Evolution 2010 Day 2 (June 26th)

For some reason the Oregon Convention Center features this large pendulum (perhaps a metaphor for academic movements?)

I spent the morning of the first full day of Evolution 2010 in a variety of talks, most of which were themed on sociality. Surprisingly, a great number of these talks tackled the difficult and controversial topic of group selection. I think that it is becoming increasingly clear that it is not as easy to mechanistically explain social behaviors without invoking a level of selection above that of the individual.

In a talk entitled “Why harm kin?”, David J. Van Dyken considered the problem of explaining siblicide in the face of kin selection. The basic problem is that kin selection models of the type first suggested by Hamilton assume that benefits will accrue from the reproductive success of one’s close relatives, so it is difficult to understand why an organism would kill its siblings. I think that this is an interesting question although the use of Hamilton’s model is a bit confusing because in the case of kin selection we assume that individuals are incurring costs in order to benefit their relatives, whereas in siblicide the individual’s cost is associated with the siblicidal effort and the benefits come from eliminating competition. Still, Van Dyken managed to use the Price Equation to create a modified model that explained how to incorporate the special case of siblicide into the larger picture of kin selection within groups. Although I would need a little more time to fully understand his conclusions, he suggested that the critical question is whether kin are competing for resources before or after dispersal. When they compete before dispersal, they are likely to compete directly, and their interaction is analogous to that of others involving interference due to resource competition.

Erol Akҫay presented another modification of the Price Equation in his talk entitled “Group optimality, behavioral responses, and kin selection in the evolution of cooperation”. Akҫay’s main focus was on the question of how behavioral responses affect the outcome of multilevel selection models. While it is well-known that high relatedness favors group adaptive behaviors, apparently there has been less work on the role of behavioral responses. Based on what I gathered from the talk, behavioral responses have to do with how individuals respond to the behavior of other group members. In this fairly simplistic model, this response can vary from exact copying to no response to the display of an opposite or contrary behavior. When copying behavior is prevalent, an effect similar to that of high relatedness makes the evolution of group-adaptive behaviors more likely. To me this sounds similar to the “tit-for-tat” strategy explored in the prisoner’s dilemma, but this connection was not made in the talk so I am not sure if it is appropriate.

Beyond these theoretical approaches there were some interesting experimentally-based talks that explored the role of cooperation in groups. Maxence Salomon presented a great talk called “Functional morphology of social groups: a case study with group living spiders”. In this talk he described differences between small and large species of the genus Anelosimus, which both build large collective webs designed to capture flying insects. These two species seem to share a very similar niche despite having very different morphology and behavior. What Salomon’s experiment showed (by providing webs in the field with bee prey) was that both individual foraging success (an individual-level trait) and the web architecture (a group-level trait) vary, and whereas the larger species relies more on speed of capture by fewer individuals (and less on web architecture), the smaller species relies heavily on its web architecture and the collective effort of individuals to subdue prey. This suggests that the niche overlap between the two species may in part be due to their use of different levels of reliance on cooperation, with differing strategies yielding similar rates of prey capture success.

Joao Muñoz-Durán presented a talk about hypercarnivory in canids (“The genome – sociome loop of mutual affectation and the evolution of sociality and hypercarnivory in canids”) that simultaneously confused and intrigued me. In explaining the evolution of the hypercarnivorous lifestyle of canids, Muñoz-Durán invoked the idea of the “sociome”. I have never heard of this term, and based on his brief talk I still can’t quite grasp what it means. So if I didn’t understand the the talk why mention it? [Please note that I sat through many more talks than I chronicle here, and the omission of some stems completely from my own ignorance and lack of conceptual understanding of the subjects presented]. The reason I am interested in this talk is because it touched on some of my favorite subjects. First, Muñoz-Durán allied his talk with others that involved higher-level selection (his term of choice is “hierarchical selection”, which I think has some strong semantic value). Second, he describe the sociome as an analog to the genome. Finally, he seemed to be trying to explain the evolution of hypercarnivory in canids as product of interactions between the sociome and the genome. A number of ideas intrigue me here, even if they were not rigorously tested. First is the idea that we ought not to think too granularly about the genetic elements that evolve. In invoking the “genome”, Muñoz-Durán seems to be highlighting the importance of considering the “community of genes” that determine the success or failure of particular genes, a common objection to the selfish gene theories forwarded by Maynard-Smith, Hamilton, and ultimately Dawkins. If the sociome is a cultural analog of the genome, perhaps using is presents a similar critique to Dawkins’ meme theory? [Muñoz-Durán did cast the idea of the sociome in contrast to the extended phenotype]. But while this combination of ideas is interesting, there was little science done to generate predictions that could be tested in extant or extinct canids. For those of us with suspicions that some of the most particulate and genetic approaches don’t fully explain the evolutionary process, it is critical that we generate testable hypotheses, not just interesting hypotheses.

