MATING PATTERN AND MATING SUCCESS OF DROSOPHILA PSEUDOOBSCURA KARYOTYPES IN LARGE EXPERIMENTAL POPULATIONS.

Abstract

The pattern of matings between genotypes is the bridge connecting genotypic frequencies at the end of one generation with those at the beginning of the next. It is consequently an essential part of models in population genetics. The simplest assumption is that mating is random. The mating process in higher animals such as vertebrates and arthropods involves the pairing of diploid males and females and is random when the pairing occurs without regard to genotype. When mating is random and forces such as selection are absent, the genotypic frequencies among zygotes are the familiar Hardy-Weinberg frequencies. Most tests for randomness of mating are based on this fact. The usual procedure is to perform a chi-square test for goodness-of-fit between the frequencies of genotypes observed among zygotes and the Hardy-Weinberg expectations. However, such tests are indirect because they are based on genotypic frequencies which are the consequences of matings among adults of the previous generation, rather than on the frequencies of the matings themselves. And it is well known that the power of the chi-square test to detect departures from Hardy-Weinberg frequencies due to inbreeding (Ward and Sing, 1970) or selection (Lewontin and Cockerham, 1959) is severely limited. A stronger means of testing the randomness of mating is by direct analysis of the mating pattern-that is, by determination of the frequencies of matings between spe-