Unlocking the Past

by Martin Jones

  The dog is a very special animal, but not alone in being treated with a particular kind of respect. In different parts of the world, other species–for example the cat, the horse and the cow–have been similarly accorded a special status in the human world. None the less, domestication of animals has less commonly been about companionship and more commonly about changing our relationships with the animals we kill and eat. Many of our victims have been the larger mammalian herbivores that our ancestors tracked and preyed upon for tens of thousands of years before domesticating and containing them. This is a group that would have had a rocky ride through the intense climatic fluctuations of the Quaternary Epoch, even without the added peril of human predation. Rapid climatic and environmental change generally favours smaller, faster growing things. It was often human predation that tipped the balance of an already fragile relationship between an animal and its changing environment. That is perhaps how the dramatic loss of indigenous large mammals from the New World is to be explained. In other instances, the human predator turned farmer or pastoralist has saved large mammals from extinction by enslaving them. Each fated animal has its own story that can be tracked by molecular methods that reveal their histories and contribute to our understanding of that very human process ofdomestication. First, let us look at some of those that neither sought nor found refuge in human enslavement, but perished instead.

  the perils of being large

  Over the last 2 million years, global temperature fluctuations gained a fresh intensity, intermittently blanketing large swathes of the planet with ice. The period has been peppered with extinctions of mammals in the larger size range. They seem to have been especially vulnerable to the periods of global ice melt, when temperature changes were particularly pronounced. The most recent period of warming saw a swathe of such extinctions. Occasionally, domestication by humans provided a lifeline to these species. Some of our largest domesticated animals, such as cows, horses and camels, have, to the best of our knowledge, lost their ancestral relatives from the wild. Sometimes we find a wild relative in remarkable condition–a frozen wild horse, preserved for 30,000 years in the Siberian ice, or a 9,000-year-old wild camelid, desiccated in the Peruvian Andes. In each case, the hair and flesh is preserved along with the skeleton, and retains its natural colour. We can also find preserved remains of the species that did not make it, and found neither a natural refuge nor the protection of domesticated subservience to humans. One of these is a familiar feature of the Quaternary muds and gravels, which frequently yield its massive teeth and sometimes its enormous tusks. From time to time from beneath the Siberian ice and permafrost an entire woolly mammoth comes to light, with flesh and hair intact.

  woolly mammoths, stopwatches, clocks and calendars

  A number of extinct animals can now also be reached through their DNA. In previous chapters we have heard about the South African quagga, and our close relative the Neanderthal. Ancient DNA studies have also been conducted on the marsupial wolf from Tasmania; the giant sloth that once roamed in America; the dodo that travellers to Mauritius so rapidly exterminated; that gigantic flightless bird from New Zealand called the moa; an island tortoise; and the wild cow, which is featured later on in this chapter. Some of these, like the quagga and wild cow, are sufficiently close to living relatives that we can use the now familiar mitochondrial control region to judge the closeness. With others, it may be far less clear. The ancient giant sloth is clearly related in some way to its diminutive living relatives, as is the woolly mammoth to its naked elephant cousins. How close is uncertain. If they are very near, the control region will measure the distance well. If they are too distant, then it may get a little confused, with reverse mutations and deletions piling up upon and overprinting one another. A more slowly evolving stretch of DNA might give a better measure. In these terms, the control region provides a stopwatch, when what is needed is a slower-paced clock.

  If we look on either side of the mitochondrial control region, stretches of DNA sequence are found that evolve at a slower rate. Immediately adjacent to the control region is a gene involved in energy management called the cytochrome b gene. Variations within this gene provide a suitable clock to compare animals that are broadly in the same genus or family. On either side of the control region are two genes involved in assembling the cell’s RNA toolkit. They are referred to as the 12S rRNA and 16S rRNA. The numbers 12 and 16 are measures of the size of the RNA molecules they are programmed to build. The 12S rRNA gene provides another clock, measuring intermediate evolutionary rates, while the 16S rRNA evolves at a slower pace still, providing a calendar rather than a clock. Such an array of timepieces can be brought together to assess evolutionary relationships on many scales, and can be assessed together if the scale is unclear.

