DISCORD BETWEEN THE PHYLOGENIES INFERRED FROM MOLECULAR VERSUS FUNCTIONAL DATA - UNEVEN RATES OF FUNCTIONAL EVOLUTION OR LOW-LEVELS OF GENEFLOW

According to measures of molecular divergence, the three species of th e Drosophila simulans clade are closely related to and essentially equ idistant from each other. We introgressed 10% of the D. sechellia X ch romosome into a pure D. simulans genetic background and found that mal es carrying this introgressed region were consistently fertile; in con trast, males carrying the same segment from D. mauritiana are sterile and suffer from incompatibilities at a minimum of four loci. Together with other recent results, these data suggest that D. simulans and D. sechellia are much more closely related to each other than either is t o D. mauritiana. How can we reconcile the phylogeny inferred from the density of hybrid sterility genes with that inferred from molecular di vergence? If the molecular phylogeny is correct, the discrepancy might be explained by uneven rates of functional evolution, resulting in th e uneven accumulation of substitutions with corresponding negative eff ects in hybrids. If the functional phylogeny is correct, then low leve ls of gene flow across nascent species boundaries, particularly for lo ci not tightly linked to a hybrid sterility gene, may have erased the original pattern of lineage splitting. We propose tests that will allo w us to discriminate between these hypotheses.