THE MOLECULAR CLOCK REVISITED - THE RATE OF SYNONYMOUS VS. REPLACEMENT CHANGE IN DROSOPHILA

Rates of synonymous and nonsynonymous substitution were investigated f or 24 genes in three Drosophila species, D. pseudoobscura, D. subobscu ra, and D. melanogaster. D, pseudoobscura and D. subobscura, two dista ntly related members of the obscura clade, differ on average by 0.29 s ynonymous nucleotide substitutions per site. D. melanogaster differs f rom the two obscura species by an average of 0.81 synonymous substitut ions per site. Using a method developed by Gillespie, we investigated the variance to mean ratio, or Index of Dispersion, R, of substitution s along the three species' branches to test the fundamental prediction of the neutral theory of molecular evolution, E(R) = 1. For nonsynony mous substitutions, the average R, R-a is 1.6, which is not significan tly different from the neutral theory prediction. Only 5 of the 24 gen es had significantly large R-a valves, and 12 of the genes had R, esti mates of less than one. In contrast, the Index of Dispersion for synon ymous substitutions was significantly large for 12 of the 24 genes, wi th an average of R-s = 4.4, also statistically significant. These find ings contrast with results for mammals, which showed overdispersion of nonsynonymous substitutions, but not of synonymous substitutions. Wea k selection acting to maintain codon bias in Drosophila, but not in ma mmals, may be important in explaining the high variance in the rate of synonymous substitutions in this group of organisms.