by Martin Jones
Like the horse, the ancestral camelids crossed the Bering straits from their American homeland around 1 million years ago. The camelids that remained in America did not completely die out, and two species live on. These are the guanaco and the vicuna. As in the Old World, there are also two domestic camelids in the New World, the alpaca and the llama. What is more, the archaeological record is much more generous here.
In the arid desert that rises up above the Pacific coast, the same special conditions that have left us some of the world’s best conserved mummified humans have also preserved the animals on which they depended. El Yaral was one of the sites from which the bodies of sacrificial llamas and alpacas were unearthed. Natural desiccation had left their skin and wool intact. Sometimes, their skin hung around the bone like a loose overcoat, but the muscles and internal organs were also frequently conserved. These animals were sacrificed by a blow to the head 1,000 years ago, and buried beneath the floors of the houses at El Yaral. A millennium later, their wool has lost neither its colour nor its texture. It is preserved so well that it can be analysed today using the same techniques as are applied to modern llama and alpaca fibres. The ancient wool was fine, finer than most produced today. From studies of these ancient fleeces, we sense that the farmers who dwelt here 1,000 years ago herded animals from highly selected breeds, breeds that no longer survive. Here was evidence that history, in particular the repression by the European conquistadors, had taken its toll. Its impact on indigenous human communities had also led to a loss of genetic diversity among their domestic animals.
At the research division of the London Zoo, Helen Stanley and Miranda Kadwell embarked on an exploration of the DNA within these specimens. They speculated that perhaps the ancient DNA in these ancient llamas would reveal that genetic diversity. If they could get at the DNA in this well-preserved specimen, perhaps it might even be possible to recover the lost genes. Unfortunately, their only successful amplification of the ancient camelids was a sequence of fewer than 100 base-pairs from the 16S rRNA calendar. It was enough to show that the ancient llamas were closer to alpacas than to Old World camels, but the missing clock and stopwatch were really needed to take the study further. One of the points reinforced by the evidence from these ancient Peruvian animals was that tissue that seems visually intact might not be the best preserver of large molecules. In the right conditions, an ordinary bone may have far better preserved DNA than one on which the colour of an animal’s hair is still visible. Even in these wonderfully preserved specimens, this remarkable molecule from the past remained elusive.
Each of the above accounts is of humans and animals interacting in parts of the world outside the Fertile Crescent. In many ways, the key molecular studies for this ancient region are of the sheep and the goats that first drew Robert Braidwood to the Fertile Crescent, but their story has yet to be told. However, a third animal from the region has revealed the most detailed molecular story of all.
the mighty urus
In the late fifteenth century, one Johann Pruess compiled the following description of a rather terrifying wild beast:
Isidor says of the urus: urus are wild cattle so strong that they can lift trees as well as armed knights with their horns. They are called urus from the Greek word oros meaning mountain … Heylandus says … In the Hercynian forest of Germany the Urus is found. These animals are nearly as large as elephants: in appearance, colour and conformation they are like cattle. The force of their horns is great and their speed is great. They spare neither man nor animal. One catches them in pits and kills them.
(Johann Pruess, Hortus Sanitatus, 1495)
None of these mountainous beasts, now referred to as ‘aurochsen’, survives today. They live on only as descriptions, illustrations, fossil footprints across an ancient beach, and in archaeological deposits, their gigantic bones and particularly their massive horns reminding us of Pruess’s words.
Something of the feelings they inspired in prehistory can be imagined at an ancient site in the Konya Plain of southern Turkey. Discovered by James Mellaart in the 1950s, Catal Huyuk remains one of the earliest sites to show evidence of cattle husbandry. As Mellaart dug down into the vast thirty-two-acre tell he uncovered a crowded cluster of mud-brick houses. Within the sediments was a diversity of artefacts unparalleled in a settlement that was just a few centuries younger than the earliest permanent farming settlements in the region. Among the finds were the massive spreading horns of aurochsen that had clearly inspired the inhabitants with great awe. In one of the rooms that Mellaart suggested was a shrine, the space was dominated by models in clay of bulls’ heads around the walls, painted in striking colours and fitted with genuine horns.
Catal Huyuk is now being re-excavated by Ian Hodder, and a team at Trinity College, Dublin, is currently attempting to amplify the DNA from the cattle bones excavated from the site. At the time of writing, the search for these particular DNA molecules has not reached completion, but Dan Bradley’s team already have more than enough molecular evidence to question the single origin of domesticated cattle. Just as Richard Meadow had looked beyond the Fertile Crescent, further to the east and into the Baluchistani mountains, the Dublin team also looked eastward.
