Earlier this year a new study identified a fossil canid, excavated from Razboinichya Cave in the Siberian Altai Mountains, as a member of the dog lineage. Indeed, mtDNA of this Pleistocene specimen of the ‘Mammoth steppe fauna’ resulted most similar to that of most modern dogs.
The origin of domestic dogs remains controversial, with genetic data indicating a separation between modern dogs and wolves in the Late Pleistocene. However, only a few dog-like fossils are found prior to the Last Glacial Maximum, and it is widely accepted that the dog domestication predates the beginning of agriculture about 10,000 years ago. In order to evaluate the genetic relationship of one of the oldest dogs, we have isolated ancient DNA from the recently described putative 33,000-year old Pleistocene dog from Altai and analyzed 413 nucleotides of the mitochondrial control region. Our analyses reveal that the unique haplotype of the Altai dog is more closely related to modern dogs and prehistoric New World canids than it is to contemporary wolves. Further genetic analyses of ancient canids may reveal a more exact date and center of domestication.
These results bolster previous ideas of an early appearance of domesticated dogs, though how they relate with their wolf kin remains shrouded in mystery. The resolution considerably exceeds previous results on Belgic samples thousands of miles away, where just 57 base pairs could be extracted from the mitochondrial control region of six large Pleistocene canids from Goyet cave (samples G1-6), and one from Trou des Nutons (sample TN-1), what unfortunately didn’t allow the reconstruction of much phylogenetic structure. The new results confirm the latter as a feature of Pleistocene dogs rather than a defect in previous results:
When investigating the phylogenetic informativeness of our dataset combining the Altai specimen, 72 extant dogs and 30 wolves, 35 prehistoric New World canids and four coyotes we also found low support for either a clearly resolved branching pattern or star-shaped evolution (Druzhkova et al., 2013)
Apparently dog evolution has an extremely long trajectory, and became even more complicated by the strong impact of much later, Holocene events. Their early divergence is also suggested by outstanding differences in morphology that distinguish dogs from wolves in their shortened muzzles, broader palates, crowded teeth and the broad, heavy frontal shields at the top of their skulls. The oldest (unsampled) Goyet dog, at 36,000 cal BP even older than the Altai specimen, was ‘not intermediate in form between the fossil wolves and the prehistoric dogs, but conforms to the configuration of the other Palaeolithic dogs, which are approximately 18,000 years younger’ (Germonpré et al., 2009). However, despite Goyet also supplied a representative morphological reference, Druzhkova’s team didn’t try to establish a link with this Belgium phenomenon of early dog occurrences. Even worse, the authors utterly ignored new insights that rather urge for a revision of the canid phylogenetic tree in the vein of Pilot et al. (2010). The traditional concept of a nested dog-clade, essential to the identification of Canis lupus familiaris (the domestic dog), and their putative geographical origin, already turned out incompatible with the results of Goyet. New inconsistencies may confirm the phylogenetic tree employed by Druzhkova’s team was already obsolete.
The view that dogs are the tamed descendants of the common wolf has always been the most popular one, though the scrutiny of investigators yielded ambiguous evidence. Prolific contacts between wolves and early modern humans rather tend to obfuscate early progress in the process of dog domestication:
The Palaeolithic sites yielding putative dog remains might be best described as extended-use camps or hunting sites and some of these are associated with dwellings made from mammoth bone (e.g., Germonpré et al., 2008, 489). Unambiguously identified wolf remains occur routinely in many of these sites, often at relatively high frequencies for a large carnivore. (Crockford & Kuzmin, 2012)
Incidental dog-like canids could already be distinguished in Upper Palaeolithic sites, without being sure these individual creatures could indeed be derived of tamed dogs in the periphery of human habitation rather than being the representatives of an extinct lineage of wolves. Interaction with Neanderthal may remain the only evidence that truly suits the popular preposition of the tamed wolf pur sang:
Trou de la Naulette is a famous Neanderthal site excavated by Dupont in the 1860s […]. According to the notes of Dupont, the fossil canid skull was found in the Second Horizon, the same containing the Neanderthal remains.
The Trou de la Naulette […] skulls have missing values and cannot be identified. (Germonpré et al., 2009)
Only the first appearance of dog-like features in fossils offered the opportunity to prove the acceptance of wolves as pets. For over a century Pleistocene fossils with a dog-like anatomy have been found littered over a wider area in northern Eurasia, and despite an incidental association with prehistoric human presence some reasonable doubts remain about the progress of their domestication. Dogs have traditionally been regarded the tamed version of Eurasian wolves, with the implicit assumption their differences could be fully explained by human interference and a domestic lifestyle. Earlier publications even went a lot further. Wayne et al. (1999) asserted on the basis of mtDNA that ’dogs and gray wolves may have diverged […] about 135,000 years before present’, adding that dogs ‘may have been domesticated earlier and have only recently changed in conformation with changing conditions associated with the shift from hunter-gather cultures to more agrarian societies about 12,000 years ago’. Genetic variety of dogs even exceeds the variety of the Holarctic wolf. Far from being their obvious “ancestral” counterparts, genetically speaking wolves falls within the range of genetic variety of dogs while on the basis of descend rather the opposite should be expected:
Greater mtDNA differences appeared within the single breeds of Doberman pinscher of poodle than between dogs and wolves. Eighteen breeds, which included dachshunds, dingoes and Great Danes, shared a common dog haplotype. Alaskan malamutes, Siberian huskies and Eskimo dogs also showed up in the common dog haplotype and were no closer to wolves than poodles and bulldogs. These data make wolves resemble another breed of dog. (Serpell, 1995, p.33)
Indeed, dogs are clearly the most diverse domestic species, both genetically and by morphology. Hence, Pleistocene dog domestication was traditionally considered a quite intuitive concept. Modern investigation tend to confirm the unique genetic identity of dogs in relation with wolves:
We identified 3,786,655 putative single nucleotide polymorphisms (SNPs) in the combined dog and wolf data, 1,770,909 (46.8%) of which were only segregating in the dog pools, whereas 140,818 (3.7%) were private to wolves (Axelsson et al., 2013)
However, the pendulum of scientific ‘consensus’ already tended to swing away from Pleistocene domestication in favor of still ill-understood Neolithic events when Druzhkova’s team caused considerable confusion with their published results on the Altai specimen, the second oldest Pleistocene dog. It is funny to watch great names on the scientific scene stumble in their enthusiasm, utterly unable to digest the full spectrum of conflicting information and grasp the consequences. Probably to many this new results were all too much in too short a time. Indeed, the Pleistocene Altai dog emerged genetically closer to modern dogs than to contemporary wolves, but there is more. The specimen also attested closely related with prehistoric New World canids, including both pre-Columbian domestic dogs and extinct eastern Beringian wolves, the latter being no less revolutionary for our understanding of dog evolution.
One glance at Druzhkova’s Consensus Neighbor Joining Tree makes clear the mentioned New World canids don’t simply refer to the pre-Columbian dogs that Native Americans brought with them when their ancestors left Asia. This study also incorporates the extensive mtDNA database of extinct eastern Beringian wolves published by Leonard at al. (2007). In 2002, Savolainen et al. published the discovery of three wolf mtDNA haplotypes found in wolves of China and Mongolia that cluster within the main dog haplogroup or clade A (‘A total of 71.3% of dogs had haplotypes belonging to clade A’), but now also this older group of ice age wolves has its mtDNA nesting within this main dog clade. This haplotype can’t be found in current American wolves, and apparently has an age and origin that overrule any special and tight association with east Asian wolves and dogs. Moreover, the eastern Beringian wolf diverged in many ways from current Holarctic wolves, while other features were more dog-like. The sub-species was specialized to consume more carrion and differed significantly in shape from ‘both the sample of modern North American wolves and Rancho La Brea (California) Pleistocene wolves’ (Leonard et al., 2007). Contrary to their contemporary southern neighbors and modern wolves, ‘[e]astern Beringian wolves tend to have short broad palates, probably adapted for producing relatively large bite forces suggesting the killing of bulky prey, such as bison and horse’ (Leonard et al., 2007). Apparently holarctic wolf subspecies were once distinguished along ecological lines that ceased to exist:
Notably, the Pleistocene C. lupus from eastern Beringia by the skull shape, tooth wear, and isotopic data is also reconstructed as specialized hunter and scavenger of extinct North American megafauna (Leonard et al., 2007). In addition, East-Beringian wolves genetically differ from any modern Northern American wolf, and instead they appear most closely related to Late Pleistocene wolves of Eurasia. This uniquely adapted and genetically distinct wolf ecomorph suffered extinction in the late Pleistocene, along with other megafauna.(Baryshnikov et al., 2009)
For sure the origin of the modern ‘Holarctic’ grey wolves lays in more southern latitudes, without much certainty if the wolf that ultimately replaced the variety of more northern ‘ice age’ latitudes was already part of a single wolf population that encompassed the southern glacial borderlands and refuges of Eurasia and North America, or that they expanded from any place in particular. Most likely the cradle of modern wolf evolution was somewhere south of the glacial habitat. Ancient strains of the grey wolf are nowadays recognized in the Egyptian jackal, otherwise often considered transitional to the golden jackal, that reveal mtDNA divergence with grey wolves up to about 800 kya. The Indian subcontinent still includes three diverse, distinct wolf lineages — Indian wolf, Tibetan wolf and Eurasian wolf. Mitochondrial DNA analysis suggests the Himalayan wolf is distinct from the Tibetan wolf, and represents the most divergent wolf lineage. Such genetic diversity is generally considered indicative of origin. We need to be cautious, though, since we don’t know the hybridization record of South Asian wolves, that shared their habitat for thousands of years with other canids. The recent detection of gray wolf mtDNA in the golden jackal of Senegal even ‘brings the delineation between the African wolf [Canis lupus lupaster] and the golden jackal [Canis aureus] into question’ (Gaubert et al., 2012). But even the Golden Jackal is genetically closer related than the coyote (Rueness et al., 2011, based on 726 bp of the Cyt b gene: a region of mitochondrial DNA commonly considered useful to determine phylogenetic relationships between organisms and within families and genera due to its sequence variability), hence its western distribution could eventually imply a southwestern Eurasian origin of the wolf. This location of origin may be further elaborated and even extended to the Holarctic wolf by the phylogenetic investigation of Pilot’s team (2010). Their results distinguished a single major subdivision of Holarctic wolves based on 230 bp mtDNA control region sequences of 947 contemporary European wolves that – contrary to Indian and Himalayan wolves – also applies to the worldwide wolf population:
Based on the phylogenetic trees and networks constructed, we defined two haplogroups, 1 and 2
The two haplogroups were separated by five mutational steps: three transitions, one transversion, and one insertion/deletion.
