420 DR. R. BROOM ON THE 



by which I have endeavoured to maintain this position ; sutfice it 

 to say that I now regard the position as untenable, since the 

 element in the Anomodonts which I have long regarded as the 

 tympanic now proves to be unquestionably the stapes. 



In the type skull of Dicynodon latifrons there is a slender rod- 

 like bone lying in the cavity which is bounded above by the 

 exoccipital, and usually by a dumbbell-like bone below. The 

 stout dumbbell-like bone which stretches from the basioccipital 

 process to the quadrate I have hitherto believed to be the 

 tympanic, and the slender rod-like bone the stapes. New 

 evidence now shows that whatever the slender bone may be — 

 perhaps a displaced part of the hyoid — tlie supposed tympanic is 

 luidoubtedly the stapes, and what I believed to be the fenestra 

 ovalis is the foramen for the viitli nerve. 



Having the imperfect occiput of a moderate sized Dicynodon, 

 in which the matrix was much hardened by epidote and the 

 bone mainly calcareous, I saw it might be possible to dissolve the 

 bone and leaye a cast of the internal ear and the basicranial 

 nei've-passages. Though on one side much of the internal ear 

 was found to be occupied by calcite, the result of the decalcification 

 on the whole was so satisfactory that the structure of the ear in 

 Dicynodon is now almost as well known as in recent animals, and 

 as much is known of the cranial nerves as is ever likely to be 

 known. 



The internal ear (PI. LVI. figs. 1-3) is unlike that of any 

 animal hitherto described, so far as I am aware. While the 

 semicircular canals are appai-ently not unlike those of recent 

 reptiles or mammals, the vestibule is very remarkably elongated 

 and there is no trace of a cochlea. 



The vestibule has an upper, somewhat triangular portion, 

 which lies between the upper end of the canal for the viith nerve 

 and the upper end of that for the ixth, xth, and xith nerves. 

 In close contact with the brain cavity thei^e is an upward 

 extension, with which is united the canal common to the anterior 

 and postei'ior semicircular canals. On the anterior and posterioi' 

 sides are irregular excrescences, which are perhaps vasculai- 

 spaces tilled with matrix. The broadest part of the vestibule is 

 where the ampullae of the anterior and posterior semicircular 

 canals are situated. From this point the vestibule passes down- 

 wards for a distance about twice as great as the maximum width. 

 To near the middle of this descending portion there is a steady 

 narrowing, but beyond the middle the vestibule again expands and 

 the fenestra ovalis is about as wide as the widest part of the vesti- 

 bule. The fenestra ovalis lies near the middle of the descending 

 basioccipital process, but looks outwards and slightly downwards. 

 Tins large descending process, so conspicuous in the Anomodont 

 skull, is formed posteriorly by the basioccipital bone, and in front 

 it is supported by the basisphenoid, but there seems considerable 

 reason for believing that the centre portion is partly prootic and 

 partly opisthotic. Down the descending part of the vestibule there 

 appears to be evidence of the suture between the two otic bones. 