Going beyond Meadow’s site of Mehrgarh and into the Indus Valley itself, we find the impressive urban sites of the more recent Harappan civilization. Harappa, Mohenjo-Daro and the recently uncovered Dholavira were large towns of the third millennium BC, with streets, public places and impressive systems of water management. We still cannot decipher their writing, but from the images on their cylinder seals we can see how very important cattle had become in their lives. But these cattle are not quite the same as the ones found further to the north and west. Two particular features distinguished them from cattle found throughout Europe and parts of northern Asia. The Harappan illustrations displayed a prominent hump above the neck, and the pendulous flap, or dewlap, below the throat. The same features can be seen in living breeds, particularly in Africa, and in south Asia where they are known as zebu cattle. The difference between zebu cattle and their ‘taurine’ counterparts has been much debated by cattle breeders. The two forms are totally inter-fertile, and we have already seen in both plants and animals that even greater differences in physical appearance than this need not imply a great genetic distance. However, the molecular research conducted by Dan Bradley and his colleagues indicated that, in this case, an easily observed physical feature of the beast was a very good indicator of what was going on at the molecular level.
They worked with the mitochondrial stopwatch, the control region, and targeted a hypervariable segment of 375 base-pairs within it. In that stretch, thirteen different breeds of cattle displayed variation at sixty-three positions. When the resulting variations were built into a tree, it was a tree with a very deep bifurcation. On one major branch were the Indian humped cattle, and on the other all the remaining breeds. It was a clear division and an ancient one. After including bison in order to root the tree, they estimated that the divergence between humped Indian cattle and the rest was three-quarters of the divergence between cattle and bison. That split is assumed from fossil evidence to have happened over 1 million years ago. On the basis of modern DNA alone, there was a strong argument for separate domestications of cattle in south Asia and at some other location. The African cattle presented a bit of a puzzle.
Some African cattle have humps and others do not, but, with or without, they all fall on the ‘un-humped’ side of the mitochondrial tree. The distinction has a pattern both in space and in time. The humped breeds are a more southerly phenomenon, the northern breeds lacking the hump. Back in time, so far as we can tell, from historical records, this distinction was even more pronounced than it is today. In prehistory, things were different. Domestic cattle were in Africa at least from the seventh millennium BC, and at that stage all lacked the characteristic hump. Before the Sahara dried up, the cattle that grazed on its natural pastures were captured as images in the Tasili rock ar
t, images also free of humps. It is not until around 2,000 years ago that the distinctive vertebrae of a humped cow had found their way into the Neolithic site of Ngamuraik in Kenya. Around three centuries later, an Axumite figure of a humped cow is found in Ethiopia.
These times correspond to a period of considerable change for the southern regions of the African continent. Where all the known evidence of human activity had indicated hunting and gathering, pottery, iron-working and farming all spread southwards across regions now occupied by Bantu-speaking peoples. This expansion and spread of farmers/metalworkers is held to have transformed the subsequent history of southern Africa. While evidence for cattle remains scanty after those first finds of humped cattle bones, they may have had a key role to play in the southerly spread of later generations of pastoralists. Humped cattle have a metabolic advantage over their unhumped relatives. They can slow their metabolism more effectively, something that may be key to survival in a savannah environment. Indeed, the adaptive strength of the savannah grasses themselves is that they can drop metabolism to zero by drying up and turning to seed. Humped cattle and savannah grasses alike are well adapted to the fluctuating environments across which those early pastoralists travelled in the south. What the archaeological evidence indicates is that there were humpless cattle in the more northerly regions, and two groups of humped cattle–one in south Asia and one in southern Africa–with a separate but uncertain past. Bradley’s group built up a larger collection of African specimens but still came up with the same result. All African cattle, regardless of their appearance, with or without humps, fell on the other side of the tree from the Indian humped cattle. So where did the genes for the hump come from?
What the mitochondria tell us is that the maternal lines of African cattle all lead back to a humpless, taurine form. The hump must have been inherited from the male line alone, from zebu bulls. Indeed, that is the only way to account for the evidence. At some stage in African prehistory, certainly by 2,000 years ago, some Asian bulls had been brought across to this continent. They looked very different, but someone must have known that the special hardiness of the Asian cattle was something that the African cattle lacked. For a long time, those breeds had a fairly localized impact, but by the time the spread of pastoralism was transforming the southern continent, the hardy genes they carried were absolutely key. This much is conjecture, based on the absence of zebu mitochondrial DNA. The proof lies in the chromosomes only transmitted through the male line.
Bradley’s group went on to study the paternally inherited Y-chromo-some in European, Asian and African cattle. Their hypothesis was confirmed. In Europe the animals were consistently taurine, in south Asia they were consistently zebu, but in Africa the Y-chromosomes were a complex mix of taurine and zebu. What seems to have happened is that zebu cattle were indeed brought to central Africa some time before 2,000 years ago. It is difficult to say how many were brought, or whether it was bulls and cows together, or bulls alone. It is the bulls that left their mark by interbreeding with the local cows. Their hybrid offspring then travelled and mingled across central and southern Africa, with the Bantu-speaking peoples of the Chifunbaze culture, whose sites appear from the late first millennium BC and proliferate through the subsequent millennium. These cattle began as a small component in their economy, but eventually grew into a major feature of it. The appearance of the zebu cattle in Africa is complementary to the appearance in third and second millennium BC India of African grain crops such as sorghum, which in turn transformed the nature of Asian agriculture. It used to be thought that such intercontinental contacts ran through the great civilizations of the Indus Valley and the Near East. These cultures may not however have had much to do with it. The relevant bio-data does not follow the inland route through the Near East, but links sub-Saharan Africa with mainland India via a few points on the intervening coast. It must have been a less celebrated maritime community that took crops and livestock back and forth, a few exchanges transforming the human ecology of both continents. Moreover, the most recent genetic results from Dublin suggest that the geographical heart of the zebu haplotype is in southern India, rather than towards the Indus valley in the north.
