OF EXTINCT MU CARDINE RODENT. 215 



the tooth for the whole of its distance ; this condition of wear is 

 also seen for instance in Leithia, Eliomys, and Xerus. Apart 

 from size, the upper molars of Leithia chiefly differ from those of 

 Hypnomys in their transverse ridges being considerably higher 

 and more sharply defined, and in the complete ones in the first 

 and second molars originating at the postero-internal border of 

 the crown as already described and figured by Mr. Lydekker*. 



Of the lower cheek-teeth the premolar is the smallest and is 

 obtusely triangular in shape ; the third molar is somewhat larger 

 with a rounded posterior margin, although the crown has a 

 tendency to become squarer in outline when much worn. The 

 first and second molars are the largest of the series, almost equal 

 in size and in shape quadrate, or with the antero- posterior length 

 slightly greater than the width. The number of ridges appears 

 to be much the same as in the upper cheek-teeth, though they 

 become broken up earlier and into a greater number of incomplete 

 ridges. In a moderately worn specimen (B.M. M 11674, PL I. 

 fi,g. 6) the first and second molars seem to have three complete 

 ridges, two distinct incomplete ones running half-way across the 

 crown from opposite sides and a shorter one coming in from the 

 antero-external corner, while the antero-internal area is broken 

 up into several rather indistinct loops. The third molar shows 

 four complete ridges and two small accessory ones. In the case of 

 aged individuals the confluent worn surface is of course along the 

 external border of the crown, it being on the internal side in the 

 upper cheek-teeth. In less worn specimens this external enamel 

 border is interrupted at two points, and the enamel between and 

 on either side of these channels, owing to its being slightly folded 

 over, is at this stage of wear more massive than in other parts of 

 the ridges. 



The number, position, and conformation of the roots of the 

 cheek-teeth in rodents form a subject of considerable interest and 

 importance, and it is one that has received a good deal of atten- 

 tion from investigators, notably Schlosser t and, later. Dr. Foi-syth 

 Major J. The specimens under discussion are no exception in 

 this respect, for their molar roots are unlike those of any Pleisto- 

 cene or recent genus with v/hich I have been able to compare 

 them with the single exception of Leithia, and that only in the 

 case of the upper molars. It may be mentioned that I have not 

 found this character previously noted in the descriptions of the 

 Maltese genus. The alveolar formula can be most easily realised 

 by reference to the accompanying text-figure (text-fig. 2), but at 

 the same time it should be borne in mind that such a formula 

 alone is of very little value for purpoess of comparison. The 

 pi^esent examples are a case in point, as, for instance, the first and 

 second upper molai's of Eliomys and Glis are three-rooted, as in 



* Pi-oc. Zool. Soc. 1895, pp. 861-3, text-fig. 1. 

 t ' Pakeontographica,' Band xxxi. 1885. 



X See, for instance, the following : " On Fossil Dormice," Geol. Mag. Nov. 1899, 

 ]>. 493, and " Ou some Miocene Squirrels," Proc. Zool. Soc. 1893, pp. 179-214. 



IG* 



