28G PROF. HUXLEY ON THE CRANIAL AND [Apr. 6, 



Icticyon and Palceocyon. The heel of this tooth also diminishes in 

 proportion to its blade. 



IV. All the known kinds of canine animals may be arranged in 

 two series, starting from Otocyon as the nearest ally of the lowest 

 member of each series. 



It is probable that when a sufficient number of specimens of each 

 species at present recognized has been examined, it will be found that 

 the forms with least-modified skulls and teeth are connected by in- 

 sensible gradations with those with most-modified skulls and teeth, 

 and that no absolute line of demarcation can be drawn between one 

 species and another in cranial and dental characters. 



There is no evidence that any one of these species is infertile when 

 crossed with any other. Such evidence as exists with respect to 

 G. cancrivoriis, the Jackal, Dog, and Wolf tends to the conclusion 

 that species of approximately the same size are capable of fertile 

 unions. 



V. The taxonomy of the Cauidae, under these circumstances, is very 

 much a matter of convention. I am disposed to think that the most 

 convenient mode of representing the facts is to regard Otocyon and 

 the Thooid and Alopecoid series respectively as genera, retaining for 

 the two latter the old names of Canis and Vulpes. 



In each of these genera a lower, a middle, and a higher section 

 may be conveniently recognized, though they are hardly susceptible 

 of strict definition. 



Thus, in the genus Vulpes, Baird has separated F. littoralis and 

 V. cinereo-argentatus under the name of Urocyon. The Corsachie 

 Foxes, V. zerda, V. caama, V. bengalensis, and V. velox may con- 

 stitute another subgeneric section, and the most differentiated Foxes, 

 such as V. lagopus, V.fulvus, and the rest, a third. 



In the genus Canis we have, in like manner, as a lowest section 

 the species of the C. cancrivorus and C. velulus type (answering pretty 

 much to the Aguarra Dogs of Hamilton Smith), the Sacaline section 

 (C. aureus, C. anthus, C. mesomelas, C. antarcticus, C. lati'ans), 

 and the Lupine section (C. lupus and all its varieties). Icticyon, 

 Lycaon, and Cyan may probably be usefully retained as subgeneric 

 names for the special modifications of the Thooid type which are 

 denoted by them. 



As for species, no one zoologist has ever yet agreed with the esti- 

 mate of another as to what should be considered species and what 

 local varieties among Wolves and Foxes ; and, as there is no criterion 

 by which the question can be decided, it is probable that such agree- 

 ment never will be attained. The suggestion that it may be as well 

 to give up the attempt to define species, and to content oneself with 

 recording the varieties of pelage and stature which accompany a 

 definable type of skeletal and dental structure in the geographical 

 district in which the latter is indigenous, may be regarded as revo- 

 lutionary ; but I am inclined to think that sooner or later we shall 

 have to adopt it. 



VI. The most notable facts in the Distribution of the Cauidae are : — 

 (I) the occurrence of the Corsacine Alopecoids and the Sacaline 



