LEttER
doi:10.1038/nature15757
Widespread exploitation of the honeybee by early
Neolithic farmers
Mélanie Roffet-Salque1, Martine Regert2, Richard P. Evershed1, Alan K. outram3, Lucy J. E. Cramp1,4, orestes Decavallas5,6,
Julie Dunne1, Pascale Gerbault7,8, Simona Mileto1,9, Sigrid Mirabaud6†, Mirva Pääkkönen1,10, Jessica Smyth1,4, Lucija Šoberl1,11†,
helen L. Whelton1, Alfonso Alday-Ruiz12, henrik Asplund10, Marta Bartkowiak13, Eva Bayer-niemeier14, Lotfi Belhouchet15,
Federico Bernardini16,17, Mihael Budja11, Gabriel Cooney18, Miriam Cubas19†, Ed M. Danaher20, Mariana Diniz21,
László Domboróczki22, Cristina Fabbri23, Jesus E. González-Urquijo19, Jean Guilaine24, Slimane hachi25, Barrie n. hartwell26,
Daniela hofmann27, Isabel hohle28, Juan J. Ibáñez29, necmi Karul30, Farid Kherbouche25, Jacinta Kiely31, Kostas Kotsakis32,
Friedrich Lueth33, James P. Mallory26, Claire Manen24, Arkadiusz Marciniak13, Brigitte Maurice-Chabard34, Martin A. Mc
Gonigle35, Simone Mulazzani36,37, Mehmet Özdoğan30, olga S. Perić38, Slaviša R. Perić38, Jörg Petrasch39, AnneMarie Pétrequin40, Pierre Pétrequin40, Ulrike Poensgen41, C. Joshua Pollard42, François Poplin43, Giovanna Radi23,
Peter Stadler44, harald Stäuble45, nenad tasić46, Dushka Urem-Kotsou47, Jasna B. Vuković46, Fintan Walsh48, Alasdair Whittle49,
Sabine Wolfram50, Lydia Zapata-Peña12‡ & Jamel Zoughlami51
The pressures on honeybee (Apis mellifera) populations, resulting
from threats by modern pesticides, parasites, predators and diseases,
have raised awareness of the economic importance and critical role
this insect plays in agricultural societies across the globe. However,
the association of humans with A. mellifera predates post-industrialrevolution agriculture, as evidenced by the widespread presence
of ancient Egyptian bee iconography dating to the Old Kingdom
(approximately 2400 bc)1. There are also indications of Stone Age
people harvesting bee products; for example, honey hunting is
interpreted from rock art2 in a prehistoric Holocene context and
a beeswax find in a pre-agriculturalist site3. However, when and
where the regular association of A. mellifera with agriculturalists
emerged is unknown4. One of the major products of A. mellifera is
beeswax, which is composed of a complex suite of lipids including
n-alkanes, n-alkanoic acids and fatty acyl wax esters. The
composition is highly constant as it is determined genetically
through the insect’s biochemistry. Thus, the chemical ‘fingerprint’
of beeswax provides a reliable basis for detecting this commodity
in organic residues preserved at archaeological sites, which we now
use to trace the exploitation by humans of A. mellifera temporally
and spatially. Here we present secure identifications of beeswax in
lipid residues preserved in pottery vessels of Neolithic Old World
farmers. The geographical range of bee product exploitation
is traced in Neolithic Europe, the Near East and North Africa,
providing the palaeoecological range of honeybees during
prehistory. Temporally, we demonstrate that bee products were
exploited continuously, and probably extensively in some regions,
at least from the seventh millennium cal bc, likely fulfilling a variety
of technological and cultural functions. The close association of A.
mellifera with Neolithic farming communities dates to the early
onset of agriculture and may provide evidence for the beginnings
of a domestication process.
1
Organic Geochemistry Unit, School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, UK. 2CEPAM – Cultures et Environnements. Préhistoire, Antiquité, Moyen Âge, UMR
7264, Université Nice Sophia Antipolis – CNRS, 06300 Nice, France. 3Department of Archaeology, University of Exeter, Laver Building, North Park Road, Exeter, Devon EX4 4QE, UK. 4Department
of Archaeology and Anthropology, University of Bristol, 43 Woodland Road, Bristol BS8 1UU, UK. 5Université Bordeaux Montaigne, 33607 Pessac, France. 6Laboratoire du Centre de Recherche
et de Restauration des Musées de France (C2RMF), UMR 171, Palais du Louvre, Porte des Lions, 14 Quai François Mitterrand, 75001 Paris, France. 7Research Department of Genetics, Evolution
and Environment, University College London, London WC1E 6BT, UK. 8Department of Anthropology, University College London, London WC1H 0BW, UK. 9Institut für Prähistorische Archäologie,
Freie Universität Berlin, Altensteinstr. 15, Berlin 14195, Germany. 10Department of Archaeology, University of Turku, 20014 Turun Yliopisto, Finland. 11University of Ljubljana, Faculty of Arts,
Department of Archaeology, Aškerčeva 2, box 580, 1000 Ljubljana, Slovenia. 12Department of Geography, Prehistory and Archaeology. University of Basque Country (EHU-UPV), Francisco Tomás
y Valiente s/n, 01006 Vitoria-Gasteiz, Spain. 13Institute of Prehistory, Adam Mickiewicz University, Umultowska 89d, 61-614 Poznań, Poland. 14Museum Quintana – Archäologie in Künzing,
Partnermuseum der Archäologischen Staatssammlung München, Osterhofener Str. 2, 94550 Künzing, Germany. 15Musée Archéologique de Sousse, Rue Marshall Tito, 4000 Sousse, Tunisia.
