Extremely biased sex ratios: the case of sibling-incestuous parasitoid wasps

In 1930, Fisher argued that, in a species that must reproduce sexually, has exactly two sexes, and with perfectly random mating (every female-male pair chosen randomly, until no more female-male pair can be made, and each chosen pair mates, producing offsprings), the optimal sex ratio is 1:1 for the selfish genes.

The argument is very simple: Suppose that, right now, the sex ratio is 

female:male = A:B

and A > B. Then by random mating, every male would be paired off and have offsprings, but some females would be unlucky and have no offsprings. Then, every gene has a selfish interest in avoiding falling into a female offspring, which might end up a genetic dead-end. As such, genes that nudge the offspring sex-ratio to make more males would be adaptive and spread through the population, decreasing the female:male ratio.

Similarly if there are more males than females. Thus, the equilibrium sex ratio is 1:1.

This argument can be made rigorous with more algebra, but the gist is there.

However, this result can be grossly wrong when mating is very much not random, which is the case in many species with extensive sib-mating (blood sibling incest). This was pointed out by Extraordinary sex ratios, (Hamilton, 1967). 

The argument is pretty simple. Consider parasitoid wasps. Those are basically the real life equivalent of Alien xenomorphs: the mother wasps lay eggs inside a living caterpillar, and the young wasp larvas eat the caterpillar as they grow up. 

In many parasitoid wasps, the caterpillar is usually their only meal in life. The male wasps would crawl out and immediately mate with the females and then die. The female wasps would fly out and search for new victims.

Since each caterpillar is usually victimized by just one or a few mother wasps, once the caterpillar dies and the wasps crawl out to mate, a lot of the matings would be sib-matings. This breaks the random mating assumption, and causes a sex-ratio biased to have a lot of females.

The reason goes like this: Consider the extreme case of only one wasp parasitizing a caterpillar. All the offsprings would then be from one mother. In such cases, sex isn't useful, since all the benefits of sexual reproduction (genetic recombination giving innovations) are gone. In such cases, the mother might as well clone itself and have an all-female brood. 

This extreme case is tempered to only a very high female-to-male ratio in the case of a few mothers parasitizing a single caterpillar.