Get access

Genetic and community similarities are correlated in endemic-rich springs of the northern Chihuahuan Desert


*Correspondence: Makiri Sei, Department of Zoology, Miami University, Oxford, OH 45056, USA. E-mail:


Aim  Species diversity and genetic diversity within a taxon are intrinsic parts of global biodiversity. These two levels of biodiversity can show strong correlation due to a variety of reasons (i.e. parallel processes affecting both communities and populations, genotypes of a numerically or functionally dominant species affecting community composition, a species assemblage selecting for a particular genotype by affecting its selection regime). We examined correlations between species and genetic biodiversity in four isolated endemic-rich spring systems in a hot desert and their potential link to environmental variables and physical isolation.

Location  Chihuahuan Desert spring systems in the Pecos River basin of New Mexico and Texas, USA.

Methods  We compared species richness of fish and benthic macroinvertebrate communities to within-population allelic richness of amphipods (monophyletic Gammarus spp.) and Pecos gambusia (Gambusia nobilis) using Mantel tests. We also compared pairwise community similarities with pairwise genetic identities of populations among the same groups. We tested correlations among diversity, similarity and environmental variables after controlling for the effects of spatial distances using partial Mantel tests. We partitioned genetic and species diversity into three spatial scales (i.e. individual springs, individual spring systems, the entire region) using AMOVA and partition.

Results  We found strong correlations between invertebrate species richness and mosquitofish allelic richness. We found even stronger correlations of amphipod and gambusia genetic identities with fish and invertebrate community similarities; these were best explained by geographic distance rather than abiotic environmental factors. Most of the taxa and communities exhibited the largest proportion of diversity at the regional level.

Main conclusions  Our results suggest that drift and migration are the mechanisms that best explain our observations, and although α-diversity among genes and species may not be strongly correlated, the pattern of species and allelic complementarity among these groups seems to be concordant at the regional level.