16
Centro Fermi, Museo Storico della Fisica e Centro di Studi e Ricerche Enrico Fermi, 00184 Rome, Italy. 17Multidisciplinary Laboratory, The Abdus Salam International Centre for Theoretical
Physics, 34151 Trieste, Italy. 18UCD School of Archaeology, University College Dublin, Dublin 4, Ireland. 19International Institute for Prehistoric Research of Cantabria, University of Cantabria,
Avd de los Castros s/n, 39005 Santander, Spain. 20Department of Archaeology, University College Galway, Galway, Ireland. 21UNIARQ-Departamento de História, Faculdade de Letras de Lisboa,
Universidade de Lisboa, 1600-214 Lisboa, Portugal. 22István Dobó Castle Museum, Vár út 1, 3300 Eger, Hungary. 23Dipartimento Civiltà e Forme del Sapere, Università di Pisa, Via Galvani
1, 56126 Pisa, Italy. 24CNRS – UMR 5608 – TRACES, Maison de la recherche, Université Toulouse Jean Jaurès, 5 Allée Antonio Machado, 31058 Toulouse cedex 9, France. 25CNRPAH, Centre
National de Recherche Préhistorique, Anthropologique et Historique, Algiers, Algeria. 26School of Geography, Archaeology and Palaeoecology, Queen’s University Belfast, Belfast BT7 1NN, UK.
27
Universität Hamburg, Archäologisches Institut, Edmund-Siemers-Allee 1, Flügel West, 20146 Hamburg, Germany. 28a.r.t.e.s. Graduate School for the Humanities Cologne, Graduiertenschule der
Philosophischen Fakultät, Aachener Str. 217, 50931 Cologne, Germany. 29IMF-CSIC, Egipciacas 15, 08001 Barcelona, Spain. 30Istanbul University, Faculty of Letters, Department of Prehistory,
34434 Laleli Istanbul, Turkey. 31Eachtra Archaeological Projects, Lickybeg, Clashmore, County Waterford, Ireland. 32School of History and Archaeology, Faculty of Philosophy, Aristotle University of
Thessaloniki, Thessaloniki 54124, Greece. 33German Archaeological Institute, Podbielskiallee 69-71, 14 195 Berlin, Germany. 34Musée Rolin, 3 rue des Bancs, 71400 Autun, France. 35John Cronin
& Associates, 28 Upper Main Street, Buncrana, County Donegal, Ireland. 36Aix-Marseille Université, CNRS, Ministère de la Culture et de la Communication, UMR 7269 LAMPEA, LabexMed, 13284
Marseille, France. 37Dipartimento di Biologia Ambientale, Università degli Studi di Roma La Sapienza, Rome 00185, Italy. 38Institute of Archaeology Belgrade, Kneza Mihaila 35/4 11000 Belgrade,
Serbia. 39Eberhard-Karls-Universität Tübingen, Institut für Ur- und Frühgeschichte und Archäologie des Mittelalters - Abt. Jüngere Urgeschichte und Frühgeschichte - Schloß Hohentübingen,
72070 Tübingen, Germany. 40Maison des Sciences de l’Homme et de l’Environnement C.N. Ledoux, CNRS & Université de Franche-Comté, 32 rue Mégevand, 25030 Besançon Cedex, France.
41
Kämpfenstr. 20, 78315 Radolfzell, Germany. 42Department of Archaeology, Faculty of Humanities, University of Southampton, Avenue Campus, Highfield, Southampton SO17 1BF, UK.
43
Muséum National d’Histoire Naturelle, 55 rue de Buffon, 75005 Paris, France. 44Department of Pre- and Protohistory, University of Vienna, 1190 Vienna, Austria. 45Landesamt für Archaeologie,
Zur Wetterwarte 7, 01109 Dresden, Germany. 46Department of Archaeology, Faculty of Philosophy, Belgrade University, 18–20 Čika Ljubina Street, 11000 Belgrade, Serbia. 47Department of
History and Ethnology, Democritus University of Thrace, Komotini, Greece. 48Irish Archaeological Consultancy, Unit G1, Network Enterprise Park, Kilcoole, County Wicklow, Ireland. 49Department
of Archaeology and Conservation, Cardiff University, John Percival Building, Colum Drive, Cardiff CF10 3EU, UK. 50State Museum of Archaeology Chemnitz, Stefan-Heym-Platz 1, 09111 Chemnitz,
Germany. 51Institut National du Patrimoine de Tunis - Musée archéologique de Carthage, Carthage, Tunisia. †Present addresses: Département des restaurateurs, Institut National du Patrimoine,
124 rue Henri Barbusse, 93300 Aubervilliers, France (S.M.); Laboratório HERCULES, Universidade de Évora, Palácio do Vimioso, Largo Marquês de Marialva 8, 7000-809 Évora, Portugal (L.S.);
BioArCh–University of York, York YO10 5DD, UK (M.C).
