FEOM THE SIVaLIK HILLS. 125 



shows well the downward and forward extension of the mandibular 

 symphysis, which is more elevated than in M. meganthereon. The 

 canine is absent ; but, from its inferred position, the diastema be- 

 tween it and the first false molar may be concluded to have been 

 shorter than either in M. neogceas or M. meganthereon ; and the 

 height of the ramus in front of the molar series is proportionately 

 greater than in these two species. These peculiarities are better 

 exhibited by the specimens to be described under the next species. 

 The crown of the first premolar is broken ofi", but it is entire in the 

 other fragment of lower jaw *. The anterior accessory cusp, which 

 is well developed in M. meganthereon^ as well as in M. neogceiis, is 

 entirely absent ; and the posterior is not so well marked as in these 

 two. The antero-posterior extent of the crown is less than in M. 

 meganthereon f. The next premolar (pm. 4) slightly overlaps the 

 carnassial, and is composed of a low anterior cusp, a deltoid middle 

 lobe, and a bilobed posterior cusp, the summits of all these pointing 

 backwards, as in the Brazilian and Prcnch species. The antero- 

 posterior length of the carnassial, like that of the preceding molar, 

 is intermediate between that of the corresponding tooth in these two 

 species. 



Mach^rodtjs pal^indicus, nob. (PI. YI. figs. 1-4.) 



The specimens just described indicate a species of Machcerodiis ad 

 large as the Jaguar J ; but the two specimens of lower jaw§ which 

 have been added to the collection since Dr. Palconer's death, as well 

 as the two crania mentioned above, must have belonged to much 

 larger individuals, nearly equalling in size the Koyal Tiger of 

 Bengal. 



One of the fragments of lower jaw exhibits the symphysial ex- 

 tension downwards very well. In the other known species of 

 Machcerodiis the extension takes place below the canine; but in. 

 M. palceindicus the downward prolongation of the outer border of 

 the ramus takes place further back, below the second false molar 

 (pm. 4). The fossa for the recej^tion of the upper canine is much 

 better defined than in any other species. In the Brazilian, as well 

 as the French form, the border of each ramus curves upwards in 

 front of the anterior premolar, and then downwards to meet the 

 symphysial extension. But in both the Sivalik forms there is no 

 such curvature ; consequently the height of the diastema (especially 

 in M. palceindicus) between the canine and the first false molar is 

 considerably greater. This, coupled with the relative shortness of 

 the diastema, gives a very formidable look to the lower jaw of 

 M. palceindims. As noticed by De Blainville in M. meg anther eon\\^ 

 the mental foramen is situated further down than in the Lion or 

 the Tiger. I have observed, in addition, that there is, as a rule, 



* B.M. no. 16554; figs. 8, 8^, unpublished plate N (F. A. Siv.). 

 t The first premolar (pm. 3) is of very variable size in M. neogaus ; and in 

 one of the specimens in the British-Museum collection it is entirely absent. 

 \ Owen, op. cit. p. 178. § B. M. nos. 4843G and 48437. 



II ' Osteographie,' genus Felis, p. 130. 



