Contents lists available at ScienceDirect Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev Drivers of sociality in Gobiodon fishes: An assessment of phylogeny, ecology and life-history Martin L. Hing a, ⁎ , O. Selma Klanten b , Marian Y.L. Wong a , Mark Dowton c a Centre for Sustainable Ecosystems Solutions, School of Earth, Atmospheric and Life Sciences, University of Wollongong, Australia b Fish Ecology Laboratory, School of Life Sciences, University of Technology Sydney, Australia c Molecular Horizons, School of Chemistry and Molecular Bioscience, University of Wollongong, Australia ARTICLE INFO Keywords: Gobiodon Coral goby Sociality Phylogeny Phylogenetic signal Phylogenetic generalized least squares Ecological factors Life-history factors Gobiodon spilophthalmus c.f ABSTRACT What drives the evolution of sociality in animals? Many robust studies in terrestrial organisms have pointed toward various kinship-based, ecological and life-history traits or phylogenetic constraint which have played a role in the evolution of sociality. These traits are not mutually exclusive and the exact combination of traits is likely taxon-specific. Phylogenetic comparative analyses have been instrumental in identifying social lineages and comparing various traits with non-social lineages to give broad evolutionary perspectives on the develop- ment of sociality. Few studies have attempted this approach in marine vertebrate systems. Social marine fishes are particularly interesting because many have a pelagic larval phase and non-conventional life-history strategies (e.g. bi-directional sex-change) not often observed in terrestrial animals. Such strategies provide novel insights into terrestrially-derived theories of social evolution. Here, we assess the strength of the phylogenetic signal of sociality in the Gobiodon genus with Pagel’s lambda and Blomberg’s K parameters. We found some evidence of a phylogenetic signal of sociality, but factors other than phylogenetic constraint also have a strong influence on the extant social state of each species. We then use phylogenetic generalized least squares analyses to examine several ecological and life-history traits that may have influenced the evolution of sociality in the genus. We found an interaction of habitat size and fish length was the strongest predictor of sociality. Sociality in larger species was more dependent on coral size than in smaller species, but smaller species were more social overall, regardless of coral size. Finally, we comment on findings regarding the validity of the species G. spilophthalmus which arose during the course of our research. These findings in a group of marine fishes add a unique per- spective on the evolution of sociality to the excellent terrestrial work conducted in this field. 1. Introduction The question of how sociality first arose in animals has attracted much attention in the fields of evolutionary ecology and animal beha- viour. Many mechanisms are thought to contribute to the evolution of sociality including ecological factors, life-history traits and phylogeny (Arnold and Owens, 1998; Emlen, 1982; Hamilton, 1964; Hatchwell and Komdeur, 2000; reviewed in Hing et al., 2017; Kokko and Ekman, 2002). These features are not mutually exclusive and may be highly dependent on each other (Arnold and Owens, 1998; Chapple, 2003). Hamilton’s rule predicts that sociality should evolve under certain combinations of relatedness and costs and benefits of social actions and is widely regarded as a universal framework to study social evolution (Bourke, 2014; Hamilton, 1964). Ecology, life-history and relatedness change the costs and benefits conferred to individuals within the group. Under this framework, individuals should receive greater inclusive fit- ness benefits if they form social groups with close relatives (Briga et al., 2012; Hughes et al., 2008). Groups consisting of unrelated individuals are also possible if ecological or life-history factors alter the direct costs and benefits of group living such that the benefits outweigh the costs (e.g. Buston et al., 2007; Riehl, 2011). Phylogenetic relationships among taxa can constrain the evolution of sociality which may predispose species to sociality (e.g. Agnarsson, 2002; Nowicki et al., 2018; Schneider and Kappeler, 2014; Smorkatcheva and Lukhtanov, 2014). However, the extant state of so- ciality may depend on various ecological and life-history conditions (Chapple, 2003; Rubenstein and Lovette, 2007; Schürch et al., 2016). For example, altered environmental conditions and extreme weather events could reduce habitat sizes for a normally social species, in- creasing animal density and increasing conflict within the group https://doi.org/10.1016/j.ympev.2019.05.020 Received 9 February 2019; Received in revised form 20 May 2019; Accepted 20 May 2019 ⁎ Corresponding author. E-mail address: martinhing@gmail.com (M.L. Hing). Molecular Phylogenetics and Evolution 137 (2019) 263–273 Available online 21 May 2019 1055-7903/ © 2019 Elsevier Inc. All rights reserved. T