NATURALIST SELECTIONS IS AN INTERVIEW SERIES PRODUCED BY THE AMERICAN SOCIETY OF NATURALISTS GRADUATE COUNCIL. WE SHOWCASE GRADUATE STUDENT AND POSTDOC AUTHORED WORK IN THE AMERICAN NATURALIST, A PREMIER PEER-REVIEWED JOURNAL FOR ECOLOGY, EVOLUTION, AND ANIMAL BEHAVIOR RESEARCH. CATCH UP ON EXCITING NEW PAPERS YOU MAY HAVE MISSED FROM THE JOURNAL, AND MEET SOME TRULY BRILLIANT EARLY CAREER NATURALISTS!

*Featured image photo credit: Rebecca Robinson

In this episode, Zach Wood talks with us about his new paper Wood et al. 2022: ‘Drivers and Cascading Ecological Consequences of Gambusia affinis Trait Variation.’ We dive into the weird world of human-created pond ecosystems and how these ponds afford a unique opportunity to study rapid contemporary eco-evolutionary dynamics in predator-prey interactions. We also discuss the challenges of dissecting all the possible causes of trait variation in the wild and explore other ecological and genetic contexts that might affect varying patterns of anti-predator evolution among mosquitofish ponds. Listen to our conversation and then read Zach’s full paper here: https://www.journals.uchicago.edu/doi/10.1086/717866.

Craving more context? Email Zach at zachary.t.wood@maine.edu!

Listen on Spotify and Apple Podcasts!

Credits

Featured Guest: Dr. Zach Wood, University of Maine

Host, Editor, Producer: Sarah McPeek, University of Virginia, US

Original Music: Daniel Nondorf, University of Virginia, US

Transcript:

You’re listening to Naturalist Selections, an interview series featuring graduate student and postdoc-authored work in The American Naturalist, produced by the American Society of Naturalists Graduate Council. I’m Sarah McPeek and today I’m talking with postdoc first-author Zach Wood about his new paper, “Drivers and cascading ecological consequences of Gambusia affinis trait variation.” Zach completed this project during his PhD work with Dr. Michael Kinnison at the University of Maine. Zach’s work with Dr. Kinnison and collaborators tackles the causes and consequences of predator-mediated trait variation in introduced mosquitofish. They traveled across the western United States collecting mosquitofish from ponds with voracious bass predators, ponds with slightly less voracious bluegill predators, and ponds with no predators. Then they raised these fish either in the presence or absence of bass predator cues to tease out the predator-mediated plastic, genetic, and evolved plastic components of morphological and behavioral trait variation in their fish populations. They then introduced these mosquitofish to pond mesocosms containing a wide diversity of invertebrate and algal prey to further measure how predator-mediated trait variation and fish population density affect the composition of mosquitofish’s own prey community. Most interestingly, they find that mosquitofish from different source ponds exhibit non-congruent plastic and evolutionary responses to the same predators. Fish from some populations exhibited strong behavioral responses such as higher flightiness in response to predator cues or lower feeding rates, while fish from other populations that also have predators exhibit strong shifts in morphology such as upturned head angles or longer tail lengths. In our conversation, Zach and I plunge deep into the complexity of these eco-evolutionary patterns and muse over other contexts that may matter for anti-predator evolution.

Sarah

I’m really curious what these mosquitofish ponds look like out in California and why you think that mosquito fish are a really good model for studying these ecological cascading effects that you’re interested in?

Zach

Yeah, that’s a good question. Mosquitofish are perceived as this panacea for vector control. So meaning if you have an area that has mosquito borne diseases, mosquitofish by the name are one of the first things that are thought of as a remediation agent. So you dump in these fish that I mean, they must be good at eating mosquito larvae. They’re called mosquitofish, and that’s a really popular practice in big parts of the United States. And so these mosquitofish are relative newcomers. Some of these populations have only existed for maybe a decade, and a lot of the other fish in these ponds are also newcomers. So there are bass and bluegill and other fish that are not native to these ponds and have also been introduced. So in terms of the community, it’s really what we’d call ecologically a novel community. It’s a lot of strange bedfellows. And so this creates a lot of really intense interactions because mosquitofish are voracious consumers of just about anything invertebrate. So whether it’s benthic chironomid larvae or zooplankton, mosquitofish are just sort of like a vacuum that takes them down. And the predators of mosquitofish like bass and bluegill also tend to be just as aggressive, which has given them their invasive label. So they’re very strange bonds in that regard, and that basically everything above the invertebrate level is often nonnative. So some of these are barely bigger than a pool, like a swimming pool, and others are what you’d think of as like, okay, that’s a little Lake. And they really range from kind of marshy, very green coastal ponds to some ultra-clear desert springs. So I’m thinking one of the ponds that we work in in this paper in Eastern California desert is like, yes, the size of my living room, Crystal clear water.

