ELEPHANTIDJE. 41 



M. angustidens stock. It is not known whether premolars were 

 developed \ 



Hab. India (Punjab and Sind). 



M. 2883. The first two ridges of the second left lower true molar, 

 in an unworn condition ; from the Derajat, Punjab, India. 

 This specimen apparently agrees precisely with the corre- 

 sponding portion of the right m72 figured in the ' Pakeon- 

 tologia Indica,' ser. ] 0, vol. i. pi. xxxiii. fig. 4. In addition 

 to its superior size and greater relative width, it differs 

 from m72 of If. angustidens by the form of the anterior 

 cingulum, by the absence of a distinct accessory tubercle 

 on the anterior side of the first inner column, by the 

 presence of a greater number of tubercles on the ridges, 

 by the form and size of the accessory tubercles in the 

 valleys, and the more decided approach to an alternate 

 arrangement of the columns. It is not improbable that 

 the specimen belongs to the same individual as an upper 

 true molar in the Indian Museum, Calcutta (No. A 327) 2 . 

 Presented by Prof. Sir B. Owen, K.C.B., 1877. 



Mastodon humboldti, Cuvier 3 . 



Syn. Dibelodon humboldti, Cope 4 . 



In the molars both the inner and outer columns (which are quite 

 separate) of each ridge are flanked by accessory tubercles, so that 

 the valleys are blocked on both sides of the median longitudinal 

 cleft; and the enamel on the ridges being vertically grooved, a very 

 complex pattern is produced on the worn ridges, there being trefoils 

 on both the inner and outer columns of each ridge, although those 

 on the inner columns of the upper and the outer of the lower are 

 the best defined ; the columns of the ridges have a tendency to an 

 alternate arrangement, and the median longitudinal cleft is well 

 marked. The valleys of the teeth are coated with a thick coat of 

 cement, and the mandibular symphysis is comparatively short and 

 without incisors in the adult, although these appear to have been 

 present in the young (No. 19952 c). Premolars have apparently 

 not been observed. 



1 The specimens figured in the ' Pakeontologia Indica' (Mem. Geol. Sui'V. 

 Ind.), ser. 10, vol. i. pi. xxxii. figs. 2, 3, under this name probably belong to 

 M. pandionis ; vide supra, p. 39. 



2 See Cat. Siwalifc Yert, Ind. Mus. Calcutta, pt. i. p. 102 (1885). 



3 Ann. d. Museum, toI. viii. p. 413 (1806).— M. humboldien. 



4 Proc. Amer. Phil. Soc. vol. xxii. p. 5 (1884). 



