« Priorities.... | Gene Expression Front Page | The Passion »
October 07, 2003


I enjoyed reading Bryan Sykes’s book The Seven Daughters of Eve, so I looked forward to his new one, Adam’s Curse: A future without men, Bantam Press, 2003.

I wasn’t disappointed. The book is a rattling good read, and I highly recommend it. Unlike Seven Daughters, this one is not primarily concerned with the reconstruction of phylogeny, so it is not a Y-chromosome counterpart to Sykes’s MtDNA studies. It has more in common with Steve Jones’s Y: the Descent of Man, but there is little overlap in detail, and overall I prefer Sykes’s book. Highlights include a fascinating history of human chromosome studies, a clear explanation of sex determination mechanisms, and a racy account of the ancestry of the Clan Donald.

But the main theme of the book is the conflict between different parts of the genome: mitochondria, Y-chromosomes, and other nuclear genes, all having different interests. Sykes argues persuasively that mitochondria or Y-chromosomes have succeeded in biasing some human lineages towards producing girls and boys respectively. He also argues that the interest of the nuclear genes in avoiding conflict between different strains of mitochondria is the main reason for the differentiation of two sexes in the first place. I’m not so convinced by this, but it’s an interesting theory.

Where I really part company with Sykes is over the deterioration of the Y-chromosome. As I pointed out in my comments on Jones, there are good reasons for the Y-chromosome to deteriorate, but this need not cause any great harm. Sykes takes a different view: it will be ‘death by a thousand cuts. Unable to repair themselves by recombination, the wounded Y-chromosomes will stagger on through succeeding generations, gradually becoming weaker and weaker... As human Y-chromosomes in general become more and more unhealthy there will be a relentless and progressive reduction in male fertility...’ Eventually, the only hope for the human race will be to replace the traditional method of reproduction by artificial insertion of genes into the egg - perhaps female genes! Hence the gloomy prediction of the book’s title. Sykes even estimates the time it will take for men as we know them to become extinct, concluding that it is about 5000 generations or 125,000 years.

It is an ingenious and alarming (or exciting, for lezzers) theory, but, as Captain Blackadder would say, it has one tiny flaw: it is bollocks.

The objections are both theoretical and empirical....

On a theoretical level, it is well known that if a genome does not recombine in sexual reproduction (which is true of the Y-chromosome itself), mutations will
accumulate, provided each mutation has only a very small effect on fitness. If the effects on fitness are small enough, all members of the population will be roughly equally affected, since they are all incurring small mutations all the time. For any individual, it is likely that another mutation will come along before the previous one is eliminated by natural selection. But if the effect on fitness is large, each mutation will be eliminated by selection before the next one comes along.

Sykes recognises this point in principle, but in his calculations he ignores it. He assumes that in each generation 1 per cent of men have a mutation which makes them 10 per cent less fertile than their fathers, and that the fertility of the population as a whole therefore declines by 0.1 per cent (1 per cent of 10 per cent) per generation. On this basis he calculates that fertility would decline to 1 per cent of its present level within 5000 generations. (Sykes doesn’t give details of his method of calculation, but I have checked sample values from his graph on page 239, and they seem consistent with simply multiplying by .999 at each generation, thus reducing fertility by 0.1 per cent of its level at that point.)

The trouble is that for individuals, a 10 per cent decline in fertility is not a small effect on fitness: it is a very big one. At least, it is if ‘decline in fertility’ means a reduction in the average number of offspring - which is surely what it ought to mean. Instead, Sykes seems to be measuring ‘fertility’ by some criterion such as sperm count or motility, and it is admittedly possible that this might decline (for a while) without greatly affecting the number of offspring. But even on this interpretation, the decline would have serious effects on reproductive fitness long before the 1 per cent mark was reached. There would then be a huge reproductive advantage to those males with the least damaged Y-chromosomes, or those who had compensating mutations on other chromosomes which restored fertility. We may also expect that if society were threatened by a fertility crash, then the most fertile males would be identified and used for artificial insemination.

