Copeoglossum Tschudi, 1845
|
publication ID |
https://doi.org/ 10.11646/zootaxa.3288.1.1 |
|
persistent identifier |
https://treatment.plazi.org/id/39191A7F-074C-FFBA-2DA9-EBD27E2CF94F |
|
treatment provided by |
Felipe |
|
scientific name |
Copeoglossum Tschudi, 1845 |
| status |
|
Genus Copeoglossum Tschudi, 1845
Neotropical Spotted Skinks
Copeoglossum Tschudi, 1845:162 . Type species: Copeoglossum cinctum Tschudi, 1845:162 , by monotypy; = Mabuya nigropunctata ( Spix 1825) ; synonymy by Avila-Pires, 1995:584.
Diagnosis. Species in this genus are characterized by (1) frontoparietals, two, (2) supraciliaries, 3–6 (usually 5–6), (3) supraoculars, four (rarely three), (4) prefrontal contact, absent (or contact rarely), (5) parietal contact, absent (or contact rarely), (6) rows of nuchals, one (occasionally no rows), (7) dorsals + ventrals, 105–120, (8) total lamellae, 196–253, (9) a dark middorsal stripe, absent, (10) dark dorsolateral stripes, absent, (11) a dark lateral stripe, present, and (12) dark ventral striping, absent (chin spotting in Copeoglossum margaritae sp. nov.). They are large, with a range of maximum body sizes among the species of 91–121 mm SVL ( Table 2).
The combination of the above characters, but especially the separation of the parietal scales (in almost all individuals) and low dorsal + ventral counts, distinguishes this genus from all other genera. In most (excluding Copeoglossum redondae sp. nov.), the sublabials extend all the way forward to the fused postmental (or nearly so), and this is another useful character, although not unique in the genus (e.g., it occurs in Notomabuya frenata ). The dark lateral stripe of species in this genus usually extends all the way back to the hindlimbs (or further), and the dorsum, which lacks dark dorsolateral stripes and a pale middorsal stripe, is usually covered with scattered dark spots, but coloration and pattern are variable.
Content. Five species are placed in the genus: Copeoglossum arajara , C. aurae sp. nov., C. margaritae sp. nov., C. nigropunctatum , and C. redondae sp. nov. ( Table 1).
Distribution. The genus is distributed throughout much of the Amazon basin and other regions of central, northern, and eastern South America, including Brazil, Bolivia, Colombia, Ecuador, Peru, Venezuela (including Isla de Margarita), Guyana, French Guiana, and Suriname. It also occurs in the Lesser Antilles (Redonda, St. Vincent, the Grenadines, and Grenada), Trinidad, and Tobago ( Figs. 1 View FIGURE 1 , 8C View FIGURE 8 , 9D View FIGURE 9 , and 11D–E, I–J View FIGURE 11 ).
Etymology. The generic name ( Copeoglossum ) is a neuter noun, derived from the Greek nouns kopeus (chisel) and glossa (tongue), in allusion to the shape of the tongue.
Remarks. Avila-Pires ( Avila-Pires 1995) redefined Copeoglossum nigropunctatum and designated a neotype from Brazil. Since then, the species has been shown to be distributed even more widely ( Miralles et al. 2005a; Miralles & Carranza 2010). However, molecular phylogenetic evidence here ( Fig. 5 View FIGURE 5 ), and evidence presented previously with most of the same sequences ( Whiting et al. 2006; Miralles and Carranza 2010), has revealed that the species is comprised of genetically diverse populations with long branch lengths, indicating the presence of unnamed species. Miralles and Carranza (2010) did not make a revision of C. nigropunctatum , and did not diagnose taxa, but instead suggested that three clades could be recognized within what is currently C. nigropunctatum , and gave them geographic names: (1) the "Occidental Clade," a northern (including Caribbean islands) and western group that presumably encompasses the type-locality of C. nigropunctatum , (2) the "Oriental Clade," extending from the Atlantic to western and southern Brazil, and (3) the "Meridional Clade," distributed in southern Brazil. Miralles and Carranza (2010) further suggested the three clades represent three species, recommending that the name Euprepis surinamensis Hallowell (1857) be used for the Oriental Clade. However, the holotype of that species is missing and presumed lost ( Dunn 1936), a serious problem for allocating the name to one of several overlapping clades (within the Oriental Clade) in northeast South America (see below). Also, Tiliqua aenea Gray (1831a) is an earlier available name for a South American (not Antillean) skink species within the Genus Copeoglossum (see discussion below, in Remarks for C. aurae sp. nov.). No molecular data are yet available for Mabuya arajara , but Miralles and Carranza (2011) suggested that it belongs to the Oriental Clade. Therefore, the Genus Copeoglossum as recognized here is synonymous with the " Mabuya nigropunctata species complex" of Miralles and Carranza (2011).
Our interpretation of the number and distribution of unnamed species in Copeoglossum differs from that of Miralles and Carranza (2010). We examined museum material of this genus, from the Occidental and Oriental clades (Appendix 2) and Caribbean islands, and diagnose three new species from Caribbean islands. DNA sequences are not available for two of those ( C. margaritae sp. nov. and C. redondae sp. nov.) but sequences from the other species ( C. aurae sp. nov.) indicate that it belongs to the Occidental Clade and differs from C. nigropunctatum sequences in that clade (samples 1–2, 5–9, 31; presumably C. nigropunctatum sensu stricto) by 5% sequence divergence (cyt b) and from a Venezuelan sample (sample 41) by 2.6% ( Fig. 6 View FIGURE 6 ). In terms of degree of morphological and molecular divergence, this is consistent with our results for species differences in other genera of mabuyine skinks ( Figs. 5–7 View FIGURE 5 View FIGURE 6 View FIGURE 7 ), and therefore, in contrast to Miralles and Carranza (2010), we conclude that each of these three clades of Copeoglossum contain multiple species, instead of a single species.
