INTRODUCTION

Phalotris Cope, 1862 is a speciose genus of Neotropical dipsadid snakes, with 15 small to medium sized species (< 1 m) that are widely distributed in open areas from Northeastern Brazil to Southern Argentina (Ferrarezzi, 1993; Costa & Bérnils, 2018). Together with the genera Apostolepis Cope, 1862, Coronelaps Lema & Deiques, 2010, Elapomorphus Wiegmann, 1843, they constitute the Elapomorphini Jan, 1862 tribe, a clade of approximately 50 cryptozoic and fossorial species (Ferrarezzi, 1993; Zaher, 1999; Zaher et al., 2009; Grazziotin et al., 2012; Entiauspe-Neto et al., 2019). The Elapomorphini tribe can be diagnosed on the basis of a short dentigerous dentary teeth process, U-shaped fronto-parietal suture, two or less pterygoid palatine teeth, usually six or less supralabial scales, and an entire nasal plate (Ferrarezzi, 1993; Zaher, 1999; Zaher et al., 2009). The morphological diagnosis of Phalotris is based on a single (fused) prefrontal scale, separated from two internasals (Ferrarezzi, 1993).

As in other Elapomorphini genera, Phalotris has been long subjected to taxonomic controversy. Some of its species are known based on a single or a small series of specimens (e.g. Phalotris concolor Ferrarezzi, 1994 [n = 4], Phalotris multipunctatus Ferrarezzi, 1994 [n = 4], Phalotris nigrilatus Ferrarezzi, 1993 [n = 15], Phalotris normanscotti Cabral & Cacciali, 2015 [n = 3]) even after long time periods since their description (Atkinson et al., 2019; Cacciali et al., 2020); others have had conflicting views among authors regarding their delimitations and validity (e.g. Phalotris bilineatus Duméril, Bibrón & Duméril, 1852; Phalotris lemniscatus Duméril, Bibrón & Duméril, 1852; Phalotris reticulatus Peters, 1860), with various interpretations (see Lema, 1979; Ferrarezzi, 1993).

Ferrarezzi (1994) described Phalotris concolor on the basis of an adult female specimen (IBSP 55018) from Cristália, Minas Gerais state, in the Cerrado of Southeastern Brazil. It was distinguished from its congeners by a suite of morphological characters, and said to be “distanced” from others due to a higher number of ventral scales. It remained known from its holotype (and subsequently, lost in the Instituto Butantan fire of 2010), until Moura et al. (2013) reported its rediscovery, based on two specimens from Brasilândia de Minas and Urucuia, both at Cerrado localities in Minas Gerais, Southeastern Brazil, while also providing a detailed redescription and an emended diagnosis. Freitas et al. (2016) reported its first record to Bahia, Northeastern Brazil, based on a specimen from the Cerrado of Cocos, close to the border with Minas Gerais.

Recently, Silveira (2020) described Phalotris cerradensis, on the basis of the same specimen reported by Freitas et al. (2016), a subadult male individual (CHUNB 51553) from Cocos, Bahia, Northeastern Brazil, providing only brief comments on its morphology and allocating it in the Phalotris nasutus species group, together with P. concolor Ferrarezzi, 1993; P. labiomaculatus Lema, 2002; P. lativittatus Ferrarezzi, 1994; P. nasutus Gomes, 1915; and P. nigrilatus Ferrarezzi, 1994. Silveira (2020) also states P. cerradensis is “phenotypically similar” to P. concolor. In this work, we reanalyze the holotype of P. cerradensis, and review its taxonomic status.

MATERIAL AND METHODS

We examined a total of 38 Phalotris specimens from the following collections: Coleção Herpetológica Universidade de Brasília (CHUNB), Coleção Herpetológica Universidade Federal do Rio Grande (CHFURG), Coleção Herpetológica Museu de Ciências e Tecnologia PUCRS (MCP), Fundacíon Miguel Lilo (FML), Museo Nacional de Historia Natural de Paraguay (MHNP), Museu de Zoologia João Moojen de Oliveira (MZUFV), Coleção Zoológica Universidade Federal de Santa Maria (ZUFSM). A list of the examined material is provided in Appendix. The geographic distribution of the Phalotris nasutus group was inferred based on records from Ferrarezzi (1994), Freitas et al. (2016), Nogueira et al. (2019), and Cacciali et al. (2020).