‡Deceased.
0 0 M o n t h 2 0 1 5 | Vo L 0 0 0 | nAt U R E | 1
RESEARCH LETTER
The honeybee holds a unique place in human culture. Notwith
standing its presentday economic importance, it has been revered
over the millennia for the sheer beauty and complexity of the social
organization within its colonies. For these reasons the honeybee is the
most researched of the social insects, with its origin being regularly
considered5. The last Ice Age would have had a major effect on the
honeybee with the ice sheets restricting European populations to the
northern Mediterranean hinterlands6. With the glacial retreat, the pop
ulation would have subsequently expanded northwards. However, due
to the lack of a Holocene fossil record7, the honeybee is ecologically
invisible for most of the past 10,000 years.
Intriguingly, this is the period during which Neolithic agriculture
emerged and spread out of southeastern Anatolia and the Levant,
with some human population movement into ecological zones also
conducive to the honeybee. Indeed, progressive woodland clearances
by pioneer prehistoric farmers may have opened up forests, favouring
lightdemanding shrubs, herbs and fruit trees (for example, Rosaceae)8.
Whether this would have exerted negative or positive effects on honey
bee populations is unknown8,9. Given the latter, an opportunity exists to
investigate the presence and early exploitation of the honeybee by pre
historic farming communities through the cultural materials recovered
ALK
a
100
b
0
100
from Neolithic sites, namely their recently invented pottery vessels,
and in doing so, to assess the palaeoecological range of the honeybee
in the Holocene.
Although the most obvious reason for exploiting the honeybee would
be for honey, a rare source of sweetener for prehistoric people, bees
wax would likely have been an equally important material for various
technological, ritual, cosmetic and medicinal applications10. Indeed,
beeswax has been regularly detected in later archaeological and historic
periods in lipid extracts from the fabric of unglazed pottery vessels11
where it is assumed to be a residue of honey use in cooking, or from
the use of vessels for processing wax combs12–14, with beeswax being
absorbed through repeated contacts. Beeswax has also been detected
as a fuel in lamps and in larger vessels used as protobeehives, for
example Roman Greece (second century bc to fourth century ad)15,16
and applied as a postfiring treatment to waterproof vessels17.
The detection of beeswax in archaeological and historic contexts
rests on its complex chemistry providing a unique and relatively
recalcitrant chemical signature. Fresh beeswax comprises a complex
mixture of aliphatic compounds consisting of series of homologues
differing in chainlength by two methylene groups18. Mediumchain
nalkanes range from C23 to C31 (with C27 dominating in A. mellifera),
OH
FA
WE
HWE
TIC
IS
27
29
31
25
c
0
20
AL27
24
18
16
m/z 85
33
23
Relative abundance (%)
O
OH
14
17
15
d
26
18:1
19
22
20
28
21
0
100
32
m/z 73
34
FA24
30
32
24
16
e
30
23
28
OH
26
m/z 103
34
18
0
100
OH30
46
48
O
44
O
42
40
m/z 257
50
f
0
40
WE46
48
50
O
OH
O
46
44
42
52
m/z 117
HWE48
0
12
14
16
18
20
22
24
26
28
30
32
Retention time (min)
Figure 1 | High-temperature gas chromatography/mass spectrometry
chromatograms of total lipid extract of a sherd from Çayönü Tepesi
(6500–6000 cal bc) containing beeswax. a–f, Partial total ion current
chromatogram (a) and mass chromatograms (b–f) displaying ion masses
of characteristic fragments from the main compound classes comprising
the extract (m/z 85, 73, 103, 257 and 117, respectively) with the molecular
structure of the most abundant component for each compound class.
2 | nAt U R E | Vo L 0 0 0 | 0 0 M o n t h 2 0 1 5
Squares, nalkanes (ALK); circles, nalkanoic acids (fatty acids, FA);
triangles, nalkanols (OH); black asterisks, fatty acyl monoesters (WE);
grey asterisks, hydroxyl fatty acyl monoesters (HWE); IS, internal standard
(ntetratriacontane); number n and n:i, acyl carbon number with zero or i
degrees of unsaturations. Compounds shown with a grey background are
interpreted as originating from mammalian animal fats.
