Genetic diversity and population differentiation in the rare serpentine endemic, San Benito evening primrose (Camissonia benitensis; Onagraceae)

Background/Question/Methods

Conservation of rare species often requires establishing new populations. During a reintroduction, every effort should be made to maintain the limited amount of existing genetic diversity in order to increase the probability of long-term species persistence. However, determining which source material to use during reintroductions depends on the distribution of genetic diversity within and among existing populations. Herein, we investigated the genetic makeup of the Federally Threatened San Benito evening primrose (Camissonia benitensis), a serpentine endemic plant from the Inner South Coast Range of central California. Historically, C. benitensis was thought to exist only on serpentine alluvial sediments deposited near creeks and rivers, but recent field surveys located numerous additional populations on upland geologic transition zones at the edge of serpentine outcrops. The distribution of genetic diversity within and between these two spatially distinct habitat types has yet to be established. Using seven microsatellite markers, we compared genetic diversity in 29 C. benitensis populations with that of its two closest relatives, C. contorta (non-serpentine) and C. strigulosa (serpentine tolerator). Within C. benitensis, evidence of divergence between stream terrace and transition zone habitats was examined. Due to the putatively autogamous mating system (flowers appear to be pollinated in bud) and its potential genetic consequences, we also documented pollinator visitation rates.

Results/Conclusions

Genotyping results of 317 individuals indicate that C. benitensis exhibits moderate levels of genetic diversity, but exceptionally high inbreeding (F_IS = 0.7480) in comparison to its closest relatives (C. contorta F_IS = 0.3614; C. strigulosa F_IS = -0.1826). Despite weak differentiation between habitat types (pairwise F_ST = 0.0433), the genetic data better fit a model with three clusters of individuals (k = 3). There does not appear to be hybridization between C. benitensis and C. strigulosa as there is no genetic overlap between the two co-occurring populations. The levels of inbreeding in C. benitensis are consistent with a primarily self-pollinating mating system, but pollinator observations revealed a diversity of visitors representing four insect orders. Pollinator visitation was highest over five hours after flower opening, yet flowers dehisced pollen onto stigmas in bud. Thus, floral visitors are unlikely to be effective pollinators - consistent with the genetic data estimated from the microsatellites. Reintroduction efforts should maximize the limited amount of remaining genetic diversity, but seeds can be sourced independent of habitat type.