Thursday, 2 September 2010

Theories of "altruistic" behaviour in eusocial insects

An article in Wired about a new spat over whether the theory of kin selection is justified, or necessary, sees E. O. Wilson doing battle once again against the kin selection brigade.

Something about this article set me thinking. I've never liked this idea of "altruism" in the eusocial insects, but I've never quite been able to say why, but tonight something — perhaps something in the language of the article — suddenly struck me.

Perhaps it was this: "according to the idea of kin selection, workers without young more than compensate by sharing in the reproductive success of relatives, with whom they share genes". I don't see how sharing in the success of a relative with whom you share less than 100% of your genes can ever even equal having your own offspring, never mind improve upon it.

Then it struck me what I don't like about it: there's no mechanism for selection if you look at the workers, because the workers don't reproduce! If you look at a bee hive, only one individual reproduces, and that's the queen. So rather than asking how or why workers would "evolve" a particular behaviour (because how can they "evolve" anything, when they have no offspring?) we should be asking how can it come about that one individual can "hijack" the labour of her offspring?

So I'm not quite convinced that there even *is* an "altruistic" behaviour here that needs to be explained. Quite the opposite: the evolved behaviour that needs to be explained is how the queens coopted their offspring and prevented them from laying eggs of their own.


We already know of course that it's mostly done by pheromones. The point is, selection pressure must act on the queen bee, not on the workers.

We also know that it's not quite true that the queen always directs when and where new future queens are reared. Though it mostly happens that the queen will lay eggs in cells specially marked for queens, at a time of her choosing (she leaves them empty when she doesn't feel that the conditions are quite right), the workers retain a behaviour, used in emergencies such as the premature death of the queen, to feed royal jelly to any larvae and thereby turn them into queens. It's interesting though that the queens so formed are not as large or as vigorous as the normal ones.

It would be interesting to know if the queen can somehow differentiate between the sperm of the males that she has mated with (10 to 25 males, over 2 to 3 days), stored and kept viable in her body for up to three years. Some males might be selected for producing workers, others for producing queens. Just a thought.

Has anyone looked at human menopause in relation to kin selection? I'm not talking about looking at how grandmothers may improve the reproductive success of grandchildren, I know that sort of stuff has already been done. I'm interested in the *timing* of menopause. Now it's clear that menopause happens while the woman still has eggs in her fallopian tubes. It's also known that the fertility of those eggs decreases as she ages. It strikes me that, for any particular woman, there's going to be a declining curve of "potential reproductive success" against age, based on her own egg fertility. There's also going to be a constant, horizontal line of potential reproductive success *for her* based on the reproductive success of her offspring. (OK, it's not going to be exactly a straight line, but as a first approximation...). Now at some point those lines are going to cross, and I think the kin selection people might be on to somethign if they could show that menopause happens precisely when those lines do cross, i.e. precisely when the mother benefits more from her *offspring's* offspring than she does from having more herself.

No comments:

Post a Comment