The early hunting dog from Dmanisi with comments on the social behaviour in Canidae and hominins

Abstract

The renowned site of Dmanisi in Georgia, southern Caucasus (ca. 1.8 Ma) yielded the earliest direct evidence of hominin presence out of Africa. In this paper, we report on the first record of a large-sized canid from this site, namely dentognathic remains, referable to a young adult individual that displays hypercarnivorous features (e.g., the reduction of the m1 metaconid and entoconid) that allow us to include these specimens in the hypodigm of the late Early Pleistocene species Canis (Xenocyon) lycaonoides. Much fossil evidence suggests that this species was a cooperative pack-hunter that, unlike other large-sized canids, was capable of social care toward kin and non-kin members of its group. This rather derived hypercarnivorous canid, which has an East Asian origin, shows one of its earliest records at Dmanisi in the Caucasus, at the gates of Europe. Interestingly, its dispersal from Asia to Europe and Africa followed a parallel route to that of hominins, but in the opposite direction. Hominins and hunting dogs, both recorded in Dmanisi at the beginning of their dispersal across the Old World, are the only two Early Pleistocene mammal species with proved altruistic behaviour towards their group members, an issue discussed over more than one century in evolutionary biology.

Full text



Scientic Reports | (2021) 11:13501 | 
www.nature.com/scientificreports
The early hunting dog
from Dmanisi with comments
on the social behaviour in Canidae
and hominins
Saverio Bartolini‑Lucenti1,2*, Joan Madurell‑Malapeira3,4,
Bienvenido Martínez‑Navarro5,6,7*, Paul Palmqvist8, David Lordkipanidze9,10 &
Lorenzo Rook1
The renowned site of Dmanisi in Georgia, southern Caucasus (ca. 1.8 Ma) yielded the earliest direct
evidence of hominin presence out of Africa. In this paper, we report on the rst record of a large‑sized
canid from this site, namely dentognathic remains, referable to a young adult individual that displays
hypercarnivorous features (e.g., the reduction of the m1 metaconid and entoconid) that allow us
to include these specimens in the hypodigm of the late Early Pleistocene species Canis (Xenocyon)
lycaonoides. Much fossil evidence suggests that this species was a cooperative pack‑hunter that,
unlike other large‑sized canids, was capable of social care toward kin and non‑kin members of its
group. This rather derived hypercarnivorous canid, which has an East Asian origin, shows one of its
earliest records at Dmanisi in the Caucasus, at the gates of Europe. Interestingly, its dispersal from
Asia to Europe and Africa followed a parallel route to that of hominins, but in the opposite direction.
Hominins and hunting dogs, both recorded in Dmanisi at the beginning of their dispersal across
the Old World, are the only two Early Pleistocene mammal species with proved altruistic behaviour
towards their group members, an issue discussed over more than one century in evolutionary biology.
Wild dogs are medium- to large-sized canids that possess several hypercarnivorous craniodental features and
complex social and predatory behaviours (i.e., social hierarchic groups and pack-hunting of large vertebrate prey
typically as large as or larger than themselves). Two extant species of wild dogs survive in the Old World, the
Indian dhole, Cuon alpinus (Pallas, 1811), and the African hunting dog, Lycaon pictus (Temminck, 1820). Both
are nowadays endangered or critically endangered according to the IUCN red list of threatened species1,2. e
African hunting dog, known also as painted dog, and the dhole are among the top predators in their respective
habitats3,4 thanks to the combination of several dental hypercarnivorous traits, skeletal adaptations to cursorial
pack hunting and their highly developed social behaviour.
e evolution of these hypercarnivorous canids is still unknown and open to debate5,6.
Furthermore, there is also a great deal of confusion in the taxonomy of the extinct large-sized and hypercar-
nivorous canids, which use to be referred to dierent systematic denominations (see Supplementary Informa-
tion). Such names oen hint implied or proposed anities to extant taxa, yet seldomly based on phylogenetic
analyses. Considering the results of molecular phylogenies7,8, from which it is evident that Lycaon and Cuon
are sister taxa of the crown group of Canis, and that the large-sized members of the genus Xenocyon might be
OPEN
             
Italy.              Institut Català
           
Department of Geology,
              
del Vallès, Barcelona, Spain. IPHES, Institut Català de Paleoecologia Humana i Evolució Social,
           Area de Prehistòria,
            
            

 *
icrea.cat
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2021) 11:13501 | 
www.nature.com/scientificreports/
related to both Lycaon and Cuon, here we prefer to avoid names suggestive of a closer relationship to any of both
genera, privileging the more parsimonious denomination Canis (Xenocyon) (for an in-depth discussion of the
taxonomical issues, see the Supplementary Information).
e earliest record of a species of this group of hypercarnivorous canids corresponds to Canis (Xenocyon)
cf. dubius (Teilhard de Chardin, 1940), which is represented by a single hemimandible6 from the Zanda Basin
(3.81–3.42Ma; Fig.1). e species C. (Xenocyon) dubius is generally related to the lineage of Cuon6,9. A younger
but more complete specimen from Fan Tsun (Taigu10) was ascribed to Canis (Xenocyon) antonii (ca. 2.5Ma)11.
e latter canid is large-sized and displays evident dental features hinting to an incipient adaptation to a hyper-
carnivorous diet. Other records of large-sized canids with hypercarnivorous features are rather scanty across
Eurasia and are of dicult attribution, considering the presence of hypercarnivorous Canis s.s. in Asia during
the Early Pleistocene, e.g., Canis chihliensis Zdansky, 1924; Canis teilhardi Qiu etal., 2004; or Canis yuanmoensis
You & Qi, 1973.
Around 2.0–1.8Ma, dierent forms appeared in several parts of the Old World. ese forms showed distinc-
tive dental features (i.e., broad and stoutly-built carnassials with enlarged buccal cuspids), coupled with cranio-
mandibular ones (robust mandibles and developed frontal sinuses). eir large size combined to these dental
adaptations could have determined an advantage over the contemporaneous, medium-sized mesocarnivorous
canids, as testied by the westward dispersion and radiation of Canis (Xenocyon) falconeri (Forsyth Major, 1877)
in Western Europe and of Canis (Xenocyon) africanus (Pohle, 1928) from Olduvai Bed I (Tanzania) or Ain Hanec
(Algeria) in Africa. A record of a primitive wild dog attributed to C. (Xenocyon) cf. falconeri was also reported
from deposits of Tamagawa (near Tokyo13), correlated to 2.1–1.6 Ma14. A close relationship between both taxa
was suggested by5,11 who regarded them as the ancestor of modern L. pictus. However, such interpretation has not
been shared by other researchers15. Recently, a new large-sized taxon was described as Lycaon sekowei Hartstone-
Rose etal., 2010, based on fragmented cranial material from Cooper’s Cave in South Africa (ca. 1.9Ma) and
an almost complete skeleton from Gladysvale (ca. 1.0Ma)16. Some of the morphologies of the holotype from
Cooper’s Cave (i.e., the high-crowned upper premolars, their mesial occlusal morphology, the lingual projec-
tion of P4 protocone, and the relative buccolingual length of the M1) cast doubts on its taxonomical attribution
and its actual relation with Canis (Xenocyon)’s group. Moreover, the upper teeth resemble those of the Asian C.
chihliensis, a large-sized canid possibly belonging to a hypercarnivorous lineage of Canis10.
