It started with monogamy
November 5, 2015 8:46 AM Subscribe
Phylogenetic analysis (using DNA to figure out the family tree of life) of Hymenoptera (ants, bees and wasps) has shown that eusociality (societies in which the many give up their own reproduction to support the reproduction of the few) first evolved, in every case studied so far, in monogamous species, in which all offspring have the same mother and father.
However, as we learned previously, when all Hymenoptera offspring have the same father, the resulting workers have an incentive to kill the queen.
However, as we learned previously, when all Hymenoptera offspring have the same father, the resulting workers have an incentive to kill the queen.
The paper I linked to is actually 7 years old. I just thought it was interesting. :-)
One thing I've wondered about is herds (e.g. ungulates) where one male is dominant for many years and impregnates most females over that time. Do those herds have a higher level of cooperation as a result of higher levels of relatedness than herds where mating is more evenly spread around?
posted by clawsoon at 3:18 PM on November 5, 2015
One thing I've wondered about is herds (e.g. ungulates) where one male is dominant for many years and impregnates most females over that time. Do those herds have a higher level of cooperation as a result of higher levels of relatedness than herds where mating is more evenly spread around?
posted by clawsoon at 3:18 PM on November 5, 2015
I vaguely recall the Selfish Gene (1976) mentioning that eusociality evolved like 11 independent times, but always in species haploid males, but presumably it's based on much earlier work.
It's logical that a species would become eusocial if it were both monogomous and had haploid males : Sisters are more related to one another than to their own offspring, so they evolve to forgo raising children to farm their mother. There is however a sticking point that not all those species are monogamous, which botches the math.
In this article, they examine the genetics of various non-monogomous eusocial species, proving that they all descend from monogomous ones. Yet another victory for the gene's eye view.
Now nobody has ever explained to me how males evolved to be haploid.
posted by jeffburdges at 4:15 PM on November 5, 2015 [1 favorite]
It's logical that a species would become eusocial if it were both monogomous and had haploid males : Sisters are more related to one another than to their own offspring, so they evolve to forgo raising children to farm their mother. There is however a sticking point that not all those species are monogamous, which botches the math.
In this article, they examine the genetics of various non-monogomous eusocial species, proving that they all descend from monogomous ones. Yet another victory for the gene's eye view.
Now nobody has ever explained to me how males evolved to be haploid.
posted by jeffburdges at 4:15 PM on November 5, 2015 [1 favorite]
Do those herds have a higher level of cooperation as a result of higher levels of relatedness than herds where mating is more evenly spread around?
Ooh, cool question. Here are two references that could be useful to *sort of* address this question: Lung & Childress 2003 and Hand et al. 2014. Specifically the Lung and Childress paper does compare elk herds that likely vary in the degree of relatedness, but not because of differences in the degree to which a single male monopolizes reproduction in the herd, but rather because of differences in the sex composition across herds - I'll get to why this affects relatedness in a little bit.
The Lung and Childress paper looks at patterns of vigilance behavior, which is a fairly well-studied example of cooperation. Perhaps unsurprisingly, vigilance is affected by such factors as actually having offspring in the group and the risk of predation. Figure 5 gets to where they really present data relevant to your question: they compare groups that are composed of mostly females and groups that are composed of mostly males. These groups are likely to differ in relatedness, mainly because of sex-biased dispersal (juvenile female elks are philopatric - they remain with their natal group, while the males usually leave the group -this is often the case in mammals but, interestingly, not in birds). See the Hand et al. paper for genetic confirmation of female philopatry in elk. As you can see in figure 5a in the Lung and Childress paper, the pattern of vigilance is consistent with your prediction - female groups, which should be more closely related to each other, show higher degrees of vigilance. While this pattern is consistent with the general kin selection prediction, I bet some folks could argue for other life-history type differences between males and females that could also account for this pattern and not require a kin-selection mechanism. Still, I'd say the data are consistent with your intuition, though.
