Diploid hybrid speciation in Penstemon (Scrophulariaceae).

Abstract

Hybrid speciation has played a significant role in the evolution of angiosperms at the polyploid level. However, relatively little is known about the importance of hybrid speciation at the diploid level. Two species of Penstemon have been proposed as diploid hybrid derivatives based on morphological data, artificial crossing studies, and pollinator behavior observations: Penstemon spectabilis (derived from hybridization between Penstemon centranthifolius and Penstemon grinnellii) and Penstemon clevelandii (derived from hybridization between P. centranthifolius and P. spectabilis). Previous studies were inconclusive regarding the purported hybrid nature of these species because of a lack of molecular markers sufficient to differentiate the parental taxa in the hybrid complex. We developed hypervariable nuclear markers using inter-simple sequence repeat banding patterns to test these classic hypotheses of diploid hybrid speciation in Penstemon. Each species in the hybrid complex was genetically distinct, separated by 10-42 species-specific inter-simple sequence repeat markers. Our data do not support the hybrid origin of P. spectabilis but clearly support the diploid hybrid origin of P. clevelandii. Our results further suggest that the primary reason diploid hybrid speciation is so difficult to detect is the lack of molecular markers able to differentiate parental taxa from one another, particularly with recently diverged species.