1888.] NATURAL SCIENCES OF PHILADELPHIA. 207 



British jMuseum, has enabled the writer to determine fully the struc- 

 ture of the molars and premolars and to correct a previous error. 



In examining the first and second molars of the type specimen 

 under a strong lens, an external cusp was detected directly between 

 the internal pair, a discovery of great interest, since, in connection 

 with the last genus, it adds two important types to the tritu])ercular 

 series. This external cusp is probably the one referred to by Owen, 

 (op. cit., p. 14) in describing the penultimate molar of the second 

 specimen of Amphithermm.^ He si:»eaks of the latter tooth as the 

 posterior molar, but one can detect the tips of a molar behind this, 

 just bi'eaking through the jaw. 



The molar of Amphitherium is thus apparently similar to that of 

 Peramus wuth the exception that the external cusp, in the type species 

 at least, is less lofty. This observation led to a reexamination of the 

 jaw in the British ^luseum, No. 36822. This unquestionably belongs 

 to Amphitherium, as previously determined, (Osborn, op. cit., p. 192, 

 fig. 2.; Lydekker, op. cit., p. 374), but presents the inner face of the 

 right ramus instead of the outer face of the left ramus as previously 

 described. This is proved by the double internal cusps, l)y the cin- 

 gulum upon the premolar, and by the faint mylohyoid groove,' near 

 the lower border, Avhich was previously overlooked. The individual 

 is much smaller and younger than the two Oxford types, which are 

 nearly of the same size, and the tips of the para- and metacone? are 

 entirely unworn." 



In the Memoir, the formula of ^4?jy9/iifAermm was doubtfully given 

 as p)ih, »*6- (following Lydekker). An examination of the 

 Oxford types shows that Prof. Owen was more nearly correct in 

 putting it, jrwg, m^. If we deduct the foremost bifanged tooth 

 which he naturally reckoned with the premolar series, but which is 

 probably the canine, we have Ci, pm-^, vig. In the second Oxford 

 specimen there are undoubted traces of three premolars in front of 



1 "The posterior molar shows a middle internal and part of a larger external 

 cusp." This observation shows the keenness of the observer, for the molar referred 

 to is in a very fractured condition. 



- The teeth in the Stonesfield Slate specimens are much more brittle than those 

 in the Purbeck series, but it would 1 e well to run the risk (if injuring one of these 

 molars to expose the external cone. 



