Bachia oxyrhina, Rodrigues, Miguel Trefaut, Camacho, Agustín, Nunes, Pedro Murilo Sales, Recoder, Renato Sousa, Jr, Mauro Teixeira, Valdujo, Paula H., Ghellere, José Mário B., Mott, Tamí & Nogueira, Cristiano, 2008

Rodrigues, Miguel Trefaut, Camacho, Agustín, Nunes, Pedro Murilo Sales, Recoder, Renato Sousa, Jr, Mauro Teixeira, Valdujo, Paula H., Ghellere, José Mário B., Mott, Tamí & Nogueira, Cristiano, 2008, A new species of the lizard genus Bachia (Squamata: Gymnophthalmidae) from the Cerrados of Central Brazil, Zootaxa 1875, pp. 39-50 : 40-47

publication ID	https://doi.org/ 10.5281/zenodo.184077
DOI	https://doi.org/10.5281/zenodo.5684781
persistent identifier	https://treatment.plazi.org/id/039D5067-A94C-5152-FF08-73B9FE459C94
treatment provided by	Plazi
scientific name	Bachia oxyrhina
status	sp. nov.

Treatment

Bachia oxyrhina , sp. nov.

( Figs. 1 View FIGURE 1 , 2 View FIGURE 2 )

Holotype: MZUSP 98086 ( Fig. 1 View FIGURE 1 ), an adult male from Morro do Fumo, Estação Ecológica Serra Geral do Tocantins (10°51’58.41”S, 46°49’9.07”W), Mateiros municipality, state of Tocantins, Brazil, collected by the authors of this paper on February 15th, 2008. Field number PHV 2208.

Paratypes: MZUSP 98080 (11°6’7.92”S, 46°45’57.60”W), MZUSP 98082 (11°7’25.68”S, 46°47’26.88”W), MZUSP 98083 ( Fig. 2 View FIGURE 2 A) (11°7’12.36”S, 46°47’12.48”W) (females); MZUSP 98081 (11°6’7.92”S, 46°45’57.60”W), and MZUSP 98084 ( Fig. 2 View FIGURE 2 C) (11°18’41.29”S, 46°56’6.36”W) (males); all collected by the authors of this paper between January 28th and February 4th 2008, all localities in Estação Ecológica Serra Geral do Tocantins, Almas municipality, state of Tocantins, Brazil.

Etymology: The specific name derives from the Greek “ oxy ” (sharp, spatulate, wedge shaped), and “ rhino ” (nose) being a reference to the pronounced, wedge shaped nose of this species, an adaptation to life in the sandy habitats where it occurs.

Diagnosis: ( Table 1 View TABLE 1 ) A species of the bresslaui group having lanceolate dorsal and lateral body scales, quadrangular and juxtaposed smooth ventrals, tail scales lanceolate, imbricate, keeled, 1–1 femoral pores and 1–1 preanal pores in males (only preanal pores in females), interparietal, supraoculars and superciliaries present, 42–45 dorsals, 34–36 ventrals, and 29–30 scales around midbody. Snout highly prominent and wedge shaped, distinctively projecting over lower jaw. Fore limb and hind limb rudimentary, stiliform, ending in one apical scale. Five supralabials; fifth the largest and the highest, contacting or not parietal. Anterior portion of nasal scale fused with first supralabial. One enlarged temporal scale contacting or not postocular. Two supraoculars; second small, restricted to the lateral face of head, allowing extensive contact between parietal and first supraocular. Width of first supraocular less than 1/3 of the anterior margin of frontal.

Bachia oxyrhina can be immediately distinguished from B. panoplia and B. pyburni by the absence of prefrontals, present and in contact at midline in both latter species. In Bachia scolecoides , prefrontals are also present but widely separated and reduced in size. Species of Bachia of the bresslaui group without prefrontals are B. bresslaui , B. cacerensis , B. micromela B. psamophila , and B. oxyrhina . Except for Bachia oxyrhina that has five supralabials, all other species of bresslaui group have six. Bachia oxyrhina is also unique in having the anterior portion of nasal scale fused to first supralabial and the lower number of scales around body (29– 30 vs. 33–47 in other species). Bachia oxyrhina resembles more closely B. psamophila and B. micromela , especially the former by its pronounced and prominent wedge shaped snout. Besides the above mentioned characters Bachia oxyrhina differs from the latter species by presenting the second pair of chin shields broadly separated at midline (respectively in slight and broad contact at midline in B. psamophila and B. micromela ). In Bachia bresslaui , B. scolecoides , B. panoplia , and B. pyburni there is no contact between supralabials and parietal. Contact between supralabials and parietal is variable in Bachia cacerensis , B. psamophila , B. micromela , and B. oxyrhina . In one of the three known specimens of Bachia cacerensis there is slight contact between parietal and 6th labial. In Bachia psamophila the 6th supralabial is the largest and highest and contacts the parietal; in B. micromela it is the 5th supralabial that contacts the parietal. The condition varies in Bachia oxyrhina where either it is the 5th supralabial that contacts punctually the parietal or their contact is prevented by a slight contact between postocular and temporal. This is a different condition than the one observed in Bachia psamophila where there are six labials and an enlarged postocular separates the 5th supralabial from parietal. Bachia cacerensis is unique in the bresslaui group in having four unclawed finger-like apical scales in the forelimb; all other species, B. oxyrhina , B. psamophila , B. bresslaui and B. micromela have just one apical scale in the forelimb, in the last species ending in an ungual sheath. The hind limb of Bachia oxyrhina , like those of B. bresslaui and B. cacerensis ends with one apical scale, whereas there are two in B. micromela and four clawed toes in B. psamophila . A comparative list of characters are shown in Table 1 View TABLE 1 .

