma: Editori Riuniti. Jouffroy, H. 1986. La construction publique en Italie et dans l’Afrique romaine. Groupe de recherche d’histoire romaine de l’Université des sciences humaines de Strasbourg, Études et travaux 2. Kvamme, K. L. 1999. Recent Directions and Developments in Geographical Information Systems. Journal of Archaeological Research 7:2, 153–201. Judson, S., and A. Kahane. 1963. Underground Drainageways in Southern Etruria and Northern Latium. Papers of the British School at Rome XXXI, 74–99. Kahane, A., L. Murray Threipland, and J. Ward Perkins. 1968. The Ager Veientanus, North and East of Rome. Papers of the British School at Rome XXXVI. Kahane, A., and J. Ward-Perkins. 1972. The Via Gabina. Papers of the British School at Rome XL, 91–126. Lafon, X. 2001a. Villa Marittima. Recherches sur les villas littorales de l’Italie romaine. Bibliothéque des Écoles françaises d’Athènes et de Rome 307. Lafon, X. 2001b. Le Suburbium. In La ville de Rome sous le Haut-Emipre. Nouvelles connaissances, nouvelles rélexions. Colloque organisé par l’École Française de Rome et la Société des Professeurs d’Histoire ancienne de l’Université, Rome 5–8 Mai 2001, 199–214. Pallas 55. Lake, M., ed. 2007. Viewing Space. World Archaeology 39:1. 241 bibliography Lambeck, K., M. Anzidei, F. Antonioli, A. Benini, and A. Esposito. 2004. Sea Level in Roman Time in the Central Mediterranean and Implications for Recent Change. Earth and Planetary Science Letters 224, 563– 575. Lanciani, R. 1880. Topograia di Roma antica: i comentarii di Frontino intorno le Acque e gli Acquedotti. Silloge epigraica aquaria. Roma: Salviucci. Lanciani, R. 1902–1912. Storia degli Scavi di Roma e notizie intorno le collezioni romani di antichità. Vol. I–VI. Lanciani, R. 1909. Wanderings in the Roman Campagna. London: Constable & Co. Langley, M. M. 2006. Est in agris: A Spatial Analysis of Roman uillae in the Region of Monforte, Alto Alentejo, Portugal. Revista Portuguesa de Arqueologia 9:2, 317–328. Laurence, R. 1998. Land Transport in Roman Italy: Costs, Practice and the Economy. In Trade, Traders and the Ancient City, eds. H. Parkins and C. Smith, 129–148. London, New York: Routledge. Laurence, R. 1999. The Roads of Roman Italy. Mobility and Cultural Change. London, New York: Routledge. Leach, E. Winsor. 1997. Oecus on Ibycus. Investigating the Vocabulary of the Roman House. In Sequence and Space in Pompeii, eds. S. E. Bon and R. Jones, 50–72. Oxford: Oxbow Books. Leen, A. 1991. Cicero and the Rhetoric of Art. American Journal of Philology 112:2, 229–245. Lega, C. 1995. Topograia dei culti delle divinità protettrici dell’agricoltura e del lavoro dei campi nel suburbio di Roma. In Agricoltura e commerci nell’Italia Antica, eds. L. Quilici and S. Quilici Gigli, 115–125. Atlante Tematico di Topograia Antica, Supplemento 1. eds. G. Barker and R. Hodges, 289–305. Papers in Italian Archaeology 2. BAR International Series S102. Lock, G., and Z. Stančič, eds. 1995. Archaeology and Geographical Information Systems: A European perspective. Bristol: Taylor & Francis. Lomas, K. 1997. The Idea of a City: Élite Ideology and the Evolution of Urban Form in Italy, 200 BC – AD 100. In Roman Urbanism. Beyond the Consumer City, ed. H. M. Parkins, 21– 41. London: Routledge. Lowe, J. J., C. A. Accorsi, M. Bandini Mazzanti, A. Bishop, S. van der Kaars, L. Forlani, A. M. Mercuri, C. Rivalenti, P. Torri, and C. Watson. 1996. Pollen Stratigraphy of Sedimental Sequences from Lakes Albano and Nemi (Near Rome) and from the Central Adriatic, Spanning the Interval from Oxygen Isotope Stage 2 to the Present. In Palaeoenvironmental Analysis of Italian Crater Lake and Adriatic Sediments, eds. P. Guilizzoni and F. Oldield, 71–98. Memorie dell’Istituto italiano di idrobiologia 55. Lugli, G. 1957. La tecnica edilizia Romana. Con particolare riguardo a Roma e Lazio. Vol. I–II. Roma: G. Bardi. MacKinnon, M. 2004. Production and Consumption of Animals in Roman Italy: Integrating the Zooarchaeological and Textual Evidence. Journal of Roman Archaeology Supplementary Series 54. MacMullen, R. 1974. Peasants during the Principate. Aufstieg und Niedergang der römischen Welt II:1, 253–261. Majbom Madsen, J. 2003. Signs of Prosperity in Roman Villas in South of Italy during the Third Century. Analecta Romana Instituti Danici XXIX, 29–53. Malone, C., and S. Stoddart, eds. 1994. Territory, Time and State. The Archaeological Development of the Gubbio Basin. Cambridge: Cambridge University Press. Lega, C. 2003. Ratio marmoraria. Iscrizioni estemponaree sulle pareti di un ambiente di età romana nel complesso del Barco Borghese a Monte Porzio Catone (Rm). Mélanges d l’École française de Rome. Antiquité 115, 563–592. Manacorda, D. 1982. Cento anni di ricerche archeologiche italiane: il dibattito sul metodo. Quaderni di Storia VIII:16, 85–119. Leotta, M. C.1993. Alcune classi ceramiche dall’aniteatro romana di Tivoli. Atti e Memorie della Società Tiburtina di Storia e d’Arte LXVI, 13–48. Manacorda, D. 1985. L’interpretazione della villa. Dai Sestii agli Imperatori. In Setteinestre. Una villa schiavistica nell’Etruria romana, ed. A. Carandini, 101–106. Modena: Panini. Leotta, M. C.1995. Ceramica ellenistica a rilievo dall’aniteatro di Tivoli. Quaderni di archeologia etrusco-italica 23. Archeologia Laziale XII:1, 453–458. Mangianti, F., and F. Leone. 