I think this invalidates Sykes’s gloomy prediction on theoretical grounds. On an empirical level it is also shaky. No-one disputes that the Y-chromosome has accumulated damage, but there is no evidence that this affects male fertility. The X-chromosome takes over the work of the damaged parts, and the major remaining function of the Y is to determine sex, which it still does perfectly well. Of course, there is evidence of male fertility problems in the late 20th century, among both humans and other animals, as measured by low sperm counts, etc., but this is due to modern environmental factors such as chemicals in the water supply. Some of the worst affected animals are fish, amphibians, and reptiles (like the alligators of the Everglades), which don’t have a Y-chromosome at all!

Moreover, if the Sykes theory were right, it would apply to the whole range of mammal species (and many others), and numerous species would go extinct due to male infertility. Sykes recognises this implication, and boldly claims that infertility has caused many extinctions, though he admits he cannot prove it (page 294). But if he were right, we ought to be able to find species in all stages of Y-chromosome infertility, including some on the brink of extinction from this cause. He offers no evidence whatever for this. There are plenty of species on the brink of extinction, but for entirely different reasons.

But don’t let this put you off the book. The gloom and doom is only a minor part of the book, and confined to the last few chapters.


Posted by David B at 02:34 AM

cavalli-sforza rips into syke's interp of some mtDNA facts (and yes, they are facts) with a certain self-promoting slant in the new paper. i had some of the same questions as cavalli-sforza when reading 7 daughters of eve. sykes seems like a fine scientist, but his ego is really out of control....

Posted by: razib at October 7, 2003 11:50 AM

The claim that men are going to disappear soon or that Y-based infertility will drive our species extinct has always sounded like little more than a way to grab attention. Even if the Y were to vanish completely, the male-determining genes could survive by hopping to other chromosomes. That's what happened in the past, and in some mammals who have lost the Y, it's already happened again. I wrote about some research on this in an article last year ago. ( http://www.carlzimmer.com/articles/2002/VanishingY2002.html )

Posted by: Carl Zimmer at October 7, 2003 11:58 AM

also, remember that sykes isn't just a scientist, he's a businessman. self-promotion is more than something that can satisfy the ego for him, it can make him a rich man....

Posted by: razib at October 7, 2003 12:12 PM

The argument that the Y chromosome will disappear in a few thousand years doesn't make much sense to me. Assuming one is not a creationist who believes each mammal species was separately created with a pristine new Y chromosome, the Y chromosomes of all mammals must all trace back to a common ancestor living hundreds of millions of years ago--what does Sykes think has changed that would suddenly make our Y chromosomes unviable? Is he arguing that this is an effect of civilization, medicine, and the consequent relaxed selection pressures on harmful mutations in humans?

Posted by: Jesse at October 8, 2003 01:07 AM

Ah, I see at the end of David Burbridge's review that Sykes thinks this argument applies to other species too, so he can't just be thinking of relaxed selection in humans due to civilization. But that makes me think this idea really is completely crackpot--like I said, unless you're a creationist it's not as if a new species will somehow reset its Y-chromosome when it springs into being, it will inherit whatever level of degradation was present in the species it branched off from. The proper unit for thinking about progressive degradation of a gene would be a clade, not a species--any degradation of the Y-chromosome would have been continually building up from the last common ancestor of all animals with a Y-chromosome (which I think would be the common ancestor of all mammals, which would have lived about 200 million years ago). If our Y-chromosome has survived for that many millions of generations, it just doesn't make sense to predict it'll be gone in 5000 more generations unless you can point to some factor that's changed the selection process significantly in recent times.

Posted by: Jesse at October 8, 2003 01:22 AM

Jesse: the same argument occurred to me (see my comments on Jones), but maybe Sykes would answer it along the lines that in the formation of a new species there is a selective 'bottleneck' of some kind. But I can't find any explicit statement in his book.

I agree that Sykes is a self-publicist, but I also genuinely think it is (overall) an excellent book. But maybe best to wait for the paperback!

Posted by: David B at October 8, 2003 02:51 AM