from the Base of the Red Crag of Suffolk. 291 



of the anterior talon and the separation of the inner and outer 

 columns being decisive." 



In the fourth volume of the Catalogue of Fossil Mammalia, 

 published hy the Trustees of the British Museum (1886); p. 65, 

 Mr. Lydekker records the cast presented by me as No. 48,427 in the 

 collection, and declares it to be a cast of the first three ridges of the 

 second or third left upper true molar of the Tetralophodont Mastodon 

 longirostris of Kaup. He states that the original is described and 

 figured by Lankester as the complete molar of aTrilophodont species, 

 and proceeds — "Tlie specimen has, however, lost a ridge, and agrees 

 precisely with the corresponding portion of the molars of the present 

 species (ilf. longirostris)." In a note Mr. Lydekker refers to Lartet's 

 opinion of the ' specimen.' But as a matter of fact Mr. Lartet had 

 never seen the specimen any more than had Mr. Lydekker. Both of 

 them based their opinions on a very incomplete cast of a specimen, 

 which was still largely obscured by matrix, and had never been seen 

 by either of them. 



There never was any doubt in my own mind as to the 

 completeness of the enamel crown under discussion, and I am glad 

 to be able to state that on examining the actual specimen, lent to me 

 by the authorities of the York Museum last autumn, Mr. Lydekker 

 at once admitted that the enamel crown is complete, and that his 

 contention that it was not so is devoid of any foundation. 



During its sojourn in the British Museum (Natural History) our 

 skilled ' preparateurs ' have removed from the 'Baker' molar the 

 mass of matrix which filled up the two valleys. It is now, therefore, 

 a more hopeful task than was the case thirty years ago, to attempt 

 to arrive at a conclusion as to the species of Trilophodont Mastodon 

 to which it should be referred. 



The clearing out of the matrix from the valleys in our specimen 

 has shown that the transverse ridges and valleys have by no means 

 the simple and clean-cut character which is found in those of 

 Mastodon Borsoni. The semi-detached columns on the sides of the 

 first and second ridges and the sharp median incision traversing the 

 whole tooth antero-posteriorly place this tooth in that middle 

 group of Mastodons (so far as ridge structure is concerned) occupied 

 by Trilovliodon angustidens and TetralopJiodon longirostris, a group 

 lying between the simpler TrilopJiodon Americanus and Trilophodon 

 Borsoni and the complicated Trilophodon Eumholdti and Tetra- 

 lopJiodon Arvernensis. 



The TrilopJiodon angustidens of Cuvier is a Middle Miocene form 

 to which several other species are related, perhaps only as local 

 specializations. M. Pyrenaicus of Falconer and M. tapiroides 

 (turicensis) are not remote from it. Specimens are preserved in the 

 various museums of Europe which afford an almost complete 

 transitional series between these three species. The most striking 

 difference between M. angustidens and the two others is, that it has 

 really narrow molars whilst the others have broad ones. Our Suffolk 

 molar is remarkably broad in proportion to its length, broader, I 

 should say, than any penultimate upper molar I have seen. 



