100,000 years on your dad's tail

It is just 15 years since the first paper was published in Nature suggesting that the evidence of DNA samples taken from living populations around the world could allow the reconstruction of the descent of humankind from a notional African Eve who lived more than 100,000 years ago.

That remarkable and controversial assertion was not based on the study of the DNA taken from the nuclei of cells, from the chromosome, where most of the genetic material is found. Nuclear DNA is subject to recombination from generation to generation so, in general, it is not possible to trace lines of descent, since the genetic material gets thoroughly mixed. It was based rather on mitochondrial DNA, found elsewhere in the cell, which has the remarkable property of being inherited entirely from the mother, so is passed on exclusively through the female line. It is therefore well suited to the investigations of maternal lineages. The sequencing of mitochondrial DNA has allowed for the detection and study of the rather rare mutations that have occurred and that serve to distinguish one lineage from another.

The investigation of paternal lineages is even more recent. It is based on the study of the non-recombinant portion of the Y-chromosome, which is passed on from father to son. The definitive papers by Peter Underhill and his research group at Stanford University (with which Spencer Wells, the author of this book, has been associated) are just three years old.

Yet already the broad outlines are becoming clear of a descent structure revealing pattern of mutations. As one traces the path back up the descent-tree of lineages, the branches converge until one gets back to an African Adam some 60,000 years ago. There emerges a series of population histories that allows one to document the diffusion of our species out of Africa not long after that time. The "pathways" are marked by a series of mutations that can be followed in the Y-chromosome DNA sequences in populations in different parts of the world today. As befits the descendants of the "home team", African populations still show the greatest Y-chromosome diversity.

Wells' The Journey of Man deals specifically with the picture that has recently emerged from this Y-chromosome research, so that the reference to "Man" in the title is not a lapse in political correctness but the recognition that some of the most clear-cut new information comes from the Y-chromosome record of male lineages, which can now be set alongside the female lineages derived from mitochondrial DNA. His book comes at just the right moment: the broad outlines have become clear. The mitochondrial and Y-chromosome data are in good agreement, and it is evident that more molecular genetic data will soon be available to fill out these broad outlines.

Partly for that reason, The Journey of Man is the best account available of the story of human origins and dispersals, based on the newly emerging discipline of archaeo-genetics. Wells has himself contributed to some of the work discussed, and so writes with authority about the molecular genetics, and he succeeds well in establishing the broader intellectual context in which the work is situated, so that the book carries more conviction than the good overviews by science writers and journalists that are becoming available. It is also more up to date than existing popular accounts by respected leaders in the field such as Luca Cavalli-Sforza.

The story deals mainly with the Palaeolithic period, the time of the hunter-gatherers, over the past 100,000 years. Molecular genetics has not yet been able to tell us very much about our earlier hominid ancestors, since their DNA has not yet been recovered from the preserved and fossilised remains. It is remarkable enough that ancient DNA has been recovered from the 40,000-year-old remains of the Neanderthalers, who prove to be less closely related to our own species Homo sapiens than had previously been thought.

Wells is perhaps less at ease when dealing with events and processes over the past few thousand years. Most professional linguists, for instance, will raise their eyebrows over his support for the view that there may be a linguistic macro-family embracing the North Caucasian languages, the Basque language and Sumerian. This is speculation, and the arguments, although interesting, do not carry the same authority as other parts of the book.

In general this is a first-class account of a whole new approach to the human story that allows human population history to be reconstructed in an unexpected and convincing way. Who would have thought, 20 years ago, that we should have a better reconstruction of human descent from the study of the DNA of living populations than from the combined efforts of the world's archaeologists? In reality, the strength of the reconstruction comes from the cooperation between archaeo-geneticists and archaeologists, biological anthropologists and climatologists. This is an exciting time to be alive for an archaeologist, and Wells gives a lucid account of the science in his explanation of why this should be so.

Lord Renfrew is professor of archaeology, University of Cambridge.