  Erika Hagelberg settled for the cytochrome b clock when she decided to establish the presence of ancient DNA in these frozen mammoths. She took samples from two of the beasts that had come to light in the 1970s. One, from the Taymyr Peninsular, the northernmost extension of the Siberian mainland, was carbon dated to at least 40,000-50,000 years old. The other, from further to the east and 16m down into the permafrost around the Allaikha River, could have been over 150,000 years old, judging from associated fossil evidence. From both mammoths, she managed to amplify sequences from the cytochrome b gene of over 300 base-pairs in length. They may well be the oldest confirmed specimens to yield sound, amplifiable DNA.

  The sequences were fairly close to those of living African and Asian elephants, and Hagelberg concluded that they formed a fairly tight group in evolutionary terms. Svante Pääbo was also attracted to the mammoth problem, but decided to work on the 16S rRNA gene calendar. He extended the sample to include five distinct mammoths. Four of these gave successful amplifications, allowing comparison with a range of different ungulates, including the elephants. The mammoths did indeed cluster with the elephants, but what was really striking about them was how much they varied. The two living genera, the Indian and the African elephant, differed in only two positions along the 93-base-pair sequence studied. This gene is after all a calendar, not a stopwatch. However, even in the small sample of mammoths studied by Pääbo’s Munich team, there were differences along this stretch in up to five positions along the sequence. A tiny glimpse was caught of the great genetic diversity of these extinct mammoths, apparently dwarfing the diversity of living elephants. Looking just at the whole fossils, the impression gained was that two out of three elephants survived, all but the woolly one. On the basis of DNA evidence Svante Pääbo and his colleague Matthias Hoss were emphasizing instead the diversity of ancient mammoths, of which all but a few–just a couple of naked ones–perished.

  Shortly after this work was published, the picture was further clarified by bringing another one of these large, now extinct, mammals into the picture. This was the mastodon, another hairy relative of the modern elephant that was found throughout the world and probably preyed upon by the first Americans. Edward Golenberg, whose ancient Magnolia leaf had been given a rough ride by Pääbo and others, now used mastodon evidence to take a critical look at Pääbo’s own results with mammoths. Golenberg had sampled a well-preserved mastodon skeleton from a Michigan bog, along with a few extra mammoth fossils. What he and his colleagues were able to show was that, taken alone, the ancient mammoth and modern elephant DNA could generate a variety of virtual volutionary bushes, according to slight fine-tuning of the computer program employed. It lacked robustness on its own. When another dimension of data was incorporated in the form of mastodon DNA, however, that robustness was gained. Those virtual bushes, when properly rooted in this way, resolved themselves into a single pattern. This pattern had the mastodon out on a limb as the distant cousin, and the mammoths clustering closer to the Asian than to the African elephant. This did not necessarily mean that Hoss and Pääbo were wrong about the diversity of extinct mammoths, but that there was more pattern that could be gleaned by bringing the DNA of other extinct species into the pict
ure.

  Beyond the mastodons and the mammoths in the ice, a range of other large mammals struggled against the uncertainties of climatic change. One that can also be recovered from the Siberian ice has survived within the human sphere, although its ancestors are gone from the wild.

  horse-riders of the steppes?

  The frozen wild horses of Siberia, whose hair and flesh has remained intact for 30,000 years or more, are the most recent in a much older sequence of horse fossils, mostly surviving as bones, and going back tens of millions of years. These are the fossils that, in the late nineteenth and early twentieth centuries, provided one of the best sequences to support Charles Darwin’s idea of gradual evolution by natural selection. The earliest members in this sequence, going back some 55 million years, are found in North America. By 2 million years ago, a fairly modern looking wild horse roamed the plains of North and South America as the fluctuating climate opened up vast stretches of grassland. Moving forward another million years, some of these horses made the journey across the Bering Straits, when the sea level was suitably low, to the grassy expanses of Siberia, eventually giving rise to our 30,000-year-old specimen. From Siberia they spread to Mongolia and Asia, arriving in Europe 750,000 years ago.

