Non-random dispersal can promote evolutionary divergence, even in the absence of spatial variation in selection. However, in most systems we lack the empirical data on movement behavior needed to assess this mechanism of divergence. Capitalizing on movement data for two stream-associated plethodontid salamanders, we tested the hypothesis that non-random dispersal promotes genetic and phenotypic divergence in upland, headwater areas – hotspots of plethodontid diversity.

Results/Conclusions

Gyrinophilus porphyriticus and Eurycea bislineata show contrasting patterns of non-random movement along streams: movement is upstream-biased in G. porphyriticus and downstream-biased in E. bislineata. Consistent with predictions of how these biases interact with stream slope to affect divergence, genetic distance increased with slope in G. porphyriticus and decreased with slope in E. bislineata over a standardized distance of 1 km along six streams in the Hubbard Brook Watershed, New Hampshire. At the same small scale, phenotypic divergence in relative trunk length was positively related to genetic divergence in both species. Our results underscore the value of direct data on movement for predicting population connectivity in complex landscapes, and the importance of addressing dispersal explicitly in studies of genetic and phenotypic differentiation. This study also provides novel insight on the contribution of non-random movement behavior to fine-scale evolutionary divergence.