Abstract
Over the last decade we have pushed the boundaries of evolutionary biology, discovering adaptive responses at extremely short time and small spatial scales. I aimed to contribute to the growing body of literature by examining whether selection is effective enough to repeatedly generate trait variation in heterogeneous environments despite panmixia. I studied a natural population of Fundulus heteroclitus inhabiting a highly heterogeneous habitat for which subtle but significant genetic divergence among residents of distinct environmental patches has been shown. I build on this previous work by demonstrating significant divergence in complex physiological traits among individuals residing in dissimilar microhabitats. This trait divergence seems to be consistent with the local environmental conditions and is unlikely due to phenotypic plasticity. Using a genomic sampling approach to quantify allele frequency shifts across two time points, I detected aggregate signals that were most parsimoniously explained by polygenic selection among microhabitat residents. Nevertheless, extensive whole-genome sequencing replicated across two years did not reveal convincing patterns of genomic divergence among microhabitat residents nor of allele frequency shifts that could be attributed to divergent polygenic selection. Genomic variants associated with the diverged traits did not show evidence of being under selection, suggesting selection is unlikely to be responsible for the observed trait differences. In light of the evidence, I cannot support the hypothesis of selection driving significant trait divergence among F. heteroclitus inhabiting distinct microhabitats. However, the observed trait variation among microhabitats is of substantial magnitude and could have considerable impacts on eco- evolutionary dynamics. Ignoring such variation among microhabitats could restrict our ability to predict evolutionary outcomes and define meaningful, adaptive divergence in heterogeneous habitats.