The unique nature of island ecosystems have fascinated generations of naturalists, ecologists, and evolutionary biologists. Studying island systems led to the development of keystone biological theories including: Darwin and Wallace's theories of natural selection, Carlquist's insights into the biology of adaptive radiations, MacArthur and Wilson's theory of island biogeography, and many others. Utilizing islands as natural laboratories allows us to discover the underlying fabric of ecology and evolutionary biology. This dissertation represents my attempt to contribute to this long and storied scientific history by thoroughly investigating two aspects of island biology: 1. the role of island climate in shaping drought tolerance of woody plants, and 2. the absence of mammalian herbivores from insular environments and its effects on woody plant defenses.

These goals were accomplished by quantifying functional trait patterns, seasonal water relations, and plant defenses among closely-related species pairs of chaparral shrubs from matched field sites on Santa Catalina Island and the adjacent Santa Ana Mountains in southern California. This experimental design allowed me to test for

repeated evolutionary divergences across island and mainland environments and to examine the evolutionary trade-offs between traits.

Chapter 1 focuses on differences in dry season water availability and hydraulic safety between island and mainland chaparral shrubs by measuring seasonal water relations and cavitation resistance. My results suggest that island plants are more buffered than mainland relatives from the harsh summer drought conditions that characterize the Mediterranean type climate region of California. Furthermore, island plants exhibit increased hydraulic safety margins that suggest island plants may fare better than mainland relatives during episodes of increasing aridity.

Chapter 2 examines an exhaustive suite of 12 functional traits that characterize the drought-related functional strategies of island and mainland chaparral shrubs. Island plants have more mesomorphic leaf and canopy traits than mainland relatives. However, stem hydraulic traits are surprisingly similar between the island and mainland environments despite large differences in seasonal water relations. The differences between patterns at the leaf and stem levels may be related to the existence of evolutionary correlations for leaf traits but not for stem traits. Multivariate principal component analyses suggest that island plants are employing a very different suite of functional traits than their mainland relatives that allows them to take advantage of the more moderate conditions that characterize the island environment without sacrificing increased vulnerability to drought at the stem level.

Chapter 3 tests the hypothesis that the absence of mammalian herbivores throughout most of Santa Catalina Island's history has selected for plants that are less defended and more palatable than mainland relatives that have experienced more consistent browsing pressure. My results confirm that island plants have fewer morphological

defenses and are more preferred by mammalian herbivores compared to close relatives from the mainland. These findings also suggest that island plants are more vulnerable to browsing by introduced mammalian herbivores. This vulnerability should be taken into account when making management decisions concerning introduced herbivores on islands.

In conclusion, chaparral shrubs on Santa Catalina Island have different levels of drought tolerance and herbivore defenses compared to mainland relatives that affect how they are likely to be impacted by climate change and other anthropogenic alterations of the insular environment. Furthermore, the pattern of evolutionary divergences between island and mainland plants reported in this dissertation offer new insights into how drought tolerance and herbivore defenses are shaped by environmental factors.