TRIASSIC FISHES FROM SPITZBERGEN 
175 
conditions described in B. mougeoti, we may now expect the region in question of the 
primordial neurocranium to be like that of the Crossopterygii even in the most primitive 
Actinopterygii (cf. p. i 3 g above). 
Finally the account given here ought to show that in primitive forms among the 
Teleostomes the prootic did not extend farther forward than to or at any rate a very 
little in front of the facialis foramen. The extension of the prootic in the majority of 
the Teleostei in front of this foramen ought, to jugde from the facts so far known, to 
be looked upon as secondary. It seems to have taken place with a simultaneous 
reduction of the most posterior part of the sphenoid or of corresponding ossifications 
in case the sphenoid was divided into several bones. 
The membrane bones of the primordial neurocranium. 
Parasphenoid. The primordial neurocranium is covered ventrally by a wide, 
powerful parasphenoid ( Psph, text figs. 61, 62, 64; PI. 20, fig. 6; PL 22, figs, x, 3 ; PI. 23 , 
fig. 2), the two end parts of which are not preserved. The anterior end has undoubtedly 
been situated under the ethmoidal region, the posterior one has at least reached as far 
as the anterior part of the occipital region. The greatest width is under the anterior, 
cartilaginous part of the labyrinth region and to some extent under the posterior end 
of the sphenoid. On the boundary between the orbitotemporal and the labyrinth region 
a large processus ascendens (pr. asc) text figs. 61, 64; Pi. 22, figs. 1, 3 ) is sent out on 
either lateral surface of the neurocranium. From the posterior end of the bone a deep 
median sinus, rounded anteriorly, has extended forward (P. iyi, text fig. 65). A similar 
sinus is found in Acipenser (Parker 1882, PI. 16, fig. 3 ; Woodward, 1895 b, fig. 8, p. 36 ), 
Scaphirhynchus (Ivanzow, 1887, PI. I, fig. 2), Gyrosteus (Woodward, 1895 b, fig. 7, p. 36 ) and 
the Saurichtliys species described below. In the recent ones of these forms we find that 
the sinus in question contained the arteria dorsalis. It is therefore probably correct to 
conclude that this artery in Birgeria, Gyrosteus and Saurichthys also had a similar position. 
A bifurcation of the posterior end of the parasphenoid by a sinus also occurs in 
Polypterus and a number of other forms. The anterior end of the sinus is, however, not 
rounded off but has an acute angle. In such cases the sinus is occupied by the basi- 
occipital and has nothing direct to do with the arteria dorsalis. 
As has already been pointed out above (p. 154) the parasphenoid in the labyrinth 
region shows very special conditions. From its dorsal surface there issues a paired 
longitudinal lamella (pi, text figs. 59, 60, 62; PI. 20, fig. 6; PL 22, fig. 1, 2) in a dorsal 
direction. This lamella, called by me the profound lamella of the parasphenoid, is on 
either side situated between the prootic-opisthotic and the basioccipital in the posterior 
part of the labyrinth region (text fig. 59), but extends somewhat higher up than this 
last-mentioned bone, so that it has presumably come into direct contact with cartilage 
dorsally and medially. At about the middle part of the labyrinth region the profound 
lamella has in tranverse section the appearance shown in text fig. 60. At this place it 
is also situated medially of the prootico-opisthotic but did not come into direct 
contact with this bone, probably being separated from it partly by the labyrinth recess, 
partly by cartilage. Its relations medially to possibly existing cartilage and to the 
cranial cavity cannot be made out. In the anterior part of the labyrinth region it 
