Building behavior does not drive rates of phenotypic evolution in spiders

Abstract

Do animals set the course for the evolution of their lineage when manipulating their environment? This heavily disputed question is empirically unexplored but critical to interpret phenotypic diversity. Here, we tested whether the macroevolutionary rates of body morphology correlate with the use of built artifacts in a megadiverse clade comprising builders and nonbuilders-spiders. By separating the inferred building-dependent rates from background effects, we found that variation in the evolution of morphology is poorly explained by artifact use. Thus natural selection acting directly on body morphology rather than indirectly via construction behavior is the dominant driver of phenotypic diversity.

Keywords: Araneae; animal architecture; extended phenotype; niche construction.

Fig. 1.

Posterior probability of niche construction effects on the rate of morphological evolution in spiders. Marginal posterior distribution of state-dependent rates (ζ) and posterior probability for each hypothesis (Inset bar plot, with H₀ ζ[builder] = ζ[nonbuilder]; H₁ ζ[builder] > ζ[nonbuilder]; H₂ ζ[builder] < ζ[nonbuilder]), from two combined reversible-jump Markov chain Monte Carlo analyses with 0.5 million generations each. Below mapping of the average branch-specific and state-dependent (smaller Inset) evolutionary rates from the posterior distribution. Tip labels indicate the state of the behavioral character (red, builder; light blue, nonbuilder). (A) Body size evolution, 815 species. (B) Evolution of body size and shape, 749 species. (C) Evolution of body size, body shape, cephalothorax height, mouth part size, eye size, and leg length; 340 species. For a definition of traits, see SI Appendix, Extended Methods.