814 MR. K. I. POCOGK OX THE 



111 Lyncodon also the pinna is as much reduced as in Conepatus, 

 standing away from the head as a shallow short ridge, the lower 

 end of which does not reach nearly to the aclitus infe7nor. There 

 is no bursa, but the structure of the ridges could not be deter- 

 mined on dried skins. (Text-fig. 30, C.) 



The pinna of Mellivora surpasses even that of Conepatus and 

 Lyncodon in reduction in size. It is represented merely by a 

 thickening of the integument, without any free edge, round the 

 small cavity; and, judging from the descriptions, the pinna of 

 Mydaus is very similar to that of Mellivora, but I am unable to 

 say whether the orifice is exposed as in Mephitis, Spilogcde, and 

 Gonepatxis or concealed by a tragal thickening. 



The following table shows briefly the principal modifications in 

 the external ear : — 



a. Bursa retained. 



h.. Bursa marginal, upper end of its posterior Hap hidden by 



that of anterior tlap Mustela, Gale, Flesiogale, 



Piitorins, Martes, Charronia, Gulo, Melogale. 

 h' . Bursa remote from margin, shallow, its posterior flap in 



no way concealed bj^ anterior flap Tai/ra, Grison, Grisonella. 



«'. Bursa absent. 



c. Auditory orifice concealed, usually by tragal thickening. 

 d. Pinna laminate, standing awa}- from head. 



e. Supratragus a well-developed valvular flap Ictonyx, Foecilictis, 



Fceciloqale. 

 e' . Supratragus at most a ridge with rounded thickening. 



f. Pinna moderately large Ileles, Taxidea. 



f . Pinna much reduced Lutra, Amhlonyx, etc. 



d' . Pinna a thickened rim of integument Ilellivora. 



c' . Auditory orifice exposed, not concealed by tragal thicken- 

 ing Spilogale, Mej>h itis, Conepatus. 



The Feet. 



In describing the feet of the Mustelidse I start with two 

 assumptions. First, that the primitive feet were adapted for 

 running or climbing, or both combined, and not for digging or 

 swimming, which hamper, if they do not prevent, the first- 

 mentioned activities. Second, that the feet were naked below, 

 provided with short curved claws, interdigital webbing up to the 

 proximal end of the pads, large 4-lobed plantar pads, double 

 carpal and metatarsal pads, and a hairy heel ; and that the digits 

 were unequal in length, the first being the shortest, the second 

 and fifth next and approximately equal in length, and the third 

 and fourth the longest and also approximately equal, so that the 

 digital pads, when not spread, formed a strongly curved line 

 round the anterior margin of the plantar pad. And possibly in 

 the hind foot the third and fourth digits were more closely tied 

 together than to the second and fifth respectively. 



The genera whose feet most nearly approach these hypothetical 

 feet are Tayra and Grison. It is therefore needless to describe 

 them in detail. From the material examined I have not been 



