Nothing Special   »   [go: up one dir, main page]

Academia.eduAcademia.edu

Crossbow fibulae from Gamzigrad (Romuliana)

2010, Starinar

SOFIJA PETKOVI] Institute of Archaeology, Belgrade CROSSBOW FIBULAE FROM GAMZIGRAD (ROMULIANA) UDC: 904:739.2"652"(497.11) DOI: 10.2298/STA1060111P e-mail: spetkovi@ai.sanu.ac.rs Received: August 23, 2010 Accepted: November 30, 2010 Original research article Abstract. – In the site of Gamzigrad – Felix Romuliana 31 crossbow fibulae (Zwiebelknopffibeln) were found. Except the finds from large Roman necropolis, the crossbow fibulae from Gamzigrad compose a large collection from one site in Serbia. The most of them were found in the exactly stratified archaeological units of two horizons of life in Romuliana from the beginning of 4th to the middle of 5th century. Two kinds of analysis of the crossbow fibulae from Romuliana were made: morphological – typological and statigraphical – chronological. Conclusions about the function and production of the crossbow fibulae have been made according the results of these analysis. Key words. – Late Roman period, Tetrarchy, Gamzigrad, Felix Romuliana, Crossbow fibulae, Typology of fibulae, Function of fibulae, Production of fibulae. I n the site of Gamzigrad – Felix Romuliana thirty-one Roman, crossbow or »bulbous« fibulae (crossbow brooches, Zwiebelknopffibeln)1 have been found. Considering that archaeological explorations on this site have lasted for more then half a century, not a big number of fibulae have been discovered. However, with the exception of those types of finds from large Roman necropolises2, crossbow fibulae from Gamzigrad constitute a comprehensive collection from one site in Serbia. Furthermore, the majority of these finds originate from the exactly stratified archaeological units, divided into two horizons of life in Romuliana3: the first horizon of construction of the Galerius’ palace, from the beginning of the 4th century; and the second horizon of the Late Roman fortification of Romuliana from the last decades of the 4th and the first half of the 5th century. Even though this type of dating in Roman provinces in present-day Serbia is mostly based on analysis of fibulae from graves4, their chronology also highly depends on finds from reliable archaeological units, from settlements and fortifications, including Romuliana. Considering the specific function and chronological sensitivity of fibulae type 34, the finds examined in 1 Petkovi} type 34 (Petkovi} 2008 a, 392–467, table 10, map 16, T. LIII–LXXI). 2 The biggest number of crossbow fibulae from one archaeological site in Serbia comes from the necropolises of Viminacium (Rexi}, Rai~kovi}, Milanovi} 2006; Red`i} 2007, 65–73, type XXXVI; Petkovi} 2008a, 393; Spasi}-\uri} 2008). 3 For stratigraphy of cultural layers and horizons of life in Late Roman Romuliana see: Srejovi} 1983, 14–16; Jankovi} 1983a, 99 etc.; Jankovi} 1983b, 120 etc.; Srejovi}, Vasi} 1994, 56–59; Petkovi} 2004, 127 etc.; Petkovi} 2006; Petkovi} 2008c; Petkovi} 2008d; Petkovi} 2010a; Petkovi} 2010b. * The article results from the project: Urbanization and Transformation of the City Centres of Civil, Military and Residential Character in the Region of the Roman Provinces Moesia, Pannonia, Dalmatia (no 147001) funded by the Ministry of Science and Technological Development of the Republic of Serbia. 111 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) 1 STARINAR LX/2010 2 4 3 0 3 cm 5 6 7 8 Plate I – 1) Cat. 4, type 34 A 1; 2) Cat. 9, type 34 A 2; 3) Cat. 13, type 34 C 3b; 4) Cat. 20, type 34 D 2; 5) Cat. 22, type 34 D 2; 6) Cat. 25, type 34 D 2; 7) Cat. 27, type 34 D 2; 8) Cat. 28, type 34 D 2 Tabla I – 1) Kat. 4, tip 34 A 1; 2) Kat. 9, tip 34 A 2; 3) Kat. 13, tip 34 C 3b; 4) Kat. 20, tip 34 D 2; 5) Kat. 22, tip 34 D 2; 6) Kat. 25, tip 34 D 2; 7) Kat. 27, tip 34 D 2; 8) Kat. 28, tip 34 D 2 112 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) Gamzigrad complete the picture of Romuliana over the significant period of its history, from the end of the 3rd to the end of the 5th century. Two kinds of analysis of the crossbow fibulae from Romuliana were made: morphological-typological and stratigraphical-chronological. Conclusions about the function and production of the crossbow fibulae have been made according to the results of this analysis. The text is accompanied by a catalogue of analyzed finds. STARINAR LX/2010 Type 34 B fibulae in Serbia are present in Pannonia Secunda and Moesia Prima (Singidunum, Viminacium, Horreum Margi), whereas one silver specimen comes from Dardania (Zaskok near Uro{evac) (Map 1). A type 34 C fibula, which belongs to a rare variant of »the imperial fibulae«, was found in Gamzigrad.9 A bow with a transverse bar was hollow-mold cast from bronze, whereas other parts of the fibula, bulbs, foot and cylindrical pin holder were made from tin bronze and applied afterwards. Crossbow fibulae of this type are gilded, whereas punched and/or incised decoration on the bow is usually filled with niello. Type 34 C fibulae have a short, curved bar, plastic knobs shaped as massive bulbs on the head, a wide, short bow and a rectangular foot with a cylindrical pin holder, both of trapezoid cross-sections. Depending on decoration on the bow and foot the following sub-variants are identified: the first sub-variant (34 C 1) with a bow and foot decorated with a longitudinal strip with a »fir branch« motif; the second subvariant (34 C 2) with a bow decorated with geometrical (rhombs, circles with inscribed crosses) and vegetable (vine, rosettes) motifs or a »fir branch« motif combined with portrait medallions and/or metopes performed in the niello technique and a rectangular foot, decorated similarly as the bow with an impressed motif of peltae along the edges; the third sub-variant (34 C 3) decorated similarly as the previous, with foot-edges curved by a string of peltae motifs (cat. 13, Figs. 12–13). Among the type 34 C fibulae the most interesting are those with »imperial portraits«, one of which is a specimen from grave 6/06 from the Late Roman necropolis of Romuliana (Figs. 13–15).10 Type 34 C fibulae have been found at archaeological sites Sirmium, Singidunum, Viminacium, Idimum, 1.1 MORPHOLOGICAL-TYPOLOGICAL ANALYSIS Seven fibulae type 34 from Gamzigrad belong to an early, A 1 variant (cat. 1–7), characterized by a cylindrical transverse bar, of a circular or polygonal crosssection, pine-shaped plastic knobs on the head and a rectangular, faceted foot, which mildly narrows down at the end (Figs. 1–6, Pl. I, 1).5 Variant A 2 (cat. 8–9) differs from the previous one by a simple decoration incised on its bow or foot, occasionally filled with niello (Figs. 7–9. Pl. I, 2). Among the fibulae of this variant there are luxurious ones, such as a golden specimen from Romuliana discovered in a tomb of one of Galerius’ dignitaries (cat. 9, Fig. 8, Pl. I, 2).6 Type 34 A fibulae in Serbia mostly come from sites in Pannonia Secunda and the Danube limes of Moesia Prima and Dacia Ripensis (Map 1). Three fibulae from Gamzigrad belong to the type 34 B (cat. 10–12)7, characterized by a cylindrical transverse bar, of a circular or polygonal cross-section, occasionally with plated protuberances on both sides of the bow (Figs. 9–10), volutes, (Fig. 11), or bird protomes. There are large plastic knobs on the head in form of fluted pines (Figs. 10–11) or poppy pods (Fig. 9), the bow is of triangular or trapezoidal cross-section and the rectangular foot sometimes narrows at the end. According to the ornaments, the second sub-variant (34 B 2) was identified, with a foot decorated with punched circles or incisions (cat. 10–11, Figs. 9–10) and the third sub-variant (34 B 3), with a foot decorated with an impressed peltae motif (cat. 12, Fig. 11). Variant B fibulae were often luxuriously crafted, gilded and decorated with niello, especially the last subvariant, whose foot is decorated with two or three pairs of peltae. On some specimens there are inscriptions with a name of cesar, augustus, or vota and utere felix formulas, representative of propaganda during the rule of Licinius and Constantine I, at the beginning of the 4th century8. 4 Besides the aforementioned necropolises of Viminacium, those of Srem, Svilo{ and Be{ka are important for dating type 34, as well as the necropolis in Jagodin Mala in Ni{ (Jovanovi} 1975; Marijanski-Manojlovi} 1987; Dautova-Ru{evljan 2003). 5 Keller 1971, 32–35, type 1A, Abb. 11, 1; Pröttel 1988, 349–352, tip 1 A, Abb. 1, 3–5; Petkovi} 2008a, 395, type 34 A, T. LIII–LIV. 6 Petkovi} 2009, 353, Figs. 8–9. 7 Keller 1971, 35–37, type 2 A–B, Abb. 11, 3–4; Pröttel 1988, 353–357, type 2 A–C, Abb. 2, 1–5; Petkovi} 2008a, 395–396, type 34 B, T. LV–LVIII. 8 Noll 1974, 226 etc.; Ivanovski 1987, 81–90, Fig. 3, 1; Mirkovi} 1989, 39 etc.; Vasi} 2001a, 178 etc.; Vasi} 2001b, 93 etc. 9 Keller 1971, 41, Type 5, Abb. 11, 12, Pröttel 1988, 364–369, Type 5, Abb. 6; Petkovi} 2008a, 397, cat. 1420–1432, T. LIX–LX. 10 Laur-Belart 1959; Ivanov 1972; Jovanovi} 1976; Migotti 2008. 113 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 1. Vizi} (B) 2. Sremski Karlovci (B) 3. Sirmium (B) 4. ^alma (B) 5. Svilo{ (A, B) 6. Surduk – Rittium (B) 7. Be{ka (A,B) 8. Ora{ac ([abac) (A, B) 9. Glu{ci ([abac) (A) 10. [titar ([abac) (A) 11. Pri~inovi}i ([abac) (B) 12. U{}e (Obrenovac) (A,B) 13. Singidunim (B) 14. Ritopek – Castra Tricornia (A,B) 15. Brestovik (B) 16. Viminacium (A,B) 17. Sapaja – Translederata (B) 18. ^ezava – Castrum Novae (A) 19. Donje Butorke (B) 20. Rtkovo (B) 21. Romuliana (A, B) 22. Horreum Margi (B) 23. Braljina (Kru{evac) (A) 24. Naissus (B) 25. Zaskok (Uro{evac) (B) Type 34 (Tip 34) Map 1. Distribution of the early crossbow fibulae (type 34, var. A–B) in Serbia Au Golden and gilded fibulae / Zlatne i pozla}ene fibule Ag Silver fibulae / Srebrne fibule Karta 1. Rasprostrawenost ranih lu~nih krstobraznih fibula (tipa 34, var. A–B) u Srbiji section, decorated along the sides by incisions (Pl. II, 1–7), and a long foot, either trapezoid or rectangular adorned with facets and impressed »eyelets« (Figs. 16–25, Pl. I, 4–8). Motifs of impressed »eyelets« can be symmetrically distributes along the edge of the foot (Figs. 22–23, Pl. I, 4–5, Pl. II, 11–13) or grouped in pairs at the beginning and the end (Figs. 16–21, Pl. I, 6–8, Pl. II, 8–10). Analysis of different decorative patterns of »eyelets« on feet of this variation of fibulae has not shown any chronological differentiation among them (Table 2). However, there is a possibility that different distribution of »eyelets« had a symbolic value, meaning that the decoration on the Ravna–Campsa, Prahovo–Aquae, Romuliana, ]i}evac, Naissus and Ulpiana (Map 2).11 A discovery of three specimens in the hoard of bronze artifacts found in Ni{ Fortress, stored around 378, as well as finds in graves in Viminacium, from the »Pe}ine« necropolis, and Late Roman necropolis in Jagodin Mala in Ni{, are important for chronological determination of this type of crossbow fibulae in Serbia. Based on the archaeological context, the fibulae can be dated to a period between the second third, and the end of 4th century. In historical context that is the period of dynasties of Roman emperors Constantine I and Valentine I, lasting from the death of Constantine I until the death of Theodosius I, i.e. from 337 to 395.12 The largest number of crossbow fibulae from Romuliana, eighteen of them (cat. 14–31), belong to type 34 D 2,13 characterized by a massive, triangular transversal bar curved on both sides of the bow with volutes or schematized bird protomes, distinct bulb-shaped plastic knobs on the head, a bow of triangular or trapezoid cross- 11 Petkovi} 2008a, 397, map 16; Popovi} 2009a. Petkovi} 2008a, 397, 517, table 10. 