During the late Early Pleistocene (i.e., Calabrian stage: 1.8–0.8Ma), while other more primitive species
remained in Africa [e.g., Canis (Xenocyon) atrox Broom in Broom & Schepers, 1946 from Kromdraai A; Fig.1,
although possibly synonym of C. (Xenocyon) africanus11] a more derived form of Canis (Xenocyon) appeared
and became widespread across the whole Old World (Fig.1). Canis (Xenocyon) lycaonoides (Kretzoi, 1938) was
a large-sized canid that resembled C. (Xenocyon) gr. falconeri but with more derived craniodental features (e.g.,
the P4 protocone tends to attach to the tooth; the M1 metaconule is crest-like; the M1 talon is reduced; the m1
hypoconid is enlarged and tends to be centred in the talonid, which functionally represents a lengthening of the
trenchant condition of the trigonid; the entoconid is reduced, being represented by a small crest-like cuspulid;
and the m3 is single cusped). Its earliest record appears to be that of Venta Micena (Spain5, Fig.1). In spite of its
uncertain chronology, this early occurrence suggests an eastern Asian origin for this hypercarnivorous species.
Subsequently, during the late Early Pleistocene and the base of the Middle Pleistocene, from ca 1.6 to 0.7Ma,
C. (Xenocyon) lycaonoides became one of the most common and important members of the carnivoran palae-
oguild of Eurasia (Fig.1). Moreover, C. (Xenocyon) lycaonoides dispersed also in Africa, where it is documented
in the northern and eastern part of the continent (e.g., Olduvai Bed II; Fig.1). Considering the overall cranial
morphology and its dental features, which conrm the original interpretation by Kretzoi17, Martínez-Navarro
& Rook5 deemed C. (Xenocyon) lycaonoides as strictly related to extant L. pictus. Although some scholars do not
favour this interpretation16,18, similar conclusions were shared by several other scholars10,19–21, who supported
also a Eurasian origin for the living African hunting dog.
Among extant Carnivora, Lycaon pictus has one of the most complex, structured and unique social
behaviours3,22. As one of the closest relatives to L. pictus, C. (Xenocyon) lycaonoides, the Eurasian hunting dog,
might have had comparable complex sociality. Carbone and co-authors23 showed that the metabolic energy
requirements for large-sized species (> 21.5kg) force them to predate on prey larger than themselves and thus, in
hypercarnivorous Canidae, to hunt cooperatively. As such, this element allows us to gure the social behaviour
of extinct hypercarnivorous canids, even with limited direct evidence. Nevertheless, apart from indirect and
inferred evidence, direct proof of social behaviour in the Eurasian hunting dog have been reported24,25.
Here we report the rst occurrence of wild dogs from the Georgian site of Dmanisi (Fig.1; 1.77–1.76 Ma26; see
Supplementary Information). is locality preserves an outstanding fossil record, both in terms of abundance,
completeness of skeletal remains and preservational status, as testied by the recently described molecular
phylogeny based on a fossil rhino tooth27. In this paper, we describe the newly discovered remains, identifying
them taxonomically and interpreting in the frame of Early Pleistocene diversity of Canis (Xenocyon). Moreover,
the site of Dmanisi has yielded the earliest direct evidence of hominin presence out of Africa in their dispersal
throughout Eurasia28,29 with also indication of complex sociality among individuals of this population30,31. e
co-occurrence of two highly social species in the same locality around 1.8Ma, a time of extreme diversica-
tion and expansion of the two clades from their centres of origin5,6, raises interest in the role played by social
behaviour and by mutually-benecial cooperation and reciprocity in the geographic expansion of these species.
Questions to be explored in this paper.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2021) 11:13501 | 
www.nature.com/scientificreports/
Results
Implications for fossil hunting dogs diversity. e nding of a large-sized canid in the Georgian site of
Dmanisi represents an important discovery, which adds valuable information to the current knowledge of canid
radiation during the second half of the Early Pleistocene (early Calabrian). Despite the fragmented nature of
Figure1. Map and chronology of Canis (Xenocyon) occurrences. (a) Resuming chronological scheme of the
known occurrences of fossil wild dogs in the Old World. Abbreviations: AHan, Ain Hanec (Algeria); APL1,
Apollonia-1(Greece); CVict, Cueva Victoria (Spain); EVT, Vallparadís Estació (Spain); FTs, Fan Tsun (China);
KromdA, Kromdraai A (South Africa); OH1-GR1, Oulad Hamida1-Grotte des Rhinoceros (Morocco);
Olduvai I, Olduvai Bed I (Tanzania); Olduvai II, Olduvai Bed II (Tanzania); PN, Pirro Nord (Italy); SSMZ,
Shanshenmiaozui (China); 1-GH, omas 1 Quarry-Grotte des Hominides (Morocco); VM, Venta Micena
(Spain); Westbury sM, Westbury-sub-Mendip (Great Britain). (b,c) Maps showing the Old-World occurrences
of fossil wild dogs described in the text. (c–e detailed view of respectively Europe and Circum-Mediterranean
area, eastern Asia and southern Africa). Localities: 1, Fonelas-P1 (Spain); 2, Venta Micena (Spain); 3, Cueva
Victoria (Spain); 4, Vallparadís Estació (Spain); 5, Ceyssaguet (France); 6, Vallonnet (France); 7, Westbury-sub-
Mendip (Great Britain); 8, Upper Valdarno (Italy); 9, Collecurti (Italy); 10, Pirro Nord (Italy); 11, Mosbach II
(Germany); 12, Würzburg-Schalksberg (Germany); 13, Untermassfeld (Germany); 14, Koněprusy C178 (Czech
Republic); 15, Stránská Skála (Czech Republic); 16, Gombasek (Slovakia); 17, Beta (Romania); 18, Trlica
(Montenegro); 19, Apollonia-1 (Greece); 20, Margaritovo (Russia); 21, Akhalkalaki (Georgia); 22, Dmanisi
(Georgia); 23, Tighennif/Terne (Algeria); 24, Ain Hanec (Algeria); 25, ‘Ubeidiya (Israel); 26, Lakhuti-2
(Tajikistan); 27, Campbellpore (Pakistan); 28, Zanda Basin (China); 29, Longdan (China); 30, Yunxian (China);
31, Loc. 33 in Zdansky (1924) (China); 32, Fan Tsun/Taigu (China); 33, Ma Fang (China); 34, Zhoukoudian 18
(China); 35, Zasukino (Russia); 36, Nalaikha (Mongolia); 37, Tamagawa (Japan); 38, Olyorian fauna (Russia);
39, Olduvai Bed I (Tanzania); 40, Olduvai Bed II (Tanzania); 41, Cooper’s Cave (South Africa); 42, Kromdraai
A (South Africa); 43, Gladysvale (South Africa); 44, Hopeeld (South Africa). Symbol and colors code (see
also graphic legend): red star, Dmanisi site; dark red circles, C. (Xenocyon) ex gr. falconeri; blue triangles, C.
