Metabolic rate and climatic fluctuations shape continental wide pattern of genetic divergence and biodiversity in fishes

Abstract

Taxonomically exhaustive and continent wide patterns of genetic divergence within and between species have rarely been described and the underlying evolutionary causes shaping biodiversity distribution remain contentious. Here, we show that geographic patterns of intraspecific and interspecific genetic divergence among nearly all of the North American freshwater fish species (>750 species) support a dual role involving both the late Pliocene-Pleistocene climatic fluctuations and metabolic rate in determining latitudinal gradients of genetic divergence and very likely influencing speciation rates. Results indicate that the recurrent glacial cycles caused global reduction in intraspecific diversity, interspecific genetic divergence, and species richness at higher latitudes. At the opposite, longer geographic isolation, higher metabolic rate increasing substitution rate and possibly the rapid accumulation of genetic incompatibilities, led to an increasing biodiversity towards lower latitudes. This indicates that both intrinsic and extrinsic factors similarly affect micro and macro evolutionary processes shaping global patterns of biodiversity distribution. These results also indicate that factors favouring allopatric speciation are the main drivers underlying the diversification of North American freshwater fishes.

Figure 1. Plots of genetic divergence against midpoint latitude and mass specific metabolic rate using raw data.

Each sign represents a different order (○ = Cypriniformes, × = Perciformes, Δ = Cyprinodontiformes, ∇ = Scorpeiniformes, ⊠ = Semionotiformes, * = Siluriformes, ◊ = Salmoniformes, ⊞ = Esociformes). Fitted generalized linear mixed models show a significant relationship between variables in all comparisons (see Table 1).

Figure 2. Plots of phylogenetically independent sister pair comparisons of substitution rate against mass specific metabolic rate.

The X values correspond to the natural logarithm of the mass specific metabolic rate of the first species divided by the mass specific metabolic rate of the second species. The Y values correspond to the natural logarithm of the branch length of the first species divided by the branch length of the second species. Regression line was forced through the origin.