In all the trees and networks constructed, haplotypes from the two groups were clearly separated, although they did not always constitute two monophyletic clades
Haplotypes of Indian and Himalayan wolves were separated from all the other haplotypes by more than 6 mutational steps
In ancient European wolves, haplogroup 2 was predominant. All ancient samples from Belgium, Germany, Czech Republic, Hungary, and Ukraine, ranging in age from 44,000 to 1,400 years B.P., belonged to this haplogroup. Only one haplotype of ancient wolf (w7), sampled in western Russia and dated from 2,700 – 1,200 years ago, belonged to haplogroup 1
Our results are consistent with Germonpré et al. , who showed that the ancient European haplotypes are placed in one part of the wolf haplotype network rather than being scattered across the complete network.
[…] contemporary data may suggest substantial changes in haplogroup frequencies over the last 40,000 years from the predominance of haplogroup 2 to the predominance of haplogroup 1.
[…] mtDNA haplotypes of Pleistocene wolves from eastern Beringia belonged to a distinct haplogroup that does not occur in contemporary North American wolves. This haplogroup corresponds to haplogroup 2 in our study […] and some of the ancient European and Beringian wolves even shared a common haplotype (Pilot et al., 2010)
The ambiguous assignment of only one haplotype (w20, from Turkish Trakia) and the intermediate position of Balkan wolves in general (figure 1 in Pilot et al., 2010) is in agreement with the aforementioned southwest Eurasian origin of Holarctic wolves, and even more specific to the origin of modern wolves whose mtDNA cluster in haplogroup 1. Instead, dogs cluster to ancient European wolves and their relatives in the Pleistocene arctic region, whose mtDNA cluster in haplogroup 2. The ancient predominance of dog-like mitochondrial DNA in wolves thus can’t be exclusively associated with some ‘Beringian-like’ tendencies in the Pleistocene Eurasian wolf phenotype, that otherwise has all appearance to be intrusive from the east. This counter-intuitive geographic separation of ancestral Holarctic wolves in the west, nearby old centers of human occupation, and dog-like haplotypes far away from any human interference at all, isn’t necessarily irreconcilable to dog domestication. Actually, wild animals that had the opportunity to co-evolve with humans never experienced the same magnitude of extinction that happened e.g. in the Americas: ‘It has long been thought that extinctions in Africa were less severe than in other regions of the world due to the long term coevolution of humans and megafauna’ (Louys et al., 2006). In general, such animal populations being situated closer to the known centers of human evolution turned out less prone to domestication as well. Coevolution apparently involved wild animals getting wilder and less prone to domestication. Animals that remained curious and less cautious for humans met their end during and after the great megafaunal extinctions. The Beringian wolf phenotype may have belonged to the latter category, and became extinct. However, before this happened some individuals were attested to have already mixed with western wolf populations. Probably it was no coincidence this phenotype emerged as the prime target of human interaction and, eventually, domestication.
The nature of Holocene extinction, that started at the end of the last Ice Age, is contested, though generally considered due to both climate change and the proliferation of modern humans. The impact of this phenomenon was not limited to the extinction of numerous species. Hofreiter (2010) stressed that ‘studies like that by Pilot et al. are starting to reveal that at least some surviving species are also depleted in genetic diversity compared with their Pleistocene ancestors.’ Somehow the dogs were exempted from this depletion, for they still are ‘a genetically diverse species that likely originated from a large founding stock possibly derived from wolf populations existing in different places and at different times’ (Vilà et al., 1999) – even though most of these ancestral wolf populations are now extinct, and replaced by new wolf populations that are not representative to the founding stock. The effect of this Late Pleistocene change of grey wolf populations was huge. Meachen’s team (2012) ‘suggest that Pleistocene coyotes may have been larger and more robust in response to larger competitors and a larger-bodied prey base’, being consistent with the view that even the early American grey wolves in the tar pits could thrive in their niche only after this post-megafaunal development of the coyote evolving away from wolf-like forms. Leonard et al. (2007) relied ‘on an early arrival of the more gracile wolf from the Old World and migration to areas below the Wisconsin ice sheet’ to explain the presence of two distinct Pleistocene gray wolves in North America, though actually the southern type is rare or problematic before 10,000 BP while eastern Beringian wolves attest too many infinite dates and genetic diversity to have simply evolved locally and derive from this hypothesized population of immigrant American common ancestors. Instead, sudden grey wolf competition may have supplied an additional reason for the sudden post-Glacial downsizing of the coyote, while also current hybridization patterns in North America tend to confirm the modern grey wolf’s more recent introduction from Eurasia:
[…] extensive admixture zones in North America, where four morphologically distinguishable wolf-like canids can potentially interbreed: the gray wolf of Old World derivation, the coyote and red wolf (both of which originated in North America), and the Great Lakes wolf. (VonHoldt et al., 2011)
Meanwhile, the genetic tie of eastern-Beringian wolves with Eurasian gray wolves – and early dogs! – discourages premature conclusions on dog domestication in the Americas, not in the least because this tie considerably precedes the attested Native American immigration ~15,000 years BP. More likely, eastern Beringian wolves managed to preserve their dog-like genetic legacy longer than elsewhere on the northern hemisphere, until worldwide disaster struck the ice age megafauna:
the eastern-Beringian ecomorph was hypercarnivorous with a craniodental morphology more capable of capturing, dismembering, and consuming the bones of very large mega-herbivores, such as bison. When their prey disappeared, this wolf ecomorph did as well, resulting in a significant loss of phenotypic and genetic diversity within the species. Conceivably, the robust ecomorph also was present in western Beringia in the Late Pleistocene, but specimens were not available for this study. (Leonard et al., 2007)
The ecological and derived morphological segregation of Pleistocene wolves in a northern ice age subspecies and the subspecies of southern ice age refuges appears to have had a genetic component that survives in dogs and wolves respectively:
According to Hofreiter , this may imply that Pleistocene wolves across Northern Eurasia and America may have represented a continuous and almost panmictic population that was genetically and probably also ecologically distinct from the wolves living in that area today. (Pilot et al., 2010)
The southern subspecies may have comprised several isolated gray wolf populations in Eurasian and American ice age refuges, not all of which necessarily survived the global wave of post-glacial loss of diversity, or extinction. Indeed, ‘[…] most of the genetic diversity of megafaunal animals may have been lost at the end of the Pleistocene, even in surviving species’ (Hofreiter 2007). Some of the general post-glacial loss of diversity may be due to more recent expansions and replacements, and since most of the sampled eastern Beringian wolves attest ‘infinite dates’ (Leonard et al., 2007 sup.) it is likely their stock was already present in the region for a longer time. However, at least the northern subspecies is strongly indicated to have been a single Pleistocene population that was also genetically distinct:
[Leonard et al., 2007] found evidence that the Beringian wolves were morphologically different from modern North American wolves and from Pleistocene wolves from more southern regions. Moreover, the differences in morphology suggest that the Pleistocene Beringian wolves were adapted to hunting and scavenging members of the now extinct megafauna, a conclusion supported by isotope analysis. Finally, these wolves not only represented a different ecomorph, they were also genetically distinct. Not a single sequence of their mitochondrial DNA haplotypes exactly matched sequences found in modern and historical wolves identified to date. However, some of the sequences perfectly matched, albeit only for short stretches, sequences obtained from Eurasian Pleistocene wolves, from as far west as the Czech Republic. Thus, Pleistocene wolves across Northern Eurasia and America may actually have represented a continuous and almost panmictic population that was genetically and probably also ecologically distinct from the wolves living in this area today. Despite their high mobility, these wolves did not escape the megafaunal extinctions at the end of the Pleistocene, even though the causes of their extinction are unclear. The specialized Pleistocene wolves, thus, did not contribute to the genetic diversity of modern wolves. Rather, modern wolf populations across the Holarctic are likely be the descendants of wolves from populations that came from more southern refuges as suggested previously for the North American wolves (Hofreiter, 2007)
Germonpré (2009) noted similar trends like Beringian wolves in skulls from Trou Baileux, Trou des Nutons, Mezin 5469, Mezin 5488 and Yakutia, that all are considered to be from fossil wolves. Leonard et al. (2007) analyzed the eastern Beringian wolves genetically and attributed the disappearance of this robust prehistoric ecomorph to ecological changes that occurred after the Last Glacial Maximum. Even though her team also based themselves on a database of pre-Columbian dogs when they rejected domestication of modern North American wolves for the dogs of Native Americans, only Druzhkova’s team combined both data sets to make a genetic comparison between Beringian wolves and dogs. His unfortunate inadequacy to compare his results on the Altai dog with Germonpré, whose results on Pleistocene Belgian dogs of comparable age and morphology were hampered by the low resolution of just 57 bp of the mitochondrial control region, may be solved by the results of Pilot’s team he was not acquainted with. Like the Pleistocene Belgium dogs of Goyet Cave, Altai dog of Razboinichya Cave clearly cluster with ancient Holarctic wolves rather than modern wolves, but their close genetic relationship with modern dogs is based on the virtual extinction of the Holarctic wolf source population rather than some advanced stage of dog domestication. As such, criticism against the existence of Pleistocene ‘dogs’ is still valid:
One of the features that M. Germonpré et al.’s so-called ‘Palaeolithic dogs’ (and the Razboinichya specimen) have in common is that they represent one or a few dog-like individuals found amongst many typical wolf individuals.