An emended diagnosis, as well as meristic and morphometric characters followed the nomenclature used by Entiauspe-Neto et al. (2020) for Apostolepis, as follows: head length, measured from the center of rostral to the corner of mouth; head width, measured at the corner of mouth; snout-vent length, ventrally measured from the center of rostral to the posterior margin of cloacal scale; tail length, measured from the posterior margin of cloacal scale to the terminal scale. Head and tail measurements were taken with a dial caliper to the nearest 0.01 mm; for others, a flexible ruler was used. Scale counts follow Dowling (1951). Sex was determined through a subcaudal incision.

RESULTS

Upon examination of the holotype of P. cerradensis, it becomes evident that the correct specimen number is in fact “CHUNB 51553”, and not “CHUNB 52553”, as stated in its description. According to Silveira (2020), the type specimen differs from its congeners by the following characters: a prominent rostral scale (projected), but with a rounded apex; presence of contact between the rostral and the prefrontal, which separates the pair of internasals, prefrontal with a more pronounced angle (evident) on its anterior edge; 1+1 temporals, fifth labial (supralabial) separated from the parietal for its anterior temporal; the number of ventral scales relatively high (202 in a male); the number of subcaudal scales also relatively high (36 pairs in a male); dorsum with uniform coloration, yellow in preservative and reddish-orange in life (with a vestigial dark vertebral stripe); venter with clear (sic) immaculate coloration, cream in preservative, with black blotches on chinshields; a pronounced and narrow anterior nuchal white collar, one to two wide dorsal at vertebral and paravertebral rows (smaller than its black nuchal collar); a pronounced and wide black nuchal collar, three to five wide dorsals at vertebral and paravertebral rows (Silveira, 2020: 54).

Furthermore, Silveira (2020) provides a comparison of P. cerradensis with P. concolor (characters in parentheses), distinguishing it from the latter by the presence of black dorsal and lateral head surfaces, with white supralabial blotches (versus darker color of dorsal surface of the head, pale or cream supralabial surfaces); white nuchal collar that is narrower than black nuchal collar (versus white nuchal collar broader than black nuchal collar).

Both characters, head coloration and nuchal collar size, are also regarded as highly polymorphic in Elapomorphini species (see Entiauspe-Neto et al., 2020). There is only minor difference in the extent of nuchal collars over dorsal scale rows in P. cerradensis (white nuchal collar 1-2 scales wide, black nuchal collar 3-5 scales wide) and P. concolor (white nuchal collar 2-3 scales wide, black nuchal collar 1-3 scales wide).

The holotype of Phalotris cerradensis shares several similarities with the known specimens of P. concolor (Figs 1, 2): (1) rostral projected in lateral view, rounded in dorsal view; (2) temporals 1+1, fifth supralabial separated from parietal by anterior temporal; (3) a high number of ventrals in males (202 in the male P. cerradensis, 212 in the male P. concolor); (4) a high number of subcaudals in males (36 pairs in the male P. cerradensis, 34 in the male P. concolor); (5) orange or red dorsal coloration in life; (6) immaculate cream venter.

Both type localities (Cristália and Cocos) are situated within the Cerrado biome, and Phalotris concolor has been recorded close to P. cerradensis (Fig. 3). The only remaining difference is in coloration characters that are generally highly variable in these snakes. It should also be noted that P. concolor is known from fewer than 5 individuals, and these differences likely constitute the extremes of morphological variation for these characters (Table 1). Considering the lack of specific diagnostic characters and a significant overlap in morphology, we propose that P. cerradensis should be considered as a junior synonym of P. concolor. Phalotris concolor can be diagnosed based on (1) 15-15-15 dorsal scales; (2) preocular present, contacting nasal; (3) loreal absent; (4) temporals 1+1 (rarely 1+1+1); (5) supralabials six, 2nd-3rd in contact with orbit; (6) infralabials seven, 1st-4th in contact with anterior chinshields; (7) ventrals 202-224 (220-224 in females, 202-212 in males); (8) subcaudals 28-36 (28-29 in females, 34-36 in males); (9) dorsal pattern uniformly red or orange, vertebral stripe, if present, vestigial; (10) ventral pattern uniformly cream, head and gular region white, diffuse black pigmentation on infralabials, chinshield, gulars, and symphisal; (11) white nuchal collar covering 1-3 rows, black nuchal collar covering 1-5 rows; (12) supralabial blotch large or medium, covering four to six supralabials; (13) SVL 184-522 mm, TL 23-60 mm.