2008. Analisi climatiche delle temperature e delle precipitazioni a Roma. In La geologia di Roma dal centro storico alla periferia II, eds. R. Funiciello, A. Praturlon, and G. Giordano, 169–186. Memorie descritteve della Carta Geologia d’Italia LXXX. Leotta, M. C.1997. Fornaci Tiburtine della tarda repubblica 1. Atti e Memorie della Società Tiburtina di Storia e d’Arte LXX, 13–70. Leotta, M. C.1998. Fornaci Tiburtine della tarda repubblica 2. Atti e Memorie della Società Tiburtina di Storia e d’Arte LXXI, 7–64. Leotta, M. C.1999. Fornaci Tiburtine della tarda repubblica 3. Atti e Memorie della Società Tiburtina di Storia e d’Arte LXXII, 7–47. Lewit, T. 1991. Agricultural Production in the Roman Economy A. D. 200–400. BAR International Series S568. Lilli, M. 2001. Lanuvium: avanzi di ediici antichi negli appunti di R. Lanciani. Occasional Papers of the Nordic Institutes in Rome 2. Liverani, P. 1992. Villa romana alla Cecchignola (scavi 1828, 1939). Rendiconti. Atti della Pontiicia Accademia Romana di Archeologia LXII, 173–183. Livi, V. 2006. Località San Basilio. Via pollenza. Villa Romana. In Roma. Memorie dal sottosuolo. Ritrovamenti archeologici 1980/2006, ed. M. A. Tomei, 255. Milano: Electa. Llobera, M. 2001. Building Past Landscape Perception With GIS: Understanding Topographic Prominence. Journal of Archaeological Science 28, 1005–1014. Lloyd, J., and G. Barker. 1981. Rural Settlement in Roman Molise: Problems of Archaeological Survey. In Archaeology and Italian Society. Prehistoric, Roman and Medieval Studies, 242 Mangiatordi, A. 2003. Le ville di Cicerone, fra innovazione e tradizione. Annali della Facoltà di Lettere e Filosoia Università degli Studi di Bari 46, 213–251. Mansuelli, G. A. 1958. Le ville del mondo Romano. Quaderni di cultura materiale I. Marcelli, M. 2002. Tor Bella Monaca. Indagini archeoloiche (Municipio VIII). Bullettino della Commissione Archeologica Comunale di Roma, 241–245. Marcone, A. 1997. Storia dell’agricoltura romana. Dal mondo arcaico all’età imperiale. Roma: Nuova Italia Scientiica. Mari, Z. 1983a. Tibur. Pars tertia. Forma Italiae Regio I:XVII. Mari, Z. 1983b. Materiale epigraico dalla Valle dell’Aniene e dalla Sabina meridionale. Atti e Memorie della Società Tiburtina di Storia e d’Arte LVI, 21–88. Mari, Z. 1984. La villa tiburtine detta di Cassio: nuove acquisizioni. Rivista dell’Istituto Nazionale d’Archeologia e Storia dell’Arte VI–VII, 97–131. Mari, Z. 1988. I resti della cosiddetta villa dei Coponii sulla via Empolitana. Atti e Memorie della Società Tiburtina di Storia e d’Arte LXI, 131–144. Mari, Z. 1991. Tibur. Pars quarta. Forma Italiae Regio I:XXXV. Mari, Z. 1992. Una cava romana all Acquoria (Tivoli). Analecta Romani Instituti Danici 20, 31–42. Mari, Z. 1994. Scoperte archeologiche nel territorio Tiburtino bibliography e nella valle dell’Aniene (IV). Atti e Memorie della Società Tiburtina di Storia e d’Arte LXVII, 145–179. Mari, Z. 1996. Scoperte archeologiche nel territorio tiburtino (V). Atti e Memorie della Società Tiburtina di Storia e d’Arte LXIX, 105–134. Mari, Z. 2004. La cava romana del Barco: stato attuale e prospettive di valorizzazione. Atti e Memorie della Società Tiburtina di Storia e d’Arte LXXVII, 201–230. Mari, Z. 2005. La villa romana di età repubblicana nell’ager Tiburtinus e Sabinus: tra fonti letterarie e documentazione archeologica. In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B. Santillo Frizell and A. Klynne, 75–96. The Swedish Institute in Rome Projects and Seminars 2. Mari, Z., and F. Boanelli. 1991. Tivoli. La villa di Quintilio Varo. Bollettino di Archeologia 10, 37–50. Mari, Z., and E. Moscetti. 1992. Via Tiburtina. Rinvenimenti lungo la via 28bis (Guidonia–Montecelio). Bullettino della Commissione Archeologica Comunale di Roma, 100–115. Mari, Z., and E. Moscetti. 1993. Scoperte archeologiche nel territorio tiburtino. Atti e Memorie della Società Tiburtina di Storia e d’Arte LXVI, 109–146. Mari, Z., E. Moscetti, and L. Rustico. 1995. Via Tiburtina/Via Nomentana. Rinvenimenti fra le vie Tiburtina e Nomentana (Guidonia–Montecelio). Bullettino della Commissione Archeologica Comunale di Roma, 215–239. Suburbium: Il suburbio di Roma dalla crisi del sistema delle ville a Gregorio Magno, eds. P. Pergola, R. Santangeli Valenzani, and R. Volpe, 25–46. Collection de l’École française de Rome 311. Messineo, G. 2005. Ville a Tor di Quinto e nelle Tenute di Grottarossa e Acquatraversa. In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B. Santillo Frizell and A. Klynne, 49–54. The Swedish Institute in Rome Projects and Seminars 2. Messineo, G., and R. Sorella. 1990a. Via Nomentana/Via Salaria. Via delle Vigne Nuove, km 1,06 (circ. IV). Bullettino della Commissione Archeologica Comunale di Roma, 218–222. Messineo, G., and R. Sorella. 1990b. Via Nomentana/Via Salaria. Via delle Vigne Nuove, km 2,10 (circ. IV). Bullettino della Commissione Archeologica Comunale di Roma, 222–224. Messineo, G., and R. Sorella. 1991. I resti di una villa romana a Tor Cervara. Bollettino della Unione storia ed arte XXXIV, 29–38. Mielsch, H. 1987. Die römische Villa. Architektur und Lebensform. München: C. H. Beck. Molinaro, A., R. Leonardi, V. Pisano, and C. Villani. 2001. Geologia, paleoambiente e risorse naturali. In Luoghi e paesaggi archeologici del suburbio orientale di Roma, eds. S. Musco, L. Petrassi, and S. Pracchia, 27–90. Roma: Pegaso. Moltesen, M. 1978. La Giostra – Tellenae? Quaderni del Centro di studio per l’archeologia etrusco-italica 1. Archeologia Laziale I, 60–63. Martins, C. 2003. Becoming Consumers: Looking Beyond Wealth as an Explanation for Villa Variability. In TRAC 2002: Proceedings of Twelth Annual Theoretical Roman Archaeology Conference which took place at The University of Kent at Canterbury, 5–6 April 2002, eds. G. Carr, E. Swift, and J. Weekes, 84–100. Oxford: Oxbow Books. Moltesen, M. 1980. La Giostra (Roma): Rapporto preliminare sullo scavo di sondaggio del novembre 1976. Notizie degli Scavi di Antichità, 51–58. Marzano, A. 2007. Roman Villas in Central Italy. A Social and Economic History. Columbia Studies in the Classical Tradition 30. Moltesen, M., and J. R. Brandt. 