In those cases where we do find a mixture of sexual and asexual organisms competing for the same resources, then sex must have a short-term advantage, not to the population as a whole but really an advantage at the individual level. And this is what George Williams calls a 'balance argument', and I think it's a very good argument. And essentially, the advantages we can identify fall into two classes. And I think we're still really quite uncertain about which of them is the more important, and different people you ask would take different views. But the categories are, first of all, the advantage of sex is it enables the population to evolve more rapidly to meet changed circumstances. And this is actually what I did most of my own mathematics on, when I was thinking about the problem, just showing under what circumstances and by how much the exchange of genes did enable a population to evolve more rapidly. But the essential argument goes back to the 1930s, to HJ Muller, I think, and to RA Fisher, and it simply is that if you have a sexual population, if two different, favourable mutations occur in that population, both can increase in frequency, and then, by sexual recombination, one can produce a population containing both of them. If you don't have sex and you have two individual favourable mutations, there's no way they can get together, the population could have one or the other but not both. And it has to wait for them to reoccur, to occur twice. And when one does the sums, it turns out that sex can really give a quite substantial advantage in rate of evolution. The problem is whether one can, so to speak, turn that into an individual advantage. Can one show that, actually, selection even acting at the level of an individual, could favour sexual reproduction? I think there are still quite formidable difficulties about that, although people have been struggling with it for a long time. The other question associated with it is; is it really true that the environment changes all that much? I mean, after all, it doesn't change all that quickly, surely we can't be evolving so fast that we're really having to chase.

[Q] You can calculate how quickly it ought to change in order to justify sex.

That's right, yes. And I think that the main - I mean, most people you spoke to about this, and I think I go along with it, to a considerable extent, feel that probably the main reason why these populations are having to evolve all the time, they can't just sit there and say, I've adapted myself to the environment, let me just sit here and not change too much. The main reason has to do with diseases and parasites. Parasites are constantly evolving new ways of attacking us, and we are having to evolve new ways of getting rid of them and of resisting them. And there is accumulating quite a bit of evidence tending to show that sexual populations and sexual individuals have an advantage in resisting parasitic infection, much greater than asexual individuals. And I think the data there is quite good, and a lot of people are certainly working on that, both theoretically and empirically.