As persistent readers of this blog will know I’ve been working for some time on a book on Joshua trees, and one of the more interesting facets of the Joshua tree’s lifestyle is the tree’s reproductive partnership with two species of moths, Tegeticula synthetica (which works with the western population of trees, these days dubbed subspecies Yucca brevifolia brevifolia) and Tegeticula antithetica (the partner of the eastern subspecies, Yucca brevifolia jaegeriana). The tree can’t reproduce without the moth, and the moth can’t reproduce without the tree.
This partnership, which the Joshua tree has in common with all other Yucca species, is pretty much the canonical example of mutualism, a kind of relationship between two species in which each benefits from the other. The yucca-moth relationship has been cited in biology texts almost since Charles Valentine Riley first described the relationship in the 1870s. Despite the exemplary prominence of the relationship, there wasn’t much detailed study made of yucca moths after Riley until late in the twentieth century, when researcher (and Coyote Crossing reader) Olle Pellmyr took the moths on as a project, accomplishing among many other things the splitting of the sprawling taxon Tegeticula yuccasella into a swarm of 13 species in 1999.
Research continues at Pellmyr’s lab at the University of Idaho, Moscow. This week, researcher Jeremy Yoder published a new paper — with Pellmyr and Christopher Irwin Smith as co-authors — entitled “How to become a yucca moth: minimal trait evolution needed to establish the obligate pollination mutualism.” In the words of the paper’s abstract:
The origins of obligate pollination mutualisms, such as the classic yucca–yucca moth association, appear to require extensive trait evolution and specialization. To understand the extent to which traits truly evolved as part of establishing the mutualistic relationship, rather than being pre-adaptations, we used an expanded phylogenetic estimate with improved sampling of deeply-diverged groups to perform the first formal reconstruction of trait evolution in pollinating yucca moths and their nonpollinating relatives. Our analysis demonstrates that key life-history traits of yucca moths, including larval feeding in the floral ovary and the associated specialized cutting ovipositor, as well as colonization of woody monocots in xeric habitats, may have been established before the obligate mutualism with yuccas. Given these pre-existing traits, novel traits in the mutualist moths are limited to the active pollination behaviours and the tentacular appendages that facilitate pollen collection and deposition. These results suggest that a highly specialized obligate mutualism was built on the foundation of pre-existing interactions between early Prodoxidae and their host plants, and arose with minimal trait evolution.
Yoder et al took samples of about fifty moth species in the Prodoxidae, the family to which yucca moths belong. The samples included individuals from the two yucca moth genera Tegeticula and Parategeticula; from the so-called “bogus” yucca moth genus Prodoxus, whose members lay eggs in yucca fruit without providing pollination services; and from several other related genera of moths which feed on plants belonging to a number of other families. They analyzed the relationships among the moths and built a cladogram (an “evolutionary tree”) that described the lineage of each species with approximate dates of divergence at each branching point. GIven the characteristic morphologies and behaviors of each sampled species and the likely characteristics of their common ancestors, the researchers were then able to estimate when yucca moth behaviors and other traits arose. They found that even before yuccas and yucca moths started working together, the ancestors of today’s yucca moths already had most of the physical traits and behaviors their descendants now use in partnership with yuccas. Aside from active pollination (yucca moths deliberately pack pollen into the flowers’ stigmas instead of transferring it “accidentally”), a behavior unique to yucca moths and fig wasps, and the fascinating tentacles female yucca moths possess with which they perform that active pollination, yucca moths were already pretty much yucca moths before they started hanging out with yuccas.
Jeremy Yoder and I caught up with each other on Twitter in the wake of the Scienceblogs.com debacle last week, and he was kind enough to agree to answer a few questions about the paper.
Jeremy, thanks for agreeing to an interview. Tell us a little bit about yourself.
Let me start by thanking you for your interest in the new paper, and for inviting me to answer questions about it. I’m a doctoral candidate in the lab of Olle Pellmyr at the University of Idaho. I study the evolutionary consequences of species interactions—how predators and prey, hosts and parasites, or plants and pollinators shape each others’ evolutionary history. In my spare time, I do a lot of running — I’m signed up for my second marathon this fall — and I write at Denim & Tweed.
How did you come to decide to work on yucca moths?
Well, the short answer is that I’m drawn to cool natural history stories. The longer answer is that after I finished undergrad, I spent a year as an ecology intern at the Western Pennsylvania Conservancy, doing, among other things, a lot of plant community ecology surveys. In the research I did for that project, as well as reading on my own time, I realized I was really interested in the evolution of species interactions—and one of the interactions that kept showing up in my literature searches was the yucca-yucca moth mutualism. So I wrote to the author of all those papers, Olle Pellmyr, and asked if he wanted a new graduate student, and it turned out he did.
