THE ANTARCTIC WOLF, 383 



But in the summer of 1912, I received for review from the Editor 

 of ' Nature ' a copy of Dr. R. F. Scharfi's volume, ' Distribution 

 and Origin of Life in America,' 1911; and when I found it 

 definitely stated tlierein that C. antarcticus is closely related to 

 C. latraiis, and when I saw the obvious diificulties in which 

 Dr. Scharff was involved in his attempt to explain, on geo- 

 graphical grounds, this singular affiliation, I ventui'ed to reassure 

 him by remarking, in efiect, that his belief was devoid of morpho- 

 logical foundation. 



Now, an author who compiles a volume on zoology of the size 

 and scope of the ' Distribution and Origin of Life in America ' 

 cannot be expected to verify all the statements of earlier and con- 

 temporary writers. Nor in the present instance could Dr. Scharff 

 be justly criticised for not travelling to London to examine for 

 himself the preserved material of 0. antarcticus, of which, I take 

 it, there is no specimen in Dublin. Yery naturally, therefore, 

 he trusted to the verdict of others, and promptly replied to my 

 remark with a request for my reasons for making it. But since 

 1 could not ask the Editor of ' Nature ' to give me the necessary 

 space for justifying the statement I had made, I pledged myself 

 to do this elsewhere, and the matter that follows is an attempt 

 to redeem that promise. 



The acknowledged source of Dr. ScharfF's opinion about the 

 mutual affinities of C. aiitarcticus and C. latrans was the following 

 passage in Mr. Lydekker's ' Geographical History of Mammals,' 

 1896 : — "Of the two indigenous mammals, the most remarkable 

 is the Falkland Island Wolf (Canis antarcticus), which differs 

 markedly from all the Canidte of the mainland and is apparently 

 closely allied to the North American Coyote (0. latrans) " (p. 140). 

 I therefore wrote and asked Mr. Lydekker if he would kindly 

 tell me his reasons for this conclusion, and he informed me that 

 he took it from Prof. Huxley's classic paper upon the cranial and 

 dental characters of the Oanidse, published in the ' Proceedings ' of 

 this Society, 1880, pp. 238-288. Upon looking up this paper 

 I find the following passages referring to the two species under 

 discussion and bearing upon the question at issue : — 



(1) But sometimes there is a well-defined though com- 

 paratively narr.ow sagittal area, from the centre of which 

 a low sagittal crest rises. This is well seen in some 

 Jackals, and especially in G. antarcticus (p. 250). 



(2) In the large size of the upper molars C antarcticus 



presents the closest approximation to some specimens of 

 C. latrans (p. 266). 



(3) From the range of variation of C. cancrivorus it can hardly 



be doubted that the examination of moi-e extensive 

 materials will prove the existence of an uninterrupted 

 series of gradations from C. vetulus to C. antarcticus and 

 C.jubatus (p. 266). 



(4) Seven ci-ania of C. latrans, when measured, exhibit a con- 



siderable range of variation, though probaljly less than 

 a larger series would show. But, as thej are, I must 

 confess myself unable to find an important break in the 