Sarah

Oh, wow.

Zach

Yeah. Oh, there’s golf courses. Some of these are golf course ponds. So they’re like, as green as you can imagine. So ecologically, they vary quite a lot, too. And I’m sure that has something to do with some of the results that we’ve seen that we’ll talk about later.

Sarah

Sure. Yeah. So when you are selecting your ponds to pull fish from in these experiments, were you trying to maximize that ecological variation or were you trying to lump ponds into three different categories. So is that something that you were focusing on when you were designing the experiment?

Zach

Yeah, we were mostly limited by the different predator regimes. So trying to find ponds that just had bluegill but not bass, ponds that just had bass but not bluegill, and ponds that had neither. And of course, then there’s the more logistical concerns, like access, just being able to get in there and not having to fight a housing association or something to get into the pond in someone’s backyard. That can be tricky because some of these variables, like predator presence, are correlated with other environmental variables in the pond, like water quality but also proximity to humans. So we found through our work that the closer pond is to an urban area, the more likely it is to have those bass and bluegill introduced predators. And, of course, there’s this whole suite of other environmental impacts that are associated with being close to urban areas. So there’s definitely a lot of environmental variation we couldn’t control for in these source ponds coming into the experiment.

Sarah

Yeah. That leads very well into my next question, which is about the experimental design itself. I was blown away by it when I read it. I thought it was so elegant for the kinds of questions that you were trying to address and disentangling what are the causes of this trait variation? But you were also able to ask questions about what you called congruence or noncongruence of these adaptations to predator regimes. And you also ask questions about the ecological effects of these trait changes. So did you have all these questions in mind when you were setting this up with your lab, or were there some things that you found in the data that allowed you to ask those questions?

Zach

Yeah, I think this may be surprising, but we did design the experiment to test for all those different factors you talked about, and this wasn’t some spark of inspiration. We’ve got it. We’ll measure all these things once. This is what a year of friendly arguing amongst a bunch of collaborators interested in different things gets you for some scope and just a great example of how long science takes. This paper just came out in 2022, but we started designing this project in the fall of 2016, and then we did the fieldwork in summer of 2017, and then I finally started doing the sample processing, the data analysis, and grappling with the results in 2018. And then it was the writing review back and forth after that. Very long process. But no, we were interested in everything at once, but not just in this experiment has to do everything. It was like if we don’t measure this as well, we might miss something important. Like if we measure the effects of traits of an evolving species on the community. So measure the effect of mosquitofish traits on these pond mesocosms, and we don’t compare it to the effects of mosquitofish density change, then we don’t know how big it is. So we have to add density in. And if we don’t have population level replication, we won’t know how generalizable our findings are. And if we don’t separate genetic effects and plastic effects, we won’t know what is evolution and what’s not. The experiment as it looks. It’s really just us almost giving up, actually. All we have to do it all. There’s no way around it.

Sarah

Yeah. That is something that I picked up a lot in this paper, too, is just the amount of work that you guys had to do to disentangle all of these effects, both with measuring them in the field and in the lab and statistically. And I’m wondering if you think that that is the future of disentangling complex these eco-evolutionary patterns. I think there are a lot of well established patterns that come out in your data, like the strong effects of density on lower trophic levels and the combination of plastic versus genetically evolved trait differences that we see. So do you think that the only solution to understanding these complex patterns is to measure everything, or is there anything that we can do to compensate for that effort?