Description of the holotype: ( Fig. 1 View FIGURE 1 ) Body elongate, with a slight cervical constriction on head, snout highly prominent and wedge shaped, tail longer than body. Rostral broad, prominent, contacting first supralabial, nasal and frontonasal. Viewed from above the rostral is about twice as wide as high; on lateral view it projects broadly anteriorly toward, forming a horizontal surface ventrally, at the level of ventral surface of the upper lips. Frontonasal trapezoidal, wider than long, wider posteriorly, contacting rostral, nasal, first supraocular and frontal. Prefrontals absent. Frontal pentagonal, longer than wide, with anterior margin straight, as wide as and in broad contact with frontonasal; lateral margins straight to slightly concave, in contact on each sides with first supraocular; posteriorly angulose, broadly contacting parietals and in short contact with interparietal. Frontal more than four times wider than anterior supraocular. Frontoparietals absent. Interparietal narrow, longer than wide, subrectangular, slightly wider posteriorly, shorter than frontal and parietals. Parietals very large, longer than wide, slightly longer and wider than frontal, roughly pentagonal, their anterior margin deeply indented and in broad contact with frontal, externally contacting first and second supraoculars, the postocular, the sixth supralabial, a large and long temporal and the dorsals; internally it contacts frontal and interparietal. Posterior borders of interparietal and parietals and dorsals coincide with a slight transverse cervical constriction in the occipital region. Two supraoculars, first largest, about three times longer than wide, contacting frontal, frontonasal, nasal, loreal, first superciliary, second supraocular and parietal. Second supraocular smaller, above second superciliary, longer than wide, separated from frontal by the slight contact between parietal and first supraocular. Two superciliaries, the first longer, its sutures coincide with that between supraoculars. Nasal large, longer than high, their suture with posterior part of first supralabial complete, anteriorly to nostril fused to first supralabial forming an incomplete nasolabial. Nasal portion of this incomplete nasolabial clearly viewed from above; above first and second supralabials. Nostril in the first third of lower margin of nasal, deeply indenting the suture with the posterior part of first supralabial. Loreal roughly squared, in contact with nasal, first supraocular, first superciliary, preocular, subocular and second and third supralabials. Frenocular absent. Five supralabials, third and fourth under the orbital region, fifth the highest and largest, contacting punctually parietal. One long subocular, widest posteriorly. Eyelid present with an undivided semitransparent disc. A large and elongate postocular between fourth and fifth supralabials and parietal. An enlarged, longer than wide, temporal scale between parietal and fifth labial, in broad contact with parietal. Ear opening absent. All head scales smooth and juxtaposed with scattered sensorial organs.

Mental roughly trapezoidal, wider than long, slightly longer than the ventral surface of rostral. Postmental heptagonal, as wide as long. Two pairs of chin shields, both contacting infralabials; the anterior pair smaller, in broad contact at midline; second pair broadly separated by an enlarged pair of symmetric flat and diagonally disposed elongate pregulars. Five infralabials, first the smallest, second, third and fourth about the same size. Gulars smooth, imbricate, rounded posteriorly, in eight transversal rows; scales of gular rows increasing gradually in size toward interbrachial region. Interbrachial region with four scales, the central ones largest, twice longer than wide. Lateral scales of neck subrectangular, smooth, imbricate, slightly rounded posteriorly and longer than wide, disposed in regular transverse rows and becoming gradually similar to adjacent dorsal or ventral scales. Collar fold absent.

