The identity of Stenorhabdium temporale Werner, 1909 (Serpentes: Colubroidea)

Re-examination of the type material of Ligonirostra Stuhlmanni Pfeffer, 1893 (original spelling, now Prosymna stuhlmanni) and comparison with the sole type specimen of its synonym Stenorhabdium temporale Werner, 1909 revealed a number of significant morphological differences between these taxa. Detailed analyses of pholidosis and osteology of comparative material show that S. temporale is in fact a subjective junior synonym of Pseudorabdion longiceps (Cantor, 1847). A lectotype and a paralectotype of Ligonirostra stuhlmanni are designated and described.


Introduction
In 1909 franz Werner described from a single specimen the new monotypic genus and species Stenorhab dium temporale.According to the original description it is assumed that the holotype was collected in "Ostafrika" [East Africa] and was given by "Stud.SChWarzkoPf" to the collection of the then "Königliches Naturalienkabinett in Stuttgart" [now Staatliches Museum für Naturkunde Stuttgart] where it is catalogued under inventory number SMNS 3204.Since its erection the name Stenorhabdium temporale has only been mentioned sporadically in the literature and can be found in Werner's synopses (1923Werner's synopses ( , 1929)), in encyclopedic catalogues of generic and subgeneric names (e.g.SChulze et al. 1937;naeve 1940) as well as the Zoological Record (Boulenger 1911).
loveridge (1957 p. 265; 1958 p. 127) assumed that the holotype of S. temporale was lost, and synonymized the taxon with Prosymna ambigua stuhlmanni (Pfeffer, 1893) (now Prosymna stuhlmanni, considered as specifically distinct from P. ambigua by Broadley (1992)) based on Werner's original description.Subsequently, this classification was followed by anyone who worked on the genus Prosymna, as well as by authors mentioning the genus name Stenorhabdium or the taxon S. temporale (e.g. loveridge 1958;Broadley 1980;1983;WilliamS & WallaCh 1989;SChlüter & hallermann 1997;Wal-laCh et al. 2014).However, in 1997 SChlüter & hallermann published the type catalogue of the herpetological collection of the Staatliches Museum für Naturkunde Stuttgart, which showed that Werner's holotype of S. tem porale had actually not been lost.Nevertheless, a re-examination of the specimen to confirm the status of S. temporale was never conducted.
During examination of different species of the genus Prosymna gray, 1849 we recently re-examined the holotype of Stenorhabdium temporale Werner, 1909.A direct comparison with one of the syntypes of Ligo nirostra stuhlmanni Pfeffer, 1893 showed that by no means could S. temporale be considered conspecific with L. stuhlmanni, since the holotype of Werner's taxon shows a number of significant morphological differ-ences that preclude its assignment to the genus Prosym na.In this paper, we discuss the present allocation of S. temporale as synonym of P. stuhlmanni and clarify its actual identity.

Material and Methods
This contribution is based on the study of selected preserved specimens of the genera Stenorhabdium, Prosym na and Pseudorabdion (see below), as well as information from the literature.Dorsal scale rows were counted at three points along the trunk, i.e. at one head length posterior to the end of the head, at midbody, and at one head length anterior to the anal scute.Dorsal scale row reduction formulae are based on doWling (1951a).The reduction formulae for the supracaudal scales is used analogous to the dorsal scale row reduction scheme.
Head length is defined as the length from the tip of the snout to the end of the parietals.The midbody scale count was taken at half of the total number of ventral scales.Ventrals were counted according to doWling (1951b).The terminal scale is not included in the subcaudal count.Values for symmetric characters are given as left/right.Sex of specimens were determined by the presence or absence of hemipenes, inspected through a subcaudal incision at the tail base.
Colour descriptions are based on preserved specimens.
The heads of types of Stenorhabdium temporale and Ligonirostra stuhlmanni as well as comparative material of Pseudorabdion were subjected to micro-tomographic analysis at the Museum für Naturkunde Berlin, using a Phoenix nanotom X-ray|s tube at 70 kV and 150 or 200 μA, generating 1000 projections with 750 ms per scan.The different kV-settings depended on the respective specimen size.Effective voxel size, i.e. resolution in three-dimensional space, ranged from 4.2 -10 μm.The cone beam reconstruction was performed using the datos|x-reconstruction software (GE Sensing & Inspection Technologies GMBH phoenix|x-ray datos|x 2.0) and the data were visualised in VG Studio Max 2.2.

