The transition from outcrossing to self-fertilization is one of the most common evolutionary changes in plants, yet only 15 percent of flowering plants are considered highly selfing. To explain this phenomenon, Stebbins (1957) proposed that selfing may be an evolutionary dead end. According to his hypothesis, transitions from outcrossing to selfing are irreversible, and selfing lineages suffer from an increased risk of extinction due to a reduced potential for adaptation. I will present results from ecological, genetic and genomic studies using the self-compatible genus, Collinsia, highlighting conditions in the wild that favor selfing and provide evidence that selfing may drive species divergence and be reversible.