You might answer that apes are 'animals', while humans are humans, and that is enough. An ethical code for treating humans should not be extended to an 'animal', no matter what percentage of its genes it shares with us, and no matter what its capacity for social relationships or for feeling pain. That is an arbitrary but at least self-consistent answer that cannot be lightly dismissed. In that case, learning more about our ancestral relationships will not have any ethical consequences, but it will still satisfy our intellectual curiosity to understand where we come from. Every human society has felt a deep need to make sense of its origins, and has answered that need with its own story of the Creation. The Tale of Three Chimps is the creation story of our time.

For centuries it has been clear approximately where we fit into the animal kingdom. We are obviously mammals, the group of animals characterized by having hair, nursing their young, and other features. Among mammals we are obviously primates, the group of mammals including monkeys and apes. We share with other primates numerous traits lacking in most other mammals, such as flat fingernails and toenails rather than claws, hands for gripping, a thumb that can be opposed to the other four fingers, and a penis that hangs free rather than being attached to the abdomen. Already by the Second Century AD, the Greek physician Galen deduced our approximate place in Nature correctly when he dissected various animals and found that a monkey was 'most similar to man in viscera, muscles, arteries, veins, nerves and in the form of bones'. It is also easy to place us within the primates, among which we are obviously more similar to apes than to monkeys. To name only one of the most visible signs, monkeys sport tails, which we lack along with apes. It is also clear that gibbons, with their small size and very long arms, are the most distinctive apes, and that orangutans, chimpanzees, gorillas, and humans are all more closely related to each other than any of them is to gibbons. But to go further with our relationships-proves unexpectedly difficult. It has provoked an intense scientific debate, which revolves around three questions including the one that I posed in the first paragraph of this chapter: What is the detailed family tree of relationships among humans, the living apes, and extinct ancestral apes? For example, which of the living apes is our closest relative? When did we and that closest living relative, whichever ape it is, last share a common ancestor? What fraction of our genes do we share with that closest living relative? At first, it would seem natural to assume that comparative anatomy had already solved the first of those three questions. We look especially like chimpanzees and gorillas, but differ from them in obvious features such as our larger brains, upright posture, and much sparser body hair, as well as in many more subtle points. However, on closer examination these anatomical facts are not decisive. Depending on what anatomical characters one considers most important and how one interprets them, biologists differ on whether we are most closely related to the orangutan (the minority view), with chimps and gorillas having branched off our family tree before we split off from orangutans, or whether we are instead closest to chimps and gorillas (the majority view), with the ancestors of orangutans having gone their separate way earlier. Within the majority, most biologists have thought that gorillas and chimps are more like each other than either is like us, implying that we branched off before the gorillas and chimps diverged from each other. This conclusion reflects the common-sense view that chimps and gorillas can be lumped in a category termed 'apes', while we are something different. However, it is also conceivable that we look distinct only because chimps and gorillas have not changed much since we shared a common ancestor with them, while we were changing greatly in a few important and highly visible features like upright posture and brain size. In that case, humans might be most similar to gorillas, or humans might be most similar to chimps, or humans and gorillas and chimps might be roughly equidistant from each other, in overall genetic make-up.

Hence, anatomists have continued to argue about the first question, the details of our family tree. Whichever tree one prefers, anatomical studies by themselves tell us nothing about the second and third questions, our time of divergence and genetic distance from apes. Perhaps fossil evidence might in principle solve the questions of the correct ancestral tree and of dating, though not the question of genetic distance. If we had abundant fossils, we might hope to find a series of dated proto-human fossils and another series of dated proto-chimp fossils converging on a common ancestor around ten million years ago, converging in turn on a series of proto-gorilla fossils twelve million years ago. Unfortunately, that hope for insight from the fossil record has also been frustrated, because almost no ape fossils of any sort have been found for the crucially relevant period between five and fourteen million years ago in Africa.

The solution to these questions about our origins came from an unexpected direction: molecular biology as applied to bird taxonomy. About thirty years ago, molecular biologists began to realize that the chemicals of which plants and animals are composed might provide 'clocks' by which to measure genetic distances and to date times of evolutionary divergence. The idea is as follows. Suppose there is some class of molecules that occurs in all species, and whose particular structure in each species is genetically determined. Suppose further that that structure changes slowly over the course of millions of years because of genetic mutations, and that the rate of change is the same in all species. Two species derived from a common ancestor would start off with identical forms of the molecule, which they inherited from that ancestor, but mutations would then occur independently and produce structural changes between the molecules of the two species. The two species' versions of the molecule would gradually diverge in structure. If we knew how many structural changes occur on the average every million years, we could then use the difference today in the molecule's structure between any two related animal species as a clock, to calculate how much time had passed since the species shared a common ancestor. For instance, suppose one knew from fossil evidence that lions and tigers diverged five million years ago. Suppose the molecule in lions were ninety-nine per cent identical in structure to the corresponding molecule in tigers and differed only by one per cent. If one then took a pair of species of unknown fossil history and found that the molecule differed by three per cent between those two species, the molecular clock would say that they had diverged three times five million, or fifteen million, years ago.

Neat as this scheme sounds on paper, testing whether it succeeds in practice has cost biologists much effort. Four things had to be done before molecular clocks could be applied: find the best molecule; find a quick way of measuring changes in its structure; prove that the clock runs steady (that is, that the molecule's structure really does evolve at the same rate among all species that one is studying); and measure what that rate is.

Molecular biologists worked out the first two of these problems by around 1970. The best molecule proved to be deoxyribonucleic acid (abbreviated to DNA), the famous substance whose structure James Watson and Francis Crick showed to consist of a double helix, thereby revolutionizing the study of genetics. DNA is made up of two complementary and extremely long chains, each made up of four types of small molecules whose sequence within the chain carries all the genetic information transmitted from parents to offspring. A quick method of measuring changes in DNA structure is to mix the DNA from two species, then to measure by how many degrees of temperature the melting point of the mixed (hybrid) DNA is reduced below the melting point of pure DNA from a single species. Hence the method is generally referred to as DNA hybridization. As it turns out, a melting point lowered by one degree centigrade (abbreviated: delta T = 1 °C) means that the DNA's of the two species differ by roughly one per cent. In the 1970s most molecular biologists and most taxonomists had little interest in each other's work. Among the few taxonomists who appreciated the potential power of the new DNA hybridization technique was Charles Sibley, an ornithologist then serving as Professor of Ornithology and Director at Yale's Peabody Museum of Natural \ History. Bird taxonomy is a difficult field because of the severe anatomical constraints imposed by flight. There are only so many ways to design a bird capable, say, of catching insects in mid-air, with the result that birds of similar habits tend to have very similar anatomies, whatever their ancestry. For example, American vultures look and behave much like Old World vultures, but biologists have come to realize that the former are related to storks, the latter to hawks, and that their resemblances result from their common lifestyle.