By the time a Chifumbaze settlement just outside Fort Victoria had grown to the monumental, high-walled Great Zimbabwe after which its country was named, this trade with the east had become extensive. The site contains Near Eastern glass and pottery from as far away as Persia and China. But the most critical import from the east was not those conspicuous high status goods, but a few genes that had arrived a few centuries before the massive stone buildings were erected. A group of zebu cattle, perhaps exchanged for sacks of African grain, had transformed the great herds of cattle upon which that society depended and the human ecology of a subcontinent. Furthermore, we only know this from the evidence of the molecular tachometer encapsulated within their cells.
One final part of the story is the history of the taurine branch of the African cattle breeds. Their conventional archaeology is in no way inconsistent with a spread from the Fertile Crescent, down the Nile Valley and across land to north and west Africa. However, the growing understanding of D-loop data allowed Bradley’s group to go one stage further and look for the common ancestor of Asian, European and African cattle. What they found was that the three continents contained three haplotype clusters. As we have seen, the common ancestor of European and Asian cattle lived between 200,000 and 1 million years ago. They estimated that the equivalent common ancestor of European and African cattle lived 22,000-26,000 years ago–much closer related than taurine and zebu, but still too far back to accommodate a common single ancestor within the last 10,000 years.
By looking further into the patterning of the D-loop variations, it was possible to tease out the demographic history of these variations. Wild cows spread out from the refuges they had occupied during the coldest period of the most recent Quaternary cycle, into Europe and Africa, leading to genetically divergent populations of wild progenitors. The wild zebu had diverged during some earlier Quaternary cycle. In all three continents, cattle were independently domesticated. One group of domesticates spread westward across north Africa from 9,000-11,000 years ago. Another group spread across Europe 5,000-9,000 years ago, and a cross between African taurines and a population of imported zebu bulls spread across sub-Saharan Africa over the last 2,000 years.
The genetic precision of this story is impressive, but we must remember that the chronological precision is still no better than the molecular clock on which it is based. To reiterate: the basis of the clock used in this case is based on two assumptions. First, it is assumed that Bos (the cattle genus) and bison diverged 1 million years ago, and second, that the mutation within the sequence of 370 base-pairs studied is relatively uniform. It works out in fact at one base-pair mutation, on average, every 4,000 years. Even if both these assumptions are sound, the difference between African and European cattle is only two base-pair mutations away from a single domestication. Once the molecular clock is measuring differences of hundreds of thousands of years, then it is on firmer ground. Once the differences of interest reduce to a few thousand years, then what are needed are more precise dates from archaeology, and DNA sequences from ancient bones, both domesticated and wild.
The bones of domesticated cattle are widespread, and from time to time the bones of the ancestral wild aurochsen turn up, for example in Upper Palaeolithic cave sites. The Oxford group of ancient DNA researchers tracked down a number of these aurochs skulls, and attempted to extract and amplify aurochs DNA. Four samples from caves in southern England, ranging in age between 11,000 and 12,300 years, were assayed. The European aurochsen were far closer to the modern European taurine cow than either were to the south Asian zebu, a result subsequently corroborated by DNA from a number of other aurochsen. The genetics of these extinct beasts was adding clear support to the multiple domestication hypothesis.
These data from modern and ancient cattle, domesticated and wild, allowe
d another tree to be built, rich in information about the cattle’s past. A principal feature of the tree was the three distinct clusters, corresponding to distinct domestications in south Asia, the Near East and north Africa. The internal diversity within each cluster indicated several millennia of independent divergence in each. Then the trajectories of each could be independently followed into different regions of the Old World.
One region of particular interest was Europe. Here the taurine cow from the Near East could be traced, with a strong cluster including such breeds as Charolais, Simmental and Friesian bearing very similar mitochondrial sequences. Outside this cluster, another group of European cattle was out on a limb, and a limb that retained a surprisingly high diversity. These were the breeds from the north-western islands of Europe, including British and Channel Island breeds. Their mitochondrial divergence suggested a considerable antiquity, which puzzled Bradley’s team. One possible explanation is that some of the domesticated bulls on these islands had crossbred with indigenous island aurochsen, from which their divergent mitochondrial haplotypes had arisen. There is no problem with this genetically, now we know how very closely related the two were, but there is more to coupling than DNA. Rereading Isidor’s and Heylandus’s observations of the strength, size, speed and massive horns of the aurochs brings to mind how brave that diminutive ancient proto-Friesian bull would have needed to be to seek out such a mother for his calves. The idea of crossing with aurochsen was further questioned by the sequences now coming from a range of aurochs skeletons. Their mitochondria did not match up with those of the island breeds. It was more likely that the spread of domesticated cattle came in several stages, with successive waves sometimes–but not always–over-stamping earlier genetic patterns.