18
16
a
TAG48
IS
Relative abundance (%)
0
50
20
26
24 29
30
28
26 31
DAG34
24
28
40
46 48
DAG36
23
DAG32
17
32
MAG18
14
MAG16
30
42
44
IS
b
TAG54
100
TAG50
TAG52
LETTER RESEARCH
46
27
48
44
29
24
25
30
31 28
26
23
0
100
40
32
50
42
33
46
c
48
30
44
16
17
18
19
20 25 2122
22 23
23
25
25
26
27 27
40 42
32
28
31
29 26
27 24 24
IS
28 33
30
31
50
32 34
0
12
14
16
18
20
22
24
26
28
30
32
Retention time (min)
Figure 2 | Partial gas chromatograms of total lipid extracts from
Neolithic sherds from each geographical region. a, Mağura (5500–5200
cal bc). b, Niederhummel (5360–5220 cal bc). c, Gueldaman (fifth
millennium bc). a is interpreted as mixture of animal fats and beeswax;
b and c as pure beeswax. MAG, monoacylglycerols; DAG, diacylglycerols,
TAG, triacylglycerols. Other peak attributions as in Fig. 1.
and nalkanoic acids from C20 to C36. Monoesters comprise predom
inantly alkyl palmitates (C38 to C52), with characteristic hydroxy
monoesters comprising longchain alcohols (C24 to C38) esterified
mainly to hydroxypalmitic acid, ranging between C40 and C54 (ref. 18).
The hydrophobic nature of beeswax makes it relatively resistant to
degradation. Hence, if protected from extensive microbial attack
and/or exposure to high temperatures during anthropogenic manip
ulation, the aforementioned chemical characteristics can be used in
assessing its presence10,19 (Figs 1 and 2).
Adopting this lipid biomarker approach, we now explore the asso
ciation of the honeybee with the spread of early Old World farmers
based on lipid residue analyses of more than 6,400 pottery vessels
(Supplementary Information sections 1 and 2). Combining our new
findings with published occurrences of beeswax in prehistoric pot
tery allows the association between honeybees and early farmers to
be mapped spatially and temporally through prehistory (Figs 3 and 4).
The oldest evidence for beeswax comes from Neolithic sites in
Anatolia dating from the seventh millennium cal bc, as these sites are the
locations of the oldest pottery vessels in Europe and Eurasia. Most of the
assemblages investigated comprised globular or bowl shape ‘cooking’
vessels, an interpretation supported by the finding of ruminant and
porcine animal fats in significant numbers of vessels. No beeswax resi
dues were detected during the intensive investigations of > 380 vessels
from the Levant, although only 34 residues were detected20. Moving
into eastern Anatolia, the site of Çayönü Tepesi revealed two beeswax
residues from 83 vessels from the seventh millennium including an
exceptionally wellpreserved residue containing all the biomarkers of
beeswax (Fig. 1b–f). The free nalkanols, dominated by C30 and C32
homologues, do not occur in fresh beeswax but are a feature of aged
wax, due to hydrolysis of the wax esters. The high abundance of C18:0
fatty acid suggests mixing with mammalian animal fat, the latter being
common in other sherds in the assemblage20. The second sherd from
this site contained a lower concentration of beeswax but all the bio
markers were clearly evident. These two residues establish the easterly
limit of the beeswax detected in this investigation and provide the old
est unequivocal evidence, to our knowledge, of honeybee exploitation
by early Neolithic farmers.
In central Anatolia, extensive investigations of organic residues in
650 vessels, mainly from the site of Çatalhöyük, revealed abundant
animal fat residues. Only one residue showed tentative evidence for
beeswax based on wax esters, dominated by C46 and C48 homologues;
however, the nalkanols do not exhibit the familiar distribution.
nAlkanes were detectable but the distribution is skewed towards
the higher homologues compared to that expected in fresh beeswax,
although such distributional changes are frequently seen in historical
and archaeological beeswax, assumed to arise by sublimation during
ageing or heat treatment10. The tentative identification of this very early
beeswax residue at Çatalhöyük is supported by the discovery of a strik
ing depiction of a honeycomblike pattern painted on a wall at the site21.
Analyses of approximately 570 cooking vessels from northwestern
Anatolia revealed 72 lipid residues of which 4 were identified as con
taining beeswax, from Aşaği Pinar and Toptepe, dating to 5500–5000
cal bc. Although the overall purity of the beeswax (two were mixed with
ruminant fat) and lipid concentrations (20 to 220 μg per gram of sherd)
were quite variable, the distributions were unmistakable. One of the
beeswax finds from Toptepe is well preserved, albeit with ageing evident
from the hydrolytically released free nalkanols and slight distortions
of the various homologous series, through loss of lower homologues.
The most abundant evidence for honeybee exploitation by early
farmers was seen in the rest of the Balkan Peninsula. The full range
of beeswax biomarkers was identified in sherds from bowls, pans and
sieves from the Late Neolithic sites of Paliambela, Greece (4900–4500
cal bc), Măgura, Romania (Fig. 2a; 5500–5200 cal bc) and Drenovac
Turska Česma, Serbia (5300–4700/4600 cal bc). A large number of
beeswax residues were found in Neolithic potsherds (11 residues out
of 81 sherds analysed) from Attica, the Peloponnese and the Cyclades
(Aegean Islands), dating between 5800 and 3000 cal bc, firmly estab
lishing the long tradition of bee exploitation in this region. Overall,
the incidence of beeswax residues is highest in the Balkan Peninsula,
where of the 1,915 Neolithic sherds analysed, 473 yielded lipid residues,
of which 5.5% contained beeswax.
