Canis familiaris var. papuensis Ramsay, 1879b

Parnaby, Harry E., Ingleby, Sandy & Divljan, Anja, 2017, Type Specimens of Non-fossil Mammals in the Australian Museum, Sydney, Records of the Australian Museum 69 (5), pp. 277-420 : 345-347

Canis familiaris var. papuensis Ramsay, 1879b

Proc. Linn. Soc. N.S.W. (ser.1) 3(3): 242. (?Jan–? April 1879).

Common name.?Papuan Village Dog.

Current name. Canis familiaris Linnaeus, 1758 ; following rationale of Jackson & Groves (2015: 288) and Jackson et al. (2017); also see Dwyer & Minnegal (2016) for a review of the relationship between New Guinean village and feral dog populations.

Holotype. A.3652, designated here. Skull ( Fig. 28 ), sex not recorded, purchased from Andrew Goldie, registered December 1878. Collection date and collector not given in the A Register. No specific locality is entered against this specimen, but an earlier register entry indicates that it was part of a large collection from the “Port Moresby district” purchased from Andrew Goldie. In his original account, Ramsay states that the specimen was obtained on Andrew Goldie’s second expedition to New Guinea. No associated skeletal elements or skin have yet been located in the collection but it is possible that the holotype was originally a mounted skeleton, as “1 dog, from New Guinea” is listed amongst mammal specimens articulated during 1881 in the AM Annual report ( Ramsay, 1882c). Alternatively, the mounted specimen may have been a specimen acquired from another source, possibly from Miklouho-Maclay. The skull has the registration number written in what is likely to be Troughton’s handwriting.

Condition. Cranium missing upper right 1st and 2nd premolars (possibly lost while the animal was alive), missing medial side of right auditory bulla, anterior tips of nasal bones eroded, hole in occipital. Left dentary has two holes in ramus, right dentary complete.

Cranial measurements (mm). A.3652: GL, 165.25; ConL,154.36; BasL, 147.39; NasL, 50.78*; NasB, 17.22; UC1–C1 (alv.), 30.25; APV, 10.58; PAL, 82.42; UPM (alv.), 16.73; UMR (alv.), 16.32; ZB, 88.73; POC, 24.86; BUL, 19.95; MB, 58.74; DL (condyl.), 123.28; LPM (alv.), 10.14; LMR (alv.), 29.92. [* = estimate, nasal tips broken].

Type locality. Port Moresby district ( Ramsay, 1879b), Central Province, Papua New Guinea.

Comments. Volume 3 issue 3 of the Proc. Linn. Soc. N.S.W. is the only issue for which Fletcher (1896) could not determine the publication date; it was probably between January–May 1879, based on publication dates of conjoint issues.

Previous assessments of the status of this taxon have presumably been made without examination of Ramsay’s original material that has remained unreported in the AM Collection since its description. Troughton (1971) reported that Ramsay’s specimen had been destroyed but we are confident that the dog skull labelled A.3652 is the specimen reported by Ramsay. Although he did not specify the collection locality, sex, or provide any measurements of the specimen, Ramsay stated that it was obtained from Goldie and his wording implies a single specimen. Only one “ Canis papuensis ” is listed in the AM specimen acquisitions for 1878, purchased from Goldie and listed, remarkably, under “Collection of fish specimens from New Guinea (in spirits)” ( Ramsay, 1879c).

Specimen A.3652 was registered in December 1878 as part of a large consignment of specimens in spirit purchased from Goldie; these were entered in the register under the heading “Collection of spirit specimensAndrew Goldie not in the best state” and many were later marked as “destroyed”. Taken along with Ramsay’s comment that the dog specimen was received from Goldie in a poor state of preservation in spirits, these entries may have influenced Troughton (1971) to state that Ramsay’s holotype was no longer in the collection. However, A.3652 was actually marked as being transferred to the Palaeontology Dept. (no date given) where it was found in 2012 during preparation of this paper. Troughton (1957) initially treated papuensis as a nomen nudum but he later equivocated and even commented that the name had been ignored by most authors ( Troughton, 1971).

The ratio zygomatic breadth/condylobasal length (87.8/ 168.5 mm) of 0.52 for A.3652 falls below the range 0.64–0.74 given for Koler-Matznick et al. (2003) for New Guinea Singing Dogs “ C. hallstromi ” and it is also below 0.61–0.62 given by those authors for Papuan village dogs.

Ramsay stated that his knowledge of this taxon, including its small body size and inability to bark, drew from the observations of Morton and Broadbent about Papuan dogs. Although it is not known who collected the holotype during Goldie’s expedition, it is possible that the specimen came from a moderate to high elevation in the Owen Stanley Range behind Port Moresby, given that collecting occurred quite early over a wide range of altitudes ( Moore & Mullins, 2012; Morton, 1885a). In an account of his work with Goldie, Morton (1885b) indicates in relation to “the New Guinea dogs” that “we did not find them in a wild state”, only in villages. However, Broadbent had a separate collecting camp and might have obtained the holotype as a feral animal. It is possible that the holotype was a village dog but that Ramsay’s description reported observations of the singing calls heard by field workers.

Figures

Figure 28. AM A.3652, holotype skull of Canis familiaris var. papuensis Ramsay, 1879b. (Photography by Stuart Humphreys).