Muscular evolution of hemipenis in Imantodini snakes (Squamata: Dipsadidae)

Muscular evolution of hemipenis in Imantodini snakes (Squamata: Dipsadidae) Abstract The anatomy of squamates’ hemipenis has historically been an important source of characters for phylogenetic and taxonomic studies. Herein, we use new comparative evidence from the retractor penis magnus (RPM) muscle to revisit Myers models of hemipenial evolution and current homology hypotheses for the sulcus spermaticus of Imantodini (Imantodes, Leptodeira). We compare the topological and connectivity relationships of RPM muscle to other hemipenial structures in 35 species of Imantodini, all tribes of Dipsadidae, and other alethinophidian lineages, analysing everted, inflated, dissected and stained hemipenes. To trace the evolution of hemipenial morphology in dipsadids, we performed a maximum likelihood analysis of a molecular data set with 45 species from the aforementioned clades. These analyses supported the hypothesis that the morphology of the RPM and the sulcus evolved at the same nodes in dipsadid phylogeny. The morphology of this muscle in Leptodeira is similar to that in other taxa, with a bifurcated sulcus spermaticus, supporting the hypothesis that their condition evolved from an ancestral state of a bifurcated sulcus. This unexpected anatomical relationship is congruent with indirect embryonic data published elsewhere. We conclude that the sulcus in Imantodini evolved together with the RPM, independent of evolutionary changes in other structures. comparative anatomy, development, Macrostomata, Serpentes, Squamata INTRODUCTION The male reproductive organ of Squamata, hemipenis, is a hollow structure with an external sulcus spermaticus where the seminal fluids flow, and an internal space housing spongy tissues and musculature (Dowling & Savage, 1960). Hemipenial characters are useful in squamate systematics for differentiating at the species level and higher. Their importance arises from a broad consensus recognizing the similarities between hemipenial parts and a consequent congruence across hypotheses of homology (Patterson, 1982; De Pinna, 1991) and nomenclature of hemipenial structures (Dowling & Savage, 1960; Myers, 1974; Zaher, 1999; Myers & Mcdowell, 2014). The hemipenis was first used as evidence to support squamate evolutionary relationships by Cope (1886, 1900) in his work on reptile classification. From then, multiple authors used hemipenial variation in their own snake classification hypotheses (Dunn, 1928; Vellard, 1928, 1946; Bogert, 1940). The hemipenis became an intrinsic tool for snake phylogeny only after Dowling & Savage (1960) proposed procedures for preparing (evertion and inflation), describing and comparing hemipenial characters. These methods became standard among snake biologists. The external features of hemipenes are surprisingly variable at all taxonomic levels of snakes (e.g. Zaher, 1999; Zaher & Prudente, 2003). Today, descriptions of snake hemipenes and associated discussions of systematics are mostly based on variations in shape (e.g. single or divided), sulcus spermaticus orientation (e.g. centrifugal, centrolineal position) and morphology (e.g. simple or bifurcated), ornamentation (e.g. calyces, flounces, papillae, spines) and apical differentiations (e.g. apical disk, capitulum, capitular groove) (Dowling & Savage, 1960; McDowell, 1961; Savage, 1997; Zaher, 1999; Myers & McDowell, 2014). Despite the importance of the hemipenis in snake classification, few detailed descriptions and comparisons have been carried out on its embryology (Clark, 1945; Raynaud & Pieau, 1985; Gredler et al., 2014; Leal & Cohn, 2015), muscles and internal tissues (Beuchelt, 1936; Dowling & Savage, 1960; Ziegler & Böhme, 1997; Porto et al., 2013). The studies of hemipenial embryology include eight species representing three lizard families (Anguidae, Lacertidae, Polychrotidae) and four macrostomatan families (Dipsadidae, Lamprophidae, Natricidae, Pythonidae) (Supporting Information, Table S1). These analyses describe and compare the development of the sulcus spermaticus and hemipenial body and muscles. The retractor penis magnus (RPM) muscle and the sulcus spermaticus developed in the same position at the base of the hemipenis and share the growing pattern proximodistally. The RPM developed before the sulcus, and when the muscle bifurcates the sulcus bifurcates soon after (Clark, 1945; Raynaud & Pieau, 1985; Gredler et al., 2014; Leal & Cohn, 2015). These data have been used to review the hemipenial anatomy and to discuss its evolution in the modern phylogenetic framework for snakes. Elsewhere, data comparing embryonic skulls have been used to define and discuss the evolution and relationships of major lineages of snakes (Bellairs & Kamal, 1981; Cundall & Irish, 2008). Ontogenetic and embryonic information in phylogenetic analyses have been proposed for two stages of the analyses: homology hypothesis proposition (including character codification and ordering) and recovering the parsimonious tree as a rooting criteria (De Pinna, 1994; Mabee, 2000). However, it is difficult to including embryonic information in ancestral reconstructions, since it is difficult to adjust the resulting information so it can be included in phylogenetic methods (Alberch, 1985; Bininda-Emonds et al., 2002; Werneburg, 2009). The hemipenial musculature is defined by the movements achieved through contraction in the propulsor and the retractor muscles. Three groups of retractor muscles have been described in snakes: retractor lateralis anterior, retractor lateralis posterior and RPM (Dowling & Savage, 1960; Arnold, 1984; Keogh, 1996). The retractor lateralis anterior and posterior originate on the horizontal septum of the caudal vertebrae transverse process. The retractor lateralis anterior is inserted in front of the vent while the lateralis posterior is inserted at the base of the hemipenis. The RPM originates at the transverse process of the caudal vertebrae and inserts into the inner surface of the free distal hemipenial apex (Porto et al., 2013). The RPM is the longest and most important muscle in hemipenial withdrawal. In species with bilobed or divided organs, this muscle is divided at some distance posterior to its attachment to the hemipenis and each branch is inserted into its apex (Dowling & Savage, 1960). In Dipsadidae, one of the largest American radiations of higher snakes, the clade Dipsadinae has been congruently recovered in comparative molecular and morphological analyses (Cadle, 1984a; Jenner & Dowling, 1985; Myers & Cadle, 1994; Kraus & Brown, 1998; Zaher, 1999; Slowinski & Lawson, 2002; Pinou et al., 2004; Vidal et al., 2007; Zaher et al., 2009; Grazziotin et al., 2012). The morphological diagnosis of Dipsadinae includes the following hemipenial characters: hemipenis unilobed or with reduced bilobation, unicapitate, sulcus spermaticus bifurcated either at hemipenial base or after the capitular groove (Myers, 1974; Cadle, 1984b; Zaher, 1999). Dipsadinae includes the tribes Imantodini and Dipsadini and one unnamed clade that includes species of the genus Atractus (Grazziotin et al., 2012). Imantodini includes two genera of Neotropical snakes (Imantodes and Leptodeira), and the clade is supported solely by molecular characters, while interpretation of their hemipenial morphology (single, unilobed, capitate, sulcus spermaticus simple) is subject to discussion (see Myers & McDowell, 2014). Imantodes comprises a group of eight primarily arboreal species that occur at low to moderate elevations (up to 2200 m above sea level) in rainforest and savanna environments (Peters & Orejas-Miranda, 1970; Myers, 1982; Torres-Carvajal et al., 2012; Missassi & Prudente, 2015). Leptodeira includes 11 species and 12 subspecies ranging from southern USA to northern Argentina and Paraguay, the east coast of Brazil and the islands of Aruba, Margarita, Tobago and Trinidad. Some species include geographical variations and are arranged into three groups of subspecies (L. annulata, L. septentrionalis, L. splendida) (Duellman, 1958a, b; Dowling & Jenner, 1987; Mulcahy, 2007; Daza et al., 2009; McCranie, 2011; Reyes-Velasco & Mulcahy, 2010). In Imantodes species, the hemipenes have a single simple sulcus ending at the hemipenial apex (Myers, 2011; Missassi & Prudente, 2015). However, Missassi and Prudente (2015) described a new hemipenis condition for the genus in Imantodes guane: a sulcus spermaticus with an elongated, expanded terminal area. The condition in I. guane may be interpreted as a transitional evolutionary state before the one present in some Leptodeira groups (Myers, 2011; Missassi & Prudente, 2015). In Leptodeira, the hemipenis has a sulcus spermaticus that terminates in an expanded area without the medial sulcar walls and without any ornamentation (Myers, 2011: 22). The homology hypothesis interpreting these conditions in sulcus spermaticus morphology has long been a controversial issue. For some authors, Leptodeira has a simple sulcus spermaticus (Dunn, 1928: 23; Duellman, 1958a: 17; Cadle, 1984b: 24; Dowling & Jenner, 1987: 198), whereas for others it is bifurcated (Underwood, 1967: fig. 10; Dowling, 1975: 198; Myers & Cadle, 1994: 27; Savage, 2002: 610). The proposed models for loss of bifurcation in the sulcus spermaticus hemipenial evolution in dipsadid snakes (Myers, 2011) stated that the most probable morphological scenario for the condition in Leptodeira might be the shortening and/or loss of a single hemipenial lobe combined with the ‘weakening’ and loss of the medial sulcus walls, ‘leaving only a tissue divide separating the lateral branches in situ; ornamentation such as calyces between the branches may also be lost during this process’. Here, we review the hemipenial variation in species of Imantodini using the available embryonic data and a new data set of hemipenial RPM of dipsadids to reconstruct the ancestral state of Imantodini and discuss the evolution of the muscle and the sulcus spermaticus in Imantodes and Leptodeira. The embryo data was used to propose the homology hypothesis and character codification for the RPM and the sulcus spermaticus. The comparative evidence for the RPM was used to revisit Myers’s models for dipsadids hemipenial evolution and current homology hypotheses for the sulcus spermaticus in Imantodini. Were described the topological and connectivity relationships of the RPM with other hemipenial structures and compared it with those of representative taxa from major dipsadid lineages to discuss the evolution of the sulcus spermaticus in Imantodini. MATERIALS AND METHODS Taxonomic sampling and DNA data set Our sampling was based on the taxonomic arrangements of Zaher et al. (2009) and Grazziotin et al. (2012) and included species representing the following clades of alethinophidian snakes: Aniliidae, Boidae, Colubridae, Dipsadidae (Dipsadini, Elapomorphini, Echinantherini, Imantodini, Philodryadini, Pseudoboini, Psomophiini, Tachimenini, Xenodontini), Elapidae and Viperidae. We selected 35 species to represent these clades in the comparative analyses of hemipenial anatomy (Appendix). In addition, we created a DNA character matrix using the available data from GenBank for 45 species of the aforementioned alethinophidian clades. We gathered DNA data from gene fragments of four nuclear (c-mos, bdnf, ntf3, rag1) and four mitochondrial genes (cytb, nd4, 12s, 16s) (Supporting Information, Table S1). Hemipenial nomenclature We used the following hemipenial nomenclature proposed by Arnold (1986), Dowling & Savage (1960) and Zaher (1999): asulcate face – on the hemipenis external layer, the side without the sulcus spermaticus; absulcate layer – the functionally internal surface of the everted hemipenis where the RPM attaches; external layer – hemipenial surface composed of epidermis and dermis; inner layer – internal tissues of hemipenis including the blood and lymphatic sinuses and muscle; base – proximal region to the cloaca and usually constricted compared to the hemipenial body; apical end – free tip of the hemipenis; lobe – division of the hemipenial body; capitulum – apical portion separated from the hemipenial body by a deep groove or capitular groove; sulcus spermaticus – longitudinal groove on the external layer of the everted hemipenis; RPM – muscle with origin on the caudal vertebrae and with insertion on the apical end of the hemipenis; RPM insertion – inside the absulcate layer; RPM origin – on the transverse process of the caudal vertebrae. We also propose a new term, cavitas hemipenis, to describe the internal central space created and delimited by the absulcate layer, containing the distal portion of the RPM when the hemipenis is everted. Eversion, dissection and staining We used two complementary hemipenial preparation techniques: fully everted and everted-inflated. As stated by Myers & Cadle (2003), there is a difference between hemipenis preparations when ‘fully everted’, as opposed to ‘maximally expanded or inflated’. In a maximally expanded preparation, the RPM muscle is sectioned at its insertion and the hemipenis is everted and inflated to the fullest extent allowed by its original elasticity, which can vary when working with preserved specimens (Myers & Cadle, 2003; Zaher & Prudente, 2003). The hemipenis was everted and inflated following the techniques proposed by Zaher & Prudente (2003). We used the fully everted hemipenis of preserved museum specimens to analyse the topology and connectivity relationships between hemipenial internal and external parts (RPM, sulcus spermaticus, capitular groove, apical end). Analysing the hemipenis in this condition provided a view of the muscle groups, ornamentation structures and the sulcus spermaticus, including its terminal region (Myers & Cadle, 2003). To access RPM morphology in the everted hemipenis, we dissected it in the following manner: (1) open the ventral surface of the tail at the midline, posterior to the cloaca; (2) isolate the hemipenis and the RPM from surrounding tissues and cut the muscle close to its origin; (3) remove the organ from the specimen; and (4) dissect the asulcate surface at its medial side, following the longitudinal axis from the hemipenial base to its apical end. Afterwards, we stained each hemipenis with an alizarin-ethanol solution (Springer & Johnson, 2000). This procedure creates contrast between tissues and highlights RPM fibres, facilitating the recognition and description of the structures. The absulcate surface is surrounded by tissue walls, which prevent the diffusion of the staining solution over muscular tissue. We dissected the tissue of the lymph and blood sinuses to expose the muscles to the staining solution. Phylogenetic analysis We generated a phylogenetic hypothesis based on our sampling of the morphological characters to reconstruct the ancestral state at the node Imantodini, instead of using the large sampled matrices available (e.g. Pyron, Burbrink & Wiens, 2013; Figueroa et al., 2016). Using these matrices, or their recovered trees, as the basis for reconstructing the ancestral states would have generated a large amount of missing data at our morphological matrix and undesirable noise in the reconstruction at the Imantodini node. We generated a maximum likelihood (ML) phylogenetic