  There are no wild descendants of our frozen horse. Truly wild horses disappeared from their ancestral homeland in America between 10,000 and 8,000 years ago. In Asia, they gradually dwindled in number, coming close to extinction in the last century. Those that entered domestication flourished and diversified, giving mobility to humans to a degree unmatched until the invention of the combustion engine. Overland travel on foot through unmanaged vegetation was always extremely slow. Although many prehistoric communities would have covered large distances during their lives, the rate at which they did so would have borne no comparison with that of the earliest horse-riders who traversed vast expanses of the later prehistoric world. We can get a sense of the scale of the transformation by following their sixteenth-century reintroduction into the New World. This fresh arrival of horses heralded an entirely new cultural tradition among the Plains Indians, and an entirely new set of ecological relations with the prairie and the buffalo. The great mobility of horse-riders has underpinned a number of hypotheses about some of the ‘great journeys’ of prehistory, to which we shall turn in the next chapter. Some investigators have followed Maria Gimbutas, in holding the horse, or more precisely its riders, responsible for the spread of Indo-European languages. She saw the domestication of the horse, in the wooded steppe north of the Black and Caspian Seas, as both the cause of, and the means for, a series of westward waves of horse-riding warrior societies. These waves took them across Europe, where they overran the more peaceful and harmonious matriarchal societies of Neolithic Europe. Prehistoric settlements such as Dereivka in Russia and Botai in Kazakhstan were spilling over with horse bones. From here her three waves of ‘kurgan’ (mound-building) cultures moved westward, spreading a new language, culture, technology and masculine ideology across western Eurasia.

  This account of the radiation of the horse from an Asian centre of origin, taking with it the cultural package of its riders, is open to question on various fronts. Modern breeds seem to be very diverse, something that has led others to speculate about two, perhaps three separate domestications. These would separately account for breeds described as ‘light’ or ‘warm-blooded’, and those described as ‘heavy’ or ‘cold-blooded’. The first group includes the rather elegant, fast and spirited animals such as the Arabian so favoured by horse-racers. The second includes the massive hard workers such as the Shire horse and the Suffolk Punch. Between these lie many crosses and intermediates, and there are the so-called ‘native ponies’ on the moors of south-west England, for example. According to one argument, the light breeds derive from wild horses from the heart of Asia, the heavy breeds arising independently from a European wild forest horse. Another argument gives native ponies their own local ancestry. This was clearly a problem to be tackled by genetics and ancient molecular evidence.

  The palaeontologist Adrian Lister was keen to crack the problem and teamed up with geneticist Helen Stanley to do so. They collected ancient DNA from a variety of sources. Their most ancient material was the 30,000-year-old horse frozen in the Siberian ice, and bones of a similar age from Kazakhstan. Wild horse bones were collected from where they had lain buried for 12,000-14,000 years, in a cave in southern England, and in the tar pits of California. Others were selected from the 5,000-year-old site of Botai in Kazakhstan. The wild horse had lingered far longer than the woolly mammoth, and some available specimens were a great deal younger. The youngest ‘ancient’ samples were less than a century old, and came from the most recent horse breed to disappear from the wild. This was the ‘tarpan’, a small dun-coloured animal with a flowing mane and tail, which had survived through the Middle Ages in small herds in remote parts of central and eastern Europe. Some say there were two types of tarpan, a forest and a steppe form, but as the last known individual died in 1918, this is open to speculation. Fragments of its teeth and skin survived for Lister’s and Stanley’s analyses.