(Xenocyon) lycaonoides; yellow squares, C. (Xenocyon) dubius. Chronological scale edited by S. Bartolini-Lucenti
in Inkscape ver. 0.92 (https:// inksc ape. org/) from Bartolini-Lucenti & Madurell-Malapeira12. Georeferenced
maps (points and background) made in Simplemappr (https:// www. simpl emappr. net/) and modied in
Inkscape ver. 0.92.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2021) 11:13501 | 
www.nature.com/scientificreports/
the specimens, the set of features possessed by D6327 (Fig.2a–f and Augmented Reality content) allow a con-
dent attribution to Canis (Xenocyon) lycaonoides (see Supplementary Information), the plausible ancestor of the
extant African hunting dog5,19. As such, this record is the oldest occurrence of Eurasian hunting dogs and pre-
cedes the burst of dispersal that the species experienced across the entire Old World during the Calabrian5,10,19.
Dietary preferences of the Dmanisi hunting dog. In order to test the dietary adaptations of the
Dmanisi hunting dog and other Early Pleistocene forms, a linear discriminant analysis was performed over
the extant canids (32 species, 247 specimens; craniodental measurements kindly provided by B. Van Valken-
burgh), which were grouped in two feeding groups: (i) omnivores (i.e., meso- and hypocarnivores; 27 extant
species, 210 specimens), in which vertebrate esh represents less than 70% of their dietary requirements; and (ii)
hypercarnivores (four extant species, 34 specimens), which diet consists almost entirely of vertebrate esh and
are pack-hunters of prey as large as or larger than themselves. Seven metric variables of this dataset for which
the measurements were available in the Dmanisi specimens were used in the analysis: length and breadth of the
third lower premolar (p3L and p3B, respectively), length and breadth of the trigonid basin of the lower carnassial
(m1trigL and m1trigB, respectively), length and breadth of the talonid basin of the lower carnassial (m1talL and
m1talB, respectively), and jaw depth measured at the limit between p3 and p4 (JDp4). e linear discriminant
function was obtained with the direct method for inclusion of all variables. Reclassication of specimens to each
dietary group were derived by cross validations using the leave one out method. Aer cross-validation, the dis-
criminant function (Fig.3) correctly allocated 98.8% of the specimens to their feeding group.
Indeed, all omnivores and all hypercarnivores, apart from the four specimens of the small-sized S. venati-
cus, were correctly classied in their feeding groups (Fig.3). According to the loadings of the variables in the
discriminant function, the hypercarnivores show third premolars that are relatively mesiodistally shorter and
buccolingually narrower compared to those of omnivorous species, as well as a carnassial with an enlarged
trigonid blade and a reduced talonid basin, and a deeper, more stoutly-built mandibular corpus, which is in
agreement with previous analyses of adaptations in canids towards hypercarnivory33,34. is function reclassied
Figure2. Canis (Xenocyon) lycaonoides from Dmanisi. (a–c) D6327a, le corpus with p1-p3 in buccal (a),
lingual (b) and occlusal (c) views. (d)–(f), D6327b, le lower m1 in buccal (d), lingual (e) and occlusal (f)
views. QR code and Augmented Reality (AR) marker showing 3D comparison between the lower rst molar
morphologies of Canis (Xenocyon) from Dmanisi (red), Canis (Xenocyon) lycaonoides from Venta Micena
(green) and Canis (Xenocyon) falconeri from Upper Valdarno (gray). Instructions: Scan the QR code on the
le; open the link; allow the browser to access the camera of your device; point the camera toward the marker
(on the right); and wait for the model to load (up to 10s). It is possible to turn the device around the marker
(or to move the marker) to see dierent parts of the model. Best visualization performances can be achieved
by printing the markers, rather than pointing at them on screens. For common issues refer to Supplementary
Information and Bartolini-Lucenti etal.32. Photos of the fossil specimens elaborated in Photoshop CC2019
(https:// www. adobe. com/). Line drawing of C. (Xenocyon) and gure composition made by S. Bartolini-Lucenti
in Inkscape ver. 0.92 (https:// inksc ape. org/). AR content made in Visual Studio Code ver. 1.50.0 (https:// code.
visua lstud io. com/) and GitHub Desktop ver. 2.6.6 (https:// deskt op. github. com/).
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2021) 11:13501 | 
www.nature.com/scientificreports/
unequivocally the individual from Dmanisi (values of the variables obtained from D6327) in the group of hyper-
carnivores (Fig.3), with a probability of pertinence of 0.97. e two specimens of C. (Xenocyon) lycaonoides
from Venta Micena (a site that is slightly younger in age than Dmanisi, ca 1.6Ma) for which these measurements
were available were also classied as hypercarnivores. However, they show higher scores in the discriminant
function, close to the group centroid of hypercarnivores. Similarly, the single specimen from Untermassfeld, a
site of Jaramillo age (ca. 1.0Ma), shows the highest score among the fossil hunting dogs, which reects its more
advanced adaptations towards hypercarnivory, like those of extant African hunting dogs. ese results conrm
that the craniodental morphological features of the Eurasian hunting dog from Dmanisi (Fig.2) were well suited
for a diet consisting exclusively of vertebrate esh. Moreover, they show that there was a gradual evolution of
these craniodental adaptations in C. (Xenocyon) lycaonoides from the oldest members analyzed of the lineage
(Dmanisi) to the most derived ones (Untermassfeld), conrming the morphological evidence pointed out by
other scholars5,10,19,35.