Dorsal scales imbricate and disposed in regular transversal rows; smooth, subrectangular and wider in occipital region, becoming progressively narrower, more elongate and rounded towards the level of the forelimbs and then on longer, hexagonal, lanceolate, strongly keeled, with lateral sides almost juxtaposed. Fortyfive tranverse rows between interparietal and the level of hind limbs. Lateral scales about the same size as dorsals but smooth and less acuminate; those closer to ventrals slightly wider. A distinctive area with granular scales surrounds the area of arm insertion and the posterior part of leg insertion. Thirty scales around midbody. Ventral scales smooth, longitudinally imbricate, laterally juxtaposed, almost squared just after the interbrachial row, becoming gradually longer than wide, rounded posteriorly, those after midbody narrower; 34 transverse rows from interbrachials (excluded) to preanals. Five scales in the posterior part of preanal plate, central one largest. One preanal and one femoral pore on each side.

Scales of tail similar to midbody dorsals, keeled, lanceolate, strongly imbricate.

Fore limbs rudimentary, stiliform, covered by smooth and imbricate scales, ending in a single apical scale.

Fore limb length equal to width of one row of lateral scales. Hind limb also stiliform but larger than fore limb, covered by smooth, large and imbricate scales ending by one apical scale; length equivalent to one and onehalf scale rows. One femoral pore present at each side.

Background dorsal and lateral surfaces of body and tail almost uniformly cream to light brown with scattered melanophores. Ventral parts of body and tail immaculate cream.

Hemipenes partially everted, bilobed, sulcus spermaticus central in position, extending from the base of organ to lobes, sulcate and assulcate faces smooth and without papillae or spines.

Measurements of the holotype: Snout-vent-length: 65 mm; tail length (regenerated): 85 mm.

Variation: the type series is fairly homogeneous, showing little variation in number of dorsal scales (42– 45), ventral scales (34–36), scales around body (29–30), and gular scale rows (6–8). Like the holotype all other specimens have the first supralabial partially fused with nasal, two superciliaries, five supralabials, five infralabials, five preanal scales, one preanal pore, and both fore and hind limb ending by one apical scale. All males have one femoral pore at each side; these are absent in females. In two specimens (MZUSP 98080 and 98081) the first supraocular is fused to the loreal on the left side. There is also some variation regarding contact between labials and parietal. Like in the holotype, in general the fifth supralabial punctually contacts the parietal except for some cases when there is a contact between postocular and temporal. This is the case in MZUSP 98081, on the right side of the holotype, where there is a slight contact between postocular and temporal, and in MZUSP 98082 and 98083 where this contact is larger. Maximum snout-vent length varies between 61–80 mm, the largest specimen is a female. Although the holotype is almost homogeneously immaculate cream, other specimens show in general two symmetrical dorsolateral light stripes two to two and one-half scales wide, and extending from the parietal scale to the end of tail, showing a darker overall color pattern. Between them there is a conspicuously wider but irregular brown dorsal stripe also extending from the posterior border of the interparietal scale to the tip of the tail. Longitudinal dark and light stripes are more conspicuous anteriorly where the light stripes are limited below and above by a darker border. Lateral parts of body are cream with a scattered light brown reticulum.

Hemipenes of one of the paratypes (MZUSP 98084) were partially everted during preservation and the left hemipenis was prepared ( Fig. 2 View FIGURE 2 B). The organ is relatively small, extending along approximately four subcaudal rows (4 mm). It is slightly bilobed, with the distal tip of the retractor muscle divided. The hemipenial body is globose, slightly Y-shaped, ending in two small and symmetric lobes surrounded by a thin and fleshy fold. The sulcus spermaticus is broad, central in position, originates at the base of the organ, and proceed in a straight line towards the lobes. At the distal tip of the hemipenial body the sulcus is divided in two branches. In the lobes, branches of the sulcus run among folds in an S-shape condition, ending at their tip. Sulcate and asulcate faces of the organ are completely smooth, with no evident plicae, papillae, ridges, calyces, mineralized spines or spinules, even after immersion on an Alizarin Red solution for 24 hours. The absence of hemipenial ornamentation is similar to the condition reported for Bachia trisanale and B. intermedia ( Presch 1978) .

