TRIASSIC FISHES FROM SPITZBERGEN 
29 
under the name of Acrodus minimus. This makes it very desirable to have a microscopical 
investigation of all the types of teeth referred to this species. I did not have the material 
to carry this out myself. 
Polyacrodus pyramidalis n. sp. 
(Pi. i, figs. 21—26.) 
The material of which I have established this species consists of 12 incomplete 
teeth (P. yi 3 — 717)- With a few exceptions it is only crowns or parts of crowns that 
are present. They are usually so embedded in the stone that only one side is accessible 
for investigation. 
The smallest tooth is scarcely 2 mm long; the maximum height of its crown is 
somewhat over 1 mm. In the largest teeth the length may be estimated at about 8 mm, 
while the crown has a maximum height of 2-5 —3 mm. 
The teeth that I group here under the species name Polyacrodus pyramidalis are 
of at least three types, mutually rather different, and the dentition must therefore have 
been rather heterodont if my view is correct. Only a very vague idea can be obtained 
of the position of the different types of teeth in relation to each other when they were 
in situ in the jaws. To judge from their shape it seems probable, however, that they 
were arranged in somewhat the same way as in Acrodus and the Palaeobates species 
described below. 
The crown {P.717 a) figured in text fig. 12 A and in PL 1, fig. 21 is about 2 mm 
long and its greatest height is somewhat more than 1 mm. Even towards the well rounded 
ends the height is strikingly great. The crown is not quite straight but shows, as the 
figures indicate, a slight bend in the vertical plane through its longitudinal axis in the 
same way as several of the above-described Hybodus and Acrodus teeth. It has a broad, 
relatively high and rather sharp principal cone, but this is situated somewhat towards 
one end of the crown. Between the principal cone and this end there are two fairly low, 
blunt lateral cones. On the other side of the principal cone there are, on the contrary, 
no lateral cones at all. A longitudinal keel seems, as far as one can judge, to have ex¬ 
tended between the two ends over the three cones. On the side accessible for investi¬ 
gation — probably the lateral side — the sculpture consists of delicate, sparse striae, 
which issue from the longitudinal keel and pass off in a basal direction. They do not 
nearly reach the basal margin of the crown, however, but between this and their basal 
ends there is a sculptureless stretch. The striae often show a certain convergence towards 
the distal ends of the cones. — Only fragments of the root are preserved. 
The tooth figured in text fig. 10B and PI. 1, fig. 22 (P. 714a) resembles very closely 
in all respects the one just described. It is, however, larger and has lower and blunter 
lateral cones. The root is partly preserved. As in Hybodus and Acrodus a number 
of Haversian canals open on its surface. 
Another type of teeth is represented by five crowns (P.712 a, P.7i4b, P.715 a and 
b, P. 717b), all of which resemble each other very closely. None of them is completely 
preserved. The largest one was probably about 8 mm long and about 2^5 mm high, 
while in the case of the smallest these measurements were 6 mm and 2 mm respectively. 
They have all been slightly bent (text figs. 12 C, D; PI. 1, figs. 23 , 26) in the same way 
