'2QG Mr. R. I. Pocock on the 



foramen, moreover, is cut off from the periotic by bone, a 

 short straight suture alone indicating its original continuity 

 with the space between the periotic and the basisphenoid. 



In her paper upon Nandinia Miss Albertina Carlssoa 

 marks the carotid groove as running between the antero- 

 internal portion of the tympanic bone and the basioccipital. 

 This must, I think, be a mistaken inference. At all events, 

 the artery did not take that course in the fresh example of 

 Nandinia that I examined (see Zool. Jahrb. Svst. xiii. 

 pp. 509-528, pi. xxxvi. fig. 1, 1900). 



It may be added that there is no partition, either carti- 

 laginous, membranous, or osseous, in the bulla of Nandinia. 

 When the tympanic membrane is cut away, a probe can 

 be passed in the uncleaned skull through the external auditory 

 meatus to the posterior wall of the cartilaginous portion of 

 the bulla. 



In the Felidse, in conformity with the homogeneity of the 

 family, the carotid canal is much less variable than in the 

 Viverridse^. The canal is almost always apparent as a short 

 shallow groove notching the tympanic bulla close to the 

 basioccipital, and not infrequently set so far back that it 

 lies within the depression which leads to the forameti lacerum 

 posticum. Occasionally, however, the notch or groove lies 

 just in front of that depression, as in a skull of Felis jagua- 

 rondi I possess ; but it is never set nearly so far forward as 

 the middle of the inner surface of the bulla. Only quite 

 exceptionally, and as an individual peculiarity, is the notch 

 converted into a bony tube, with a rounded orifice, by the 

 extension and fusion of its edges, so that the basioccipital 

 forms no part of the carotid canal. This is the case on one 

 side, but not on the other, in a skull of Felis uncia, in which 

 the posterior orifice of the canal is, as in Mungos., a round 

 hole in the bulla. In this skull, as in that of F.jaguarondi, 

 the canal is placed in front of the foramen lacerum posticum. 



In all cases the canal descends f to the edge of the con- 

 cealed iuturned portion of the tympanic above the periotic, 

 where it ceases. From that point the artery apparently 

 runs along the periotic close to the basioccipital and the 

 adjacent portion of the tympanic, and in some cases this 

 portion of the tympanic is longitudinally grooved J ; but I 



* In this paper the significance applied to the term Viverridse by 

 Mivart and Flower is, without prejudice, adopted. 



t From the point of view of the spectator, when the skull is examined 

 with its base uppermost. 



t I have not, however, traced the course of the artery within the 

 bulla of any of the Felidaj. 



