TRIASSIC FISHES FROM SPITZBERGEN 
l83 
The palatoquadrate and mandibula were very long'. The mouth was consequently 
large and extended so far backwards that its posterior end was situated caudally of 
the orbital entrance, as in the Palaeoniscids in general. 
It is impossible to ascertain to what extent cartilage was present in the palato¬ 
quadrate. Of its membrane bones two are preserved, one of which I interpret as an 
ectopterygoid (Ecpt), the other as an entopterygoid (Enpt). 
The ectopterygoid {Ecpt, text fig. 65, PI, 23 , figs. 1, 2) is. a long bone situated along 
the lateral margin of the palatoquadrate. In its present state of preservation it is defec¬ 
tive medially, anteriorly, and to some exent also laterally, so that its width and its 
extension forwards cannot be ascertained with certainty. Backwards it extended to the 
quadrate. It occupies such a position that one of its surfaces is directed ventro-medially, 
the other dorso-laterally. The dorso-laterally directed one is weakly concave in a trans¬ 
versal direction the ventro-medially directed one correspondingly convex. The lateral 
margin is almost straight. The bone substance is strikingly thin. The centre of ossifi¬ 
cation is situated towards the posterior end of the bone rather near the lateral margin. 
The entopterygoid {Enpt, text fig. 65; PI. 23 , fig. 2) is represented by a small portion, 
situated between the anterior end of the ectopterygoid and the parasphenoid (Psph). 
Medially it is in direct contact witht he latter bone. The preserved part is rather narrow. 
It may possibly have had a considerable extension backwards as in other allied forms. 
Along the lateral margin of the ectopterygoid are found slight remains of the 
maxilla {Mx, text fig. 65; PI. 23 , fig. 2). It is evident that this bone was very long and 
that it presumably extended as far as the pars quadrata of the palatoquadrate, covering 
this on the outside. In its present state of preservation a posterior and longer part of 
it is situated close to the ectopterygoid without coming into connection with it. In the 
anterior part, on the other hand, its medial edge has been covered with a thin lamella 
from the ectopterygoid, and it also seems probable that this lamella existed farther 
posteriorly. That in this case we are not concerned with an accidental covering due to 
displacement of the two bones in question towards each, is shown by the fact that the 
lamella of the ectopterygoid that covers the maxilla is definitely distinguished from the 
rest of the ectopterygoid by the absence of teeth or any indication of them. 
I have previously shown (Stensio 1919) that the maxilla in B. mougeoti is typically 
Palaeoniscid-like in its general shape. I have further been able to show here that it was 
firmly attached medially to the ectopterygoid. Traquair (1877 a, p. 18) also surmised that 
a connection of this sort existed in Palaeoniscids. Woodward (1889 b, p. 3 oi) has described 
in Birgeria acuminata on the medial side of the maxilla a «palatal plate* that seems 
apparently to have been covered by the ectopterygoid. In Chondrosteus, (Traquair, 1887, 
pp. 258—256) as in the recent Acipenserids and Polyodontids, the maxilla is also closely 
joined to the palatoquadrate and this is also the case in Saurichthys. In the last- 
mentioned form there occurs, as we shall see (cf. Part II), as in B. acuminata, a medial 
lamella or list that is covered dorsally by the ectopterygoid. Finally we also find, as 
is known, a firm connection between the palatoquadrate and the maxillary in Polypterus, 
a connection which is formed posteriorly in the same way as in B. mougeoti and Saurichthys, 
but is due farther anteriorly, on the other hand, to the so-called maxilla covering the anterior 
end of the so-called ectopterygoid. 