In Central Europe, pure beeswax was recovered from potsherds
from Linearbandkeramik (LBK) sites occupied by the earliest farmers
of Austria and Germany (oldest LBK) including the sites of Brunn am
Gebirge (5500–5400 cal bc) and Niederhummel (5360–5220 cal bc),
pushing back the date for bee exploitation in this region by approxi
mately 1,500 years13 (Fig. 2b). Beeswax was also detected in late sixth
millennium LBK sites of Ludwinowo 7 and Wolica Nowa, Poland17. In
France, the exploitation of bee products is evident during the second
half of the fifth millennium at Chasséen sites (FontJuvénal, Chassey
leCamp and Bercy10) and fourth millennium at the Lake Village sites
of ClairvauxlesLacs (3900 to 3700 bc) and Chalain 3 (ref. 22) and
4 (3200 to 3100 bc and 3040 to 2990 bc). High incidence of bees
wax (approximately 15% of the detectable residues) was identified in
0 0 M o n t h 2 0 1 5 | Vo L 0 0 0 | nAt U R E | 3
RESEARCH LETTER
a
60° N
j′k′
i′
u
g′h′
50° N
a′
t d′ z
v
y
c′e′f′
s
wb′
x
k
e
i
h f
r g
pl
d
q m
o
40° N
n
l′
a
j
b
c
30° N
b
10° W
0°
10° E
20° E
30° E
40° E
a. Çatalhöyük
b. Çayönü Tepesi
c. Tepe Sofalin 11
d. Liménaria 14
e. Măgura
f. Toptepe
g. Dikili Tash 10
h. Aúa÷ı Pınar
i. Drenovac Turska Cesma
j. Ftélia
k. Vinþa Belo Brdo
l. Paliambela
m. Rachmani
n. Kouvéléikès A
o. Profitis Ilias Rizoupolis
p. Vassilara Rachi
q. Théopetra
r. Balkan Export
s. Brunn am Gebirge
t. Niederhummel
u. Kuyavia region 17
v. Chassey-le-Camp
w. Moverna vas 23
x. Font-Juvénal
y. Lonche *
z. Künzing-Unternberg
a′. Bercy 10
b′. Ajdovska jama 23
c′. Clairvaux XIV
d′. Ergolding Fischergasse 13
e′. Chalain 3 22
f′. Chalain 4
g′. Eton rowing lake 12
h′. Runnymede bridge 12
i′. Bulford Torstone
j′. Åle 24
k′. Bjørnsholm 24
l′. Gueldaman
7000
6000
5000
4000
3000
2000
Time (cal. years BC)
Figure 3 | Geographical distribution of prehistoric sites in the date
range 7500 and 2000 cal bc yielding beeswax residues. a, Locations of
archaeological sites. b, Chronology of beeswax use in the Near East, the
Balkan Peninsula, mainland Europe, Scandinavia, the UK and northern
Africa. Neolithic finds in black, preNeolithic (huntergatherer contexts)
in light grey and Bronze Age in dark grey. * Dental filling reexamined
after ref. 31.
fifth millennium sherds from two Slovenian sites (Ajdovska jama and
Moverna vas)23.
Around 130 sherds have so far been analysed from the Iberian
Peninsula. However, no beeswax residues have yet been detected,
although the overall preservation of organic residues was poor. Further
investigations will likely reveal examples of beeswax in Neolithic
pottery from this region.
The northerly limit of bee exploitation in northern Europe appears
to be Denmark with two beeswax finds in late Mesolithic and Neolithic
contexts24. Around 5° to the south in southern Britain, beeswax is
evident in 7 vessels amongst the approximately 670 Neolithic vessels
analysed. These findings clearly counter any arguments for a late intro
duction of the honeybee into the British Isles8,25. Interestingly, however,
investigations of nearly 1,200 Mesolithic and Neolithic vessels from
Ireland, Scotland and Fennoscandia26,27 have failed to reveal any con
clusive evidence of beeswax (Supplementary Information section 3).
Given that organic residue preservation in these regions is excellent,
the lack of beeswax would seem to establish the ecological limit of
A. mellifera at that time. Similar arguments are likely to account for
the absence of beeswax residues from >350 prehistoric pottery vessels
from the Eurasian Steppe28.
Finally, we report the first evidence for bee exploitation by Neolithic
pastoralists in North Africa. The analysis of 71 sherds from the Algerian
site of Gueldaman revealed a single wellpreserved beeswax residue
(fifth millennium bc). The preservation is again exceptional with
nalkanes, nfatty acids and fatty acyl wax ester distributions providing
an unequivocal identification of beeswax. The presence of free long
chain nalkanols and lack of hydroxy fatty acid wax esters are indicative
of diagenesis and/or userelated alteration. However, the overall dis
tribution indicates the wax residue derives from A. mellifera (Fig. 2c).