Anyway, elk are just an OK system to study these types of questions. Most of the hard-core empirical work where they really collect data that can test the models well goes on with bees - which can range from solitary to facultatively social to eusocial systems. Its not my speciality, but my take is that while there is lots that is still being highly debated (especially with regards to models of reproductive skew or the sharing of the proportion of reproduction), the kin selection predictions are largely borne out - increasing relatedness increases the degree of cooperation.
posted by cnanderson at 7:59 PM on November 5, 2015 [3 favorites]
Ooh, cool question. Here are two references that could be useful to *sort of* address this question: Lung & Childress 2003 and Hand et al. 2014. Specifically the Lung and Childress paper does compare elk herds that likely vary in the degree of relatedness, but not because of differences in the degree to which a single male monopolizes reproduction in the herd, but rather because of differences in the sex composition across herds - I'll get to why this affects relatedness in a little bit.
The Lung and Childress paper looks at patterns of vigilance behavior, which is a fairly well-studied example of cooperation. Perhaps unsurprisingly, vigilance is affected by such factors as actually having offspring in the group and the risk of predation. Figure 5 gets to where they really present data relevant to your question: they compare groups that are composed of mostly females and groups that are composed of mostly males. These groups are likely to differ in relatedness, mainly because of sex-biased dispersal (juvenile female elks are philopatric - they remain with their natal group, while the males usually leave the group -this is often the case in mammals but, interestingly, not in birds). See the Hand et al. paper for genetic confirmation of female philopatry in elk. As you can see in figure 5a in the Lung and Childress paper, the pattern of vigilance is consistent with your prediction - female groups, which should be more closely related to each other, show higher degrees of vigilance. While this pattern is consistent with the general kin selection prediction, I bet some folks could argue for other life-history type differences between males and females that could also account for this pattern and not require a kin-selection mechanism. Still, I'd say the data are consistent with your intuition, though.
Anyway, elk are just an OK system to study these types of questions. Most of the hard-core empirical work where they really collect data that can test the models well goes on with bees - which can range from solitary to facultatively social to eusocial systems. Its not my speciality, but my take is that while there is lots that is still being highly debated (especially with regards to models of reproductive skew or the sharing of the proportion of reproduction), the kin selection predictions are largely borne out - increasing relatedness increases the degree of cooperation.
posted by cnanderson at 7:59 PM on November 5, 2015 [3 favorites]
jeffburdges: I vaguely recall the Selfish Gene (1976) mentioning that eusociality evolved like 11 independent times, but always in species haploid males...
Eusociality is more common among species with haploid males, but it's not limited to those species. The naked mole-rat, for example, is a (fascinating) eusocial mammal. I'm pretty sure the eusocial termites and shrimp don't have haploid males, either, but I'll admit that I haven't investigated them in depth.
Now nobody has ever explained to me how males evolved to be haploid.
That's a fascinating question. I would also like to know the answer!
posted by clawsoon at 6:59 AM on November 6, 2015
Eusociality is more common among species with haploid males, but it's not limited to those species. The naked mole-rat, for example, is a (fascinating) eusocial mammal. I'm pretty sure the eusocial termites and shrimp don't have haploid males, either, but I'll admit that I haven't investigated them in depth.
Now nobody has ever explained to me how males evolved to be haploid.
That's a fascinating question. I would also like to know the answer!
posted by clawsoon at 6:59 AM on November 6, 2015
IANA biologist, but I'd vaguely understood that eusociality in termites, etc. were substantially different from the species that evolved the haploid male based eusociality. I know some termites have multiple queens for example. Also, termite queens live extremely long lives, which sounds relevant, i.e. they're more very long lived insects with shorter lived slave offspring. Amusingly, I suppose the Formics from Ender's Game sound more like termites than like bees, etc.
posted by jeffburdges at 8:11 AM on November 6, 2015
posted by jeffburdges at 8:11 AM on November 6, 2015
Sorry--didn't click the article, that's egg on my face! :) I kind of just assumed "well, yeah, this is boring" but then I work in integrative biology and have a passing interest in eusociality, and I even wrote a (unpublished) synthesis paper on the evolution thereof in undergrad. Kind of failed to consider that maybe not everyone had picked up those papers when I had.
Eusociality is perhaps best studied and most common in the haplodiploid species of the Hymenoptera, but it's by no means only something that happens in those species. Off the top of my head, there are also eusocial species of gall thrips, naked mole rats, snapping shrimps, termites (which are, as jeffburdges points out, not haplodiploid*), aphids, and parasitic flatworms. (No, really!) Most of those aren't haplodiploid, which has cast quite a bit of doubt on the strong association between haplodiploidy and the evolution of eusociality that existed in evolutionary biologist thought for a long time.
