.OF THE KATEL AND THE WOLVERENE. 181 



The claim of Gido to rank as a subfamily may be set aside for 

 the present, but it appears to me that Miller's guess at the 

 affinities of the genus with Mellivora, expressed in the following 

 passage, is very wide of the mark. He wrote {op. cit. p. 433) : — 

 " Tlie subfamily Guloninse, consisting of the genus Gulo alone, 

 is well characterised by its peculiarities of skull, teeth, and 

 external form. Though usually regarded as a near relative of 

 the Mustelinae, the genus more probably finds its true affinities 

 in the African genus llellivora." This means that in Miller's 

 opinion Gido is more nearly allied to Mellivora than it is to 

 Mustela or Martes. The main purpose of this paper is to refute 

 that idea. 



External form. 



Apart from being heavily and powerfully built and provided 

 Avith comparatively short tails, the two genera are quite unlike 

 in shape. Mellivora essentially resembles Meles or Taocklea, being- 

 low on the legs, plantigrade, broad across the back and flat 

 along the spine, the body being rather depressed than com- 

 pressed. The form is that of a terrestrial fossorial beast, without 

 power to leap and with running capacity reduced almost to the 

 Carnivore minimum. The hair is everywhere short, sleek, and 

 coarse. 



Gulo stands comparatively high on tlie legs, is digitigrade, 

 comparatively narrow across the back, and arched along the 

 spine, the body being compressed rather than depressed. The 

 form is that of a terrestrial and arboreal beast, capable of running 

 at some speed and of leaping to a moderate extent. The hair is 

 eveiywhere, except on the face, soft, furry, and mixed with wool. 



The Head. 



The top of the head in Mellivora is high and arched ; in Gulo 

 it is low and flat. (Text-fig. 14, A, D.) 



The tufts of facial vibrissce are alike in the two genera 

 in the sense that they are the same in number and position 

 as in typical predatory Carnivores like the Canidas, "Viverridse, 

 Mungotidae, and most Mustelidse. (Text-fig. 14, A, D.) 



The rhmarium of Mellivora recalls that of Meles in having a 

 well-defined area encii-cling the nostrils below and laterally, 

 although this area is considerably shallower than in that genus. 

 Also it is not continued inferiorly as a philtriim dividing the 

 upper lip. In Gulo the rhinarium is like that of Cants, being 

 continued inferiorly as a distinct philtrum dividing the upper 

 lip, and the nostrils are encircled laterally by an area of naked 

 moist skin continuous with the philtrum in front. (Text-fig. 14, 

 B,E.) 



The external ear in Mellivora has been desci-ibed as absent 

 The truth is that there is no definite laminate pinna standing 

 away from the head, the cavities of the ear being merely 



