TRIASSIC FISHES FROM SPITZBERGEN 
4 1 
found close to the root, run parallel to the anterior edge .of the spine, the .result of 
which is that the most posterior ones are rather short, as in their course distally they 
meet the postero-lateral edge fairly soori; The number of rows of tubercles thus 
gradually decreases in this way towards the distal end of the spine. 
It is noteworthy with regard to the tubercles that on closer inspection they appear 
to be flattened, with an obvious point, directed postero-dorsally. They also have a 
sculpture of fine ridges and in all respect much resemble dermal denticles. 
It is specially worthy of attention that the posterior surface is quite without 
denticles or formations corresponding to these. It is impossible, however, to decide 
whether this condition is primary or is due to the state of preservation. The pulp 
cavity opens on the basal half of the posterior surface in the usual way. 
The outer layer of the spine consists of trabecular-dentine and, at least in thl 
root, this layer is the only one present, as was also the case in 
spines described above belonging to Hybodus and Acrodus. As in 
these, so here too the layer of trabecular-dentine has on its surface 
the usual fine striation that is due to the microscopic structure. 
The two remaining spines (P. 4S, P. 44) are poorly preserved 
(PI. 1, figs. 12, i 3 ). The most complete one of them (P. 44) consists 
of a fragment of the root with an adjacent piece of the exposed 
part. The other is represented only by a fragment from the anterior 
half of the exposed part. Both belong in the same species. 
The spine P. 44 has a maximum width (measured from the 
anterior to the posterior edge) of about 10 mm and a maximum 
thickness (measured from one lateral surface to the other) of about 
5 mm. The greatest thickness is* situated in the posterior half 
other spine seems to have been smaller, but it is impossible to 
measurements of it. 
A transverse section through the spine P. 44 at the proximal end of the exposed 
part (text fig. 17) shows that all the edges -— both the anterior one and the postero¬ 
lateral ones — are only slightly marked and the different surfaces consequently merge 
into each other without any very distinct boundaries. The same cross section also 
shows that the posterior surface is strongly convex in the transversal direction. 
The anterior edge has no enamel list. The lateral surfaces have on the exposed 
part a tubercular sculpture closely resembling that described on the fin-spine P. 3 y 
although it differs from that by being developed quite close to the anterior edge 
(PI. 1, figs. 12, i 3 ). Further differences are that the tubercles in the two fin-spines 
discussed here are coarser than those of the spine P. 3 5 and that, contrary to what 
was the case in this spine, they do not seem to be arranged in distinct rows. 
The posterior surface has no trace of denticles, at least in the preserved part, 
but it cannot be decided whether this condition is primary or secondary. 
With regard to the microstructure we find that the basal part of both the spines 
consists only of trabecular-dentine around the large pulp cavity. Farther distally the 
spine P- 44 shows a rather reduced irregular pulp cavity (p, text fig. 17), which is 
almost surrounded by a thick and, to judge from the macroscopical appearance, compact 
Stensio, Triassic Fishes from Spitzbergen. 6 
certain of the fin- 
,o 
Text fig. 17. 
Cross' section through 
fin-spine P. 44. 1/1. 
The section has proceeded 
through the proximal end of 
the exposed part of the spine. 
p, pulp cavity; o, trabecular- 
dentine ; d, layer consisting, 
at least most centrally, of 
ortho-dentine. 
of the spine. The 
obtain any certain 
