THE CONTRIBUTION OF NEW MUTATIONS TO GENOTYPE-ENVIRONMENT INTERACTIONFOR FITNESS IN DROSOPHILA-MELANOGASTER

Many studies have documented the existence of genotype-environment int eraction (GEI) for traits closely related to fitness in natural popula tions. A type of GEI that is commonly observed is changes in the fitne ss ranking of genetic groups (families, clones, or inbred lines) in di fferent environments. We refer to such changes in ranking as crossing of reaction norms for fitness. A common interpretation of crossing of reaction norms for fitness is that selection favors different alleles in the different environments (i.e., that ''trade-offs'' exist). If th is is the case, selection could maintain genetic variation, and even l ead to reproductive isolation between subpopulations using different e nvironments. Even if the stunt alleles are favored in every environmen t, however, deleterious mutations that vary in the magnitude of their effect depending on environment could cause reaction norms for fitness to cross. If deleterious mutations with environment-dependent effects are responsible for maintaining much of the Variation leading to cros sing of reaction norms for fitness in natural populations, it should b e possible to observe crossing of reaction norms for fitness among oth erwise genetically identical lines bearing newly arisen spontaneous mu tations. We examined the contribution of new mutations to GEI for fitn ess in Drosophila melanogaster. Eighteen lines were derived from a com mon, highly inbred base stock, and maintained at a population size of 10 pairs far over 200 generations, to allow them to accumulate spontan eous mutations. Because of the small population size of the lines, sel ection against mildly deleterious mutations should have been relativel y ineffective. The lines were tested for productivity (number of survi ving adult progeny from a standard number of parents) in five differen t environmental treatments, comprising different food media, temperatu res, and levels of competition. The lines showed highly significant GE I for productivity, owing largely to considerable changes in ranking i n the different environments. We conclude that mutations that are dele terious on average, but whose quantitative effects depend on environme nt, could be responsible for maintaining much of the Variation leading to crossing of reaction norms for fitness that has been observed in s amples of D. melanogaster from the wild.