302 - ON FOSSIL FISH-TEETH. 



lovrer surface of all the teeth is concave and hollow for the entire length 

 of the tooth. (PI. xvi, Fig. 1). Two teeth are shewn in contact in 

 Figs. 6 and 7, PI. xiv ; but unfortunately the artist has reversed what we 

 suppose to have been their true position and has put the lower-jaw tooth 

 uppermost in both cases. 



Among all the specimens the only instance in which the teeth were 

 found attached to the bones of the jaw is shewn in Figs. 1 and 7, PI. xiv ; 

 and although very imperfect the general form of the bone ma.y be made 

 out. The inner curve of the edge of the bone is seen, and gives some 

 key to determine the general form of the jaw. The bone was thin and 

 concave below, a well marked and widely rounded depression or furrow 

 traversing its entire length. Unfortunately the posterior portion of the 

 bone is broken, but it can be seen that it continued to extend for some 

 distance back, and was probably expanded laterally. 



This species more nearly resembles Ceratodus serratus and C. Phil- 

 lipsii, of Agassiz than any other figured by that author. But the 

 distinctions are obvious on comparison. I am disposed to think that both 

 the species referred to have been figured from specimens of lower-jaw 

 teeth ; and it does not seem at all improbable that both may even belong 

 to the same species, but of different ages. One marked difference be- 

 tween our Indian specimens and the teeth figured by Agassiz is the 

 rounded and obtusely blunt point of the anterior horn, which, in his 

 species, is represented very sharp or even bicuspid ; unless, as may be 

 the case, the two first points of Agassiz'a Fig. 18, PI. xix. Vol. iii, in 

 reality belong to one horn or ridge, the surface of which has been worn 

 down in the centre, so as to give a somewhat fallacious appearance of 

 division into two points. In our species also the ridges are narrower, 

 sharper, and more prominent, and divided by wider, and deeper farrows. 



There is, however, a close relationship between the Keuper C- 

 serratus and the Stonesfield-slate C. Phillipsii and our C. Hislopianus, 

 C. runcinatus (Plieninger), is even more closely allied, but differs in the 