  On the margins between ancient and modern DNA, they also had blood samples from a Mongolian wild horse hovering on the very edge of extinction. The Russian traveller, Nikolai Przewalski, had encountered this stocky animal, with its distinctive dorsal stripe, when travelling in the late 1870s through western Mongolia in search of undiscovered plants and animals. Thirty years on, twelve of these animals were captured and transferred to zoos, which is just as well, as the most recent expeditions in the area have failed to rediscover Przewalski’s horse in the wild. The descendants of those twelve animals have provided blood samples to add to Lister’s and Stanley’s specimens. In addition to these various ancient specimens, they took samples from a range of living light and heavy breeds, as well as from a series of native ponies.

  They studied both the mitochondrial stopwatch (the control region) and the corresponding calendar (the 16S rRNA gene) to get a sense of variation on different scales. From the stopwatch, they built a series of trees and networks that displayed a now-familiar pattern of deep-seated diversity. Working from a horse-zebra split between 2-4 million years ago, in this case based on a rich and detailed fossil record, the various living horse breeds were seen to stem from a common maternal ancestor that lived in the region of 880,000 years ago. Here was yet another phylogeny too deep to be contained within the archaeologically attested time scale for domestication. Those early central Asian sites, yielding the earliest archaeological evidence for intensive horse management, go back 5,000 years at most. There is no way that the sequence divergence measured by Lister and Stanley could be contained within this abbreviated time scale. Was this a split between light and heavy breeds? Apparently not. Two heavy breeds, the Suffolk Punch and Shire, are on completely different branches, and such breeds as the thoroughbred and Shetland pony are scattered among different branches. The Przewalski’s horses do indeed cluster together, but not in a way that might suggest a long and separate history.

  Lister’s and Stanley’s key finding was that the date for the common maternal ancestry of all living horses, including Przewalski’s horse, corresponds well with the fossil record date for the first entry of the horse into Eurasia. Following this entry, and the genetic bottleneck associated with it, a huge range soon developed, and over the millennia a great deal of variation built up in their mitochondrial DNA, which is rather randomly distributed among modern horse breeds. Once again, the evidence for a major animal species in the human orbit is inconsistent with a very localized single domestication event.

  Going back to those early sites in central Asia, we can see that diffuse and gradual process going on within particular culture groups. When such sites as Dereivka and Botai first came to light, there was a tendency to seek within them the ‘big event’, the revolutionary transition from hunting wild animals to husbanding domestic livestock. A range of markers, from bone structure and wear on teeth to
the simple quantity of horse bones, was used to argue the case. Marsha Levine has recently looked more carefully and closely at these sites, the population structure of the horses, and the evidence of stress and disease in the bones. She has arrived at a more complex story. The prehistoric inhabitants of these sites had a very long-standing and intimate knowledge of their horses. They understood their behaviour and movements, utilized them in a variety of ways and, judging from their elegantly carved horse bones, held them in veneration. But there is no particular evidence that the horses were other than wild in the conventional sense, and the point at which they were fully contained and bred within the domestic sphere is still open to debate. As late as the Iron Age, when horses were clearly being harnessed, ridden and widely used, the population structure of southern British horses suggests that many were still rounded up from a feral state for training and use. Against a behavioural background of this kind, the rather dispersed picture of domestication coming from the evidence of mitochondrial DNA need not surprise us.

  crossing mountain and desert

  More mysterious than either the mammoth or the horse is the camel. The two kinds of Old World camels have been fundamental to the human exploitation of the world’s regions of greatest aridity and poorest vegetation. The two-humped Bactrian camel can be seen carrying loads throughout the highlands of central Asia. The one-humped dromedary also supplies meat, milk and wool from India across to the Sahara. Yet neither has left a trace of its ancestry in the present or the past. Nikolai Przewalski was convinced that he had found a wild camel as well as a wild horse, and one or two feral examples exist. None however has convinced the zoologists of its being truly wild and ancestral. To find any wild relatives, we have to trace their presumed journey out of Asia and back to the continent where the camelid line evolved.