While behavior determines how organisms interact with their environment and may “drive” evolution and local adaptation in novel habitats, regulatory behaviors may actually inhibit evolutionary change. When populations inhabit different thermal environments, thermoregulatory behaviors may reduce expected differences in body temperature, and mediate selective pressures on thermal physiology. Huey et. al (2003) termed this the “Bogert effect” and predicted that divergent selection should be weak, and gene flow high, across thermal gradients for thermoregulating species, whereas selection should be strong, and gene flow low, for thermoconforming species. In this study, I use targeted Sanger sequence data and genome-wide single nucleotide polymorphisms (SNPs) to investigate population genetic structure and gene flow in Anolis cristatellus, a thermoregulating species that inhabits a range of thermal environments on Puerto Rico. I find multiple genetic populations of A. cristatellus, some of which may correlate with distinct thermal environments. Furthermore, I demonstrate uni-directional gene flow between populations that span a thermal gradient, a result that has important implications for thermal adaptation and that is only partially predicted by the “Bogert effect”. These results are discussed with regards to adaptation and selection on thermal physiology, as well as to alternative “non-physiological” models.