The title of your paper — How to Become a Yucca Moth — is intriguing, and in relatively plain English compared to many such papers. Is there a similarly plain-English answer to the question your title poses? How does a moth species become so closely dependent on a plant partner?
Well, to be honest, no single paper can answer that question! But the results we present offer some significant clues:
First, that the ancestors of yucca moths had probably already been feeding on the developing seeds (the floral ovaries) of their host plants for a long time before they became yucca moths. This suggests that the active pollination behavior, and the physical adaptations to support it, probably evolved to help ensure a food source for the moths’ larvae. Active pollinators are much more efficient than most other pollinating insects—they apply the pollen directly to the flower, and don’t eat any of it—so you could imagine that the plants would then evolve to rely on the moths’ pollination services.
Second, our analysis suggests that the ancestral moths fed on members of the family Rosaceae, which includes roses (of course), but also everything from blackberries to apples. This is interesting because the members of this family used by modern members of the yucca moth family are mostly adapted to cooler, wetter environments—nothing like the places where most yuccas grow. So in order to colonize yuccas, the ancestors of the yucca moths must have had to make a fairly big ecological transition as well. Maybe moving from moist mountainsides to deserts somehow contributed to the evolution of the obligate mutualism.
How old is the partnership between yuccas and their moth colleagues?
We don’t really know how old the relationship itself is, and the clues we have don’t give us a very clear answer. The last phylogenetic study of the yucca moth family (PDF available here) estimated that the genus Tegeticula, which contains most of the yucca moths, is about 40 million years old. On the other hand, a more recent analysis (PDF here) estimated that the age of the genus Yucca is between 6 and 10 million years.
So maybe Tegeticula was around for quite a while before it colonized yuccas; or maybe a new analysis with more data and up-to-date statistical methods would find that the yucca moths originated more recently than 40 million years ago. And neither of these estimates really tell us when the obligate pollination mutualism originated; Tegeticula might have used Yucca for a long time before evolving the mutualism, or it might even have had a similar kind of relationship with a different group of plants before it colonized yuccas.
Your work describes the phylogenetic relationship between the two sister genera of yucca moths — Tegeticula and Parategeticula — and their closest relatives the “bogus” yucca moths, in genus Prodoxus. Bogus yucca moths use developing yucca fruit as nurseries for their larvae, but without providing pollination services to the yucca, in effect parasitizing the relationship betweem yuccas and “true” yucca moths. Before now, my understanding had been that that kind of “cheater” behavior evolved in Prodoxus moths as a way of gaming the existing relationship between Yucca and the “true” yucca moths. However, your work indicates that yucca-moth mutualism arose after true and bogus yucca moths diverged. Do you have any thoughts as to how the common ancestor of true and bogus yucca moths might have made its living, and how cheater behavior might have evolved from that behavior?
That’s a good question. What we see in our reconstruction is that the moths who would evolve into pollinating and “bogus” yucca moths were already eating the developing fruit of their host plants, so the “parasitic” aspect of the way Prodoxus makes a living is really the ancestral condition. That is, Prodoxus only looks like a cheater because mutualists evolved. However, our reconstruction also suggests that Prodoxus has actually changed what part of the plant it uses—the ancestral moths used floral ovaries, and most members of Prodoxus use mature fruit, or the floral stem, or (in one case) leaves.
This might be evolution in response to the origin of the pollination mutualism. We know that yuccas kill off flowers that receive too much damage from pollinators’ egg laying, which helps prevent the pollinators from over-exploiting the mutualism. Maybe Prodoxus evolved to bypass that check on moth behavior by laying eggs in other parts of the plant, or in fruit that’s already well developed.
What struck you as the most surprising result of the work that went into this paper?
I think it might be the finding I just described above—that the Prodoxus habit of feeding in non-floral tissues is probably not the ancestral condition. The other members of the yucca moth family in our analysis use a lot of different parts of the host plant, and I might have said, before the analysis, that feeding in floral ovaries was closely tied to the pollination mutualism in Tegeticula. Now it looks like yucca moths were doing this long before they became mutualists.
What are you working on now?
I’m wrapping up data collection for a study of how yucca moth host preferences might be shaping gene flow between two different forms of Joshua trees—part of an ongoing study of Joshua tree and its pollinators. I describe the latest published results in that system here. That project is the capstone to my dissertation work, which includes the “How to become a yucca moth” study, and some evolutionary theory work; I’m lining up to graduate next spring, and looking at options for postdoctoral positions.