Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Ecology and Evolutionary Biology

Major Professor

Daniel Simberloff, Christine R. B. Boake

Committee Members

David Buehler, Sergey Gavrilets, Louis J. Gross


A central tenet of the competition paradigm is that community structure is governed by deterministic rules. The competition paradigm pervades nearly all subdisciplines and extends to the broadest, deepest questions in ecology. To determine whether patterns of co-occurrence, nestedness, and morphology in avian communities are consistent with a competition hypothesis, I use null models to compare observed patterns to patterns expected in the absence of competition.

I use presence-absence matrices of birds in three archipelagoes to test whether species exhibit exclusive distributions. Congeneric birds co-occur significantly less frequently than predicted in two archipelagoes, consistent with a competition hypothesis. However, when examined separately, most genera do not exhibit patterns that differ from random expectations. Furthermore, species differences in habitat preference and barriers to dispersal within archipelagoes are two alternate hypotheses that can explain many exclusive patterns and are supported by available data. Distributional evidence alone does not implicate competition. I argue that the range of conditions under which competition is likely to generate exclusive distributions across islands is narrow.

Although nested patterns are viewed as the converse of competitive exclusion, meta- analyses of presence-absence matrices simultaneously support the ubiquity of exclusive distributions and indicate that most ecological systems are nested. I show why these apparently contradictory patterns are not mutually exclusive. Patterns of nestedness can result from multiple processes; without understanding mechanisms, nestedness analysis is unlikely to prove useful for conservation.

I study algorithms that randomize presence-absence matrices. Two commonly used algorithms generate biased statistical distributions and should be abandoned.

I test for community-wide character displacement in wing length and talon size among Accipiter hawks. I also assess whether subspecific variation in wing length is predictable from community composition. I find limited evidence of community-wide character displacement for wing length, but talon size, which relates directly to resource use, shows larger minimum and more equal size ratios than predicted. I find no evidence of competitively driven change in wing length. These findings demonstrate the importance of examining traits directly related to function. I cannot determine whether community-wide character displacement of talon size results from ecological character displacement, species assortment, or both.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."