Version classiqueVersion mobile
OpenEdition Books

Les vaisseaux du désert et des steppes

Damien Agut-Labordère
Bérangère Redon

Domestication, diffusion and hybridization of the Bactrian camel

A zooarchaeological perspective

Rémi Berthon, Marjan Mashkour, Pamela Burger et Canan Çakırlar

Texte intégral

  • 1 See for instance the articles of L. Cousin, H.‑P. Francfort and B. Lafont in this volume.

1The two‑humped camel, which the Greeks and Romans called the Bactrian camel, is a mystery, when it comes to its zooarchaeological record. The archaeological evidence for its pre‑domestication biogeography, domestication centre, diffusion, and finally its hybridization with the dromedary is extremely scarce. This situation is in great contrast with the ubiquity of its iconographic representations from Scythia to China to Persia since the Bronze Age.1 In this chapter we discuss the state of current research into the zooarchaeology of the Bactrian camel, with specific reference to research problems.

A brief “natural” history of the two‑humped camel

  • 2 Van der Made et al. 2002.
  • 3 Wu et al. 2014.
  • 4 Przewalski 1878.
  • 5 Gentry, Clutton‑Brock and Groves 2004.
  • 6 Ji et al. 2009; Jianlin et al. 2000; Silbermayr et al. 2010.
  • 7 Burger 2016.

2The ancestors of the old‑world camels migrated from North America to Asia via the Bering land bridge. Camels spread into the Old World after 6.3 Ma and before 5.8 Ma.2 The divergence between one‑humped and two‑humped camels is estimated, using the molecular clock, to around 4.4 Ma.3 Two extant forms of two‑humped camels are recognized: the wild one, and the domestic one. Nikolaï Przewalski first described wild Bactrian camels,4 and the International Commission of Nomenclature fixed in 2003 the first available specific name based on a wild population as Camelus ferus to the wild Bactrian camel.5 The domestic Bactrian camel is named Camelus bactrianus following Linnae (1758). It was however heavily debated if the last wild two‑humped camels in the Mongolian Gobi and the Chinese Taklimakan and Lop Nor deserts were feral or an evolutionary unique unit. Molecular genetic analyses demonstrated the divergence between wild and domestic Bactrian camels and estimated the time of separation between 0.7 and 1.5 Ma in the Pleistocene,6 long before any domestication event. This long‑term divergence excludes the present‑day wild two‑humped camels as direct ancestors of modern domestic Bactrian camels.7

3Bactrian camels were therefore domesticated from a distinct, and now extinct, wild population. The big question is when, where and how this domestication occurred.

The domestication of the two‑humped camel

4Although Burger’s (2016) finding that the extant wild and domestic populations of the two‑humped camel have nothing to do with each other is an important breakthrough to advance investigations into the origins of Bactrian camels, there is really little much else to get further grip on. The tools available to us are rare specimens of camel bones and teeth, DNA and ancient DNA.

  • 8 Steiger 1990.
  • 9 Compagnoni and Tosi 1978.
  • 10 Peters and Driesch 1997.
  • 11 Buckley 2018.

5As a preamble to the presentation of the current knowledge about Bactrian camel domestication, it should be stressed that the study of the domestication process is difficult due to the generally rare occurrence of camel bones in the archaeological sites. Given the association of camels with nomadic life styles, even today, this is perhaps to be expected. Archaeological sites which could potentially yield early camel remains tend to be shallow, representing ephemeral settlements rather than multi‑period occupational remains like tell sites. Moreover, it is clear that there are large archaeological research lacunae in regions where we should expect Bactrian domestication, such as the Central Asian former Soviet Union states, Afghanistan, and even Mongolia. In these areas, the density of excavated Neolithic or Chalcolithic occupations is very low. In the absence of large assemblages to study, it is difficult to establish the osteomorphological criteria to sort out the wild Bactrian camels remains from the domestic ones. Furthermore, the distinction between the bones of Bactrian camel and dromedary (and their hybrids) is also challenging. The first comprehensive study that aimed at sorting the Bactrian camels’ skeletal elements from the dromedaries’ ones dates back to the late 1980s.8 As a consequence, any Bactrian camel remains published before at least 1990 without mention of specific identification criteria should be checked. For example, a calcaneus supposedly from Phase 6 (ca 2600‑2500 BCE) at Shahr‑i Sokhta, a key settlement site for the early evidence for domestic Bactrians in the archaeological record of Iran,9 almost certainly belongs to the dromedary10 and likely represents intrusive remains from later occupation levels. Such stratigraphic issues could be solved by direct radiocarbon dating of remains identified as early domestic Bactrian camels. Mass spectrometry analysis of collagen fingerprints11 can also be employed for assessing the taxonomic status of Camel remains predating any possible hybridization event (see below).

