Repeated patterns in the evolution of size in island plants

For reasons not fully understood, animals often evolve predictably on islands. For example, radiations of large, flightless birds are a common element of many island biotas. However, our understanding of how plants evolve on islands is comparatively poor. Further, an investigation into the evolution of island plants could help resolve unanswered questions about island animals. This thesis investigates insular size changes in a range of plant functional traits. First (Chapter 2), I explored size changes in 9 species of vines that have colonized islands from the New Zealand and Australian mainland. I asked whether leaf–stem allometry prohibits leaves and stems from evolving independently from one another. Island populations consistently produced larger leaves than did mainland populations. Moreover, changes in leaf size were not associated with concomitant changes in stem size, suggesting that trait allometry does not govern trait evolution on islands. Next (Chapter 3), I asked whether plants obey the infamous island rule, a putative trend in island evolution wherein small animals become large on islands and large animals become small. I demonstrate that plant stature and leaf area obey the island rule, and seed size does not. My findings illustrate that the island rule is more pervasive than previously considered, but that support for its predictions vary among plant functional traits. Third (Chapter 4), I demonstrate that the island rule results from evolutionary drift along bounded trait domains. The island rule has long been hypothesized to result from a suite of selective pressures. Applying my model to island plants, I show that evolutionary drift is the most parsimonious explanation for the island rule pattern. Finally (Chapter 5), to explore insular patterns in leaf size evolution, I conducted a large-scale, macroevolutionary analysis of leaf size on 98 of New Zealand’s offshore islands. Leaf gigantism was emblematic of island populations, and was most prominent in taxa with variable leaf morphologies on the mainland. Further, leaf gigantism was greatest in populations inhabiting old, distant islands, suggesting that time since divergence is a direct predictor of morphological differentiation between mainland and island populations. Overall, this thesis reveals novel patterns, and helps disentangle the distinct roles of natural selection and drift, in the evolution of plant form and function on islands. Finally, this thesis illustrates how investigating the changes in plant traits can help identify the evolutionary mechanisms operating on islands