THREE EXTINCT SPECIES OF ELEPHANT. 245 



or, as it were, elongated in the antero-posteiior direction, and perhaps more so in the 

 African than in the Asiatic species, as is, in fact, in some measure, shown in pi. vii. 

 fig. 3 of the ' Ossemens Fossiles ; ' but in the present case this compression appears to 

 be carried to the extreme. There does not appear to be anything distinctive in the 

 form or size of the tuberosity ; but the bicipital groove, even as compared with that of 

 the African Elephant, is remarkable for its great width and extreme shallowness. In 

 fact, after making every possible allowance for the portion of bone which has been de- 

 tached from the anterior border of the head, it would seem that there could scarcely 

 have been any distinct bicipital groove, and certainly none at all comparable in depth 

 with that in E. primigenius, indicus, and afncanus, as may be seen in Cuvier's 

 figures above cited. And from these, as well as from an observation in the text 

 (torn. ii. p. 218), it would appear that this groove is still narrower in the Mammoth 

 than it is in H. indicus. It should also be mentioned that both borders of the groove 

 arch over it in the mature E. indicus ; whilst in U. africanus, and still more so in 

 E. melifensis, there is no incurvation of the kind on either side. 



3. JJbia. — Of this bone the collection contains seven well-recognizable fragments, 

 three of which from their size would appear to belong to E. melitensis, and four to the 

 smaller species. One of the best-marked specimens of the former is represented in 

 PI. XL VIII. figs. 24 and 24 a. It is a portion, about four inches long, of the upper part 

 of the right ulna, which has been fractured transversely through the shaft, at about that 

 distance below the articular surface. The olecranon is broken off obliquely downwards, 

 on a level with the horizontal part of the articulation. The remaining part of the 

 articular surface is nearly entire, being only slightly eroded at the anterior part of the 

 outer cusp. The surface of the bone elsewhere, except at the upper part of the pos- 

 terior angle, as above noticed, is quite uninjured. 



The bone is evidently that of a fully mature animal ; and from its colour and con- 

 dition, both externally and within, it is not unreasonable to consider that it may have 

 belonged to the same indi\ddual as that which owned the upper end of the humerus 

 just described. The form of the articular surface is shown in the figui'e. The tr. d. 

 of the inner condyloid facet, measured from the middle of the radial sulcus, is 

 l"-65, and itsap. d. at i-ight angles to the same line, l"-65; so that it is circular; whilst 

 the radius of the curve of the concavity is about I"' 3*. The trausverse diameter 

 of the upper end of the bone at the level of the lip of the articular surface is 2"-6. 

 The transverse section at the lower end of the fragment exhibits a nearly equilateral 

 triangle, the anterior side («, fig. 5) of which rises into an angular eminence towards the 

 inner side ; the outer side of the triangle is slightly convex, and the internal is nearly 

 straight or slightly con^'ex, — the respective lengths of the three sides being: — internal 



* The tr. d. of the outer facet cannot be exactly determined, as it is partly destroyed ; but its extreme 

 length, measured from the bottom of the radial sulcus, may be estimated at 1""1, and its width, from the same 

 point in a line at right angles to its length, is about 0"-05. 



2 M 2 



