TRIASSIC FISHES FROM SPITZBERGEN 
17 
part one cannot decide with certainty whether a transversal crista has existed. At any 
rate, however, it is certain that, if it existed, it must have been short. The primary 
longitudinal crista is distinct. On the medial side the striae extend continuously from 
the base of the crown to the primary longitudinal crista, which they meet at right 
angles. They are all straight on this side and equally thick throughout their length. 
Between their distal ends there issue here and there from the primary longitudinal 
crista a few very short striae. On the lateral side of the same crista, on the contrary 
a great number of such short striae are sent out. The rest of the striae on the lateral 
side are all long and their distal ends are curved towards the highest part of the 
crown, and run together into a few. Each of these in its turn then continues a 
long way towards the middle of the tooth, being in this course at first rather 
parallel to the primary crista, then curving towards and finally joining this crista. This 
relation between the striae gives us a good idea of the origin of the secondary lon¬ 
gitudinal cristae. And, if one likes, one can here speak of a secondary longitudinal 
crista subdivided into several sections. All the transversal striae of the lateral side are, 
however, not of equal length basally. Only every second or every third one reaches 
right down to the basal margin of the crown (PI. 2, fig. 19). These longer striae are 
apparently thickened at their proximal ends; this is due to the fact that they really 
send out fine branches on both sides (PL 2, fig. 19 c). 
The root, like the crown, is highest and thickest at the middle. Its basal surface 
is about as oblique as that of the tooth P. 100 and like all other teeth in which this 
is oblique, it faces ventro-laterally. On the medial side the crown and root form together 
a notch which is developed most strongly at the middle part of the tooth. — A cross 
section through the lateral part of the tooth in question is shown in text fig. 4 F. 
Microscopical structure. — The teeth included here under the name of 
A. spitzbergensis show normal conditions of microscopic structure a fact that it is 
important to observe, as the teeth in question are partly rather distinctly differentiated 
from the usual Acrodus type by their macroscopical appearance (cf. Jaekel, 1889). 
Remarks. — Although the teeth described above sometimes show rather great 
mutual deviations, there seems to be no doubt, in my opinion, that the majority of 
them belong to the same species and, as we have seen, I think I can identify this with 
the species A. spitzbergensis founded by Hulke. A few of the teeth in question may 
perhaps belong to the species A. vermiformis described below, although with our present 
incomplete knowledge of this species they cannot be kept separate from A. spitzbergensis. 
It is not improbable that this is the case, for instance, with such teeth as P. 98 b and 
P. 99 c. If we leave out A. vermiformis, A. spitzbergensis seems to be well differentiated 
from most of the other known species. Such teeth as P. 98j, P. 98 1 and P. 98 m certainly 
bear a rather great resemblance to the A. pulvinatus type (Schmid, 1861, PI. 2, figs 2, 3 ; 
Woodward, 1889a, p. 2gg), but this is still too little known for us to venture to draw 
any conclusions as to its closer relationship to the Spitzbergen species in question. To 
a certain extent it also seems as if we could find some distant resemblance between, 
Stensio, Triassic Fishes from Spitzbergen. 3 