[…] the ‘Palaeolithic dogs’ […] may simply be rather ‘short-faced’ wolf individuals that lived within a population of typical wolves that interacted in various ways with human hunters. While the dog-like morphology of some Late Pleistocene wolves may have arisen due to persistent interactions with people over varying lengths of time, it is misleading to call this relationship ‘domestication.’ (Crockford & Kuzmin, 2012)
The similarity of Pleistocene dogs with certain types of Pleistocene wolves may be deceiving. Not unlike Beringian wolves, the Paleolithic dogs tend to have short, broad palates with still large carnassials:
Compared to wolves, ancient dogs exhibit a shorter and broader snout (Lawrence, 1967; Olsen, 1985; Sablin and Khlopachev, 2002). All Palaeolithic dogs in our study conform to this pattern. (Germonpré et al., 2009)
European Palaeolithic dogs, characterized by, compared to wild wolves, short skulls, short snouts, wide palates and braincases, and even-sized carnassials (Germonpré, 2012a)
There exists an older notion of a late Pleistocene lineage of dag that might have existed during the late Pleistocene until these became globally extinct during the last 20,000 years. This possibility of ‘an aberrant lineage of dog-like canids might have existed throughout the northern hemisphere during the late Pleistocene’ was recently explored by Thalmann, to the somewhat unsettling result that no exact mtDNA matches could be retrieved in modern dogs or modern wolves. At least the sudden Neolithic reduction of the carnassials in the perceived dog lineage may indicate a recent bottleneck that could explain such an extinction of mtDNA lineages for dogs, while the extinction of these mtDNA lineages in wolves has already been cited as a typical post-Glacial phenomenon: also ‘the mtDNA of the Belgian fossil wolves shows a large amount of genetic diversity that has not been described for modern wolves’ (Germonpré et al., 2009).
This ‘aberrant lineage of canids’ probably existed in the ecomorph stock that had its gravity in NE Asia and northern America, and incidentally reached far into the west. Being closely related with Eurasian wolves, an early tendency towards speciation was arrested by the Late Glacial events that led to the predominance of the Holarctic wolf. Mere extinction may have been the result, or advanced hybridization processes that eventually lead to ‘the fusion of the parent species’ gene pool and a loss of species.’ (Reece et al., 2011), not unlike the ABC Island brown bears that are recently identified as the descendants of a polar bear population that was gradually converted into brown bears via (male-dominated) brown bear admixture: ’This process of genome erosion and conversion may be a common outcome when climate change or other forces cause a population to become isolated and then overrun by species with which it can hybridize’ (Cahill et al., 2013). By the time the Eurasian ice age wolf ecomorph became extinct, human interaction or ‘dog domestication’ was already ongoing, leading to the preservation of much of the genetic diversity that has disappeared in wolves. The now attested genetic continuity in the Pleistocene Altai dog supplies evidence it didn’t take such a long time since the Pleistocene to get nowhere. On the contrary, the origin of dogs can now be perceived firmly rooted in the Pleistocene whatever the type of domestication applicable to the earliest interactions with humans. Germonpré (2012b) announced a forthcoming investigation of stable isotopes on Pleistocene dogs to verify if their diet can be related to that of prehistoric humans from the same time and region. Instead, Belgian Pleistocene wolves were already attested to consume the same kind of bulky prey that Leonard et al. (2007) deduced for eastern Beringian wolves based on their bite, such as horse and large bovids. Could the Pleistocene abundance of meat offer an alternative explanation for the unaltered size of the carnassials of Pleistocene dogs? If so, would there exist any other criteria that suffice to prove the veracity of Pleistocene dog domestication at all?
The persistence of large carnassials up to Neolithic times would also imply a dependency of fresh meat that may have been ever harder to obtain. On the eve of the Neolithic transition, when increased population pressures virtually forced human culture into the direction of food production, a dog must have been an expensive asset only a few could afford. Until recently everywhere in Polynesia starch-rich poi from the taro corm was used to fatten the dogs for use as food because meat was too valuable to be used as dog food. This transition of breeds like the Hawaiian Poi dog and the Maori Kuri dog, both extinct, from predators to fattened up food resources, was only possible in a Neolithic society with an abundance of starch-rich products. The Neolithic change to starch-rich diets induced some dogs to adapt genetically (Axelsson et al., 2013). This important event in dog evolution apparently caused an explosion of population growth and genetic replacement – until very recently confused with a one and only domestication event in the Neolithic rather than already during the Pleistocene.
Only nowadays dog domestication is being appreciated as a true evolutionary process. Their divergence from wolves wasn’t just a superficial accumulation of qualities favored by human preferences, such as color and character, hardly worthy of major biological interest. Instead, novel adaptations including brain functions, starch digestion and fat metabolism indicate more fundamental changes, each having evolutionary advantages all of their own that tend to set dogs apart from wolves. One amazing result of current investigation is these changes must have been closely associated with dog populations in Southeast Asia, even to the point that dogs were hypothesized to originate from an otherwise unspecified Asiatic wolf, albeit ‘with minor genetic contributions from dog–wolf hybridization elsewhere’ – for this occasion assumed to have inhabited eastern Asia south of the Yangtze River (Ding et al, 2011). Actually, there is no shred of evidence wolves or other members of the “Canis” genus were present in this area, thus adding up to a major inconsistency in dog evolution:
The earliest archaeological evidence of ancient dogs was discovered in Europe and the Middle East, some 5–7 millennia before that from Southeast Asia. However, mitochondrial DNA analyses suggest that most modern dogs derive from Southeast Asia, which has fueled the controversial hypothesis that dog domestication originated in this region despite the lack of supporting archaeological evidence. (Sacks et al., 2013)
Once again high genetic variability emerged as a poor advice to pinpoint the geographic origin of a species. Not unlike the results on human DNA, for dogs the archeological record is all out of tune with genetic evidence, or the way it is currently interpreted. Also for dogs scientific efforts to understand the origin were increasingly directed away from observations considered inconvenient to the traditional ‘scientific’ scenarios, and aberrated in the realm of increasingly unintelligible constructs. Intriguing parallels with the Recent Out of Africa (ROA) paradigm may be elucidated, where high human genetic variability of the purported African homeland is alternately considered evidence for a recent origin of modern humans – or the result of archaic hybridization (Hammer et al., 2011). Likewise, modern science erroneously sought the putative center of dog domestication where genetic variability reaches peak values, that for dogs could be found in Southeast Asia. Dogs even mimic the genetic signals of change in modern humans:
Sets of functionally related genes show highly similar patterns of evolution in the human and dog lineages. This suggests that we should be careful about interpreting accelerated evolution in human relative to mouse as representing human-specific innovations (for example, in genes involved in brain development), because comparable acceleration is often seen in the dog lineage. (Lindblad-Toh et al., 2005)
Rather than being deterred by the lack of archeological evidence, modern investigation contented itself with pursuing false positives to discard Pleistocene dogs like this Altai canid as an incipient dog that did not give rise to late Glacial – early Holocene lineages and probably represents wolf domestication disrupted by the climatic and cultural changes associated with the LGM (Ovodov et al, 2012). For convenience of the naysayers this fate was shared by an abundance of other ‘incipient dogs’ allegedly up to 36,000 BP whose remains were found littered over a wide area, not only including the canid skulls from Goyet (Belgium), that so far counts as the oldest specimen, but also from disparate places like Predmostí (the Czech Republic), and Mezin and Mezhirich (the Ukraine). Four out of thirteen ambiguous finds fished from the bottom of the North Sea (Southern Bight), were firmly identified as Pleistocene dogs in the thesis of Datema, 2011. In the deepest part of the famous Chauvet cave, France, a track of footprints from a large canid could be associated with the one of a child (Garcia, 2005), whose torch wipes were dated at c. 26,000 BP. Still other investigators remained to even discard the notion of these Pleistocene dogs being incipient at all – to the chagrin of Germonpré’s team that as late as 2012 felt obliged ’ to remedy some errors, misunderstandings and misrepresentations’ on the case. Now, the results of Druzhkova’s team blow away this apparent prevalence of a Southeast Asian origin of dog domestication in more recent scientific investigation, that not unlike the Recent Out of Africa paradigm became genetically supported by unfortunate conclusions on variability of mtDNA (Savolainen et al., 2002) and of Y-DNA (Ding et al., 2010) respectively. However, unlike ROA, archeological and fossil evidence that a common ancestor was present in the putative geographic origin, i.e. traces of early domestication and a habitat for prehistoric wolves in Asia South of Yangtze River (ASY) as proposed by Pang et al. (2009), is still dearly missing.