1994. Excavations at La Giostra. A Mid-Republican Fortress Outside Rome. Analecta Romani Instituti Danici, Supplementum 21. Mastrantonio, G., A. P. Viola, S. Argentini, G. Fiocco, L. Giannini, L. Rossini, G. Abbate, R. Ocone, and M. Casonato. 1994. Observations of Sea Breeze Events in Rome and The Surrounding Area by a Network of Doppler Sodars. BoundaryLayer Meteorology 71, 67–80. Montalcini De Angelis d’Ossat, M. 1983. Via Prenestina. Rinvenimenti in località Lunghezzina. In Roma, Archeologia e progetto. Mercati Traianei, Roma, 23 maggio–30 giugno 1983. A cura di M. Mattei and M. Wappner, 30–31. Roma: Multigraica. Moltesen, M. 1988. Via Appia/Via Ardeatina. Località La Giostra. Bullettino della Commissione Archeologica Comunale di Roma, 559–566. Mattei, D. B. 1711. Memorie istoriche dell’antico Tuscolo, oggi Frascati. Roma: Stamperia di Gio. Francesco Buagni. Morelli, C. 1984. Via Labicana. Intervento n. 1: Loc. Carcaricola. Bullettino della Commissione Archeologica Comunale di Roma, 91–93. Mattingly, D. J. 1996. First Fruit? The Olive in the Roman World. In Human Landscapes in Classical Antiquity: Environment and Culture, eds. G. Shipley and J. Salmon, 213–253. Leicester– Nottingham Studies in Ancient Society 6, 1996. Morelli, C., and S. Musco. 1984. Via Labicana. Interventi nn. 7–10. Bullettino della Commissione Archeologica Comunale di Roma, 96–97. Mayer, J. W. 2005. Imus ad villam. Studien zur Villeggiatur im stadtrömischen Suburbium in der späten Republik und frühen Kaiserzeit. Geographica Historica 20. Moretti, L. 1986. Due epigrammi greci inediti di Roma. Rendiconti. Atti della Pontiicia Accademia Romana di Archeologia 57, 233–246. McCracken, G. E. 1935. Cicero’s Tusculan Villa. Classical Journal 30, 261–277. Morley, N. 1996. Metropolis and Hinterland. The City of Rome and the Italian Economy 200 B.C. – A.D. 200. Cambridge: Cambridge University Press. McCracken, G. E. 1942. The Villa and Tomb of Lucullus at Tusculum. American Journal of Archaeology XLVI, 325–340. Morley, N. 2004. Theories, Models and Concepts in Ancient History. London: Routledge. McKay, A. C. 1975. Houses, Villas and Palaces in the Roman World. London: Thames and Hudson. Moscetti, E. 1991. Proposta di un parco archeologico-naturale in Guidonia–Montecelio. Atti e Memorie della Società Tiburtina di Storia e d’Arte LXIV, 139–179. Melis, M., and S. Vardaro. 1993. Gabii: storia di una città. Roma: I. T. C. “Botticelli.” Mercuri, A. M., C. A. Accorsi, and M. Bandini Mazzanti. 2002. The Long History of Cannabis and its Cultivation by the Romans in Central Italy, Shown by Pollen Records from Lago Albano and Lago di Nemi. Vegetation History and Archaeobotany 11, 263–276. Messineo, G. 1992. Via Tiburtina. Via Antenore (Tor Cervara) (circ. V). Bullettino della Commissione Archeologica Comunale di Roma, 88–91. Messineo, G. 2003. Via Flaminia tra V and VI miglio. In Moscetti, E. 1994. In rinvenimento del gruppo scultureo della Triade Capitolino nella villa romana dell’Inviolata (Guidonia– Montecelio). Atti e Memorie della Società Tiburtina di Storia e d’Arte LXVII, 181–193. Moscetti, E. 1995. Scoperte archeologiche nel territorio Nomentano. Annali dell’Associazione Nomentana di Storia e Archeologia 1, 16–20. Moscetti, E. 1996. Notiziario archeologico. Annali dell’Associazione Nomentana di Storia e Archeologia 2, 57–60. Moscetti, E. 1997. Notiziario archeologico. Annali 243 bibliography dell’Associazione Nomentana di Storia e Archeologia 3, 141– 158. eds. J. Carlsen, P. Ørsted, and J. E. Skydsgaard, 107–114. Analecta Romani Instituti Danici, Supplementum 22. Moscetti, E. 1999a. Notiziario archeologico. Annali dell’Associazione Nomentana di Storia e Archeologia 5, 123– 135. Naval Intelligence Division. 1944. Italy. Vol. 1. Geographical Handbook Series 517. Moscetti, E. 1999b. Scoperte archeologiche nella villa romana della Triade Capitolina a Guidonia. Annali dell’Associazione Nomentana di Storia e Archeologia 6, 35–39. Moscetti, E. 2001. Notiziario archeologico. Annali dell’Associazione Nomentana di Storia e Archeologia 7, 112– 120. Moscetti, E. 2002. I bolli laterizi dell’Antiquarium comunale e del Museo della Via Cornicolana a Guidonia. Annali dell’Associazione Nomentana di Storia e Archeologia 8, 65–87. Moscetti, E. 2003. Notiziario archeologico. Annali dell’Associazione Nomentana di Storia e Archeologia 9, 139– 158. Moscetti, E. 2004. Notiziario archeologico. Annali dell’Associazione Nomentana di Storia e Archeologia 10, 168– 187. Neeve, P. W. de. 1984. Colonus. Private Farm-tenancy in Rome during the Republic and the Early Principate. Amsterdam: J. C. Gieben. Neeve, P. W. de. 1990. A Roman Landowner and His Estates. Pliny the Younger. Athenaeum 78, 363–402. Neudecker, R. 1988. Die Skulpturausstattung römischer Villen in Italien. Beiträge zur Erschließung hellenistischen und kaiserzeitlicher Skulptur und Architektur 9. Nibby, A. 1819. Viaggio antiquario ne’ contorni di Roma. Vol. I–II. Reprint Biblioteca istorica della antica e nuova Italia 158. Nibby, A. 1848–1849. Analisi storico-topograico antiquaria della carta de’ dintorni di Roma. Vol. I–III. Nicodemi, M. A. ca. 1585–1589 [1926]. Storia di Tivoli, eds. A. Bussi and V. Paciici. Studi e Fonti per la storia della regione tiburtina IV. Nielsen, I. 1994. Hellenistic Palaces: Tradition and Renewal. Studies in Hellenistic Civilization V. Moscetti, E., and A. La Porta. 2004. Tra Nomentum e Ficulea. Il territorio archeologico di Fonte Nuova. In Fonte Nuova entra nella storia. Con il saggio “Tra Nomentum e Ficulea. Il territorio archeologico di Fonte Nuova,” S. G. Vicario, 137–185. Roma: Istituto Poligraico e Zecca dello Stato Nissinen, L. 2010. Cubicula diurna, nocturna – Revisiting Roman cubicula and Sleeping Arrangements. Arctos XLIII, 85–107. Motta, L. 2002. Planting the Seed of Rome. Vegetation History and Archaeobotany 11, 71–77. Norberg-Schulz, C. 1980. Genius loci: Towards Phenomenology of Architecture. New York: Rizzoli. Musco, S. 1984. Via Labicana. Intervento n. 13. Bullettino della Commissione Archeologica Comunale di Roma, 98–100. Ohlig, C. P. J. 2001. De aquis pompeiorum. Das castellum aquae in Pompeji: Herkunft, Zuleitung und Verteilung des Wassers. Circumvesuviana 4. Musco, S., and P. Zaccagni. 