Mammals from N.H. New Guinea. 671 
Skull quite as in keysseri. 
Dimensions of the type (measured on skin) :— 
Head and body 700 mm.; tail 420; hind foot (wet) 136 ; 
ear (wet) 54, 
Skull: greatest length 110 ; condylo-basal length 108 ; 
zygomatic breadth 58; nasals 41 x 17:5; intertemporal breadth 
12°6 ; palatal length 66 ; length of p* 7:2; combined length 
Gtons = * N83). 
Hab. as above. 
Type. Adult male. Original number 8 (Keysser number 
24). Collected August 1914. 
This Kangaroo, as is shown by its skull, is evidently nearly 
allied to AZ. keysseri, of which it would appear to be a high- 
altitude race, distinguished by its long woolly fur and some- 
what different colour. 
In making this comparison I have had, by Lord Rothschild’s 
kindness, the advantage of examining the type, an old male, 
of M. keyssert, which was collected by Mr. Keysser in the 
Bulung region, inland of the Huon Gulf, at an altitude of about 
1800-2000 m. 
* Tn deseribing this and other Marsupials of the present collection, my 
attention has again been drawn to the inconvenience systematic workers 
suffer from the present absence of a common nomenclature of the teeth. 
This absence is largely due to my own desertion of the ancient Marsupial 
formula of P.8, M.4, on account ofits being possibly erroneous. For the 
correct formula was by some authors thought to be P. 4, M.38, asin other 
mammals, the seven postcanine teeth being then serially and individually 
homologous with each other inthe two groups. This latter was the view 
taken in a paper on the nomenclature of the teeth published in 1905 f, and 
since that date I have not ventured definitely to assign any Marsupial 
cheek-tooth to its serial place, and in giving descriptions and measure- 
ments I have used words, such as ‘ molariform tooth,” correct on either 
theory. 
Nee however, on reviewing the whole subject, it seems to me that it 
would be better to revert to the old notation, that used in the ‘ Catalogue 
of Marsupials,’ until such time as more definite proof is brought forward 
of the incorrectness of this notation. 
The four premolars, with the last changing, of the Mesozoic Triconodon, 
not to mention the four present in the abnormal Phascogale on which I 
largely based my 1887 { paper, seem to me to speak very strongly in 
favour of the old view, even if some arguments may be found against it. 
I therefore now propose, in systematic descriptions, to revert to the 
Catalogue notation, with the secator reckoned as p*, and the “ three ante- 
rior molariform teeth” called, as in that work, m'-m’. 
The paper of 1892 § would, therefore, again fairly represent the views 
I now hold on the various theories which have been put forward in 
regard to the subject of Marsupial tooth-homologies, 
+ P. Biol. Soc. Wash. xviii. p. 194 (1905). 
¢ Phil. Trans. 1887, p. 443. 
§ Ann. & Mag. N. H., April 1892, p. 308. 