  • 12 Trinks et al. 2012.
  • 13 Peters and Driesch 1997.
  • 14 Vigne and Debue forthcoming.

6A palaeogenetic study using mitochondrial DNA evidenced a high homogeneity between Late Bronze Age and Iron Age samples from Siberia and Uzbekistan, and modern domesticated Bactrian camels from China and Mongolia. It led to the conclusion that there was a single domestication for Bactrian camels.12 Where this domestication process took place remains uncertain but it is closely related to the question of the geographical distribution of the wild Bactrian camels prior to domestication. It has been stressed by J. Peters and A. von den Driesch that no Bactrian camel remain has been found in Mesolithic and Neolithic archaeozoological assemblages from southern and western Turkmenistan, western Iran, southeastern Uzbekistan and northern Afghanistan.13 Domestication should have taken place further east such as southern Kazakhstan, northwestern China or western Mongolia. More recently, wild Bactrian camel remains were discovered in the sixth millennium BCE site of Ajakagytma in central Uzbekistan.14 The area of distribution of the wild Bactrian camels and therefore the potential location of its centre of domestication should be extended up to the Kyzylkum desert.

7Since we are lacking archaeological sites where the domestication of the Bactrian camel could have occurred, we need to use another approach to determine when the Bactrian camel was domesticated. Looking at the spread of the Bactrian camel outside the distribution area of its wild ancestor can provide a terminus ante quem date to the beginning (or start) of domestication.

The westward and southward spread of the domestic Bactrian camel

  • 15 Moore, Ermolova and Forstern 2003.
  • 16 Kasparov 2006.
  • 17 Compagnoni and Tosi 1978.
  • 18 Mashkour 2001.

8The earliest Bactrian camel remains, outside the distribution area of the wild Bactrian camel, are to be found in southern Turkmenistan. At the site of Anau, a few Bactrian camel remains were encountered in levels dated as far back as the early fourth millennium BCE.15 Remains of domestic Bactrian camels were also present in the nearby site of Chong‑depe during the late fourth millennium BCE.16 Since none of the above‑mentioned domestic Bactrian camel remains were directly radiocarbon dated, the chronology of the spread thus remains hypothetic. Further south, in eastern Iran, the site of Shahr‑i Sokhta yielded camel bones as well as camel dung and camel fibers from layers dated to the early third millennium BCE.17 Contemporary to the Shahr‑i Sokhta findings, three Bactrian camel bones were identified at Tepe Qabrestan in northern Iran.18 If the spread of the domestic camel towards adjacent territories of the domestication center can be dated as early as the fourth millennium BCE, evidence for its expansion further west is very patchy.

  • 19 Zeder 1991.
  • 20 Becker 2008.
  • 21 Sapir‑Hen and Ben‑Yosef 2013.