1985. Caratteri e forme di insediamenti rustici e residenziali nel suburbio orientale tra il IV ed il I secolo a.C. In Misurare la terra: centuriazione e coloni nel mondo romano. Città, agricoltura, commercio: materiali di Roma e del Suburbio, eds. R. Bussi and V. Vandelli 90–106. Modena: Panini. Musco, S. 2001. L’attività della Soprintendenza Archeologica di Roma in un settore del Suburbio orientale. In Luoghi e paesaggi archeologici del suburbio orientale di Roma, eds. S. Musco, L. Petrassi and S. Pracchia, 149–236. Roma: Pegaso. a Olcese, G. 1997. Notizia preliminare sulle analisi di laboratorio della ceramica “italo-megarese” di Tivoli. Atti e Memorie della Società Tiburtina di Storia e d’Arte LXX, 71–74. Oliver-Smith, P., and W. M. Widrig. 1981. Roma. Loc. Tor Bella Monaca: Villa rustica romana. Relazione preliminare sulle campagne di scavo 1976 e 1977 nell’agro romano. Notizie degli Scavi di Antichità, 99–114. Ortolani, F., and S. Pagliuca. 1994. Variazioni climatiche e crisi dell’ambiente antropizzato. Il Quaternario 7, 351–356. Musco, S., and M. Cima. 2006. La villa della Tenuta della Lunghezza e la testa marmorea di Augusto. In Roma. Memorie dal sottosuolo. Ritrovamenti archeologici 1980/2006, ed. M. A. Tomei, 311–312. Milano: Electa. Ortolani, F., and S. Pagliuca. 1996. Variazioni climaticoambientali nell’area mediterranea durante il periodo storico. Evidenze geoarcheologiche di cicliche crisi ambientali tipo “effetto serra.” Il Quaternario 9:1, 209–212. Musco, S., and A. Delino. 2002: Via Labicana/Via Prenestina. Ritrovamenti archeologici 1997–2002 (Municipio VIII). Insediamenti nr. 15. Bullettino della Commissione Archeologica Comunale di Roma, 256–258 nr. 15. Ortolani, F., and S. Pagliuca. 2003. Variazioni climatiche cicliche e modiicazioni ambientali nel periodo storico. In Variazioni climatico-ambientali e impatto sull’uomo nell’area circumMediterranea durante l’Olocene. = Climatic-environmental Variations and Impact on Man in the Circum-Mediterranean Area During the Holocene, eds. C. Albore Livadie and F. Ortolani, 165–170. Territorio storico e ambiente 3. Musco, S., L. De Michelis, and E. Brienza. 2002. Via Labicana/Via Prenestina. Ritrovamenti archeologici 1997– 2002 (Municipio VIII). Insediamenti nr. 16. Bullettino della Commissione Archeologica Comunale di Roma, 259 Nr. 16. Pala, C. 1976. Nomentum. Forma Italiae Regio I:XII. Musco, S., C. Morelli, and M. Brucchetti. 1995. Ager Gabinus; note di topograia storica. Quaderni del Centro di studio per l’archeologia etrusco-italica 23. Archeologia Laziale XII:2, 275–292. Panciera, S. 1999. De formis locorum. Corpus inscriptionum Latinarum VI, 4665–4666. Musco, S., L. Petrassi, and S. Pracchia, eds. 2001. Luoghi e paesaggi archeologici del suburbio orientale di Roma. Roma: Pegaso. Pantano, D. 1998. Tenute di Sant’Agata e Capobianco. Bullettino della Commissione Archeologica Comunale di Roma, 310–312. Muzzioli, M. P. 1970. Praeneste. Forma Italiae Regio I:VIII. Myers, K. S. 2000. Miranda ides: Poet and Patrons in Paradoxographical Landscapes in Statius’ Silvae. Materiali e discussioni per l’analisi dei testi classici 44, 103–138. Panimolle, G. 1968. Gli acquedotti di Roma antica. Roma: Abete. Pantano, D. 2001. Ipotesi per la collocazione di Ficulea. Annali dell’Associazione Nomentana di Storia e Archeologia 7, 69–71. Myers, K. S. 2005. Docta Otia: Garden Ownership and Conigurations of Leisure in Statius and Pliny the Younger. Arethusa 38, 103–129. Parenti, R. 1988a. Le tecniche di documentazione per una lettura stratigraica dell’elevato. In Archeologia e restauro dei monumenti. I ciclo di lezioni sulla ricerca applicata in archeologia, Certosa di Pontignano (Siena), 28 settembre – 10 ottobre 1987, eds. R. Francovich and R. Parenti, 249–279. Firenze: All’Insegna del Giglio. Mørch, H. F. C. 1994. Agricultural Landscape: A Geographer’s Considerations on the Past. In Landuse in the Roman Empire, Parenti, R. 1988b. Sulle possibilità di datazione per una lettura stratigraica dell’elevato. In Archeologia e restauro 244 bibliography dei monumenti. I ciclo di lezioni sulla ricerca applicata in archeologia, Certosa di Pontignano (Siena), 28 settembre – 10 ottobre 1987, eds. R. Francovich and R. Parenti, 280–304. Firenze: All’Insegna del Giglio. Paribeni, R. 1926. Marino: Rinvenimenti nell’area dell’antica Bovillae. Notizie degli Scavi di Antichità, 206–209. Patterson, H., ed. 2004. Bridging the Tiber. Approaches to Regional Archaeology in the Middle Tiber Valley. Archaeological Monographs of the British School at Rome 13. Patterson, H., H. Di Giuseppe, and R. Witcher. 2004. Three South Etruscan “Crises:” First Results of the Tiber Valley Project. Papers of the British School at Rome LXXII, 1–36. Patterson, J. R. 1987. Crisis: What Crisis? Rural Change and Urban Development in Imperial Apennine Italy. Papers of the British School at Rome LV, 115–146. Pavese, M. P. 2004. Fundus cum vadis et alluvionibus. Gli incrementi luviali. Fra documenti della prassi e rilessione giurisprudenziale romana. Minima epigraphica et papyrologica Supplementa IV. Pavolini, C., S. Dinuzzi, C. Cupitò, and U. Fusco. 