Stegonotus muelleri Duméril et al., 1854 :682

Kaiser, Christine M., Kaiser, Hinrich & O’Shea, Mark, 2018, The taxonomic history of Indo-Papuan groundsnakes, genus Stegonotus Duméril et al., 1854 (Colubridae), with some taxonomic revisions and the designation of a neotype for S. parvus (Meyer, 1874), Zootaxa 4512 (1), pp. 1-73 : 16-19

Stegonotus muelleri Duméril et al., 1854:682

Taxonomic status. Valid.

Synonyms. Spilotes samarensis Peters, 1861 .

Original name. Stegonotus Mülleri Duméril et al. 1854:682 . The authors listed in their synonymy for the species the name Herpetodryas Mülleri with reference to the “Musée de Leyde,” which indicates that there must have been a specimen label with this name in that museum. The species was named for the German physiologist Johannes Müller (1801–1858), whose Handbuch der Physiologie [Handbook of Physiology] was the physiology textbook of choice for decades and made him one of the most influential physiologists of the 19 th Century. The species description was presented in French.

Holotype. MNHN 848 ( Fig. 7 ; Table 1), an adult female.

Type locality. “Java” [Java Island, Indonesia]. This locality is in error (see below). The specimen was certainly collected in the southern Philippines, where the species is known only from the islands of Dinagat, Leyte, Mindanao, and Samar (University of Kansas collection data, accessed online via VertNet).

Collection. The provenance of this specimen is problematic. In the original description, Duméril et al. (1854) stated that the origin of the specimen was Java (without further elaboration), and recounted that it had arrived in Paris from the RMNH in 1845. At the RMNH, the specimen had arrived some time earlier as a donation from Johannes Müller in Berlin, along with a specimen of a species Duméril et al. (1854) described as Lycodon Mülleri from Java, a species known to exist only in the northern Philippines (see Leviton 1965).

As professor for anatomy and physiology, Müller was a member of the anatomy department at the University of Berlin and received material for study from students, former students, commercial dealers, and via institutional exchanges. Examination of the collection data available at the museums in Paris, Leiden, and Berlin regarding the acquisition specifics of this specimen has, unfortunately, proven unsuccessful. It is therefore not possible to trace its provenance beyond Berlin. It should be noted, however, that with the strong Dutch presence on Java much, if not most, of the shipments from the far reaches of Southeast Asia would at the time have come through the port at Batavia (now Jakarta) on Java. It is therefore likely that the proximal shipping port (Batavia) was given as a shipment’s origin instead of a true collecting locality, leading to the locality error.

Compounding the issue of erroneous type localities for species named muelleri is the division of specimens into jars at the Paris museum. There are currently two tall glass jars ( Fig. 8A,B ) holding four specimens. The taller jar (numbered C684) contains two specimens (MNHN 848, 1320), and the shorter one (numbered C684/1) also contains two specimens (MNHN 5764, 5764A). Jar C684 has an affixed label identifying it as containing Stegonotus mulleri [sic] and including the citation of Duméril et al. (1854) and collection information given as “Java: J. Muller” ( Fig. 8C ). It is in this jar where both the holotypes of S. muelleri (MNHN 848) and L. muelleri (MNHN 1320) are currently housed. Jar C684/1 has an external label identifying the specimens as Lycodon mülleri [sic] and giving the collecting locality as “Luçon. Montano Rey” ( Fig. 8E ) and an internal label identifying the specimens as Stegonotus mulleri [sic] from “Ile Luçon,” providing two additional numbers, 1880-217 and 1880- 218 ( Fig. 8D ).

It is clear from the text of the original descriptions of the two muelleri species that each was written using a single reference specimen. Based on specimen length and body coloration, it is also clear that MNHN 848 is the holotype of Stegonotus muelleri . Using the same characteristics, it is possible to identify MNHN 1320 as the holotype of Lycodon muelleri . The two specimens in Jar C684/1 are clearly not representatives of S. muelleri , as the internal label would suggest, but merely two additional specimens of L. muelleri without type status, probably added to the collection in 1880.

Key characteristics of the holotype. 1002 (1006) mm SVL + 30 (302) mm TL = 1032 (1308) mm TTL. In the original description, the length values were given as “ 1 m 002”, “0 m,030”, and “ 1 m,032”, respectively [1002 mm, 30 mm, and 1032 mm, respectively]. It is apparent that the second value in this list is a typo and should read 0 m 300, which would result in a TTL of 1302 mm. V ♀ = 220 (217), SC ♀ = 100 (97), SCR ♀ = 0.31 (0.31), D = O-17-O (19- 17-15), SL E = 4+5 (4+5), SL = 8 (8), IL = O (9), IL G = O (4).

Key characteristics of the species. We have been able to examine 32 specimens of S. muelleri . Of these, one (♂) was collected on Dinagat, three on Leyte (2 ♀♀, 1 ♂♂), 12 on Mindanao (7 ♀♀, 4 ♂♂, one unsexed), and 16 on Samar (3 ♀♀, 12 ♂♂, one unsexed). The latter include the holotype of Spilotes samarensis (see below). Two additional conspecifics have no specific locality data. Characteristics include V ♀ = 217–234 (227 ± 4.7), V ♂ = 226– 241 (235 ± 3.1); SC ♀ = 86–103 (97 ± 5.3), SC ♂ = 85–108 (97 ± 7.0); SCR ♀ = 0.29, SCR ♂ = 0.30; D = 19-17-15 (100%); SL E = 4+5 (100%); SL = 8 (100%); IL = 10 (100%); IL G = 4 (100%).

Unlike in S. modestus , there does not appear to be any sexual dimorphism with respect to V and SC counts. Values of SCR are within 0.01 of each other, and the female holotype has an SCR value equal to the average male SCR. As a rule, V> 220 and SC> 80, with SL 4+5 touching the eye, an SL count of 8 and an IL count of 10 appear to be diagnostic of this species. This species appears to be the only one in the genus, in which the slightly greater dorsal count right behind the head extends for more than one head length beyond the head. The usual 17-17-15 dorsal count commences in the region between one and two head lengths behind the head.

Comments. In his listing of S. muelleri, Boulenger (1893) reported a total length for the type specimen of 2070 mm, with a tail length of 390 mm. We believe that in the measurement for total length two typos occurred (perhaps due to errors in the transcription of handwriting), such that the number 1070 for total length was replaced by 2070, and the number 300 for tail length by 390. We do not believe that Boulenger’s source, François Mocquard (1834– 1917), at the time of Boulenger’s writing curator of the Paris museum, would have made such egregious measurement errors.

Figures

FIGURE 7. Holotype of Stegonotus muelleri Duméril et al., 1854, MNHN 848, an adult female from Mindanao Island, The Philippines. Shown are the entire specimen (TTL = 1308 mm) in (A) dorsal and (B) ventral views, and its head in (C) dorsal, (D) right lateral, and (E) ventral views.

FIGURE 8. Jars and labels related to Stegonotus muelleri and Lycodon muelleri at the MNHN. (A) A tall glass jar (numbered C684) holding two snake specimens (MNHN 848, 1320). (B) A second tall glass jar (numbered C684/1), also holding two snake specimens (MNHN 5764 and 5764A). (C) Close-up of the label affixed to the jar shown in (A), identifying the specimens as Stegonotus mulleri [sic] and providing the correct citation for S. muelleri with an incorrect locality of Java. (D) A label from inside the jar shown in (B), listing the specimens correctly by their number but in the wrong genus. Both specimens are representatives of Lycodon muelleri. (E) Close-up of the external label from the jar shown in (B), listing the correct genus and locality.