The study of VonHoldt et al. (2010) had already revealed disproportionate hybridization especially for the ancient breeds in east- and southeast Asia by Chinese and Middle Eastern wolves (supplemental Figure 5 at k=5). Apparently hybridization inflated heterozygousity of south-east Asian nuclear DNA and lured the scientific community into an “Out of Southeast Asia hypothesis. Instead, Sacks et al. (2013) introduced a new interpretation of this Southeast Asian hypothesis by stating that ‘isolation of Neolithic dogs from wolves in Southeast Asia was a key step accelerating their phenotypic transformation’, to offer a scenario where a new type of dog emerged in the Neolithic that virtually pushed earlier Palaeolithic dogs into extinction:
Archaeological and ancient DNA evidence suggesting that late Palaeolithic dogs were replaced by Neolithic immigrants in regions as disparate as Japan, the Middle East, and North America
The apparent origins of most modern dog matrilines from Southeast Asia has been interpreted as evidence that dogs were first domesticated in this region (Savolainen et al. 2002; Pang et al. 2009). However, the lack of archaeological evidence of dogs in Southeast Asia until some 5,000–7,000 years later than in central and western Eurasia, suggests either that the single genealogical history reflected in mtDNA could be misleading (e.g. VonHoldt et al. 2010) or that most modern dogs trace their ancestry proximately to Southeast Asia, but as a secondary center of diversification associated with Neolithic rather than Palaeolithic peoples (Brown et al. 2011).
[…]our findings support the hypothesis for a massive Neolithic expansion of dogs from Southeast Asia rather than a Palaeolithic origin of dogs from this region. (Sacks et al., 2013)
Thus accelerated Neolithic evolution and the absence of wolves in Southeast Asia to mate with, are now considered the main agents for the rapid change of Palaeolithic dogs into modern dogs, the latter being especially successful in increasing their numbers on a worldwide scale.
Inadvertedly, hybridization was here mentioned as a factor to be dealt with. Actually, recurrent hybridization events already prevented Canis evolution to be a straightforward linear succession towards present-day species. This process should also have blurred out the nascent differences typical of a more detailed branching tree full of evolutionary dead-ends, as traditionally employed also for canid evolution. Again, not unlike human evolution, mosaic evolution was a feature also for canids that can be traced back even to the early divergence of wolves and jackal:
considering C. etruscus and C. arnensis as wolf-like and jackal-like dogs, respectively, is an oversimplification not always valid, because C. arnensis is more similar to C. lupus than C. etruscus regarding some cranial characters (e.g., Fig. 1b), while C. etruscus seems to exhibit a broader set of peculiar features. Recent genetic studies confirm this hypothesis, as jackals do not constitute a homogeneous genetic group despite their great morphological affinity, albeit different analyses do not fully agree on their phylogenetic relationships (Cherin et al., 2013)
Some prehistoric canid characteristics are simply ancestral, what back in time can be illustrated by some degree of convergence to rather coyote-like forms, suggesting even jackals evolved partly parallel with the ancestors of the gray wolf:
This paper reports a new species of dog (Canis accitanus nov. sp.) from the Fonelas P-1 site (dated close to the Plio-Pleistocene boundary) in Granada, Spain. This new taxon shows cranial features more similar to coyote-like dogs (C. lepophagus, C. priscolatrans, C. arnensis or C. latrans) than to wolf-like dogs (C. etruscus, C. mosbachensis or C. lupus), such as a long and narrow muzzle, a little-developed sagittal crest and frontal bones raised only a little above the rostrum. (Garrido & Arribas, 2008)
As raised in a previous blog, in some cases mosaic evolution may be the result of hybridization events or cross-species gene flow during some extended period during speciation, or even afterwards as long as cross-breeding may still result in viable offspring. According to Gaubert et al. (2012) ‘hybridization among Canis taxa has proved to be common and to involve significant phenotypic changes in hybrid generations, reaching fixation in several cases’.
Naturally, cross-breeding may result in increased variability – as long as the elimination of deleterious hybrid fitness components by selection doesn’t purge most of the polymorph alleles. However, admixture with modern wolves can’t fully explain the genetic deviation of modern dogs since much of their polymorph alleles simply can’t be found in wolves. This does not contradict wolf admixtures in a later stage, what apparently occurred in the ancestors of Neolithic dogs. Since ‘[…] the European and American breeds clustered almost entirely within the Southeast Asian clade, even sharing many haplotypes, suggesting a substantial and recent influence of East Asian dogs in the creation of European breeds’ (Brown et al., 2011), it shouldn’t come as any surprise that ‘morphologic comparisons suggest that dogs are closest phenotypically to Chinese wolves’ (Wayne et al., 1999). The evolution of Neolithic dogs must have been highly impacted by Pleistocene dogs near East Asia, what suggests late-Pleistocene admixture to explain that ‘one osteological feature diagnostic of dogs is also found among Chinese wolves’ (Savolainen et al., 2002). However, the more complex hybrid character of all descendant dog breeds may be illustrated by the ‘higher proportion of multi-locus haplotypes unique to grey wolves from the Middle East, indicating that they are a dominant source of genetic diversity for dogs rather than wolves from east Asia, as suggested by mitochondrial DNA sequence data’ (VonHoldt et al., 2010).
Hybridization between dog breeds require a broad genetic base to start with. However, the direct ancestors of modern wolves and their deduced range of genetic variability don’t supply a sufficient source for the total range of dog variability. Still, an abundance of other traits attest the impact of hybridization processes. Neoteny, the retention of juvenile characteristics into adulthood, is often considered an important characteristic of dogs. Serpell considered this a result of hybridization:
Our feeling on the development of breeds is expressed by Haldane (1930, pp. 138), writing about the evolution of species, when he stated that there is ‘every reason to believe that new species may arise quite suddenly, sometimes by hybridization, sometimes perhaps by other means. Such species do not arise, as Darwin thought, by natural selection. when they have arisen they must justify their existence before the tribunal of natural selection’. (Serpell, 1995, p.42)
Brown’s team (2011) does a remarkable job in distinguishing a Southeast Asian development in recent dog evolution, though his classification doesn’t suffice to reveal the origin of dog Y chromosome haplotypes. Genomic analysis of prehistoric canids, modern dogs and wolves indicate a basal placement of some dog haplotypes in the phylogeny. One middle-eastern haplotype actually resulted “ancestral”, being attested also in Dhole, black-backed jackal and a wolf from China (haplotype 12). Middle-eastern dog Y-chromosome haplotypes resulted shared with wolves and/or Southeast Asian dogs (haplotypes 8, 10 and 11) or none (haplotypes 7 and 9, though both being rather distantly related with wolves). Only Southeast Asian dogs were easily distinguished from all other related canines combined. Despite alignment bias being possibly the prime reason for these results, more recent wolf admixture and even cross-species hybridization can’t be ruled out. Indeed, the most basal members of Canis tend to be more “doggish” in looks and behavior, while the more “wolfish” phenotype of the northern hemisphere seems to have made its appearance only about 300,000 years ago. In other words, canids are increasingly “doggish” with their genetic distance from true grey wolves, what can be illustrated by the incremental series Indian wolves; Egyptian wolves; Golden jackal; coyote; Ethiopian wolf (Charles Darwin equivocally hypothesized this species gave rise to greyhounds); “the other jackals” (a polyphyletic group that despite great morphological affinity miss any further taxonomic integrity); and Dholes respectively, the latter having a phenotype most similar to the dingo while being of the genus “Cuon,” and as such not even part anymore of genus “Canis.” The wide-spread “doggish” appearance among non-wolf canids might indeed raise the question of ancient attempts to achieve interspecies hybrids.
The confusion extends to other domesticated animals: also the domestication of sheep, amongst the first species to be domesticated by man, is characterized by considerable genetic variability and a certain lack of wild matches. Kijas et al. (2012) state that despite a common origin of all domestic breeds of sheep, their ‘analysis revealed this domestication process must have involved a genetically broad sampling of wild stock.’ Moreover, like with dogs, the genetic variability of sheep has a weak global population structure and lack of association between genetic diversity and distance from the perceived domestication center, interpreted thus:
This suggests a highly heterogeneous predomestication population was recruited, and the genetic bottleneck which took place was not as severe during the development of sheep as for some other animal domesticates. It is also possible that cross-breeding with wild populations persisted following the initial domestication events to generate the diversity observed. (Kijas et al. (2012)
However, a purported single origin certainly contradictory with a broad genetic base. The possibility of cross-breeding with wild populations, as raised by Kijas’ team, virtually depends on the assumption that ‘high levels of gene flow have occurred between populations following domestication’. However, admixture appears contradictory with the observed homogeneity:
For SNP pairs separated by 10 kb or less, a high degree of conservation of LD phase was observed between all breeds […] Given that LD at short haplotype lengths reflects population history many generations ago, this also supports a common ancestral origin of all domestic breeds of sheep. The result is in contrast to cattle, where two distinct groups emerge from a similar analysis, even at haplotype lengths of 0–10 kb, reflecting the Bos taurus taurus and Bos taurus indicus sub-species and their separate domestication events (Kijas et al., 2012)
Instead, Pedrosa’s team rather adhered to the view that ‘introgression is generally ruled out as the cause of clearly differentiated maternal lineages in livestock, since introgression via females seems quite improbable.’ Indeed, just like with dogs, hybridization can’t be easily assumed just because of the theoretic possibility offered by closely related species or sub-species in the wild.