hypothesis of dipsadid snakes based on the molecular data available from GenBank that included, whenever possible, species for which we have hemipenial morphology (Supporting Information). As the data available on GenBank were not complete for the selected species and genera, we included sequences from congeneric species that shared at least one gene with our ‘morphological terminal’. Thus, hemipenial morphology was available for 23 out of the 45 taxa used in our phylogenetic analysis. The only taxon without an available sequence was Apostolepis quinquelineata. Therefore, for the purposes of the analysis, we considered the sequence for Apostolepis dimidiata as equivalent to A. quinquelineata, placing them at the same terminal. The proportion of taxon represented within each gene matrix varied: 82% of the 45 taxon had c-mos, 49% had bdnf and ntf3, 33% had rag1, 87% had cytb, 51% had nd4, 87% had 12s and 89% had 16s. We downloaded and visualized all sequence fragments with Geneious v 9.1.6 (Kearse et al., 2012). We aligned the sequences with MAFFT v 7.222 (Katoh et al., 2002), using the FFT-NS-I algorithm with default parameters, by c-mos, bdnf, ntf3, cytb, and E-INS-i by rag1, nd4, 12s, 16s. The sequences were concatenated with SequenceMatrix v 1.7.8 (Vaidya, Lohman & Meier, 2011). The best scheme for partitions was selected using PartitionFinder v1.1.1 (Lanfear et al., 2012). The decision criterion was set to BIC (Bayesian Information Criterion), the search algorithm set to ‘greedy’ and branch lengths set to ‘linked’. ML analyses were conducted with the CIPRES Science Gateway (Miller, Pfeiffer & Schwartz, 2010) using RAxML 8.0 (Stamatakis, 2014). We recovered the best ML tree by running 100 independent searches and verified similarity in the likelihood values, thus avoiding the use of suboptimal trees. Statistical support for clades was obtained from 1000 non-parametrical pseudoreplications. Clades with bootstrap values higher than 70% were considered well supported (Hillis & Bull, 1993). Ancestral state reconstructions We used our phylogenetic framework to reconstruct the evolutionary history of the external and internal hemipenial morphology using parsimony (PA) and ML methods. PA reconstructions were performed with states ordered and unordered and with all transformations equally weighted, using Mesquite v2.75 (Maddison & Maddison, 2011). ML reconstructions were also performed in Mesquite v2.75, using the Markov k-state 1 parameter (Mk1) model of evolution (Schluter et al., 1997; Pagel, 1999), which gives equal probabilities for changes between any two character states. Based on the available embryonic data (Supporting Information, Table S1) and our comparative anatomy results, we coded two characters for the sulcus spermaticus and three for the RPM muscle. For some taxa with missing data, we included the descriptions by Zaher (1999). These are our hypotheses of homology and their character states: Sulcus spermaticus condition: 0 – simple, 1 – bifurcated and 2 – bifurcated but ending in expanded area without ornamentation (ordered linear additive – during the embryology the sulcus is a simple structure that bifurcates subsequently in the taxa with two branches). Sulcus spermaticus length on the everted and inflated hemipenis: 0 – extended from hemipenial base to the apical end, and 1 – extended from hemipenial base to the medial portion of hemipenial body and never reaching apical end. RPM muscle in cavitas hemipenis, morphology: 0 – single and 1 – bifurcated. RPM in absulcate layer, insertion position: 0 – at the distal end of hemipenis and 1 – before the apical end of hemipenis. RPM in cavitas hemipenis, muscle fibres bifurcation position to the position of the sulcus spermaticus bifurcation: 0 – before bifurcation of the sulcus, 1 – underneath the bifurcation of the sulcus and 2 – after the bifurcation of the sulcus. RESULTS Hemipenial morphology in Imantodini In an inflated state, L. annulata subspecies had a single, cylindrical, slightly curved, capitate and calyculate hemipenis (Fig. 1). Capitulum was longest on the asulcate face, with calyces and small spines mineralized at the corners of calyces walls (Fig. 1A). Calyces and spines altered in shape from the capitular groove to the apical end. The calyces increased in length while the spines reduced in height. Capitular groove deep and well defined on sulcate face (Fig. 1A) and interrupted at the asulcate face by perpendicular folds continuous with two paired rows of spines (Fig. 1B). The capitular groove between the two folds was deeper. The sulcus spermaticus single, with spinules from the base of organ to the capitular groove, terminating soon after crossing the capitular sulcus, on the base of capitulum (Fig. 1A). The sulcus spermaticus lacked lips at its terminus, ending in a large area that is without ornamentation (Fig. 1C). The end of the sulcus did not reach the apical end of hemipenis. Hemipenial body had larger, hooked and mineralized spines, arranged in parallel transverse rows that followed the hemipenial body laterally (Fig. 1A); at the sulcate face, the spines reducing in length and height towards the base of the organ and the capitular groove. Asulcate face contained two longitudinal and parallel rows of large, hooked spines, separated by a broad, denuded central region (Fig. 1B). Base of the organ covered by small spines of different sizes that were irregularly arranged, on both faces. Figure 1. View largeDownload slide Everted and inflated hemipenis in adults of Imantodini. A–C, Leptodeira a. annulata (MPEG 20016) sulcate face, asulcate face and naked area in detail. D, E, Imantodes cenchoa (MPEG 17442). Abbreviations: cg, capitular groove; na, naked area; ss, sulcus spermaticus. Scale bar: A, B, D, E = 5.0 mm; C = 2.0 mm. Figure 1. View largeDownload slide Everted and inflated hemipenis in adults of Imantodini. A–C, Leptodeira a. annulata (MPEG 20016) sulcate face, asulcate face and naked area in detail. D, E, Imantodes cenchoa (MPEG 17442). Abbreviations: cg, capitular groove; na, naked area; ss, sulcus spermaticus. Scale bar: A, B, D, E = 5.0 mm; C = 2.0 mm. The everted dissected hemipenis in all L. annulata subspecies (Appendix) showed RPM fibres flexed at a 45° angle at the base of the hemipenis. The single RPM was arranged centrally in the cavitas hemipenis (Fig. 2A). Apically, the RPM was bent and inserted at the medial region of the capitulum underneath the distal end of the naked area of the sulcus spermaticus (Fig. 2B). At its insertion, the RPM formed large, fan-like fascia, bilaterally arranged. The side facing the hemipenial asulcate face (Fig. 2A) had fibres disposed continuously, forming a short longitudinal array. On the side opposite of the sulcate surface, the muscular fibres of the RPM divided in two groups by short fascia (Fig. 2B). In L. annulata subspecies, the RPM was bifurcated at its insertion forming two muscle anchorage areas. Figure 2. View largeDownload slide Everted and dissected hemipenis in adults of Imantodini. A, B, Leptodeira a. ashmeadi (MZUSP 10654). C, D, Imantodes cenchoa (MPEG 26270). Abbreviations: cg, capitular groove; in, insertion of RPM; rpm, retractor penis magnus muscle; rpp, retractor penis parvus muscle; ss, sulcus spermaticus. Scale bar: 5.0 mm. Figure 2. View largeDownload slide Everted and dissected hemipenis in adults of Imantodini. A, B, Leptodeira a. ashmeadi (MZUSP 10654). C, D, Imantodes cenchoa (MPEG 26270). Abbreviations: cg, capitular groove; in, insertion of RPM; rpm, retractor penis magnus muscle; rpp, retractor penis parvus muscle; ss, sulcus spermaticus. Scale bar: 5.0 mm. Imantodes cenchoa (Fig. 1D, E) and I. lentiferus had the hemipenis unilobed, capitate and calyculate; calyces in transverse rows, conspicuous in sulcate, lateral, and asulcate sides of capitulum; asulcate side with small naked pocket located at the base of capitulum; asulcate side with two rows of small spines separating capitulum from the hemipenial body; sulcus spermaticus single, reaching the apical end of the hemipenis, sulcus margins bordered with spinules from the base of hemipenial body to the beginning of capitulum; hemipenial body with four rows of large spines disposed transversally and dispersed spinules concentrated on the basal region of organ. Both species had a single RPM muscle in cavitae hemipenes with its insertion at the apical end and underneath the terminus of the sulcus spermaticus (Fig. 2C, D). Hemipenial morphology in Dipsadidae and other alethinophidians In other groups of species of Dipsadinae (Supporting Information, Appendices S1–S3), including representatives of the genus Atractus and species of the tribe Dipsadini, the hemipenis had two short lobes with a bifurcated sulcus spermaticus. In these species, the sulcus ended at the apical end of the hemipenis. The RPM in the cavitas hemipenis bifurcated underneath the bifurcation of the sulcus spermaticus (Atractus albuquerquei, A. latifrons, Dipsas catesbyi, D. indica, Sibynomorphus mikanii) (see Fig. 3), or after the sulcus bifurcation (Sibon nebulatus). Each muscle ran through the lobe and inserted at the apical end of the hemipenis underneath the end of the sulcus spermaticus. Close to the insertion region, the fibres transformed into short fascia with a fan-like shape, but at the insertion the RPM remained a single muscle. Figure 3. View largeDownload slide Everted and dissected hemipenis in adults of alethinophidian families. A, Viperidae (Bothrops atrox, MPEG 18985). B, Elapidae (Micrurus lemniscatus, MPEG 26392). C, Colubridae (Tantilla melanocephala, MPEG 24568). D, Dipsadidae (Taeniophallus brevirostris, MPEG 5033). Abbreviations: cg, capitular groove; in, insertion of the RPM; rpm, retractor penis magnus muscle; ss, sulcus spermaticus. Scale bar: 5.0 mm. Figure 3. View largeDownload slide Everted and dissected hemipenis in adults of alethinophidian families. A, Viperidae (Bothrops atrox, MPEG 18985). B, Elapidae (Micrurus lemniscatus, MPEG 26392). C, Colubridae (Tantilla melanocephala, MPEG 24568). D, Dipsadidae (Taeniophallus brevirostris, MPEG 5033). Abbreviations: cg, capitular groove; in, insertion of the RPM; rpm, retractor penis magnus muscle; ss, sulcus spermaticus. Scale bar: 5.0 mm. In species of Xenodontinae, the hemipenes were simple (e.g. A. quinquelineata, Taeniophallus brevirostris, T. nicagus) or bilobed (e.g. Erythrolamprus reginae, Oxyrhopus guibei, O. petolarios, Phylodryas olfersii, Psomophis jobarti, Pseudoboa neuwiedii, Xenopholis scalaris) (Fig. 3). But in all taxa the sulcus spermaticus was bifurcated, except in Taeniophallus nicagus, which had a simple sulcus. In all taxa the single RPM was bifurcated distally in two main muscles, except in T. nicagus, where the RPM remained a single muscle. The position of the RPM bifurcation varied. For a subgroup of xenodontines (Philodryas olfersii, P. neuwiedii, Psomophis joberti, X. scalaris), the bifurcation point was underneath the bifurcation of the sulcus spermaticus. In another subset (A. quinquelineata, E. reginae, Oxyrhopus petolarius, T. brevirostris), the division was after the sulcus bifurcation. However, in all species, the RPM inserted by short fascia at the apical region of the hemipenis underneath the end of the sulcus spermaticus. At its insertion the RPM did not divide. The representative species of Aniliidae (Anilius scytale), Boidae (Corallus hortulanus), Elapidae (Micrurus lemniscatus) and Viperidae (Bothrops atrox) had a hemipenis with a bilobed body and bifurcated sulcus spermaticus. The colubrid species, Chironius exoletus, Leptophys ahaetulla and Tantilla melanocephala, had a simple hemipenis with a single sulcus spermaticus. However, in all species of these families, the sulcus spermaticus ended at the apical end of the hemipenis. In A. scytale, the RPM was formed by two separate muscle groups. In the cavitas hemipenis, each muscle group was inserted at the end of each lobe underneath the terminus of the sulcus. In boid, elapid, viperid and colubrid species, the RPM was a single muscle at its origin, but it was divided in two at the cavitas. In the boid species C. hortulanus and the viperid B. atrox (Fig. 3A), the bifurcation of the muscle was underneath the bifurcation point of the sulcus spermaticus. In the elapid M. lemniscatus, the division of the RPM occurred before the bifurcation of the sulcus (Fig. 3B). In colubrid species (C. exoletus, L. ahaetulla, T. melanocephala) (Fig. 3C), the RPM was a single muscle at the cavitas hemipenis and inserted at the apical portion of the hemipenis. There was a short fascia inserted the muscle underneath the end of the single sulcus spermaticus. Phylogenetic relationships of Dipsadidae Our recovered tree is congruent with the hypotheses generated with larger matrices, such as in research by Grazziotin et al. (2012), Pyron et al. (2013) and Figueroa et al. (2016). Our hypothesis for dipsadid interrelationships was based on the ML analysis of 4906 characters over 45 species (Fig. 4). We recovered monophyly for Dipsadidae (bootstrap > 70) and four additional clades: Thermophis species (bootstrap = 100); Heterodon, Farancia and Diadophis (bootstrap = 32); Dipsadinae (bootstrap = 100); and Xenodontinae (bootstrap = 100). Thermophis appeared as a sister clade to all dipsadids (bootstrap < 70), and the clade composed of Heterodon, Farancia. Diadophis appeared as a sister group of Dipsadinae and Xenodontinae (bootstrap > 70). Thus, the phylogenetic position of Diadophis, Farancia, Heterodon and Thermophis is unstable and controversial. Supports were above 95 for the Dipsadinae clade and its internal relationships among Imantodini, Dipsadini and the ‘Unamed Clade’ (grouping Atractus species, see Grazziotin et al., 2012). The phylogenetic relationships among dipsadines revealed Imantodini as a sister group to the clade comprising Dipsadini together with the ‘Unamed Clade’. Figure 4. View largeDownload slide Phylogenetic relationship of the family Dipsadidae based on the maximum likelihood analysis of molecular data and the variation of hemipenial forms and sulcus spermaticus. Bootstrap values next to nodes. Figure 4. View largeDownload slide Phylogenetic relationship of the family Dipsadidae based on the maximum likelihood analysis of molecular data and the variation of hemipenial forms and sulcus spermaticus. Bootstrap values next to nodes. Ancestral state reconstructions in Imantodini In reconstructions of PA and ML, a hemipenis with a bifurcated sulcus was the ancestral condition for Imantodini (Fig. 5A; Supporting Information, Fig. S1A). A simple sulcus arose independently in Imantodes species and T. nicagus (Echiantherini). The L. annulata group evolved an autopomorphic state for all dipsadids with a sulcus ending in an expanded area. PA and ML reconstructed the RPM for the ancestral node of Imantodini in a bifurcated state (Fig. 5B; Supporting Information, Fig. S1B). The RPM evolved into one muscle independently in Imantodes taxa and T. nicagus. Species of the L. annulata group retained the plesiomorphic state of a bifurcated muscle. Figure 5. View largeDownload slide Mirrored parsimony character reconstructions in Dipsadidae. Circles on nodes denote the reconstructed states for the characters. A, sulcus spermaticus condition. B, muscle retractor penis magnus in