Outcrossing plant species are more likely to exhibit autonomous selfing in marginal habitats to ensure reproduction under conditions of limited pollinator and/or mate availability. Distyly is a classical paradigm that promotes outcrossing; however, little is known about the variation in floral traits associated with distylous syndrome in marginal populations. In this study, we compared the variation in floral traits including stigma and anther height, corolla tube length, herkogamy, and corolla diameter between the central and peripheral populations of the distylous Primula wannanensis, and assessed the variation of floral traits at early and late florescence stages for each population. To evaluate the potential consequences of the variation in floral traits on the mating system, we investigated seed set in each population under both open-pollinated and pollinator-excluded conditions. The flower size of both short- and long-styled morphs was significantly reduced in late-opening flowers compared with early opening flowers in both central and peripheral populations. Sex-organ reciprocity was perfect in early opening flowers; however, it was largely weakened in the late-opening flowers of peripheral populations compared with central populations. Of these flowers, disproportionate change in stigma height (elongated in S-morph and shortened in L-morph) was the main cause of reduced herkogamy, and seed set was fairly high under pollinator-excluded condition. Our results provide empirical support for the hypothesis on the evolution of delayed autonomous selfing in marginal populations of distylous species. Unsatisfactory pollinator service is likely to have promoted reproductive assurance of distylous plants with largely reduced herkogamy mimicking “homostyles.”

No abstract

No abstract

Fingerlings of two different Schizothorax species, S. o'connori Lloyd and S. waltoni Regan, were exposed to copper for 264 h in a series of static toxicity tests. The mortality rates of these two species increased as the exposure concentrations increased and the exposure time was prolonged. Estimated 96 h median lethal concentration (LC50) values were 0.31 and 0.28 mg/L for the two species, respectively. These findings indicated that fingerlings of these species were sensitive to copper, and that both species may be suitable for use as local biomonitors of copper pollution in the high altitude environment. However, S. waltoni may be more useful in indicating the safe concentrations of copper used in aquaculture for therapeutic purposes. Further quantitative studies with measured concentrations of copper are required to verify the results observed in the present study.