Comparative genetic structure and demographic history in endemic galapagos weevils

Abstract

The challenge of maintaining genetic diversity within populations can be exacerbated for island endemics if they display population dynamics and behavioral attributes that expose them to genetic drift without the benefits of gene flow. We assess patterns of the genetic structure and demographic history in 27 populations of 9 species of flightless endemic Galápagos weevils from 9 of the islands and 1 winged introduced close relative. Analysis of mitochondrial DNA reveals a significant population structure and moderately variable, though demographically stable, populations for lowland endemics (F(ST) = 0.094-0.541; π: 0.014-0.042; Mismatch P = 0.003-0.026; and D((Tajima)) = -0.601 to 1.203), in contrast to signals of past contractions and expansions in highland specialists on 2 islands (Mismatch P = 0.003-0.026 and D((Tajima)) = -0.601 to 1.203). We interpret this series of variable and highly structured population groups as a system of long-established, independently founded island units, where structuring could be a signal of microallopatric differentiation due to patchy host plant distribution and poor dispersal abilities. We suggest that the severe reduction and subsequent increase of a suitably moist habitat that accompanied past climatic variation could have contributed to the observed population fluctuations in highland specialists. We propose the future exploration of hybridization between the introduced and highland endemic species on Santa Cruz, especially given the expansion of the introduced species into the highlands, the sensitivity to past climatic variation detected in highland populations, and the potentially threatened state of single-island endemics.

Figure 1

Map of the Galápagos archipelago indicating distribution per island of Galapaganus species, and in parentheses, the number of collecting sites per species included from each island. Inset A: relative position of the archipelago, inset B: detailed view of Santa Cruz island, and inset C: 2 northernmost islands.

Figure 2

Minimum spanning network of all combined COI, COII, CytB, and 12S haplotypes for the 9 island endemic species. Haplotypes belonging to the same species are surrounded by dashed lines, and circle sizes are proportional to the number of individuals sharing that haplotype. Numbers on the lines indicate the number of mutational steps between haplotypes, italic indicates steps between haplotypes from different islands, and bold indicates steps between haplotypes from different species. Shaded circles correspond to the simplified ecological zonation scheme used in the AMOVA analysis. White: lowland habitats, light gray: midelevation habitats, and dark gray: highland habitats; when there are 2 or more populations of the same species from a certain habitat, haplotypes belonging to one of the populations are marked with a dot.

Figure 3

Mismatch distributions for the concatenated mitochondrial data set for some of the endemics and the introduced species. When a species’ range spans more than one island, mismatch distributions are presented for population groups within one island as indicated in parenthesis after species names. Galapaganus ashlocki, G. galapagoensis, and G. williamsi are single-island endemics as indicated by SIE. Curves of expected values were obtained by simulating 10 000 data sets under a coalescent algorithm by implementing parameter estimates based on a sudden demographic expansion (Schneider and Excoffier 1999). P values represent the probability that the variance of the simulated data set is equal or greater than the observed one.