cavitashemipenis. C, sulcus spermaticus length. D, insertion of the muscle retractor penis magnus. Figure 5. View largeDownload slide Mirrored parsimony character reconstructions in Dipsadidae. Circles on nodes denote the reconstructed states for the characters. A, sulcus spermaticus condition. B, muscle retractor penis magnus in cavitashemipenis. C, sulcus spermaticus length. D, insertion of the muscle retractor penis magnus. As for length of the sulcus spermaticus, in species of Leptodeira, PA and ML reconstructed a sulcus extending from the base of the hemipenis to the medial portion of the hemipenial body (Fig. 5C; Supporting Information, Fig. S1C). The insertion of the RPM showed a similar phylogenetic history, reconstructing as inserted at the apical end at the Dipsadidae node (however, ML reconstruction was unable to calculate its likelihood), with a single transformation in Leptodeira to an insertion before the apical end (Fig. 5D; Supporting Information, Fig. S1D). DISCUSSION Our comparative analysis of the RPM muscle in Dipsadidae supports the hypothesis that the condition of sulcus spermaticus in Imantodini evolved from an ancestral bifurcated state. Imantodes and some Leptodeira species seem to share a simple, unbranched sulcus spermaticus. However, the species of the L. annulata group as well as L. septentrionalis had an expanse without sulcar walls at the terminus of the sulcus (Myers, 2011). In Leptodeira taxa, the RPM is divided distally into two muscles and inserts underneath the expanded naked area after the end of the sulcus spermaticus. In all other snakes, the RPM is inserted below the apical end of the sulcus spermaticus. All taxa with bifurcated sulcus have a divided RPM muscle. Thus, the morphology and position of the insertion of the RPM in Leptodeira are very similar to that observed in other species with bifurcated sulcus spermaticus (Fig. 2A–C). The sulcus spermaticus in Leptodeira terminates close to the capitular groove in an expanded naked area without medial sulcar lips (Myers, 2011: 22). In Imantodes, the morphology of the sulcus and the RPM is different from that in Leptodeira. For Imantodes, a single sulcus extends over the whole hemipenial length with its terminus at the apical end; at its terminus it is not expanded, retaining sulcar lips. Its single RPM muscle is inserted at the apical region, underneath the sulcus terminus. Underwood (1967: 45) was the first to ascertain that the apomorphic condition of the sulcus in Leptodeira evolved by reduction of a plesiomorphic bifurcated sulcus. Myers (2011: 23) refined and extended Underwood’s homology hypothesis, proposing at least four different evolutionary models for the loss of bifurcation in dipsadid snakes: (1) simultaneous shortening of both branches, probably associated with reduction or loss of hemipenial bilobation; (2) shortening of a single branch; (3) shortening and/or loss of a single hemipenial lobe; (4) weakening and loss of the medial sulcus lips, leaving a dividing tissue between lateral branches, with the subsequent or simultaneous loss of ornamentation between the branches. Our results suggest that in the L. annulata group, the expanded naked area is homologous to the terminus of a bifurcated sulcus, with reduced or undeveloped medial sulcar walls or lips. Myers (2011, 2014: 65) suggested that models 1 and 4 apply to the evolution of the sulcus in Leptoderia. He assumed that the ancestral plesiomorphic state was a bifurcated sulcus that evolved through the shortening of both branches in conjunction with the atrophy of tissue forming the medial sulcar lips. Comparative anatomy of the RPM in snakes verified the bifurcation and insertion of the RPM in the same region of bifurcation and terminus of the sulcus spermaticus, providing evidence for the topographic relationship and connectivity between sulcus and this muscle. Our reconstructions of the evolutionary history of the sulcus spermaticus and RPM muscle in Dipsadidae show that both evolved simultaneously in the tree (Fig. 5). In Leptodeira, the insertion of the RPM and the terminus of the sulcus spermaticus shifted to the middle hemipenial region. In addition, the reconstructions show (Fig. 5A, B) that in the species of Imantodes, T. nicagus and the observed Neotropical species of Colubridae, the lack one of the branches of the sulcus coincides with the lack in these taxa of one of the RPM muscles. Imantodes guane is the only species of Imantodes that presents hemipenial morphology with an expanded terminal area, which may be interpreted as homologous with that found in some Leptodeira species (Missassi & Prudente, 2015). In their description of I. guane, Missassi and Prudente found an expansion of the sulcus spermaticus at the apical region beyond the capitular groove, and interpreted this as a new condition for the genus. The terminus of the sulcus of I. guane is at the apex of the apical region, where it preserves the sulcar walls (Prudente & Missassi, 2015: fig. 6). In Leptodeira, medial position and loss of sulcar walls or ‘sulcus lips’ are morphological definitions of their expanded naked area. Since we do not have descriptions of the RPM and its morphology at the cavitas hemipenis of I. guane, we can only surmise that its naked area is not homologous to that in Leptodeira. We observed indirect evidence for the associated evolution of RPM and sulcus spermaticus morphology in T. nicagus and in Neotropical species of Colubridae. Species of the tribe Echinantherini have a hemipenis with a simple body and bifurcated sulcus spermaticus, with the exception of T. nicagus (Myers & Cadle, 1994). In the T. brevirostris sister clade of T. nicagus (Fig. 4), this simple hemipenis has a bifurcated sulcus spermaticus and a bifurcated RPM muscle. The Colubridae is a diverse clade diagnosed by having a simple sulcus, which evolved by the reduction and loss of its left branch (McDowell, 1961; Zaher et al., 2009; Zaher et al., 2012). Most colubrids have a simple hemipenis but in taxa with a bilobed organ (e.g. Panterophis, Masticophis) the sulcus remains a simple structure with its terminus on the right side of the hemipenial apical region. Unfortunately, there is no available information on the internal hemipenial anatomy of these taxa. Our reconstruction analyses support an evolutionary model for dipsadid snakes wherein the evolution of the hemipenial body may have occurred independently of the evolution of the sulcus spermaticus, while the RPM and the sulcus have a linked phylogenetic history. Our connectivity hypothesis is supported by embryological evidence from Squamata (Clark, 1945; Raynaud & Pieau, 1985; Gredler et al., 2014; Leal & Cohn, 2015). The gathered available data on squamates’ hemipenial development (Table 1) suggest the evolution of common developmental patterns. The timing, embryonic position of precursor cells and morphogenetic pattern of the internal and external hemipenial parts are all congruent with our homology hypotheses based on adult morphology. Table 1. Comparative embryonic development in Squamata of the sulcus spermaticus and the muscle retractor penis magnus (RPM) Family  Species  Reference  Sulcus spermaticus origin  Sulcus spermaticus anlage position  Sulcus spermaticus development timing  Sulcus spermaticus development pattern  RPM origin  RPM anlage position  RPM development pattern  Polychrotidae  Anolis carolinensis  Gredler, Sanger & Cohn (2015)  Ectodermal invagination  ?  After RPM  ?  ?  ?  ?  Anguidae  Anguis fragilis  Raynaud & Pieau (1985)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  ?  Lacertidae  Lacerta viridis  Raynaud & Pieau (1985)  Ectodermal invagination  ?  ?  ?  ?  ?  ?  Pythonidae  Python regius  Leal & Cohn (2015)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  Hemipenis base  Proximodistally  Natricidae  Natrix tessellata  Raynaud & Pieau (1985)  Ectodermal invagination  ?  After RPM  ?  ?  Hemipenis base  Proximodistally  Natricidae  Thamnophis sirtalis  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Colubridae  Lampropeltis triangulum  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Dipsadidae  Diadophis punctatus  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Family  Species  Reference  Sulcus spermaticus origin  Sulcus spermaticus anlage position  Sulcus spermaticus development timing  Sulcus spermaticus development pattern  RPM origin  RPM anlage position  RPM development pattern  Polychrotidae  Anolis carolinensis  Gredler, Sanger & Cohn (2015)  Ectodermal invagination  ?  After RPM  ?  ?  ?  ?  Anguidae  Anguis fragilis  Raynaud & Pieau (1985)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  ?  Lacertidae  Lacerta viridis  Raynaud & Pieau (1985)  Ectodermal invagination  ?  ?  ?  ?  ?  ?  Pythonidae  Python regius  Leal & Cohn (2015)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  Hemipenis base  Proximodistally  Natricidae  Natrix tessellata  Raynaud & Pieau (1985)  Ectodermal invagination  ?  After RPM  ?  ?  Hemipenis base  Proximodistally  Natricidae  Thamnophis sirtalis  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Colubridae  Lampropeltis triangulum  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Dipsadidae  Diadophis punctatus  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  View Large Table 1. Comparative embryonic development in Squamata of the sulcus spermaticus and the muscle retractor penis magnus (RPM) Family  Species  Reference  Sulcus spermaticus origin  Sulcus spermaticus anlage position  Sulcus spermaticus development timing  Sulcus spermaticus development pattern  RPM origin  RPM anlage position  RPM development pattern  Polychrotidae  Anolis carolinensis  Gredler, Sanger & Cohn (2015)  Ectodermal invagination  ?  After RPM  ?  ?  ?  ?  Anguidae  Anguis fragilis  Raynaud & Pieau (1985)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  ?  Lacertidae  Lacerta viridis  Raynaud & Pieau (1985)  Ectodermal invagination  ?  ?  ?  ?  ?  ?  Pythonidae  Python regius  Leal & Cohn (2015)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  Hemipenis base  Proximodistally  Natricidae  Natrix tessellata  Raynaud & Pieau (1985)  Ectodermal invagination  ?  After RPM  ?  ?  Hemipenis base  Proximodistally  Natricidae  Thamnophis sirtalis  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Colubridae  Lampropeltis triangulum  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Dipsadidae  Diadophis punctatus  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Family  Species  Reference  Sulcus spermaticus origin  Sulcus spermaticus anlage position  Sulcus spermaticus development timing  Sulcus spermaticus development pattern  RPM origin  RPM anlage position  RPM development pattern  Polychrotidae  Anolis carolinensis  Gredler, Sanger & Cohn (2015)  Ectodermal invagination  ?  After RPM  ?  ?  ?  ?  Anguidae  Anguis fragilis  Raynaud & Pieau (1985)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  ?  Lacertidae  Lacerta viridis  Raynaud & Pieau (1985)  Ectodermal invagination  ?  ?  ?  ?  ?  ?  Pythonidae  Python regius  Leal & Cohn (2015)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  Hemipenis base  Proximodistally  Natricidae  Natrix tessellata  Raynaud & Pieau (1985)  Ectodermal invagination  ?  After RPM  ?  ?  Hemipenis base  Proximodistally  Natricidae  Thamnophis sirtalis  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Colubridae  Lampropeltis triangulum  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Dipsadidae  Diadophis punctatus  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  View Large In the initial stages of development, the hemipenis is a solid protuberance of mesenchymal cells covered by ectodermal cells (Table 1). The development of the RPM starts by condensation and differentiation of mesenchymal cells in the middle region of the hemipenis base. Then, the RPM grows in the proximal and apical directions. The sulcus spermaticus developed after the development of the RPM as an invagination of the ectodermal cells (Clark, 1945; Raynaud & Pieau, 1985; Gredler et al., 2014; Leal & Cohn, 2015). Usually the sulcus anlage appears at the hemipenis base and grows proximally and apically. In Diadophis punctatus and Python regius, the distal subdivision of the RPM occurs before the bifurcation of the sulcus (Clark, 1945; Leal & Cohn, 2015). The inclusion of this embryological information in our character reconstruction analysis adds explanatory power to the primary homology hypothesis of the sulcus spermaticus condition. When we used the embryonic developmental sequence to order the character sulcus spermaticus, the ancestral state reconstructed at the Imantodini node was a bifurcated sulcus (Fig. 5A). However, using the character sulcus spermaticus unordered, the retrieved ancestral condition at the Imantodini node was ambiguous with three possible states: simple, bifurcated and bifurcated but ending in expanded area without ornamentation. Revisiting the proposed models for the evolution of sulcus spermaticus in dipsadid snakes (Myers, 2011), taking into account the comparative hemipenial embryological data, internal anatomy and phylogenetic reconstructions, leads us to conclude that the sulcus spermaticus in Imantodini evolved together with the RPM muscle, independent of evolutionary changes in other structures (e.g. hemipenial body). With this new perspective on hemipenial comparisons, we would be able to morphologically test the monophyletic hypotheses of the tribe. It will be necessary to include groups positioned closer to the tribe node in the analysis. These groups were recovered in the PhD thesis by Costa (2014), namely: Imantodes inornatus, I. gemmistratus, Leptodeira nigrofasciata, L. punctatus, L. uribei, L. frenata. The inclusion of these species in a future analysis will provide a more robust conclusion on the evolution of RPM and the sulcus. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s website: Table S1. Species used for the molecular analysis, used gene fragments and their GenBank accession number. Appendix S1. Molecular character matrix used in the phylogenetic analysis. Appendix S2. Ancestral reconstruction matrix. Appendix S3. Dipsadidae tree newick format. Figure S1. Mirrored maximum likelihood character reconstructions in Dipsadidae. Circles on nodes denote the reconstructed states for the characters. A, sulcus spermaticus condition. B, muscle retractor penis magnus in cavitas hemipenis. C, sulcus spermaticus length. D, insertion of the muscle retractor penis magnus. ACKNOWLEDGEMENTS J.C.L.C. thanks the Biomatters development team for the license Geneious 9.1.6. copyright license. R.A.G.