Discussion
Dmanisi, located in the Caucasus at the gates of Europe and near the crossway between Africa and Eurasia, is
a key site to explain the dispersal of large mammal species, in a time of great faunal turnovers in the whole Old
Worl d36,37. is Georgian site also records the earliest direct evidence of hominins presence out of Africa and
their dispersal into Eurasia, at ca. 1.8Ma. Here, we report the record of the Eurasian hunting dog, C. (Xenocyon)
lycaonoides, which testies to the beginning of the dispersal of this more derived, frankly hypercarnivorous canids
from its eastern Asia region of origin, similarly to Canis borjgali Bartolini-Lucenti etal., 2020 (the mesocarnivo-
rous, wolf-like species also recorded in Dmanisi32). During the Calabrian, C. (Xenocyon) lycaonoides became
a common element of the entire Old-World faunas in the late Early-early Middle Pleistocene19, when it even
reached North America10. In this dispersal, the Eurasian hunting dog followed at the same time the same disper-
sal pattern of hominins, just in the opposite direction. e co-occurrences of both species along their dispersal
routes together with some other large-sized carnivore taxa, for instance the dirk-toothed cat of African origin
Megantereon whitei (Broom, 1937)38,39, suggest that the ecological conditions favoured the dispersal of these spe-
cies at that time. Large-sized carnivorans like this felid has been recognized as important supplier of scavengeable
resources for the hominins in direct competition with the large-sized scavenger Pachycrocuta brevirostris40,41.
Social behaviour of Canis (Xenocyon) and Homo in the late Early Pleistocene. “ere is, at the
same time, as much, or perhaps even more, of mutual support, mutual aid, and mutual defense: Sociability is
as much a law of nature as mutual struggle”42. Probably, the most relevant common feature between the extinct
hominins and the fossil hunting dogs is the fossil evidence on the mutually-benecial cooperation, reciprocity
and social behaviour43 of both species. is is well documented in Dmanisi by the nding at this site of an eden-
tulous individual of Homo erectus (composite skull D 3444/D 3900) who lost all but one of its teeth several years
before the time of its death, as evidenced by extensive bone loss in the maxilla and mandible due to resorption
of the tooth alveoli. is old individual, probably a female given the relative gracile condition of the skull, could
not chew hard or coriaceous food by itself, which means that its survival aer the loss of the majority of its teeth
probably relied on the assistance from other members of the family group30 (Fig.4a). As it has been noted30,31,
this kind of altruistic behaviour is beyond forms of biological altruism, proper of non-primate mammals or even
“non-human primates”31. is suggests that altruistic behaviour and care of the elderly might have developed
very early in hominins, at least two million years ago30,31. Among Carnivora, social behaviour is frequent, con-
sidering the numerous benets that cooperation oers to carnivorans (increased breeding success and individual
survival; enhanced hunting success; ability to kill larger prey; deterrent and strength against kleptoparasites; help
for the rearing of pups44,45).
Canidae have some of the best-known examples of social organization of all mammals (e.g., the grey wolf,
C. lupus46). Probably less known, yet interesting is the case of the African hunting dogs. is hypercarnivore
species display a more complex and peculiar set of behaviour, unique among Canidae, if not carnivorans. is
includes exclusive cooperative hunt, obligate cooperative breeding47, prioritized access of the pups to the kills3,
widest variety of vocal repertoire in canids48 and consensus decision making via “sneezing”49. Many authors50
Figure3. Discriminant analysis using metric measurements of lower teeth (p3 and m1) and jaw between the
living omnivorous (i.e., meso- and hypocarnivorous) and hypercarnivorous canids (see metric data available
on the online repository at the following link: https:// dx. doi. org/ dx. 10. 5281/ zenodo. 47043 27). e scores of the
fossil specimens, including Dmanisi, Venta Micena and Untermassfeld, are shown. Graph made in Photoshop
CC2019 (https:// www. adobe. com/).
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2021) 11:13501 | 
www.nature.com/scientificreports/
Figure4. Two social species at Dmanisi. (a) altruistic behaviour of a group of Homo erectus sharing food with
an individual who lived several years without teeth (as evidenced by edentulous skull D3444 and associated
mandible D3900). is severe masticatory impairment would limit the diet of the individual to foodstus that
did not require heavy chewing (e.g., so plants, animal brain and marrow) or that were orally processed before
by others. (b) a pack of hunting dogs chasing a prey (goat Hemitragus albus) by at Venta Micena, a site where
a pathological skull (cranium and associated mandible VM-7000) of Canis (Xenocyon) lycaonoides showing
marked bilateral asymmetry and agenesia of several teeth was unearthed. e disabled dog, whose absence
of an upper canine probably made it useless for hunting, is drawn running far behind the pack. Given that
the individual managed to survive until a relatively advanced age, as indicated by tooth wearing, this suggests
that the other members of its family group would have allowed it to feed on the prey captured by the hunting
pack. Remains of this hypercarnivorous canid species are also preserved in the assemblage of large mammals
from Dmanisi, as shown in this paper. Artwork made by Mauricio Antón with the scientic supervision by the
authors of the manuscript.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2021) 11:13501 | 
www.nature.com/scientificreports/
noted a reduced degree of aggressivity between pack members in comparison to other social Canidae (C. lupus
and C. alpinus46), even during the consumption of the kill3.
Sociality in fossil canids had been investigated by numerous authors51 and Carbone and coauthors23 proved
the necessity for large canids, weighing more than 21.5kg, to hunt cooperatively to kill on prey larger than them-
selves. e Eurasian hunting dog C. (Xenocyon) lycaonoides was indeed a large-sized hypercarnivorous species.
