TRIASSIC FISHES FROM SPITZBERGEN 
31 
On all the teeth in question the crown consists entirely, apart from its thin 
peripheral layer of enamel' (cf. Tomes, 1914, pp. 50—59), of ortho-dentine, as in 
Palaeobates. As has already been pointed out, the root consisted of ordinary trabe¬ 
cular-dentine (cf. Tomes, p. 192; Rose, 1898, p. 42). 
* 
Remarks. — The types of teeth dealt with here have so many common charac¬ 
teristics that, in spite of their rather great differences, I feel justified in combining them 
in one species. I have suggested the name P. pyramidalis for this species on account of 
the shape of the principal cone. 
As far as is known P. pyramidalis seems to be most closely related to P. angiilaris 
described below. It is well differentiated from P. polycyphus (Agassiz) — cf. Jaekel, 1889. 
Geological occurrence and locality. — Polyacrodus pyramidalis is only 
known from Mt Bertil, where it occurs in a bone-bed belonging to the upper Triassic 
(the upper sandstone series). According to barometrical measurements the bone-bed is 
situated about 263 m above the fish level, but on account of the slope of the strata this 
figure is probably somewhat too low. 
Polyacrodus angulatus n. sp. 
(PI. 1, fig. 27.) 
Under this name I have provisionally grouped five detached crowns of teeth, 
belonging to the material collected by Salomon in 1910 and in his possession. None of 
the crowns is accessible on both sides for investigation. They are all very small, their 
length varying between about 1*5 to 2 mm and the height between 0*5 to 075 mm. 
As in P. pyramidalis, so here one finds teeth of at least three types. 
The crown which is shown in text fig. i 3 A is the only one known of its type. 
It is nearly straight and has a powerful, rather high and somewhat blunt principal 
cone, wfiich is symmetrically situated between both ends of the crown. The principal 
cone and the part basally of it are very much thickened in comparison with the other 
parts of the crown and this thickened part also forms a swelling on the basal margin 
in the same way as in the type of tooth figured in text fig. 12 E belonging to P. pyramidalis 
though not so much prominent as in this latter. There is only an indication of a lateral 
cone on one side of the principal cone. 
A longitudinal crista seems to be developed and probably also extended over the 
principal cone. On the side accessible for investigation the crown is nearly quite 
smooth or has at least no striae clearly discernible. 
Another type is also represented by one crown (text fig. i 3 B). This is bent in 
the vertical plane through its axis of length in the usual way, so that its base is concave 
in the longitudinal direction and its dorsal contour is correspondingly convex. It has 
a principal cone that is certainly not high but all the same very distinct; this cone is 
bluntly rounded and is situated somewhat asymmetrically on the side of the transversal 
middle axis. A lateral cone is very faintly indicated on each side of the main cone by 