13 Keller 1971, 37–41, type 3 B and type 4 A, Abb. 11, 6, 9; Pröttel 1988, 357–364, type 3/4 B, Abb. 4a, 3–5; Petkovi} 2008a, 397, type 34 D 2, T. LXII–LXIII, LXIV, 1–2. 12 114 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 1. Sirmium 2. Singidunum 3. Viminacium 4. Ravna - Campsa 5. Prahovo - Aquae 6. Romuliana 7. Medve|a - Idimum 8. ]i}evac 9. Naissus 10. Ulpiana Map 2. Distribution of the »imperial« crossbow fibulae (type 34, var. C) in Serbia Karta 2. Rasprostrawenost »carskih« krstobraznih fibula (tipa 34, var. C) u Srbiji Type 34 C (Tip 34 C) 1. Variants A and B fibulae dated to the Tetrarchic period and the time of Flavian dynasty of Constantine I, mainly around the second half of the 3rd – the first half of 4th century.17 2. Variants C 3 and D 2 fibulae dated to the time of Valentinian dynasty until the battle of Hadrianopolis, or foot marked the rank or branch in the military or administrative hierarchy. Variant D fibulae stand second in prevalence among crossbow fibulae in Serbia and constitute over a quarter of all type 34 finds, i. e. 23.14%. They are distributed over all Roman provinces in Serbia, both in the Danube limes and further inland. According to analysis of grave units containing finds of this variant of fibulae, mostly from the necropolises in Be{ka, Svilo{ and Viminacium, use of type 34 can be limited to a two-decade period: from rise to power of Valentinian I and Valens until the big collapse of the Roman army in the battle of Hadrianopolis, i. e. the start of Theodosius’ I rule.14 Analysis of type 34 D fibulae from Roman provinces on the territory of Bulgaria confirmed the same chronological framework.15 Crossbow fibulae of this variant from Salona have a somewhat broader chronological framework.16 According to the typological analysis of type 34 variants of fibulae from Romuliana, two groups of such finds can be identified. 14 Petkovi} 2008a, 517, table 10, map 16. Haralambieva 1990, pp. 86–88, type 3 A, Figs 3, 4–6, T. I, 4–6, T. II, 1–3, 6; Gen~eva 2004, 64, type 24d, T. XXIII, 1. 16 Iv~evi} 2000, 149–158, cat. 32–36, T. VI–VIII – the author dates crossbow fibulae of this type from Salona according to the typologies of E. Keller (see Keller 1971, 34–36, Typ 3, Abb. 11, 6–8) and F. M. Pröttel (see Pröttel 1988, 357–359, Type 3/4 A–B, Abb. 4a): one in a 310–340 period, five artifacts in 340–390 and nine in 340–360 period. It can be concluded that the most type 34 D fibulae originate from the middle of 4th century. 17 Finds of type 34 A in Serbia are dated to the time of the First Tetrarchy until the end of joint rule of Lucius and Constantine I, from 293 to 324 AD (Petkovi} 2008a, 395, 517, table 10), a type 34 B to the first half of the 4th century (Petkovi} 2008a, 396, 517, table 10). 15 115 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 Table 1. Chronological table of the types of crossbow fibulae from Gamzigrad (Romuliana) accordint to dated archaeological units Tabela 1. Hronolo{ka tabela tipova krstobraznih fibula sa Gamzigrada (Romuliana) prema datovanim arheolo{kim celinama 116 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 crossed over the stomach.24 There was a military belt (cingulum militae) with a bronze buckle laid by the side of his lower left leg, together with a bag made of organic material, either leather or cloth (decomposed), containing his personal belongings which survived: iron metal, flint and seven bronze coins. One coin belonged to the mintage of Emperor Constans between 340 and 350 AD, whereas the other six were minted during the reign of Emperor Valens, from 357 to 378.25 Analysis of grave finds laid next to the lower left leg of the deceased indicates that the grave should be dated to the last quarter of the 4th century: The bronze buckle of the military belt had a rectangular plating decorated by faceting, and a spike with the end shaped as a snake’s head, which exceeds its frame. The rectangular frame of the buckle, decorated with incised grooves along its edges and notches on the rims, was fastened on the strap by three rivets. This type of buckle is dated to the last quarter of the 4th or the first half of the 5th century.26 the beginning of the reign of Theodosius I, mainly during the last third of the 4th century.18 1.2. STRATIGRAPHICAL-CHRONOLOGICAL ANALYSIS Based on the stratigraphy of cultural layers on Gamzigrad and conditions of finds, i. e. archaeological units, the both typological groups of crossbow fibulae from Romuliana have been analyzed (Table 1). Among the total number of crossbow fibulae (31) only four specimens have not been found in a definite archaeological unit. Those are two type 34 A 1 fibulae, accidental finds from an extra muros area of Galerius’ fortified palace (cat. 6–7), and two type 34 D 2 fibulae, discovered during initial excavation campaigns in Gamzigrad (cat. 15–16). The rest of twenty-seven fibulae mostly come from systematic archaeological excavations in the fortified imperial palace. Several were found during research of a sacral-memorial complex on Magura (two specimens, cat. 4–5) and trench excavations outside the fort’s walls (three specimens, cat. 8–9, 13). Dating of archaeological units was given in accordance with currently accepted cultural stratigraphy of the archaeological site of Gamzigrad.19 Archeological units, explored before the 2002 excavations, were incorporated into the existing stratigraphy of cultural layers, based on documented data from archaeological excavations on this site.20 Particularly relevant are the grave finds of two type 34 fibulae, which come from the Late Roman necropolis of Romuliana, explored between 2005 and 2006, south of the fortified palace, because the grave units (the tomb was explored in 2005 and the grave 6/06 in 2006) contained other finds, including money. The type 34 C 3 specimen decorated with »imperial portraits«, found in grave 6/06,21 belongs to the second typological group of crossbow fibulae from Gamzigrad (cat. No. 13, Figs. 12–15, Pl. I, 3).22 A rectangular pitgrave had been dug along the outer side of foundation of the southern rampart of Galerius’ palace, in the area between polygonal towers 13 and 15. The grave is oriented in the west to east direction, and its left (northern) longitudinal line is a part of the foundation of the southern rampart. Even though stratigraphical situation of this area is not quite clear,23 it is certain that the layer from which the grave had been dug was above the foundation zone of the southern rampart, i.e. that the later fortification of Romuliana had already existed at the time of the burial. An adult man had been buried in an extended position, lying on his back, with arms bent at the elbows and 18 Petkovi} 2008a, 397, table 10. Even though type 34 C fibulae cover a wider chronological span, sub-variant 3, which the specimen from Gamzigrad falls under, whose foot-edges are decorated with a line of peltae, is dated to the Valentinian period. 19 See fn. 3. However, it should be taken into account that not all movable finds from years of systematic research have been processed, most importantly, numerous ceramic materials. Publication of a full archaeological database from Gamzigrad should complete, or perhaps, change the present cultural stratigraphy of this multi-layered site. 20 Documentation of the Archaeological Institute in Belgrade, Documentation of the National Museum in Zajecar. 21 Petkovi} 2009, 266–267, Fig. 41–47; @ivi} 2009, 284–285, Cat. 50, Pl. X. 22 See fn. 18. 23 In 1996, layers of rubble waste from the outside area of southern rampart were removed to the assumed level from the period of Galerius’ palace by construction machines. On that occasion solid layer of crumbed waste, made of stone, tegulae fragments and pestled limestone plaster, was formed against the rampart by bulldozers. This 20 cm thick coating forms a recent layer, which closes the pit of grave 6/06. That recent layer was, inaccurately, associated by dr. Gerda von Bülow with the level of construction of southern rampart and digging of a »fortification trench« along its side, and it was concluded that the burial in grave 6/06 is terminus post quem for the construction of later fortification of Romuliana (v. Bülow, Schüler 2009, 236). 24 I am grateful to my colleague Dr. Nata{a Miladinovi}-Radmilovi} for preliminary anamnesis of the deceased. 25 Vasi} 2009, 313. 26 Sommer 1984, 23, Taf. 61, Anm. 55; Gomolka-Fuchs 1982, 174, cat. 230; Gomolka-Fuchs 1991, 187, cat. 773; Gomolka-Fuchs 1995, 85, Taf. 1, 16–17; Uenze 1992, 175, Taf. 9, 24; Tejral 1997, 323, Abb. 1, 1, 8. 117 STARINAR LX/2010 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (###–###) 1 2 8 3 9 4 5 10 11 6 12 7 13 Plate II – Ornamental motives on fibulae of type 34 D 2 from Gamzigrad (Romuliana): 1) motive 1 on the bow; 2) motive 2 on the bow; 3) motive 3 on the bow; 4) motive 4a on the bow; 5) motive 4b on the bow; 6) motive 5 on the bow; 7) motive 6 on the bow; 8) motive 1 on the foot; 9) motive 2 on the foot; 10) motive 3 on the foot; 11) motive 4a on the foot, 12) motive 4b on the foot; 13) motive 5 on the foot Tabla II – Ukrasni motivi na fibulama tipa 34 D 2 sa Gamzigrada (Romuliana): 1) motiv 1 na luku; 2) motiv 2 na luku; 3) motiv 3 na luku; 4) motiv 4a na luku; 5) motiv 4b na luku; 6) motiv 5 na luku; 7) motiv 6 na luku; 8) motiv 1 na stopi; 9) motiv 2 na stopi; 10) motiv 3 na stopi; 11) motiv 4a na stopi; 12) motiv 4b na stopi; 13) motiv 5 na stopi orated along its rims by an incised string of three pairs of peltae and volutes, on the beginning and the end, and along the middle by an inlaid strip of tin silver with a decoration in niello technique. The entire fibula was gilded except for the silver strip on the foot. In the centre of the decoration on this strip there was a square metope with an »imperial portrait« surrounded by a multiple »fir branch« motif. On the beginning and the end of the bow there were incised rectangular metopes with »imperial portraits«, which might have been filled with The rectangular iron metal was not sufficiently preserved for its type to be precisely established, but together with the flint pointed to the »barbarian« EastGerman population, whose presence in Dacia Ripensis should not have been expected before the end of the 4th century.27 The numismatic finds suggest the year 367 as terminus ante quem non for the burial in grave 6/06. However, the money was not laid in the grave as a tribute; it was the property of the deceased, which he had carried along in the bag on his belt. Therefore, the last year of mintage is the most probable approximate time of death.28 A crossbow fibula was found on the right shoulder of the deceased, foot pointing upwards, in a position used for fastening a military cloak (paludamentum), made of white woolen cloth, judging by the threads preserved on the inner side of the bow and on the corroded iron pin of this brooch. A long rectangular foot was dec- 27 During the period after the battle of Hadrianopolis 379–381, no later than 383, Theodosius I gave receptio to groups of »barbarians«, Goths, Huns, Alans in diocese of Dacia (Burns 1994, 43–72). 