One thing I've wondered about is herds (e.g. ungulates) where one male is dominant for many years and impregnates most females over that time. Do those herds have a higher level of cooperation as a result of higher levels of relatedness than herds where mating is more evenly spread around?
This also depends heavily on the species. For one thing, in many of those--especially ungulates!--harem compositions tend to be pretty random, because they're composed basically of whatever females the male can keep herded into a small band without said females wandering off. And those are composed of whatever females happen to be in a particular territory at the beginning of the breeding season. Ungulates (especially those which are solitary for part of the year, like deer, or which live and breed in large groups, like zebra) tend not to, in my experience, have very specific bonds between individuals. So while you do (as cnanderson points out) see things like differing levels of vigilance among herds depending on relatedness, you tend not to see altruism in ungulates which is as strong as you see in social animals which have more stable relationships.
On the other hand, in species like lions or some primates where the group of females is more specific and less prone to changing, there's generally a lot more obvious cooperation and helping. But that's usually distributed along matrilines more than anything else--groups of females who pay attention to their mothers and sisters (via that mother) and daughters and aunts, and preferentially give help to members of the same matriline over females from other matrilines in the same group.
I have no earthly idea about how the evolution of haplodiploidy works, but I agree it's a very interesting question!
*Although there's this weird system of genetic linkage going on with termites that led people to think that maybe they were maybe kind of haplodiploid for a long time, before people started really delving into eusociality into species outside of the Hymenoptera.
posted by sciatrix at 9:38 AM on November 6, 2015 [2 favorites]
Eusociality is perhaps best studied and most common in the haplodiploid species of the Hymenoptera, but it's by no means only something that happens in those species. Off the top of my head, there are also eusocial species of gall thrips, naked mole rats, snapping shrimps, termites (which are, as jeffburdges points out, not haplodiploid*), aphids, and parasitic flatworms. (No, really!) Most of those aren't haplodiploid, which has cast quite a bit of doubt on the strong association between haplodiploidy and the evolution of eusociality that existed in evolutionary biologist thought for a long time.
One thing I've wondered about is herds (e.g. ungulates) where one male is dominant for many years and impregnates most females over that time. Do those herds have a higher level of cooperation as a result of higher levels of relatedness than herds where mating is more evenly spread around?
This also depends heavily on the species. For one thing, in many of those--especially ungulates!--harem compositions tend to be pretty random, because they're composed basically of whatever females the male can keep herded into a small band without said females wandering off. And those are composed of whatever females happen to be in a particular territory at the beginning of the breeding season. Ungulates (especially those which are solitary for part of the year, like deer, or which live and breed in large groups, like zebra) tend not to, in my experience, have very specific bonds between individuals. So while you do (as cnanderson points out) see things like differing levels of vigilance among herds depending on relatedness, you tend not to see altruism in ungulates which is as strong as you see in social animals which have more stable relationships.
On the other hand, in species like lions or some primates where the group of females is more specific and less prone to changing, there's generally a lot more obvious cooperation and helping. But that's usually distributed along matrilines more than anything else--groups of females who pay attention to their mothers and sisters (via that mother) and daughters and aunts, and preferentially give help to members of the same matriline over females from other matrilines in the same group.
I have no earthly idea about how the evolution of haplodiploidy works, but I agree it's a very interesting question!
*Although there's this weird system of genetic linkage going on with termites that led people to think that maybe they were maybe kind of haplodiploid for a long time, before people started really delving into eusociality into species outside of the Hymenoptera.
posted by sciatrix at 9:38 AM on November 6, 2015 [2 favorites]
Great synthesis paper, sciatrix, thanks!
posted by clawsoon at 12:37 PM on November 6, 2015 [1 favorite]
posted by clawsoon at 12:37 PM on November 6, 2015 [1 favorite]
« Older “Houellebecq’s name is so rich with associations... | N is for Neo-Otyugh Newer »
This thread has been archived and is closed to new comments
posted by sciatrix at 2:40 PM on November 5, 2015