In conclusion, the approximately 50 new finds of beeswax residues
considered above provide evidence for the widespread exploitation
of the honeybee by the early agriculturalists and pastoralists of the
Near East, Europe and North Africa dating back nearly 9,000 years
4 | nAt U R E | Vo L 0 0 0 | 0 0 M o n t h 2 0 1 5
LETTER RESEARCH
Online Content Methods, along with any additional Extended Data display items and
Source Data, are available in the online version of the paper; references unique to
these sections appear only in the online paper.
11. Mayyas, A. S., Al-Qudah, M. A., Douglas, K. A. & Al-Ajlouny, F. K. Beeswax
preserved in archaeological ceramics: function and use. Annals of Faculty of
Arts. Ain Shams University 40, 343–371 (2012).
12. Copley, M. S. et al. Dairying in antiquity. III. Evidence from absorbed lipid
residues dating to the British Neolithic. J. Archaeol. Sci. 32, 523–546 (2005).
13. Heron, C., Nemcek, N., Bonield, K. M., Dixon, D. & Ottaway, B. S. The chemistry
of Neolithic beeswax. Naturwissenschaften 81, 266–269 (1994).
14. Decavallas, O. in Cooking up the Past: Food and Culinary Practices in the
Neolithic and Bronze Age Aegean (eds Mee, C. & Renard, J.) 148–157 (Oxford
Books Limited, 2007).
15. Evershed, R. P., Vaughan, S. J., Dudd, S. N. & Soles, J. S. Fuel for thought?
Beeswax in lamps and conical cups from Late Minoan Crete. Antiquity 71,
979–985 (1997).
16. Evershed, R. P., Dudd, S. N., Anderson-Stojanovic, V. R. & Gebhard, E. R. New
chemical evidence for the use of combed ware pottery vessels as beehives in
ancient Greece. J. Archaeol. Sci. 30, 1–12 (2003).
17. Salque, M. et al. Earliest evidence for cheese making in the sixth millennium BC
in northern Europe. Nature 493, 522–525 (2013).
18. Aichholz, R. & Lorbeer, E. Investigation of combwax of honeybees with
high-temperature gas chromatography and high-temperature gas
chromatography–chemical ionization mass spectrometry: I. high-temperature
gas chromatography. J. Chromatogr. A 855, 601–615 (1999).
19. Garnier, N., Cren-Olivé, C., Rolando, C. & Regert, M. Characterization of
archaeological beeswax by electron ionization and electrospray ionization
mass spectrometry. Anal. Chem. 74, 4868–4877 (2002).
20. Evershed, R. P. et al. Earliest date for milk use in the Near East and
southeastern Europe linked to cattle herding. Nature 455, 528–531 (2008).
21. Mellaart, J. Excavations at Çatal Hüyük, 1962: second preliminary report.
Anatolian Studies 13, 43–103 (1963).
22. Regert, M., Dudd, S. N., Van Bergen, P. F., Pétrequin, P. & Evershed, R. P.
Investigations of solvent extractable lipids and insoluble polymeric
components: organic residues in Neolithic ceramic vessels from Chalain (Jura,
France). British Archaeological Reports 939, 78–90 (2001).
23. Šoberl, L. et al. Neolithic and Eneolithic activities inferred from organic residue
analysis of pottery from Mala Triglavca, Moverna vas and Ajdovska jama,
Slovenia. Documenta Praehistorica 41, 149–179 (2014).
24. Heron, C., Craig, O. E., Forster, M. & Stern, B. & Andersen, S. H. in Shell Middens
in Atlantic Europe (eds Milner, N., Craig, O. E. & Bailey, G. N.) Ch. 7, 78–85
(Oxbow Books, 2007).
25. Carreck, N. Are honey bees (Apis mellifera L.) native to the British Isles?
J. Apic. Res. 47, 318–322 (2008).
26. Cramp, L. J. E. et al. Immediate replacement of ishing with dairying by the
earliest farmers of the northeast Atlantic archipelagos. Proc. R. Soc. Lond. B
281, 20132372 (2014).
27. Cramp, L. J. E. et al. Neolithic dairy farming at the extreme of agriculture in
northern Europe. Proc. R. Soc. Lond. B 281, 20140819 (2014).
28. Outram, A. K. et al. The earliest horse harnessing and milking. Science 323,
1332–1335 (2009).
29. Bartlein, P. J. et al. Pollen-based continental climate reconstructions at 6 and
21 ka: a global synthesis. Clim. Dyn. 37, 775–802 (2011).
30. Abou-Shaara, H. F. The foraging behaviour of honey bees, Apis mellifera: a
review. Vet. Med. (Praha) 59, 1–10 (2014).
31. Bernardini, F. Beeswax as dental illing on a Neolithic human tooth. PLoS ONE
7, e44904 (2012).
Received 23 April; accepted 29 September 2015.
Supplementary Information is available in the online version of the paper.
0
0.02
0.05
60° N
0.1
0.15
0.2
0.25
50° N
40° N
30° N
1
10° W
0°
10° E
20° E
30° E
40° E
Figure 4 | Regional distribution of beeswax residues in potsherd
lipid extracts. Interpolated map of Old World beeswax occurrences
(proportion of beeswax residues per number of residues in pottery
sherds, in percentages) during the Neolithic (including the Mesolithic
sites available). Colours and colour key show the proportions of beeswax
residues estimated by surface interpolation, where collection locations are
represented by dots (n = 154).
(Fig. 3). In all these regions the new data have either provided the first
evidence of honeybee exploitation in a region, as in North Africa, or
pushed the chronology of human–honeybee association to substantially
earlier dates, as in Anatolia and Central Europe (Fig. 3b). The lack of
evidence for beeswax use at Neolithic sites north of the 57th parallel
North may suggest an ecological limit to the natural occurrence of
honeybees. Indeed, harsh highlatitude conditions, even with temper
atures warmer than today29, would affect the foraging capabilities of
honeybees30. Critically, in the absence of a Holocene fossil record for
A. mellifera7 these findings provide the first ancient biomoleculebased
palaeoecological map of the distribution of an economically and cul
turally important animal (Fig. 4).
1.
Crane, E. The World History of Beekeeping and Honey Hunting (Duckworth,
1999).
2. Dams, M. & Dams, L. R. Spanish art rock depicting honey gathering during the
Mesolithic. Nature 268, 228–230 (1977).
3. d’Errico, F. et al. Early evidence of San material culture represented by organic
artifacts from Border Cave, South Africa. Proc. Natl Acad. Sci. USA 109,
13214–13219 (2012).
4. Larson, G. & Fuller, D. Q. The evolution of animal domestication. Annu. Rev.
Ecol. Evol. Syst. 45, 115–136 (2014).
5. Wallberg, A. et al. A worldwide survey of genome sequence variation provides
insight into the evolutionary history of the honeybee Apis mellifera.
Nature Genet. 46, 1081–1088 (2014).
6. Ruttner, F. Biogeography and Taxonomy of Honeybees (Springer, 1987).
7. Buckland, P. I. & Buckland, P. C. in Versions: BugsCEP v7.63; Bugsdata v8.01;
BugsMCR v2.02; BugStats v1.22 (2006).
8. Limbrey, S. in Archaeological Aspects of Woodland Ecology (eds Bell, M. &
Limbrey, S.) 279–286 (British Archaeological Reports, 1982).
9. Wilson, E. O. The Insect Societies (Belknap Press of Harvard Univ. Press,
1971).
10. Regert, M., Colinart, S., Degrand, L. & Decavallas, O. Chemical alteration and
use of beeswax through time: accelerated ageing tests and analysis of
archaeological samples from various environmental contexts. Archaeometry
43, 549–569 (2001).
Acknowledgements We thank the UK Natural Environment Research Council
for partial funding of the mass spectrometry facilities at Bristol (contract no.
R8/H10/63; http://www.lsmsf.co.uk) and English Heritage, European Research
Council, Leverhulme Trust, Ministère de la Culture et de la Communication,
Ministère de l’Enseignement Supérieur et de la Recherche (ACI Jeunes
Chercheurs), Natural Environment Research Council, Région PACA, Royal
Society and Wellcome Trust for funding.
Author Contributions M.R.-S., M.R., R.P.E. and A.K.O. conceived and planned the
project about beeswax in prehistory. M.R.-S., M.R. and R.P.E. wrote the paper.
M.R.-S., M.R., L.J.E.C., O.D., J.D., S.Mil., S.Mir., M.P., J.S., L.S., H.L.W., M.Bart. and
D.U.-K. undertook planning of regional lipid residue analyses projects, sampling,
analytical work and data analysis. P.G. created Figure 4 and Supplementary
Information section 3. All other authors either directed excavations or provided
expertise in relation to pottery collections and essential insights into the study
region and sites.
Author Information Reprints and permissions information is available
at www.nature.com/reprints. The authors declare no competing financial
interests. Readers are welcome to comment on the online version of the
paper. R code is available upon request to P.G. (p.gerbault@ucl.ac.uk).
Correspondence and requests for materials should be addressed to
M.R.-S. (melanie.salque@bristol.ac.uk), M.R. (martine.regert@cepam.cnrs.fr)
or R.P.E. (r.p.evershed@bristol.ac.uk).
0 0 M o n t h 2 0 1 5 | Vo L 0 0 0 | nAt U R E | 5
RESEARCH LETTER
METHODS
Lipid residue analyses. All solvents used were HPLC grade (Rathburn) and the
reagents were analytical grade (typically > 98% of purity).
