We test the hypothesis that differences in selfing ability across populations result from differential selection from pollinator service. Although this hypothesis is commonly invoked to explain variation in selfing ability, this study is one of few to address the hypothesis by correlating pollinator service and selfing ability across populations. We examine relationships between multiyear estimates of pollinator service, autogamous selfing ability, outcrossing rates, and inbreeding depression across 11 populations across the geographic range of Collinsia verna. The specific goals of this study were 1) quantify the degree of genetic variation in selfing ability across populations and identify floral traits that are correlated with selfing ability, 2) test the prediction that selfing ability is negatively correlated to pollinator service across populations, and 3) measure outcrossing rates and inbreeding depression across populations and test the prediction that selfing ability is negatively correlated to these parameters. We found evidence that differentiation of selfing ability is adaptive. However, pollinator service was not correlated to outcrossing rates, as expected. Selfing ability, combined with pollinator service, may determine outcrossing rates if selfing interferes with the ability of plants to outcross. Because pollinator service was variable in the two years of this study, we suggest that within population variability may affect differentiation of selfing ability across populations.