August, 1903.] 



KNOWLEDGE. 



171 



animiil, but each of its " iutermediate " raolars was three- 

 ridged. 



A more primitive type is indicated by the four-tusked 

 mastodons, in some of which the " intermediate " cheek- 

 teeth are three-ridged, while in others they are four- ridged. 

 In most or all the members of this group the lower pair of 

 tusks are short, and pi'oject from the extremity of a long 

 trough-like extension of the lower jaw. Moreover, in a 

 skeleton of Mastodon angiistidens now mounted in the 

 Paris Museum, the upper tusks are bent downwards so as 

 to cross and project far below those of the lower jaw. As 

 regards its head, this mastodon, which is one of the oldest 

 of its kind, occurring in the Miocene strata of the Continent, 

 must have been very different looking to an elephant. To 

 correspond with the trough-like elongation of the lower 

 jaw, the upper part of the muzzle was probably prolonged 

 in proportion ; and consequently the pi-oboscis was rela- 

 tively short. 



Contemporaneously with the species last mentioned 

 lived another elephant-like creature — the dinotherium — 

 characterised by the presence of a pair of downwardly- 

 bent tusks in the lower jaw, and the more ordinary 

 structure of its cheek-teeth, of which all five pairs were in 

 vise at the same time. These teeth recall, indeed, to a 

 considerable extent those of a tapir, the two hinder pairs 

 in each jaw being sunnouuted with a couple of 

 transverse ridges. The tooth in advance of these 

 has, however, three ridges, like the " inter-_ 

 mediate " molars of some mastodons, and the 

 same is the case with the last milk-molar, which 

 in a young animal is situated immediately in 

 advance of the permanent three-ridged tooth. 

 In the adult the three-ridged milk, or baby-tooth, 

 is, however, vertically replaced by a successional 

 tooth of simpler structure, as in ordinary 

 mamuials. In build, the dinotherium was pro- 

 bably very like a short-truuked elephant ; it 

 may, perhaps, have obtained its food by wading 

 mid-leg deep in lakes and marshes, and thus 

 bringing its mouth within reach of the water- 

 plants. 



Although the dinotherium tends to connect 

 mastodons and elephants with ordinary hoofed 

 mammals, so far as its cheek-teeth are concerned, 

 yet the peculiar form of its lower tusks, and the 

 absence of weapons of this nature in the upper 

 jaw, show that it is not the ancestral type of the 

 former. On the contrary, elephants and mastodons 

 on the one hand, and the dinotherium on the 

 other, form branches of a common ancestral stock 

 which till lately was quite unknown. Indeed, 

 the seemingly sudden appearance of the pro- 

 boscideans, in the form of both these branches, in 

 the Miocene strata of Europe, had long lieen one of the 

 ])uzzles of palfeontology, and remained so till light was 

 thrown on the subject from an unexpected quarter. 



The region whence this light has come is the Fayum 

 district of Egypt, from the Lower Tertiary, or Eocene, 

 deposits of which remains of \uikuown forms of vertebrates 

 have lieen recently exhnmed. Among these, are jaws and 

 tc'th of a small and primitive mastodon, related in many 

 respects to Mastodon aiHjustidens, of which it may have 

 been the ancestral form. In this connection it may be 

 mentioned that remains of the latter sjx'cies have been met 

 with at Moghara, to the north-west of the Fayum. 



The lower jaw of this early mastoilon rescnililos that of 

 M. aiKjuxtidfiis in the elongation of the trough-like anterior 

 portion, and also in the cheek-teeth, which are thiee-ridged. 

 There is, however, the important difference that the last 

 molar rcseiiibles those in front of it in the luimbcr of its 



ridges, in place of having four ridges, and also that the 

 whole five cheek-teeth are in use at once. Nor is this all, 

 for the anterior two of these teeth are premolars, instead 

 of milk-molars ; that is to say, they have vertically replaced 

 deciduous predecessors. Clearly, then, this Egyptian 

 Eocene mastodon — palseomastodon, as it is called — is a 

 proboscidean but little removed as regards the characters 

 of its cheek-teeth from ordinary hoofed mammals, although 

 in the production of the lower jaw, and the reduction of 

 the front teeth to a single pair of tusks in each jaw, it 

 resembles typical mastodons. Apparently, the limb-bones 

 conform to the ordinary proboscidean type, from which it 

 may be inferred that the legs were constructed in much 

 the same manuer as in an elephant. The fact that such a 

 comparatively small animal has a conformation of limb 

 similar to that supposed to have l^een developed for the 

 purpose of carrying the enormous weight of an elephant, 

 throws doubt on the theoiw of the secondary origin of this 

 type of limb-structure to which allusion has been already 

 made. 



An earlier formation in the Fayum district has yielded 

 remains of an allied animal — the moeritherium — so named 

 from the ancient Lake Moeris, near the bed of which it 

 was discovered. This creature, which was about the size 

 of a tapir, departs widely from a mastodon, and still more 



Fio. 3.— The Imperfect Skull (A and B) and Lower Jaw (C) o£ the fore 



runner of tlie Elephants (the Mteritherium). i'— »\ incisoi-s - — 



pm-—pm*, premolars; m' — m'\ molars. (l?.v permission of 

 Andrews the describer ! 



canine ; 



, , ,. ,.... Dr. C. W. 



I joint-discoverer of the ancestral proboscideans.) 



from an elephant. Nevertheless, there are many features 

 in its structure pointing to the conclusion that it is an 

 ancestral member of tlip i>roboscidean stock. Each jaw is 

 furnished with six pairs of cheek-teeth, all of which ai-e in 

 use at the same time ; the series consisting of three pairs 

 of premolars and as many molars. The latter arc not 

 unlike the corresponding teeth of many early-hoofed 

 mammals— some of the extinct pigs, for instance-and, 

 ^vith the exception of the last in the lower jaw (, which is 

 three-ridged), carrv two transverse ridges. They display, 

 however, a tendency to the derelopmeut of thive ridges, 

 thus foreshadowing" the mastodon type. 



Very remarkable are the front teeth. In the upper jaw 

 these "comprise four pairs, of which the second is much 

 larger than either of the others. Tlie first three pairs 

 coiTespond to the incisors of a pig ; while the fourth pair 

 represent the canines. Obviously the thive small pairs of 