2003. L’area compresa fra il Tevere, l’Aniene e la via Nomentana. In Suburbium: Il suburbio di Roma dalla crisi del sistema delle ville a Gregorio Magno, eds. P. Pergola, R. Santangeli Valenzani, and R. Volpe, 47–95. Collection de l’École française de Rome 311. Pellegrini, E. 1997. Scavi archeologici a Centocelle – Roma. AIAC News 14,2–3. Pensabene, P. 1999. Terrecotte del Museo Nazionale Romano I. Gocciolatoi e protomi da sime. Studia Archeologica 101. Percival, J. 1976. The Roman Villa: An Historical Introduction. London: B.T. Batsford. Perego, L. 1990. Via Salaria. Tenuta di Redicicoli (circ. IV). Bullettino della Commissione Archeologica Comunale di Roma, 244. Pergola, P., R. Santangeli Valenzani, and R. Volpe, eds. 2003. Suburbium: Il suburbio di Roma dalla crisi del sistema delle ville a Gregorio Magno. Collection de l’École française de Rome 311. Perkins, P. 1999. Etruscan Settlement, Society and Material Culture in Central Coastal Etruria. BAR International Series S788. Petracca, L., and L. M. Vigna. 1985. Le fornaci di Roma e suburbio. In Misurare la terra: centuriazione e coloni nel mondo romano. Città, agricoltura, commercio: materiali di Roma e del Suburbio, eds. R. Bussi and V. Vandelli, 131–137. Modena: Panini. Pirro, M. 2002. Un esempio di archeosismologia nel territorio della Campagna Romana. Annali dell’Associazione Nomentana di Storia e Archeologia 8, 102–106. Pisani Sartorio, G. 1995, Via Maresciallo Pilsudski – Via P. De Coubertin (circ. II). Rinvenimenti archeologici nel corso dei saggi di scavo per le fondazioni dell’Auditorium di Roma (ex parcheggio Flaminio al Villaggio Olimpico). Bullettino della Commissione Archaeologica Comunale di Roma, 276–280. Pitkäranta, R. 2001. Suomi–latina–suomi-sanakirja. Helsinki: Werner Söderström Osakeyhtiö. Potter, T. W. 1976. Valleys and Settlement: Some New Evidence. World Archaeology 8:2, 207–219. Potter, T. W. 1979. The Changing Landscape of South Etruria. London: P. Elek. and K. Lomas, 151–179. London: UCL Press. Pracchia, S. 2001. Note per un’archeologia dei paesaggi agrari. (Con contributi di F. M. Cifarelli e R. Zaccagnini). In Luoghi e paesaggi archeologici del suburbio orientale di Roma, eds. S. Musco, L. Petrassi, and S. Pracchia, 237–332. Roma: Pegaso. Quilici, L. 1969. Inventario e localizzazione dei beni culturali archeologici del territorio del comune di Roma. Urbanistica 54–55, i–xx. Quilici, L. 1974a. Collatia. Forma Italiae Regio I:X. Quilici, L. 1974b. La campagna romana come suburbio di Roma antica. Parola del Passato 29, 410–438. Quilici, L. 1979. La villa nel suburbio romano: problemi di studio e di inquadramento storico-topograico. Archeologia Classica XXXI, 309–317. Quilici, L. 1989. Via Appia da Porta Capena ai Colli Albani. Roma: Flli. Palombi. Quilici, L. 1994. Centuriazione e paesaggio agrario nell’Italia centrale. In Landuse in the Roman Empire, eds. J. Carlsen, P. Ørsted, and J. E. Skydsgaard, 127–134. Analecta Romani Instituti Danici, Supplementum 22. Quilici, L., and S. Quilici Gigli. 1980. Crustumerium. Latium Vetus III. Quilici, L., and S. Quilici Gigli. 1986. Fidenae. Latium Vetus V. Quilici, L., and S. Quilici Gigli. 1991. Tusculum ed il parco archeologico. Roma: XI Comunità Montana del Lazio “Castelli romani e prenestini.” Quilici, L., and S. Quilici Gigli. 1993. Ficulea. Latium Vetus VI. Quilici, L., and S. Quilici Gigli. 1995. Un grande santuario fuori la porta occidentale di Tusculum. Quaderni del Centro di studio per l’archeologia etrusco-italica 23. Archeologia Laziale XII:1, 509–534. Quilici, L., and S. Quilici Gigli. 2004. Introduzione alla topograia antica. Bologna: Il Mulino. Quilici Gigli, S. 1987. Su alcuni segni dell’antico paesaggio agrario presso Roma. Quaderni di archeologia etrusco-italica 14. Archeologia Laziale VIII, 152–166. Rasch, J. J. 1998. Das Mausoluem der Kaiserin Helena in Rom und der ”Tempio della Tosse” in Tivoli. Spätantike Zentralbauten in Rom und Latium 3. Ravelli, F., and P. Howarth. 1989. Irrigazione, drenaggio e sanità negli scritti dei georgici latini. Irrigazione e drenaggio. Edagricole Bologna XXXVI:2. Rawson, E. 1976. The Ciceronian Aristocracy and its Properties. In Studies in Roman Property, ed. M. I. Finley, 85–102. Cambridge, New York: Cambridge University Press. Ratilainen, T. 2001. Stratigrainen muurien lukeminen – esimerkkinä Hattulan Pyhän Ristin kirkko. SKAS 4, 4–13. Rea, R. 1985a. Via Prenestina. Località Torre Spaccata: ville rustiche (circ. VIII). Bullettino della Commissione Archeologica Comunale di Roma, 102–111. Rea, R. 1985b. Gli impianti del settore Est. In Misurare la terra: centuriazione e coloni nel mondo romano. Città, agricoltura, commercio: materiali di Roma e del Suburbio, eds. R. Bussi and V. Vandelli, 119–121. Modena: Panini. Rea, R. 2003. Via Latina. In Suburbium: Il suburbio di Roma dalla crisi del sistema delle ville a Gregorio Magno, eds. P. Pergola, R. Santangeli Valenzani, and R. Volpe, 241–266. Collection de l’École française de Rome 311. Relph, E. C. 1976. Place and Placelessness. London: Pion. Pucci, G. 1985. Il sistema della villa nell’Italia centrale. La villa nel suo insieme. In Setteinestre. Una villa schiavistica nell’Etruria romana, ed. A. Carandini, 15–21. Modena: Panini. Ricci, A. 1986. Via Appia. La Villa dei Quintili (circ. XI). Bullettino della Commissione Archeologica Comunale di Roma, 607–615. Pucci, G. 1994. La prova in archeologia. Quaderni storici 85, 59–74. Ricci, A., ed. 1998. La villa dei Quintilii. Fonti scritte e fonti igurate. Roma: Lithos editrice. Purcell, N. 1995. The Roman villa and the Landscape of Production. In Urban Society in Roman Italy, eds. T. J. Cornell Ricciardi, M. 2002. Via Grotta Perfetta. Bullettino della 245 bibliography Commissione Archeologica Comunale di Roma, 377–384. Ricciardi, M. 2005. La villa di Casale Novelli: recenti indagini di scavo in via Grotta Perfetta (suburbio sud di Roma). In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B. Santillo Frizell and A. Klynne, 197–210. The Swedish Institute in Rome Projects and Seminars 2. Ricotti, E. Salza Prina. 