Distribution and natural history: The study area is situated near the borders of states of Tocantins, Maranhão, Piauí, and Bahia and is usually referred to as the Jalapão region. The Jalapão, comprising approximately 53300 km 2, consists of extensive quartzitic sand depressions resulting from erosion of arenitic plateaus of the Serra Geral and Chapada das Mangabeiras, and is drained by headwaters of the Tocantins and São Francisco river basins ( SEPLAN 2003). Isolated and highly eroded sandstone relicts are frequent in the depositional plains, and represent evidence of the formerly continuous plateau (Ribeiro et al. in press). The Jalapão is considered one of the three best preserved areas in the Cerrado domain, Central Brazil ( Cavalcanti & Joly 2002). Estação Ecológica da Serra Geral do Tocantins (EESGT) is a conservation unit created in 2001 that protects 7163.06 Km2 of the depositional plains and part of the Serra Geral, being the second largest federal protected area in the Cerrado. Vegetation in Jalapão and EESGT is typical of the Cerrado domain, and is characterized by extensive grasslands, open savannas and typical “cerrado” savannas (campo sujo, campo cerrado and cerrado sensu stricto, respectively, Oliveira-Filho & Ratter 2002) ( Figs. 3 View FIGURE 3 A, 3B), interspersed by palm marshes (dominated by Mauritia flexuosa palms) and gallery forests along water courses.

Two main regions were sampled at EESGT: a southern, lower portion, located mostly at Almas municipality, state of Tocantins, and a northeastern portion, located at Formosa do Rio Preto municipality, in state of Bahia, in the higher portions of EESGT on the Serra Geral plateau. Nine pitfall-trap lines, containing five sets of traps each, were installed in each region. Each set of traps consisted in four 35 liter buckets arranged in Y shape and connected by drift fences. A total of 180 buckets were used in each site, which remained opened for nine consecutive days at the southern (27/January to 04/Febuary/2008) and at the northeastern sites (08–16/ Febuary/2008), totaling an effort equivalent to 3240 buckets/day. The traps covered the main physiognomies present in EESGT region, ranging from grasslands, with few scattered small trees, to typical Cerrado, with discontinuous herbaceous layer and leaf litter patches. Palm marshes and gallery forests were also sampled. In addition, active search was performed around the lines and along the roads that cross the EESGT. Although we saw tracks of Bachia oxyrhina ( Fig. 3 View FIGURE 3 C) along the different sampled habitats at the southern part of EESGT, except in palm marshes and gallery forests, only four specimens were collected in pitfall traps, all at lower elevations in the south-central part of the station ( Fig. 4 View FIGURE 4 ). Two additional individuals were collected manually by digging through the sand surface, one in the southern site, and another in the central part of the station, adjacent to “Morro do Fumo”, one of the relictual and highly eroded sandstone plateaus of the area. Although sandy soil grasslands and denser savanna habitats, similar to those of the southern site were sampled with similar effort in the northeastern part of the station, no specimens or tracks were found. The absence of Bachia oxyrhina at the northeastern site and characterized by higher elevations, suggests that the species is restricted to the southern part of the Jalapão. This is reinforced by the absence of the species in all other previous herpetological surveys conducted at the northern part of Jalapão region ( Vitt et al. 2002; Vitt et al. 2005; Mesquita et al. 2006; Nogueira 2006). The single previous lizard sampling within EESGT was conducted in March–April 2003 by one of us (CN) as part of a study on Cerrado lizard diversity ( Nogueira 2006; Nogueira et al. in press), in the northernmost limits of the protected area, all north of the Morro do Fumo site, in depressions associated with the Rio Novo drainage, one of the main watercourses in the Jalapão. Using the same overall sampling method (35 liter bucket traps with drift fences, in different habitat types), 490 lizard specimens of 14 species were collected (see results in Nogueira et al. in press), after 3600 bucket/days of pitfall trapping and manual collection in fifteen days of fieldwork. In this previous sampling, no specimens of Bachia were obtained. Furthermore, no tracks were observed at these sites within the Rio Novo depressions, suggesting that the new species is rare or absent outside the southern portion of EESGT, which was previously unsampled.

Bachia oxyrhina seems to inhabit the superficial layer of sandy soils, as many tracks can be easily found at exposed sandy soils in the station. Frequently, tracks were observed on open ground heading from one bush to another, indicating sub-superficial ground locomotion. During the survey, we measured temperatures hourly using a HOBO data logger, with a sensor placed 1 cm deep in the exposed sand, next to one of the lines where two specimens of Bachia oxyrhina were captured in pitfall traps. Mean temperature was 26.32ºC, ranging from 21.33ºC to 42.46ºC (sd = 4.83ºC, n = 192). These data represent a gross estimate of the temperatures that the species may encounter in the sand.

One Scotinus sp. larva (Tenebrionoidea, Coleoptera ) was regurgitated by the holotype during fixation.

The fact that only four individuals were obtained with pitfall-traps, despite the relatively large effort, is not surprising given that individuals of the genus Bachia are usually not abundant in herpetological surveys ( Colli et al. 1998; Nogueira 2006; Pavan 2007; Rodrigues et al. 2007). However, as many tracks were seen through great part of the station it is difficult to say if this species is rare. Considering that it was raining during part of the time that the pitfalls remained open, we prefer do not use our pitfall sampling results to infer the real density of this species.