9Domestic Bactrian camels reached south‑western Iran in late second millennium BCE as it is attested by 14 remains from the Qaleh Phase at Tall‑i Malyan.19 Although camel could be invisible in the archaeozoological records if they were not consumed, our current knowledge points towards a very slow spread of the Bactrian camel towards southwestern Near East. This might reflect difficulties to adapt the animal to new environmental conditions. Bactrian camels reached Mesopotamia roughly at the same time as south‑western Iran. At the site of Tell Sheikh Hamad in eastern Syria, nine domestic Bactrian camel bones were found in contexts dated to the 13th century BCE.20 At the same site, dromedaries appear only in the subsequent phase dated to the 9th‑6th century BCE. The arrival of dromedary in northern Mesopotamia from the 9th century BCE onwards is consistent with the dating of the arrival of dromedary in the southern Levant, not earlier than the last third of the 10th century BCE.21

10Once Bactrian camel and dromedary have reached the same territories, the question of their hybridization raises.

The hybridization of the domestic Bactrian camel and the dromedary

  • 22 Kolpakow 1935.
  • 23 Tapper s.d.
  • 24 Potts 2004.
  • 25 See litterature in Çakırlar and Berthon 2014.
  • 26 Mashkour 1997; Uerpmann 1999.
  • 27 Ruiz et al2015.
  • 28 Galik et al. 2015.

11The Bactrian camel (Camelus bactrianus) and the dromedary (Camelus dromedarius) belong to different species. Their hybridization is however possible and the first‑generation hybrids are mostly fertile.22 The most common hybrid is the offspring of Bactrian camel studs and dromedary females. Hybrids are particularly large and combine boons from both parents. Thus, they are favored for the caravan trade.23 The occurrence of hybrids is extremely likely for any time period and region where the human‑induced geographical overlap between the two species is expected. As shown above, this is the case in Mesopotamia from the beginning of the first millennium BCE onwards.24 Although bones of hybrid specimens are particularly large and present mixed non‑metric morphological features, the identification of hybrid in faunal assemblages from archaeological settlements is a challenge.25 The earliest hybrids identified so far were found in Mleiha (United Arab Emirates) and dated to the first two centuries AD.26 That is quite late and far from the area where dromedaries and Bactrian camels first encountered. It is therefore necessary to use new approaches for providing direct evidence of hybridization on rare and fragmented camelid remains from the 13th century BCE onwards. Much is to be expected from palaeogenetic analyses,27 which already helped confirming the hybrid status of an Ottoman army camel dated to the 17th century AD and found in the vicinity of Vienna in Austria.28 Furthermore, ancient DNA analyses can inform on the variety (i.e. paternal and maternal lineage) of the hybrid specimen. Such data is crucial to understand hybridization strategies in the past and cannot be achieved with anatomical and morphometrical analyses. But the low preservation of aDNA poses a problem especially in arid environments where camel remains typically occur, leaving enough ground not to rely solely on aDNA to identify hybrids and purebreds in the archaeological record.


12There is not much to summarize and conclude from the above zooarchaeological review. Region of domestication is not clear yet. The diffusion rate of the Bactrian camel outside of the distribution area of its wild ancestor is somewhat clearer despite the little evidence available. For later periods the picture is blurred by hybridization with the dromedary. Our knowledge of the domestication and early spread of the Bactrian camel will not improve without the discovery of new material. Osteometric approaches can be efficient and palaeogenetic approaches are very promising but we need bones to study. Our research questions depend on an increase of archaeological and zooarchaeological investigations on fifth and early fourth millennium settlements in central Asia and northeastern Iran. It is however good to remember that purebred Bactrian males, whatever our ignorance of their domestication history, are highly‑prized, even today, as studs.


Becker C. 2008, “Die Tierknochenfunde aus Tall Seh Hamad/Dur‑Katlimmu – eine zoogeographisch‑haustierkundliche Studie”, in H. Kühne (ed.), Umwelt und Subsistenz der assyrischen Stadt Dur‑Katlimmu am unteren Habur, Wiesbaden, p. 61‑131.

Buckley M. 2018, “Zooarchaeology by mass spectrometry (ZooMS) collagen fingerprinting for the species identification of archaeological bone fragments”, in C.M. Giovas and M.J. LeFebvre (ed.), Zooarchaeology in practice: Case studies in methodology and interpretation in archaeofaunal analysis, p. 227‑247.

Burger P.A. 2016, “The history of Old World camelids in the light of molecular genetics”, Tropical Animal Health and Production 48, p. 905‑913.

