HOW MUCH OF THE VARIATION IN ADAPTIVE DIVERGENCE CAN BE EXPLAINED BY GENE FLOW? AN EVALUATION USING LAKE-STREAM STICKLEBACK PAIRS

• Published in: Evolution Vol. 58; no. 10; pp. 2319 - 2331
• Main Authors: Hendry, Andrew P, Taylor, Eric B
• Format: Journal Article
• Language: English
• Published: United States Society for the Study of Evolution 01.10.2004; Oxford University Press
• Subjects: Adaptation; Adaptation, Biological; Animal populations; Animals; Body Weights and Measures; British Columbia; Cluster Analysis; dispersal; Ecological genetics; ecological speciation; Estimation methods; Evolution; Evolutionary genetics; Fish; Fresh Water; Freshwater fishes; Gasterosteus aculeatus; Gene flow; Genetic diversity; Genetic Variation; Genetics, Population; Geography; Gills - anatomy & histology; Inlets; Lakes; Microsatellite Repeats - genetics; microsatellites; migration; natural selection; Phenotypic traits; population mixing; REGULAR ARTICLES; Selection, Genetic; Smegmamorpha - anatomy & histology; Smegmamorpha - genetics; Species Specificity; Streams; Watersheds; British Columbia
• ISSN: 0014-3820; 1558-5646
• DOI: 10.1554/04-376

How much of the variation in adaptive divergence can be explained by gene flow? The answer to this question should objectively reveal whether gene flow generally places a substantial constraint on evolutionary diversification. We studied multiple independent lake-stream population pairs of threespine stickleback (Gasterosteus aculeatus). For each pair, we quantified adaptive divergence based on morphological traits that have a genetic basis and are subject to divergent selection. We then estimated gene flow based on variation at five unlinked microsatellite loci. We found a consistent and significant pattern for morphological divergence to be positively correlated with genetic divergence and negatively correlated with gene flow. Statistical significance and the amount of variation explained varied within and among traits: 36.1–74.1% for body depth and 11.8–51.7% for gill raker number. Variation within each trait was the result of differences among methods for estimating genetic divergence and gene flow. Variation among traits likely reflects different strengths of divergent selection. We conclude that gene flow has a substantial effect on adaptive divergence in nature but that the magnitude of this effect varies among traits. An alternative explanation is that cause and effect are reversed: adaptive divergence is instead constraining gene flow. This effect seems relatively unimportant for our system because genetic divergence and gene flow were not correlated with ecologically relevant habitat features of lakes (surface area) or streams (width, depth, flow, canopy openness).