Dr. C. W. Andrews — Notes on Egyptian Eocene Mammals. 159 



angle of the anterior one, from the outer angle of which another 

 ridge runs forwards and inwards to the anterior face of the tooth. 

 The pattern assumed in wear is shown in Fig. 2. These molars in 

 some respects resemble those of some species of Cort/jjhodon (e.g. 

 C. simus^}, but are more hypselodont. 



The premolars are much more compressed laterally than the 

 molars ; they seem to consist essentially of a pair of crescents, 

 but the details of their structure cannot be discussed here. The 

 -canines and incisors are simple columnar teeth wearing to a common 

 level and forming a closed series both in the middle line and with 

 the premolars. Altogether the dentition in this genus seems to be 

 one of the most remarkable known, at least among the Ungulata. 

 The teeth here specially referred to and figured are those of the 

 type-specimen of Arsinoitherium andrewsi, Lankester. 



The skeleton is almost completely known, but in the case of 

 some of the bones there may be some danger of confusion with 

 those of Palceomastodon. 



The axis has a blunt peg-like odontoid process ; its centrum and 

 still more those of the cervical vertebrae behind it are very broad 

 and short, so that the neck must have been nearly as short as in 

 the elephants. 



The scapula is much like that of Binoceras as figured by Marsh 

 in his monograph of the Dinocerata. 



The humerus differs considerably both from that of Elephas and 

 of Dinoceras. It is especially remarkable for the extreme antero- 

 posterior compression of the lower part of its shaft and distal 

 end, and for the presence of a very prominent deltoid process. 



The radius and ulna are very short and stout, and do not differ 

 widely in their main features from those of Elephas, while in 

 some points, e.g. the distal articulation of the radius, they are unlike 

 the corresponding parts of Dinoceras. The distal articulation of 

 the ulna is still larger in proportion to that of the radius than 

 in the elephants. In these latter, in some cases, the lunar has 

 a surface for the trapezoid as well as for the magnum, there being 

 apparently some displacement of the proximal row of carpals to 

 the pre-axial side, instead of post-axially as usual. Whether this 

 is so in Arsinoitherium or not in the case of the lunar is not known 

 at present, but there is some evidence that the cuneiform extended 

 pre-axially a short distance over the magnum. Weithofer ascribes the 

 peculiar displacement in the elephants to the preponderating size of 

 the ulnar articulation, and the same cause may have been efficient here. 



The short stout metacarpals are somewhat displaced outwards ; 

 the third has a small contact with the unciform which entirely 

 supports the fourth and fifth. 



The femur is chiefly remarkable for the great antero-posterior 

 compression of its shaft, the outer border of which is a thin sharp 

 edge without any distinct projection representing the third trochanter. 

 The distal articulation is much as in Dinoceras. 



1 See Osborn, "Evolution of the Amblypoda," pt. i: Bull. Amer. Mus. Nat. 

 Hist., vol. X (1898), p. 192, %. 16. 