Çakirlar C. and Berthon R. 2014, “Caravans, camel wrestling and cowrie shells: towards a social zooarchaeology of camel hybridization in Anatolia and adjacent regions”, Anthropozoologica 49, p. 237‑252.

Compagnoni B. and Tosi M. 1978, “The camel: its distribution and state of domestication in the Middle East during the third millennium BC in light of finds from Shahr‑I Sokhta”, in R.H. Meadow and M.A. Zeder (ed.), Approaches to faunal analysis in the Middle East, Cambridge, p. 91‑103.

Galik A., Mohandesan E., Forstenpointer G., Scholz U.M., Ruiz E., Krenn M. and Burger P. 2015, “A sunken ship of the desert at the River Danube in Tulln, Austria”, PLoS ONE 10, p. e0121235.

Gentry A., Clutton‑Brock J. and Groves C.P. 2004, “The naming of wild animal species and their domestic derivatives”, Journal of Archaeological Science 31, p. 645‑651.

Ji R., Cui P., Ding F., Geng J., Gao H., Zhang H., Yu J., Hu S. and Meng H. 2009, “Monophyletic origin of domestic bactrian camel (Camelus bactrianus) and its evolutionary relationship with the extant wild camel (Camelus bactrianus ferus)”, Animal Genetics 40, p. 377‑382.

Jianlin H., Mburu D., Ochieng J., Kaufmann B., Rege J.E.O. and Hanotte O. 2000, “Application of New World Camelidae microsatellite primers for amplification of polymorphic loci in Old World camelids”, Animal Genetics 31, p. 404‑406.

Kasparov A. 2006, Skotovodstvo i okhota epokhi neolita – paleometalla v Juzhnom turkmenistane, Moscow.

Kolpakow V.N. 1935, “Über Kamelkreuzungen”, Berliner Tierärztliche Wochenschrift 51, p. 617‑622.

Linnaeus C. 1758, Systema Naturae per regna tria naturae, secundum classis, ordines, genera, species cum characteribus, differentiis, synonymis, locis, vol. 1, Stockholm (10th ed.).

Mashkour M. 1997, “The funeral rites at Mleiha (Sharja‑U.A.E.): The camelid graves”, Anthropozoologica 25‑26, p. 725‑736.

Mashkour M. 2001, Chasse et élevage du Néolithique à l’âge du Fer dans la plaine de Qazvin (Iran). Étude archéozoologique des sites Zagheh, Qabrestan et Sagzabad, PhD, University of Paris I Panthéon‑Sorbonne (unpublished).

Moore K.M., Ermolova N.M. and Forstern A. 2003, “Animal herding, hunting, and the history of animal domestication at Anau Depe”, in F.T. Hiebert and K. Kurbansakhhatov (ed.), A Central Asian village at the dawn of civilization, excavations at Anau, Turkmenistan, Philadelphia, p. 154‑159.

Peters J. and Driesch A. von den 1997, “The two‑humped camel (Camelus bactrianus): new light on its distribution, management and medical treatment in the past”, Journal of Zoology 242, p. 651‑679.

Potts D.T. 2004, “Camel hybridization and the role of Camelus bactrianus in the Ancient Near East”, Journal of the Economic and Social History of the Orient 47, p. 143‑165.

Przewalski N.M. 1878, From Kul’dzha through Tyan’‑Shan’ to Lob‑Nor (in Russian), St Petersburg.

Ruiz E., Mohandesan E., Fitak R.R. and Burger P.A. 2015, “Diagnostic single nucleotide polymorphism markers to identify hybridization between dromedary and Bactrian camels”, Conserv Genet Resour 7, p. 329‑332.

Sapir‑Hen L. and Ben‑Yosef E. 2013, “The introduction of domestic camels to the Southern Levant: Evidence from the Aravah Valley”, Tel Aviv 40, p. 277‑285.

