Geographic variation of contact calls suggest distinct modes of vocal transmission in a leaf-roosting bat

Abstract: Populations that have historically been isolated from each other are expected to differ in some heritable features. This divergence could be due to drift (and other mechanisms of neutral evolution) or differential adaptation of populations to local conditions. Discriminating between these two evolutionary trajectories can be difficult, but when possible, such data provides critical insight into the evolutionary history of a species. Here, we examine the patterns of geographic variation of two contact signals regularly produced by disc-winged bats, Thyroptera tricolor, and discuss possible processes leading to the observed patterns of differentiation. We compared allopatric populations separated by an elevational barrier, and estimated genetic distance using nuclear microsatellite loci. Our findings revealed that the mountain ridge is permeable to gene flow. Acoustic divergence was significantly explained by genetic and spatial factors, supporting the notion that stochastic factors are the main drivers of signal divergence. Yet, we found different patterns of geographic variation between the two types of calls. We examine how this variability in the patterns of acoustic divergence may suggest distinct modes of signal transmission within and between populations (i.e., social vs genetic transmission). This work provides further support of the role of random change shaping communication systems in mammals, and highlights the importance of studying multiple elements of a species repertoire to evaluate the underlying processes driving signal evolution. Significance statement: Despite increasing interest in studying patterns of acoustic divergence, the relative contribution of adaptive and stochastic processes underlying variation of acoustic signals remain poorly understood, particularly in mammals. Our study examines signal divergence in Spix’s disc-winged bats, Thyoptera tricolor, with the goal of understanding the underlying processes driving signal evolution. Specifically, we studied whether the patterns of geographic variation of two social signals regularly emitted by T. tricolor are congruent with patterns of genetic distance among populations separated by a geographic barrier. We demonstrate that genetic and spatial distance explains acoustic variation, which points to stochastic processes as major drivers of signal divergence in T. tricolor. Notably, we found that the patterns of geographic variation differ between the two types of calls studied. We suggest that this variation results from distinct modes of vocal transmission within populations. Comparison of different signal types provides additional insight of social pressures shaping call design.