Measurement of fitness at the esterase-5 locus in Drosophila pseudoobscura.

Abstract

KNOWLEDGE of the amount of genic variation in natural populations is the first step towards understanding how such variation is maintained. The first systematic studies of the amount of genic variation in natural populations of Drosophila pseudoobscura were reported by HUBBY, LEWONTIN, and co-workers (HUBBY and LEWONTIN 1966; LEWONTIN and HUBBY 1966; PRAKASH, LEWONTIN and HUBBY 1969) who estimated that 40% of the loci are polymorphic and that on the average 12% of loci per individual are heterozygous in North American populations of the species. Similar estimates of heterozygosity have now been made for several organisms, including mice (SELANDER and YANG 1969) and humans (HARRIS 1969; LEWONTIN 1967). All these estimates indicate that the proportion of polymorphic loci in natural populations may be much greater than was previously supposed. Were these polymorphisms maintained primarily by heterosis operating independently at each locus, populations probably would not be able to tolerate the huge genetic load involved (see LEWONTIN and HUBBY 1966). SVED, REED and BODMER (1967) have attempted to solve this dilemma by postulating that there is a limit to maximum fitness because of gene interaction. Owing to the rarity of the maximally fit genotypes, the amount of reduction in the selective advantage of heterozygote over homozygote at individual loci is extremely small. KING (1967) proposed a different threshold model in which he assumed that a certain proportion of the population with the “worst combination of genes, environment, and luck” will be eliminated by natural selection. The effect of KING’S model is similar to that of SVED, REED and BODMER (1967) both in the sense that maximal fitness is approached asymptotically as heterozygosity increases, and that there are thousands of different genotypes with essentially the same fitness. According to these models a large number of polymorphic loci can be maintained without increasing genetic load to an excess. However, both hypotheses fail to explain the relatively small loss of fitness found when natural populations are inbred. The hypothesis of neutrality or very small selection coefficients of allozymes as proposed by KIMURA (1968) may be adequate to explain the maintenance of large amounts of polymorphism by assuming a rather high mutation rate, migration rate, and population size, but it does not appear to be an adequate explanation of