Body-size estimates suggests that this canid was similar to L. pictus (whose average weight is 20–25 kg52) if not
larger (estimated weight of C. (Xenocyon) lycaonoides = 28 kg24). e individual from Dmanisi, despite being a
young adult, would have been rather robust (around 30kg, applying the regression equation for body mass on
lower carnassial length53). Such a body mass, coupled with its marked hypercarnivorous features, support the
idea that C. (Xenocyon) lycaonoides adopted cooperative hunting strategies, similar to the extant canids C. lupus,
C. alpinus and L. pictus. Further support of a highly social group organization is provided by fossil pathological
specimens. Recently, Tong etal.25 described injuries in the sample of Shanshenmiaozui, Nihewan Basin, dated to
1.2Ma. One of the specimen records a dental infection likely inicted by processing hard food, such as bone; the
other suered a displaced fracture of its tibia and, despite such a severe injury (which would represent a death
sentence for a solitary predator) it managed to survive the trauma to heal. e long period that was presumably
required for healing the compound fracture, as well as the incapacitating nature of this trauma for a cursorial
predator during the rest of its life (as the healed tibia was considerably shortened), suggests social hunting
strategies and provision by other members of the family/pack (primarily food-sharing). Similar pathologies
have been also detected in the Late Pleistocene population of Canis dirus Leidy, 1858 (recently reassigned to
Aenocyon dirus54) from La Brea tar pits in southern California55. is is not surprising considering that packs
of extant canids temporarily support wounded or sick members of their group, as reported by many authors in
both extant C. lupus and L. pictus45, despite the cost in terms of eciency of the group56. Nevertheless, in the
case of African hunting dogs, several studies describe the tolerance by group members not only for injured, but
also for disabled or old individuals at the kills45,57. Furthermore, disabled or old African hunting dogs receive
food by fellow pack members via regurgitation45,58, a way of food-sharing that other canids reserve exclusively
to kin, very rarely non-kin, pups and to the breeding female. e fossil record yields evidence of similar behav-
iour in extinct hunting dog as well. An altruistic behaviour of food provisioning to disabled individuals was
documented in C. (Xenocyon) lycaonoides at the site of Venta Micena (Fig.4b). Here a nearly complete cranium
with a mandible preserved in anatomical connection were unearthed (skull VM-7000)24. is skull belonged to
a 7–8years-old individual (considering the moderate-heavy dental wear of its teeth). By far the most sticking
features of this specimen are the high degree of cranial uctuating asymmetry and several tooth anomalies,
including dental agenesia of the upper right canine, the P3 and m3. ese teeth were not broken or lost during
the life of the individual, as showed by CT scans and radiographs of the cranium24. e dental alveolus of the
right upper canine is completely absent, as for the other teeth24. Moreover, the right m2 is missing and its alveolus
is partially reabsorbed. e malformations of the C. (Xenocyon) lycaonoides from Venta Micena were probably
due to developmental instabilities resulting from a high level of genetic homozygosity in the relatively small
population of wild dogs that inhabited the Baza Basin during late Early Pleistocene times24: anodontia (tooth
losses) and cranial bilateral asymmetry have been both documented in extant populations of C. lupus of small
size subject to severe bottlenecks and inbreeding, for example the wolf population of the Białowieża Primeval
Forest in Poland59,60. In the case of modern L. pictus, a study of museum skulls that span a period of a hundred
years, which records the dramatic decline in the populations of the species in sub-Saharan Africa during the
last century, has shown a marked increase in uctuating asymmetry as a result of increasing levels of population
homozygosity61. is suggests that the malformations of the C. (Xenocyon) lycaonoides skull from Venta Micena
would reect developmental instabilities resulting from a high level of genetic homozygosity in the relatively
small population of hunting dogs of the Baza Basin, which was geographically (and genetically) isolated from
other populations. Moreover, the eective population size of modern painted dogs is typically reduced to 20–35%
of the censused population size by reproductive suppression of subordinates and uneven sex ratios62. In the case
of Venta Micena, this would have also promoted further inbreeding and homozygosity. However, despite the
numerous congenital disabilities, the individual VM-7000 was able to reach adulthood, which probably aected
or even precluded its ability in the pack-hunting activities (Fig.4b). is suggests that cooperative behaviour
and food provisioning from other members of the family group were the only way for this individual to survive
until this age24. Similarly to the old human from Dmanisi, who managed to reach such an old age thanks to
the altruistic help and care of other family members (Fig.4a), this hunting dog reached adulthood. is truly
altruistic behaviour probably applies also to the hunting dog population of Dmanisi, although the scarce record
of this species in the site precludes a direct inference.
erefore, these ndings seem to suggest that increased cooperation and altruistic behaviour may have been
important factors for the survival and dispersal of both humans and large social carnivorans in the open envi-
ronments of Africa, Eurasia and North America. Interestingly, hunting dogs and hominins are up to now the
only late Early Pleistocene highly-social species with proved altruistic behaviour towards other members of their
group, including food sharing to group members. As noted before, such a behaviour is specially developed in the
extant African hunting dog, where individuals with limitations resulting from genetic abnormalities, pathologies
and/or advanced age are helped and sustained by the other members of the family group45,49,50. Canis (Xenocyon)
lycaonoides showed a similar pattern of cooperative and altruistic behaviour towards pack-members24,25. e
occurrence of the Eurasian hunting dog in Dmanisi marks one the rst and better chronologically-constrained
record of this large-sized, pack-hunting canid. e success of this wide-ranging dispersion across continents5,10,
unprecedented and never reached by any other large-sized canids, might be correlated also to the advantages
of the mutually-benecial cooperation and altruistic nature of these extinct hunting dogs, as the result of an
evolutionary trend leading to co-operation among members of a species: “the best pathway to advantage for
individuals”63.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2021) 11:13501 | 
www.nature.com/scientificreports/
It would not be necessary, but we have here a new evidence of the importance of Dmanisi for that, para-
phrasing Dawkins64, Homo and highly social Canidae both are descended from highly social ancestors and their
ancestors lived in groups; this was not an option but an essential survival strategy and from this mutual aid arose.
Materials and methods
e present study is based on the comparative morphological analysis of the large-sized Canis (Xenocyon) from
Dmanisi and other Plio-Pleistocene hypercarnivorous canids of the Old World. e described fossils are housed
at the S. Janashia Museum of Georgia, Georgian National Museum (Tbilisi) (MG-GNM). As comparative fossil
material, the Villafranchian and Epivillafranchian canids from the Old World and North America housed at
theAmerican Museum of Natural History, New York (United States), Earth Science Dept. of the Aristotle Uni-
versity of essaloniki (essaloniki, Greece), Institut Català de Paleontologia Miguel Crusafont,Universitat
Autonoma de Barcelona (Barcelona, Spain), Museo di Geologia e Paleontologia, Università degli Studi di Firenze
(Italy), and Musée Nationald’Histoire Naturelle (Paris, France) were studied. is fossil comparative sample
includes specimens of Canis (Xenocyon) dubius from Zhoukoudian Loc. 1865. Canis (Xenocyon) falconeri from
Upper Valdarno Basin and Tamagawa15. Canis (Xenocyon) lycaonoides from Apollonia-166; Campbellpore11;
Chukochya35, Zanushino35; Cripple Creek Sump10; Cueva Victoria, Vallparadís Estació19; Lakhuti-211,35; Ma
Fang11; Nalaikha35; Olduvai Bed II5; Pirro Nord67; Shanshenmiaozui6,25, Tighennif68; Trlica69; Untermassfeld35;
Venta Micena24; Westbury-sub-Mendip70. Canis chihliensis from Yushe Basin11. e relevant literature on these
canids was reviewed6,10,13,14,35,57,58,65,72,73.