-F. and J.C.L.C. are supported by fellowships from Programa de Capacitação Institucional (PCI/MPEG/MCTIC), grant numbers 300065/2016-7 and 312847/2015-7; A.F.R.M. is supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (PROTAX grant number 440413/2015-0). A.L.C.P. is supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (Pq. 305475/2014-2; PROTAX 440413/2015-0) and Fundação de Amparo a Estudos e Pesquisa (2016-111.449). Glen Sheppard for the English revision. The authors thank the anonymous reviewers and the Associate Editor for helping to improve the manuscript. REFERENCES Alberch P. 1985. Problems with the interpretation of developmental sequences. Systematic Zoology  34: 46– 58. Google Scholar CrossRef Search ADS   Arnold EN. 1984. Variation in the cloacal and hemipenial muscles of lizards and its bearing on their relationships. Symposia of the Zoological Society of London  52: 47– 85. Arnold EN. 1986. Why copulatory organs provide so many useful taxonomic characters: the origin and maintenance of hemipenial differences in lacertid lizards (Reptilia: Lacertidae). Biological Journal of the Linnean Society  29: 263– 281. Google Scholar CrossRef Search ADS   Bellairs A, Kamal AM. 1981. The chondrocranium and the development of the skull in recent reptiles. In: Gans C, Parsons TS, eds. Biology of the Reptilia . London & New York: Academic Press, 1– 263. Beuchelt H. 1936. Bau, Funktion und Entwicklung der Begattungsorgane der ma ̈nnlichen Ringelnatter (Natrix natrix L.) und Kreuzotter (Vipera berus). Morphologisches Jahrbuch  78: 445– 516. Bininda-Emonds ORP, Jeffery JE, Coates MI, Richardson MK. 2002. From Haeckel to event-pairing: the evolution of developmental sequences. Theory in Biosciences  121: 297– 320. Google Scholar CrossRef Search ADS   Bogert CM. 1940. Herpetological results of the Vernay Angola expedition. Bulletin of the American Museum of Natural History  77: 1– 107. Cadle JE. 1984a. Molecular systematics of Neotropical xenodontine snakes: I. South American xenodontines. Herpetologica  40: 8– 20. Cadle JE. 1984b. Molecular systematics of Neotropical xenodontine snakes: II. Central American xenodontines. Herpetologica  40: 21– 30. Clark H. 1945. The anatomy and embryology of the hemipenis of Lampropeltis, Diadophis and Thamnophis and their value as criteria of relationship in the family Colubridae. Proceedings of the Iowa Academy of Sciences  51: 411– 445. Cope ED. 1886. An analytical table of the genera of snakes. Proceedings of the American Philosophical Society  23: 479– 499. Cope ED. 1900. The crocodilians, lizards and snakes of North America. Report of the U.S. National Museum  1898: 153– 1270. Costa JCL. 2014. Análise filogenética de Leptodeira Fitzinger, 1843 e taxonomia das espécies do clado do sul do complexo Leptodeira annulata/septentrionalis (Serpentes, Dipsadidae) . Unpublished D. Phil. Thesis , Universidade Federal do Pará. Cundall D, Irish F. 2008. The snake skull. In: Carl G, Abbot SG, Adler K, eds. Biology of the Reptilia, Vol. 20 . Salt Lake City, UT: Society for the Study of Amphibians and Reptiles, 349– 692. Daza JM, Smith EN, Paez VP, Parkinson CL. 2009. Complex evolution in the Neotropics: the origin and diversification of the widespread genus Leptodeira (Serpentes: Colubridae). Molecular Phylogenetics and Evolution  53: 653– 667. Google Scholar CrossRef Search ADS PubMed  De Pinna MCC. 1991. Concepts and tests of homology in the cladistic paradigm. Cladistics  7: 367– 394. Google Scholar CrossRef Search ADS   De Pinna MCC. 1994. Ontogeny, rooting, and polarity. In: Scotland RW, Siebert DJ, Williams DM, eds. Models in phylogeny reconstruction, Vol. 52 . Oxford: Clarendon Press, 157– 721. Dowling HG. 1975. The Neartic snake fauna. In: Dowling HG, ed. 1974 Yearbook of Herpetology . New York: Herpetological Information Search System Publications, 191– 202. Dowling HG, Jenner JV. 1987. Taxonomy of American xenodontine snakes. II. The status and relationships of Pseudoleptodeira. Herpetologica  43: 190– 200. Dowling HG, Savage JM. 1960. A guide to the snake hemipenis: a survey of basic structure and systematic characteristics. Zoologica  45: 17– 27. Duellman WE. 1958a. A monographic study of the colubrid snake genus Leptodeira. Bulletin of the American Museum of Natural History  114: 1– 183. Duellman WE. 1958b. Systematic status of the colubrid snake, Leptodeira discolor Günther. University Kansas Publications of Museum of Natural History  11: 1– 9. Dunn ER. 1928. A tentative key and arrangement of the American genera of Colubridae. Bulletin of the Antivenin Institute of America  2: 18– 24. Figueroa A, McKelvy AD, Grismer LL, Bell CD, Lailvaux SP. 2016. A species-level phylogeny of extant snakes with description of a new colubrid subfamily and genus. PLoS One  11.9: e0161070. Google Scholar CrossRef Search ADS   Grazziotin FG, Zaher H, Murphy RW, Scrocchi G, Benavides MA, Zhang YP, Bonatto SL. 2012. Molecular phylogeny of the New World Dipsadidae (Serpentes: Colubroidea): a reappraisal. Cladistics  1: 1– 23. Gredler ML, Larkins CE, Leal F, Lewis AK, Herrera AM, Perriton CL, Sanger TJ, Cohn MJ. 2014. Evolution of external genitalia: insights from reptilian development. Sexual Development  8: 311– 326. Google Scholar CrossRef Search ADS PubMed  Gredler ML, Sanger TJ, Cohn MJ. 2015. Development of the cloaca, hemipenes, and hemiclitores in the green anole, Anolis carolinensis. Sexual Development  9: 21– 33. Google Scholar CrossRef Search ADS PubMed  Hillis DM, Bull JJ. 1993. An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis. Systematic Biology  42: 182– 192. Google Scholar CrossRef Search ADS   Jenner JV, Dowling HG. 1985. Taxonomy of American xenodontine snakes: the tribe Pseudoboini. Herpetologica  41: 161– 172. Katoh K, Misawa K, Kuma K, Miyata T. 2002. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Research  30: 3059– 3066. Google Scholar CrossRef Search ADS PubMed  Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T, Ashton B, Mentjies P, Drummond A. 2012. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics  28: 1647– 1649. Google Scholar CrossRef Search ADS PubMed  Keogh JS. 1996. Origin of the hemipenis retractor penis magnus muscles: evidence for utility in snake systematics. The Snake  27: 114– 119. Kraus F, Brown WM. 1998. Phylogenetic relationships of colubroid snakes based on mitochondrial DNA sequences. Zoological Journal of the Linnean Society  122: 455– 487. Google Scholar CrossRef Search ADS   Lanfear R, Calcott B, Ho SYW, Guindon S. 2012. PartitionFinder: combined selection of partitioning schemes and substitution models for phylogenetic analyses. Molecular Biology and Evolution  29: 1695– 1701. Google Scholar CrossRef Search ADS PubMed  Leal F, Cohn MJ. 2015. Development of hemipenes in the ball python snake Python regius. Sexual Development  9: 6– 20. Google Scholar CrossRef Search ADS PubMed  Mabee PM. 2000. The usefulness of ontogeny in interpreting morphological characters. In: Wiens JJ, ed. Phylogenetic analysis of morphological data . Washington, DC: Smithsonian Institution Press, 84– 114. Maddison WP, Maddison DR. 2011. Mesquite: a modular system for evolutionary analysis . Available at: http://mesquiteproject.org/mesquite/mesquite.html McCranie JR. 2011. The Snakes of Honduras – Systematics, Distribution, and Conservation. Michigan: Society for the Study of Amphibians and Reptiles. McDowell SB. 1961. Systematic division and evolution of the colubrid snake genus Natrix, with comments on the subfamily Natricinae, by Edmond V. Malnate [review]. Copeia  1961: 502– 506. Google Scholar CrossRef Search ADS   Miller MA, Pfeiffer W, Schwartz T. 2010. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. Proceedings of the Gateway Computing Environments Workshop (GCE), New Orleans, LA, 1– 8. Missassi AFR, Prudente ALC. 2015. A new species of Imantodes Duméril, 1853 (Serpentes, Dipsadidae) from the Eastern Cordillera of Colombia. Zootaxa  3980: 562– 574. Google Scholar CrossRef Search ADS PubMed  Mulcahy DG. 2007. Molecular systematics of Neotropical cat-eyed snakes: a test of the monophyly of Leptodeirini (Colubridae: Dipsadinae) with implications for character evolution and biogeography. Biological Journal of the Linnean Society  92: 483– 500. Google Scholar CrossRef Search ADS   Myers CW. 1974. The systematics of Rhadinaea (Colubridae), a genus of New World snakes. Bulletin of the American Museum of Natural History  153: 1– 262. Myers CW. 1982. Blunt-headed vine snakes (Imantodes) in Panama, including a new species and other revisionary notes. American Museum Novitates  2738: 1– 50. Myers CW. 2011. A new genus and new tribe for Enicognathus melanauchen Jan, 1863, a neglected South American snake (Colubridae: Xenodontinae), with taxonomic notes on some Dipsadinae. American Museum Novitates  3715: 1– 33. Google Scholar CrossRef Search ADS   Myers CW, Cadle JE. 2003. On the snake hemipenis, with notes on Psomophis and techniques of eversion: a response to Dowling. Herpetological Review  34: 295– 302. Myers CW, Cadle JE. 1994. A new genus for South American snakes related to Rhadinaea obtusa Cope (Colubridae) and resurrection of Taeniophallus Cope for the “Rhadinaea” brevirostris group. American Museum Novitates  3102: 1– 33. Myers CW, McDowell SB. 2014. New taxa and cryptic species of Neotropical snakes (Xenodontinae), with commentary on hemipenes as generic and specific characters. Bulletin of the American Museum of Natural History  385: 1– 112. Google Scholar CrossRef Search ADS   Pagel M. 1999. The maximum likelihood approach to reconstructing ancestral character states of discrete characters on phylogenies. Systematics Biology  48: 612– 622. Google Scholar CrossRef Search ADS   Patterson C. 1982. Morphological characters and homology. In: Joysey KA, Friday AE, eds. Problems of phylogenetic reconstruction . London: Academic Press, 21– 74. Peters JA, Orejas-Miranda B. 1970. Catalogue of the Neotropical Squamata Part 1. Snakes. United States National Museum Bulletin  297: 1– 347. Pinou T, Vicario S, Marschner M, Caccone A. 2004. Relict snakes of North America and their relationships within Caenophidia, using likelihood-based Bayesian methods on mitochondrial sequences. Molecular Phylogenetic and Evolution  32: 563– 574. Google Scholar CrossRef Search ADS   Porto M, de Oliveira MA, Pissinatti L, Rodrigues RL, Rojas-Moscoso JA, Cogo JC, Metze K, Antunes E, Nahoum C, Mónica FZ, de Nucci G. 2013. The evolutionary implications of hemipenial morphology of rattlesnake Crotalus durissus terrificus (Laurent, 1768) (Serpentes: Viperidae: Crotalinae). PLoS One  8: 1– 8. Google Scholar CrossRef Search ADS   Pyron RA, Burbrink FT, Wiens JJ. 2013. A phylogeny and revised classification of Squamata, including 4161 species of lizards and snakes. BMC Evolutionary Biology  13: 93. Google Scholar CrossRef Search ADS PubMed  Raynaud A, Pieau C. 1985. Embryonic development of the genital system. In: Gans C, Billett F, Maderson PFA, eds. Biology of the Reptilia: development B . New York: John Wiley and Sons, 151– 300. Reyes-Velasco J, Mulcahy DG. 2010. Additional taxonomic remarks on the genus Pseudoleptodeira (Serpentes: Colubridae) and the phylogenetic placement of “P. uribei”. Herpetologica  66: 99– 110. Google Scholar CrossRef Search ADS   Savage JM. 1997. On terminology for the description of the hemipenis of squamate reptiles. Journal of Herpetology  7: 23– 25. Savage JM. 2002. The amphibians and reptiles of Costa Rica. Chicago, IL: The University Chicago Press. Schluter D, Price T, Mooers AO, Ludwig D. 1997. Likelihood of ancestor states in adaptive radiation. Evolution  51: 1699– 1711. Google Scholar CrossRef Search ADS PubMed  Slowinski JB, Lawson R. 2002. Snake phylogeny: evidence from nuclear and mitochondrial genes. Molecular Phylogenetics and Evolution  24: 194– 202. Google Scholar CrossRef Search ADS PubMed  Springer VG, Johnson GD. 2000. Use and advantages of ethanol solution of alizarin red S dye for staining bone in fishes. Copeia  2000: 300. Google Scholar CrossRef Search ADS   Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics  30: 1312– 1313. Google Scholar CrossRef Search ADS PubMed  Torres-Carvajal O, Yánez-Muñoz MH, Quirola D, Smith EN, Almendáriz A. 2012. A new species of blunt-headed vine snake (Colubridae, Imantodes) from the Chocó region of Ecuador. ZooKeys  244: 91– 110. Google Scholar CrossRef Search ADS   Underwood G. 1967. A contribution to the classification of snakes . London: Trustees of the British Museum (Natural History). Vaidya G, Lohman DJ, Meier R. 2011. SequenceMatrix: concatenation software for the fast assembly of multi-gene datasets with character set and codon information. Cladistics  27: 171– 180. Google Scholar CrossRef Search ADS   Vellard J. 1928. Importance des caractères fournis par l’hemipénis pour la classification des ophidiens. Bulletin de la Societé Zoologique de France  53: 406– 418. Vellard J. 1946. Morphologia del hemipenis y evolución de los ofidios. Acta Zoologica Lilloana  3: 263– 288. Vidal N, Delmas AS, David P, Cruaud C, Couloux A, Hedges SB. 2007. The phylogeny and classification of caenophidian snakes inferred from seven nuclear protein-coding genes. Comptes Rendus Biologies  330: 182– 187. Google Scholar CrossRef Search ADS PubMed  Werneburg I. 2009. A standard system to study vertebrate embryos. PLoS One  6: e5887. Google Scholar CrossRef Search ADS   Zaher H. 1999. Hemipenial morphology of the South American xenodontine snakes, with a proposal for a monophyletic Xenodontinae and a reappraisal of colubroid hemipenes. Bulletin of the American Museum of Natural History  240: 1– 168. Zaher H, Prudente ALC. 2003. Hemipenes of Siphlophis (Serpentes, Xenodontinae) and techniques of hemipenial preparation in snakes: a response to Dowling. Herpetological Review  34: 302– 307. Zaher H, Grazziotin FG, Cadle JE, Murphy RW, de Moura-Leite JC, Bonatto SL. 2009. Molecular phylogeny of advanced snakes (Serpentes, Caenophidia) with an emphasis on South American xenodontines: a revised classification and descriptions of new taxa. Papéis Avulsos de Zoologia  49: 115– 153. Google Scholar CrossRef Search ADS   Zaher H, Grazziotin FG, Graboski R, Fuentes RG, Sánchez-Martinez P, Montingelli GG, Zhang YP, Murphy RW. 2012. Phylogenetic relationships of the genus Sibynophis (Serpentes: Colubroidea). Papéis Avulsos de Zoologia  52: 141– 150. Ziegler T, Böhme W. 1997. Genitalstrukturen und Paarungsbiologie bei squamaten Reptilien, speziell den Platynota, mit Bemerkungen zur Systematik. Mertensiella  8: 1– 207. APPENDIX Hemipenial anatomy examined. Aniliidae: Anilius scytale MPEG 25828, MPEG 62373. Boidae: Corallus hortulanus MPEG 18920. Viperidae: Bothrops atrox MPEG 18985. Elapidae: Micrurus lemniscatus MPEG 26392. Colubridae: Chironius exoletus MPEG 24296; Leptophis ahaetulla MPEG 21275; Tantilla melanocephala MPEG 24568. Dipsadidae: Apostolepis quinquelineata MPEG 1700; Atractus albuquerquei MBS 191; Atractus latifrons MPEG 17837; Dipsas catesbyi MPEG 23291; Dipsas indica MPEG 24209; Erythrolamprus reginae MPEG 25458; Helicops angulatus MPEG 4402, MPEG 22297; Hydrops triangularis MPEG 2955; Imantodes cenchoa MPEG 26270, MPEG 26275, MPEG 26269; Imantodes lentiferus MPEG 23572; Leptodeira annulata annulata MPEG 2421; Leptodeira annulata ashmeadi MZUSP 10654; Leptodeira annulata pulchriceps MZUSP 9558; Leptodeira annulata rhombifera MZUSP 17435; Oxyrhopus petolarius MPEG 12266; Philodryas olfersii MPEG 24499; Pseudoboa neuwiedii MPEG 20693; Psomophis joberti MPEG 24868; Sibon nebulatus MPEG 1785; Sibynomorphus mikanii MPEG 21694; Taeniophallus brevirostris MPEG 5033; Taeniophallus nicagus MPEG 23312; Taeniophallus quadriocelatus MPEG 22098; Xenopholis scalaris MPEG 23163. © 2017 The Linnean Society of London, Zoological Journal of the Linnean Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Zoological Journal of the Linnean Society Oxford University Press

Muscular evolution of hemipenis in Imantodini snakes (Squamata: Dipsadidae)

Loading next page...
