Fig. 4.

Analysis of the effect of mating system and EPP on the rate of syllable repertoire size evolution. a Mating system and syllable repertoire: We generated 1000 stochastic character maps—simulations of the evolutionary history of monogamy and polygyny mapped onto the phylogeny—and then we tested whether syllable repertoire size evolved at different rates in monogamous vs. polygynous branches of the tree. From all runs that converged out of 1000 total runs of the Brownie algorithm, we plot the distribution of the rate of evolution of syllable repertoire size in monogamous lineages (blue) and the rate of evolution of syllable repertoire size in polygynous lineages (red). Distributions are kernel density plots generated using the R function density with a Gaussian smoothing kernel. In all panels, the dashed line indicates the rate of evolution estimated when the song characteristic is assumed to evolve at the same rate regardless of mating behavior. We find that syllable repertoire size evolves significantly faster in polygynous branches (Brownie likelihood-ratio test p = 0.006). b Mating system and song duration: The rate of evolution of song duration also differed, evolving significantly faster in monogamous lineages. c EPP and syllable repertoire: We performed a similar analysis with high and low rates of EPP mapped onto the phylogeny, and tested whether syllable repertoire size evolved at different rates in periods of high (red) vs. low (blue) rates of EPP. We do not reject the null hypothesis that syllable repertoire size evolved at the same rate in high-EPP and low-EPP branches of the tree