A subsample (1 to 3 g) from archaeological potsherds was cleaned with a model
ling drill to remove any exogenous lipids (from the soil and handling) and crushed
with a solventwashed mortar and pestle. An internal standard (ntetratriacontane,
typically 20 μ g) was added to the powdered sherd to enable the quantification of
lipid extract. Ground samples of sherds were extracted with CHCl3/MeOH (2:1
(vol/vol), 2 × 10 ml) using ultrasonication. Both supernatants were combined and
the solvent was removed under a gentle stream of nitrogen at 40 °C. Aliquots of the
total lipid extract (TLE) were treated with 40 μl of N,Obis(trimethylsilyl)trifluoro
acetamide (BSTFA) containing 1% trimethylchlorosilane (Sigma Aldrich) for 1 h
at 70 °C and the BSTFA in excess evaporated under a gentle stream of nitrogen.
The trimethysilylated TLE was diluted in hexane (typically 50 to 150 μ l) and sub
mitted to analysis by hightemperature gas chromatography (HTGC) and high
temperature gas chromatographymass spectrometry (HTGC/MS) to identify the
major compounds present.
All TLEs were initially screened in a HewlettPackard 5890 Series II gas chro
matograph equipped with a fusedsilica capillary column (15 m × 0.32 mm) coated
with dimethyl polysiloxane stationary phase (DB1HT; film thickness, 0.1 μ m;
Agilent Technologies). Derivatized extracts (1.0 μl) were injected oncolumn using
a cool oncolumn inlet in track oven mode. The temperature was held isothermally
for 2 min at 5 °C and then increased at a rate of 10 °C min−1 and held at 350 °C for
10 min. The flame ionization detector (FID) was set at a temperature of 350 °C.
Helium was used as a carrier gas and maintained at a constant flow of 4.6 ml min−1.
Data acquisition and processing were carried out using the HP Chemstation soft
ware (Rev. B.03.02 (341), Agilent Technologies).
HTGC/MS analyses of trimethylsilylated aliquots were performed using a
Thermo Scientific Trace 1300 gas chromatograph coupled with an ISQ single
quadrupole mass spectrometer. Diluted samples were introduced using a PTV
injector in split mode (split flow of 30 ml min−1, split ratio of 6.0) onto a 0.53 mm
fused silica precolumn connected to a 15 m × 0.32 mm i.d. fusedsilica capil
lary column coated with dimethyl polysiloxane stationary phase (Rxi1HT; film
thickness, 0.1 μ m; Restek). The initial injection port temperature was 50 °C with
an evaporation phase of 0.05 min, followed by a transfer phase from 50 °C to
380 °C at 0.2 °C min−1. The oven temperature was held isothermally for 2 min at
50 °C, increased at a rate of 10 °C min−1 to 280 °C, then at a rate of 25 °C min−1 to
380 °C and finally held at 380 °C for 5 min. Helium was used as a carrier gas and
maintained at a constant flow 5 ml min−1. The mass spectrometer was operated
in the electron ionization (EI) mode (70 eV) with a GC interface temperature of
380 °C and a source temperature of 340 °C. The emission current was 50 μA and the
mass spectrometry set to acquire in the range of m/z 50–950 Daltons at two scans
per second. Data acquisition and processing were carried out using the Thermo
XCalibur software (version 3.0.63). Peaks were identified on the basis of their mass
spectra, gas chromatography (GC) retention times, by comparison with the NIST
mass spectral library (version 2.0) and by comparison with modern beeswax (from
the Loire department, France).
Construction of Fig. 4. The total number of archaeological sites investigated is
166, but only 154 of these fell within the geographical area of interest (longitude
− 10° to 42° and from latitude 25° to 62°, see Supplementary Information section
1). To estimate the distribution of beeswax residues in continuous space from irreg
ularly spaced data, linear interpolation was performed in the triangles bounded
by data points32,33. The output grid was made of 530 × 380 points evenly spaced
over the range of latitude and longitude. No extrapolation was being used. Kriging
was used to narrow the interpolation values to locations around data points (and
not show interpolation values where there is no data). Kriging allows to obtain
weights of the prediction locations based on the distance between data points, with
lower variance where data points are and higher variance where there is no data.
Interpolation, kriging and plotting were all performed in R version 2.15.1 (ref. 34).
Interpolation was performed using the function ‘interp’ from the package ‘akima’
(CRAN repository, http://cran.rproject.org/web/packages/akima/akima.pdf).
Kriging was performed using the function 'krige.conv' from the package ‘geoR'’
(CRAN repository, http://cran.rproject.org/web/packages/geoR/geoR.pdf, further
information on the package ‘geoR’ can be found at http://www.leg.ufpr.br/geoR).
R code available upon request to P.G.
32. Akima, H. A method of bivariate interpolation and smooth surface itting for
irregularly distributed data points. ACM Trans. Math. Softw. 4, 148–159
(1978).
33. Akima, H. Algorithm 761: scattered-data surface itting that has the accuracy
of a cubic polynomial. ACM Trans. Math. Softw. 22, 362–371 (1996).
34. The R Core Team. R: a Language and Environment for Statistical Computing.
(R Foundation for Statistical Computing, 2012).
LETTER RESEARCH
Web
summary
Detection of molecular biomarkers characteristic of beeswax in pottery vessels at archaeological
sites reveals that humans have exploited bee products (such as beeswax and honey) at least
9,000 years ago since the beginnings of agriculture.