CHARACTER DISPLACEMENT FOR SEXUAL ISOLATION BETWEEN DROSOPHILA MOJAVENSIS AND DROSOPHILA ARIZONENSIS.

Abstract

According to Brown and Wilson (1956) character displacement exists between two closely related species when their allopatric populations are very similar and their sympatric populations are quite distinct in one or more characters. The disparate characters could be either morphological, physiological, ecological, or behavioral. They viewed the phenomenon as a result of species interaction in sympatry, where divergence would decrease competition or reduce hybridization, and thus be at a selective advantage. They gave several instances which have since been widely cited as classical, textbook examples. More recently, Grant (1972, 1975) reexamined the phenomenon of interspecific interaction, including convergent character displacement as well as the divergent character displacement described by Brown and Wilson. Restricting himself to morphological attributes, Grant reviewed the literature and concluded that there really was no good example of character displacement. Where divergent character displacement had been proposed to exist, Grant argued, the data were incomplete or else the reason for the divergence was not a result of interspecific competition but rather other ecological factors. We feel that many of these problems could be avoided if one could demonstrate character displacement for sexual isolation. Although ethological barriers to interspecific mating are expected to be developed in allopatric populations (Muller, 1939, 1942), their reinforcement in sympatric populations can only be attributed to interspecific interaction and not to other correlated factors. Dobzhansky (1940) proposed on theoretical grounds that sexual isolation between closely related species should be greater in sympatric than in allopatric populations. His arguments essentially are that if there is selection against the hybrids, those individuals which are involved in interspecific crosses would be wasting their gametes. Any gene which improved the ability of the species to discriminate would be advantageous and would increase in frequency, or be fixed, in the region of sympatry. Reinforcement of sexual isolation by means of artificial selection against the hybrids can sometimes be accomplished in the laboratory (Koopman, 1950; Thoday and Gibson, 1962; Kessler, 1966; Ehrman, 1971, 1973; Soans et al., 1974). However, despite its theoretical importance and despite the moderate degree of success obtained in the laboratory, evidence for reproductive character displacement in natural populations is poor and controversial (see e.g., Loftus-Hills, 1975 and Jones, 1975). It has only been shown to exist in some anurans (Fouquette, Jr., 1975; Littlejohn and Loftus-Hills, 1968), some grasshoppers (Cohn and Cantrall, 1974), possibly in two species of damselfly (Waage, 1975), and between races of Drosophila paulistorum (Ehrman, 1965). In 1973 we undertook an investigation on the sexual isolation between natural populations of Drosophila mojavensis and Drosophila arizonensis species. The advantages of using Drosophila for this purpose are manifest in the availability of techniques which would not only enable us to detect reinforcement if it exists, but also to measure its strength and perhaps determine its genetic basis. D. mojavensis