 
/lp/ou_press/muscular-evolution-of-hemipenis-in-imantodini-snakes-squamata-Wu4Rsk6oo6
Publisher
Oxford University Press
Copyright
© 2017 The Linnean Society of London, Zoological Journal of the Linnean Society
ISSN
0024-4082
eISSN
1096-3642
D.O.I.
10.1093/zoolinnean/zlx091
Publisher site
See Article on Publisher Site

Abstract

Abstract The anatomy of squamates’ hemipenis has historically been an important source of characters for phylogenetic and taxonomic studies. Herein, we use new comparative evidence from the retractor penis magnus (RPM) muscle to revisit Myers models of hemipenial evolution and current homology hypotheses for the sulcus spermaticus of Imantodini (Imantodes, Leptodeira). We compare the topological and connectivity relationships of RPM muscle to other hemipenial structures in 35 species of Imantodini, all tribes of Dipsadidae, and other alethinophidian lineages, analysing everted, inflated, dissected and stained hemipenes. To trace the evolution of hemipenial morphology in dipsadids, we performed a maximum likelihood analysis of a molecular data set with 45 species from the aforementioned clades. These analyses supported the hypothesis that the morphology of the RPM and the sulcus evolved at the same nodes in dipsadid phylogeny. The morphology of this muscle in Leptodeira is similar to that in other taxa, with a bifurcated sulcus spermaticus, supporting the hypothesis that their condition evolved from an ancestral state of a bifurcated sulcus. This unexpected anatomical relationship is congruent with indirect embryonic data published elsewhere. We conclude that the sulcus in Imantodini evolved together with the RPM, independent of evolutionary changes in other structures. comparative anatomy, development, Macrostomata, Serpentes, Squamata INTRODUCTION The male reproductive organ of Squamata, hemipenis, is a hollow structure with an external sulcus spermaticus where the seminal fluids flow, and an internal space housing spongy tissues and musculature (Dowling & Savage, 1960). Hemipenial characters are useful in squamate systematics for differentiating at the species level and higher. Their importance arises from a broad consensus recognizing the similarities between hemipenial parts and a consequent congruence across hypotheses of homology (Patterson, 1982; De Pinna, 1991) and nomenclature of hemipenial structures (Dowling & Savage, 1960; Myers, 1974; Zaher, 1999; Myers & Mcdowell, 2014). The hemipenis was first used as evidence to support squamate evolutionary relationships by Cope (1886, 1900) in his work on reptile classification. From then, multiple authors used hemipenial variation in their own snake classification hypotheses (Dunn, 1928; Vellard, 1928, 1946; Bogert, 1940). The hemipenis became an intrinsic tool for snake phylogeny only after Dowling & Savage (1960) proposed procedures for preparing (evertion and inflation), describing and comparing hemipenial characters. These methods became standard among snake biologists. The external features of hemipenes are surprisingly variable at all taxonomic levels of snakes (e.g. Zaher, 1999; Zaher & Prudente, 2003). Today, descriptions of snake hemipenes and associated discussions of systematics are mostly based on variations in shape (e.g. single or divided), sulcus spermaticus orientation (e.g. centrifugal, centrolineal position) and morphology (e.g. simple or bifurcated), ornamentation (e.g. calyces, flounces, papillae, spines) and apical differentiations (e.g. apical disk, capitulum, capitular groove) (Dowling & Savage, 1960; McDowell, 1961; Savage, 1997; Zaher, 1999; Myers & McDowell, 2014). Despite the importance of the hemipenis in snake classification, few detailed descriptions and comparisons have been carried out on its embryology (Clark, 1945; Raynaud & Pieau, 1985; Gredler et al., 2014; Leal & Cohn, 2015), muscles and internal tissues (Beuchelt, 1936; Dowling & Savage, 1960; Ziegler & Böhme, 1997; Porto et al., 2013). The studies of hemipenial embryology include eight species representing three lizard families (Anguidae, Lacertidae, Polychrotidae) and four macrostomatan families (Dipsadidae, Lamprophidae, Natricidae, Pythonidae) (Supporting Information, Table S1). These analyses describe and compare the development of the sulcus spermaticus and hemipenial body and muscles. The retractor penis magnus (RPM) muscle and the sulcus spermaticus developed in the same position at the base of the hemipenis and share the growing pattern proximodistally. The RPM developed before the sulcus, and when the muscle bifurcates the sulcus bifurcates soon after (Clark, 1945; Raynaud & Pieau, 1985; Gredler et al., 2014; Leal & Cohn, 2015). These data have been used to review the hemipenial anatomy and to discuss its evolution in the modern phylogenetic framework for snakes. Elsewhere, data comparing embryonic skulls have been used to define and discuss the evolution and relationships of major lineages of snakes (Bellairs & Kamal, 1981; Cundall & Irish, 2008). Ontogenetic and embryonic information in phylogenetic analyses have been proposed for two stages of the analyses: homology hypothesis proposition (including character codification and ordering) and recovering the parsimonious tree as a rooting criteria (De Pinna, 1994; Mabee, 2000). However, it is difficult to including embryonic information in ancestral reconstructions, since it is difficult to adjust the resulting information so it can be included in phylogenetic methods (Alberch, 1985; Bininda-Emonds et al., 2002; Werneburg, 2009). The hemipenial musculature is defined by the movements achieved through contraction in the propulsor and the retractor muscles. Three groups of retractor muscles have been described in snakes: retractor lateralis anterior, retractor lateralis posterior and RPM (Dowling & Savage, 1960; Arnold, 1984; Keogh, 1996). The retractor lateralis anterior and posterior originate on the horizontal septum of the caudal vertebrae transverse process. The retractor lateralis anterior is inserted in front of the vent while the lateralis posterior is inserted at the base of the hemipenis. The RPM originates at the transverse process of the caudal vertebrae and inserts into the inner surface of the free distal hemipenial apex (Porto et al., 2013). The RPM is the longest and most important muscle in hemipenial withdrawal. In species with bilobed or divided organs, this muscle is divided at some distance posterior to its attachment to the hemipenis and each branch is inserted into its apex (Dowling & Savage, 1960). In Dipsadidae, one of the largest American radiations of higher snakes, the clade Dipsadinae has been congruently recovered in comparative molecular and morphological analyses (Cadle, 1984a; Jenner & Dowling, 1985; Myers & Cadle, 1994; Kraus & Brown, 1998; Zaher, 1999; Slowinski & Lawson, 2002; Pinou et al., 2004; Vidal et al., 2007; Zaher et al., 2009; Grazziotin et al., 2012). The morphological diagnosis of Dipsadinae includes the following hemipenial characters: hemipenis unilobed or with reduced bilobation, unicapitate, sulcus spermaticus bifurcated either at hemipenial base or after the capitular groove (Myers, 1974; Cadle, 1984b; Zaher, 1999). Dipsadinae includes the tribes Imantodini and Dipsadini and one unnamed clade that includes species of the genus Atractus (Grazziotin et al., 2012). Imantodini includes two genera of Neotropical snakes (Imantodes and Leptodeira), and the clade is supported solely by molecular characters, while interpretation of their hemipenial morphology (single, unilobed, capitate, sulcus spermaticus simple) is subject to discussion (see Myers & McDowell, 2014). Imantodes comprises a group of eight primarily arboreal species that occur at low to moderate elevations (up to 2200 m above sea level) in rainforest and savanna environments (Peters & Orejas-Miranda, 1970; Myers, 1982; Torres-Carvajal et al., 2012; Missassi & Prudente, 2015). Leptodeira includes 11 species and 12 subspecies ranging from southern USA to northern Argentina and Paraguay, the east coast of Brazil and the islands of Aruba, Margarita, Tobago and Trinidad. Some species include geographical variations and are arranged into three groups of subspecies (L. annulata, L. septentrionalis, L. splendida) (Duellman, 1958a, b; Dowling & Jenner, 1987; Mulcahy, 2007; Daza et al., 2009; McCranie, 2011; Reyes-Velasco & Mulcahy, 2010). In Imantodes species, the hemipenes have a single simple sulcus ending at the hemipenial apex (Myers, 2011; Missassi & Prudente, 2015). However, Missassi and Prudente (2015) described a new hemipenis condition for the genus in Imantodes guane: a sulcus spermaticus with an elongated, expanded terminal area. The condition in I. guane may be interpreted as a transitional evolutionary state before the one present in some Leptodeira groups (Myers, 2011; Missassi & Prudente, 2015). In Leptodeira, the hemipenis has a sulcus spermaticus that terminates in an expanded area without the medial sulcar walls and without any ornamentation (Myers, 2011: 22). The homology hypothesis interpreting these conditions in sulcus spermaticus morphology has long been a controversial issue. For some authors, Leptodeira has a simple sulcus spermaticus (Dunn, 1928: 23; Duellman, 1958a: 17; Cadle, 1984b: 24; Dowling & Jenner, 1987: 198), whereas for others it is bifurcated (Underwood, 1967: fig. 10; Dowling, 1975: 198; Myers & Cadle, 1994: 27; Savage, 2002: 610). The proposed models for loss of bifurcation in the sulcus spermaticus hemipenial evolution in dipsadid snakes (Myers, 2011) stated that the most probable morphological scenario for the condition in Leptodeira might be the shortening and/or loss of a single hemipenial lobe combined with the ‘weakening’ and loss of the medial sulcus walls, ‘leaving only a tissue divide separating the lateral branches in situ; ornamentation such as calyces between the branches may also be lost during this process’. Here, we review the hemipenial variation in species of Imantodini using the available embryonic data and a new data set of hemipenial RPM of dipsadids to reconstruct the ancestral state of Imantodini and discuss the evolution of the muscle and the sulcus spermaticus in Imantodes and Leptodeira. The embryo data was used to propose the homology hypothesis and character codification for the RPM and the sulcus spermaticus. The comparative evidence for the RPM was used to revisit Myers’s models for dipsadids hemipenial evolution and current homology hypotheses for the sulcus spermaticus in Imantodini. Were described the topological and connectivity relationships of the RPM with other hemipenial structures and compared it with those of representative taxa from major dipsadid lineages to discuss the evolution of the sulcus spermaticus in Imantodini. MATERIALS AND METHODS Taxonomic sampling and DNA data set Our sampling was based on the taxonomic arrangements of Zaher et al. (2009) and Grazziotin et al. (2012) and included species representing the following clades of alethinophidian snakes: Aniliidae, Boidae, Colubridae, Dipsadidae (Dipsadini, Elapomorphini, Echinantherini, Imantodini, Philodryadini, Pseudoboini, Psomophiini, Tachimenini, Xenodontini), Elapidae and Viperidae. We selected 35 species to represent these clades in the comparative analyses of hemipenial anatomy (Appendix). In addition, we created a DNA character matrix using the available data from GenBank for 45 species of the aforementioned alethinophidian clades. We gathered DNA data from gene fragments of four nuclear (c-mos, bdnf, ntf3, rag1) and four mitochondrial genes (cytb, nd4, 12s, 16s) (Supporting Information, Table S1). Hemipenial nomenclature We used the following hemipenial nomenclature proposed by Arnold (1986), Dowling & Savage (1960) and Zaher (1999): asulcate face – on the hemipenis external layer, the side without the sulcus spermaticus; absulcate layer – the functionally internal surface of the everted hemipenis where the RPM attaches; external layer – hemipenial surface composed of epidermis and dermis; inner layer – internal tissues of hemipenis including the blood and lymphatic sinuses and muscle; base – proximal region to the cloaca and usually constricted compared to the hemipenial body; apical end – free tip of the hemipenis; lobe – division of the hemipenial body; capitulum – apical portion separated from the hemipenial body by a deep groove or capitular groove; sulcus spermaticus – longitudinal groove on the external layer of the everted hemipenis; RPM – muscle with origin on the caudal vertebrae and with insertion on the apical end of the hemipenis; RPM insertion – inside the absulcate layer; RPM origin – on the transverse process of the caudal vertebrae. We also propose a new term, cavitas hemipenis, to describe the internal central space created and delimited by the absulcate layer, containing the distal portion of the RPM when the hemipenis is everted. Eversion, dissection and staining We used two complementary hemipenial preparation techniques: fully everted and everted-inflated. As stated by Myers & Cadle (2003), there is a difference between hemipenis preparations when ‘fully everted’, as opposed to ‘maximally expanded or inflated’. In a maximally expanded preparation, the RPM muscle is sectioned at its insertion and the hemipenis is everted and inflated to the fullest extent allowed by its original elasticity, which can vary when working with preserved specimens (Myers & Cadle, 2003; Zaher & Prudente, 2003). The hemipenis was everted and inflated following the techniques proposed by Zaher & Prudente (2003). We used the fully everted hemipenis of preserved museum specimens to analyse the topology and connectivity relationships between hemipenial internal and external parts (RPM, sulcus spermaticus, capitular groove, apical end). Analysing the hemipenis in this condition provided a view of the muscle groups, ornamentation structures and the sulcus spermaticus, including its terminal region (Myers & Cadle, 2003). To access RPM morphology in the everted hemipenis, we dissected it in the following manner: (1) open the ventral surface of the tail at the midline, posterior to the cloaca; (2) isolate the hemipenis and the RPM from surrounding tissues and cut the muscle close to its origin; (3) remove the organ from the specimen; and (4) dissect the asulcate surface at its medial side, following the longitudinal axis from the hemipenial base to its apical end. Afterwards, we stained each hemipenis with an alizarin-ethanol solution (Springer & Johnson, 2000). This procedure creates contrast between tissues and highlights RPM fibres, facilitating the recognition and description of the structures. The absulcate surface is surrounded by tissue walls, which prevent the diffusion of the staining solution over muscular tissue. We dissected the tissue of the lymph and blood sinuses to expose the muscles to the staining solution. Phylogenetic analysis We generated a phylogenetic hypothesis based on our sampling of the morphological characters to reconstruct the ancestral state at the node Imantodini, instead of using the large sampled matrices available (e.g. Pyron, Burbrink & Wiens, 2013; Figueroa et al., 2016). Using these matrices, or their recovered trees, as the basis for reconstructing the ancestral states would have generated a large amount of missing data at our morphological