Extant specimens housed at theAmerican Museum of Natural History (New York, United States), Museo di
Zoologia "La Specola", Università degli Studi di Firenze (Italy), Institut Català de PaleontologiaMiguel Crusa-
font(Barcelona, Spain), Royal Museum for Central Africa (Tervuren, Belgium) and MG-GNMwere also used for
morphological and metrical comparisons. We examined specimens of Canis lupus Linnaeus, 1758, and Lycaon
pictus (Temminck, 1820). Moreover, a wide data set of craniodental measurements taken in modern canids (247
specimens from 32 species) by Prof. Blaire Van Valkenburgh was used also in some statistical comparisons,
including a discriminant analysis between omnivorous (i.e., meso- and hypocarnivorous) and hypercarnivorous
canids, in order to deliver palaeoecological inferences for the Dmanisi wild dogs and also for others from dif-
ferent (and younger) sites, like Venta Micena in Spain and Untermassfeld in Germany. Analyses and graphs on
dental values present in the supplementary were made in R ver. 3.6.1. (https:// cran.r- proje ct. org/) using package
ggplot2 ver. 3.2.1 (http:// ggplo t2. tidyv erse. org)73.
Cranial and dental measurements were taken with a digital calliper to the nearest 0.1mm following von den
Driesch74 with minor modications.
Data availability
All data generated or analyzed during the study are included in this published article, in its Supplementary Infor-
mation Files and on the online repository Zenodo at the following link https:// doi. org/ 10. 5281/ zenodo. 47043 27.
Received: 3 February 2021; Accepted: 16 June 2021
References
1. Jdeidi, T., Masseti, M., Nader, I., de Smet, K., & Cuzin, F. Lycaon pictus. e IUCN Red List of reatened Species 2010. doi:https://
doi. org/ 10. 1371/ journ al. pone. 00738 56 (2010).
2. Kamler, J. F. et al. Cuon alpinus. e IUCN Red List of reatened Species 2015. e.T5953A72477893 (2015.).
3. McNutt, J. W. & Woodroe R. Lycaon pictus African Wild Dog (Painted Dog, Hunting Dog) In Mammals of Africa: Volume V:
Carnivores, Pangolins, Equids and Rhinoceroses, (Ed. Kingdon, J.) 51–59 (A&C Black, 2014).
4. Nurvianto, S., Eprilurahman, R., Imron, M. A. & Herzog, S. Feeding habits of pack living dhole (Cuon alpinus) in a dry deciduous
forest of East Java, Indonesia. Taprobanica 8, 10–20 (2016).
5. Martínez-Navarro, B. & Rook, L. Gradual evolution in the African hunting dog lineage systematic implications. C. R. Palevol 2,
695–702 (2003).
6. Wang, X., Li, Q. & Xie, G. Earliest record of Sinicuon in Zanda Basin, southern Tibet and implications for hypercarnivores in cold
environments. Quat. Int. 355, 3–10 (2014).
7. Koepi, K. P. et al. Genome-wide evidence reveals that African and Eurasian golden jackals are distinct species. Curr. Biol. 25,
2158–2165 (2015).
8. Gopalakrishnan, S. et al. Interspecic gene ow shaped the evolution of the genus Canis. Curr. Biol. 28, 3441-3449.e5 (2018).
9. Qiu, Z., Deng, T. & Wang, B. Early Pleistocene mammalian fauna from Longdan, Dongxiang, Gansu, China. Paleontol. Sin. 27,
1–252 (2004).
10. Tedford, R. H., Wang, X. & Taylor, B. E. Phylogenetic systematics of the North American fossil Caninae (Carnivora: Canidae).
Bull. Am. Museum Nat. Hist. 325, 1–218 (2009).
11. Rook, L. e Plio-Pleistocene Old World Canis (Xenocyon) ex gr. falconeri. Boll. Soc. Paleontol. Ital. 33, 71–82 (1994).
12. Bartolini-Lucenti, S. & Madurell-Malapeira, J. Unraveling the fossil record of foxes: An updated review on the Plio-Pleistocene
Vulpes spp. from Europe. Quat. Sci. Rev. 236, 106296 (2020).
13. Koizumi, A. First record of the Plio-Pleistocene Hypercarnivorous canid, Canis (Xenocyon) falconeri from Tama River. Quat. Res.
42, 105–111 (2003).
14. Matsukawa, M., Kakinuma, H., Baba, K. & Ohira, H. Stratigraphy and correlation of the Plio-Pleistocene strata along the western
ank of the Kwanto Plain, Japan. Bull. Tokyo Gakugei Univ. 58, 173–202 (2006).
15. Lewis, M. E. & Werdelin, L. Patterns of change in the Plio-Pleistocene carnivorans of eastern Africa. In Hominin Environments in
the East African Pliocene: An Assessment of the Faunal Evidence, (Eds. R. Bobe et al.) 77–105 (2007).
16. Hartstone-Rose, A., Werdelin, L., De Ruiter, D. J., Berger, L. & Churchill, S. E. e Plio-Pleistocene ancestor of Wild Dogs, Lycaon
sekowei n. sp. J. Paleontol. 84, 299–308 (2010).
17. Kretzoi, M. Die Raubtiere von Gombaszög nebst einer übersicht der Gesamtfauna (Ein beitrag zur stratigraphie des Altquartaers).
Ann. Mus. Natl. Hung. 31, 88–157 (1938).
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2021) 11:13501 | 
www.nature.com/scientificreports/
18. Zrzavý, J. et al. Phylogeny of the Caninae (Carnivora): Combining morphology, behaviour, genes and fossils. Zool. Scr. 47, 373–389
(2018).
19. Madurell-Malapeira, J. et al. e latest European painted dog. J. Vertebr. Paleontol. 33, 1–6 (2013).
20. Stiner, M. C. et al. Outside Africa: Middle Pleistocene Lycaon from Hayonim Cave, Israel. Boll. Soc. Paleontol. Ital. 40, 293–302
(2001).
