Variation in Social Structure, Dispersal, and Genetic Differentiation in Response to Local Habitat Quality in a Metapopulation of the Eastern Collared Lizard

Date of Award

Summer 8-15-2013

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Evolution, Ecology & Population Biology)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Predicting population response to changes in climate or habitat quality depends upon understanding which habitat features will significantly affect evolutionary forces likes gene flow or genetic drift. Features that are closely associated with social behavior have the potential to significantly impact both.

Here the hypothesis that the degree of clustering of resources within a deme was an honest signal of environmental potential for polygyny was tested using six focal populations of the Eastern collared lizard (Crotaphytus collaris collaris). Populations subsisting on habitat with greater clustering of exposed bedrock exhibited social network structures with higher potential for monopolization than populations on evenly distributed habitat. Males on clustered habitat also exhibited greater variance in aggressive displays.

Evaluating the potential for variation in local social structure to influence patterns of dispersal and gene flow is made difficult by the assumption underlying most landscape ecology tools, that local populations are, for all intensive purposes, homogenous. The relationship between within-glade social structure and dispersal was tested for each glade on Stegall Mountain using mark/recapture records. A metric for glade quality with respect to social behavior was created based on the six glade focal study, and a method was developed for measuring it from aerial photographs. This expanded the analysis from six populations to 132. Habitats with high potential for polygyny produced fewer dispersers.

Finally, a dispersal model was developed that incorporates both variation in local social structure as well as variation in quality of the intervening matrix. The predicted dispersal patterns and resistance distances were compared with observed networks of collared lizard dispersal and observed levels of genetic differentiation to test which models most closely resemble observed dispersal behavior. Incorporating both landscape and local habitat into predicted dispersal models seems to be most closely correlated with patterns of genetic differentiation.


English (en)

Chair and Committee

Alan R Templeton

Committee Members

James M Cheverud, Fiona Marshall, Jane Phillips-Conroy, David Queller, Joan Strassmann


Permanent URL: https://doi.org/10.7936/K7959FH8

This document is currently not available here.