matrix and undesirable noise in the reconstruction at the Imantodini node. We generated a maximum likelihood (ML) phylogenetic hypothesis of dipsadid snakes based on the molecular data available from GenBank that included, whenever possible, species for which we have hemipenial morphology (Supporting Information). As the data available on GenBank were not complete for the selected species and genera, we included sequences from congeneric species that shared at least one gene with our ‘morphological terminal’. Thus, hemipenial morphology was available for 23 out of the 45 taxa used in our phylogenetic analysis. The only taxon without an available sequence was Apostolepis quinquelineata. Therefore, for the purposes of the analysis, we considered the sequence for Apostolepis dimidiata as equivalent to A. quinquelineata, placing them at the same terminal. The proportion of taxon represented within each gene matrix varied: 82% of the 45 taxon had c-mos, 49% had bdnf and ntf3, 33% had rag1, 87% had cytb, 51% had nd4, 87% had 12s and 89% had 16s. We downloaded and visualized all sequence fragments with Geneious v 9.1.6 (Kearse et al., 2012). We aligned the sequences with MAFFT v 7.222 (Katoh et al., 2002), using the FFT-NS-I algorithm with default parameters, by c-mos, bdnf, ntf3, cytb, and E-INS-i by rag1, nd4, 12s, 16s. The sequences were concatenated with SequenceMatrix v 1.7.8 (Vaidya, Lohman & Meier, 2011). The best scheme for partitions was selected using PartitionFinder v1.1.1 (Lanfear et al., 2012). The decision criterion was set to BIC (Bayesian Information Criterion), the search algorithm set to ‘greedy’ and branch lengths set to ‘linked’. ML analyses were conducted with the CIPRES Science Gateway (Miller, Pfeiffer & Schwartz, 2010) using RAxML 8.0 (Stamatakis, 2014). We recovered the best ML tree by running 100 independent searches and verified similarity in the likelihood values, thus avoiding the use of suboptimal trees. Statistical support for clades was obtained from 1000 non-parametrical pseudoreplications. Clades with bootstrap values higher than 70% were considered well supported (Hillis & Bull, 1993). Ancestral state reconstructions We used our phylogenetic framework to reconstruct the evolutionary history of the external and internal hemipenial morphology using parsimony (PA) and ML methods. PA reconstructions were performed with states ordered and unordered and with all transformations equally weighted, using Mesquite v2.75 (Maddison & Maddison, 2011). ML reconstructions were also performed in Mesquite v2.75, using the Markov k-state 1 parameter (Mk1) model of evolution (Schluter et al., 1997; Pagel, 1999), which gives equal probabilities for changes between any two character states. Based on the available embryonic data (Supporting Information, Table S1) and our comparative anatomy results, we coded two characters for the sulcus spermaticus and three for the RPM muscle. For some taxa with missing data, we included the descriptions by Zaher (1999). These are our hypotheses of homology and their character states: Sulcus spermaticus condition: 0 – simple, 1 – bifurcated and 2 – bifurcated but ending in expanded area without ornamentation (ordered linear additive – during the embryology the sulcus is a simple structure that bifurcates subsequently in the taxa with two branches). Sulcus spermaticus length on the everted and inflated hemipenis: 0 – extended from hemipenial base to the apical end, and 1 – extended from hemipenial base to the medial portion of hemipenial body and never reaching apical end. RPM muscle in cavitas hemipenis, morphology: 0 – single and 1 – bifurcated. RPM in absulcate layer, insertion position: 0 – at the distal end of hemipenis and 1 – before the apical end of hemipenis. RPM in cavitas hemipenis, muscle fibres bifurcation position to the position of the sulcus spermaticus bifurcation: 0 – before bifurcation of the sulcus, 1 – underneath the bifurcation of the sulcus and 2 – after the bifurcation of the sulcus. RESULTS Hemipenial morphology in Imantodini In an inflated state, L. annulata subspecies had a single, cylindrical, slightly curved, capitate and calyculate hemipenis (Fig. 1). Capitulum was longest on the asulcate face, with calyces and small spines mineralized at the corners of calyces walls (Fig. 1A). Calyces and spines altered in shape from the capitular groove to the apical end. The calyces increased in length while the spines reduced in height. Capitular groove deep and well defined on sulcate face (Fig. 1A) and interrupted at the asulcate face by perpendicular folds continuous with two paired rows of spines (Fig. 1B). The capitular groove between the two folds was deeper. The sulcus spermaticus single, with spinules from the base of organ to the capitular groove, terminating soon after crossing the capitular sulcus, on the base of capitulum (Fig. 1A). The sulcus spermaticus lacked lips at its terminus, ending in a large area that is without ornamentation (Fig. 1C). The end of the sulcus did not reach the apical end of hemipenis. Hemipenial body had larger, hooked and mineralized spines, arranged in parallel transverse rows that followed the hemipenial body laterally (Fig. 1A); at the sulcate face, the spines reducing in length and height towards the base of the organ and the capitular groove. Asulcate face contained two longitudinal and parallel rows of large, hooked spines, separated by a broad, denuded central region (Fig. 1B). Base of the organ covered by small spines of different sizes that were irregularly arranged, on both faces. Figure 1. View largeDownload slide Everted and inflated hemipenis in adults of Imantodini. A–C, Leptodeira a. annulata (MPEG 20016) sulcate face, asulcate face and naked area in detail. D, E, Imantodes cenchoa (MPEG 17442). Abbreviations: cg, capitular groove; na, naked area; ss, sulcus spermaticus. Scale bar: A, B, D, E = 5.0 mm; C = 2.0 mm. Figure 1. View largeDownload slide Everted and inflated hemipenis in adults of Imantodini. A–C, Leptodeira a. annulata (MPEG 20016) sulcate face, asulcate face and naked area in detail. D, E, Imantodes cenchoa (MPEG 17442). Abbreviations: cg, capitular groove; na, naked area; ss, sulcus spermaticus. Scale bar: A, B, D, E = 5.0 mm; C = 2.0 mm. The everted dissected hemipenis in all L. annulata subspecies (Appendix) showed RPM fibres flexed at a 45° angle at the base of the hemipenis. The single RPM was arranged centrally in the cavitas hemipenis (Fig. 2A). Apically, the RPM was bent and inserted at the medial region of the capitulum underneath the distal end of the naked area of the sulcus spermaticus (Fig. 2B). At its insertion, the RPM formed large, fan-like fascia, bilaterally arranged. The side facing the hemipenial asulcate face (Fig. 2A) had fibres disposed continuously, forming a short longitudinal array. On the side opposite of the sulcate surface, the muscular fibres of the RPM divided in two groups by short fascia (Fig. 2B). In L. annulata subspecies, the RPM was bifurcated at its insertion forming two muscle anchorage areas. Figure 2. View largeDownload slide Everted and dissected hemipenis in adults of Imantodini. A, B, Leptodeira a. ashmeadi (MZUSP 10654). C, D, Imantodes cenchoa (MPEG 26270). Abbreviations: cg, capitular groove; in, insertion of RPM; rpm, retractor penis magnus muscle; rpp, retractor penis parvus muscle; ss, sulcus spermaticus. Scale bar: 5.0 mm. Figure 2. View largeDownload slide Everted and dissected hemipenis in adults of Imantodini. A, B, Leptodeira a. ashmeadi (MZUSP 10654). C, D, Imantodes cenchoa (MPEG 26270). Abbreviations: cg, capitular groove; in, insertion of RPM; rpm, retractor penis magnus muscle; rpp, retractor penis parvus muscle; ss, sulcus spermaticus. Scale bar: 5.0 mm. Imantodes cenchoa (Fig. 1D, E) and I. lentiferus had the hemipenis unilobed, capitate and calyculate; calyces in transverse rows, conspicuous in sulcate, lateral, and asulcate sides of capitulum; asulcate side with small naked pocket located at the base of capitulum; asulcate side with two rows of small spines separating capitulum from the hemipenial body; sulcus spermaticus single, reaching the apical end of the hemipenis, sulcus margins bordered with spinules from the base of hemipenial body to the beginning of capitulum; hemipenial body with four rows of large spines disposed transversally and dispersed spinules concentrated on the basal region of organ. Both species had a single RPM muscle in cavitae hemipenes with its insertion at the apical end and underneath the terminus of the sulcus spermaticus (Fig. 2C, D). Hemipenial morphology in Dipsadidae and other alethinophidians In other groups of species of Dipsadinae (Supporting Information, Appendices S1–S3), including representatives of the genus Atractus and species of the tribe Dipsadini, the hemipenis had two short lobes with a bifurcated sulcus spermaticus. In these species, the sulcus ended at the apical end of the hemipenis. The RPM in the cavitas hemipenis bifurcated underneath the bifurcation of the sulcus spermaticus (Atractus albuquerquei, A. latifrons, Dipsas catesbyi, D. indica, Sibynomorphus mikanii) (see Fig. 3), or after the sulcus bifurcation (Sibon nebulatus). Each muscle ran through the lobe and inserted at the apical end of the hemipenis underneath the end of the sulcus spermaticus. Close to the insertion region, the fibres transformed into short fascia with a fan-like shape, but at the insertion the RPM remained a single muscle. Figure 3. View largeDownload slide Everted and dissected hemipenis in adults of alethinophidian families. A, Viperidae (Bothrops atrox, MPEG 18985). B, Elapidae (Micrurus lemniscatus, MPEG 26392). C, Colubridae (Tantilla melanocephala, MPEG 24568). D, Dipsadidae (Taeniophallus brevirostris, MPEG 5033). Abbreviations: cg, capitular groove; in, insertion of the RPM; rpm, retractor penis magnus muscle; ss, sulcus spermaticus. Scale bar: 5.0 mm. Figure 3. View largeDownload slide Everted and dissected hemipenis in adults of alethinophidian families. A, Viperidae (Bothrops atrox, MPEG 18985). B, Elapidae (Micrurus lemniscatus, MPEG 26392). C, Colubridae (Tantilla melanocephala, MPEG 24568). D, Dipsadidae (Taeniophallus brevirostris, MPEG 5033). Abbreviations: cg, capitular groove; in, insertion of the RPM; rpm, retractor penis magnus muscle; ss, sulcus spermaticus. Scale bar: 5.0 mm. In species of Xenodontinae, the hemipenes were simple (e.g. A. quinquelineata, Taeniophallus brevirostris, T. nicagus) or bilobed (e.g. Erythrolamprus reginae, Oxyrhopus guibei, O. petolarios, Phylodryas olfersii, Psomophis jobarti, Pseudoboa neuwiedii, Xenopholis scalaris) (Fig. 3). But in all taxa the sulcus spermaticus was bifurcated, except in Taeniophallus nicagus, which had a simple sulcus. In all taxa the single RPM was bifurcated distally in two main muscles, except in T. nicagus, where the RPM remained a single muscle. The position of the RPM bifurcation varied. For a subgroup of xenodontines (Philodryas olfersii, P. neuwiedii, Psomophis joberti, X. scalaris), the bifurcation point was underneath the bifurcation of the sulcus spermaticus. In another subset (A. quinquelineata, E. reginae, Oxyrhopus petolarius, T. brevirostris), the division was after the sulcus bifurcation. However, in all species, the RPM inserted by short fascia at the apical region of the hemipenis underneath the end of the sulcus spermaticus. At its insertion the RPM did not divide. The representative species of Aniliidae (Anilius scytale), Boidae (Corallus hortulanus), Elapidae (Micrurus lemniscatus) and Viperidae (Bothrops atrox) had a hemipenis with a bilobed body and bifurcated sulcus spermaticus. The colubrid species, Chironius exoletus, Leptophys ahaetulla and Tantilla melanocephala, had a simple hemipenis with a single sulcus spermaticus. However, in all species of these families, the sulcus spermaticus ended at the apical end of the hemipenis. In A. scytale, the RPM was formed by two separate muscle groups. In the cavitas hemipenis, each muscle group was inserted at the end of each lobe underneath the terminus of the sulcus. In boid, elapid, viperid and colubrid species, the RPM was a single muscle at its origin, but it was divided in two at the cavitas. In the boid species C. hortulanus and the viperid B. atrox (Fig. 3A), the bifurcation of the muscle was underneath the bifurcation point of the sulcus spermaticus. In the elapid M. lemniscatus, the division of the RPM occurred before the bifurcation of the sulcus (Fig. 3B). In colubrid species (C. exoletus, L. ahaetulla, T. melanocephala) (Fig. 3C), the RPM was a single muscle at the cavitas hemipenis and inserted at the apical portion of the hemipenis. There was a short fascia inserted the muscle underneath the end of the single sulcus spermaticus. Phylogenetic relationships of Dipsadidae Our recovered tree is congruent with the hypotheses generated with larger matrices, such as in research by Grazziotin et al. (2012), Pyron et al. (2013) and Figueroa et al. (2016). Our hypothesis for dipsadid interrelationships was based on the ML analysis of 4906 characters over 45 species (Fig. 4). We recovered monophyly for Dipsadidae (bootstrap > 70) and four additional clades: Thermophis species (bootstrap = 100); Heterodon, Farancia and Diadophis (bootstrap = 32); Dipsadinae (bootstrap = 100); and Xenodontinae (bootstrap = 100). Thermophis appeared as a sister clade to all dipsadids (bootstrap < 70), and the clade composed of Heterodon, Farancia. Diadophis appeared as a sister group of Dipsadinae and Xenodontinae (bootstrap > 70). Thus, the phylogenetic position of Diadophis, Farancia, Heterodon and Thermophis is unstable and controversial. Supports were above 95 for the Dipsadinae clade and its internal relationships among Imantodini, Dipsadini and the ‘Unamed Clade’ (grouping Atractus species, see Grazziotin et al., 2012). The phylogenetic relationships among dipsadines revealed Imantodini as a sister group to the clade comprising Dipsadini together with the ‘Unamed Clade’. Figure 4. View largeDownload slide Phylogenetic relationship of the family Dipsadidae based on the maximum likelihood analysis of molecular data and the variation of hemipenial forms and sulcus spermaticus. Bootstrap values next to nodes. Figure 4. View largeDownload slide Phylogenetic relationship of the family Dipsadidae based on the maximum likelihood analysis of molecular data and the variation of hemipenial forms and sulcus spermaticus. Bootstrap values next to nodes. Ancestral state reconstructions in Imantodini In reconstructions of PA and ML, a hemipenis with a bifurcated sulcus was the ancestral condition for Imantodini (Fig. 5A; Supporting Information, Fig. S1A). A simple sulcus arose independently in Imantodes species and T. nicagus (Echiantherini). The L. annulata group evolved an autopomorphic state for all dipsadids with a sulcus ending in an expanded area. PA and ML reconstructed the RPM for the ancestral node of Imantodini in a bifurcated state (Fig. 5B; Supporting Information, Fig. S1B). The RPM evolved into one muscle independently in Imantodes taxa and T. nicagus. Species of the L. annulata group retained the plesiomorphic state of a bifurcated muscle. Figure 5. View largeDownload slide Mirrored parsimony character reconstructions in Dipsadidae. Circles on nodes denote the reconstructed states for the