The center of sheep domestication currently encompasses the natural environment of at least three subspecies of Ovis, whose genetic boundaries are sometimes difficult to conceive: Argali (O. Ammon), Urial (O. Vignei), and Mouflon (O. Orientalis), the most western variety. Mitochondrial DNA of sheep doesn’t show any close relationship with argali or urial sequences (Pedrosa, 2006), what is remarkable since hybridization between the various subspecies has been observed in the wild. However, if hybrid admixture was limited to the domestication event this would almost insinuate a domestication purpose! For sheep cross-breeding may have been recognized from the start as predominantly detrimental to their specific qualities in the process of domestication, what should be a compelling argument to discard the persistence of cross-breeding with wild populations as a valid explanation for the observed genetic diversity of the domesticated form. On the contrary, the existence of widely spread and divergent mtDNA lineages without any current match in wild populations rather seems to indicate ancient hybridization events that preceded the fixation of domesticated sheep:
Time since divergence of types B and A estimated from the Cyt b gene [for sheep lineages] (around 150 000 to 170 000 years ago) agrees with the values obtained from Cyt b for goat lineages by Luikart et al. (2001; around 200 000 years ago) as well as those obtained for cattle (Bradley et al. 1996). Lineage C proved to have diverged earlier (between 450 000 and 700 000 years ago). (Pedrosa, 2006)
Another positive for early hybridization “by design” involves the domestication of chicken, that may descend from up to four Gallus species and their subspecies:
Four species of genus Gallus inhabit Southeast Asia: […] Red junglefowl has a strong sexual dimorphism with males having red fleshy wattles, and it is most widely distributed over the area. La Fayette’s junglefowl morphologically resembles red junglefowl, but it inhabits only Sri Lanka. Gray junglefowl has body plumage on a gray background color and is distributed from southwest to central India. Morphologically distinct green junglefowl is limited to Java and its immediate vicinity, Bali and Lombok [corr.: the Lesser Sunda Islands]. It has been debated whether any single species of the four, especially red junglefowl, predominantly contributed to the genome of domestic chickens (a single-origin hypothesis) or whether multiple species of the four made a substantial genetic contribution to domestic chickens (a multiple-origin hypothesis). (Sawai et al., 2010)
Actually, the Green junglefowl has a lot more eastern penetration into the Lesser Sunda islands, what somewhat invalidates Sawai’s ‘Javan’ origin hypothesis for this species. However, its unique isolation resulted in a genetically divergent species whose ability to hybridize with chicken is highly restricted.
Evidence abounds that Red junglefowl is the prime progenitor of domesticated chicken:
Haplogroup E is predominant among Indian, Middle Eastern, and European chickens and is an indication that the roots of European chickens lie within the Indian subcontinent. (Gongora, 2008b)
However, not unlike dogs, the calculated mtDNA divergence date with this purported progenitor by far exceeds archeological evidence of domestication:
According to archeological findings, the divergence time of domestic chickens from junglefowls is estimated to be on the order of 10,000 years. The MCMC method reveals, however, that the extent of nucleotide divergence after the split of red junglefowl from the chicken ancestor is […] 58,000 +/- 16,000 years ago. This dating is nearly six times older than what the archeological remains suggest. (Sawai et al., 2010)
Like with dogs it is tempting to attribute this divergence discrepancy to an ‘aberrant wild subspecies’ as the progenitor of the domestic form. Despite the historic distribution of Red junglefowl is limited to Southeast Asia in the range between the Indus valley, Yunnan (southern China) and Lombok, the oldest undisputed domestic chicken remains, dated 5400 BC, have been recovered from archaeological sites in northern China. Though chicken near that age (5000 BC) were also attested in cultural contexts in the Ganges region of India, this has led to debate about whether the natural range of Red Junglefowl reached much further north in prehistory. However, unlike dogs, this genetic diversity doesn’t extend to mtDNA, despite the lack of a major mismatch between the mtDNA haplotypes of domesticated and wild subspecies.
Only 56 out of all 206 Red junglefowl mtDNA sequences, recovered from the control region (CR) and available to a 2013 study of Miao’s team, were not found in domestic chicken. Of these, only seven from Yunnan and the island of Hainan (China) could be accommodated as new haplogroups W-Z within the monophyletic tree of domestic chicken, thus probably confirming rather Chinese roots of domesticated chicken (in the vein of Liu and Oka). An additional unclassified 21 haplotypes from Vietnam, Sumatra and Haryana (Northern India) appear not too distantly related, while 28 other haplotypes from Haryana and Indonesia could only be classified as ‘divergent’:
As for the red junglefowl, 76.2% (157/206) of the CR haplotypes were assigned to haplogroups in this genealogy (Figure 4; see Supplementary dataset 1). Apart from the common haplogroups A–G, the wild fowl harbored haplogroups W–Z, which were not detected in domestic chickens. Of the remaining sequences (49/206) not classified in the genealogy, 28 haplotypes from India and Indonesia (for example, ‘outgroups’ in Figure 1) had too many variants to be assigned; variation included many transversions (see Supplementary dataset 1). This suggested that they were remotely related to the other chickens (Miao et al., 2013)
This sequence divergence is such that natural hybridization with grey and green junglefowl respectively has been the suggested source of these divergent mtDNA lineages, though Miao’s team justly stresses this hypothesis requires a more comprehensive survey.
In Bangladesh the ‘phylogenetic tree showed low genetic distance and close relationship within and between the chicken populations of Bangladesh, which were closely related with G. g. murghi of Indian origin’ while a minority of domestic chicken and one red junglefowl were related with G. g. bankiva and G. g. gallus, ‘implying the origin of gene flow to Bangladesh’ (Islam & Nishibori, 2012).
This shocking lack of evidence for a truly unadmixed wild ancestral population has severe repercussions on the origin question of domestic chicken. The first three Gallus g. Bankiva accession numbers ever investigated (AB009430, AB009431 and AP007718) were found sufficiently different to originally exclude the Javanese wild subspecies from the group of domesticated chicken ancestors altogether:
On the whole, the analyzed data fit into two main clades […]: one formed by the continental red jungle fowl subspecies and all domestic chicken samples, which we named the continental clade, and another exclusively constituted by G. g. bankiva samples from Java that we named the island clade. (Liu et al., 2006a)
However, Oka et al. (2007) modified the interpretation of AP007718, and attributed a subsequent bankiva sample (AP003323) to one of the known chicken haplogroups, thus bringing also the Gallus g. bankiva subspecies into the fold of ancestral chicken:
G. g. bankiva [~Javanese red junglefowl] sequence ([AP007718, corr.]), which is distantly related to domestic chickens, was located on the outside of the Type C clade. However, another G. g. bankiva sequence ([AP003323, corr.]) and two G. g. gallus sequences [AP003322 and AB007725] were included in Type C. One G. g. spadiceus [~Burmese red junglefowl] sequence (AB007721) was close to the domestic chicken and very similar to Type A. The other G. g. spadiceus sequence (AP003323) was included in Type E. (Oka et al. – 2007).
Despite a genetic distance of no less than 30 mutation steps (Figure 3, Oka et al. 2007), AP007718 is slightly closer related with haplogroup C as defined by Oka’s team. It was not too farfetched to attribute an ancestral status to this chicken haplogroup:
[…] it is suggested that Types D, F and G diverged from Type C, and that Type E diverged from Type B.
[…] we suggest that Type F was derived from a group of Indonesian native chickens
It is suggested that Types A and B chickens (i.e. chickens of Chinese and Korean origin) were derived from Type C (i.e. they are of Southeast Asian origin) – Oka et al. (2007)
Such an all-encompassing phylogenetic system for chicken mtDNA, that includes ~75% of all wild specimen, seems irreconcilable with the extant picture of low genetic distances for chicken mtDNA:
It should be noticed that all the haplotypes that shared by or restricted to the red jungle fowls in clades A, B, E, and F diverged from the potential root in each clade by no more than 4-mutation distance, which was within the mutation distance observed between the domestic chicken and the potential wild progenitor G. g. gallus (Liu et al., 2006a)
Apparently, for all chicken and continental Red junglefowl together, the calculated mtDNA age is far less than nuclear DNA would suggest. However, since the monophyletic tree of chicken mtDNA can’t convincingly reflect the evolutionary history of the Gallus parent species, this should imply that current Red junglefowl populations simply lost their original mtDNA diversity quite recently due to introgression from feral chicken.