Zach

Yeah, it’s very tempting always at the end of the paper to say, oh, you just have to measure everything, and you have to do it exactly the way we did it, of course. But I don’t think that’s always necessary. I think probably in a new system where you’re trying to uncover the predominant dynamics and just, like, see what is happening, you probably have to measure a lot. But there are different applications of this kind of work where you don’t necessarily need to know everything. So if you just wanted to know the effects of traits on the environment, you don’t necessarily need to untangle genetic and plastic change because they’re happening at the same time. And if you are just interested in the effect of involving species on one particular taxa, like maybe you’re trying to manage a threatened or endangered species and you just care about evolution of its competitor, you don’t initially have to measure the whole ecosystem per se. But if your goal is to trace the effect of genetic evolution through a whole food web. Yeah, hate to break it to you got to measure everything.

Sarah

Yeah, I think there’s a lot of value to that in that we, as human observers, have certain ideas about how these patterns would play out. But you’re absolutely right that you could miss something really crucial by not taking a certain measurement, not accounting for a certain factor that could really be driving patterns in your data.

Zach

Yeah, I think so. And we’ve been burned by that before, which is why we designed the experiment this way.

Sarah

How so?

Zach

Well, we have figured out in mosquitofish, that density and traits, of course, interact when evolution is occurring. And the same processes that give you new traits can often make the population more adapted, which can change its density. And we found that that’s happening at the same time: the density change and the trait change, and in some cases, they even mask each other. So we really wanted to tease apart those two, so we wouldn’t get burnt by that. Again, same thing with the teasing apart genetic and plastic effects. So we’ve for a long time focused on just the trade outcomes in mosquitofish evolution. So the total genetic and plastic contributions to phenotype and evolutionary biologists hate that. We got a lot of pushback that we needed to figure out what was genetic and what wasn’t. So we built that into this experiment as well for that reason.

Sarah

What for you was most surprising about your results?

Zach

Yeah. I think the non-congruence element was what’s really cooking my brain, and it still is something I’m really grappling with and trying to figure out why, because there’s so many different reasons we could have seen this non-congruence. And the non congruence is just when you have what you think is the same important selective agent in the population, but they exhibit very different trait responses as a result. That one is. I’m still grappling with that.

Sarah

I think you did a great job in your discussion laying out different possibilities, like other ecological factors of the ponds that could be contributing. One that really stuck out to me that I think as an ecologist, I probably don’t often consider enough is the genetic background of these ponds. And so if there’s a lot of genetic variation in behavior, but not much for morphology, you could get very divergent responses. I thought that was really interesting.

Zach

Yes. And this is next on our menu of things we want to try and mosquitofish, but we don’t have great genetic data on these populations. But we do know, as we said in the paper, they go through some wild bottlenecks where the populations can be absolutely hammered in the wintertime, and then I sort of think of it as breathing. Right. They expand out to these big populations in the summer and then they shrink down every winter, and that’s a pretty strong bottleneck to create over and over and over again. So that could potentially have some effect on limiting genetic diversity in mosquitofish. But yeah, if I had to speculate heavily, speculate on what I think is the explaining factor here, I do think it’s the pond environment. I think that anti predator evolution is just such a more complex, multifaceted thing than we give it credit for, that I think elements of the local environment really play a big role, if I were to guess so, like, hiding is a really feasible defense in some of these ponds, particularly the ones that are really weedy.

Sarah

Sure.

Zach

Just hide in the weeds and freeze or hide in the shallows and freeze. But some of these ponds, like I said, they’re like crystal clear swimming pools. Like where is a fish going to hide? So you have a very different potentially adaptive anti predator response in that kind of environment, which might be crypsis or being able to burst swim really fast to get away from a predator that’s chasing you. And, of course, there’s all these other elements, like food availability and like the sensory environment, water clear. There’s just so much going on. Looking back it seems almost naive to think, oh, well, it’s just whether there’s bad predators or not that will shape all the evolution. Nope!

Sarah

Yeah. We thought we measured it enough, but clearly we didn’t. This is going to be really hard for you, but if you could only pick one ecological factor of the ponds to measure, what would you want to look at first?

Zach

Oh, that’s a good question. I think it would be water clarity for a few reasons. I think that it varies a lot pond to pond, so it could be driving some of these dynamics. I think it’s role with the sensory evolution in the mosquitofish and in their predators is potentially strong, and it’s impacted a lot by humans. And that just generally interests me. And should probably interest all of us. So that’s what I would measure if I had to pick one thing.

