Vol.:(0123456789) 1 3 Oecologia https://doi.org/10.1007/s00442-020-04683-z COMMUNITY ECOLOGY – ORIGINAL RESEARCH Relative contributions of ecological drift and selection on bat community structure in interior Atlantic Forest of Paraguay Richard D. Stevens 1,2  · Jenna R. Grimshaw 1 Received: 25 July 2019 / Accepted: 9 June 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020 Abstract Despite extensive focus on numerous mechanisms that potentially structure Neotropical bat communities, understanding of the relative importance of any is still illusive. Recently, it has been demonstrated that all mechanisms used to explain community organization can be conceptualized as one or a combination of the few higher-level processes of dispersal, drift, selection and speciation. These four higher-level processes have not been addressed equally by Neotropical bat community ecologists. In particular, predictions formulated from a hypothesis of ecological drift have not been tested for any Neotropi- cal bat community. Herein we contrast efcacy of predictions based on the higher-level processes of drift and selection in describing community structure of bats in the Atlantic Forest of eastern Paraguay. Predictions apply to species-environment interactions, patterns of trait variation and beta-diversity, predictability of dominant competitors and responses to seasonal- ity. At best, there was inconsistent support for the operation of either drift or selection within this bat community. Selection, however, had more various forms of support including strong species-environment relationships, predictable patterns of dominant competitors and strong responses to seasonality. Despite stronger support for selection, a number of predictions of drift were supported as well. It is likely that a combination of both of these processes operates across the variable envi- ronments experienced in Atlantic Forest. Predictions of both processes are difcult to make operational. Support for drift often comes from failure to demonstrate a signifcant pattern and should not be considered strong support of a prediction. Similarly, many predictions of selection predict phenotypic patterns among species without specifying a particular trait. This is problematic because the phenotype is multifaceted and a lack of pattern in one measured trait might mask a strong pattern in some other unmeasured trait. Distilling mechanisms of community organization into four higher level processes is a substantial innovation in community ecology. Nonetheless, eforts need to be made to develop a suite of mutually exclusive and falsifable predictions to facilitate future and more rapid understanding of community organization. Keywords Atlantic forest · Bats · Community structure · Ecological drift · Neutral theory · Selection Introduction Attempts to understand the relative contributions to com- munity organization of diferent ecological mechanisms has been an active area of research for decades. For example, determination of the mechanistic basis to the structure of communities, especially in the Neotropics, has been the focus of hundreds of published manuscripts (Carson and Schnitzer 2008, Mittelbach and McGill 2019, Stevens and Estrada-Villegas 2020). Despite much research, little con- sensus exists regarding the relative contributions of diferent ecological processes on the structure of bat communities. Recently, Vellend (2016) has distilled all disparate lower- level processes potentially structuring communities as the product of the four higher-level processes of dispersal, drift, Communicated by Thomas Lilley. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00442-020-04683-z) contains supplementary material, which is available to authorized users. * Richard D. Stevens richard.stevens@ttu.edu 1 Department of Natural Resources Management, Texas Tech University, Lubbock, TX 79409, USA 2 Museum of Texas Tech University, Lubbock, TX 79409, USA