Better survival of ancestral mtDNA haplogroups in sheep and dogs could be due to evolutionary strain elsewhere that applied exclusively on their wild progenitors after the domestication events. Currently, gene flow between wolves and dogs, or between sheep and their wild progenitors, virtually doesn’t happen in the wild. Wolves would rather eat than mate a dog, and sheep don’t even survive to meet a wild partner in its habitat. Red junglefowl didn’t experience or acquire such an efficient natural barrier against gene flow from feral domesticated forms. Up to modern times this situation still didn’t affect the competitiveness of Red junglefowl populations in the wild. In general, in a natural habitat admixture of wild animals with their domesticated kin tends to compromise the competiveness of their feral offspring. Genetically there is few to gain and much to lose from domestic chicken that e.g. lost their ability to fly; feral sheep don’t even exist and only wolves evolved to abstain from cross-breeding with feral domesticated kin – even without strict biological barriers. However, since modern wolves predominantly attest another mtDNA clade, it is more than likely that evolutionary adaptation included less vulnerability to gene flow and ultimately led to the extinction of dog-introgressed wolves that were purportedly available in the Pleistocene (Germonpré et al., 2012a). The current vulnerability of Red junglefowl is unknown, but all indicates gene flow and introgression from domesticated chicken is rampant. The success of Green junglefowl west of the Wallace biogeographic division where they share the habitat with Javan Red junglefowl, while the latter hardly penetrated further east than Lombok (except for some unreliable observations). Wallace’s Line is the western border of a transitional region between Asiatic and Australian floras and faunas, where organisms show a high degree of isolation and endemism. Here, the Lesser Sunda Islands chain was definitely accessible from the west in the Pleistocene, what is attested even by the elephant that made an appearance up to Timor. West of this line, Java never ceased to be part of the Asian ecozone and hence was never truly isolated to justify Green junglefowls divergence from Red junglefowl populations. Instead, current cohabitation in Java together with the local subspecies of Red junglefowl could be indicative of future replacement tendencies for Gallus that mimic the gray wolf divergence from dogs towards the modern Holarctic wolf.
However, chicken introgression could even be indicated for the vastly divergent Green junglefowl, on the basis of nuclear DNA, attesting that ‘shared haplotypes are evenly distributed over all samples of red junglefowl, whereas this is not the case for those of green junglefowl’ (Sawai et al., 2010).
Purportedly introgressed chicken haplotypes in the green junglefowl were found to significantly increase Green junglefowl nucleotide diversity. This result is remarkable, since green junglefowl allegedly diverged 3.6 million years ago from the common ancestor of chicken and red junglefowl (Sawai et al., 2010), and first generation female hybrids were found notably infertile. Introgression of mtDNA would thus be close to impossible here. To break natural barriers in the forests, introgression of nuclear DNA may only be achieved for next generation hybrid offspring, that indeed must be readily available on a local basis in feral chicken populations. Indeed, Green junglefowl hybrids are still around being in high demand and part of an old Indonesian tradition.
Meanwhile, ‘attempts to investigate domestication and dispersal using mtDNA data from modern chickens have been confused by the tangled phylogenies which reflect millennia of overlapping dispersals and over a century of interbreeding for both commercial lines and show breeds’ (Storey et al,. 2012). Hence, only nuclear DNA remains to offer more insight into the origin of genetic diversity in domesticated chicken, while mtDNA diversity may only supply information on the world-wide history of the domestication process itself. Interestingly, ‘regional distribution of the clades was observed, which indicates some geographic structuring in chicken populations’ (Liu et al., 2006a). Structure, that thus should tell us the tale of domestic chicken from a single source rather than multiple domestication events. Also this feature is structurally different from domesticated animals like dogs and sheep, where high mtDNA diversity was preserved. This may be explained by a quite distinguished purpose of chicken domestication, as ‘it has also been proposed based on comparative morphology, historical depictions, and genetic relatedness that egg type chickens are the most ancient breed […] Therefore, the phylogeographic assumption that females have greater geographic inertia may be violated in the study of chickens by the widespread use of eggs’ (Storey et al., 2012). No other domestic vertebra experienced selective processes that applied so exclusively on females, what inevitably contributed to the severe reduction of mtDNA diversity as being observed.
The original proposal that haplogroup C ultimately derives from China, where prehistoric Red junglefowl populations could merely be assumed, might be as good as any, and actually reminds to the proposal of an ‘aberrant lineage of wolf’ for the ancestor of domestic dogs:
Although Japanese chickens displayed the highest nucleotide diversity […] for the clade C (Fig. 2D), the absence of red jungle fowl samples in clade C favors that this clade originated from South China. A recent domestication of clade D or gene Xow from domestic into the wild red jungle fowl population are two possible explanations for the fact that clade D mainly contained of red jungle fowl and gamecocks. These distinct patterns combined with archaeological records as well as with the geographic distribution of G. gallus are consistent with clades C and D originating relatively recently, perhaps in South and Southwest China and/or surrounding areas (Liu et al., 2006a)
Deviation from a simple star-like distribution pattern of this geographically restricted grouping may be due to the success of more recent breeds that instead propagated the few international mtDNA lineages, such as Haplogroup E. Due to the selective forces of human intervention such perpetual replacements completely obfuscated, for instance, the debate on early Polynesian arrivals in the Americas, where allegedly pre-Columbian chicken and their potential offspring happen to group predominantly with international mtDNA rather than the Polynesian haplogroups of more limited distribution:
The single ancient sequences reported from Tonga Mele Havea and Ha’ateiho, Samoa Fatuma Futi, Hawaii Kualoa, Niue Paluki, one sequence from Easter Island Anakena (early settlement phase Cal AD 1270–1400), and the putatively pre-Columbian (Cal AD 1304–1424) Chilean El Arenal-1 sequence belong to haplotypes 8 and 5 within haplogroup E (Fig. 1). Haplotype 8 equates to the E1 haplotype reported in Liu et al. [2006a], which is ubiquitous worldwide and 100% identical to haplotypes from European Barred Plymouth Rock, White Plymouth Rock, White Leghorn, and New Hampshire as well as native chicken sequences from [India, corr.], Central Asia, and China. In contrast, five of the other contemporaneous archaeological chicken sequences from Easter Island cluster with haplotypes 145 (n = 4) and 148 (n = 1), which are part of an uncommon group comprising mostly Indonesian chickens within haplogroup C (Fig. 1). Ancient Easter Island haplotype 145 is identical to one sequence of Red Junglefowl from the Philippines. Within other modern chickens, the closest related sequences have been recorded from Lombok and Java in Indonesia and the Philippines. Given their unique phylogenetic position and their pre-European contact dates, haplotypes 145 and 148 presumably represent a record of early Polynesian chicken transport, potentially overwritten subsequently in the western Pacific. The noticeably less star-like pattern of haplogroup C, centered on the less frequent haplotypes 91 and 95, is likely to be an artifact of incomplete sampling or a different population history. (Gongora et al., 2008a)
Nowadays, the haplogroup C denomination is confined to a much smaller group of continental chicken, while Oka’s and Gongora’s haplogroup C definition has been currently passed over to Liu’s (ancestral) haplogroup D, once tentatively proposed by Liu’s team for a loose group of cock-fighting chicken and Red junglefowl though currently thought to be closely associated with an Austronesian legacy in Indonesia, Madagascar and (hence?) Africa, and Polynesia:
Among the three Guamanian haplogroup D samples, three lineages were observed. One matched contemporary samples from the Philippines, Indonesia, and Japan. The other lineages were unique, but were closely related to contemporary samples from Indonesia, Japan, the Philippines, and China as well as India, Sri Lanka, Thailand, and Madagascar. Among the 40 haplogroup D samples from Vanuatu, seven lineages were observed. Four were exact matches to contemporary samples from Indonesia, Japan, the Philippines, and Southern China, as well as prehistoric Easter Island samples. The major Vanuatu lineage (n = 17), also found in a Red Jungle Fowl from the Philippines, was distributed across all four islands sampled (Dancause et al., 2011)
Conceivably the oldest ‘chicken’ haplogroup, the Indonesian hotspot of mtDNA haplogroup D distribution opens up an entirely new perspective on chicken domestication. Near the Island Southeast Asian origin of Polynesian Haplogroup D chicken, Bekisars are still the first generation (F1) hybrid offspring with Green junglefowl. Bekisars, whose colorful roosters have a glossy blackish green plumage, were the mascots of the original inhabitants of the Sunda Islands. Indeed, fresh chicken hybrids of the green junglefowl are still in demand. Seafaring cultures in the neighborhood allegedly took advantage of their unique crowing sound, loud enough to be heard for two miles over the sea, and consequently hybrids were part of their traveling gear. Because of the location, results of hybridization experiments on local chicken may be most noticeable in chicken populations whose history was especially related to mtDNA haplogroup D.
Currently, the role attributed to green junglefowl hybrids in the earliest Polynesian explorations is a popular theme. This has everything to do with the radiocarbon dating of apparently pre-Columbian chicken of El Arenal, Chile, that allegedly lived sometime between ad. 1321 and 1407 (Storey et al., 2007), and its purportedly ancestral relation with a peculiar local breed:
It has been suggested that the unique type of chicken known as the Araucana, which has no tail and lays blue eggs, is descended from pre-European stock bred by the Mapuche (formerly called Araucanos) people of Southern Chile (Storey et al., 2007)
An Oceanic origin, still contested, has been partly substantiated:
All ancient West Polynesian samples, early samples from Anakena, Easter Island and Kualoa, Hawaii, and the El Arenal sample share a single unique point mutation (a T to C transition) at site 214. One of the modern Araucana feather samples also shares this unique mutation. Three other SNPs (all transitions) at sites 278, 303, and 339 are shared by these West Polynesian, early Anakena and Hawaii, and the Chilean bone samples and sequences reported from modern chickens in Southeast Asia, specifically samples from the Yunnan region of China and Vietnam (Storey et al., 2007)
Indeed, in 1532, Spanish conquistador Francisco Pizarro recorded the presence of chickens in Peru, where the Inca used them in religious ceremonies. ‘That suggests chickens had already been there for a while’ (Storey in Archeology, Volume 61 Number 1, 2008).
