Body elongation, lateral shape, and body size do not show a consistent effect on diversification rates of ray-finned fishes at broad phylogenetic scale
摘要
Species richness differences among lineages and habitats, as well as the variety of their traits, have led to the association of intrinsic and extrinsic traits of species with diversification rates. Certain traits may be relevant for all species within a large clade and may thus influence the diversification dynamics of the whole clade. Here, we used the most extensive molecular phylogeny and morphological dataset available for ray-finned fishes (Actinopterygii), to test the influence of three ecologically relevant and widely sampled morphological traits – body elongation (bel), body lateral shape (bls), and maximum body size (len) – on the diversification rates at this broad phylogenetic scale. We also included habitat type (freshwater/marine) as a trait that could interact with the morphological traits and influence diversification rates. We applied the SecSSE model of trait-dependent diversification based on discrete traits, which allows evaluating the effect of examined and concealed trait states on diversification. We complemented this analysis with the QuaSSE and ES-sim models based on continuous traits. Under SecSSE, our results do not provide evidence for an effect of the examined morphological traits or their interactions with habitat, on the diversification rates of Actinopterygii. Instead, these results suggest the possibility that other, unmeasured trait(s) may contribute to their diversification dynamics. However, this finding could be limited by the variation loss imposed by the discretization of continuous traits that could potentially mask biological signal. In contrast, QuaSSE recovered an effect of all traits on diversification, although with high Type I error rate, whereas ES-sim detected a significant negative relationship with diversification only for body elongation. Taken together, our results do not provide consistent support for an effect of the examined traits on diversification rates, highlighting the sensitivity of inferences to methodological assumptions and data structure. Our findings do not exclude the possibility that the examined traits may have clade-specific effects, influencing the diversification of particular lineages (e.g., orders or families) within the Actinopterygii.