2001. Villa Adriana. Il sogno di un imperatore. Bibliotheca archaeologica 29. Riera, I. 1994. Le testimonianze archeologiche. In Utilitas necessaria. Sistemi idraulici nell’Italia Romana, ed. I. Riera, 163–468. Milano: Progetto Quarta Dimensione. Riggsby, A. M. 1997. “Public” and “Private” in Roman Culture: The Case of the Cubiculum. Journal of Roman Archaeology 10, 36–56. Riggsby, A. M. 1998. Self and Community in the Younger Pliny. Arethusa 31:1, 75–97. Rinkewitz, W. 1984. Pastio villatica. Untersuchungen zur intensiven Hoftierhaltung in der römischen Landwirtschaft. Europäische Hochschulschriften 234. Rodger, A. 1972. Owners and Neighbours in Roman Law. Oxford: Clarendon Press. Romizzi, L. 2001. Ville d’otium dell’Italia antica (II sec. a.C. – I sec. d.C.). Aucnus X. Rosaio, P. 1994. Slaves and Coloni in the Villa System. In Landuse in the Roman Empire, eds. J. Carlsen, P. Ørsted, and J. E. Skydsgaard, 145–158. Analecta Romani Instituti Danici, Supplementum 22. Rossiter, J. J. 1978. Roman Farm Buildings in Italy. BAR International Series S52. Rossiter, J. J. 1981. Wine and Oil Processing at Roman Farms in Italy. Phoenix 35, 345–361. Schubert, C. 1996. Land und Raum in der römischen Republik. Die Kunst des Teilens. Darmstadt: Wissenschaftliche Buchgeschellschaft. Schädler, U. 1998. Scavi e scoperte nella villa dei Quintilii. In La villa dei Quintilii. Fonti scritte e fonti igurate, ed. A. Ricci, 29–79. Roma: Lithos editrice. Scotoni, L. 1993. Deinizione geograica della campagna romana. Rendiconti. Atti della Accademia Nazionale dei Lincei IV:4, 647–667. Sfameni, C. 2004. Residential Villas in Late Antique Italy: Continuity and Change. In Recent Research on the Late Antique Countryside, eds. W. Bowden, L. Lavan, and C. Machado, 335– 375. Late Antique Archaeology 2. Shatzman, I. 1976. Senatorial Wealth and Roman Politics. Collection Latomus 142. Shiel, R. S. 1999. Reconstructing Past Soil Environments in the Mediterranean Region. In Environmental Reconstruction in Mediterranean Landscape Archaeology, eds. P. Leveau, F. Trément, K. Walsh, and G. Barker, 68–79. The Archaeology of Mediterranean Landscapes 2. Sirago, V. A. 1995. Storia Agraria Romana. Vol. I. Fase ascensionale. Napoli: Liguori. Sirago, V. A. 1996. Storia Agraria Romana. Vol. II. La dissoluzione. Napoli: Liguori. Slaska, M. 2002. Via Nomentana. Le fosse e le trincee agricole. Bullettino della Commissione Archeologica Comunale di Roma, 307. Smith, J. T. 1997. Roman Villas: A Study in Social Structure. London, New York: Routledge. Solin, H. 1975. Epigraphische Untersuchungen in Rom und Umgebung. Annales Academiae Scientiarum Fennicae B:192, 63–73. Rostovzeff, M. I. 1904. Pompeianische Landschaften und römische Villen. Jahrbuch des Kaiserlich Deutschen Archäologischen Instituts XIX, 10–126. Sorella, R. 1998. Dal Torraccio della Cecchina al Casale delle Vittorie. In Roma oltre le mura. Lineamenti storico topograici del territorio della V circoscrizione, ed. C. Calci, 156–175. Roma: Associazione Culturale “Roma oltre le mura.” Rostovzeff, M. I. 1911. Die hellenistisch-römische Architekturlandschaft. Mitteilungen des Deutschen Archäologischen Instituts, Römische Abteilung 26, 1–185. Soren, D., and N. Soren, eds. 1999. A Roman Villa and a Late Roman Infant Cemetery: Excavation at Poggio Gramignano, Lugnano in Teverina. Bibliotheca archaeologica 23. Rostovzeff, M. I. 1926. The Social & Economic History of the Roman Empire. Oxford: Clarendon Press. Spera, L. 1999. Il paesaggio suburbano di Roma dall’età tardoantica al medioevo. Il comprensorio tra le vie Latina ed Ardeatina dalle mura Aureliane al III miglio. Biblioteca Archeologica 27. Rua, H. 2009. Geographic Information Systems in Archaeological Analysis: A Predictive Model in the Detection of Rural Roman villae. Journal of Archaeological Science 36, 224–235. Sadori, L., and F. Susanna. 2005. Hints of Economic Change During the Late Roman Empire Period in Central Italy: A Study of Charred Plant Remains from “La Fontanaccia,” near Rome. Vegetation History and Archaeobotany 14, 386–393. Sallares, R. 2002. Malaria and Rome. A History of Malaria in Ancient Italy. Oxford: Oxford University Press. Santangeli Valenziani, R., and R. Volpe. 1980. Tentativo di ricostruzione di una sistemazione agricola di età repubblicana nei dintorni di Roma. Archeologia Classica XXXII, 206–215. Santangeli Valenziani, R., and R. Volpe. 1988. Via Ardeatina. Zona fra via di Grottaferrata, via di Vigna Murata, via Ardeatina (circ. XII). Bullettino della Commissione Archeologica Comunale di Roma, 544–559. Scheidel, W. 1994. Grain Cultivation in the Villa Economy of Roman Italy. In Landuse in the Roman Empire, eds. J. Carlsen, P. Ørsted, and J. E. Skydsgaard, 159–166. Analecta Romani Instituti Danici, Supplementum 22. Spera, L., and S. Mineo. 2004. Via Appia. Vol. 1 da Roma a Bovillae. Antiche strade, Lazio. Roma: Istituto Poligraico e Zecca dello Stato, Libreria dello Stato. Spurr, M. S. 1986. Arable Cultivation in Roman Italy, c. 200 BC – c. AD 100. Journal of Roman Studies Monographs 3. Stedman, R. C. 2003. Is It Really Just a Social Construction?: The Contribution of the Physical Environment to Sense of Place. Society & Natural Resources 16:8, 671–685. Steinby, E. M. 1981. Appendice a CIL XV,1. Bullettino della Commissione Archeologica Comunale di Roma, 55–88. Sterry, M. 2007. Searching for Identity in Italian Landscapes. In TRAC 2007: Proceedings of the Seventeenth Annual Theoretical Roman Archaeology Conference. London 2007, eds. C. Fenwick, M. Wiggins, and D. Wythe, 31–43. Oxford: Oxbow Books. Storey, G. R. 1999. Archaeology and Roman Society: Integrating Textual and Archaeological Data. Journal of Archaeological Science 7:3, 203–248. Swoboda, K. M. 1918. Römische und romanische Paläste. Eine architekturgeschichtliche Untersuchung. Wien: A. Schroll. Schipani, S., ed. 2008. Statistiche ambientali. Roma: Istituto Nazionale di Statistica. Syme, R. 1983. Spaniards at Tivoli. Ancient Society 13–14, 241–264. Schneider, K. 1995. Villa und Natur. Eine Studie zur römischen Oberschictkultur im letzten vor- und ersten nachchristlichen Jahrhundert. Quellen und Forschungen zur antiken Welt 18. Talbert, R. J. A. 2000. Barrington Atlas of the Greek and Roman World. Princeton: Princeton University Press. 