28 Burial in the grave 6/06 along the southern rampart can be dated to a period after the battle of Hadrianopolis, in time of conflict with »barbarians« on the territory of Dacia Ripensis (see fn. 27). 118 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) niello. The iron pin, which was damaged and corroded, had been additionally applied.29 Crossbow fibulae, embellished with portraits were a sign of military and official honor. They were presented personally by the emperor during the state holydays (dies imperii) and the ruler’s anniversaries (vota).30 They were manufactured in imperial workshops, which traveled throughout the empire together with the emperor.31 Earlier, the fibulae of this type had been interpreted as means of propaganda for the Flavian dynasty of Constantine I and his heirs.32 Dr. Branka Miggoti concluded, in her monography dedicated to these types of finds, that after considering all known theories of portrait interpreting on type 34 C fibulae, it was impossible to give their definite interpretation.33 In a new study of these fibulae, after analyzing their ornaments, Dr. Ivana Popovi} asserts that medallions with portraits (imagines clipetae) represent one of the favorite motifs of the Late Roman style, witnessed in mosaics, frescos, silverware and glassware, luxurious belt sets, and objects made of ivory. According to the author, the portraits represent allegories and lesser deities.34 However, it is unlikely that the portraits on the official fibulae, manufactured at imperial workshops, had a simple decorative character; especially if other decorative motifs, connected with the cult of Dionysus (fir branch, vine, ivy leaf), or early Christian symbols (cross and Christogram) are taken into account. I assume that ornaments on type 34 C 3 fibulae carried an explicit ideological and political message of their time, the meaning of which eludes us over a distance of millennium and a half. A specimen from the village of Kolarci in Bulgaria is interesting in terms of portrait interpretation. On the end of a foot, whose edges have been lined with peltae, or more precisely on the pin holder of trapezoidal crosssection, there was only one engraved portrait.35 A ¾ profile of a man was shown, turned to his left, wearing a pageboy haircut and an militry overcoat (paludamentum), fastened by a flat, discoid fibula on the right shoulder, framed on both sides by leafy vines. As mentioned before, the decoration in form of incised pairs of peltae on of the foot on this fibula, as well as on the specimen found in Gamzigrad, is no older than the last third of the 4th century.36 It can be concluded that the individual was an emperor considering the circular, plate fibula shown on his right shoulder.37 In that case, »the emperor« shown on the fibula from Kolarci ruled alone. That could have been the depiction of Julian the Apostate (361–36.), from the second half of the 4th century, whereas the Dionysian motif of vine is understandable in the context of pagan restoration attempt during his short rule. An STARINAR LX/2010 almost identical portrait was shown in a rectangular metope on the foot of the fibula from Gamzigrad, decorated along its length by a motif of a double »fir branch«. On the beginning and the end of the bow there were two more portraits in rectangular metopes, and there were three visible circular medallions with crossing lines (Christograms?), which further complicate the interpretation by suggesting Christian symbolism.38 Finally, it is possible that portraits on type 34 C 3 fibulae are cult depictions (Dionysus, allegories and/or personification, Christ, the Apostles). Combinations of pagan and Christian iconographic symbols that appear on this type of fibulae are common in the 4th century, during the establishment of Christianity and intensive Christological debates. Equally, medallions and metopes with portraits on these fibulae represent divinized portraits of reigning emperors or deceased rulers of the same dynasty. That is suggested by nimbuses on some of the portraits.39 Dionysus shown on the mosaic floor of Triclinium of Palace I of Romuliana, who can probably be identified with a divinized emperor, also has a blue nimbus, as well as analog depictions on third-century mosaics that could have been a model for this motif.40 Likewise, Constantine I solidi, minted in 316/317 in Ticinum and Siscia, show an emperor with nimbus as Sol Invictus.41 In any case, a nimbus can be understood as 29 Pins on the crossbow fibulae made of non-ferrous metals, gold, silver or bronze, were often replaced due to bending and breaking caused by tightening of heavy military coats. 30 Laur-Belart 1959, 68; Keller 1971, 44; Jovanovi} 1976, 48–49; Migotti 2008, 61–67. 31 Vasi} 2001a, 195–197; Migotti 2008, 69–71. 32 Laur-Belart 1959, 64 etc.; Ivanov 1972, p. 9 etc.; Jovanovi} 1976, 2 etc. 33 Migotti 2008, 16–22. 34 Popovi} 2009a, 108–106. 35 Ivanov 1972, 21, Fig. 15 – author regards this portrait as divinized representation of Constantine I. 36 Crossbow fibulae of late type 34 F variant, which can be dated in the end of 4th and the first half of the 5th century, have a foot with this kind of decoration (Petkovi} 2008a, p. 400, T. LXII, 1–2). 37 Janes 1998, 388–391. 38 Dr. Branka Migotti notices strong Christian symbolism in later specimens of fibulae type 34 C (Migotti 2008, 38–39). 39 Portrait of a male in toga (togatus) at the beginning of the bow of the fibula from Romuliana has a nimbus, undoubtedly just like the portrait at the end of bow of the fibula from Drnovo (Petru, Petru 1978, T. IV, 3) as well as the one on the elevated end of the foot of the specimen from the British Museum (Laur-Belart, Abb. 44, 2 a–b; Migotti 2008, T. 4, E 14). 40 Jeremi} 2006, 52, Abb. 8–9. 41 Popovi} 2010a, 150–151, Fig. 3. 119 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 or on ivory diptychs, like The Diptych of Stilicho from Monza and diptychs from Louvre and Berlin.48 The prevalence of fibulae of second chronological group in relation to dated archaeological units is interesting. In archaeological units dated to the last quarter of the 4th century, five fibulae were discovered (18.52%), three fibulae in units dated between the end of the 5th century and the beginning of the 5th century (11.11%), 7 fibulae from units from between last quarter of the 4th to the first half of the 5th century (25.93%) and four fibulae in units from the first half of the 5th century (18.81%). This ratio points to the fact that the largest number of crossbow fibulae from Romuliana belong to the horizon of settlements which appeared after the abandonment of construction of the imperial palace, in the last third of the 4th century, and that they were equally discovered in units from the last quarter of the 4th and first half of the 5th century. I stress that fibulae of the second chronological group are absent in archaeological units from the end of the 3rd to the beginning of the 4th century (Table 1). Until today, no crossbow fibulae have been found in archaeological units dating from the 6th and the beginning of the 7th century in Gamzigrad. In addition to that, I emphasize that the Early Byzantine horizon of Romuliana, which spans between the end of the 5th and the end of the 6th/beginning of the 7th century, contains ample archaeological finds, including bronze and iron fibulae. a sign of divinization. Furthermore, the imitatio deorum motif is frequent on portraits of Roman emperors shown on objects used as means of political propaganda, such as money and jewelry.42 Judging by the manner of production of fibula from grave 6/06 from Gamzigrad, decorations with multiple, incised peltae on the foot, as well as the style of depicted »imperial portraits«, I assume that it was made during the last third of the 4th century.43 In case the fibula was made in one of the imperial workshops so that it would be presented during a special occasion, and if the portraits represented emperors of the Valentinian dynasty, two possibilities occur: that the depicted emperors were Valentinian I, Valens and Gratian, if the fibula was made in honor of Gratian’s ascension to the throne in 367, or that it was Valens, Gratian and Valentinian II shown in the metopes, if the reason was that the latter was promoted to Caesar. Two typological groups of crossbow fibulae from Romuliana generally fit into two chronological groups of these finds, determined by analysis of archaeological context they were found in (Table 1).44 Two chronological groups of type 34 fibulae in Gamzigrad are: – a group of crossbow fibulae dated between the end of 3rd to the beginning of the 4th century; – a group of crossbow fibulae dated between the last quarter of the 4th to the first half of the 5th century. The first chronological group is made from fibulae of the first typological group, discovered in the archaeological units dated between the last decades of the 3rd and the first quarter of the 4th century: types 34 A 1, 34 A 2, 34 B2 and 34 B 3 (cat. 1, 4–5, 8–12). The second chronological group contains fibulae of type 34 D 2 (cat. 14–28), i. e. type 34 D (cat. 29–31), dated to the last third of the 4th century, one type 34 C 3 b fibula from a grave dating between the end of 4th to the beginning of the 5th century (cat. 13), and two earlier specimens of type 34 A 1, discovered in Tower 19 (cat. 2–3). It is possible that a certain percentage of earlier types of crossbow fibulae (11.11% in Gamzigrad) were found in later archaeological units, because those objects symbolized status in the public service, army or administration, and they were awarded for certain merits,45 so their owners kept and worn them for years. It was common to be buried with them,46 as well as showing them on sarcophagi, stelas or frescoes in tombs, wearing a crossbow fibula on the right shoulder (Fig. 26).47 Likewise, highranking officials and military leaders were shown on artistic depictions wearing crossbow fibulae on the right shoulder, like on Teodosius’ obelisk in Constantinople, 42 Popovi} 2001, 377 etc. For the way of manufacturing see: Petkovi} 2008a, 394–395, 511; for decoration on the long foot in form of a line of incised peltae, see: Buora 1997, 249; for stylistic analysis of portraits and »haircut« see: Migotti 2008, 54–56. 44 By archaeological context I imply stratigraphy of cultural layers and layers of life, which refers to a location of each fibula, possible closed unit (grave, pit, stove etc.) and analysis of other archaeological finds from that unit. In horizons of Late Roman Romuliana, most of the cultural layers also represent closed units, because they are situated between two undisturbed levels (floors of limestone plaster or firmly packed earth). Besides that, the majority of layers of life perished in fire, so they are sealed with conflagration layer of burned soil, carbonated wood, soot and ash (Petkovi} 2004, 127–153; Petkovi} 2006, 29–45; Petkovi} 2008c, 61–63; Petkovi} 2008d, 64–67). 45 Zahbelicky 1980, 1099–1101; Theune-Grosskopf 1995, 84–89; Janes 1998, 387–388. 46 Hence the most of known crossbow fibulae from Serbia comes from necropolises (see fn. 2 and 4). 