It is generally believed that Drosophila melanogaster has no closely related species with which it can produce the viable and fertile hybrids that are essential for the genetic analysis of speciation. Following the recent report of molecular differentiation between a Zimbabwe, Africa, population and two United States populations, we provide evidence that strong sexual isolation exists between the D. melanogaster population in Zimbabwe and populations of other continents. In the presence of males of their own kind, females from most isofemale lines of Zimbabwe would not mate with males from elsewhere; the reciprocal mating is also significantly reduced, but to a lesser degree. The genes for sexual behaviors are apparently polymorphic in Zimbabwe and postmating reproductive isolation between this and other populations has not yet evolved. Whole chromosome substitutions indicate significant genetic contributions to male mating success by both major autosomes, whereas the X chromosome effect is too weak to measure. In addition, the relative mating success between hybrid and pure line males supports the interpretation of strong female choice. These observations suggest that we are seeing the early stages of speciation in this group and that it is driven by sexual selection. The genetic and molecular tractability of D. melanogaster offers great promise for the detailed analysis of this apparent case of incipient speciation.The difficulties in studying the genetics of speciation can arise from several sources. First, the species in question may have diverged beyond the incipient stage. Many genetic differences between them could have accumulated after speciation had been completed and the information on the population genetic dynamics of speciation may have been lost. Ideally, we would like to observe variation in genes of reproductive isolation that are still in the process of becoming fixed. Polymorphisms of such genes within species would suggest speciation inflagrante delicto (1). The second difficulty arises when the species of interest does not lend itself to extensive and detailed genetic analysis. The conspicuous absence of a species that could hybridize with Drosophila melanogaster to produce fertile progeny is the prime example of the second point. To some degree, all studies of the genetics of postmating reproductive isolation encounter the two problems (2). Genetic analysis of premating sexual isolation also confronts a third difficulty in that mating behaviors are often labile (3-8). Finding a system of sexual isolation associated with robust behavioral phenotypes is thus crucial for genetic studies of premating isolation.Recently, a collection of isofemale lines of D. melanogaster from Zimbabwe, Africa, was reported to show a surprisingly high level of DNA sequence divergence at several nuclear genes compared to flies from North American populations (9). In light of previous observations that flies of this species collected over a wide geographical range are very similar in their nuclear DNA polym...

Genetic divergence for characters pertaining to reproductive isolation is of considerable interest in evolutionary biology. Since most studies concentrate on sibling species (for recent reviews, see Wu et al. 1996), we would like to know how much genetic variation exists between populations that are at an incipient stage of speciation. To answer this question, we have begun measuring variations in mating preference among natural isolates of Drosophila melanogaster, represented by the cosmopolitan and Zimbabwe sexual races. We quantify the variation in mating preference and success in both sexes by using a multiple-choice design and an index that is suited to cases of strong asymmetry in mate choice. Different designs and indices for measuring sexual isolation are also discussed. These sexual traits are entirely genetically determined. Surveying four populations in southern Africa and additional cosmopolitan lines, we observe extensive genetic variation in sexual characters as well as strong correlation between sexes. The populations are highly differentiated and represent various stages of evolution between the African and the cosmopolitan type of sexual behaviors. The genetic variation and correlation for these sexual characters coupled with their geographical pattern have interesting implications for models of speciation by sexual selection.

Factors that influence the genesis and genomic distribution of microsatellite DNA are poorly understood. We have identified a novel class of Dipteran mobile elements, mini-me elements, which help elucidate both of these issues. These retroposons contain two internal proto-microsatellite regions that commonly expand into lengthy microsatellite repeats. These elements are highly abundant, accounting for approximately 1.2% of the Drosophila melanogaster genome, giving them the potential to be a prolific source of microsatellite DNA variation. They also give us the opportunity to observe the outcomes of multiple microsatellite genesis events (initiating from the same proto-microsatellite) at separate mini-me loci. Based on these observations, we determined that the genesis of microsatellites within mini-me elements occurs through two separate mutational processes: the expansion of preexisting tandem repeats and the conversion of sequence with high cryptic simplicity into tandemly repetitive DNA. These modes of microsatellite genesis can be generalized beyond the case of mini-me elements and help to explain the genesis of microsatellites in any sequence region that is not constrained by selection.

Genetic divergence for characters pertaining to reproductive isolation is of considerable interest in evolutionary biology. Since most studies concentrate on sibling species (for recent reviews, see Wu et al. 1996), we would like to know how much genetic variation exists between populations that are at an incipient stage of speciation. To answer this question, we have begun measuring variations in mating preference among natural isolates of Drosophila melanogaster, represented by the cosmopolitan and Zimbabwe sexual races. We quantify the variation in mating preference and success in both sexes by using a multiple-choice design and an index that is suited to cases of strong asymmetry in mate choice. Different designs and indices for measuring sexual isolation are also discussed. These sexual traits are entirely genetically determined. Surveying four populations in southern Africa and additional cosmopolitan lines, we observe extensive genetic variation in sexual characters as well as strong correlation between sexes. The populations are highly differentiated and represent various stages of evolution between the African and the cosmopolitan type of sexual behaviors. The genetic variation and correlation for these sexual characters coupled with their geographical pattern have interesting implications for models of speciation by sexual selection.

The distribution of wildlife parasites in a landscape is intimately tied to the spatial distribution of hosts. In parasite species, including many gastrointestinal parasites, with obligate or common environmental life stages, the dynamics of the parasite can also be strongly affected by geophysical components of the environment. This is especially salient in host species, for example humans and macaques, which thrive across a wide variety of habitat types and quality and so are exposed to a wealth of environmentally resilient parasites. Here, we examine the effect of environmental and anthropogenic components of the landscape on the prevalence, intensity, and species diversity of gastrointestinal parasites across a metapopulation of long-tailed macaques on the island of Bali, Indonesia. Using principal-components analysis, we identified significant interaction effects between specific environmental and anthropogenic components of the landscape, parsing the Balinese landscape into anthropogenic (PC1), mixed environment (PC2), and non-anthropogenic (PC3) components. Further, we determined that the anthropogenic environment can mitigate the prevalence and intensity of specific gut parasites and the intensity of the overall community of gut parasites, but that non-anthropogenically driven landscape components have no significant effect in increasing or reducing the intensity or prevalence of the community of gut parasites in Balinese macaques.