Figure 1. (a) The cross between D. melanogaster females and D. simulans males produces sterile hybrid daughters and hybrid sons that die at the larval-pupal transition. (b) The cross between D. simulans females and D. melanogaster males produces sterile hybrid sons and hybrid daughters that die as embryos. Light grey and dark grey chromosomes come from D. melanogaster and D. simulans, respectively. One pair of sex chromosomes (the Y is drawn with a hook) and one representative pair of autosomes are shown for each individual.

Figure 2. Results from the first backcross analysis of hybrid male sterility between D. pseudoobscura and D. persimilis (Dobzhansky, 1936). Testis length (a proxy for hybrid sterility) is shown for 16 backcross genotypes. Black=D. persimilis, white=D. pseudoobscura; the tiny dot-fifth chromosome is not shown. Substitution of the X-chromosome has a disproportionately large effect on hybrid male sterility.

Figure 3. The evolution of hybrid incompatibilities according to the Dobzhansky–Muller model. The A mutation arises and replaces the a allele in the species 1 lineage; similarly, the B mutation arises and displaces b allele in the species 2 lineage. Note that neither species’ lineage passes through a sterile or inviable intermediate genotype. Nevertheless, the A and B alleles, while both functional in a and b genetic backgrounds, have never been in the same genome together during their evolutionary history. Consequently, natural selection has never had the opportunity to test the A–B combination, and there is some chance that A and B will be incompatible with one another, causing hybrid sterility or inviability.

Figure 4. Two approaches to genetically mapping hybrid incompatibility genes. (a) Crossing scheme used to deficiency map recessive hybrid lethal genes between D. melanogaster (mel) and D. simulans (sim). Df, deficiency chromosome and Bal, dominantly marked balancer chromosome. Coyne et al. (1998) contrasted the viability of Df/sim versus Bal/sim hybrid females to map recessive sim factors that are incompatible with dominant mel ones. Presgraves (2003) used the Lhr rescue mutation to recover and contrast Df/sim versus Bal/sim hybrid males to map recessive autosomal sim factors that are incompatible with recessive X-linked mel ones. (b) Introgression mapping scheme used to move small chromosomal segments from D. mauritiana into otherwise D. simulans (True et al., 1996) or D. sechellia (Masly and Presgraves, 2007) genetic backgrounds. D. mauritiana segments are marked with a dominant visible marker (here a P-element bearing the white⁺ eye colour allele) and introgressed over 15 generations by backcrossing through fertile hybrid females. Introgressed segments are then made homozygous and tested for their effects on hybrid fitness.

Figure 5. The distribution of hybrid incompatibilities in D. mauritiana–D. sechellia introgression hybrids. Above the X, second and third chromosomes, triangles show the locations of different P-elements used for introgression; below the chromosomes, the estimated sizes of many introgressions are shown. Red=hybrid male sterile; black=hybrid lethal and grey=viable and fertile. The tiny dot-fourth chromosome is not shown.