Fig. 1.

Scalation of Phalotris concolor (MZUFV 1870).

Fig. 2.

Morphological variation of Phalotris concolor. Top: Juvenile male (CHUNB 51553, former holotype of P. cerradensis); Middle: Adult male (MZUFV 1890); Bottom: Adult female (MZUFV 1870).

Fig. 3.

Geographic distribution of the Phalotris nasutus species group.

Table 1.

Morphological variation of Phalotris concolor. Measurements are given in mm. Data from ¹Ferrarezzi (1994); ²Moura et al. (2013).

DISCUSSION

The number of Phalotris species is herein reduced to 14. Recent rediscoveries (e.g. P. nigrilatus, Cacciali et al., 2007, 2020; P. concolor, Moura et al., 2013; Phalotris multipunctatus, Atkinson et al., 2018) have highlighted that these secretive snakes may go undetected for long periods, even in historically well-sampled areas. More than that, these findings particularly call attention to the slight variations in meristic, morphometric and color pattern characters, even in specimens collected in the same locality. Not by chance, P. lemniscatus, the congener whose morphology has been most extensively studied (> 300 specimens), is the species that exhibits chromatic variations the most, with at least 8 recognizable patterns and some intermediate specimens between them (Noronha, 2012). If we consider the putatively diagnostic characteristics between P. cerradensis and P. concolor (head coloration and nuchal collar size) in the light of the variation of P. lemniscatus, it is possible to notice that both are very fragile. While the head color can vary from black to light brown in P. lemniscatus, the nuchal collar size is widely variable and may even be absent in many specimens (Noronha, 2012; this study). In addition, the same pattern variation in the nuchal collar has already been pointed out for more than one species of the genus Apostolepis, which suggests that this characteristic is highly variable not only in Phalotris, but possibly in Elapomorphini as a whole (Ferrarezzi et al., 2005; Entiauspe-Neto et al., 2020).

The most surprising feature in Silveira's (2020) paper, however, is that it contemplates exactly the combination of two factors that have generated numerous taxonomic problems in Elapomorphini throughout history: poor diagnoses based on a single specimen. This practice is obviously not exclusive to this group of snakes, and its potential negative consequences have been reported by different authors (e.g. Cruz-da-Silva et al., 2018; Palacios-Aguilar & García-Vázquez, 2020). However, possibly due to the difficulty in obtaining large series of specimens, together with the high chromatic variability, this practice has become frequent in Elapomorphini (Lema, 2004a, b; Lema & Renner, 2011). This problem was addressed indirectly by Ferrarezzi et al. (2005), but Entiauspe-Neto et al. (2019) clearly exposed this point of view through direct association between single-based descriptions and subsequent synonymizations, and warned: “We urge our fellow authors not to commit taxonomic malpractice and to carefully generate, rethink and analyze their data, providing compelling evidence for their claims” (our bold). Currently, the snake's systematics has evolved in order to increasingly combine sources of evidence to support taxonomic decisions (e.g. cranial osteology, muscular, glandular and hemipenial morphology, and different molecular frameworks) and minimize nomenclatural instabilities. We argue that, while many of these sources of evidence do not yet exist in the scientific literature for much of the species of Elapomorphini, this should not be seen as a reason to continue a tortuous cycle of rudimentary descriptions. On the contrary, it should be seen as an excellent opportunity to put into practice a more robust taxonomy and really provide useful information regarding the diversity and evolutionary relationships of this elusive neotropical snake lineage.

Phalotris concolor is a poorly known species, known from four specimens, and restricted to the Cerrado open areas of the Espinhaço Range, a narrow mountain formation between the states of Minas Gerais and Bahia. There is no available information on its populational trends or conservation, and very little is known about its natural history. We suggest that further fieldwork on the Espinhaço Range should uncover new specimens, and help enrich the knowledge on the variation of this elusive species.