A Study Of Adaptive Significance Of Distyly In Primula Species

Distyly is a genetic dimorphism in which plant populations are composed of two morphs that differ reciprocally in the heights of stigmas and anthers in hermorphrodite flowers. The adaptive significance of this type of heterostyly has been widely studied since Darwin (1877) first proposed a functional hypothesis of promoting the legitimate pollen transfer and improving the pollination accuracy. Todate, three different but closely related hypotheses have been proposed to explain the adaptive significance of such reciprocal herkogamy:facilitation of legitimate pollination; avoidance of sexual interference between pollen and stigma within the same flower; prevention of pollen wastage by reducing the amount of incompatible pollen deposition on stigmas. Facilitation of legitimate pollination includes both avoidance of self-pollination and promotion of intermorph pollination. Although studies have supported distyly as a mechanism to promote disassortative pollination, data from stigmatic pollen load under nature conditions and experiment conditions did not always support this view. In this study, we investigated the pollination biology of six Primula species to further explore the adaptive significance of distyly.The genus Primula is a typical example of distyly which adaptation has been investigated by Darwin. Many Primula species distribute in the Hengduan Mountains, southwest China, its species richness and abundant flowering facilitate our field study during 2010 to 2013.Firstly, we investigated the floral traits and pollen loaded on stigmas in six Primula species, the effects of floral traits on the disassortative pollination, to determine the potential factors influencing pollination accuracy. Our results indicated that legitimate pollen ratio loaded on stigmas was low in both morphs (pin 9.56% and thrum 33.6% on average) especially in pin flowers; suggesting that the reciprocal herkogamy of distylous species seemed unlikely to promote the disassortative pollination, not in accordant with the three hypotheses of adaptive significance. In addition, our correlation analysis showed that stigmatic pollen load was correlated with several floral traits, such as stigma height, anther height, anther length, floral size, stigmatic surface size, pollen production. For example, higher pollen production associated with more illegitimate pollen grains loaded on stigma, and the greater stigmatic surface size associated with more illegitimate pollen grains loaded on stigma.Secondly, we tested what factors affecting pollination accuracy in distylous species through flower manipulations, for example, we emasculated flowers to detect the source of illegitimate pollen grains, to examine whether different in pollinator body size has different pollination efficiency through estimating pollen removal and deposition by a pollinator single visit. Earier studies suggested different sources of illegitimate pollen grains, including most of the illegitimate pollen grains on stigma in distylous species might come from self-pollination, but the other studies assumed that a sizeable proportion of illegitimate pollination came from geitonogamy. In this study we found that the illegitimate pollen grains came from self-pollination, geitonogamy or intramorph as well as the same morph of different plants. The pin pollen had longer transmission distance than thrum pollen grains, and this may be caused by pollinator behavior.Pollinators had significantly influence on reproductive success. In this study, we designed experiments to compare the effect of major pollinators (bumble bees) with different body sizes on pollination accuracy and the efficiency of different pollinators, by estimating the quantity (visiting frequency) and quality (pollen removal and stigmatic pollen deposition by a single visit). Our results indicated that bumble bees of different body sizes had no effects on pollen removal, all visitors could remove a lot of pollen grains in a single visit; but had effects on stigmatic pollen deposition, small Bombus richardsi with the smallest body size having lower legitimate pollen ratio and fewer legitimate pollen grains on stigma compared with those bees with larger body size. Legitimate pollen ratio on stigma varied with the visitation frequency between different years, especially in thrum morph. These results remind us of Darwin's idea that pollinators were the main evolutionary force that drove the evolution of distyly.Thirdly, we further explored the evolution and adaptive significance of distyly through artificial pollination to test the strength of self-incompatibility given that the large number of illegitimate pollen grains loaded on stigma would reduce reproductive success of distyous species.To examine the hypothesis of promotion of disassortative pollination, we altered the relative distance between stigma and anther in the same flower to decrease the relative stigma-anther distance. The results of artificial pollination in four Primula species showed that both pin and thrum morphs in the four species had a certain degree of compatibility and different morphs in different species had the different intensity of compatibility. A certain degree of compatibility has been considered as a reproductive assurance especially in the species under pollen limitation. Our result of manipulating the relative stigma-anther distance within flowers indicated that percentage of legitimate pollen loaded on stigma was low in both pin and thrum morphs under natural condition, especially for pin flowers, only about 10%. However the percentage of legitimate pollen loaded on stigma was decreased significantly in the manipulated flowers with shortened stigma-anther distance. This floral manipulation provided evidence in support that reciprocal herkogamy did facilitate legitimate pollination. The result of change the opening direction of flowers showed similar effect on stigmatic pollen load in the two Primula species. Changed flowers had lower legitimate pollen ratio than intact flowers. This decrease of fitness in changed flowers can be explained by the behavior of pollinators especially for the flowers which were opening upward.To conclude, our study showed that the relationship between plants and pollinators in nature is complex, the sexual reproductive success was influenced by many factors. In distylous species, its pollination accuracy is influenced by numerous factors, including the floral traits and pollination behaviors. Taken together, our experimental studies in the six Primula species indicated that the reciprocal herkogamy in distyly could promote assorative pollination, reduce the waste of resources (pollen).