characters. A, sulcus spermaticus condition. B, muscle retractor penis magnus in cavitashemipenis. C, sulcus spermaticus length. D, insertion of the muscle retractor penis magnus. Figure 5. View largeDownload slide Mirrored parsimony character reconstructions in Dipsadidae. Circles on nodes denote the reconstructed states for the characters. A, sulcus spermaticus condition. B, muscle retractor penis magnus in cavitashemipenis. C, sulcus spermaticus length. D, insertion of the muscle retractor penis magnus. As for length of the sulcus spermaticus, in species of Leptodeira, PA and ML reconstructed a sulcus extending from the base of the hemipenis to the medial portion of the hemipenial body (Fig. 5C; Supporting Information, Fig. S1C). The insertion of the RPM showed a similar phylogenetic history, reconstructing as inserted at the apical end at the Dipsadidae node (however, ML reconstruction was unable to calculate its likelihood), with a single transformation in Leptodeira to an insertion before the apical end (Fig. 5D; Supporting Information, Fig. S1D). DISCUSSION Our comparative analysis of the RPM muscle in Dipsadidae supports the hypothesis that the condition of sulcus spermaticus in Imantodini evolved from an ancestral bifurcated state. Imantodes and some Leptodeira species seem to share a simple, unbranched sulcus spermaticus. However, the species of the L. annulata group as well as L. septentrionalis had an expanse without sulcar walls at the terminus of the sulcus (Myers, 2011). In Leptodeira taxa, the RPM is divided distally into two muscles and inserts underneath the expanded naked area after the end of the sulcus spermaticus. In all other snakes, the RPM is inserted below the apical end of the sulcus spermaticus. All taxa with bifurcated sulcus have a divided RPM muscle. Thus, the morphology and position of the insertion of the RPM in Leptodeira are very similar to that observed in other species with bifurcated sulcus spermaticus (Fig. 2A–C). The sulcus spermaticus in Leptodeira terminates close to the capitular groove in an expanded naked area without medial sulcar lips (Myers, 2011: 22). In Imantodes, the morphology of the sulcus and the RPM is different from that in Leptodeira. For Imantodes, a single sulcus extends over the whole hemipenial length with its terminus at the apical end; at its terminus it is not expanded, retaining sulcar lips. Its single RPM muscle is inserted at the apical region, underneath the sulcus terminus. Underwood (1967: 45) was the first to ascertain that the apomorphic condition of the sulcus in Leptodeira evolved by reduction of a plesiomorphic bifurcated sulcus. Myers (2011: 23) refined and extended Underwood’s homology hypothesis, proposing at least four different evolutionary models for the loss of bifurcation in dipsadid snakes: (1) simultaneous shortening of both branches, probably associated with reduction or loss of hemipenial bilobation; (2) shortening of a single branch; (3) shortening and/or loss of a single hemipenial lobe; (4) weakening and loss of the medial sulcus lips, leaving a dividing tissue between lateral branches, with the subsequent or simultaneous loss of ornamentation between the branches. Our results suggest that in the L. annulata group, the expanded naked area is homologous to the terminus of a bifurcated sulcus, with reduced or undeveloped medial sulcar walls or lips. Myers (2011, 2014: 65) suggested that models 1 and 4 apply to the evolution of the sulcus in Leptoderia. He assumed that the ancestral plesiomorphic state was a bifurcated sulcus that evolved through the shortening of both branches in conjunction with the atrophy of tissue forming the medial sulcar lips. Comparative anatomy of the RPM in snakes verified the bifurcation and insertion of the RPM in the same region of bifurcation and terminus of the sulcus spermaticus, providing evidence for the topographic relationship and connectivity between sulcus and this muscle. Our reconstructions of the evolutionary history of the sulcus spermaticus and RPM muscle in Dipsadidae show that both evolved simultaneously in the tree (Fig. 5). In Leptodeira, the insertion of the RPM and the terminus of the sulcus spermaticus shifted to the middle hemipenial region. In addition, the reconstructions show (Fig. 5A, B) that in the species of Imantodes, T. nicagus and the observed Neotropical species of Colubridae, the lack one of the branches of the sulcus coincides with the lack in these taxa of one of the RPM muscles. Imantodes guane is the only species of Imantodes that presents hemipenial morphology with an expanded terminal area, which may be interpreted as homologous with that found in some Leptodeira species (Missassi & Prudente, 2015). In their description of I. guane, Missassi and Prudente found an expansion of the sulcus spermaticus at the apical region beyond the capitular groove, and interpreted this as a new condition for the genus. The terminus of the sulcus of I. guane is at the apex of the apical region, where it preserves the sulcar walls (Prudente & Missassi, 2015: fig. 6). In Leptodeira, medial position and loss of sulcar walls or ‘sulcus lips’ are morphological definitions of their expanded naked area. Since we do not have descriptions of the RPM and its morphology at the cavitas hemipenis of I. guane, we can only surmise that its naked area is not homologous to that in Leptodeira. We observed indirect evidence for the associated evolution of RPM and sulcus spermaticus morphology in T. nicagus and in Neotropical species of Colubridae. Species of the tribe Echinantherini have a hemipenis with a simple body and bifurcated sulcus spermaticus, with the exception of T. nicagus (Myers & Cadle, 1994). In the T. brevirostris sister clade of T. nicagus (Fig. 4), this simple hemipenis has a bifurcated sulcus spermaticus and a bifurcated RPM muscle. The Colubridae is a diverse clade diagnosed by having a simple sulcus, which evolved by the reduction and loss of its left branch (McDowell, 1961; Zaher et al., 2009; Zaher et al., 2012). Most colubrids have a simple hemipenis but in taxa with a bilobed organ (e.g. Panterophis, Masticophis) the sulcus remains a simple structure with its terminus on the right side of the hemipenial apical region. Unfortunately, there is no available information on the internal hemipenial anatomy of these taxa. Our reconstruction analyses support an evolutionary model for dipsadid snakes wherein the evolution of the hemipenial body may have occurred independently of the evolution of the sulcus spermaticus, while the RPM and the sulcus have a linked phylogenetic history. Our connectivity hypothesis is supported by embryological evidence from Squamata (Clark, 1945; Raynaud & Pieau, 1985; Gredler et al., 2014; Leal & Cohn, 2015). The gathered available data on squamates’ hemipenial development (Table 1) suggest the evolution of common developmental patterns. The timing, embryonic position of precursor cells and morphogenetic pattern of the internal and external hemipenial parts are all congruent with our homology hypotheses based on adult morphology. Table 1. Comparative embryonic development in Squamata of the sulcus spermaticus and the muscle retractor penis magnus (RPM) Family  Species  Reference  Sulcus spermaticus origin  Sulcus spermaticus anlage position  Sulcus spermaticus development timing  Sulcus spermaticus development pattern  RPM origin  RPM anlage position  RPM development pattern  Polychrotidae  Anolis carolinensis  Gredler, Sanger & Cohn (2015)  Ectodermal invagination  ?  After RPM  ?  ?  ?  ?  Anguidae  Anguis fragilis  Raynaud & Pieau (1985)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  ?  Lacertidae  Lacerta viridis  Raynaud & Pieau (1985)  Ectodermal invagination  ?  ?  ?  ?  ?  ?  Pythonidae  Python regius  Leal & Cohn (2015)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  Hemipenis base  Proximodistally  Natricidae  Natrix tessellata  Raynaud & Pieau (1985)  Ectodermal invagination  ?  After RPM  ?  ?  Hemipenis base  Proximodistally  Natricidae  Thamnophis sirtalis  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Colubridae  Lampropeltis triangulum  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Dipsadidae  Diadophis punctatus  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Family  Species  Reference  Sulcus spermaticus origin  Sulcus spermaticus anlage position  Sulcus spermaticus development timing  Sulcus spermaticus development pattern  RPM origin  RPM anlage position  RPM development pattern  Polychrotidae  Anolis carolinensis  Gredler, Sanger & Cohn (2015)  Ectodermal invagination  ?  After RPM  ?  ?  ?  ?  Anguidae  Anguis fragilis  Raynaud & Pieau (1985)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  ?  Lacertidae  Lacerta viridis  Raynaud & Pieau (1985)  Ectodermal invagination  ?  ?  ?  ?  ?  ?  Pythonidae  Python regius  Leal & Cohn (2015)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  Hemipenis base  Proximodistally  Natricidae  Natrix tessellata  Raynaud & Pieau (1985)  Ectodermal invagination  ?  After RPM  ?  ?  Hemipenis base  Proximodistally  Natricidae  Thamnophis sirtalis  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Colubridae  Lampropeltis triangulum  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Dipsadidae  Diadophis punctatus  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  View Large Table 1. Comparative embryonic development in Squamata of the sulcus spermaticus and the muscle retractor penis magnus (RPM) Family  Species  Reference  Sulcus spermaticus origin  Sulcus spermaticus anlage position  Sulcus spermaticus development timing  Sulcus spermaticus development pattern  RPM origin  RPM anlage position  RPM development pattern  Polychrotidae  Anolis carolinensis  Gredler, Sanger & Cohn (2015)  Ectodermal invagination  ?  After RPM  ?  ?  ?  ?  Anguidae  Anguis fragilis  Raynaud & Pieau (1985)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  ?  Lacertidae  Lacerta viridis  Raynaud & Pieau (1985)  Ectodermal invagination  ?  ?  ?  ?  ?  ?  Pythonidae  Python regius  Leal & Cohn (2015)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  Hemipenis base  Proximodistally  Natricidae  Natrix tessellata  Raynaud & Pieau (1985)  Ectodermal invagination  ?  After RPM  ?  ?  Hemipenis base  Proximodistally  Natricidae  Thamnophis sirtalis  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Colubridae  Lampropeltis triangulum  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Dipsadidae  Diadophis punctatus  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Family  Species  Reference  Sulcus spermaticus origin  Sulcus spermaticus anlage position  Sulcus spermaticus development timing  Sulcus spermaticus development pattern  RPM origin  RPM anlage position  RPM development pattern  Polychrotidae  Anolis carolinensis  Gredler, Sanger & Cohn (2015)  Ectodermal invagination  ?  After RPM  ?  ?  ?  ?  Anguidae  Anguis fragilis  Raynaud & Pieau (1985)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  ?  Lacertidae  Lacerta viridis  Raynaud & Pieau (1985)  Ectodermal invagination  ?  ?  ?  ?  ?  ?  Pythonidae  Python regius  Leal & Cohn (2015)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  Hemipenis base  Proximodistally  Natricidae  Natrix tessellata  Raynaud & Pieau (1985)  Ectodermal invagination  ?  After RPM  ?  ?  Hemipenis base  Proximodistally  Natricidae  Thamnophis sirtalis  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Colubridae  Lampropeltis triangulum  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  Dipsadidae  Diadophis punctatus  Clark (1945)  Ectodermal invagination  Hemipenis base  After RPM  Proximal to distal  Mesenchymal  ?  Proximodistally  View Large In the initial stages of development, the hemipenis is a solid protuberance of mesenchymal cells covered by ectodermal cells (Table 1). The development of the RPM starts by condensation and differentiation of mesenchymal cells in the middle region of the hemipenis base. Then, the RPM grows in the proximal and apical directions. The sulcus spermaticus developed after the development of the RPM as an invagination of the ectodermal cells (Clark, 1945; Raynaud & Pieau, 1985; Gredler et al., 2014; Leal & Cohn, 2015). Usually the sulcus anlage appears at the hemipenis base and grows proximally and apically. In Diadophis punctatus and Python regius, the distal subdivision of the RPM occurs before the bifurcation of the sulcus (Clark, 1945; Leal & Cohn, 2015). The inclusion of this embryological information in our character reconstruction analysis adds explanatory power to the primary homology hypothesis of the sulcus spermaticus condition. When we used the embryonic developmental sequence to order the character sulcus spermaticus, the ancestral state reconstructed at the Imantodini node was a bifurcated sulcus (Fig. 5A). However, using the character sulcus spermaticus unordered, the retrieved ancestral condition at the Imantodini node was ambiguous with three possible states: simple, bifurcated and bifurcated but ending in expanded area without ornamentation. Revisiting the proposed models for the evolution of sulcus spermaticus in dipsadid snakes (Myers, 2011), taking into account the comparative hemipenial embryological data, internal anatomy and phylogenetic reconstructions, leads us to conclude that the sulcus spermaticus in Imantodini evolved together with the RPM muscle, independent of evolutionary changes in other structures (e.g. hemipenial body). With this new perspective on hemipenial comparisons, we would be able to morphologically test the monophyletic hypotheses of the tribe. It will be necessary to include groups positioned closer to the tribe node in the analysis. These groups were recovered in the PhD thesis by Costa (2014), namely: Imantodes inornatus, I. gemmistratus, Leptodeira nigrofasciata, L. punctatus, L. uribei, L. frenata. The inclusion of these species in a future analysis will provide a more robust conclusion on the evolution of RPM and the sulcus. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s website: Table S1. Species used for the molecular analysis, used gene fragments and their GenBank accession number. Appendix S1. Molecular character matrix used in the phylogenetic analysis. Appendix S2. Ancestral reconstruction matrix. Appendix S3. Dipsadidae tree newick format. Figure S1. Mirrored maximum likelihood character reconstructions in Dipsadidae. Circles on nodes denote the reconstructed states for the characters. A, sulcus spermaticus condition. B, muscle retractor penis magnus in cavitas hemipenis. C, sulcus spermaticus length. D, insertion of the muscle retractor penis magnus. ACKNOWLEDGEMENTS J.C.L.C. thanks the Biomatters development team for the license Geneious 9.1.6. copyright license. R.A.G.