47 Zahbelicky 1980, 1101–1103, cat. 1, cat. 4, cat. 5; TheuneGrosskopf 1995, 83–87; Abb. 55, Abb. 59; Iv~i} 2001, 165, Fig. 1; Jovanovi} 2007, 112–113, Figs. 15, 6; Pop-Lazi}, 166–167, Fig. 7. 48 Zahbelicky 1980, 101, 103–104, cat. 1, cat. 7–8; Pröttel 1988, 369, Abb. 9, 1; Theune-Grosskopf 1995, 91, Abb. 57, Abb. 60, Abb. 64–65. 43 120 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) 2. CONCLUSION: FUNCTION, PRODUCTION AND WORKSHOPS STARINAR LX/2010 and civil wars between Constantine and Maxentius. A find which speaks in favor of this theory comes from a burial in a tomb in Generala @danova street No. 11 in Belgrade (Singidunum). An adult man was buried with a type 34 B 1 fibula, together with Maximianus Herculius’ and Galerius’ money from 295–296, and six Diocletian’s coins minted between 295–298.56 Even though type 34 B fibulae from Romuliana are of modest design, similar crossbow fibulae of luxurious design appear in Serbia, gilded, silver or decorated with niello, especially a sub-variant with an ornament on the foot which consists of two or three pairs of punched peltae (Type 34 B 3). Luxurious specimens of type 34 B were mostly found in provinces of Pannonia II and Moesia I in Dardania, one gilded fibula, which originated from Naissus, and a silver one from a burial of Late Roman necropolis on the site of Zaskok near Uro{evac (Map 1).57 They could have been produced at the imperial workshop in Trier, considering that this type of luxurious fibulae had been the means of Constantine’s political propaganda since 306 when he was declared emperor.58 In that case, the number of them found in Pannonia Secunda and Moesia Prima could testify about Constantine’s territorial ambitions towards Illyricum, until the decisive victory over Licinius at Cibalae in 316. At the same time, early type 34 A–B of crossbow fibulae, as well as other artifacts from precious metal had The collection of crossbow fibulae from Romuliana facilitates the functional determination of these kinds of finds. Due to uniqueness of the multi layer settlement on Gamzigrad it was possible to follow the development of this type of fibulae. The earliest specimen belongs to the early 34 A 1 variant, which developed from hinge T–fibulae. They were discovered in appropriate archaeological units of the Tetrarchic period: in the area of a consecrative-memorial complex on Magura (cat. 4–5), in a tomb explored in 2005 south of the fortified palace (cat. 9) as well as in the earliest layers at the location of eastern and western gate of the later fortification (cat. 1 and 8). The fibulae from Magura and the golden specimen from the tomb are of special importance, because they were discovered together with money and other artifacts, precisely dated between the end of the 3rd and the beginning of the 4th century.49 Since no type 34 fibulae had been found in archaeological units of the horizon of the earlier Roman settlement, which had existed before construction of the imperial palace in the 3rd century,50 I assume that the beginning of their use can be dated to the transition from the 3rd to the 4th century. Historically speaking, it is a time of the First Tetrarchy (293–305), when administrative and military reforms took place besides other Diocletian’s reforms of the Roman Empire. The collection of crossbow fibulae from Romuliana clearly indicates the connection with the reforms mentioned previously. This type of fibulae had been carried by the members of imperial administration and military as a sign of rank and status.51 Golden and silver specimens of crossbow fibulae, characteristic for the period of the First Tetrarchy, had been presented by the emperor personally, and had been a sign of high-ranking in military and administrative hierarchy.52 At this time the emperors also wear crossbow fibulae as a sign of rank, as can be seen on the depiction of Galerius on a relief on a triumphal arch in Salonika, and some sculptural depictions of Tetrarchs.53 In archaeological units dated to the beginning of the 4th century, which originate from the time of construction of the imperial palace,54 three bronze type 34 B fibulae, characteristic for the first half of the 4th century (cat. 10–12) were found. This type of crossbow fibulae had been manufactured during the reign of Constantine I and his dynasty.55 Specimens from Gamzigrad show that their production could have begun during the first decade of the 4th century, at the time of the Second Tetrarchy 49 Petkovi} 2009, 253–261, Figs. 8–9, 12; Vasi} 2009, 309; Popovi} 2009b; Bori}-Bre{kovi} 2009; Popovi} 2010b, 156–158, Fig. 131. 50 Petkovi} 2010a. 51 Theune-Grosskopf 1995, 84–89; Janes 1998, 388–390. 52 Diaconescu 1999, 205–217 , Abb. 1, 1–2, Abb. 2,1, Abb. 4, 3, etc. 53 Theune-Grosskopf 1995, 86, Abb. 58; Tomovi} 1997, 420–421, Figs. 1–2, 5–6. 54 Te later fortification of Romuliana, or representative ramparts and towers of Galerius’ palace, were erected at the beginning of the 4th century, but they were not finished (Petkovi} 2004, 138–140, Fig. 8, T. IX, 1–2; Petkovi} 2006, 32, 40, Fig. 4). Construction of the imperial palace was abandonned no later than 316, when Constantine I won the battle over Licinius on the Cibalae and overtook his territory in the western Illyiricum (Vasi} 2008, 12–13). 55 Petkovi} 2008a, 396, 517, table 10. 56 Simi} 1997, 36, T. II, G. 32, 1. 57 Petkovi} 2008a, 396–397, cat. 1409–1410, 1412–1415, T. LVII, 1–2, T. LVIII, 1–2. 58 This type of luxurious fibulae, decorated with pairs of peltae on the foot, were dated between 308/9 and 321/2 by Dr. Miloje Vasi} (Vasi} 2001a, 180–191). The same author thinks that this type of fibulae of luxurious design, with inscriptions dedicated to Maximian Hercules and Constantine I, was manufactured in short period between years 306 and 316 (Vasi} 2001b, 93–105). 121 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 Decorative motifs on bows of type 34 D crossbow fibulae: Decorative motifs on feet of type 34 D crossbow fibulae: 1. Longitudinal groove with transverse notches (T. II, 1) 2. Longitudinal groove with diagonal notches (T. II, 2) 3. »Fir branch« (T. II, 3) 4a. Braid (T. II, 4) 4b. Braid with a longitudinal line of impressed rhombs (T. II, 5) 5. Longitudinal groove with a line of impressed peltae (T. II, 6) 6. Longitudinal line of lying S spirals (T. II, 7) 1. At the beginning of the foot 1, – at the end 2 pairs of »eyelets« (T. II, 8) 2. At the beginning and the end 2 pairs of »eyelets« (T. II, 9) 3. At the beginning the foot 2 – at the end 3 pairs of »eyelets« (T. II, 10) 4a. Along each rim of the foot a line of »eyelets« (T. II, 11) 4b. Along each rim of the foot a line of »eyelets« connected by a tangent into a flowing spiral (T. II, 12) 5. Along each rim of the foot a line of alternately impressed circles and triangles (T. II, 13) Table 2. Table of the ornamental motives of crossbow fibulae of type 34 D 2 from Gamzigrad (Romuliana) Tabela 2. Tabela ukrasnih motiva krstobraznih fibula tipa tip 34 D 2 sa Gamzigrada (Romuliana) 122 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 for contributions to the imperial service. In those terms, it is interesting to consider the possibility of forging type 34 D crossbow fibulae, suggested by the poor production of some specimens from Romuliana (Figs. 18–20, Figs. 23–24),63 as well as exuberant decorations on others, which is a sign of »barbarian taste« (Fig. 20, Fig. 24, T. I, 7–8, PL. II, 4–6, 12).64 This begs the thought that type 34 D 2 crossbow fibulae were manufactured at local workshops, perhaps within a local army unit, exclusively for its needs. It explains the poor design quality of some artifacts, or too many decorations on others, with use of a large number of techniques and motifs (incising, punching, impressing). Namely, the craftsman was a member of a local army unit, probably of »barbarian« origin, with a limited skill and knowledge of metallurgy. However, the meaning of crossbow fibulae remains the same from the end of the 3rd to the middle of the 5th century: they are symbols of engagement in state, i.e. imperial service. The person carrying a crossbow fibula works in service and under the patronage of the Roman Empire and the emperor himself, regardless of whether they are a member of administration, the army, or a group of »barbarians« bound to the Empire by a contract. (receptio, foedus).65 The lack of type 34 fibulae in the sixth-century Romuliana contradicts the opinion that crossbow fibulae had been worn by members of the army, and administrative elite of Justinian’s era.66 This view is based on artistic depictions of Emperor Justinian I and Empress Theodora with their escort, on mosaics in the San Vitale basilica in Ravenna, in which dignitaries wear robes (chlamys) fastened on the right shoulder by this type of been produced in workshops in Sirmium and Naissus, for the sake of Licinius’ propaganda.59 Based on all cited examples, I assume that type 34 A and 34 B fibulae had been worn by members of the imperial army and administration, according to clearly established rules that resulted from Diocletian’s reforms. On the other hand, that is not the case for the most numerous group of crossbow fibulae from Gamzigrad, type 34 D 2. The archaeological context, a broad chronological span, ranging from the end of the 4th until the end of the 5th century, of units where different decorative motifs were discovered, do not suggest a clearly defined official character of these fibulae. The attempt of classification and analysis of decorations incised on the bow or foot of type 34 D 2 specimens from Romuliana, do not lead to specific conclusions about their function (Pl. II, Table 2). They had been worn probably by members of military units stationed at the fortification of Romuliana. I assume that those were smaller squads of light cavalry auxiliary units.60 A question remains whether these fibulae were a mark of army rank, or an army branch, or were they awarded for special merits. The already mentioned, luxurious type 34 C 3 fibula from grave 6/06 represents an antithesis to numerous specimens of type 34 D 2. Crafted at an imperial workshop in the last third of the 4th century, and presented on a special occasion, it definitely testifies to a promotion of a person it had been given to. In that sense, this fibula represents the ornatus, but from the moment of donation it also marks a high status of its owner (ornamentum dignitatis). The owner of this fibula, symbolically buried along the foundation of the southern rampart of Romuliana, was a high-ranking official of the Roman Army, or perhaps a commander of the squad stationed in Gamzigrad. Workshops in which crossbow fibulae from Romuliana were crafted can be identified according to their official character. Luxurious specimens (cat. 9 and 13) were manufactured at imperial workshops, most likely in Naissus or Sirmium.61 Early type 34 A and 34 B fibulae are also presumed to have been produced in imperial workshops, which were state controlled production facilities, like weapons production factories, fabricae. Except in Horreum Margi,62 such workshops had to have existed in cities near legionary camps, Singidunum and Viminacium. Comparison of finds of the first and second chronological group of crossbow fibulae from Romuliana leads to the conclusion that their function evolved from a role of strictly established marks of rank in the army and administration, to decorations awarded on different occasions 59 Mirkovi} 1989; Popovi} 1997; Popovi} 2002; Popovi} 2004, 232–235. 