Sarah

Are bass and bluegill visual predators where water clarity could strongly affect their ability to catch prey?

Zach

I believe so. I know in these mosquitofish ponds they are. And that may have a lot to do with the small size of these ponds compared to the Lakes. We typically think about where there’s not necessarily these huge spaces between individuals. I know this is a sad story, but I know in some of our field collection efforts, we’ll be waiting in the shallows trying to net mosquitofish, and you startle a mosquitofish, like 2ft out of the shallow region, a bass just, boop, thank you. Eats it and swims away. They’re there. They’re watching.

Sarah

Wow.

Zach

Yeah, it’s like okay, that’s creepy.

Sarah

That is freaky!

Zach

One time we were sampling in Death Valley, in the golf course pond in Furnace Creek. It’s like 120 degrees. We’re netting mosquitofish out of the one pond in this otherwise very dry area, and there’s, like, a dozen bass just lined up like cows at a dairy farm waiting. Oh, yeah. Scare those mosquitofish out.

Sarah

You might be selecting for bluegills and bass who learn when scientists are visiting and wait for the mosquitofish then.

Zach

I hope not! But it does make me think that hiding and habitat choice and vision all have a really strong effect. This might not be something we think of initially, but there is this, like, pond edge refuge, especially in a pond that has a fairly low angle of slope on the bottom, there’s almost this little teeny angle. And mosquitofish are really not tall fish. They can just get right up in there and a bath can’t physically get in that shell of the water. Sometimes they just hang out there. So that’s another example of how, like, the specific benthic makeup of a pond can potentially affect what’s an adaptive anti predator trait and what isn’t.

Sarah

That’s really cool.

Zach

Yeah. Not for the fish.

Sarah

Not for the fish, no. But for us studying them, it’s a great opportunity. Given what you told me at the beginning of our talk about how recently a lot of these ponds are established, I think it’s really striking how much the variation that you saw in fish seemed to be genetically based. So what does that tell you about the scale on which these eco-evolutionary dynamics can occur, especially the timescale of how rapidly these effects can change the makeup of a pond?

Zach

Right. Yeah, this is such a good question. In general, there’s so much work showing that fish especially can evolve to be very different in just a handful of generations. I’m thinking of some of the work on fisheries induced evolution that if you have a heavy side selective harvest, it only takes two or three generations for you to have almost no body size overlap in the size distribution of mature fish. So it’s like, oh, these fish are half the size that they used to be. And we’ve also seen this in hatchery domestication and we’ve seen this in mosquitofish as well where ten to 20 generations after bringing fish into a hatchery they don’t look like wild fish. Their fin placement, their head shape, their body size looks different and their ecological niche changes too as a result. This is something I think about a lot is how the traits of an organism line up with their niche. So I think that these dynamics can potentially happen a lot faster than we might think. I think this is a reoccurring theme in contemporary evolution is that we’re reminded how contemporary it really is and yeah, it definitely is happening much faster than we think.

Sarah

I think that’s been one of the main insights of contemporary eco-evolutionary dynamics Is that these processes of trait evolution and ecological dynamics are occurring on the same time scale.

Zach

Yeah, definitely.

Sarah

I think historically we would have thought that plasticity would be a much faster way to change your traits than genetic evolution because we’re used to thinking about it as this long, slow, gradual process. But maybe that’s not what we understand anymore.

Zach

Yeah, in fact, one of our running jokes in our lab is like, oh, we know the one thing Darwin was really wrong about. There’s some quote in origin of the species only after the long hands of time heavy lapse. Nope, not anymore. And yeah, definitely. Evolution is contemporary with ecology and contemporary with human dynamics and human disturbance. So, you know, it not only matters for how we perceive natural systems but it matters for how we perceive our own impacts on those systems.

Sarah

Yeah, absolutely. Well, I think your paper has had a really positive impact on me and on a lot of people who read it and think about the complexities that are present in their system and other things that they might want to think about. When you’re understanding these equal evolutionary effects, It was a really well done study. I know you guys put years of arguing and effort into it.

Zach

Indeed.

Sarah

I think it definitely paid off. So Congratulations.

Zach

Thank you so much.