Clownfishes evolution below and above the species level

Abstract

The difference between rapid morphological evolutionary changes observed in populations and the long periods of stasis detected in the fossil record has raised a decade-long debate about the exact role played by intraspecific mechanisms at the interspecific level. Although they represent different scales of the same evolutionary process, micro- and macroevolution are rarely studied together and few empirical studies have compared the rates of evolution and the selective pressures between both scales. Here, we analyse morphological, genetic and ecological traits in clownfishes at different evolutionary scales and demonstrate that the tempo of molecular and morphological evolution at the species level can be, to some extent, predicted from parameters estimated below the species level, such as the effective population size or the rate of evolution within populations. We also show that similar codons in the gene of the rhodopsin RH1, a light-sensitive receptor protein, are under positive selection at the intra and interspecific scales, suggesting that similar selective pressures are acting at both levels.

Keywords: RH1; interspecific; intraspecific; macroevolution; microevolution; positive selection.

Figure 1.

(a) Phylogenetic tree showing the relationships between species of clownfishes (macroevolution dataset) and (b) coalescence process showing the relationship between 53 individuals of A. clarkii (microevolution dataset). Nodes highlighted in red are nodes less supported, with a posterior probability inferior to 0.8. (Online version in colour.)

Figure 2.

Comparison between the micro- and macroevolutionary rates and the predicted macroevolutionary rate for molecular data (RH1). The columns (micro) in yellow and (macro) in black correspond to the estimates of micro- and macroevolutionary rates obtained from RH1 sequences. The column macro_P, in blue, represents the predicted macroevolutionary rate based on A. clarkii molecular evolutionary rate described in the Material and methods. This rate was obtained by a subsampling procedure by pruning all but one A. clarkii individual in the tree containing all clownfish species and A. clarkii individuals. Estimates of molecular evolutionary rate are given in number of mutations × 10⁻³/Myrs for the microevolutionary scale and in number of substitutions × 10⁻³/Myr for macroevolutionary scale. (Online version in colour.)

Figure 3.

Comparison between the macroevolution rates estimated from empirical data and simulations (based on intraspecific parameters) of the five morphological traits. The columns (macro) in black correspond to the estimates of macroevolutionary rates observed in the data, and (macro_P) in blue, represent the predicted macroevolutionary rate of the simulations based on effective population size, trait variance and generation time (described in details in electronic supplementary material, S2). The dots represent the median and the segment represents the 95% confidence interval of the distribution of rates across the 1000 trees. For each trait, simulated and empirical rates of evolution (σ²) were obtained from best fitting model chosen only from the empirical data (BM or OU, electronic supplementary material, table S4). The predicted and observed distributions of macroevolutionary rates did not differ significantly in any of the traits (p > 0.05). (Online version in colour.)