Negative frequency-dependent selection on polymorphic color morphs in adders.

Color pattern polymorphism occurs when more than one form is found within the same population. It is widespread in a variety of taxa, leading us to ask what maintains this variation. One stabilizing mechanism is negative frequency-dependent selection, also known as apostatic selection, in which the fitness of a phenotype decreases with its frequency.1-6 Negative frequency-dependent selection has been proposed as one of the most powerful selective forces in maintaining phenotypic and genetic diversity in both plant and animal populations.1,5 Despite its importance and experimental evidence,1,2,6 no study has documented that natural selection due to predation may result in negative frequency-dependent selection in a wild undisturbed vertebrate population. Here, we report the results of a long-term study, spanning 37 years from 1984 to 2020, of two distinct color morphs, zigzag and melanistic, within a population of adult adders (Vipera berus) on the island of Hallands Väderö in southern Sweden (Figure 1). Our results strongly suggest that the color pattern polymorphism is maintained by negative frequency-dependent natural selection in both males and females.

The effect of hosptlant and parasitoids on the reproductive success of the parthenogenetic gall wasp Diplolepis rosae (Hymenoptera, Cynipidae).

The univoltine cynipid gall wasp Diplolepis rosae reproduces by an obligate homozygosity promoting system known as gamete duplication. The wasp is confined to roses (Rosa spp) on which it induces large, complex and multichambered galls. In southern Sweden, D. rosae was found to parasitize Rosa canina, R. dumalis, R. rubiginosa, R. villosa, R. sherardi and R. rubrifolia, but not R. majalis and R. rugosa. The distribution of galls shows that there are differences in the relation between wasp and hosplant with respect both to species and individual plants. There is a positive correlation between wasp size and gall (clutch) size. Parasitoid pressure was found to be high, causing D. rosae an estimated average larval loss of approximately 75%, mainly due to the attack of the ichneumonid wasp Orthopelma mediator. The very common cynipid inquiline Periclistus brandtii does not seem to have any negative effects. Overall parasitism and probability of no hatched offspring per gall decrease with increasing gall (clutch) size. The probability of loosing all of a given number of offspring decreases with the number of galls produced. It is suggested that D. rosae, in order to escape parasitoids, needs high ability to establish new colonies. Hence the production of many comparatively small galls, which increases the chance of leaving any offspring, rather than the production of few large galls, maximizing the number of offspring, should be favoured by selection.