TRIASSIC FISHES FROM SPITZBERGEN 
157 
names of the different bones. On a large Polypterus cranium at my disposal I have 
seen that the part of the parasphenoid which according to Van Wijhe and Goodrich 
corresponds to the prootic has a greater extension backwards and upwards than in the 
specimen figured by Traquair (1871). For while in this the lateral wall of the neuro¬ 
cranium between the opisthotic, the lateral margin of the parasphenoid and the process 
running dorso-caudally from the parasphenoid which is pierced by n. facialis consisted 
only of cartilage, this is not the case in the specimen investigated by me. The part of 
the parasphenoid that occupies the place of the supposed prootic here covers the 
lateral side of the bulla acustica saccularis in its anterior half and only a small amount 
of cartilage is visible behind it. I was able to observe with certainty, however, that this 
part was not attached in any very intimate way to the capsula auditiva, i. e. it is not 
to be looked upon as a replacing bone, and there is consequently no prootic component 
in it. On the other hand I was able to observe a weak remains of a much reduced 
independent prootic in the wall of the primordial neurocranium (Pro, PI. 20, fig. 5). 
More closely defined, this rudimentary prootic is situated in the anterior lower corner 
of the bulla acustica saccularis. But it does not come into direct contact with the 
labyrinth, as it is considerably thinner than the cranial wall. 
The conditions of the substitution bones in the labyrinth region in fossil Ganoids 
have hitherto been somewhat well known in Lepidotus alone. 1 ) According to Wood¬ 
ward (1893 b, p. 560; 1915, pp. 37— 3 g) the prootic and epiotic are there joined together 
here by means of a suture. One or both of them must therefore have more than the 
usual extension. Consequently it does not seem improbable that their relations to the 
labyrinth can be different from what is otherwise the rule. In this case the most 
probable assumption would be that the external semicircular canal pierced one or both 
of them. If the epiotic had a considerable extension antero-ventrally and if this alone 
was pierced by the external semicircular canal, it would seem as if it contained the 
homologue of that part of the pterotic of the Teleostei that consists of cartilage bone. 
On the other hand this last-mentioned pterotic component would of course probably 
be fused with the prootic, if this bone had a greater extension in the dorso-caudal 
direction than usual and if at the same time it could be shown that it enclosed the 
main part of the external semicircular canal. If, finally, the epiotic and the prootic had 
grown together and both helped to surround the external semicircular canal the 
conditions would be more complicated and difficult to explain. 
From the account given, without a more detailed analysis of the conditions one 
might perhaps be inclined to overestimate the differences with regard to the relation . 
between the labyrinth and the substitution bones in the capsula auditiva in different 
Teleostomes. To some extent (Sagemehl 1884, p. 207) it seems to be fairly certain that 
these different relations are due to a different degree of development of the labyrinth 
itself. They seem, however, to be chiefly more apparent than real, as they are presum¬ 
ably due to an incorrect idea as to the homologues of the different bones or to unsuitable 
z ) That the primordial neurocranium was well ossified in both Dapedius and other fossil forms is certainly 
well known (Woodward 1893b, pp.563—564; 1895 b, pp. 128, 190, 287, 33o etc.; Frost 1913, pp. 219—222), but it 
has not been possible to establish anything' in detail about the extension and relations of the different ossifications 
in these forms. 
