Hidden diversity in two widespread snake species (Serpentes: Xenodontini: Erythrolamprus) from South America

Highlights

• We infer the largest phylogeny of Erythrolamprus to date.

• Erythrolamprus epinephelus and E. reginae are not monophyletic.

• Subspecies albiventris, fraseri and lamonae are elevated to species.

• E. zweifeli from Trinidad is nested within E. reginae from Amazonian Ecuador.

Abstract

The spatial distribution of genetic diversity of widely distributed Neotropical tetrapods has been an active research field during the last decade, although it has focused on lowland (mostly Amazonian) taxa. In this paper we use phylogenetic analyses to explore the diversity within two widely distributed snake species, Erythrolamprus epinephelus and E. reginae. Specifically, we focus on Andean populations of the former and lowland samples of the latter. Our results show that Erythrolamprus epinephelus is paraphyletic and support recognition of the subspecies albiventris, fraseri and lamonae as distinct species. Regarding E. reginae, our results are in conflict with recent taxonomic proposals in that (1) E. reginae is paraphyletic, and (2) E. zweifeli from Trinidad and E. reginae from Amazonian Ecuador are not reciprocally monophyletic. Finally, we recommend caution when proposing taxonomic changes based on incomplete geographic and/or character sampling of widespread Neotropical taxa.

Introduction

Widely distributed taxa are great models to study the spatial distribution of genetic diversity in the Neotropics. Several studies on Neotropical lowland vertebrates suggest that many widely distributed species are composed of multiple distinct evolutionary linages (Bittencourt et al., 2019, Domingos et al., 2014, Funk et al., 2011, Nunes et al., 2012, Sousa-Neves et al., 2013, Werneck et al., 2012), some of which represent cryptic species. Interestingly, however, not all lowland species with broad geographic distributions are necessarily reservoirs of cryptic diversity (Cadena et al., 2011, Gehara et al., 2014, Pinto et al., 2019, Torres-Carvajal et al., 2019). Contrastingly, the genetic diversity of widely distributed Andean taxa is less understood. It is generally accepted that the complex orogeny and topography of the Andes facilitated geographic isolation and formed complex environmental gradients, thus promoting lineage diversification (Bonaccorso, 2009, Cadena et al., 2007, Chaves et al., 2011). Consequently, Andean vertebrates with low vagility, such as amphibians and reptiles, tend to have small distribution ranges (Guayasamin et al., 2010, Kizirian, 1996, Torres-Carvajal, 2007). In fact, a few studies have discovered cryptic lineages within widespread Andean species (Coloma et al., 2010, Páez and Ron, 2019, Páez-Moscoso and Guayasamin, 2012, Torres-Carvajal and Mafla-Endara, 2013). Nonetheless, the genetic diversity of many widespread Andean species remains unknown.

With 51 species, Erythrolamprus Boie 1826 is one of the most species-rich snake clades recognized as a genus in the Neotropics (Ascenso et al., 2019, Murphy et al., 2019). They occur from Honduras southward to Argentina, on both sides or along the Andes, throughout the Amazon basin, and in the Lesser Antilles, between 0 and 3500 m in elevation. They are medium-sized (∼50–150 cm total length), diurnal snakes, with fossorial, terrestrial, or semi-aquatic habits. Remarkably, some species (e.g., E. bizona Jan 1863, E. mimus Cope 1869) are coral snake mimics (Savage, 2002). One of the most widespread and variable species of Erythrolamprus is E. epinephelus Cope 1862. It occurs from Costa Rica and Panama to northern Peru, and it has the greatest altitudinal distribution among species of Erythrolamprus, from lowlands west of the Andes to highlands above 3000 m throughout the Andes of Venezuela, Colombia and Ecuador. Dixon (1983a) presented a systematic study of E. epinephelus (as ‘Liophis’), in which he recognized eight subspecies based on squamation, color patterns and maxillary teeth—epinephelus, albiventris Jan 1863, fraseri Boulenger 1894, bimaculatus Cope 1899, opisthotaenia Boulenger 1908, pseudocobella Peracca 1914, juvenalis Dunn 1937 and lamonae Dunn 1944. Even though most squamation characters overlap among subspecies, Dixon (1983a) recognized some striking differences in color patterns. For example, some have red and black bands, whereas others have a green dorsum with 2–4 longitudinal black stripes posteriorly; some have uniform yellow venters, and others have checkered ventral patterns. The most widespread species of Erythrolamprus is E. reginae Linnaeus 1758, which occupies most of cis-Andean South America from Colombia and Venezuela to Argentina, as well as the island of Trinidad, between 0 and 2000 m. As with E. epinephelus, Dixon (1983b) recognized four subspecies of E. reginae based on scale counts and color patterns—macrostoma Amaral 1935, reginae, semilineata Wagler 1824 and zweifeli Roze 1959. The latter has been recently proposed as a valid species from northern Venezuela and Trinidad based on morphology, coloration and phylogenetic relationships, although with limited taxon sampling (Murphy et al., 2019, Rivas et al., 2012).

The genetic diversity of widely distributed species of Erythrolamprus remains poorly explored. In this paper, we studied lineage boundaries among some of the subspecies of two widely distributed (both in terms of area and elevation) South American snake species, Erythrolamprus epinephelus and E. reginae. We analyzed the phylogenetic relationships among 24 (∼50%) species of Erythrolamprus including new samples from Ecuador of three subspecies of E. epinephelus, as well as E. reginae semilineata. Using the largest data sampling to date—four mitochondrial and two nuclear gene regions—, we (1) assessed the monophyly of both Erythrolamprus epinephelus and E. reginae, as well as each of the included subspecies; (2) identified lineages that could represent undescribed species; and (3) commented on current taxonomy based on the recovered phylogeny.

The species name epinephelus has been misspelled as epinephalus by some authors (Dunn, 1937, Murphy et al., 2019, Savage, 2002). Interestingly, this confusion started with the original description of the species, in which the name was spelled as both epinephelus and epinephalus (p. 76 and 78, respectively, in Cope, 1862). However, in the same volume of the Proceedings of the Academy of Natural Sciences of Philadelphia where this species was described, the name epinephalus was explicitly recognized as a misspelling of epinephelus in the “Errata and Addenda” section (p. 594). Thus, the correct spelling is epinephelus following Article 32.5.1.1 of the International Code of Zoological Nomenclature (International Commission on Zoological Nomenclature, 1999).