60 Petkovi} 1999, 227–228; Petkovi} 2008b, 361–363; Petkovi} 2010 b, 195, Fig. 168. 61 About the official character of workshops for manufacturing items from precious metals see: Popovi} 1997; Popovi} 2002; Popovi} 2008. 62 Weapon workshop in Horreum Margi, present-day ]uprija, was listed in document Notitia dignitatum (Not. dign., or. XI, 39.) 63 It appears as if some fibulae were cast in molds, made by according to the »original« such as cat. nos. 16–18, 24–25, 27–28. 64 Bows of some fibulae were extensively decorated by combinations of different geometrical motifs: grooves with transverse and slanting notches, wave patterns, braids, spirals, circles, »eyelets«, rhombs. This crowded decoration fills the entire visible surface of the bow (horror vacui) and has the same attributes of the rustic, »barbarian« style, most likely transferred from textile. 65 About luxurious specimens of early crossbow fibulae, as a status symbols of members of local elite outside the territory of Roman empire see: Werner 1989; Diaconescu 1999, 203–217. 123 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) fibulae. On the mosaic of the basilica of San Apollinare Nuovo in Ravenna from the first quarter of the 6th century, in the scene of Trial of Christ, Pilate was depicted with an army cloak fastened by a crossbow fibula on the right shoulder.67 Crossbow fibula also appears on the depiction of St. Theodore on a mosaic in the church of St. Cosmas and Damian in Rome, which dates to the first half of the 6th century.68 Likewise, on a fresco in the catacombs of San Gennaro in Naples, dated according to stylistic characteristics to the beginning of the 6th century, an individual by the name of Theotecnus is shown with a type 34 fibula on his right shoulder.69 Later depictions of crossbow fibulae from the 7th century are also well known. On the Mother of God icon from the Monastery of St. Catherine on the Sinai, dated between the end of the 6th and the beginning of the 7th century, St. Theodore and St. George are shown in chlamys with crossbow fibulae on the right shoulders.70 In one 6th–7th century Sinai icon, kept in the Kiev Lavra, Sts. Sergius and Bacchus are shown with robes fastened on the right shoulder with rudimentary crossbow fibulae, reduced to three »bulbs«.71 At the 7th century basilica of St. Dimitrius in Salonika, this saint was depicted as a »holly warrior«, with a cloak fastened on the right shoulder with a crossbow fibula.72 I regard the artistic representation of crossbow fibulae on Early Byzantine monuments of the 6th and 7th centuries as a mere traditional depiction of clothing of emperors, court and army dignitaries, as well as Christian saints during the Late Roman period, in the 4th and the 5th century. This traditionalism in the Late Roman art had been canonized in Early Byzantine period, in the 6th and 7th centuries. Besides that, the above mentioned artistic testimonies display massive gold plated fibulae with a long foot, decorated with incised pairs of peltae, known as fibulae type Keller/Pröttel 6,73 or luxurious golden crossbow fibulae with a long foot ornamented by Christian symbols in opus interrasile technique, type Pröttel 7.74 Both types are dated to the 5th century and rarely appeared on the territory of the Eastern Roman Empire.75 Five fibulae type Keller/Pröttel 6 are known to have been found on the territory of Serbia (Petkovi} type 34 F), in three sites; one in Singidunum, three from Viminacium and one from the fortification Ravna (Campsa) on the Danube.76 Two of them are particularly important for the chronology of this type in Serbia. They come from graves G–851 and G–1033 from the Viminacium necropolis of Pecine.77 In the first grave (G–851), which contained dislocated bones of two deceased,78 besides a type 34 F fibula, a bronze military belt buckle was discovered, STARINAR LX/2010 with a rectangular plating and a B-shaped frame with a spike that exceeds it, and money from the Constantine I, Constantine II and Constancius II period, minted between 335/6 and 361.79 In the other grave (G–1033) a type 34 F crossbow fibula was found on the right shoulder of a (male ?) deceased; parts of a silver military belt were found near the pelvis, with a buckle of the same type as in the previous grave, rectangular plating and a glass cup placed above the head as a tribute.80 A bow and foot of this fibula are decorated longitudinally with a »fir branch« motif, and circles with crosses inscribed, whereas at the end of the foot, or more exactly on the trapezoid pin holder, there was a ¾ male portrait, facing to his right, incised in the circular medallion.81 Even though it belongs to the 5th century variant, the fibula from grave G–1033 resembles the find from the Bulgarian village of Kolarci in regards of decoration, it can be concluded that it is an early specimen of type 34 F.82 According to the analyases of money and parts of belt sets,83 burials in graves G–851 and G–1033 can be dated to the last third of the 4th – first half of the 5th century.84 Therefore, I assume that crossbow fibulae in Roman provinces in present-day Serbia had been in use until no later than mid-5th century, i.e. until the fall of the Danube limes between 441 and 443. 66 Zahbelicky 1980, 1107–1108. Zahbelicky 1980, 1105, cat. 11. 68 Zahbelicky 1980, 1105–1106, cat. 13; Theune-Grosskopf 1995, 93, Abb. 66, 102, Abb. 74. 69 Zahbelicky 1980, 1108, cat. 10; Pröttel 1988, 370, Abb. 9. 70 Theune-Grosskopf 1995, 105, Abb. 76. 71 Zahbelicky 1980, 1106, Kat. 16; Theune-Grosskopf 1995, 105. 72 Zahbelicky 1980, 1106, Kat. 15. 73 Keller 1971, 52, Abb. 11, 13; Pröttel 1988, 369–371, Abb. 8, 2–4. 74 Pröttel 1988, 370, Anm. 166, Abb. 8, 6; Theune-Grosskopf 1995, 95–101, Abb. 69, Abb. 75. 75 Pröttel 1988, 370–371, Abb. 11; Buora 1997, 354–368. 76 Bojovi} 1983, kat. 453, T. LV, 453; Petkovi} 2008, 399, T. LXXI, 1–2. 77 Petkovi} 2008a, 465, kat. 1676–1677, T. LXXI, 1–2; Spasi}-\uri} 2008, 419–423, Sl. 8–9, T. I, 5–6. 78 Spasi}-\uri} 2008, 422. 79 I am grateful to dr. Mirjana Arsenijevic, who processed the numismatic material from the necropolises of Viminacium when she was working in the Archaeological institute in Belgrade. 80 Spasi}-\uri} 2008, 419–421. 81 Spasi}-\uri} 2008, Sl. 8. 1. 82 See fns. 35 and 36. 83 Spasi}-\uri} 2008, Sl. 8. 2, Sl. 9. 2. 84 Tejral 1997, 323–328, Abb. 1.2, 6, 11, 7, Abb. 2. 3–5, 10–11. 67 124 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) The question arises as to whether some other types had overtaken the official role, or the function of crossbow fibulae, after the restoration of the Danube limes and border provinces in Lower Danube in the 5th century (Moesia I, Dacia Ripensis, Moesia II). For now, without the typological, chronological and functional analysis of Early Byzantine crossbow fibulae from the region mentioned above, it is impossible to give an answer to this question, despite the indications that that some types with an incised cross, Christian inscriptions and schematic portraits of emperors and saints, could have marked the status in the clerical hierarchy or the Roman army.85 I observe that the Early Byzantine fibulae decorated in such a way have been found in horizon of life from the 6th – the beginning of the 7th century in Romuliana. STARINAR LX/2010 Fig. 2. Cat. 2, type 34 A 1 Sl. 2. Kat. 2, tip 34 A 1 3. Tower 19, segment II, layer E. National Museum in Zaje~ar, C 733/02. Type 34 A 1. Bronze. Dim: 5 cm. Pin is missing, and the foot is damaged. Fluted bar. Fluted pine shaped knobs on a double annular shaped base. Bow of trapezoidal cross-section longitudinally decorated with an incised wave pattern. A rectangular foot narrows at the end, faceted. Unpublished. DD: last quarter of 4th century. (Fig. 3) 3. CATALOGUE 3.1 Type 34, variant A 1. The western gate (?). National Museum in Zaje~ar, inv. G/272. Type 34 A 1. Bronze. Dim: 7 cm. Pin is missing. Fluted bar. Pine-shaped knobs on an annular base. Bow of trapezoidal cross section. Faceted, rectangular foot narrows at the end. Published: Jankovi} 1983 a, cat. 84. DD: the last quarter of the 4th century. (Fig. 1) Fig. 3. Cat. 3, type 34 A 1 Sl. 3. Kat. 3, tip 34 A 1 4. Magura. National Museum in Zaje~ar, inv. G/1575. Type 34 A 1. Bronze. Dim: 5.7 cm. Whole. Fluted bar. Pine-shaped knobs on a double annular base. Bow of trapezoidal cross-section. Short, rectangular foot narrows at the end, faceted. Published: @ivi} 2003. cat. 413. DD: the end of 3rd and the beginning of the 4th century. (Fig. 4, T. I, 1) Fig. 1. Cat. 1, type 34 A 1 Sl. 1. Kat. 1, tip 34 A 1 2. Tower 19, segment II, layer E. National Museum in Zaje~ar, C 672/02. Type 34 A 1. Bronze. Dim: 6.2 cm. Whole. Fluted bar, pine-shaped knobs. Bow of trapezoidal cross-section longitudinally decorated with a groove with transverse notches. A rectangular foot narrows at the end, faceted. Published: Petkovi} 2006, p. 35, Pl. III, 2. DD: last quarter of 4th century. (Fig. 2) 85 Haralambieva 1998, 368–370, Abb. 1–5; Jovanovi} 2007, 68–70, Sl. 7.4. 125 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 Fig. 4. Cat. 4, type 34 A 1 Fig. 6. Cat. 7, type 34 A 1 Sl. 4. Kat. 4, tip 34 A 1 Sl. 6. Kat. 7, tip 34 A 1 5. Magura. National Museum in Zaje~ar, inv. G/1654. Type 34 A 1. Bronze. Dim: 4.3 cm. Part of the bow with a foot. Bow of trapezoidal cross-section. Rectangular foot narrows at the end, faceted. Unpublished. DD: End of the 3rd – beginning of the 4th century. Dim: 4.7 cm. Pin and middle bulb missing. Fluted bar. Spherical knobs on an annular base. Bow of trapezoidal cross-section. Short, rectangular foot narrows at the end, faceted. Unpublished. DD: the end of 3rd and the beginning of the 4th century. (Fig. 7) 6. Extra muros, accidental find (field of Dimitrije Ili} from the village Gamzigrad). National Museum in Zaje~ar, inv. G/254 Type 34 A 1. Bronze. Dim: 7 cm. Pin missing Fluted bar. Pine-shaped knobs. Bow of trapezoidal cross-section. Rectangular foot narrows at the end, faceted. Unpublished. (Fig. 5) Fig. 7. Cat. 8, type 34 A 2 Sl. 7. Kat. 8, tip 34 A 2 Fig. 5. Cat. 6, type 34 A 1 9. Extra muros, tomb east of fortification. National Museum in Zaje~ar, C 1137/05. Type 34 A 2. Gold, bronze. Dim: 5.9 cm. Bronze pin missing. Fluted bar. Pine-shaped knobs with a groove in the middle fitted with a ring made of granulated gold wire. Bow of trapezoidal cross-section longitudinally decorated with a strip with transverse notches. There is a groove between the foot and the bow fitted with a ring made of granulated gold wire. Rectangular foot narrows at the end, longitudinally decorated similarly as the bow and faceted at the beginning and the end. DD: the end of 3rd and the beginning of the 4th century. Published: Petkovi} 2009, p. 253, Figs. 8–9; @ivi} 2009, p. 278, Cat. 4, Pl. IIa. (Fig. 8, T. I, 2) Sl. 5. Kat. 6, tip 34 A 1 7. Extra muros, accidental find (Petar Bo`anovi} from the village Gamzigrad). National Museum in Zaje~ar, inv. G/1603. Type 34 A 1. Bronze. Dim: 5 cm. Pin and foot missing. Fluted bar. Pine-shaped knobs on an annular base. Bow of trapezoidal cross-section. Published: @ivi} 2003. cat. 421. (Fig. 6) 8. Extra muros, east of the eastern gate. National Museum in Zaje~ar, inv. G/1027. Type 34 A 2. Bronze. 126 STARINAR LX/2010 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) Fig. 10. Cat. 11, type 34 B 2 Sl. 10. Kat. 11, tip 34 B 2 Fig. 8. Cat. 9, type 34 A 2 Sl. 8. Kat. 9, tip 34 A 2 12. Sector of »the Big temple«. National Museum in Zaje~ar, inv. G/599. Type 34 B 3. Bronze. Dim: 7.5 cm. Pin is missing as well as part of the bar with one bulb. Large fluted bulbs on an annular base. Bow of trapezoidal cross-section decorated with transverse incisions. Long rectangular foot faceted and decorated with a pair of peltae at the beginning and the end. The end of foot is curved. Published: @ivi} 2003, cat. 416. DD: the beginning of the 4th century. (Fig. 11) 3.2. Type 34, Variant B 10. Entrance to tower XIV. National Museum in Zaje~ar, inv. G/581. Type 34 B 2. Bronze. Dim: 8 cm. Pin is missing, foot is broken. Large knobs shaped like poppy pods on an annular base. Bow of trapezoidal crosssection. Long rectangular foot faceted and decorated with one or two pairs of circles. Published: Jankovic 1983 a, 114. cat. 83, DD: the beginning of the 4th century (Fig. 9) Fig. 11. Cat. 12, type 34 B 3 Sl. 11. Kat. 12, tip 34 B 3 Fig. 9. Cat. 10, type 34 B 2 3.3. Type 34, Variant C 13. Extra muros, grave 6/06. National Museum in Zaje~ar, C 81 e/06. Type 34 C 3 b. Bronze, gold, silver, iron. Dim: 8 cm. Iron pin and bulbs of tin bronze are damaged. Large, distinct bulbs shaped by hammering from thin tin bronze. Wide, hollow cast, short bow of trapezoidal cross-section, with a massive curved traverse bar. Longitudinally decorated with a strip done in niello technique: alternately placed circles with inscribed crossing lines (christogram?), and rhombs with vines; at the beginning and the end of the bow there are square spaces, metopes with depictions of male busts. At the beginning of the bow, a face in en face position was de- Sl. 9. Kat. 10, tip 34 B 2 11. Sector of »the Big temple«. National Museum in Zaje~ar, inv. G/689. Type 34 B 2. Bronze. Dim: 5 cm. The head is damaged, and a large portion of the bow is missing. Large fluted bulbs on an annular base. Bow of trapezoidal cross-section. Long rectangular foot narrows at the end, faceted and longitudinally decorated with a strip with reticulated and punched decorations. Unpublished. DD: the beginning of 4th century. (Fig. 10) 127 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 3.4. Variant 34 D 14. Between Tower 14 and western rampart of the younger fortification. National Museum in Zaje~ar, inv. G/206. Type 34 D 2. Bronze. Dim: 8 cm. A pin and part of the bar with one bulb is missing. Large, distinct bulbs. Bow of trapezoidal cross-section longitudinally decorated with a »fir branch« motif, and with a pseudo filigree on the border with the foot. A rectangular foot longitudinally decorated with two grooves, faceted on the beginning and the end with one and two pairs of »eyelets«. Unpublished. DD: the last quarter of the 4th – the first half of 5th century. (Fig. 16) picted, dressed in furnished clothes (a toga?), with a wide hat or a nimbus around their head. A person’s ¾ profile facing to his left is shown at the end of the bow, wearing pageboy hairstyle and a paludamentum fastened by a circular plate fibula on the right shoulder. A long rectangular foot, with edges cut (hemstiched) with three pairs of carved peltae and volutes (at the beginning and the end), is longitudinally decorated with an inserted tin silver strip, with a double motif of fir branch in niello technique. There is a square metope in the middle of the foot, with a male ¾ portrait facing to his left crafted in niello technique. The depiction is the same as the one at the beginning of the bow, only better crafted. Cylindrical pin holder of trapezoidal cross-section is made of tin silver. Published: Petkovi} 2009, 266–267, Figs. 46–47; @ivi} 2009, 285, Cat. 50e, Pl. X, 50e. DD: the end of 4th – beginning of 5th century. (Fig. 12–15, Pl. I, 3) Fig. 16. Cat. 14, type 34 D 2 Sl. 16. Kat. 14, tip 34 D 2 15. National Museum in Zaje~ar, inv. G/462. Type 34 D 2. Bronze. Dim: 5 cm. A pin is missing together with a part of the bar with a bulb and the middle bulb, the foot is damaged. Large, distinct bulbs on an annular base. Bow of trapezoidal cross-section longitudinally decorated with Fig. 12. Cat. 13, type 34 C 3 b Sl. 12. Kat. 13, tip 34 C 3 b Fig. 13. Cat. 13, the detail of »imperial« portrait on the foot of fibula Fig. 14 and 15. Cat. 13, the detail of »imperial« portrait on the beginning and on the end of fibula’s bow Sl. 13. Kat. 13, detaq stope sa »carskim« portretom Sl. 14 i 15. Kat. 13, detaq »carskog« portreta na po~etku i na kraju luka fibule 128 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 18. Sector of Palace I. National Museum in Zaje~ar, inv. G/591. Type 34 D 2. Bronze. Dim: 7.5 cm. A pin is missing as well as a part of the bar with one bulb and the middle bulb. Large, distinct bulbs on a pseudo filigree base. Bow of trapezoidal cross-section longitudinally decorated with a strip, rimmed by pseudo filigree, with an incised braid, and pseudo filigree on the border with a foot. Long trapezoidal foot longitudinally decorated with a strip with an incised »fir branch« motif. Faceted and decorated on the beginning and the end with two and three pairs of »eyes«. Published: @ivi} 2003, cat. 425. DD: the last quarter of 4th – the first half of 5th century. (Fig. 20) a strip with slanting incisions. Long trapezoidal foot longitudinally decorated similarly as the bow, faceted and embellished at the beginning and the end with one and two (?) pairs of »eyelets«. Unpublished. (Fig. 17) Fig. 17. Cat. 15, type 34 D 2 Sl. 17. Kat. 15, tip 34 D 2 16. National Museum in Zaje~ar, inv. G/480. Type 34 D 2. Bronze. Dim: 7.8 cm. Head and pin missing. Bow of trapezoidal cross-section longitudinally decorated with a strip with diagonal incisions, and with pseudo filigree on the border with the foot. Long trapezoidal foot longitudinally decorated similarly as the bow, faceted and decorated at the beginning and the end with two and three pairs of »eyelets«. Unpublished (Fig.18) Fig. 20. Cat. 18, type 34 D 2 Sl. 20. Kat. 18, tip 34 D 2 19. Sector of »the Big temple«. National Museum in Zaje~ar, inv. G/652. Type 34 D 2. Bronze. Dim: 8 cm. A pin is missing and a middle bulb. Large distinct bulbs. Bow of trapezoidal cross-section longitudinally decorated with a groove with transverse incisions, at the end by pseudo filigree. Long rectangular foot longitudinally decorated with two grooves, at the beginning and the end faceted and decorated with two pairs of circles. Published: @ivi} 2003, cat. 417. DD: the last quarter of 4th – the first half of 5th century. (Fig. 21) Fig. 18. Cat. 16, type 34 D 2 Sl. 18. Kat. 16, tip 34 D 2 17. Sector of Palace I. National Museum in Zaje~ar, inv. G/591. Type 34 D 2. Bronze. Dim: 7.5 cm. A pin is missing. Large, distinct bulbs. Bow of trapezoidal crosssection longitudinally decorated with a »fir branch« motif. Long trapezoidal foot longitudinally decorated with two grooves, faceted and decorated at the beginning and the end with one and two pairs of »eyelets« Published: Jankovic 1983 A, p. 114, cat 82. DD the last quarter of 4th – the first half of 5th century. (Fig. 19) Fig. 21. Cat. 19, type 34 D 2 Sl. 21. Kat. 19, tip 34 D 2 20. Sector of eastern gate. National Museum in Zaje~ar, inv. G/845. Type 34 D 2. Bronze. Dim: 6.7 cm. Pin and bulbs missing. Bow of trapezoidal cross-section longitudinally decorated with a groove with transverse incisions. Long rectangular foot with rounded end longitudinally decorated in the same way, and along the rims with a line of ten alternating circular and triangular Fig. 19. Cat. 17, type 34 D 2 Sl. 19. Kat. 17, tip 34 D 2 129 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 notches. Unpublished. DD: the last quarter of 4th – the first half of 5th century. (T. I, 4) 21. Sector of »the Small temple«. National Museum in Zaje~ar, inv. G/1506. Type 34 D 2. Bronze. Dim: 6.5 cm. Pin and middle bulb missing. Large, distinct bulbs on an annular base. Bow of trapezoidal cross-section longitudinally decorated with three grooves and transverse notches. Long rectangular foot with serrated end longitudinally decorated with a groove, and along its rims with an array of seven »eyelets«. Published: @ivi}, cat. 419. DD: the last quarter of 4th – the first half of 5th century. (Fig. 22) Fig. 23. Cat. 24, type 34 D 2 Sl. 23. Kat. 24, tip 34 D 2 25. Sector of Thermae, layer E. National Museum in Zaje~ar, C 593/05. Type 34 D 2. Bronze. Dim: 8.3 cm. Pin is missing. Large, distinct bulbs. Bow of trapezoidal cross-section longitudinally decorated with two punched lines and a groove with transverse lines in the middle. Long trapezoidal foot longitudinally decorated the same way as the bow, faceted at the beginning and the end and ornamented with two and three pairs of »eyelets«. Unpublished. DD: the last quarter of 4th – the first half of 5th century. Fig. 22. Cat. 21, type 34 D 2 Sl. 22. Kat. 21, tip 34 D 2 22. Tower 19, segment III, layer E. National Museum in Zaje~ar, inv. G/1756. Type 34 D 2. Bronze. Dim: 8.7 cm. Pin and bulbs missing. Bow of triangular cross-section. Pseudo filigree on the border with the foot. Rectangular foot longitudinally decorated with a groove, at the beginning and the end faceted and decorated with two pairs of circles. Unpublished. DD: the last quarter of 4th – the first half of 5th century. (T. I, 5) Fig. 24. Cat. 25, type 34 D 2 Sl. 24. Kat. 25, tip 34 D 2 23. Tower 19, segment I, layer E. National Museum in Zaje~ar, C87/97. Type 34 D 2. Bronze. Dim: 5 cm. Foot with a bow fragment. Bow of triangular cross-section. Rectangular foot longitudinally decorated with two grooves, at the beginning and the end faceted and decorated with two pairs of circles. Unpublished. DD: the last quarter of 4th – the first half of 5th century. 26. Sector of Thermae, layer D, house 1/07. National Museum in Zaje~ar. C 299/07. Type 34 D 2. Bronze. Dim: 4 cm. Pin and foot are missing. Large, distinct bulbs. Bow of trapezoidal cross-section. Unpublished. DD: the first half of the 5th century. 24. Tower 19, segment II, layer C. National Museum in Zaje~ar, C347/02. Type 34 D 2. Bronze. Dim: 5.1 cm. Pin and middle bulb missing. Large, distinct bulbs. Bow of triangular cross-section longitudinally decorated with a groove and a line of peltae. Long trapezoidal foot longitudinally decorated with two grooves and along the rims with an array of six eyes connected by tangents. Published: Petkovic 2006, p. 36, PL. III,6. DD: the last quarter of 4th – the first half of 5th century. 27. Sector of Thermae, layer D, house 1/07. National Museum in Zaje~ar. C 357/07. Type 34 D 2. Bronze. Dim: 8.5 cm. Pin is missing. Large, distinct bulbs. Bow of trapezoidal cross-section longitudinally decorated with an engraved braid with an array of impressed rhombs in the middle. Long trapezoidal foot has two grooves at the middle, and it is faceted at the beginning and the end with two and three pairs of »eyelets«. Unpublished. DD: the first half of the 5th century. 130 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) 28. Sector of Thermae, layer E. National Museum in Zaje~ar, C 448/07. Type 34 D 2. Bronze, iron. Dim: 8.3 cm. Iron pin is damaged. Large, distinct bulbs, with base decorated with pseudo filigree. Bow of trapezoidal crosssection, longitudinally decorated with an array of laid S spirals. Long rectangular foot has two grooves along the middle, faceted at the beginning and the end and ornamented with two and three pairs of »eyelets«. Unpublished. DD: the last quarter of 4th – the first half of 5th century STARINAR LX/2010 Fig. 25. Cat. 29, type 34 D 2 Sl. 25. Kat. 29, tip 34 D 2 29. National Museum in Zaje~ar, inv. G/1576, 1995 Type 34 D. Bronze. Dim: 5.5 cm. Middle bulb is missing, foot is broken.86 Large, distinct bulbs on an annular base. Bow of trapezoidal cross-section longitudinally decorated with a strip with an engraved »fir branch« motif. Published: @ivi} 2003, cat. 420. DD: the last quarter of 4th – the first half of 5th century (Fig.25) Fragment of transverse curved bar with two lateral bulbs. Distinct bulbs on an annular base. Unpublished. DD: the last quarter of 4th – the first half of 5th century. 31. Tower 19, segment II, layer E, under the furnace 6. National Museum in Zaje~ar, C 803/02. Type 34 D 2. Bronze. Dim: 2.4 cm. Fragment of transverse curved bar with one lateral, distinct bulb on a pseudo filigree base. Unpublished. DD: the last quarter of 4th century. 30. Tower 19, segment II, layer C. National Museum in Zaje~ar, C 147/02. Type 34 D. Dim: 4.8 cm Bronze. Fig. 26. A part of the funeral monument from the site Tilva ro{ (Bor, eastern Serbia) with the presentation of decesed persons with crossbow fibulae on their right shoulders, the end of 3rd – beginning of 4 th century (Photo from the documentation of Museum of Mining and Metallurgy in Bor) Sl. 26. Deo nadgrobnog spomenika sa lokaliteta Tilva Ro{ (Bor, isto~na Srbija) na kome su pokojnici prikazani sa krstobraznim fibulama na desnom ramenu, kraj III – po~etak IV veka 86 Large portion of the fibula was damaged during conservation, but the drawing in a terrain inventory (C 77/95) shows that the fibula had a long, rectangular, faceted foot with an incised decoration (?). 131 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) STARINAR LX/2010 BIBLIOGRAPHY: Bojovi} 1983 – D. Bojovi}, Rimske fibule Singidunuma, Beograd 1983. Bori}-Bre{kovi} 2009 – B. Bori}-Bre{kovi}, Novi pogled na numizmati~ke nalaze sa Magure, Zbornik Narodnog muzeja XIX–1, arheologija, Beograd 2009, 343–365. Buora 1997 – M. Buora, »Zwiebelkopffibeln« dell tipo Keller 6 da Aquilea, AV 48, Ljubljana 1997, 247–260. Burns 1994 – Th. S. Burns, Barbarians within the Gates of Rome, A Study of Roman Military Policy and the Barbarians, ca. 375–425 A.D., Bloomington – Indianapolis 1994. von Bülow, Schüler 2009 – G. von Bülow, T. Schüler, Geophysical and Archaeological Research at Gamzigrad – Report of the 2004–2007 Campaignes, Starinar LVII/2007, Beograd 2009, 231–249. Dautova-Ru{evljan 2003 – V. Dautova-Ru{evljan, Kasnoanti~ka nekropola kod Svilo{a u Sremu, Novi Sad 2003. Diaconescu 1999 – A. Diaconescu, Ornamenta dignitatis: Gradabzeichen und Symbole des sozialen Status bei den lokalen Eliten von Dakien nache dem Aurelianische Rückzug, Acta MN 36/I, Cluj – Napoca 1999, 203–243. Gen~eva 2004 – E. Gen~eva, Rimskite fibuli ot Bølgariý ot kraý na I v. pr. n. e. do kraý na VI v. na n.e. (E. Gen~eva, Les fibules romaines de Bulgarie de la fin du Ier s. av. J.–C. à la fin du VIe s. ap. J.–C.), Veliko Tørnovo 2004. Gomolka-Fuchs 1982 – G. Gomolka-Fuchs, Die Kleinfunde, in: Iatrus–Krivina II. Ergebnisse der Ausgrabungen 1966–1973, Berlin 1982, 149–205. Gomolka-Fuchs 1991 – G. Gomolka-Fuchs, Die Kleinfunde, in: Iatrus–Krivina IV. Ergebnisse der Ausgrabungen 1975–1981, Berlin 1991, 167–205. Gomolka-Fuchs 1995 – G. Gomolka-Fuchs, Die Kleinfunde und ihre Aussagen zur Bevölkerung von Iatrus, in: Iatrus–Krivina V. Studien zur Geschichte des Kastells Iatrus, Berlin 1995, 81–105. Haralambieva 1990 – A. Haralambieva, A. Lukovi~ni fibuli vøv Varnenski muzeç, INMV 26(41), Varna 1990, 29–41. Haralambieva 1998 – A. Haralambieva, Darstellungen christlischer Symbole, Inschriften und Heiligen auf Trachtzubehor des 4.–7. Jhs. Aus Heutigem Bulgarien, Acta XIII Congressus internationalis Archaeologiae Christianae, vol. III, Split–Vatikan 1998, 367–373. Ivanov 1972 – T. Ivanov, Obrazite na Konstantin i sinovete mu vørhu lukovi~nite fibuli ot Bølgaria, Arheologiý /1972–4, Sofiý 1972, 9–29. Ivanovski 1987 – M. Ivanovski, The Grave of Warrior from the Period of Licinius I Found at Tarane{, AI/24, Ljubljana 1987, 81–90. Iv~evi} 2001 – S. Iv~evi}, Lukovi~aste fibule iz Salone u Arheolo{kom muzeju u Splitu, VAHD 92/2000, Split 2001, 125–187. Janes 1998 – D. Janes, Brooches and Insignia and Loyality to the Late Roman State, VHAD 87–89, Acta XIII Congressus internationalis Archaeologiae Christianae, vol. III, Split–Vatikan 1998, 387–394. Jankovi} 1983a – \. Jankovi}, U sutonu antike, u: Gamzigrad. Kasnoanti~ki carski dvorac, ur: D. Srejovi}, Beograd 1983, 98–119 . Jankovi} 1983b – \. Jankovi}, Ranovizantijski Gamzigrad, u: Gamzigrad. Kasnoanti~ki carski dvorac, ur: D. Srejovi}, Beograd 1983, 120–141 . Jovanovi} 1975 – A. Jovanovi}, Krstobrazne fibule iz anti~ke zbirke Narodnog muzeja u Ni{u, ZNM VIII, Beograd 1975, 235–245. Jovanovi} 1976 – A. Jovanovi}, O problemu fibula sa portretima, ZRFF XIII, 1, Beograd 1976, 43–51. Jovanovi} 2007 – A. Jovanovi}, Ogledi iz anti~kog kulta i ikonografije, Beograd 2007. Keller 1971 – E. Keller, Die spätrömischen Grabfunde in Südbayern, MBV 14, München 1971. Laur-Belart 1959 – R. Laur-Belart, Ein frühchristlische Grab aus Bazel, Ur-Schweiz XXIII, 4, Bazel 1959, 57–71. Marijanski-Manojlovi} 1987 – M. MarijanskiManojlovi}, Rimska nekropola kod Be{ke u Sremu, Novi Sad 1987. Migotti 2008 – B. Migotti, Lukovi~aste fibule s portretima na podru~ju Rimskog carstva (The Crossbow Brooches with Portraits in the Roman Empire), Zagreb 2008. Mirkovi} 1989 – M. Mirkovi}, Natpisi na »carskim« fialama i fibulama i Licinijeva propaganda, 316–322. godine, ZRFF XVI, Beograd 1989, 37–44. Noll 1974 – R. Noll, Eine goldene »Kaiserfibel« aus Niederemmel vom Jahre 316, Bonner Jahrbuch 174/ 1974, Bonn 1974, 221–244. Petkovi} 1999 – S. Petkovi}, Meaning and Provenance of Horses’ Protomes Decoration on the Roman 132 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) Antler Combs, Starinar XLIX/1998, Beograd 1999, 213–228. Petkovi} 2004 – S. Petkovi}, Arheolo{ka iskopavawa u ju`noj kuli zapadne kapije mla|eg utvr|ewa na lokalitetu Romuliana – Gamzigrad u 2002. godini, Glasnik SAD 20/2004, Beograd 2004, 127–153. Petkovi} 2006 – S. Petkovi}, Study of Stratigraphy of Cultural Layers of Late Roman Romuliana; Case Study: South Tower of West Gate of Later Fortification, in: Felix Romuliana. 50 Years of Archaeological Excavations, Papers from the International Conference, Zaje~ar, 27th–29th October 2003, ed: M. Vasi}, Arheolo{ki institut, Odbor za arheologiju SANU i Narodni muzej u Zaje~aru, Beograd 2006, 29–45. Petkovi} 2008a – S. Petkovi}, Fibule u rimskim provincijama na tlu Srbije, od I do IV veka n.e., Doktorska disertacija odbranjena 18. 02. 2008. godine na Filozofskom fakultetu, Univerzitet u Beogradu. Petkovi} 2008b – S. Petkovi}, Unilateral Antler Combs from Romuliana, Starinar LVI–LVII/2006–2007, Beograd 2008, 353–366. Petkovi} 2008c – S. Petkovi}, Iskopavawa na lokalitetu Gamzigrad – Romuliana, 2004. godine, Arheolo{ki pregled nova serija 2/3 (2004/2005), Beograd 2008, 61–63. Petkovi} 2008d – S. Petkovi}, Iskopavawa na lokalitetu Gamzigrad – Romuliana, 2005. godine, Arheolo{ki pregled nova serija 2/3 (2004/2005), Beograd 2008, 64–67. Petkovi} 2009 – S. Petkovi}, Late Roman Necropolis of Romuliana, Area South of the Fortified Palace (Research 2005–2006), Starinar LVII/2007, Beograd 2009, 251–275. Petkovi} 2010a – S. Petkovi}, Rimski Gamzigrad pre carske palate, in: Felix Romuliana – Gamzigrad, ed: I. Popovi}, Beograd, 2010, 33–42. Petkovi} 2010b – S. Petkovi}, Romulijana u vreme posle carske palate, u: Felix Romuliana – Gamzigrad, ur: I. Popovi}, Beograd 2010, 167–200. Petru, Petru 1978 – S. Petru, P. Petru, Neviodunum (Drnovo pri Kr{kem), Ljubljana 1978. Pop-Lazi} 2009 – S. Pop-Lazi}, Late Roman Necropolis Beljnja~a in [id, Starinar LVIII/2008, Beograd 2009, 163–174. Popovi} 1997 – I. Popovi}, Les production officielles et priveés des ateliers d’ orféuverie de Naissus et de Sirmium, Antiquité tardive 5, 133–144. Popovi} 2001 – I. Popovi}, Imitatio deorum kao motiv na nakitu iz Gorwe Mezije (Summary: Imitatio deorum as a Motif in Jewelry from Upper Moesia), u: STARINAR LX/2010 Vetsigatio vetustatis Aleksandrini Cermanovi}Kuzmanovi} od prijateqa, saradnika i u~enika, ur: M. Lazi}, Beograd 2001, 371–389. Popovi} 2002 – I. Popovi}, The Production of Gold and Silver Workshops in Late Roman Sirmium, in: The Roman and Late Roman City, The International Conference (Veliko Trnovo 26–30 July 2000), ed: L. RusevaSlokoska, R. Ivanov, V. Dintchev, Sofia 2002, 383–387. Popovi} 2004 – I. Popovi}, Specific Variants of Gold and Silver Early »Zwiebelknopf« Fibulae from Eastern Serbia, Starinar LIII–LIV/ 2003–2004, Beograd 2004, 225–239. Popovi} 2008 – I. Popovi}, Naissus (Ni{) – Augusta Trevirorum (Trier) and the reciprocal relationship between late antique imperial workshops, Trierer Zeitschrift 69/70–2006/2007, Trier 2008, 191–200. Popovi} 2009a – I. Popovi}, Gilt Fibula with Christogram from Imperial Palace in Sirmium, Starinar LVII/2007, Beograd 2009, 101–112. Popovi} 2009b – I. Popovi}, Novi pogled na arheolo{ke nalaze iz konsekrativnog spomenika I na Maguri, Zbornik Narodnog muzeja XIX–1, arheologija, Beograd 2009, 315–342. Popovi} 2009c – I. Popovi}, Silver Belt Garniture from the Tomb outside Romuliana Walls, Antiquité tardive 17/2009, Brepols 2009, 317–326. Popovi} 2010a – I. Popovi}, Verska politika Konstantina Velikog i wegovi oficijelni portreti na novcu, u: Ni{ i Vizantija. Zbornik radova VIII. Osmi nau~ni skup »Ni{ i Vizantija«, dani Sv. Cara Konstantina i Carice Jelene, Ni{ 3–5. juni 2009, ur: M. Rakocija, Ni{ 2010, 147–162. Popovi} 2010b – I. Popovi}, Sakralno-funerarni kompleks na Maguri, u: Felix Romuliana – Gamzigrad, ur: I. Popovi}, Beograd 2010, 141–158. Pröttel 1988 – P. M. Pröttel, Zur Chronologie de Zwibelknopffibeln, JRGZM 35/1, Mainz 1988, 347– 372. Red`i} 2007 – S. Red`i}, Nalazi rimskih fibula na nekropolama Viminacijuma, Beograd 2007. Rexi}, Rai~kovi}, Milovanovi} 2006 – S. Rexi}, A. Rai~kovi}, B. Milovanovi}, Krstaste fibule u grobovima viminacijumskih nekropola, Arheologija i prirodne nauke 2, Beograd 2006, 24–45. Simi} 1997 – Z. Simi}, Rezultati za{titnih arheolo{kih istra`ivanja na prostoru jugoisto~ne nekropole Singidunuma, Singidunum 1, Beograd, 21–56. Sommer 1984 – M. Sommer, Die Gürtel und Gürtelbeschläge des 4. und 5. Jahrhunderts im römischen Reich, Bonner Hefte zur Vorgeschichte 22, Bonn 1994, 1–83. 