Anole biologists may take for granted that anoles compete. We have plenty of evidence, after all, for the phenomenon of interspecific competition — when comparing populations in sympatry vs. allopatry, we often see reductions in fitness, shifts in resource use, or trait evolution to minimize resource use overlap. It may surprise you, then, that we actually have relatively little insight into the mechanisms of competition in Anolis: what resources do anoles actually compete for (e.g. food or space), and how (e.g. exploitation or interference)? If you don’t believe me, pull out your copy of Lizards in an Evolutionary Tree and read page 2291.
Understanding mechanisms is important because it allows us to make predictions for the outcomes of species interactions in different contexts. I have long wanted to know whether green anoles (Anolis carolinensis) are likely to persist in Hawaii following the introduction of brown anoles (Anolis sagrei) and gold dust day geckos (Phelsuma laticauda). My initial work in this system was directly out of the classic Anolis playbook: enclosure experiments looking at traits, resource use, and fitness under controlled resource availability in allopatry vs. sympatry2,3. In other words, I was focused mainly on documenting the phenomenon of competition. At some point it clicked that in order to answer the question I cared about the most — the potential for long-term coexistence — I needed to identify the mechanisms of competition, and how species differences in competitive ability may translate into coexistence in this system.
Our latest experiment was grounded in mechanistic theory for consumer-resource interactions developed by Tilman4, R* theory. The key insights from this body of theory are that exploitation competition comes down to who can suppress shared limiting resources to the lowest levels under different conditions, and that the ability to suppress resources should be predictable from consumer traits. Designing experiments from this perspective caused me to shift from measuring resource use overlap (i.e. diet, perch height, body temperature), to measuring prey populations, lizard demographics, and lizard foraging rates.
While R* theory is foundational, general, and from the golden age of ecology — when there was great optimism about the potential for simple theory to explain community dynamics — empirical tests of R* predictions have previously been limited to microbes, microfauna, and grassland plants. This is probably because it is a ton of work to collect the necessary data, and we only had a chance at success in a system of vertebrate, mobile predators because of the many decades of work on anoles. Our core field team (myself, Spencer Alascio, and Jose Carranza), aided by a large cast of volunteers, spent multiple days in the field every single week for over a year. 29,000 arthropods measured5, 3,000+ mark-recaptures of lizards6, and over a 100 hours of focal lizard observations later7, we got our answer.
We previously found that the species differ in their clinging ability on rough vs. smooth substrates, and here we found that this results in different species being better at catching prey in different microhabitats (ground vs. vegetation). We all like to do what we’re good at, and indeed brown anoles and day geckos each had the highest prey capture rates in the microhabitat where they cling the best. These differences in microhabitat-specific performance led to spatial differences in competitive ability: brown anoles drove prey densities on the ground down to lower levels than any other species, and day geckos drove prey densities in the vegetation down to lower levels than any other species. Even though green anoles had similarly high capture rates to brown anoles on the ground, and to day geckos in the vegetation, they did not suppress prey the most in either microhabitat because they are the worst at converting prey into new offspring. As a result, green anoles achieve densities only half as high as the other two species under the same environmental conditions and in the absence of interspecific competition. This insight underscores the value of rooting empirical work in theory—we would have never measured conversion rates without the guidance of a theoretical framework.
In addition to helping us design an effective experiment, the models allow us to achieve my ultimate goal: predicting whether green anoles can persist long-term. By using our empirical data to parameterize consumer resource models for mobile predators a la Vincent et al.8, co-author Kyle Edwards was able to generate testable predictions for species coexistence under different ratios of ground and vegetation habitat. What we found, by graphical analysis of zero net growth isoclines, is that while any two of the species can coexist under some conditions, green anoles cannot persist on the landscape in the presence of both of the other species under any habitat ratios. As all three species are now distributed island-wide9, our models predict that green anoles will go extinct. This prediction is of course specific to the context in which we collected the empirical data to parameterize these models. I look forward to testing these predictions in the next enclosure experiment, and in the natural experiment playing out in real time on the landscape.
Overall, our results support the idea that trade-offs in the ability to suppress prey across different microhabitats, which we refer to as spatial heterogeneity in competitive ability, is a likely coexistence mechanism in this system.