246 bibliography Tarpin, M. 2002. Vici et pagi dans l’occident romain. Collection de l’École française de Rome 299. mosaico romano. Quaderni di archeologia etrusco-italica 8. Archeologia Laziale VI, 188–193. Tartara, P. 1988. Via Labicana. Località S. Maria (Torre Maura) (circ. VIII). I. Villa Romana. Bullettino della Commissione Archeologica Comunale di Roma, 408–414. Turner, S. D., and A. G. Brown. 2004. Vitis Pollen Dispersal in and from Organic Vineyards. I. Pollen Trap and Soil Pollen Data. Review of Palaeobotany and Palynology 129:3, 117–132. Tartara, P. 1999. Torrimpietra (IGM 149 I NO). Forma Italiae Regio I:XXXIX. Uggeri, A. 1802–1830. Journées pittoresques des édiices antiques dans les environs de Rome. Vol. I–XXIII. Taylor, R. 2000. Public Needs and Private Pleasures. Water Distribution, the Tiber River and the Urban Development of Ancient Rome. Studia archeologica 109. Valencia Hernández, M. 1991. Agricultura, comercio y ética. Ideología económica y economía en Roma (II a.e.–I d.e). Monografuas de historia antigua 7. Tchernia, A. 1986. Le vin et l’Italie romaine. Essai d’histoire économique d’après les amphores. Bibliothèque des Écoles Françaises d’Athènes et de Rome CCLXI. Valenti, M. 2003. Ager Tusculanus (IGM 150 III NE–II NO). Forma Italiae Regio I:XLI. Tessaro Pinamonti, A. 1984. Rapporti fra ambiente naturale ed ambiente architettonico nella villa romana del I sec d.C. in Italia. Rivista di Archeologia VIII, 48–67. Terrenato, N. 1992. La ricognizione della val di Cecina: l’evoluzione di una metodologia di ricerca. In Archeologia del paesaggio. IV ciclo di lezioni sulla ricerca applicata in archeologia. Certosa di Pontignana (Siena), 14–26 gennaio, 1991, ed. M. Bernardi 561–596. Firenze: All’Insegna del Giglio. Terrenato, N. 1996. Field Survey Methods in Central Italy (Etruria and Umbria). Between Local Knowledge and Regional Traditions. Archaeological Dialogues 3:2, 216–230. Terrenato, N. 1998. Fra tradizione e trend. L’ultimo ventennio (1975–1997). In L’archeologia degli italiani, M. Barbanera, 175–192. Roma: Editori Riuniti. Terrenato, N. 2000. Surface Thoughts: Future Directions in Italian Field Surveys. In The Future of Surface Artefact Survey in Europe, eds. J. L. Bintliff, M. Kuna, and N. Venclová, 21–28. Shefield Archaeological Monographs 13. Terrenato, N. 2001. The Auditorium Site in Rome and the Origins of the Villa. Journal of Roman Archaeology 14, 5–32. Terrenato, N. 2004. Sample Size Matters! The Paradox of Global Trends and Local Surveys. In Side-by-Side Survey: Comparative Regional Studies in the Mediterranean World, eds. S. E. Alcock and J. F. Cherry, 36–48. London: Oxbow Books. Thagaard Loft, G. 2004. Villa Landscapes in Pompeian Wall Painting – A Different Approach. Analecta Romana Instituti Danici XXIX, 7–27. Thomas, J. 2001. Archaeologies of Place and Landscape. In Archaeological Theory in Europe: The Last Three Decades, ed. I. Hodder, 165–186. London, New York: Routledge. Thomas, R. G., and A. Wilson. 1994. Water Supply for Roman Farms in Latium and South Etruria. Papers of the British School at Rome LXII, 139–196. Todisco, E. 2004. La percezione della realtà rurali nell’ Italia romana: i vici e i pagi. In Epigraia e territorio. Politica e società, ed. M. Pani, 161–184. Temi di antichità romane VII. Documenti e Studi 37. Tomassetti, G. 1910–1926. La Campagna Romana antica, medioevale e moderna. Vol. I–IV. Torelli, M. 1980. Industria estrattiva, lavoro artigianale, interessi economici: qualche appunto. In The Seaborne Commerce of Ancient Rome: Studies in Archaeology and History, eds. J. H. D’Arms and E. C. Kopff, 313–324. Memoirs of the American Academy in Rome 36. Torelli, M. 1990. La formazione della villa. Storia di Roma II, 123–132. Toynbee, A. J. 1965. Hannibal’s Legacy. The Hannibalic War’s Effects on Roman Life. London, New York: Oxford University Press. Treggiari, S. 1979. Sentiment and Property: Some Roman Attitudes. In Theories of Property: Aristotle to the Present, eds. A. Parel and T. Flanagan, 53–85. Waterloo: Wilfried Laurier University Press. Troccoli, M. G. 1984. Tivoli, Villa d’Este: scoperta di un Venetucci, B. P., ed. 1992a. Pirro Ligorio e le erme Tiburtine. Uomini Illustri dell’antichità I:1. Venetucci, B. P., ed. 1992b. Le erme Tiburtine e gli scavi del settecento. Uomini illustri dell’antichità I:2. Ventriglia, U. 1989. Idrogeologia della Provincia di Roma. Vol. II: Regione vulcanica Sabatina. Roma: Amministrazione provinciale di Roma, Assessorato LL.PP. viabilità e trasporti. Ventriglia, U. 1990a. Idrogeologia della Provincia di Roma. Vol. III: Regione vulcanica dei Colli Albani. Roma: Amministrazione Provinciale di Roma, Assessorato LL.PP. Viabilità e Trasporti. Ventriglia, U. 1990b. Idrogeologia della Provincia di Roma. Vol. IV Regione Orientale. Roma: Amministrazione Provinciale di Roma, Assessorato LL.PP. Viabilità e Trasporti. Vera, D. 1995a. Dalla ”villa perfecta” alla villa di Palladio: sulle trasformazioni del sistema agrario in Italia fra principato e dominato. (Prima parte.) Athenaeum 83:1, 189–211. Vera, D. 1995b. Dalla ”villa perfecta” alla villa di Palladio: sulle trasformazioni del sistema agrario in Italia fra principato e dominato. (Seconda parte.) Athenaeum 83:2, 331–356. Vicario, S. G. 1976. Notizie archeologiche nomentane. Rendiconti Atti della Accademia Nazionale dei Lincei XXXI:1– 2, 81–85. Vita-Finzi, C. 1969. The Mediterranean Valleys. Geological Changes in Historical Times. London: Cambridge University Press. Vita-Finzi, C., and E. S. Higgs. 