21. Chavez, D. E. et al. Comparative genomics provides new insights into the remarkable adaptations of the African wild dog (Lycaon
pictus). Sci. Rep. 9, 1–14 (2019).
22. Chengetanai, S. et al. Brain of the African wild dog. I. Anatomy, architecture, and volumetrics. J. Comp. Neurol. 528, 3245–3261
(2020).
23. Carbone, C., Teacher, A. & Rowclie, J. M. e costs of carnivory. PLoS Biol. 5, e22 (2007).
24. Palmqvist, P., Arribas, A. & Martínez-Navarro, B. Ecomorphological study of large canids from the lower Pleistocene of south-
eastern Spain. Lethaia 32, 75–88 (1999).
25. Tong, H. et al. Hypercarnivorous teeth and healed injuries to Canis chihliensis from Early Pleistocene Nihewan beds, China, sup-
port social hunting for ancestral wolves. PeerJ https:// doi. org/ 10. 7717/ peerj. 9858 (2020).
26. Ferring, R. et al. Earliest human occupations at Dmanisi (Georgian Caucasus) dated to 1.85–1.78 Ma. PNAS 108, 10432–10436
(2011).
27. Cappellini, E. et al. Early Pleistocene enamel proteome from Dmanisi resolves Stephanorhinus phylogeny. Nature 574, 103–107
(2019).
28. Lordkipanidze, D. et al. Postcranial evidence from early Homo from Dmanisi, Georgia. Nature 449, 305–310 (2007).
29. Lordkipanidze, D. et al. A complete skull from Dmanisi, Georgia, and the evolutionary biology of early Homo. Science 342, 326–331
(2013).
30. Lordkipanidze, D. et al. e earliest toothless hominin skull. Nature 434, 717–718 (2005).
31. Lordkipanidze, D. et al. A fourth hominin skull from Dmanisi, Georgia. Anat. Rec. 288, 1146–1157 (2006).
32. B artolini-Lucenti, S., Bukhsianidze, M., Martínez-Navarro, B. & Lordkipanidze, D. e Wolf from Dmanisi and Augmented Real-
ity: Review, Implications, and Opportunities. Front. Earth Sci. 8, 1–13 (2020).
33. Van Valkenburgh, B. Iterative evolution of hypercarnivory in canids (Mammalia: Carnivora): Evolutionary interactions among
sympatric predators. Paleobiology 17, 340–362 (1991).
34. Balisi, M. A. & Van Valkenburgh, B. Iterative evolution of large-bodied hypercarnivory in canids benets species but not clades.
Comm. Biol. 3, e461 (2020).
35. Sotnikova, M. V. Remains of Canidae from the lower Pleistocene site of Untermassfeld In Das Pleistozän von Untermassfeld bei
Meiningen (üringen) (ed. R. D. Kahlke) 607–632 (Römisch-Germanisches Zentralmuseum, 2001).
36. Rook, L. & Martínez-Navarro, B. Villafranchian: the long story of a Plio-Pleistocene European large mammal biochronologic unit.
Quat. Int. 219, 134–144 (2010).
37. Azzaroli, A., De Giuli, C., Ficcarelli, G. & Torre, D. Late Pliocene to early Mid-Pleistocene mammals in Eurasia: Faunal succession
and dispersal events. Palaeogeogr. Palaeoclimat. Palaeoecol. 66, 77–100 (1988).
38. Martínez-Navarro, B. & Palmqvist, P. Presence of the African machairodont Megantereon whitei (BROOM, 1937) (Felidae, Car-
nivora, Mammalia) in the lower Pleistocene site of Venta Micena (Orce, Granada, Spain), with some considerations on the origin,
evolution and dispersal of the genus. J. Archaeol. Sci. 22, 569–582 (1995).
39. Palmqvist, P. et al. A re-evaluation of the diversity of Megantereon (Mammalia, Carnivora, Machairodontinae) and the problem
of species identication in extinct carnivores. J. Vert. Paleontol. 27, 160–175 (2007).
40. Espigares, M. P. et al. Homo vs. Pachycrocuta: Earliest evidence of competition for an elephant carcass between scavengers at Fuente
Nueva-3 (Orce, Spain). Quat. Int. 295, 113–125 (2013).
41. Madurell-Malapeira, J. et al. Were large carnivorans and great climatic shis limiting factors for hominin dispersals? Evidence of
the activity of Pachycrocuta brevirostris during the Mid-Pleistocene Revolution in the Vallparadís Section (Vallès-Penedès Basin,
Iberian Peninsula). Quat. Int. 431, 42–52 (2017).
42. Kropotkin, P. Mutual Aid, A Factor of Evolution 348 (McClure P. & Co., 1902).
43. Boucher, D. H. e Biology of Mutualism: Ecology and Evolution (Oxford University Press, 1985).
44. Almberg, E. S. Social living mitigates the costs of a chronic illness in a cooperative carnivore. Ecol. Lett. 18, 660–667 (2015).
45. Courchamp, F. & Macdonald, D. W. Crucial importance of pack size in the African wild dog Lycaon pictus. Anim. Conserv. 4,
169–174 (2001).
46. Mech, L. D. Canis lupus. Mamm. Species 37, 1–6 (1974).
47. Woodroe, R., Lindsey, P. A., Romañach, S. S. & Ranah, S. M. O. African wild dogs (Lycaon pictus) can subsist on small prey:
implications for conservation. J. Mammal. 88, 181–193 (2007).
48. Robbins, R. L. Vocal communication in free-ranging African wild dogs (Lycaon pictus). Behaviour 137(10), 1271–1298. https://
doi. org/ 10. 1163/ 15685 39005 019261 (2000).
49. Walker, R. H., King, A. J., McNutt, J. W. & Jordan, N. R. Sneeze to leave: African wild dogs (Lycaon pictus) use variable quorum
thresholds facilitated by sneezes in collective decisions. Proc. R. Soc. B Biol. Sci. 284, (2017)
50. Schaller, G. B. e Serengeti Lion: A Study of Predator-Prey Relations (University of Chicago Press, 1970).
51. Van Valkenburgh, B., Sacco, T. & Wang, X. Chapter7: Pack hunting in Miocene borophagine dogs: evidence from craniodental
morphology and body size. Bull. Am. Mus. Nat. Hist. 279, 147–162 (2003).
52. Castelló, J. R. Canids of the World: Wolves, Wild Dogs, Foxes, Jackals, Coyotes, and eir Relatives (Princeton University Press, 2018).
