« no 'safe' time to avoid pregnancy | Gene Expression Front Page | Individual impact of affirmative action »
July 09, 2003


I've just been reading Bryan Sykes's 'The Seven Daughters of Eve'. It turned out much better than I expected. The publicity for the book emphasises the fictionalised reconstruction of the lives of the 'daughters' - Ursula, Xenia, etc - but fortunately this turns out to be a small and unimportant part of the book.

Most of the book is a very clear and readable account of the development of ancestry tracing through mitochondrial DNA. I thought I knew a bit about this already, but I understood it much better after reading the book.

The other big theme is the conflict between Sykes's thesis, based on Mito DNA, that the bulk of European ancestry is paleolithic, and the Cavalli-Sforza doctrine, based on principal components analysis of nuclear DNA, that the bulk of it is descended from neolithic immigrants from the middle-east. According to Sykes's account, Cavalli-Sforza has finally conceded that the 'paleolithic' element is much larger, and now claims that this is consistent with what he said all along! Is this right?

Just a few grumbles about the book: (a) no notes or bibliography (b) the fictionalised reconstructions don't seem to take much account of modern hunter-gatherer societies, e.g. their marriage and kinship customs, and (c) the usual politically correct flapdoodle about the 'nonsense of any biological basis for racial classifications' (page 359 in the Corgi edn.), when all he really means is the platitude that races are not 'pure' and sharply differentiated. Well, of course not.


Posted by David B at 02:22 AM

yes, the 20% figure was confirmed all around for about 1 year. recently a new estimate gives 50% for the "neolithic middle eastern" contribution to european genomic ancestry....

you can look up the info over at the THE HUMAN RACES website on the blogroll....

Posted by: razib at July 9, 2003 09:09 AM