1970. Prehistoric Economy in the Mount Carmel Area of Palestine: Site Catchment Analysis. Proceedings of the Prehistoric Society 36, 1–37. Volpe, R. 2000. Le ville del suburbio di Roma. In Aurea Roma. Dalla città pagana alla città cristiana. Catalogo della mostra (Roma, 22 dicembre 2000 – 20 aprile 2001), eds. S. Ensoli and E. La Rocca, 161–167. Roma: “L’Erma” di Bretschneider. Volpe, R., and A. Aronoldus-Huyzendveld. 2005. Interpretazione dei dati archeologici nella ricostruzione storica e ambientale del paesaggio suburbano: l’area di Centocelle nel suburbio sudorientale. In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B. Santillo Frizell and A. Klynne, 55–64. The Swedish Institute in Rome Projects and Seminars 2. Volpi, G. R. 1745. Vetus Latium profanum et sacrum. Vol. X:1–2. Walcot, P. 1975. Cicero and Private Property: Theory and Practice. Greece & Rome 22:2, 120–128. Wallace-Hadrill, A. 1994. Houses and Society in Pompeii and Herculaneum. Princeton: University Press. Wallace-Hadrill, A. 1998. The Villa as Cultural Symbol. In The Roman Villa. Villa Urbana, ed. A. Frazer, 43–53. University Museum Monograph 101, Symposium Series 9. Ward Perkins, J. 1961. Veii. The Historical Topography of the Ancient City. Papers of the British School at Rome VI. Weiss, R. 1969. The Renessaince Discovery of Classical Antiquity. Oxford: B. Blackwell. Wheatley, D., and M. Gillings. 2000. Vision, Perception and GIS: Developing Enriched Approaches to the Study of 247 bibliography Archaeological Visibility. In Beyond the Map: Archaeology and Spatial Technologies, ed. G. Lock, 1–27. NATO Science Series A: Life Sciences 231. White, K. D. 1970. Roman Farming. London: Thames & Hudson. Wilson, A. 1999. Deliveries extra urbem: Aqueducts and the Countryside. Journal of Roman Archaeology 12, 314–332. Wilson, A. 2000. Land Drainage. In Handbook of Ancient Water Technology, ed. Ö. Wikander, 302–317. Technology and Change in History 2. Widrig, W. M. 1980. Two Sites on the Ancient Via Gabina. In Roman Villas in Italy: Recent Excavations and Research, ed. K. Painter, 119–140. British Museum. Occasional Paper 24. Widrig, W. M. 1981. Two Villas and a Late Antique Horreum on the Via Gabina in Latium. American Journal of Archaeology 85, 224. Witcher, R. 2005b. The Hinterlands of Rome: Settlement Diversity in the Early Imperial Landscape of Regio VII Etruria. In Papers in Italian Archaeology VI: Communities and Settlements from the Neolithic to the Early Medieval Period. Proceedings of the 6th Conference of Italian Archaeology held at the University of Groningen, Groningen Institute of Archaeology, The Netherlands, April 15–17, 2003, eds. P. Attema, A. Nijboer, and A. Zifferero, 1045–1055. BAR International Series S1452. Witcher, R. 2006a. Settlement and Society in Early Imperial Etruria. Journal of Roman Studies 96, 88–123. Witcher, R. 2006b. Broken Pots and Meaningless Dots? Surveying the Rural Landscapes of Roman Italy. Papers of the British School at Rome LXXIV, 39–72. Yntema, D. 1993. In Search of an Ancient Countryside. The Amsterdam Free University Field Survey at Oria Province of Brindisi South Italy (1981–1983). Scrinum VI. Widrig, W. M. 1983a. Roma. Loc. Tor Bella Monaca: Excavations on the Ancient Via Gabina. Second Preliminary Report. Notizie degli Scavi di Antichità, 141–182. Zaccagni, P. 1984. Via Labicana. Tor Vergata (Circ. VIII). Programma e interventi di scavo in relazione all’insediamento della II Università di Roma. Bullettino della Commissione Archeologica Comunale di Roma, 89–91. Widrig, W. M. 1983b. Twelve Centuries of Occupation along the Via Gabina in Latium. American Journal of Archaeology 87, 269. Zanker, P. 1979. Die Villa als Vorbild späten pompejanischen Wohngeschmacks. Jahrbuch Deutschen Archäologischen Instituts 94, 460–523. Widrig, W. M. 1987. Land Use at the Via Gabina Villas. In Ancient Roman Villa Gardens, ed. E. B. MacDougall, 223–260. Dumbarton Oaks Colloquium on the History of Landscape Architecture X. Zanker, P. 1987. Augustus und die Macht der Bilder. München: Beck. Wiseman, T. P. 1970. Roman Republican Road-building. Papers of the British School at Rome XXXVIII, 122–152. Wiseman, T. P. 1987. Conspicui postes tectaque digna deo: The Public Image of Aristocratic and Imperial Houses in the Late Republic and Early Empire. In L’Urbs. Espace urbain et histoire (Ier siècle av. J.-C. – IIIe siècle ap. J.-C.). Actes du colloque international organisé par le Centre national de la recherche scientiique et l’École française de Rome (Rome, 8–12 mai 1985), 393–413. Collection de l’École française de Rome 98. Witcher, R. 2005a. The Extended Metropolis: Urbs, Suburbium and Population. Journal of Roman Archaeology 18, 120–138. 248 Zappi, G. M. ca. 1572–1583 [1920]. Annali e Memorie di Tivoli, ed. V. Paciici. Studi e Fonti per la storia della regione tiburtina I. Zarmakoupi, M. 2005, Villa Anguillara Sabazia. In Roman Villas Around the Urbs. Interaction with Landscape and Environment. Proceedings of a Conference at the Swedish Institute in Rome, September 17–18, 2004, eds. B. Santillo Frizell and A. Klynne, 151–158. The Swedish Institute in Rome Projects and Seminars 2. Zarmakoupi, M. 2006. The Roman Villa and its Cultural Landscape from the Late Republic to the Early Empire. In Common Ground: Archaeology, Art, Science and Humanities: The Proceedings of the 16th International Congress of Classical Archaeology, Boston 2003, eds. A. Brauer, C. Mattusch, and A. Donohue, 245–248. Oxford: Oxbow Books.