Until a revision is made of the South American material of Copeoglossum , beyond the scope of this study, it is not possible to assign available names or new names to clades in that region. The three clades defined by Miralles and Carranza (2010) overlap broadly, in contrast to their geographic names. Based on the distribution of localities within each clade, their relationships, and levels of sequence divergence, we propose that Copeoglossum contains at least 14 species, including the three diagnosed here from Caribbean islands. We have not used a strict sequence divergence cutoff to diagnose species. Instead, we note that our putative species show a geographic cohesiveness below about 1–2% (cyt b) divergence and lose that cohesiveness above about 2–3% divergence. This is consistent with skink species on Caribbean islands where we find that several morphologically diagnosable species differ by only 1.5% sequence divergence, and sympatric species (e.g., S. semitaeniatus and S. sloanii ) differ by as little as
Concerning the Occidental Clade, there are at least five species. We agree with Miralles and Carranza (2010) that the name Copeoglossum nigropunctatum probably goes with this clade, and most likely with samples 5–9 in our analysis. Those samples are from the state of Amazonas, Brazil, which contains the type-locality. We also include samples 1–2 (Porto Walter, Acre, Brazil) and sample 31 (Puerto Inirida, Guainia, Colombia) in this putative Species A. It is uncertain whether all of these samples represent the same species, yet they have a relatively small amount of sequence divergence (<2%) considering their great geographic separation (up to ~ 1500 km). Species B, an unnamed species, is represented in the tree by sample 41, from north-central Venezuela (Turiamo, Aragua). It is the closest relative of C. aurae sp. nov. ( Figs. 5–7 View FIGURE 5 View FIGURE 6 View FIGURE 7 ). We examined museum material from that region and elsewhere in Venezuela (Appendix 2) as part of our diagnosis of the three Caribbean species. We tentatively treat the other two Caribbean island species, C. margaritae sp. nov. and C. redondae sp. nov., as members of the Occidental clade, based largely on geography, but their relationships within the Genus Copeoglossum are not known. Atlantic ocean currents ( Hedges 1996b) also may have brought the ancestor of C. redondae sp. nov. from eastern South America (Oriental Clade).
Within the Meridional Clade, we have identified three putative species (C–E). Species C is from the western Brazilian Highlands (samples 23–24, and 26; all from Rondônia, Brazil). As noted by Miralles and Carranza (2010), there is sympatry at that locality with another putative species (our Species I, of the Oriental Clade). We tentatively include sample 10 (Brasilía, Brazil) in this species, from ~ 2000 km to the east. It shows a relatively small amount of divergence (~2%) from the Rondônia samples of Species C in Fig. 6 View FIGURE 6 , but it is missing a sequence from the fast-evolving cyt b gene, which would otherwise increase that overall divergence. Species D is represented by sample 15 (Aripuanã, Mato Grosso, Brazil). As noted by Miralles and Carranza (2010), there is sympatry at that locality with another putative species (again, our Species I, of the Oriental Clade). Species E includes Brazilian samples 13 (Niquelandia, Goias) and 16 (UHE Manso, Mato Grosso) with relatively small sequence divergence (1.4%; Fig. 6 View FIGURE 6 ) between them despite a large geographic separation (~ 800 km). The cyt b sequence divergence among species C–E is as high as 8% ( Fig. 6 View FIGURE 6 ).
Within the Oriental Clade, we have identified five putative species, each geographically distinct and with several occurring broadly and narrowly in sympatry in the Guiana Highlands, and separated from each other by 3.0–8.5% ( Fig. 6 View FIGURE 6 ). Species F is from the Brazilian Highlands (sample 21; Piaui, Brazil). Species G is broadly distributed and occurs in the central (samples 27–30; Mount Roraima) and eastern (samples 17–20, 32, and 39; southern Guyana, and Pic Coudreau, French Guiana) Guiana Highlands southward and eastward to the Amazon delta area (samples 27–30; three locations in Pará, Brazil) and over to the northeast Brazilian Highlands (samples 11–12; Ceara, Brazil). Species H is from the Guiana Highlands (samples 35–38, 40, and 42; St. Eugène and Pic Coudreau). Species I is from the western Brazilian Highlands (samples 14, 22, and 25; Mato Grosso Plateau). Species J is also from the Guiana Highlands (samples 3–4, and 33–34; Mitaraka, French Guiana and Amapá, Brazil). Two species (G and H) are sympatric at Pic Coudreau, French Guiana (samples 39–40). These two putative species are not each other's closest relatives, and have a sequence divergence (cyt b) of 4.7 % ( Fig. 6 View FIGURE 6 ).
In summary, we suggest here that the Copeoglossum contains at least 14 species, three of which are diagnosed and named below. That interpretation is consistent with morphological and genetic divergence in mabuyine skinks of Caribbean islands. Instead of applying a strict genetic divergence cutoff to arrive at that conclusion, we use phylogenetic pattern, geography, sympatry, and sequence divergence. There were insufficient samples to perform an ABGD analysis ( Puillandre et al. 2011). However, this may be a useful tool for species delimitation in mabuyines in the future, as more samples and DNA sequences become available. Until a revision is made of South American members of Copeoglossum , we recommend continued use of the name C. nigropunctatum for the unnamed putative species discussed above.
No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.