Results
Between  1897; anonymouS 1901, 1906).Today it is impossible to recover the original collector's data and inventory number in order to check if the specimen that became the holotype of S. temporale was already accessioned into the Stuttgart collection before it was sent to Werner, since the original catalogue of the herpetology collection got lost in September 1944 during World War II (SChlüter & hallermann 1997;SChlüter 2001).However, there is also a chance that the specimen was never part of the Stuttgart collection and was misplaced while it was in the hands of Werner who then erroneously assigned it to the material he had borrowed from the Stuttgart Museum.
loveridge (1957 p. 265) synonymized without further comments Stenorhabdium temporale with Prosym na ambigua stuhlmanni, but explained his decision one year later: "I suggest that its [Stenorhabdium] condition masked its true appearance and resulted in some erroneous interpretations of what may have been a slightly aberrant stuhlmanni.Twenty-five years ago when I wished to see the holotype, it could not be found".He further indicated that according to Werner (1909) the type of S. temporale shows some unusual pholidotic characters in comparison to P. a. stuhlmanni, e. g., parietal in contact with labial, internasal and prefrontal divided, but finally established: "In other respects his [Werner's] holotype, now lost I believe, conforms to a male a. stuhlmanni in its lepidosis and scale counts" (loveridge 1958 p. 128, p. 163).
After World War II, the curators of the amphibian and reptile collection in Stuttgart heinz Wermuth (since 1962) and andreaS SChlüter (since 1984) were devoted to the challenging task of reconstructing lost collection data.One result of their effort is the catalogue of type material of the herpetological collection in Stuttgart (SChlüter & hallermann 1997), which clarifies the status of the thought to be lost (according to loveridge (1957; 1958) holotype of S. temporale and confirms its conservation.
Due to the presumed synonymy of Stenorhabdium temporale with Prosymna stuhlmanni we attempted to re-determine the syntypes of Ligonirostra stuhlmanni for comparison.Broadley (1980) erroneously presumed that the type material of L. stuhlmanni was destroyed during World War II, and WallaCh et al. (2014 p. 582) stated that the description of L. stuhlmanni is based on three specimens.However, the type series actually comprises two specimens only, as unmistakably declared by Pfeffer's (1893 p. 79) indication of "Zwei Stücke" [two pieces].WallaCh et al. (l.c.) incorrectly interpret Pfeffer's "No.521 -522" as inventory numbers for the herpetological collection of the Zoologisches Museum Hamburg, and added one of the correct inventory numbers (ZMH R07910) to 521 and 522, which undoubtedly were the field numbers as assigned by the collector franz Stuhlmann.
Our current inquiry revealed that in fact only this one (ZMH R07910, old No. 1646), of the two syntypes, still exists in the collection of the Zoologisches Museum Hamburg, the second type could not be found (pers. com. J. hallermann, Feb. 2015).The original inventory number of the second syntype is unknown due to the fact that all original correspondence, catalogues, and type lists of the herpetological collection were destroyed during World War II (hallermann 2006).During an inspection of the Prosymna material in the collection of the Museum für Naturkunde Berlin we found a specimen (ZMB 11008) determined as Prosymna ambigua and collected at "Usambáa" by f.Stuhlmann.Its re-examination, a comparison with Pfeffer's original description and the associated inventory catalogue entry revealed that it represents the apparently lost second syntype of Ligoniro stra stuhlmanni Pfeffer, 1893.After its examination by georg Pfeffer, and in agreement with the collector f.Stuhlmann, the specimen was sent from the Museum Hamburg along with 28 other East-African amphibian and reptile species from Stuhlmann's collection, including types of Phrynopsis boulengeri and Megalixalus stuhlmanni described by Pfeffer (1893), to the Zoological Museum Berlin in 1893 (Zool.Mus.Sign.S II "Museum Hamburg"; Sign.S II "Stuhlmann, F.").
A comparison of both syntypes of L. stuhlmanni with the specimen depicted in Pfeffer (1893) revealed unambiguous concordance with the syntype from Hamburg.In order to fix the status, we herein designate in accordance with Articles 74.1.and 74.7. of the Code (ICZN, 1999), specimen ZMH R07910 as lectotype of Ligonirostra stuhlmanni described by Pfeffer, 1893, depicted on plate 1, figure 8 -10 of the original description.Sex and pholidosis -adult male; total length 218 mm [215 mm]; tail length 29 mm [25 mm]; head length 6.1 mm; ratio of tail length/total length 0.133; head not distinct from body and pointed; original body and tail shape difficult to determine because of the shrivelled condition; dorsal and supracaudal scales smooth and without apical pits, dorsals arranged in 15 scale rows throughout the body, outer dorsal scale row slightly enlarged.

