... The new work could help test a hypothesis first proposed nearly 20 years ago, stating that sex evolved as a way to purge harmful mutations from a population. According to this view, the random shuffling of genes through sex will sometimes have the effect of concentrating many harmful mutations into single individuals.
These individuals will be less healthy than their peers, and therefore more likely to be weeded out by natural selection, the thinking goes.
This hypothesis, called the "mutational deterministic hypothesis," is controversial though, because it assumes that single mutations by themselves are only slightly harmful, while a combination of many mutations together is much more damaging. Scientists call this phenomenon "negative epistasis."
If negative epistasis were true, it would provide a powerful explanation for why sex has managed to persist for so long despite its numerous costs. But the phenomenon has yet to be widely demonstrated in nature and scientists have yet to figure out how such a thing evolved in the first place.
A new computer model by Ricardo Azevedo of the University of Houston and colleagues provides a possible answer to this last question. According to their model, detailed in the March 2 issue of the journal Nature, negative epistasis is a natural byproduct of sex itself.
--
http://livescience.com/animalworld/060301_evolution_sex.html