OF THE RATEL AND THE WOLVEKENE. 187 



The vtdva. — I have no notes on the female generative organs 

 of Mellivora ; bnt in Gulo the vulva is a naked piriform pro- 

 luinence a little below the anns, from which it is separated by a 

 band of hair, and the orifice is a verticallv elongated slit. (Text- 

 fig. 18, A.) 



Conclusion. 



From the characters above described it is evident that theie 

 is no particular resemblance in any respect between Mellirora 

 and Gulo. The differences, on the contrary, are profound. 

 Miller's suggestion, therefore, of kinship between the two geneia 

 must be dismissed, and it appeai-s to me that the evidence on 

 this head supplied by the skulls confirms that of the external 

 characters here discussed. 



To what genera of Mustelidfe, then, are Gulo and Mellivora 

 related? Gtdo, in my opinion, might be described broadly but 

 with much truth as a gigantic heavily-built Marten (Martes). 

 I can find nothing in the structure of the skull and teeth opposed 

 to the view that these two genera are related, and tolerably 

 closely related *. There are also no differences of moment 

 between them in the structure of the ear, of the rhinarium, of 

 the upper lip, of the anus, and of the external genitalia ; and the 

 feet of Gtclo are little more than broad, short editions of those of 

 Mealies f, the claws, pads, disposition of the digits, hairiness of 

 the soles, etc., being strikingly alike in the two animals. Finally, 

 it does not appear to me that Gulo differs much more from 

 Martes than Martes difiers from Mustela or Vormela, the three 

 genera which constitute the subfamily Mustelinae as defined in 

 Miller's volume. If this be true, the subfamily Guloninfe can 

 hardly be considered a defensible group, 



Mellivora is much more difficult to classify. The position 

 assigned to the genus by authors will depend upon their views 

 regarding the plasticity of the skull and teeth as compared with 

 the plasticity of the ears, feet, and other external organs described 

 in this paper. I cannot agree with de Winton that the genus is 

 closely related to Ictonyx, and I doubt its near afl&nity with the 

 South American genus Galera ; and although the feet and 

 rhinarium and general form are very like those of Meles, and the 

 pouched anus occurs in both the genera, the structure of the 

 skull and teeth should, I think, exclude Mellivora from a place 

 in the Melinae, despite the heterogeneity of that subfamily as 

 constituted in the current text-books. The best Avay of dealing 

 with the genus at present seems to me to follow Gill in making 

 it the type and sole representative of a special subfamily, the 

 Mellivorinfe. 



* The tip of the haculum is also as3'mnietrical in structure in Martes and 

 Charronia as it is in Mustela. 



t For figures and descriptions of the feet and ears of Martes martes and 

 M.foina, see mj' paper on these two species (Proc. Zool. Soc. 1914, pp. 1062-1068). 