Stenorhabdium temporale
Supracaudal scale reduction formula: 3+4(4) 3+4( 16) 2+3(25)  The tips of all teeth are distinctly recurved posteriorly.Between the anterior 6 teeth there is an interspace of less than the base of the longest tooth (base length of tooth 6: 0.09 mm; interspace: 0.06 -0.08 mm).Posterior to tooth 6, interspaces increase to 0.12 -0.15 mm, being largest between teeth 7 through 9. Lateral to each palatine tooth is a single replacement tooth at different growth stages.The posterior 23% of the palatine bone are without teeth and articulate with the pterygoid.
21/22 pterygoid teeth, decreasing in size posteriorly.The tips of all teeth are distinctly recurved, whereas the angle of curvature decreases in posterior direction.
The interspace between the teeth decreases posteriorly, from being the length of approximately one tooth base (base length of tooth 1: 0.12 mm) to just about half of it.Lateral to each pterygoid tooth there is a single replacement tooth at different growth stages.The posterior 10% of the pterygoid bone are without teeth.14/14 mandibular teeth, slightly increasing in size until tooth 5, and slightly decreasing thereafter.All teeth are recurved with increasing curvature in posterior direction.The anterior 2 teeth are closely spaced, whereas the interspaces between the following teeth increase posteriorly, and are less than the base of the longest tooth (base length of tooth 5: 0.13 mm; longest interspace between teeth: 0.10 mm).Medial to each mandibular tooth is a single replacement tooth in different growth stages.
Colouration -Dorsal and ventral ground colour of body and tail middle brown, head light-brown; supralabials, infralabials and throat cream-white; an oblique cream coloured mark from the border of parietals to angle of mouth; a thin, light beige collar occupying half of the fifth and sixth transversal dorsal scale row, extending laterally onto the first longitudinal dorsal scale row; edges of dorsal and supracaudal scales and posterior edges of ventral and subcaudal scales brightened.Sex and pholidosis -subadult male; total length 196 mm [ca.200 mm]; tail length 30 mm; head length 7.03 mm; ratio of tail length/total length 0.153; head only slightly distinct from body with a rounded tip of snout; body and tail rounded; dorsal scales smooth, arranged in 19/15/15 rows throughout the body, with single apical pits, outer dorsal scale row slightly enlarged; supracaudal scales smooth with paired apical pits.Dorsal scale row reduction formula: 3+4(11) 3+4(28) (10) 19 ----------17 ----------15(132).

Redescription of the lectotype of Ligonirostra stuhlmanni
3+4(11) 3+4( 27) Supracaudal scale row reduction formula: Dentition -9/11 maxillary teeth, the anterior 7/9 precranterian teeth are nearly equal in size (~0.3 mm) but are followed by 2/2 large cranterian teeth (~1 mm) without diastema.All precranterian teeth are curved posteriorly shortly above the base.Medial to the precranterian teeth 1, 3, 4, 6/2, 4, 5, 7 there is a single replacement tooth at different growth stages, respectively.The cranterian teeth are flattened, posteriorly not grooved, but edged, resulting in a blade-like appearance.The tips of these 2 teeth are distinctly bend posteriorly, similar to the precranterian teeth.One to 2 replacement teeth per cranterian tooth are found posteromedially to each tooth, showing different growth stages.No significant interspace exists between the different maxillary teeth.The anterior 6.3% and posterior 11.8% of the maxilla are without teeth.5/6 palatine teeth, nearly equal in size.All are curved posteriorly shortly above the base.Lateral to teeth 2, 4/2, 4, 6 there is a single replacement tooth at different growth stages.No significant interspace exists between the different palatine teeth.The anterior 5.5% and posterior 33.3% of the palatine bone are without teeth.
The pterygoid bone is toothless.9/9 mandibular teeth, nearly equal in size.All are slightly curved posteriorly shortly above the base.The anterior 3 teeth are spaced close together.Between the following posterior teeth the respective interspace increases until it is as long as the length of one tooth base.Medial to the mandibular teeth 1, 3, 7/2, 4, 6, 8 there is a single replacement tooth at different growth stages, respectively.The anterior 13.9% and posterior 7% of the mandibula are without teeth.
Colouration -The specimen is more or less bleached out to light brown colour dorsally, particular the forebody is bleached out to whitish-cream; head and ventrum uniformly whitish coloured.

