Comparisons.—From P. s. amphichoricus, kermiti differs in: upper parts Tawny instead of Buckthorn Brown; incisive foramen longer and wider; vomerine sheath incomplete; only M3 instead of usually all molars, with four counterfolds. From P. s. liminalis, kermiti differs in: upper parts Tawny instead of Ochraceous-Tawny; aristiforms narrower; M3 only, instead of usually M2 and M3, with four counterfolds.
Specimens examined.—Only the type.
Proechimys longicaudatus (Rengger)
General characters.—Size medium to large; tail short; aristiforms long and narrow; general color on upper parts Ochraceous-Buff to Ochraceous-Orange, finely and uniformly lined with blackish and not forming evident dark band on back; feet dorsally white or gray; underparts of body and inner sides of legs white; skull elongate and slender with moderate ridges; incisive foramen of medium size; vomerine sheath complete or incomplete; bullae large and elongate; upper molariform teeth with three counterfolds; lower molariform teeth with three counterfolds but commonly one or two molars have only two although premolar may have four.
Remarks.—The identity of "Echimys longicaudatus Rengger" can be ascertained only after samples have been collected in the area indicated by Rengger: "unter dem ein und zwansigsten Breitengrade" in Paraguay. Of the samples available to me, those from Urucum, in western Brazil, are geographically nearest the type locality. North of Urucum, both in Brazil and Bolivia, two species of Proechimys live together and one of them is the same species as that at Urucum. Of the two species found to the northward in Brazil and Bolivia, the one that ranges farther south probably will occur at the locality indicated by Rengger. Provisionally, therefore, the name longicaudatus is allocated to the Urucum sample (see [Osgood], 1944:198). In fact, the lack of a type specimen and the general nature of Rengger's description make "Echimys longicaudatus" a nomen vanum. If two species are found living together in the region of northern Paraguay indicated by Rengger it probably will be impossible to be sure to which one his vague description applies.
The form from Urucum, to which the name Proechimys longicaudatus is here applied, is undoubtedly closely related to Proechimys leucomystax [Ribeiro], from Utiarití, on the Rio Papagaio and also to P. roberti and P. boimensis, all from Brazil. P. longicaudatus is used as the name of the species because it is the oldest of the four names.
Figs. 45, 48. Proechimys longicaudatus boimensis, male, MCZ no. 30881, Boim. × 1.
Figs. 41, 42. Proechimys longicaudatus longicaudatus, male, AMNH no. 37085, Urucum. × 1.