Altruistic punishment by fig trees?
As I was getting ready to write about some of the talks at the Applied Evolution Summit, I received a very interesting paper: Host sanctions and pollinator cheating in the fig tree - fig wasp mutualism, which was recently published in Proceedings of the Royal Society by Charlotte Jander and Allen Herre.
Fig-tree fruits are lined with many little flowers. Female wasps crawl inside to lay their eggs, often carrying pollen from the fig where they, themselves, hatched from an egg. Different fig species host different wasp species. Some wasp species are like many other pollinators, carrying pollen only by accident; fig trees pollinated by these species have to make lots of pollen. Other figs are pollinated by wasps that actively collect pollen and actively pollinate flowers inside fig fruits; these fig species can make less pollen, which frees resources to make more seeds.
But there is presumably some cost to the wasps of transporting pollen. Why not save this cost, travel light, and lay eggs in a fig fruit without pollinating its flowers? This is essentially the same question people in my lab have asked about rhizobia, the bacteria that provide legume plants with nitrogen: once rhizobia have reproduced inside a root nodule, why stick around and invest resources in pulling nitrogen out of the atmosphere and converting it to a form the plant can use?
In each case, as shown in our previous work and in this week's paper, the host plant imposes "sanctions" that reduce the reproductive success of partners (wasps or rhizobia) that fail to keep their end of the bargain.
Through painstaking work described in the paper, Jander kept a bunch of wasps from acquiring pollen. (Similarly, Toby Kiers and Bob Rousseau, in my lab, kept rhizobia from acquiring nitrogen, by flushing root nodules with a nitrogen-free, argon-oxygen atmosphere.) She let wasps with or without pollen lay eggs in figs and compared their reproductive success.
When the fig species Ficus citrifolia was pollinated by a pollen-free wasp, a very large fraction of the fruit were aborted by the tree, killing any wasp eggs inside. Even when fruit weren't aborted, very few of the eggs laid by pollen-free wasps hatched into adult wasps. So this species imposed very strong sanctions against what they called "cheaters." (I accept the term, assuming that there is some cost to the wasps from carrying pollen or distributing it, which seems likely but may not have been demonstrated yet.)
Passively pollinated fig species didn't impose such sanctions, whereas other actively pollinated fig species imposed weaker sanctions. In these figs, that is, the reproductive success of wasps without pollen wasn't reduced as much, relative to those with pollen. Across the five actively pollinated fig species compared, there was a strong negative correlation between the frequency of pollen-free "cheater" wasps and fig-tree sanction strength.
So, which was cause and which was effect? I wouldn't expect cheaters being rarer to select for stronger sanctions, so presumably the effect acted in the opposite direction: strong sanctions by figs imposed greater selection against cheating by wasps. For this to be an evolutionary effect, the tendency not to carry pollen would have to be at least somewhat heritable. Under this hypothesis, cheaters arise occasionally in all actively pollinating wasp species, via mutations that change wasp behavior or perhaps physical traits that affect the ability to carry pollen. In species where the host tree imposes strong sanctions, the cheaters quickly die out, while they may persist (at low to moderate levels) in species subject to less-severe sanctions.
But that raises the question hinted at in my title for this week's post: if cheaters are so rare among the pollinators of F. citrifolia, why has natural selection maintained such strong sanctions? Sure, the evolutionary effects of host sanctions benefit this fig species as a whole, by preventing cheaters from becoming more common over generations. But natural selection usually depends on benefits to individuals or close relatives, not to species as a whole. Are F. citrifolia trees more likely to be found near close relatives than other fig species are? If so, then a fig-tree gene for punishing cheaters could, through its effects on the evolution of wasps, benefit other copies of that gene in related fig trees nearby. Of course, a tendency of related trees to cluster together also makes them more likely to compete with each other, which can undermine this effect.
The whole paper is well worth reading, for additional details (e.g., on differences among species in the extent to which cheaters may escape sanctions by sharing a fig with pollinators), for perspective on the challenges of field research (some data were lost when a tree fell down), and for the way in which different statistical tests were compared to see if they led to the same conclusions. All in all, this is one of the most interesting papers I've read in the past year.
Another cool story on dealing with difficult pollinators was discussed recently in Not Exactly Rocket Science.