Mayr (1947) has presented a strong case against the hypothesis of sympatric speciation in the course of supporting the universal hypothesis of geographic (or at least microgeographic) speciation. In the summary (p. 286) of that paper he made the following statement: "It is unproven and unlikely that reproductive isolation can develop between contiguous populations." Later, Grant (1949), in a discussion of pollination systems as isolating mechanisms, presented a model for sympatric speciation in angiosperms dependent on the known flower constancy of bees. This model indicated the means by which an important mutation in flower color or form might become established within a population, thus resulting in time in a new and well isolated species to which internal isolating mechanisms are unnecessary. In passing, the latter author suggested that hybridization might also provide an example of sympatric speciation. It is the purpose of the present paper to further develop this last hypothesis on the basis of a realistic model suggested by recent studies of hybridization and floral isolation in a group of Penstemon species. It is not the purpose of this paper to refute the general conclusions reached by Mayr (I.c.) on the importance of geographic speciation, but merely to indicate that sympatric speciation may be less unlikely than previously considered. The word, species, is here used in its biological sense.