However, all (meanwhile) three purported Pre-Columbian chicken remains turned out to have their mtDNA in the most common haplogroup E. But, in accordance with ‘modern Chilean sequences [that] cluster closely with haplotypes predominantly distributed among European, Indian subcontinental, and Southeast Asian chickens, consistent with a European genetic origin’ (Gongora et al., 2008a), this common haplotype (mtDNA E) was also present in ancient Polynesian chicken. Prehistoric chicken remains from Easter Island stand out for the dominant presence of mtDNA D, however, so ‘the two lineages may have converged before they were dispersed, as a polymorphic population, to Hawaii and Easter Island. Both haplogroups appear in early period archaeological sites in East Polynesia’ (Storey et al., 2012).
The search for haplogroup D in the Americas was not all in vain:
Haplogroup D has not yet been detected in any ancient chicken bone samples from Europe or from Thailand. Thus far it has only been identified in ancient Polynesian and Micronesian chicken remains as well as a single Peruvian sample. Haplogroup D has not yet been detected in any ancient chicken bone samples from Europe or from Thailand. Thus far it has only been identified in ancient Polynesian and Micronesian chicken remains as well as a single Peruvian sample. The available sample size for this study is too small to be representative
The early date of the Peruvian assemblage from which this haplogroup D sample was recovered raises the possibility that it could represent a descendant of a chicken haplogroup introduced from Polynesia. The fact that the sequence from the chicken bone from the Torata Alta site in Peru is identical to one from Fais in Micronesia also may support a possible Pacific connection. (Storey et al., 2012)
The Green Junglefowl is the only species of junglefowl that produces grey tinted eggs, indeed a possible forerunner of the bluish Araucana eggs. Instead, Red junglefowl miniature eggs vary between pure white and a deep creamy-buff. Ancient breeds often have tinted eggs (Silkies, Sussex) that approximate Gray junglefowl, whose eggs vary from very pale cream to rich warm buff, though the latter are often freckled and even spotted. The brownish red spotted eggs of the Singhalese Gallus Lafayette, or Ceylon junglefowl, appear profoundly out of range though reminds the French marran breed. Funny detail is that brown eggs, commonly associated with ‘natural’, are actually without equivalent in natural gallus species.
Nishibori et al. (2005) elaborated molecular evidence of Gallus hybridization except for the Green junglefowl. Plain hybridization has been well substantiated with the Grey junglefowl (India):
Our data imply that carotenoids are taken up from the circulation in both genotypes but are degraded by BCDO2 in skin from animals carrying the white skin allele (W*W). Surprisingly, our results demonstrate that yellow skin does not originate from the red junglefowl (Gallus gallus), the presumed sole wild ancestor of the domestic chicken, but most likely from the closely related grey junglefowl (Gallus sonneratii). This is the first conclusive evidence for a hybrid origin of the domestic chicken, and it has important implications for our views of the domestication process (Eriksson et al. – 2008)
Thus, the most common chicken being yellow-legged, modern chicken should be descendants of hybrids that originate predominantly from India, home of both the Grey junglefowl and an Indian subspecies of Red junglefowl: ‘Recently, […] molecular data also show that Indian Red junglefowl (G. g. murghi) also contributes to the domestication’ (Sawai et al., 2010).
The wide distribution of this hybrid form may be tempting to assume also a hybrid origin of domestic chicken. However, the ancient Sussex breed, though their large lightly tinted tan eggs might still suggest Gray junglefowl influences, also has white legs and skin in every variety. Brahmas, one of the largest breeds of chicken having fully feathered, white legs, derive from Shanghai, China. Such breeds far outside the range of red junglefowl tend to corroborate to the evidence that on a world-wide scale yellow legged hybrids were not part of the oldest introduction of domesticated chicken, or that successive waves of hybrids introduced different expressions of a hybrid nature.
How come early hybridization of domestic stock still receives so little attention? Faunal interference before the successful Neolithic domestication events may be deduced from biogeographic evidence. Some animals are thought to have translocated by humans already in prehistoric times. For instance, the natural range of Asian wild dogs (or Dholes) may reached much further than nowadays, still genetic evidence also insinuates a human role in their current distribution:
the grouping of Sumatran and Javan haplotypes with those from India (south of the Ganges) and Myanmar, as opposed to those from Malaysia and Thailand, should be noted.
Further studies are required to clarify these results but in the absence of alternative explanations, these results may be suggestive of human translocation of dholes from one of these regions into Sumatra and/or Java. But since there is no documented evidence for such translocation(s), and given that dholes have been long considered vermin (and not hunted for sport, cf. red fox introductions to Australasia), this hypothesis must remain highly speculative (Iyengard et al., 2005)
At least the genetic results of Brown’s team (2011) cited above into the direction of Dhole or other admixture in domestic dogs was hastily swept under the carpet, when they found the novel Y-chromosome haplotype 12 that besides being recognized as ‘ancestral’ was also shared by Dhole, black-backed jackal, and a wolf from China. The Dhole (Cuon alpinus), also called the Asiatic wild dog, red dog, red wolf, or whistling dog, was widespread across North America, Europe and Asia during the Pleistocene. The species’ range, however, became restricted to Asia after the late Pleistocene mass extinctions c. 12 000–18 000 bp, when it became extinct across North America and Europe, along with several other large species such as mammoths and dire wolves. Like the African wild dog, they have the same number of 78 chromosomes like dogs, as do wolves, coyote and the golden jackal. However, hybridization with any of them has been attempted nor recorded. In appearance Dholes are not unlike shepherd dogs, that purportedly resemble Pleistocene dogs most:
In the PCA and DFA plots, the Palaeolithic dogs are situated near the Central Asian Shepherd dog, which was assigned in the DFA to the prehistoric dog group. This suggests that their skull shape resembles that of the latter breed which was originally used as a flock guardian and as a protector against predators such as bears, striped hyenas and wolves (Labunsky, 1994). The latter property in a dog could also have been useful for Palaeolithic people (Germonpré et al., 2009)
More features of modern dogs are intermediate between Dholes and wolves. Female dogs have 8 to 12 teats, while wolves have only 4 to 8. Female Dholes are on the other end of the spectrum having 12 to 14 teats rather than 10.
Dholes can be recognized in the fossil record by their reduced number of teeth: 40 instead of 42, due a missing molar (M3) in the lower jaw: 184.108.40.206/220.127.116.11, against dogs: 18.104.22.168/22.214.171.124. However, the number of teeth for dogs is known to vary, and in his thesis Datema (2011) uses two reference cranials of dog (Canis lupus familiaris) with exactly this Dhole-like dental formula to investigate whether some dog-like remains recently fished up from the bottom of the North Sea (Southern Bight) really belonged to Pleistocene dogs or rather to “Cuon alpinus”. Was the Dhole indeed a member of the Mammoth Steppe Fauna?
One of Datema’s fossil Canidae specimen, labeled NMR90, was suspected to belong to Cuon alpinus. In this sample M3 is missing, like in Dholes. However, at least with respect to mandible lengths a-b and c-b this specimen was rather intermediate between wolf and Dhole and most similar to dogs:
NMR90, a recently found posterior half of an extremely small jaw (with P4 and M1) of estimated Late Pleistocene age) is suspected, due to its apparent small size and dental formula, that it most probably belongs to neither wolf nor dog (nor fox, which is considerably smaller), but to Cuon alpinus
The specimen range of NMR90 falls entirely outside [below] the population range of C. l. lupus and also entirely outside [above] the population range of Cuon alpinus, but well within the range of C. l. familiaris for a-b (table 4.2 and fig. 4.2). For c-b the specimen range of NMR90 falls entirely outside [above] the population range of Cuon alpinus, entirely within the C. l. familiaris range and overlaps slightly (1.36 mm) with the C. l. lupus population range, although most of the specimen range of NMR90 (2.75 mm) falls outside [below] the C. l. lupus range. (Datema, 2011)
Even though the sample was ultimately classified as a Dutch Dhole from the Southern Bight, this comparative study reveals how tentative such results may be. On the contrary, it has all appearance that Pleistocene dogs represented much more than an aberrant species of wolf. They were close relatives to the ice age ecomorph and probably all other Holarctic Pleistocene wolves that couldn’t resist the ‘human kiss of death’, and ‘hence’ became utterly outcompeted by an aggressive new type of wolf from the south that could. And Pleistocene dogs, embraced by the eerie love of humans, might have diverged even further for being purposely hybridized from the start!