One of the fascinating things to note about the group of talks above is that they all use different terms for selection above the level of the individual. A surprising number of people used the term “group selection”, but this tended to be a “point of reference” term used to bridge the audience’s understanding to the presenter’s preferred term-of-choice. People are more likely to have heard about group selection but also more likely to have heard bad things about it, so perhaps the term “group selection” is now being used to mean “naive group selection” with the implication that more modern terms are of the advanced non-naive type. This terminological diversity is probably indicative of the fact that although there is a lot of renewed interest in this line of thinking, we still haven’t come to coherent agreement on what we want to call this kind of selection.

There were numerous talks on sexual selection on this and subsequent days, and I tried to get to as many of these as possible. Melvin Bonilla presented a talk (“Cryptic male choice: males strategically allocate sperm in response to female exposure to previous males”) which showed that male pseudoscorpions may be capable of lowering the degree to which females subject them to sperm competition by picking up on olfactory cues on females that indicate previous matings. Like many other arthropods, pseudoscorpion males provide females with a spermatophore, which sets up the potential for sexual conflict because females can hold onto the spermatophore provided in the first mating and then using subsequent matings to subject males to sperm competition. For this reason, males have the incentive to be somewhat choosy, preferring females that don’t smell like they have already mated with other males.

Another example of sexual conflict came from Jennifer McCreight, who discussed the nature of copulatory plugs in Kangaroo rats (in her talk “The potential for sperm competition and sexually antagonistic coevolution in banner-tailed kangaroo rats (Dipodomys spectabilis) revealed by DNA extracted from copulatory plugs”). Copulatory plugs are produced by males as the terminal component of their ejaculate, and benefit the male by preventing other males from successfully mating with the female. While good for the male, this “chastity belt” may not be good for the female, who might benefit from increasing the diversity and/or quality of her offspring through subsequent matings. McCreight wanted to understand the morphology and function of these plugs, so she sectioned plugs found in wild females and sequenced genetic material in each section. What she found was that many of the plugs actually include DNA from multiple males, which suggests that the effectiveness of these plugs as a male strategy may be mitigated by the behaviors of females or other males.

In case you were worried that sperm competition will be rampant anywhere in nature where is it logistically possible, Jeanette McGuire presented a talk on painted turtles entitled “Stored Sperm, Iteroparity, Multiple Paternity, and Reproductive Success of Male Painted Turtles (Chrysemys picta marginata)”. These turtles certainly have the potential for sperm competition: females store sperm for years, which ought to encourage them to dabble in competing males against each other. But what McGuire found was a strong trend towards single-male paternity, even in multiple clutches laid by the same female. Where multiple males fathered clutches, it was usually in older females.

Beyond my most-favored subjects (cooperation and sex), I also heard a bunch of good talks on different topics. I couldn’t follow his entire presentation, but James Hunt presented an interesting talk on how founding female Polistes wasps give birth to different pulses of offspring (first workers, then reproductive males, then potentially-reproductive females). I am interested in becoming more familiar with all these different systems, as they offer diverse insights into the evolution of sociality.

