EVOLUTION ON A VOLCANIC CONVEYOR BELT - USING PHYLOGEOGRAPHIC RECONSTRUCTIONS AND K-AR-BASED AGES OF THE HAWAIIAN-ISLANDS TO ESTIMATE MOLECULAR EVOLUTIONARY RATES

The Hawaiian Islands form as the Pacific Plate moves over a 'hot spot' in the earth's mantle where magma extrudes through the crust to build huge shield volcanos. The islands subside and erode as the plate carr ies them to the north-west, eventually to become coral atolls and seam ounts. Thus islands are ordered linearly by age, with the oldest islan ds in the north-west (e.g. Kauai at 5.1 Ma) and the youngest in the so uth-east (e.g. Hawaii at 0.43 Ma). K-Ar estimates of the date of an is land's formation provide a maximum age for the taxa inhabiting the isl and. These ages can be used to calibrate rates of molecular change und er the following assumptions: (i) K-Ar dates are accurate; (ii) tree t opologies show that derivation of taxa parallels the timing of island formation; (iii) populations do not colonize long after island emergen ce; (iv) the coalescent point for sister taxa does not greatly predate the formation of the colonized younger island; (v) saturation effects and (vi) among-lineage rate variation are minimal or correctable; and (vii) unbiased standard errors of distances and regressions can be es timated from multiple pairwise comparisons. We use the approach to obt ain overall corrected rate calibrations for: (i) part of the mitochond rial cytochrome b gene in Hawaiian drepanidines (0.016 sequence diverg ence/Myr); (ii) the Yp1 gene in Hawaiian Drosophila (0.019/Myr Kambyse llis et al. 1995); and (iii) parts of the mitochondrial 12S and 16S rR NA and tRNA(val) in Laupala crickets (0.024-0.102/Myr, Shaw 1996). We discuss the reliability of the estimates given the assumptions (i-vii) above and contrast the results with previous calibrations of Adh in H awaiian Drosophila and chloroplast DNA in lobeliods.