Evolution Highlight: Ornamentation and display behavior in Poecilia

The Evolution Highlights series highlights some of the interesting and varied papers published within the last few years in Evolution. The goal of these Evolution Highlights is to let our readers learn more about how the highlighted study came into existence, and to invite the authors to share stories and tips from the perspective of a recently published author. We welcome nominations and self-nominations for the Evolution Highlights. Find out how to submit a paper here.

In love and war: The morphometric and phylogenetic basis of ornamentation, and the evolution of male display behavior, in the livebearer genus Poecilia.
Goldberg, D.L., Landy, J.A., Travis, J., Springer, M.S. and Reznick, D.N.

https://onlinelibrary.wiley.com/doi/10.1111/evo.13671

Abstract:

Exaggerated male traits under sexual selection are often used for both competition and courtship, raising the question of whether ornaments evolved simultaneously for both functions, or if use in one context preceded use in another. Here, we apply a phylogenetic approach to study the evolution of ornamental dorsal fins in male poeciliid fish of the subgenera Mollienesia and Limia, which exhibit convergent development of an enlarged dorsal fin, and often direct erect‐fin displays to male and female conspecifics. Unlike prior categorical assessments of poeciliid adornments, we measure dorsal fin exaggeration with a continuous index of ornamentation. Phylogenetic logistic and generalized least squares regression analyses indicate that high index values are significantly associated with the use of two component postures of courtship and aggressive displays, dorsal fin erection and body curvature, but not with the presence of sexual dichromatism. Male displays initially evolved for male–male aggression in the common ancestor of Mollienesia and Limia, suggesting that this signal originated for competition, then became co‐opted for courtship. These results support the armament‐ornament hypothesis for evolution of exaggerated male traits, and are consistent with an evolutionary shift in the predominant mechanisms of sexual selection from intra‐ to intersexual.

Evolution Highlight by Daniel Lorenz Goldberg
School of Biological Sciences, Illinois State University

What gave you the idea for this study?

It was our senior author David Reznick who gave me this idea, when I started as a Master's student in his laboratory at the University of California Riverside, after I had been working there for two years as an undergraduate. He wanted to know if there was a relationship between male behavioral traits and the size of the male dorsal fin in mollies and Limia, and wanted someone to do a phylogenetic study of the evolution of these traits. It seemed like too good of an idea to pass up, so we quickly rounded up our UCR systematics expert, Mark Springer, who agreed to help me create the phylogenetic tree. Next on board were David Reznick's colleagues at Florida State University, Joseph Travis and his student Alex Landy, who were instrumental in devising the index of ornamentation that was the key takeaway and the real selling point of this study.

What was the big question you were trying to ask and why was it exciting?

We wanted to know whether male dorsal fin displays in mollies and Limia -- specifically two behaviors, the sigmoid S-shaped body posture, and dorsal fin erection -- originally evolved for the purpose of courting females or competing with other males in dominance displays. In other words, was the trait co-opted from its original function for use in a second, additional function; and did this co-option precede extreme trait exaggeration in a few species? (In our study, these were the sailfin mollies and the humpback Limia). This was a question that had seldom been tackled since the armament-ornament hypothesis was originally proposed in the mid 1990s, and only one prior study had been done in poeciliids before ours. So we were really in uncharted waters here, pun intended.

In what ways does this study expand or build on your previous studies?

The aforementioned poeciliid study had found that dark vertical bars flashed by male swordtail fish had originated as a mating display but then became co-opted as an aggressive display. This was, however, the opposite pattern than the one predicted by the original armament-ornament hypothesis, and we wanted to determine if mollies and Limia fit the pattern of traits used in aggression becoming larger and more flamboyant as sexually-selected courtship signals.

From your perspective, what was the most interesting or intriguing aspect of this study?

The most intriguing aspect of the study was the fact that behavioral repertoires of many mollies and Limia are not well documented in the literature. I had to sift through a variety of livebearer hobbyist magazine articles, old dissertations from past Reznick lab members, and Wischnath's Atlas of Livebearers with its color photographs. Three species that we obtained novel behavioral information from had never been observed in captivity before our study.

Were there any unexpected logistical or technical challenges you had to address?

We went without data for Pamphorichthys for the longest time, even though those five species fill the phylogenetic gap between mollies and Limia, because there were simply very few collections of them that we could obtain size measurements from. Finally, in the summer of 2016, Dr. Reznick got a couple of colleagues in Brazil to measure their museum specimens of Pamphorichthys to give us the missing size values. The sad thing is, the Pamphorichthys specimens were housed at the National Museum in Rio de Janeiro that later burned up in a massive fire, so I suspect that those alcohol-preserved fish may now only exist as measurements in our study and perhaps a few others.

