The engine and gearbox analogy is that if you have two clapped out Minis, one with no engine and the other one with no gearbox, you can actually make one working Mini by putting together the good engine and the good gearbox together in one car. And sex, by combining genes from different individuals, may be able to construct one efficient working individual from two that aren't too good. The trouble with that argument, of course, is that you might finish up with a Mini with no engine and no gearbox. And the mathematics of this, again, becomes really quite... quite hairy. There are two questions, first of all: Is it true that a sexual population can get rid of harmful mutations and combine good ones? And here we're not talking about changing environment, we're just talking about getting rid of the load... the continuous load of bad mutations that are happening. Is a sexual population really superior to an asexual one? And I think everybody now, who's been into the mathematics, would agree, yes, it is. There are contexts, several contexts, in which that is certainly true. A much harder question is: can we turn that into an individual selection explanation? Could selection, even within a population, between sexual and asexual competitors, actually favour recombination and sex. We can certainly construct models in which it is true, that it can happen. The trouble is that the models assume a very high rate of deleterious mutation. And at the moment, I'm uncertain whether, in the real world, rates of deleterious mutation are high enough to justify the theory. So although, you know, we've been arguing about this since the mid '70s, I'd guess, I think it's fair to say that anybody who is being at all balanced would [not] think that the problem had been solved, I think the problem is still quite hard.