Indeed, hybridization is one of the first recorded obsessions of modern humans. The famous Aurignacian “Venus and the Sorcerer” rock painting in Chauvet cave, Ardeche, already depicts the cross-species mating of an anthropomorphic steppe bison and a female lion, sexually united by a black female pubic triangle. Bull-related fertility and the procreation of hybrid monsters remained a recurrent theme in mythology up to the Greek Minotaur and the Frankish Quinotaur. There could be more. When the Sumerian Enkidu killed the Bull of Heaven, and hence was doomed to die, his friend Gilgamesh was in great grief and swore: “I will wander through the wilderness in the skin of a lion”. We might think this was a strange way to please the gods, though Chauvet Cave would suggest a rich mythology preceding Gilgamesh being based on this bond between species. We can’t dismiss this genuine fascination of early modern humans as purely theoretical. Actually, even the huge body of mythological themes related to shapeshifting has all appearance to be closely connected to an early human fascination for cross-breeding. We simply can’t exclude the possibility that hybridization experiments were already attempted long before the Neolithic.
- Aggarwal et al. – Mitochondrial DNA coding region sequences support the phylogenetic distinction of two Indian wolf species, 2007, link
- Albert et al. – A Comparison of Brain Gene Expression Levels in Domesticated and Wild Animals, 2012, link
- Avian Information System of India – On Ceylon junglefowl, Grey junglefowl and Red junglefowl.
- Axelsson et al. – The genomic signature of dog domestication reveals adaptation to a starch-rich diet, 2013, link
- Baryshnikov et al. – Finding of the Late Pleistocene carnivores in Taimyr Peninsula (Russia, Siberia) with paleoecological context, 2009, link
- Bay-Petersen – Competition for resources : the role of pig and dog in the Polynesian agricultural economy, 1983, link
- Brown et al. – Phylogenetic Distinctiveness of Middle Eastern and Southeast Asian Village Dog Y Chromosomes Illuminates Dog Origins, 2011, link
- Cahill et al. – Genomic evidence for island population conversion resolves conflicting theories of polar bear evolution, 2013, link
- Camerini – Evolution, Biogeography, and Maps: An Early History of Wallace’s Line, 1993, link
- Cherin et al. – Re-Defining Canis etruscus (Canidae, Mammalia): A New Look into the Evolutionary History of Early Pleistocene Dogs Resulting from the Outstanding Fossil Record from Pantalla (Italy), 2013, link
- Crockford & Kuzmin – Comments on Germonpré et al. […] “Fossil dogs and wolves from Palaeolithic sites in Belgium, the Ukraine and Russia: osteometry, ancient DNA and stable isotopes”, and […] “Palaeolithic dog skulls at the Gravettian Predmostí site, the Czech Republic”, 2012, link
- Dancause et al. – Characterizing Genetic Diversity of Contemporary Pacific Chickens Using Mitochondrial DNA Analyses, 2011, link
- Datema – Was Cuon alpinus a member of the Mammoth Steppe Fauna? Comparative morphological and osteometrical study on recent and fossil Canidae hemimandibles, 2011, link
- Ding et al. – Origins of domestic dog in Southern East Asia is supported by analysis of Y-chromosome DNA, 2011, link
- Dhole Fact File, link
- Druzhkova et al. – Ancient DNA Analysis Affirms the Canid from Altai as a Primitive Dog, 2013, link
- Eriksson et al. – Identification of the yellow skin gene reveals a hybrid origin of the domestic chicken, 2008, link
- Garcia – Ichnologie générale de la grotte Chauvet, 2005, link
- Garrido & Arribas – Canis accitanus nov. sp., a new small dog (Canidae, Carnivora, Mammalia) from the Fonelas P-1 Plio-Pleistocene site (Guadix basin, Granada, Spain), 2008, link
- Gaubert et al. – Reviving the African Wolf Canis lupus lupaster in North and West Africa: A Mitochondrial Lineage Ranging More than 6,000 km Wide, 2012, link
- Germonpré et al. – Fossil dogs and wolves from Palaeolithic sites in Belgium, the Ukraine and Russia: osteometry, ancient DNA and stable isotopes, 2009, link
- Germonpré et al. – Palaeolithic dog skulls at the Gravettian Predmostí site, the Czech Republic, 2012a, link
- Germonpré et al. – Palaeolithic dogs and the early domestication of the wolf: a reply to the comments of Crockford and Kuzmin, 2012b, link
- Gongora et al. – Indo-European and Asian origins for Chilean and Pacific chickens revealed by mtDNA, 2008a, link
- Gongora et al. – Reply to Storey et al.: More DNA and dating studies needed for ancient El Arenal-1 chickens, 2008b, link
- Hammer et al. – Genetic evidence for archaic admixture in Africa, 2011, link
- Hofreiter – Pleistocene extinctions: haunting the survivors, 2007, link
- Hofreiter & Barnes – Diversity lost: are all Holarctic large mammal species just relict populations? 2010, link
- Hooijer – Indo-Australian Insular Elephants, 1967, link
- Iyengard et al – Phylogeography, genetic structure, and diversity in the dhole (Cuon alpinus), 2005, try here
- Islam & Nishibori – Phylogenetic Analysis of Native Chicken from Bangladesh and Neighboring Asian Countries Based on Complete Sequence of Mitochondrial DNA D-loop Region, 2012, link
- Kays et al. – Rapid adaptive evolution of northeastern coyotes via hybridization with wolves, 2010, link
- Kijas et al. – Genome-Wide Analysis of the World’s Sheep Breeds Reveals High Levels of Historic Mixture and Strong Recent Selection, 2012, link
- Larson et al – Rethinking dog domestication by integrating genetics, archeology, and biogeography, 2012, link
- Leonard et al. – Ancient DNA evidence for Old World origin of New World dogs, 2002, link
- Leonard et al. – Megafaunal Extinctions and the Disappearance of a Specialized Wolf Ecomorph, 2007, link
- Lindblad-Toh et al. – Genome sequence, comparative analysis and haplotype, structure of the domestic dog, 2005, link
- Louys et al. – Characteristics of Pleistocene megafauna extinctions in Southeast Asia, 2006, link
- Liu et al. – Multiple maternal origins of chickens: out of the Asian jungles, 2006a, link
- Liu et al. – Genetic relationship of Chinese and Japanese gamecocks revealed by mtDNA sequence variation, 2006b, link
- Meachen & Samuels – Evolution in coyotes (Canis latrans) in response to the megafaunal extinctions, 2012, link
- Miao et al. – Chicken domestication: an updated perspective based on mitochondrial genomes, 2013, link, or try here, supplemental information here
- Moss & Wilson – Biogeographic implications of the Tertiary palaeogeographic evolution of Sulawesi and Borneo, 1998, link
- Napierala – A ‘new’ Palaeolithic dog from central Europe, 2010, link
- Nishibori et al. – Molecular evidence for hybridization of species in the genus Gallus except for Gallus varius, 2005, link
- Oka et al. – Analysis of mtDNA sequences shows Japanese native chickens have multiple origins, 2007, link
- Ovodov et al. – A 33,000-year-old incipient dog from the Altai Mountains of Siberia: evidence of the earliest domestication disrupted by the Last Glacial Maximum, 2011, link
- Pang et al. (2009) – mtDNA data indicate a single origin for dogs south of Yangtze River, less than 16,300 years ago, from numerous wolves, 2009, link
- Pedrosa et al. – Evidence of three maternal lineages in near eastern sheep supporting multiple domestication events, 2006, link
- Pilot et al. – Phylogeographic history of grey wolves in Europe, 2010, link
- Reece et al. – Campbell Biology, ninth edition, 2011, ISBN 9780321558237
- Rueness et al. – The Cryptic African Wolf: Canis aureus lupaster Is Not a Golden Jackal and Is Not Endemic to Egypt, 2011, link
- Sacks et al. – Y chromosome analysis of dingoes and Southeast Asian village dogs suggests a Neolithic continental expansion from Southeast Asia followed by multiple Austronesian dispersals, 2013, link
- Savolainen et al. – Genetic evidence for an East Asian origin of domestic dogs, 2002, link
- Sawai et al. – The Origin and Genetic Variation of Domestic Chickens with Special Reference to Junglefowls Gallus g. gallus and G. varius, 2010, link
- Serpell – The Domestic Dog: Its Evolution, Behaviour and Interactions with People, 1995, link
- Sharma et al. – Ancient wolf lineages in India, 2004, link
- Storey et al. – Radiocarbon and DNA evidence for a pre-Columbian introduction of Polynesian chickens to Chile, 2007, link
- Storey et al. – Pre-Columbian chickens, dates, isotopes, and mtDNA (Reply to Gongora et al., 2008a), 2008, link
- Storey et al. – Mitochondrial DNA from 3000-year old chickens at the Teouma Site, Vanuatu, 2010, link
- Storey et al. – Investigating the Global Dispersal of Chickens in Prehistory Using Ancient Mitochondrial DNA Signatures, 2012, link
- Thalmann et al. – Inferences on dog domestication – genetic analysis of the most ancient dogs utilizing DNA capture arrays, SMBE 2012, extract
- vonHoldt et al. – Genome-wide SNP and haplotype analyses reveal a rich history underlying dog domestication, 2010, extract
- VonHoldt et al. – A genomewide perspective on the evolutionary history of enigmatic wolf-like canids, 2011, link
- Vilà et al. – Multiple and ancient origins of the domestic dog, 1997, link
- Vilà et al. – Phylogenetic Relationships, Evolution, and Genetic Diversity of the Domestic Dog, 1999, link
- Wayne et al. – Origin, genetic diversity, and genome structure of the domestic dog, 1999, link