Meta-Analytical and Modeling Approaches to Understanding the Connection Between Species Interactions and Diversification

Abstract

The connection between species interactions and macroevolution has puzzled and fascinated biologists for decades. The lack of direct evidence of species interactions over macroevolutionary timescales poses challenges for getting a better understanding of this connection. However, species interactions are often traceable along branches of phylogenetic trees as traits that are associated with species interactions. For example, the evolution of plant-pollinator interactions is traceable along branches of plant phylogenies as morphological traits that are associated with specific groups of pollinators (i.e. pollination syndromes). The evolution of parasitism can be similarly traced across a phylogeny since a parasitic lifestyle is often associated with highly specialized morphological adaptations. Using traits as a proxy for species interactions, I examined two questions: (1) how do different types of interactions, such as mutualism, parasitism, herbivory, commensalism, and competition, affect the rates at which clades diversify? (2) how do species interactions of different outcomes (i.e. mutualistic versus antagonistic) affect the persistence of species interaction over macroevolutionary timescales? In the first two chapters of my dissertation, I took meta-analytic, comparative approaches to these questions, leveraging a wealth of empirical studies containing time-calibrated phylogenies. In my first chapter, I found that species interactions with unilaterally positive fitness effects tend to increase diversification rates, whereas those with unilaterally negative fitness effects tend to decrease diversification rates. In my second chapter, I found that the oldest mutualists are much older than the oldest antagonists, yet this contrast is reversed in animals. Ecological modeling, another line of inquiry towards a better understanding of the connection between species interactions and macroevolution, has seen recent advances that show promising future directions. Starting from a neutral speciation model in which speciation occurs in a metacommunity because of genetic differentiation across space due to dispersal limitation, I examined the effects of coevolution-induced stabilizing and destabilizing selection on species diversification in my third chapter. I found that, when species interactions are assumed to occur between a more dependent party and a less dependent party, coevolution shapes species diversification through two different mechanisms: coevolution-induced destabilizing selection impedes species diversification in the independent clade, whereas coevolution-induced stabilizing selection enhances species diversification in the dependent clade. Overall, the three chapters of my dissertation not only advance the current knowledge on large-scale patterns regarding the connection between species interactions and macroevolution, but also offer a mechanistic explanation for such a connection.