Emily Moriarty Lemmon presented a really compelling example of pre-zygotic reproductive isolation along a hybrid zone of frogs in her talk “Reinforcement in chorus frogs: Lifetime fitness estimates including intrinsic natural selection and sexual selection against hybrids”. Through experimental crosses and lab observations, she was able to show that females reject males with hybrid calls, effectively reinforcing pre-zygotic isolation generated by clinal variation in environment.

In his talk “Do fishery collapses reduce genetic diversity?”, Malin Pinsky presented an provocative meta-analysis of the diversity of over-fished species. Although fishery collapses are dramatic and have serious economic consequences, in terms of conservation they are not directly analogous to collapses in terrestrial animals. Over-harvested terrestrial animal populations often collapse to miniscule numbers (tens or hundreds of individuals according to Pinsky), and their specific habitat needs can isolate remaining populations. Such conditions lead to a very small effective population size, the reduced genetic diversity of which can put the population at great risk for extinction. Captive breeding programs are premised on the need to actively manage genetic diversity in the face of small population abundances. In fisheries economic extinction is very different than biological extinction: a species that is not worth harvesting due to low abundance may in fact still be comprised of many individuals. This reality, paired with the assumption that marine species are less isolated by habitat fragmentation, has led some to suggest that low genetic diversity is not a problem in collapsed and (presumably) recovering fisheries. But what Pinksy’s meta-analysis showed through pairwise comparison of closely-related species (one overfished, one not) is that diversity is in fact in decline within collapsed fishery species. Although he did not suggest a mechanism for this decline his work suggests that marine species censused population sizes may be much larger than the effective population size, a trend that has serious implications for the conservation of these species.

The final talk of the evening was the American Society of Naturalists presidential address, given by Jonathan Losos. These presidential addresses are interesting, because they simultaneously empower the current president with a bully-pulpit and burden him with presenting something that represents the mission of the society as a whole. Losos took this charge bravely, presenting ideas that were bound to challenge the members of his society. As the title of his talk (“Seeing the Forest for the Trees: Limitations on the Use of Phylogenies in Comparative Biology”) suggests, his focus was on phylogenies and their use in current-day evolutionary biology. He started by acknowledging the great contribution that phylogenetic methods have made since they were introduced in the mid-1980’s, most notably by Joseph Felsenstein. Prior to the introduction of rigorous methods for inferring historical evolutionary relationships, systematics was what Losos described as a “scientific backwater”. The use of statistical methods allowed for more reliable tree generation; these methods, paired with the emerging science of gene sequencing, ushered in the current era of systematic biology wherein phylogenies have become an indispensable tool to understanding the evolutionary process. While Losos does not disagree with the importance of this change, he has come to the conclusion that “the pendulum has swung too far”. Instead of being confined to their valuable uses, phylogenies are now being overused and over-interpreted in a manner that could be leading the field down some dead ends. He appropriately reminded the audience that phylogenetic trees are the result of likelihood methods, making every tree the “most probable” reconstruction of evolutionary history. This means (of course) that a certain percentage of all “most likely” trees will in fact be the wrong trees. For this reason we need to be very careful about how we interpret their results. There are also a number of invalid inferences that can be drawn from even the most reliable tree. First, we need to be careful about inferring a “phylogenetic effect”, as convergent evolution can create the illusion of homology. This suggests that “deep traits” (those held in common by extant species sharing a very ancient common ancestor) are reliable indicators of a “phylogenetic effect”, whereas “shallow traits” which have evolved more recently are not. Losos also took to task the contemporary use of the term “phylogenetic effect”, pointing out that shared phylogeny alone is not the cause of evolution, but rather an indication that common selective forces have shaped a clade.

Evolution meeting participants come and go from the picnic at the Oregon Zoo

Following Losos’ talk, most of the participants in the meeting headed over to the Oregon Zoo, located across the river and at the top of a large hill, for a picnic. This was a really nice touch, as it was a way of informally fostering new friendships: I ended up meeting several really interesting people at the picnic and whilst wandering around the Zoo, which was closed to all but Evolution participants.

Mountain goats on display at the Oregon Zoo