Silbermayr K., Orozco‑Terwengel P., Charruau P., Enkhbileg D., Walzer C., Vogl C., Schwarzenberger F., Kaczensky P. and Burger P.A. 2010, “High mitochondrial differentiation levels between wild and domestic Bactrian camels: a basis for rapid detection of maternal hybridization”, Animal Genetics 41, p. 315‑318.

Steiger C. 1990, Vergleichend morphologische Untersuchungen an Einzelknochen des poscranialen Skeletts der Altweltkamele, PhD, Ludwig‑Maximilians‑Universität (unpublished).

Tapper R. s.d., One hump or two? Hybrid camels and pastoral cultures: An update.

Trinks A., Burger P.A., Benecke N. and Burger J. 2012, “Simulations of populations ancestry of the two‑humped camel (Camelus bactrianus)”, in E. Knoll and P. Burger (ed.), Camels in Asia and North Africa. Interdisciplinary perspectives on their significance in past and present, Vienna, p. 79‑86.

Uerpmann H.‑P. 1999, “Camel and horse skeletons from protohistoric graves at Mleiha in the Emirate of Sharjah (U.A.E.)”, Arabian Archaeology and Epigraphy 10, p. 102‑118.

Van Der Made J., Morales J., Sen S. and Aslan F. 2002, “The first camel from the Upper Miocene of Turkey and the dispersal of the camels into the Old World”, Comptes Rendus Palevol 1, p. 117‑122.

Vigne J.‑D. and Debue K. forthcoming, “La faune de vertébrés d’Ajakagytma (Ouzbékistan; Kel’teminar, 8500‑6000 cal BP)”, in F. Brunet and M. Khudzhanazarov (ed.), La néolithisation en Ouzbékistan. Étude de la culture de Kelteminar (VIIe‑Ve millénaires). Le site d’Ajakagytma, Paris.

Wu H., Guang X., Al‑Fageeh M.B., Cao J., Pan S., Zhou H., Zhang L., Abutarboush M.H., Xing Y., Xie Z., Alshanqeeti A.S., Zhang Y., Yao Q., AlShomrani B.M., Zhang D., Li J., Manee M.M., Yang Z., Yang L., Liu Y., Zhang J., Altammami M.A., Wang S., Yu L., Zhang W., Liu S., Ba L., Liu C., Yang X., Meng F., Wang S., Li L., Li E., Li X., Wu K., Zhang S., Wang J., Yin Y., Yang H., AlSwailem A.M. and Wang J. 2014, “Camelid genomes reveal evolution and adaptation to desert environments”, Nature Communications 5, p. 5188.

Zeder M.A. 1991, Feeding cities: Specialized animal economy in the Ancient Near East, Washington.


1 See for instance the articles of L. Cousin, H.‑P. Francfort and B. Lafont in this volume.

2 Van der Made et al. 2002.

3 Wu et al. 2014.

4 Przewalski 1878.

5 Gentry, Clutton‑Brock and Groves 2004.

6 Ji et al. 2009; Jianlin et al. 2000; Silbermayr et al. 2010.

7 Burger 2016.

8 Steiger 1990.

9 Compagnoni and Tosi 1978.

10 Peters and Driesch 1997.

11 Buckley 2018.

12 Trinks et al. 2012.

13 Peters and Driesch 1997.

14 Vigne and Debue forthcoming.

15 Moore, Ermolova and Forstern 2003.

16 Kasparov 2006.

17 Compagnoni and Tosi 1978.

18 Mashkour 2001.

19 Zeder 1991.

20 Becker 2008.

21 Sapir‑Hen and Ben‑Yosef 2013.

22 Kolpakow 1935.

23 Tapper s.d.

24 Potts 2004.

25 See litterature in Çakırlar and Berthon 2014.

26 Mashkour 1997; Uerpmann 1999.

27 Ruiz et al2015.

28 Galik et al. 2015.


MNHN, Unité Archéozoologie, archéobotanique: sociétés, pratiques et environnements

Institute of Population Genetics, Department of Biomedical Sciences, University of Veterinary Medicine, Vienna, Austria

Groningen Institute of Archaeology, Netherlands