Skipping across the tropics: The evolutionary history of sawtail surgeonfishes (Acanthuridae: Prionurus) William B. Ludt a,⇑ , Luiz A. Rocha b , Mark V. Erdmann b,c , Prosanta Chakrabarty a a Ichthyology Section, Museum of Natural Science, Department of Biological Sciences, 119 Foster Hall, Louisiana State University, Baton Rouge, LA 70803, United States b Section of Ichthyology, California Academy of Sciences, 55 Music Concourse Dr., San Francisco, CA 94118, United States c Conservation International Indonesia Marine Program, Jl. Dr. Muwardi No. 17, Renon, Bali 80361, Indonesia article info Article history: Received 15 October 2014 Revised 22 December 2014 Accepted 23 December 2014 Available online 14 January 2015 Keywords: Anti-tropical Anti-equatorial Ancestral range Biogeography Systematics abstract Fishes described as ‘‘anti-equatorial’’ have disjunct distributions, inhabiting temperate habitat patches on both sides of the tropics. Several alternative hypotheses suggest how and when species with disjunct distributions crossed uninhabitable areas, including: ancient vicariant events, competitive exclusion from the tropics, and more recent dispersal during Pliocene and Pleistocene glacial periods. Surgeonfishes in the genus Prionurus can provide novel insight into this pattern as its member species have disjunct distributions inhabiting either temperate latitudes, cold-water upwellings in the tropics, or low diversity tropical reef ecosystems. Here the evolutionary history and historical biogeography of Prionurus is examined using a dataset containing both mitochondrial and nuclear data for all seven extant species. Our results indicate that Prionurus is monophyletic and Miocene in origin. Several relationships remain problematic, including the placement of the Australian P. microlepidotus, and the relationship between P. laticlavius and P. punctatus. Equatorial divergence events between temperate western Pacific habitats occurred at least twice in Prionurus: once in the Miocene and again in the late Pliocene/early Pleistocene. Three species with tropical affinities, P. laticlavius, P. punctatus, and P. biafraensis, form a clade that originated in the Pliocene. These results suggest that a variety of mechanisms may regulate the disjunct distribution of temperate fishes, and provide support for both older and younger equatorial crossing events. They also suggest that interspecific competitive exclusion may be influential in fishes with ‘‘anti-equatorial’’ distributions. Ó 2015 Elsevier Inc. All rights reserved. 1. Introduction A variety of terms describe related species that are distributed outside of, and on opposing sides of, the tropics: ‘‘amphitropical,’’ ‘‘bipolar,’’ ‘‘bitemperate,’’ and ‘‘bianti-tropical’’ are some examples. Many of these terms are specific to certain latitudes, but for marine fishes all temperate regions (warm-temperate, temperate, and polar) are encapsulated by the term ‘‘anti-tropical’’ (Hubbs, 1952). Several authors have summarized this distribution pattern over time for fishes (Hubbs, 1952; Randall, 1981; Briggs, 1987; Burridge, 2002). Hubbs’ (1952) original list included mostly pelagic species and was later expanded by the inclusion of many coastal and shallow water species by Randall (1981). Furthermore, Randall (1981) included several species that occurred within tropical latitudes, but which are restricted to cooler waters (such as those found in upwelling areas) whose temperatures are more typical of warm-temperate regions. As these examples did not strictly fit within the term ‘‘anti-tropical’’ as defined by Hubbs (1952), Randall (1981) coined the more inclusive term ‘‘anti- equatorial.’’ The timing and mechanisms responsible for creating these disjunct distributions restricted to colder waters are poorly under- stood. However, several hypotheses have been suggested, includ- ing: (1) dispersal during cooler interglacial periods (Berg, 1933; Ekman, 1953; Lindberg, 1991) or via deeper, cooler waters (Hubbs, 1952; Poortvliet et al., 2013); (2) competitive exclusion of once widespread species by younger species in the tropics isolat- ing ‘‘relict’’ populations at higher latitudes (Théel, 1885; Briggs, 1987); and (3) vicariance – either from the separation of larger supercontinents, or from former continuous habitats being divided during sea-level shifts (Crame, 1993). Importantly, it is possible to distinguish between the likelihood of these various hypotheses by examining the timing of divergence events. Glacial dispersal events http://dx.doi.org/10.1016/j.ympev.2014.12.017 1055-7903/Ó 2015 Elsevier Inc. All rights reserved. ⇑ Corresponding author. E-mail addresses: wbludt@gmail.com (W.B. Ludt), LRocha@calacadamy.org (L.A. Rocha), mverdmann@gmail.com (M.V. Erdmann), prosanta@lsu.edu (P. Chakrabarty). Molecular Phylogenetics and Evolution 84 (2015) 166–172 Contents lists available at ScienceDirect Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev