The onset and conservation of genetic polymorphism is a major question in evolutionary ecology. The influence of temporal fluctuations in the environment was invoked by early theorists such as J.B.S. Haldane and S. Jayakar in a controversial article published in, 1963, but their frequently cited model has almost never been used with empirical evidence. In this paper, we present a simulation model inspired by the biology of common gorse (Ulex europaeus), a species which shows polymorphism of flowering phenology: long flowering plants produce flowers from winter to spring and short flowering plants only flower in the spring. The early fruits of the former run the risk of frost, but largely escape seed predation, while those of the latter escape the risk of frost but are subject to a strong risk of seed predation. These two selection pressures vary unpredictably from year to year, making this flowering phenotype a good candidate to test Haldane and Jayakar’s model. Assuming that both flowering types are determined by a single major locus, we devise a simulation model firstly in a diploid form, and secondly by taking into account the hexaploid characteristic of gorse. Our results show that the combination of the two selective pressures acting on gorse flowering phenology can lead to fitness values meeting the Haldane and Jayakar’s conditions on geometric and arithmetic means, and to long term maintenance of polymorphism. In addition, the values of the parameters allowing polymorphism persistence and the relative proportions obtained are in agreement with values observed in natural populations. We also show that hexaploidy strongly increases the range of parameters in which polymorphism is self sustaining. These results are discussed in the context of climatic change, where increases of both mean temperature and its variance are predicted.
see on external website