Seasonal timing of reproduction affects the outcome of interactions between plants and their pollinators, grazers and seed-predators, as well as with their local abiotic environment. Phenotypic selection has generally been found to favour early flowering, especially in seasonal environments where time for growth and reproduction is limited. Yet, there is little evidence from natural populations that this corresponds to genotypic selection, or that it results in evolutionary change.
In this study, we experimentally assessed both phenotypic and genotypic selection on flowering time in a natural population of the perennial herb Lathyrus vernus. Specifically, we asked: 1) Is there phenotypic selection for early flowering? 2) If so, through what fitness components does selection act and which are the agents of selection? 3) What is the heritability of flowering time? And, 4) To what extent does phenotypic selection correspond to genotypic selection?
We transplanted sibling Lathyrus vernus (Fabaceae) individuals, obtained through controlled crosses to a common field site, where we recorded fitness and phenology in terms of first flowering day, as well as grazing and seed predation.
Results/Conclusions
Early flowering plants produced more seeds than later flowering individuals. An earlier first flowering day was associated with a decreased probability of escaping grazing, but with an increased fruit and seed set, while seed predation was not affected. Heritability for first flowering day was very low and no part of the covariance between first flowering day and fitness could be explained by plant genotype.
Like several previous studies, we found phenotypic selection for earlier flowering. However, we found no support for genotypic selection on flowering time. Our results suggest that commonly observed patterns of higher fitness in early flowering plants do not correspond to selection on genotype values, and thus are not expected to result in evolutionary change.