133 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) Spasi}-\uri} 2008 – D. Spasi}-\uri}, Prilog prou~avawu pozla}enih krstobraznih fibula iz Viminacijuma, Glasnik SAD 24/2008, Beograd 2008, 401–430. Srejovi} 1983 – D. Srejovi}, Uvod, u: Gamzigrad. Kasnoanti~ki carski dvorac, ur: D. Srejiovi} Beograd 1983, 5–16. Srejovi}, Vasi} 1994 – D. Srejovi}, ^. Vasi}, Imperial Mausolea and Consecration Memorials in Felix Romuliana (Gamzigrad, East Serbia), Belgrade 1994. Tomovi} 1997 – M. Tomovi}, O jednoj gorwomezijskoj skulptorskij predstavi »tetrarhijskog cara« (Summary: An Upper Moesian sculptoral representation of the »Tetrachial emperor«), u: Uzdarje Dragoslavu Srejovi}u ( Dragoslavo Srejovi}), ur: M. Lazi}, Beograd 1997, 475–482. Tejral 1997 – J. Tejral, Neu Aspekte der frühvölkerwanderungszeitlischen Chronologie im Mitteldonauraum, in: Neue Beiträge zur Erforschung der Spätantike im mittleren Donauraum, Hrsg: J. Tejral, H. Friesinger, M. Kazanski, Brno 1997, 321–362. Theune-Grosskopf 1995 – B. Theune-Grosskopf, Zwiebalknopffibeln und ihre Träger–Schmuck und Rangabzeichen, in: Die Schraube zwischen Macht und Pracht, Das Gewinde in der Antike, Sigmaringen 1995, 77–112. Uenze 1992 – S. Uenze, Die spätantiken Befestigungen von Sadovec (Bulgarien), MBV 43, München 1992. STARINAR LX/2010 Vasi} 2001a – M. Vasi}, Osvrt na nalaz IV veka iz Star~eva, ZNM XVII–1, Beograd 2001, 175–201. Vasi} 2001b – M. Vasi}, Neka pitawa o fibulama sa carskim natpisima, u: Vestigatio vetustatis Aleksandrini Cermanovi}-Kuzmanovi} od prijateqa, saradnika i u~enika, ur: M. Lazi}, Beograd 2001, 91–107. Vasi} 2008 – M. Vasi}, Prolasci i boravci rimskih imperatora kroz Ni{ krajem III i u IV veku (Roman Emperors Passing through or staying in Ni{ in Late 3rd and 4th Centuries), in: Naissus I, Ni{ 2008, 7–27. Vasi} 2009 – M. Vasi}, Findings of Coins from Romuliana, unearthed in 2005 and 2006, outside the Fortified Palace, Starinar LVII/2007, Beograd 2009, 309–314. Werner 1989 – J. Werner, Zu den römischen Mantelfibeln zweiter Kriegergräbern von Leune, Jahresschrift für Mitteldeutsche Vorgeschicht 72/1989, Halle/Saale 1989, 121–134. Zahbelicky 1980 – H. Zahbelicky, Zwiebelknopffiebeln als Kennenzehchen von Soldaten, in: Roman Frontier Studies 1979, ed: W. S. Hanson, L. J. F. Keppie, B. A. R. Int. Ser. 71, Oxford 1980. @ivi} 2003 – M. @ivi}, Felix Romuliana. 50 godina odgonetawa (Felix Romuliana. 50 Years of Solving), Beograd 2003. @ivi} 2009 – M. @ivi}, Catalogue of Small Finds from Excavations outside the Fortified Palace Romuliana (2005–2007), Starinar LVII/2007, Beograd 2009, 277–307. 134 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) Rezime: STARINAR LX/2010 SOFIJA PETKOVI], Arheolo{ki institut, Beograd KRSTOOBRAZNE FIBULE SA GAMZIGRADA (ROMULIANA) Kqu~ne re~i. – Kasna antika, tetrarhija, Gamzigrad, Felix Romuliana, krstoobrazne fibule, tipologija fibula, funkcija fibula, proizvodwa fibula. tim »okcima« (sl. 16–25, T. I, 4–8). Motiv »okaca« mo`e biti simetri~no raspore|en du` rubova stope (sl. 22–23, T. I, 4–5) ili grupisan u parovima na wenom po~etku i kraju (sl. 16–21, T. I, 6–8). Fibule varijante D su druge po zastupqenosti me|u krstobraznim fibulama u Srbiji i ~ine gotovo ~etvrtinu svih nalaza tipa 34, odnosno 23,14%. Rasprostrawene su u svim rimskim provincijama u na{oj zemqi, kako na dunavskom limesu, tako i u unutra{wosti. Tipolo{kom analizom varijanti fibula tipa 34 iz Romulijane uo~avaju se dve grupe ovih nalaza: 1. Fibule varijanti A i B, koje se mogu opredeliti u tetrarhijski period i vreme Flavijevske dinastije Konstantina I, generalno u kraj III – prvu polovinu IV veka. 2. Fibule varijante C 3 i D 2, koje se datuju u vreme Valentinijanske dinastije do Hadrijanopoqske bitke ili do po~etka vladavine Teodosija I, generalno u posledwu tre}inu IV veka. Na osnovu stratigrafije kulturnih slojeva na Gamzigradu analizirani su uslovi nalaza krstoobraznih fibula obe tipolo{ke grupe iz Romulijane (Tabela 1). Od ukupnog broja krstoobraznih fibula (31), samo ~etiri primeraka nije imalo siguran arheolo{ki kontekst. (kat. 6–7, kat. 15–16) Ostale fibule uglavnom poti~u sa sistematskih arheolo{kih iskopavawa u utvr|enoj carskoj palati. Nekoliko je prona|eno pri istra`ivawima sakralno-memorijalnog kompleksa na Maguri (kat. 4–5) i sondirawu van bedema utvr|ewa (kat. 8–9, 13). U okviru druge tipolo{ke grupe krstoobraznih fibula sa Gamzigrada je i primerak tipa 34 C 3 b na|en u grobu 6/06 (kat.br. 13). Na osnovu analize nalaza polo`enih uz levu nogu pokojnika grob se datuje u posledwu ~etvrtinu IV veka. Novac na|en u grobu daje 367. godinu za terminus ante quem non sahrane, a posledwa godina kovawa (378.) najverovatnije je i pribli`an datum smrti pokojnika. Na osnovu na~ina izrade, ukrasa vi{estrukih izrezanih pelti na stopi i stila predstavqanih »carskih portreta«, fibula iz groba 6/06 se datuje u posledwu tre}inu IV veka. Dve tipolo{ke grupe krstoobraznih fibula iz Romulijane uklapaju se generalno u dve hronolo{ke grupe ovih nalaza (Tabela 1): 1. grupa krstoobraznih fibula datovana u kraj III – po~etak IV veka i 2. grupa krstoobraznih fibula datovanih u posledwu ~etvrtinu IV – prvu polovinu V veka. Prvu hronolo{ku grupu ~ine fibule prve tipolo{ke grupe, na|ene u arheolo{kom kontekstu datovanom od po- Na nalazi{tu Gamzigrad – Felix Romuliana do sada je otkrivena tridesetjedna krstoobrazna ili »lukovi~asta« fibula (Crossbow Brooches, Zwiebelknopffibeln). Izuzimaju}i nalaze sa velikih rimskih nekropola, krstoobrazne fibule sa Gamzigrada ~ine obiman uzorak sa jednog nalazi{ta u Srbiji. Ve}ina ovih fibula poti~e iz jasno stratificiranih arheolo{kih celina, opredeqenih u dva horizonta `ivota u Romulijani: 1. horizont izgradwe Galerijeve palate s po~etka IV veka i 2. horizont kasnoanti~kog utvr|enog naseqa iz posledwih decenija IV i prve polovine V veka. Krstoobrazne fibule iz Romulijane analizirane su na dva na~ina: formalno-tipolo{ki i stratigrafsko-hronolo{ki. Na osnovu navedene analize dati su zakqu~ci o wihovoj funkciji i proizvodwi. Tekst je propra}en katalogom nalaza analiziranih u tekstu. Sedam fibula tipa 34 sa Gamzigrada pripada ranoj varijanti A 1 (kat. 1–7, sl. 1–6, T. I, 1). Varijanta A 2 (kat. 8–9) razlikuje se od prethodne jednostavnim urezanim ukrasom na luku i stopi (sl. 7–9, T. I, 2). Ovoj varijanti pripada i luksuzan zlatan primerak, na|en 2005. godine u grobnici Galerijevog velikodostojnika (kat. 9, sl. 8, T. I, 2). Fibule tipa 34 A iz Srbije uglavnom poti~u sa lokaliteta u Drugoj Panoniji i sa dunavskog limesa Prve Mezije i Priobalne Dakije (Karta 1). Tri fibule sa Gamzigrada pripadaju tipu 34 B (kat. 10–12): dva primerka podvarijanti 2, sa stopom ukra{enom utisnutim kru`i}ima ili urezima (kat. 10–11, sl. 9–10) i jedan podvarijanta 3, sa stopom ukra{enom utisnutim motivom pelti (kat. 12, sl. 11). U Srbiji su fibule tipa 34 B zastupqene u Drugoj Panoniji i Prvoj Meziji (Singidunum, Viminacium, Horreum Margi), a u Dardaniji jedan primerak je na|en u Naisu i jedna srebrna fibula poti~e sa nekropole Zaskok kod Uro{evca (Karta 1). Na Gamzigradu je na|ena jedna fibula tipa 34 C, podvarijanta 3 malobrojnih »carskih fibula«, sa lukom ukra{enim nielo tehnikom geometrijskim i vegetabilnim motivima i portretnim medaqonima i/ili metopama, sa pravougaonom stopom ~iji su rubovi profilisani (a`urirani) nizom pelti (kat. 13. sl. 12–15, T. I, 3). Fibule tipa 34 C na|ene su u Srbiji na lokalitetima Sirmium, Singidunum, Viminacium, Idimum, Ravna–Campsa, Prahovo–Aquae, Romuliana, ]i}evac, Naissus i Ulpiana (Karta 2). Najve}i broj krstoobraznih fibula iz Romulijane (kat. 14–31), pripada tipu 34 D 2, ~ija je duga stopa, trapezoidnog ili pravougaonog oblika, ukra{ena fasetama i utisnu- 135 Sofija PETKOVI], Crossbow Fibulae from Gamzigrad (Romuliana) (111–136) sledwih decenija III do prve ~etvrtine IV veka: tipovi 34 A 1, 34 A 2, 34 B 2 i 34 B 3 (kat. 1, 4–5, 8–12). Druga hronolo{ka grupa sadr`i fibule tipa 34 D 2 (kat. 14–31), datovanog u posledwu tre}inu IV veka, fibulu tipa 34 C 3 b iz groba s kraja IV veka (kat. 13) i dva starija primerka tipa 34 A 1, na|ena u Kuli 19 (kat. 2–3). Logi~no je da se stariji tipovi krstoobraznih fibula u odre|enom procentu (na Gamzigradu 11,11%) javqaju i u mla|im arheolo{kim celinama, jer se radi o predmetima koji su ozna~avali status u vojsci ili administraciji, te su ih vlasnici ~uvali i nosili niz godina (sl. 26). Najve}i broj krstoobraznih fibula iz Romulijane pripada drugoj hronolo{koj grupi, horizontu naseqa nastalog u posledwoj tre}ini IV veka posle napu{tawa izgradwe carske palate, i podjednako su zastupqene u celinama posledwe ~etvrtine IV veka i prve polovine V veka (Tabela 1). Pore|ewe nalaza prve i druge hronolo{ke grupe krstoobraznih fibula iz Romulijane dovodi do zakqu~ka da je wihova funkcija evoluirala od striktno utvr|enih oznaka ~ina u vojsci i administraciji do odlikovawa za individualne zasluge u carskoj slu`bi, dodeqivanih razli~itim povodom. Najstariji primerci pripadaju ranoj varijanti 34 A, koja se razvila iz zglobnih T – fibula na prelazu III u IV vek, u vreme Prve tetrarhije (293–305.) Zlatan primerak iz grobnice (kat. 9) i fibule sa Magure (kat. 3–4) ukazuju na oficijelan karakter tipa 34 A. U arheolo{kim celinama nastalim u vreme izgradwe carske palate po~etkom IV veka, na|ene su tri bronzane fibule tipa 34 B (kat. 10–12), karakteristi~ne za period vladavine Konstantina I i wegove dinastije. Primerci sa Gamzigrada pokazuju da je wihova proizvodwa mogla po~eti ve} tokom prve decenije IV veka, u vreme Druge tetrarhije i gra|anskih ratova izme|u Konstantina i Maksencija. Luksuzna fibula tipa 34 C iz groba 6/06, izra|ena u carskoj radionici u posledwoj tre}ini IV veka i darovana posebnom prigodom, bez sumwe predstavqa odlikovawe, a od trenutka donacije ozna~ava visok status wenog vlasnika (ornamentum dignitatis). Mu{karac sahrawen uz temeq ju`- STARINAR LX/2010 nog bedema Romulijane, sa vojnim ogrta~em (paludamentum) pri~vr{}enim ovom fibulom na desnom ramenu, imao je visok ~in u rimskoj vojsci, mo`da zapovednika odreda stacioniranog u utvr|ewu. Arheolo{ki kontekst i {iroko datovawe celina u kojima su nala`eni primerci tipa 34 D 2, kao i raznovrsnost wihovog ornamenta, ne ukazuju na jasno izdiferenciran oficijelni karakter ovih fibula. Na osnovu klasifikacije i analize ukrasa na ne mo`e se ni{ta zakqu~iti o wihovoj funkciji (Tabela 2). Pretpostavqam da su ih kao oznaku ~ina ili roda u rimskoj vojsci nosili pripadnici odreda stacioniranih u utvr|ewu Romulijane. Radionice u kojima su izra|ivane krstoobrazne fibule iz Romulijane mogu se identifikovati shodno wihovom oficijelnom karakteru. Luksuzni primerci (kat. 9 i 13) izra|eni su u carskim radionicama u Naisu ili Sirmijumu. Rane krstoobrazne fibule, tipa 34 A i 34 B, proizvodile su oficijelne radionice u Sirmijumu i Naisu, a najverovatnije i u Singidunumu, Viminacijumu i Horreum Margi. Krstoobrazne fibule tipa 34 D su izra|ivane u lokalnim radionicama, u okviru odre|ene vojne jedinice iskqu~ivo za wene potrebe. O tome svedo~i lo{ kvalitet izrade i prenatrpan ukras sa velikim brojem upotrebqenih motiva i tehnika. Smisao krstoobraznih fibula ostaje isti od kraja III do sredine V veka: one su oznake anga`mana u carskoj slu`bi. Osoba koja nosi krstobraznu fibulu deluje pod pokroviteqstvom i u slu`bi Rimske imperije i samog cara, bilo da je pripadnik administracije, vojske ili »varvara« povezanih ugovorom sa Carstvom. U horizontu `ivota VI veka u Romulijani nedostaju nalazi fibula tipa 34. Mo`e se pretpostaviti da je posle Justinijanove obnove dunavskog limesa i pograni~nih provincija u Dowem Podunavqu neki drugi tip preuzeo oficijelnu funkciju i zna~ewe krstoobraznih fibula. Postoje indicije da su pojedinii tipovi, sa ranohri{}anskim natpisima i simbolima ili portretima svetaca ili careva, ozna~avali status u crkvenoj hijerarhiji ili rimskoj vojsci. Ovako ukra{ene ranovizantijske fibule nala`ene u celinama VI do po~etka VII veka u Romulijani. 136