ÁñÁ«ÊÓƵ

Evolutionists sometimes dig deep searching for humanity's roots. The anatomist Frederick Wood-Jones long argued a tarsier origin for humans, and even in the mid-1960s some thought hominids to have differentiated 30-odd million years ago whereas others opted for a more modest 12 million to 14 million years. In contrast, all evidence accumulated over the past 30 years indicates a significantly more recent origin, about 6 million to 8 million years ago. Of course, it is possible to seek out older, pre-hominid ancestral forms without necessarily following Gilbert's Mikado back to the original protoplasmal, primordial atomic globule.

In The Hunt for the Dawn Monkey , Chris Beard, curator at Pittsburgh's Carnegie Museum, describes the search for the earliest anthropoid and so the common ancestor of all "higher" primates - monkeys, apes and humans - as opposed to lemurs, lorises and tarsiers. Beard has to review a bewildering array of fossil primates that usually consist of little more than jaw fragments with a few teeth, so needs to gently acquaint readers with the more esoteric details of dental anatomy. He succeeds very well, conveying a clear sense of the unfolding fossil record and our expanding knowledge of primate evolution, the subject's contemporary big beasts and their personalities, and the major issues that divide them.

Compared with their modern descendants, fossil primates were widely distributed, with abundant finds from the US, Europe, North Africa, China and Mongolia - in fact, from pretty well everywhere that modern primates (other than ourselves) are not found. With undoubted anthropoids known by final Eocene and Oligocene times, until recently most workers saw the group evolving in the later Eocene, perhaps 36 million to 40 million years ago, from either adapids (lemur-like forms) or omomyids (tarsier-like primates).

But more than a decade ago, Beard argued for anthropoids evolving at least 50 million years ago. Fossils from China of Eosimias (dawn monkey) and North African Algeripithecus display early anthropoid features and help push the record back to about 45 million years. Subsequent discoveries indicate to Beard an even earlier anthropoid origin in Asia, reaching Africa by the final Palaeocene. Needless to say, other workers do not necessarily agree with this view of anthropoid origins, and the book provides a good summary of alternative schemes. All in all, this is an excellent account: Beard succeeds in providing an engaging survey of a subject difficult to explain in a non-technical but suitably rigorous way.

The Chosen Species and Seven Million Years cover much the same ground - the origin and subsequent radiation of hominids, and the evolution of Homo sapiens. Juan Luis Arsuaga and Ignacio Martinez are palaeoanthropologists associated with many of the spectacular discoveries at the Spanish site of Atapuerca, including, at about 800,000 years, the earliest European human fossils, and the 300,000-year-old collection of probable Neanderthal ancestors from the nearby "Pit of Bones". Douglas Palmer is an experienced science writer and broadcaster with interests in the pre-mammalian fossil record, so bringing deep-time perspectives to his account of hominid evolution.

Both books are less detailed and less demanding than Beard's, and both present a mix of fossil, archaeological and genetic evidence bearing on human origins, supplemented by ecological and behavioural models of early hominids based on modern apes and humans. They are as concerned with functional and adaptive interpretations of the fossil record as with evolutionary issues, whereas for Beard the latter predominate. In fact, Beard's concerns are scarcely more than a footnote to these accounts:

"Finally, some middle Eocene fossils from China and Algeria (about 45 million years old) have been identified as anthropoids, though not all authors recognise them as such" is how Arsuaga and Martinez summarise the position, while in Palmer's account Eosimias and its relatives do at least get a couple of paragraphs to themselves.

The Chosen Species is deftly and clearly written, with some jolly reconstructions of fossil hominids. The title, with its connotations of grand design and a directed process culminating in ourselves, is unfortunate, since Arsuaga and Martinez make it clear that this is not what they intend. They are sceptical about Homo rudolfensis (based on the 1470 specimen) arguing that it can be included within Homo habilis , but they accept the distinctiveness of African erectus-like forms as Homo ergaster , with erectus proper confined to Asia. And they argue the case for the early Atapuerca fossils as Homo antecessor ("Pioneer Man"), ultimately derived from Africa via Asia, with a unique evolutionary status as the common ancestor of Neanderthals (evolving in Europe) and modern humans (evolving in Africa from later "archaics", themselves the descendants of as yet undiscovered African antecessor populations). We thus have two "out of Africa" expansions: a Lower Pleistocene one evolving into Homo erectus in the Far East and a later Homo antecessor - Homo heidelbergensis - Neanderthal cluster in Europe, and an expansion about 100,000 years ago, when modern humans moved out of their African heartland, leading to the extinction of archaic humans elsewhere. The spectacular finds from Dmanisi, Georgia, are crucial to modelling the phase-one expansion, and so it is surprising to find hardly any reference to them in Arsuaga and Martinez's account. Surprising that is, until one realises that this is the English translation of a Spanish text published in 1998, before all but the first of the Dmanisi discoveries. For the same reason there is no mention of other important finds from the past few years, such as the South African Little Foot - the most complete early hominid skeleton known - or the much more recent Flores "hobbit".

These and other omissions reduce the book's value, although the authors' virtues of clarity and enthusiasm and their balanced discussion of functional and adaptive aspects make for a useful account. Seven Million Years (the estimated time since chimpanzee-human divergence) is as current as any text can be, and so Dmanisi, Flores and other discoveries receive due attention. Palmer adopts a strongly historical/biographical approach to explaining the changing significance of the unfolding evidence, and the context for its discovery and initial, usually incorrect, evaluation. The Leakey family's outstanding contributions to African prehistory rightly feature prominently here, while some murkier undercurrents of palaeoanthropological politics are touched on when reviewing fossil hunting elsewhere. Palmer emphasises the often crucial impact individual fossils may have in forcing us to recast established views on the processes of human evolution. His point is well taken, but it is a pity that the book isn't better illustrated. There are some rather dull drawings of fossil crania, depictions of stone tools and a few maps, but not much more. Like Arsuaga and Martinez, Palmer veers towards distinguishing African ergaster from Asian erectus, but is cautious about Homo antecessor as a distinct species and its supposedly central role in both Neanderthal and modern human ancestry. His summary of the significance of molecular anthropology for reconstructing human evolution is a particular strength of the final chapter. But Beard's account of earlier primate evolution is the pick of this trio.

Alan Bilsborough is professor of anthropology and pro vice-chancellor of Durham University.