53. Van Valkenburgh, B. in Body size in mammalian paleobiology: estimation and biological implications (eds. Damuth, J. D. &
MacFadden, B. J.) 181–206 (Cambridge University Press, 1990).
54. Perri, A. R. et al. Dire wolves were the last of an ancient New World canid lineage. Nature 591(7848), 87–91 (2021).
55. Lawler, D. F., Widga, C. & Smith, G. K. Observations of the acetabulum and proximal femur of the dire wolf (Canis dirus, Leidy
1854). J. Vet. Anat. 10, 73–83 (2017).
56. Courchamp, F., Rasmussen, G. S. & Macdonald, D. W. Small pack size imposes a trade-o between hunting and pup-guarding in
the painted hunting dog Lycaon pictus. Behav. Ecol. 13, 20–27 (2002).
57. Schneeberger, K. Food sharing and nonhuman reciprocal altruism. In Encyclopedia of Evolutionary Psychological Science Publisher
(Eds. Shackelford, T. K. & Weekes-Shackelfork, V. A.) (Springer International Publishing Switzerland Editors, 2016).
58. Estes, R. D. & Goddard, J. Prey selection and hunting behaviour of the African wild dog. J. Wildl. Mgmt. 31, 52–70 (1967).
59. Buchalczyk, T., Dynowski, J. & Szteyn, S. Variations in number of teeth and asymmetry of the skull in the wolf. Acta eriolog. 26,
23–30 (1981).
60. Vilà, C., Urios, V. & Castroviejo, J. Tooth losses and anomalies in the wolf (Canis lupus). Can. J. Zool. 71, 968–971 (1993).
61. Edwards, C. T. T., Rasmussen, G. S. A., Riordan, P., Courchamp, F. & Macdonald, D. W. Non-adaptive phenotypic evolution of the
endangered carnivore Lycaon pictus. PLoS ONE 8(9), e73856 (2013).
62. Creel, S. & Creel, N. M. Six ecological factors that may limit African wild dogs. Lycaon pictus. An. Cons. 1, 1–9 (1998).
63. Gould, S. J. Kropotkin was no crackpot. Nat. Hist. 106, 12–21 (1997).
64. Dawkins, R. e Selsh Gene (Oxford University Press, 1976).
65. Teilhard de Chardin, P. e fossils from locality 18, near Peking. Geol. Survey China 9 (1940).
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2021) 11:13501 | 
www.nature.com/scientificreports/
66. Koufos, G. New material and revision of the Carnivora, Mammalia from the Lower Pleistocene Locality Apollonia 1, Greece.
Quaternary 1, 6 (2018).
67. Petrucci, M., Cipullo, A., Martínez-Navarro, B., Rook, L. & Sardella, R. e Late Villafranchian (Early Pleistocene) carnivores
(Carnivora, Mammalia) from Pirro Nord (Italy). Palaeontogr. Abteilung A Palaozoologie - Stratigr. 298, 113–145 (2013).
68. Geraads, D. A. A revision of the fossil Canidae (Mammalia) of north-western Africa. Palaeontology 54, 429–446 (2011).
69. Vislobokova, I. A. & Agadjanian, A. K. New data on large mammals of the Pleistocene Trlica fauna, Montenegro, the Central
Balkans. Paleontol. J. 49, 651–667 (2015).
70. Bishop, M. J. e mammal fauna of the early Middle Pleistocene cavern inll site of Westbury-sub-Mendip, Somerset. Spec. Pap.
Palaeontol. 28 (1982).
71. Del Campana, D. I cani pliocenici di Toscana. Palaeontogr. Ital. XIX, 189–254 (1913).
72. Sher, A. V. Olyorian land mammal age of Northeastern Siberia. Palaeontogr. Ital. 74, 97–112 (1986).
73. Wickham, H. ggplot2: Elegant Graphics for Data Analysis. Use R. (Springer, 2009).
74. Von den Driesch, A. A guide to the measurement of animal bones from archaeological sites: as developed by the Institut für Palaeo-
anatomie, Domestikationsforschung und Geschichte der Tiermedizin of the University of Munich (Peabody Museum Press, 1976).
Acknowledgements
e authors are indebted to the kindness and availability of the curators who granted access to the collections of
their institutions and museums: M. Bukhsianidze of theMG-GNM; P. Agnelli of Museo di Zoologia "La Specola",
Università degli Studi di Firenze; E. Cioppi of Museo di Geologia e Paleontologia, Università degli Studi di
Firenze; J. Galkin, and J. Meng of the American Museum of Natural History; G. Koufos of Aristotle University
of essaloniki and E. Gilissen and W. Wendelen from Royal Museum for Central Africa. Dmanisi research is
supported by Shota Rustaveli Georgian National Science Foundation, Laboratory equipment is provided by
Alexander von Humboldt Foundation.is study is framed within a wider Georgian-Italian collaborative pro-
ject (bilateral agreement between the University of Florence and the Tbilisi State University “I. Javakhishvili”
/ Georgia National Museum) supported by the Italian Embassy in Georgia. e latter is acknowledged for the
continuous support to L.R. and S.B.L. while working in Tbilisi. e authors are indebted to M. Bukhsianidze for
the fruitful discussion on a previous version of this manuscript that helped improve the text, and to L. Salimei for
the rewarding conversations on mutually benecial cooperation and reciprocity. e Italian Ministry for Foreign
Aairs (DGPCC-V) is acknowledged for nancially supporting Italian paleontological research in Georgia. is
study has been funded by the University of Florence (Fonti di Ateneo to L.R.and Fondi di Internazionalizzazione
to L.R. and S.B.L.), the Spanish Agencia Estatal de Investigación (grants CGL2016-78577-P, CGL2016-80975-P,
CGL2017-82654-P, AEI/FEDER-UE) and the Generalitat de Catalunya (CERCA Program GENCAT 2017SGR
859; SGR 416 GRC, AGAUR, Generalitat de Catalunya). Part of this research was also funded by the Synthesys
project to J.M.-M. (BE-TAF-5471).
Author contributions
S.B.L., J.M.-M., B.M.N., P.P., L.R. and D.L. conceived and designed the experiments. S.B.L., J.M.-M., B.M.N.,
P.P. and L.R. wrote the paper and prepared gures and tables. All authors analyzed the data and reviewed dras
of the paper.
Competing interests
e authors declare no competing interests.
Additional information
Supplementary Information e online version contains supplementary material available at https:// doi. org/
10. 1038/ s41598- 021- 92818-4.
Correspondence and requests for materials should be addressed to S.B.-L.orB.M.-N.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access is article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons licence, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
© e Author(s) 2021
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com