-F. and J.C.L.C. are supported by fellowships from Programa de Capacitação Institucional (PCI/MPEG/MCTIC), grant numbers 300065/2016-7 and 312847/2015-7; A.F.R.M. is supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (PROTAX grant number 440413/2015-0). A.L.C.P. is supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (Pq. 305475/2014-2; PROTAX 440413/2015-0) and Fundação de Amparo a Estudos e Pesquisa (2016-111.449). Glen Sheppard for the English revision. The authors thank the anonymous reviewers and the Associate Editor for helping to improve the manuscript. REFERENCES Alberch P. 1985. Problems with the interpretation of developmental sequences. Systematic Zoology  34: 46– 58. Google Scholar CrossRef Search ADS   Arnold EN. 1984. Variation in the cloacal and hemipenial muscles of lizards and its bearing on their relationships. Symposia of the Zoological Society of London  52: 47– 85. Arnold EN. 1986. Why copulatory organs provide so many useful taxonomic characters: the origin and maintenance of hemipenial differences in lacertid lizards (Reptilia: Lacertidae). Biological Journal of the Linnean Society  29: 263– 281. Google Scholar CrossRef Search ADS   Bellairs A, Kamal AM. 1981. The chondrocranium and the development of the skull in recent reptiles. In: Gans C, Parsons TS, eds. Biology of the Reptilia . London & New York: Academic Press, 1– 263. Beuchelt H. 1936. Bau, Funktion und Entwicklung der Begattungsorgane der ma ̈nnlichen Ringelnatter (Natrix natrix L.) und Kreuzotter (Vipera berus). Morphologisches Jahrbuch  78: 445– 516. Bininda-Emonds ORP, Jeffery JE, Coates MI, Richardson MK. 2002. From Haeckel to event-pairing: the evolution of developmental sequences. Theory in Biosciences  121: 297– 320. Google Scholar CrossRef Search ADS   Bogert CM. 1940. Herpetological results of the Vernay Angola expedition. Bulletin of the American Museum of Natural History  77: 1– 107. Cadle JE. 1984a. Molecular systematics of Neotropical xenodontine snakes: I. South American xenodontines. Herpetologica  40: 8– 20. Cadle JE. 1984b. Molecular systematics of Neotropical xenodontine snakes: II. Central American xenodontines. Herpetologica  40: 21– 30. Clark H. 1945. The anatomy and embryology of the hemipenis of Lampropeltis, Diadophis and Thamnophis and their value as criteria of relationship in the family Colubridae. Proceedings of the Iowa Academy of Sciences  51: 411– 445. Cope ED. 1886. An analytical table of the genera of snakes. Proceedings of the American Philosophical Society  23: 479– 499. Cope ED. 1900. The crocodilians, lizards and snakes of North America. Report of the U.S. National Museum  1898: 153– 1270. Costa JCL. 2014. Análise filogenética de Leptodeira Fitzinger, 1843 e taxonomia das espécies do clado do sul do complexo Leptodeira annulata/septentrionalis (Serpentes, Dipsadidae) . Unpublished D. Phil. Thesis , Universidade Federal do Pará. Cundall D, Irish F. 2008. The snake skull. In: Carl G, Abbot SG, Adler K, eds. Biology of the Reptilia, Vol. 20 . Salt Lake City, UT: Society for the Study of Amphibians and Reptiles, 349– 692. Daza JM, Smith EN, Paez VP, Parkinson CL. 2009. Complex evolution in the Neotropics: the origin and diversification of the widespread genus Leptodeira (Serpentes: Colubridae). Molecular Phylogenetics and Evolution  53: 653– 667. Google Scholar CrossRef Search ADS PubMed  De Pinna MCC. 1991. Concepts and tests of homology in the cladistic paradigm. Cladistics  7: 367– 394. Google Scholar CrossRef Search ADS   De Pinna MCC. 1994. Ontogeny, rooting, and polarity. In: Scotland RW, Siebert DJ, Williams DM, eds. Models in phylogeny reconstruction, Vol. 52 . Oxford: Clarendon Press, 157– 721. Dowling HG. 1975. The Neartic snake fauna. In: Dowling HG, ed. 1974 Yearbook of Herpetology . New York: Herpetological Information Search System Publications, 191– 202. Dowling HG, Jenner JV. 1987. Taxonomy of American xenodontine snakes. II. The status and relationships of Pseudoleptodeira. Herpetologica  43: 190– 200. Dowling HG, Savage JM. 1960. A guide to the snake hemipenis: a survey of basic structure and systematic characteristics. Zoologica  45: 17– 27. Duellman WE. 1958a. A monographic study of the colubrid snake genus Leptodeira. Bulletin of the American Museum of Natural History  114: 1– 183. Duellman WE. 1958b. Systematic status of the colubrid snake, Leptodeira discolor Günther. University Kansas Publications of Museum of Natural History  11: 1– 9. Dunn ER. 1928. A tentative key and arrangement of the American genera of Colubridae. Bulletin of the Antivenin Institute of America  2: 18– 24. Figueroa A, McKelvy AD, Grismer LL, Bell CD, Lailvaux SP. 2016. A species-level phylogeny of extant snakes with description of a new colubrid subfamily and genus. PLoS One  11.9: e0161070. Google Scholar CrossRef Search ADS   Grazziotin FG, Zaher H, Murphy RW, Scrocchi G, Benavides MA, Zhang YP, Bonatto SL. 2012. Molecular phylogeny of the New World Dipsadidae (Serpentes: Colubroidea): a reappraisal. Cladistics  1: 1– 23. Gredler ML, Larkins CE, Leal F, Lewis AK, Herrera AM, Perriton CL, Sanger TJ, Cohn MJ. 2014. Evolution of external genitalia: insights from reptilian development. Sexual Development  8: 311– 326. Google Scholar CrossRef Search ADS PubMed  Gredler ML, Sanger TJ, Cohn MJ. 2015. Development of the cloaca, hemipenes, and hemiclitores in the green anole, Anolis carolinensis. Sexual Development  9: 21– 33. Google Scholar CrossRef Search ADS PubMed  Hillis DM, Bull JJ. 1993. An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis. Systematic Biology  42: 182– 192. Google Scholar CrossRef Search ADS   Jenner JV, Dowling HG. 1985. Taxonomy of American xenodontine snakes: the tribe Pseudoboini. Herpetologica  41: 161– 172. Katoh K, Misawa K, Kuma K, Miyata T. 2002. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Research  30: 3059– 3066. Google Scholar CrossRef Search ADS PubMed  Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T, Ashton B, Mentjies P, Drummond A. 2012. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics  28: 1647– 1649. Google Scholar CrossRef Search ADS PubMed  Keogh JS. 1996. Origin of the hemipenis retractor penis magnus muscles: evidence for utility in snake systematics. The Snake  27: 114– 119. Kraus F, Brown WM. 1998. Phylogenetic relationships of colubroid snakes based on mitochondrial DNA sequences. Zoological Journal of the Linnean Society  122: 455– 487. Google Scholar CrossRef Search ADS   Lanfear R, Calcott B, Ho SYW, Guindon S. 2012. PartitionFinder: combined selection of partitioning schemes and substitution models for phylogenetic analyses. Molecular Biology and Evolution  29: 1695– 1701. Google Scholar CrossRef Search ADS PubMed  Leal F, Cohn MJ. 2015. Development of hemipenes in the ball python snake Python regius. Sexual Development  9: 6– 20. Google Scholar CrossRef Search ADS PubMed  Mabee PM. 2000. The usefulness of ontogeny in interpreting morphological characters. In: Wiens JJ, ed. Phylogenetic analysis of morphological data . Washington, DC: Smithsonian Institution Press, 84– 114. Maddison WP, Maddison DR. 2011. Mesquite: a modular system for evolutionary analysis . Available at: http://mesquiteproject.org/mesquite/mesquite.html McCranie JR. 2011. The Snakes of Honduras – Systematics, Distribution, and Conservation. Michigan: Society for the Study of Amphibians and Reptiles. McDowell SB. 1961. Systematic division and evolution of the colubrid snake genus Natrix, with comments on the subfamily Natricinae, by Edmond V. Malnate [review]. Copeia  1961: 502– 506. Google Scholar CrossRef Search ADS   Miller MA, Pfeiffer W, Schwartz T. 2010. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. Proceedings of the Gateway Computing Environments Workshop (GCE), New Orleans, LA, 1– 8. Missassi AFR, Prudente ALC. 2015. A new species of Imantodes Duméril, 1853 (Serpentes, Dipsadidae) from the Eastern Cordillera of Colombia. Zootaxa  3980: 562– 574. Google Scholar CrossRef Search ADS PubMed  Mulcahy DG. 2007. Molecular systematics of Neotropical cat-eyed snakes: a test of the monophyly of Leptodeirini (Colubridae: Dipsadinae) with implications for character evolution and biogeography. Biological Journal of the Linnean Society  92: 483– 500. Google Scholar CrossRef Search ADS   Myers CW. 1974. The systematics of Rhadinaea (Colubridae), a genus of New World snakes. Bulletin of the American Museum of Natural History  153: 1– 262. Myers CW. 1982. Blunt-headed vine snakes (Imantodes) in Panama, including a new species and other revisionary notes. American Museum Novitates  2738: 1– 50. Myers CW. 2011. A new genus and new tribe for Enicognathus melanauchen Jan, 1863, a neglected South American snake (Colubridae: Xenodontinae), with taxonomic notes on some Dipsadinae. American Museum Novitates  3715: 1– 33. Google Scholar CrossRef Search ADS   Myers CW, Cadle JE. 2003. On the snake hemipenis, with notes on Psomophis and techniques of eversion: a response to Dowling. Herpetological Review  34: 295– 302. Myers CW, Cadle JE. 1994. A new genus for South American snakes related to Rhadinaea obtusa Cope (Colubridae) and resurrection of Taeniophallus Cope for the “Rhadinaea” brevirostris group. American Museum Novitates  3102: 1– 33. Myers CW, McDowell SB. 2014. New taxa and cryptic species of Neotropical snakes (Xenodontinae), with commentary on hemipenes as generic and specific characters. Bulletin of the American Museum of Natural History  385: 1– 112. Google Scholar CrossRef Search ADS   Pagel M. 1999. The maximum likelihood approach to reconstructing ancestral character states of discrete characters on phylogenies. Systematics Biology  48: 612– 622. Google Scholar CrossRef Search ADS   Patterson C. 1982. Morphological characters and homology. In: Joysey KA, Friday AE, eds. Problems of phylogenetic reconstruction . London: Academic Press, 21– 74. Peters JA, Orejas-Miranda B. 1970. Catalogue of the Neotropical Squamata Part 1. Snakes. United States National Museum Bulletin  297: 1– 347. Pinou T, Vicario S, Marschner M, Caccone A. 2004. Relict snakes of North America and their relationships within Caenophidia, using likelihood-based Bayesian methods on mitochondrial sequences. Molecular Phylogenetic and Evolution  32: 563– 574. Google Scholar CrossRef Search ADS   Porto M, de Oliveira MA, Pissinatti L, Rodrigues RL, Rojas-Moscoso JA, Cogo JC, Metze K, Antunes E, Nahoum C, Mónica FZ, de Nucci G. 2013. The evolutionary implications of hemipenial morphology of rattlesnake Crotalus durissus terrificus (Laurent, 1768) (Serpentes: Viperidae: Crotalinae). PLoS One  8: 1– 8. Google Scholar CrossRef Search ADS   Pyron RA, Burbrink FT, Wiens JJ. 2013. A phylogeny and revised classification of Squamata, including 4161 species of lizards and snakes. BMC Evolutionary Biology  13: 93. Google Scholar CrossRef Search ADS PubMed  Raynaud A, Pieau C. 1985. Embryonic development of the genital system. In: Gans C, Billett F, Maderson PFA, eds. Biology of the Reptilia: development B . New York: John Wiley and Sons, 151– 300. Reyes-Velasco J, Mulcahy DG. 2010. Additional taxonomic remarks on the genus Pseudoleptodeira (Serpentes: Colubridae) and the phylogenetic placement of “P. uribei”. Herpetologica  66: 99– 110. Google Scholar CrossRef Search ADS   Savage JM. 1997. On terminology for the description of the hemipenis of squamate reptiles. Journal of Herpetology  7: 23– 25. Savage JM. 2002. The amphibians and reptiles of Costa Rica. Chicago, IL: The University Chicago Press. Schluter D, Price T, Mooers AO, Ludwig D. 1997. Likelihood of ancestor states in adaptive radiation. Evolution  51: 1699– 1711. Google Scholar CrossRef Search ADS PubMed  Slowinski JB, Lawson R. 2002. Snake phylogeny: evidence from nuclear and mitochondrial genes. Molecular Phylogenetics and Evolution  24: 194– 202. Google Scholar CrossRef Search ADS PubMed  Springer VG, Johnson GD. 2000. Use and advantages of ethanol solution of alizarin red S dye for staining bone in fishes. Copeia  2000: 300. Google Scholar CrossRef Search ADS   Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics  30: 1312– 1313. Google Scholar CrossRef Search ADS PubMed  Torres-Carvajal O, Yánez-Muñoz MH, Quirola D, Smith EN, Almendáriz A. 2012. A new species of blunt-headed vine snake (Colubridae, Imantodes) from the Chocó region of Ecuador. ZooKeys  244: 91– 110. Google Scholar CrossRef Search ADS   Underwood G. 1967. A contribution to the classification of snakes . London: Trustees of the British Museum (Natural History). Vaidya G, Lohman DJ, Meier R. 2011. SequenceMatrix: concatenation software for the fast assembly of multi-gene datasets with character set and codon information. Cladistics  27: 171– 180. Google Scholar CrossRef Search ADS   Vellard J. 1928. Importance des caractères fournis par l’hemipénis pour la classification des ophidiens. Bulletin de la Societé Zoologique de France  53: 406– 418. Vellard J. 1946. Morphologia del hemipenis y evolución de los ofidios. Acta Zoologica Lilloana  3: 263– 288. Vidal N, Delmas AS, David P, Cruaud C, Couloux A, Hedges SB. 2007. The phylogeny and classification of caenophidian snakes inferred from seven nuclear protein-coding genes. Comptes Rendus Biologies  330: 182– 187. Google Scholar CrossRef Search ADS PubMed  Werneburg I. 2009. A standard system to study vertebrate embryos. PLoS One  6: e5887. Google Scholar CrossRef Search ADS   Zaher H. 1999. Hemipenial morphology of the South American xenodontine snakes, with a proposal for a monophyletic Xenodontinae and a reappraisal of colubroid hemipenes. Bulletin of the American Museum of Natural History  240: 1– 168. Zaher H, Prudente ALC. 2003. Hemipenes of Siphlophis (Serpentes, Xenodontinae) and techniques of hemipenial preparation in snakes: a response to Dowling. Herpetological Review  34: 302– 307. Zaher H, Grazziotin FG, Cadle JE, Murphy RW, de Moura-Leite JC, Bonatto SL. 2009. Molecular phylogeny of advanced snakes (Serpentes, Caenophidia) with an emphasis on South American xenodontines: a revised classification and descriptions of new taxa. Papéis Avulsos de Zoologia  49: 115– 153. Google Scholar CrossRef Search ADS   Zaher H, Grazziotin FG, Graboski R, Fuentes RG, Sánchez-Martinez P, Montingelli GG, Zhang YP, Murphy RW. 2012. Phylogenetic relationships of the genus Sibynophis (Serpentes: Colubroidea). Papéis Avulsos de Zoologia  52: 141– 150. Ziegler T, Böhme W. 1997. Genitalstrukturen und Paarungsbiologie bei squamaten Reptilien, speziell den Platynota, mit Bemerkungen zur Systematik. Mertensiella  8: 1– 207. APPENDIX Hemipenial anatomy examined. Aniliidae: Anilius scytale MPEG 25828, MPEG 62373. Boidae: Corallus hortulanus MPEG 18920. Viperidae: Bothrops atrox MPEG 18985. Elapidae: Micrurus lemniscatus MPEG 26392. Colubridae: Chironius exoletus MPEG 24296; Leptophis ahaetulla MPEG 21275; Tantilla melanocephala MPEG 24568. Dipsadidae: Apostolepis quinquelineata MPEG 1700; Atractus albuquerquei MBS 191; Atractus latifrons MPEG 17837; Dipsas catesbyi MPEG 23291; Dipsas indica MPEG 24209; Erythrolamprus reginae MPEG 25458; Helicops angulatus MPEG 4402, MPEG 22297; Hydrops triangularis MPEG 2955; Imantodes cenchoa MPEG 26270, MPEG 26275, MPEG 26269; Imantodes lentiferus MPEG 23572; Leptodeira annulata annulata MPEG 2421; Leptodeira annulata ashmeadi MZUSP 10654; Leptodeira annulata pulchriceps MZUSP 9558; Leptodeira annulata rhombifera MZUSP 17435; Oxyrhopus petolarius MPEG 12266; Philodryas olfersii MPEG 24499; Pseudoboa neuwiedii MPEG 20693; Psomophis joberti MPEG 24868; Sibon nebulatus MPEG 1785; Sibynomorphus mikanii MPEG 21694; Taeniophallus brevirostris MPEG 5033; Taeniophallus nicagus MPEG 23312; Taeniophallus quadriocelatus MPEG 22098; Xenopholis scalaris MPEG 23163. © 2017 The Linnean Society of London, Zoological Journal of the Linnean Society

Journal

Zoological Journal of the Linnean SocietyOxford University Press

Published: Dec 13, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off