Speciation and reinforcement:
Hybridization between closely related species in nature is a common phenomenon. Such hybrids often have low evolutionary fitness because of either genetic incompatibilities between the species pairs or because they are a poor ecological match with the niches in their environment. The ultimate fate of species pairs that come into contact and start to hybridize is a controversial question, with profound evolutionary and conservation implications. There are four potential outcomes to such contact, either 1) one species can drive the other extinct, 2) the species pair can fuse into a single species, 3) a stable condition of a low rate of hybridization can be maintained, or 4) the species can evolve reproductive isolating barriers that will prevent future hybridization.
The last of these processes, called “reinforcement”, is a very intriguing one. In reinforcement, the low fitness of hybrids itself drives the evolution of mechanisms such as conspecific mate choice, allowing the speciation process to proceed by reducing or eliminating gene flow between the species pair. Reinforcement was a very popular idea when it was first introduced in the late 1930s, but became controversial when biologists developed several verbal arguments as to why the process might not work.
One of my main research areas focuses on building mathematical models of reinforcement and related speciationprocesses. To date, these models have demonstrated that there are indeed broad ranges of biological conditions under which reinforcement can operate. Several studies have systematically explored biological factors that would make reinforcement more or less likely. These include how reinforcement is affected by different geographical patterns of gene flow, by chromosomal sex-linkage of genes involved in the reinforcement process, and by different forms of non-random mating and selection against hybridization. Studies have also assessed the way reinforcement may interact with other mechanisms that can drive speciation, such as direct selection on preferences, local adaptation, and conspecific gamete precedence, the preferential usage of conspecific gametes following both con- and heterospecific matings. Additional work has focused on sympatric speciation, a process related in many ways to reinforcement.
I am continuing to develop this body of research along several fronts, including examining the effects of asymmetries in selection against hybrids, comparing reinforcement when males instead of females chose mates, and studying the interaction of reinforcement with condition dependence (with Steve Proulx), the selection of mates with genes for high overall genetic quality. My current work in this area also includes a fusion of speciation theory with another exciting area of research in behavioral and evolutionary biology, the evolution and effects of learning (described below).
My research on reproductive isolation between species has led me to an interest in the evolution of the fundamental mechanisms of mate choice that occur within species. Both the evolution of mate choice and its most important consequence, a type of evolutionary selection called sexual selection, have been major areas of research in evolutionary biology since the earliest days of the discipline. Despite this intense study, many questions remain unanswered. My work on mate choice evolution focuses on two topics, learning (described below) and male mate choice. Both of these topics, while rich areas of research in their own right, tie directly into the speciation projects described above.
Male mate choice is a fascinating topic that is remarkably understudied. It has long been recognized that males generally have more mating opportunities than females, because females invest greater amounts of energy into each reproductive attempt. This leads to the general outcome of females being the choosing sex, while males compete over females. However, males obviously do not have an infinite amount of energy to put into mating. It would thus be logical if males discriminated in some way between females when deciding whom to court. The vast majority of models of sexual selection, however, make the simplifying assumption that males court indiscriminately.
In my research, I have been asking the question of what would happen evolutionarily if males made courtship choices. Interestingly, females often chose males on the basis of traits such as plumage characteristics or songs that have no obvious bearing on the “quality” of the male (so called “arbitrary” traits). Males seldom, however, seem to choose on the basis of similar traits in females; for many species this is why males are ornamented while females are dull. I have therefore concentrated in my initial work on determining why this difference between the sexes exists. My models thus far have provided unexpected, yet logical and very interesting explanations for why male mate choice for “arbitrary” traits should not be able to evolve, even if the genetic variation for this type of choice were present. I have also been able to determine sets of circumstances under which male choice should evolve, including some that tie into my speciation research. Further work in the lab (with Jonathan Rowell) addresses the effects of directed movement on the evolution of male choice. There are many exciting fronts remaining to be researched in this area, which I will continue to pursue.
Although it has long been recognized that both mating signals and mating preferences can be affected by learning, the vast majority of theoretical models of mate choice and sexual selection have assumed that these traits are genetically controlled. There are cases, however, in which learning can have a profound effect on the evolutionary outcome of these processes. Both preferences and signals can be learned in various ways that can be categorized by the individuals that are being learned from (for example, fathers, mothers, or neighbors). In past and current research I have investigated the evolution of preferences to determine whether learned preferences are likely to evolve, and if so, of what type. I plan to continue this work for both preferences and signals, as well as investigating the implications of these results for the process of speciation.
Learning can affect speciation and species maintenance in several ways. Females often discriminate between mates of different species based on learned signals, such as bird song. Mating preferences can also be learned, through a process called sexual imprinting. Learning in these contexts is widespread taxonomically but has been seriously understudied in speciation research. I have conducted several studies examining the effects of learned signals and preferences on the speciation process. The goal of my current and future studies in this area is to build upon this previous work to create a comprehensive picture of the interactions of learning and speciation.
I am broadly interested in the evolution of behavior and other topics in behavioral ecology. Prior and current projects in these areas include the evolution of warning coloration and evolution during brood parasitism. Previous work also includes collaborations on methods of phylogeny reconstruction and species delimitation.