262 
" to those of the existing African elephant, in breadth, lozenge-shaped " 
"outline and mesial expansion, but when examined in detail, there are " 
"obvious points of distinction. In the living species the lozenges are" 
" more distinctly rhomb-shaped ; the salient edge of enamel is distinctly " 
" crimped ; the lateral terminations of the rhombs are flattened; and the " 
" mesial angles of the contiguous discs are either more approximated or " 
" overlap each other laterally. In E. [Loxodon) prisctis the discs are" 
" rounded at their lateral terminations and broader. Although the " 
" mesial expansion is quite as great as in the African elephant, it is less " 
" sudden, and in the general outline there is a tendency to a reniform or " 
" obsolete crescentic shape, the anterior enamel boundary of each disc" 
" being concave and the posterior convex" (1868, vol. ii., p. 97, pi. viii., 
figs. 3, 4). 
In E. ztilu the lozenge pattern is less distinct than in these excep- 
tional teeth of E. antiqnus, and the enamel is even more crimped and 
thicker than in E. africamts, but there is a strong resemblance to both 
species. 
Measuremenfs. 
Third lower molar, length '265 
width -085 
,, ,, ,, height of posterior portion "oSi 
From a single pair of teeth it would be a hazardous undertaking 
to determine the phylogenetic significance of E. zu/u, yet, so far as 
the evidence reaches, it gives us no reason to suppose that this species 
was not the direct ancestor oi E. africanus. In most discussions con- 
cerning the descent of the various species of elephants it is usually 
taken for granted that evolutionary progress in molar structure has 
always been by the addition of plates or discs. In many, perhaps 
most, cases this is probably true, but it is not necessarily true of all. 
The example of certain phyla, such as the Rodents, shows us that, 
within the limits of a single family, dental evolution may proceed, now 
by increased complication and number of elements, now by simplifica- 
tion and reduction of elements already present. 
Should E. zuhi eventually prove to be the actual ancestor of 
E. africamis, it would tend to give the latter a less isolated position, 
connecting it with Asiatic and European species. At all events, it is 
extremely interesting and important to find in South Africa an 
elephant with so many points of resemblance to species characteristic 
of the northern hemisphere. 
PAPERS QUOTED. 
Adams, A. Leith, 1877. — Monograph of British Fossil Elephants. Palteontographical 
Society, London, 1877-1881. 
Falconer, H., 1868. — Palseontological Memoirs. London, 1868. 
Pohlig, H., 1888. — Dentition und Kranologie des Elephas antiquus Falc, &c. Nova 
Acta der Ksl. Leop.-Carol. Deutschen Akademie der Natur- 
forscher., bd. liii., No. i. 
Thomas, O., 1895.— An Analysis of the Mammalian Generic Names in Dr. C. W L 
Gloger's Naturgeschichte. Ann. and Mag. Nat. Hist. 6th ser., 
vol. XV., p. 189. 
