Dr. C. W. Andrews — Fossil Mammalia from Ecjijpt. 403 



The dimensions of the atlas are : 



mm. 



Greatest ■width across articulation for skull 140 



Greatest width across articulation for axis ... ... ... ... 153 



Antero-posterior width of base of the transverse process ... ... 94 



Height of neural canal ... ... ... ... ... ... ... 70 



Width of neural canal C2 



In the imperfect right scapula, probably belonging to the same 

 species, the supra-scapular border, the upper edge of the spine, and 

 the upper portion of the glenoidal border have been broken away. 

 The most notable points about this bone are : (1) the straightness 

 of the coracoid border, which for the greater part of its length is 

 parallel to the base of the spine ; and (2) the comparative narrow- 

 ness of the prescapular fossa. In both these points the fossil 

 scapula resembles that of It. bicornis more nearly than those of 

 other living species, but its coracoid border is still straighter and 

 the spine is less inclined backward than in that species. In 

 B. unicornis the width of the bone above the coracoid tuberosity 

 is relatively considerably greater, the coracoid border more convex, 

 and the prescapular fossa somewhat wider than in the fossil ; but 

 on the other hand, in the general backward curvature of the blade 

 there is considerable similarity in the two forms. The scapula of 

 B. sondaiciis is widely different, being much more expanded and 

 having a strongly convex coracoid border. In B. simus also the 

 coracoid border is convex, the prescapular fossa wider, and the 

 coracoid tuberosity much more massive. Another point which dis- 

 tinguishes this scapula from those of recent types is the form of the 

 glenoid articulation, which is ovate, with the narrow end towards 

 the coracoid border ; in recent forms this surface is nearly circular, 

 although in B. indicus there is some indication of an ovate form. 



Comparison with fossil forms is difficult, owing to the fact that 

 the scapula is seldom figured, and that there are scarcely any 

 specimens in the Natural History Museum. Amongst the scapult© 

 of which figures are available our fossil seems to approach most 

 nearly to that of B. pachjgnathus, from the Upper Miocene of 

 Pikermi, figured in Gaudry's " Animaux fossiles et Geologie de 

 I'Attique," pi. xxxi, and from this it differs in much the same points 

 as from the scapula of B. bicornis. 



The scapula is too incomplete to give any accurate measurements, 



but its approximate dimensions are as follows : — 



mm. 



Glenoid surface, long diameter 77 



Glenoid surface, short diameter... ... ... ••■ ••• ••• 62 



"Width immediately above coracoid tuberosity .. . ... •■• ■•• 95 



Approximate length ... 360 



As was remarked above, the remains here noticed are far too 

 incomplete and the material for comparison too scanty, for it to be 

 possible to say more than that the bones are those of a rhinoceros ; 

 but at the same time the general similarity of the scapula to that 

 of B. bicornis and B. pachjgnathus suggests the probability that this 

 Miocene type will prove to belong to the same series of which those 

 species are members. 