Did your work take you to any interesting locations? If so, what were they like?

I myself did not need to venture away from UCR to conduct this research, but some of the live fish we observed in captivity came from interesting locations. Manfred Schartl, who sent us never-before-seen videos of the Liberty molly, a species that I determined possessed courtship, had his fish shipped live from Germany to the United States!

Tell us something about collecting these data that people might not know or think about.

It is much easier to obtain dorsal fin measurements from anesthetized live fish (which Alex Landy did using the Ptacek collections at the University of Clemson) than it is to obtain the same measurements for preserved museum fish (which I did for most of the other specimens). The formalin and ethanol really dry out the fins and make them relatively brittle, so you cannot spread them fully. Luckily, all I needed was the length of the longest fin ray, so I did not end up damaging the specimens to collect this information.

Did you have to learn any new analytical techniques for this study?

Absolutely! All of the systematics software that I used to create the phylogenetic tree was new to me, so Dr. Springer had to train me in how to operate those programs.

What was the most challenging part of analyzing your data?

The MatLab software that I used to run the phylogenetic logistic and least-squares regressions was fickle for its long processing times and very specific code that had to be entered to run the programs. I can't tell you how many times I had to rerun those programs and reenter data values to do so!

What was your writing process like?

The manuscript was adapted from my Master's thesis, which was over twice the length that the final journal article ended up reaching. I had to trim down the thesis and add some new data, which took me most of the summer of 2016, and then I had to complete this on and off over the next few years, given that I was working with a totally different study system for my Ph.D. research in Illinois.

What was the biggest challenge about writing the manuscript?

Keeping it at the proper length! There was a lot I wanted to communicate and a limited amount of words that I could use, so I went through several revisions of having to take out and replace certain ideas, especially in the Discussion, but it all came together well in the end.

How did writing up this study compare to previous paper writing experiences?

This was by far the longest article that I had written to date, so it was definitely a labor of love.

What was the review / revision process like?

I was amazed that the editors thought the article had a decent chance right off of the bat, and it only took a few submissions before they finally accepted it, which I am eminently grateful for. Most of their suggested changes were easy to make. The biggest challenge came with their request for a concise but detailed description of the ornamentation index and how we calculated, for which I ultimately got Joe Travis to help write a short version of its rationale. Shortly after that, the editors accepted our manuscript.

Do you have any writing tips for others?

The literature is your friend, period. Read a lot! Collect a motley assortment of journal articles (via Zotero, for example) and save them in folders on your computer based on their topic. That way, you can refer back to when you are writing about a certain topic. If I hadn't scoured the literature for studies testing the armament-ornament hypothesis, I never would have realized that this hypothesis was the perfect way to frame our study of ornament evolution in mollies and Limia.

Was this study collaborative with other groups? If so, what was valuable about that experience?

It was collaborative with Dr. Reznick's Brazilian colleagues, as well as a variety of molly researchers and aficionados who I reached out to. It was a valuable experience because I got to know people who I would later meet in person at academic conferences!

Looking back, what is one thing you learned from doing this study (other than your main findings) that you’ll use in future projects?

Using phylogenetic software and making the molly and Limia family tree really whetted my appetite for systematic studies, particularly those focused on the evolution of animal behavior. The NCBI Taxonomy database and the program it interfaces with, Geneious, were so helpful in my eyes that I went back to use them for one of my Ph.D. dissertation projects!

What do you think will be the lasting impact of this study?

It serves to offer an example of the most species-rich group of animals to date that show co-option of an armament into an ornament, and furthermore illustrate how the gain of a display function for sexual selection can go hand-in-hand not only with extreme trait exaggeration, but also rapid evolution of many species in a group that is now experiencing increased sexual selection.

What was the most difficult part of this study and / or the most rewarding?

The most rewarding part of this study was getting it published! Not every Master's student can achieve that with his or her research, and I feel very blessed that Evolution allowed me to publish my work in such a prestigious journal. I don't mean to boast, but it was also rewarding to get what I think is a genuinely original idea out there into the science sphere!

How would you explain your study to a five-year-old?

There are these fish called mollies and Limia, and males of these fish have big fins on their backs. But we didn't know why their fins are so big. We compared many different fish and found that those with the biggest fins use their fins not only to appear attractive, but also to fight each other! And the most interesting thing of all is that the fins appear to have first been used for fighting, before they were also used by some of these fish to look pretty.

Courting Limia species