Comparison
After comparing the type material of the two taxa, regarded as conspecific until now, it becomes clear that they obviously belong to different genera and cannot be considered synonyms any longer.In addition, the collection locality was erroneously assigned, which is why we consulted character matrices and determination keys of several authors to resolve the true identity of the type specimen of S. tem porale (Boulenger 1893(Boulenger , 1894(Boulenger , 1896;;Chan-ard et al. 2015;Cogger 1994;david & vogel 1996;rooiJ 1917;ernSt & ernSt 2003;leviton et al. 1992;manthey & groSSmann 1997;marx & raBB 1972;mCCranie 2011;meirte 1992;o'Shea 1996;PeterS & oreJaS-miranda 1986;Smith 1943;Smith & taylor 1945;SzCzerBak 2003;taylor 1922;zhao & adler 1993).Because of several characteristic pholidotic traits such as missing loreal and anterior temporal scales, prefrontals in contact with supralabials, 15 DSR without apical pits throughout the body as well as an aglyph dentition and lack of enlarged posterior teeth the specimen could exclusively be assigned to the genus Pseudorabdion Jan, 1862.Within the genus the traits match the best with those listed by various authors for Pseudorabdion longiceps (e.g.We examined and compared the skull morphology of specimens of Pseudorabdion longiceps from Malacca, Sumatra, Java, Kalimantan and Sarawak.With regard to the following characters we found most similarities between the holotype of S. temporale and specimens from Java (ZMB 13111; ZMB 13024 see Figs. 7 III a -b; 8 III a -e), i.e. (1) nasals slightly longer than frontals vs. nasals shorter than frontals in ZMB 4966 from Kalimantan and in ZMB 7091 from Sarawak; (2) posterior border of frontals has a wide U-shape vs. posterior border of frontals has a soft W-shape in ZMB 31404 from Sumatra and ZMB 4966 from Kalimantan or a wide M-shape in ZMB 7091 from Sarawak; (3) suture between exoccipitals longer than length of the center of the supraoccipital vs. suture between exoccipitals shorter in ZMB 3026 from Malacca, ZMB 31404 from Sumatra, ZMB 4966 from Kalimantan and in ZMB 7091 from Sarawak; (4) higher number of mandibular teeth (13 or 14) vs. lower number, i.e. 10/10 in ZMB 31404 from Sumatra, 11/11 in ZMB 4966 from Kalimantan and 10/9 in ZMB 7091 from Sarawak; (5) largest interspace between teeth at the maxilla equal or slightly longer than socket of the longest tooth (equal -16% longer) vs. largest interspace distinctly longer (24% -47% longer) in ZMB 3026 from Malacca, ZMB 4966 from Kalimantan and in ZMB 7091 from Sarawak; (6) toothless posterior part of the maxilla less than 24% vs. toothless posterior part more than 25% in ZMB 31404 from Sumatra, ZMB 4966 from Kalimantan and in ZMB 7091 from Sarawak; and (7) medial limb of ectopterygoid almost half the length of lateral limb vs. medial limb distinctly longer in ZMB 3026 from Malacca and ZMB 4966 from Kalimantan.
Based on the morphological evidence at hand, we therefore remove Stenorhabdium temporale Werner, 1909 from the synonymy of Prosymna stuhlmanni (Pfeffer, 1893) and regard it as a subjective junior synonym of Pseudorabdion longiceps (Cantor, 1847); the generic name Stenorhabdium Werner, 1909 becomes a synonym of Pseudorabdion Jan, 1862.

Discussion
The bad condition of the holotype of S. temporale probably explains Werner's misjudgments particularly regarding characteristics of the head scalation, which in combination with the incorrect allocation of the collection locality misleadingly resulted in the description of a new genus and species.That loveridge (1957) did not recognize the differences of the specimens and synonymized S. temporale with P. stuhlmanni is certainly due to the fact that the holotype was not accessible to him for reexamination and the collecting locality was not doubted.Nevertheless, it is surprising that loveridge (1957,1958) does not comment on the maxillary dentition, which Werner (1909) uses and defines in the genus diagnosis for Stenorhabdium.At least here doubts could have been raised concerning the assignment to Prosymna stuhl manni.Werner (1909 p. 59) describes the maxillary dentition as follows: "Oberkiefer lang, mit 7 soliden Zähnen, die mittleren am längsten, doch nicht wesentlich länger als die übrigen" [Upper jaw long, with 7 solid teeth, the middle ones being the tallest, but not substantial longer than the remaining ones].In contrast, P. stuhlmanni possesses toothless anterior parts of the maxilla and 8 -10, rarely 11 maxillary teeth in total (this study and Broadley 1980).The anterior teeth increase in size evenly but slightly, and are followed without diastema by two distinctly enlarged and curved, at the basis broadened, laterally flattened teeth.They are posteriorly not grooved, but edged, resulting in a blade-like appearance.Some species of Prosymna are known to feed almost exclusively on reptile eggs (Serpentes and Gekkonidae) (Broadley 1979) just like species of the Asian genus Oligodon whose maxillary dentition in a similar way is character-

Dentition
-10/12 maxillary teeth, increasing in size posteriorly until position 8, from position 10 onwards slightly decreasing in size [7 maxillary teeth, the central ones being the tallest].The tips of all teeth are distinctly recurved posteriorly (fide Wright et al. 1979, fig. 1 A).The anterior 3 teeth are spaced closely together, the interspace between them is slightly increasing posteriorly until it is longer than the base of the longest tooth (base length of tooth 8 and 9: 0.16 mm; interspace between tooth 11 and 12: 0.19 mm).Medial to each maxillary tooth is a single replacement tooth at different growth stages.The posterior 22% of the maxilla are without teeth.10/10palatine teeth, slightly increasing in size posteriorly until position 6, the following slightly decreasing.