i66 
ERIK A : SON STENSIO 
relation to the ventral parts of the diencephalon. Ventrally it is covered by the para- 
sphenoid. Whether we are to see in it only a secondary fenestration or we are to take 
it as a remaining part of the fenestra basicranialis posterior between the anterior ends of 
the parachordals (cf. Gaupp 1905, p. 663) is of course difficult to establish, but the latter 
alternative seems most probafile. It seem certain that the region pc of the pars basalis 
of the sphenoid, on account of its position in relation to the fossa hypophyseos, belongs 
to the parachordal section of the neurocranium, so that in the continuation I shall call 
it the basal parachordal plate of the sphenoid. In the Coelacanthids it seems to have its 
equivalent in the basisphenoid corpus and this is also the case in the Lacertilians, 
where the basisphenoid is formed in the crista sellaris (cf. Gaupp 1905, p. 759, fig. 382; 
Allis 1909 a, pp. 191 —192). 
In the most posterior part of the lateral surface of the pars basalis of the sphenoid 
there is in B. mougeoti a paired fossa ( pry , text figs. 62, 63 A, B; PI. 20, fig. 60; PI. 21, 
figs. 1, 3), which I have not mentioned above. As will be seen from the figures quoted, 
this fossa is low and oval in shape. In the anterior end of it the canalis transversus has its 
exit and the two fossae are thus in communication with each other here, while the 
parts situated posteriorly are separated by the median parts of the basal parachordal 
plate (pc). Dorsally each fossa is bounded by this plate and the process c (text figs. 62, 
63 A, B, C; PL 20, fig. 60; PI. 21, figs. 1, 3), rostrally by the posterior surface of the 
region bsph and ventrally by the parasphenoid. The posterior boundary was partly 
formed the by basal parachordal plate and partly by cartilage belonging to the most 
anterior part of the labyrinth region. Finally it is also noteworthy in this connection 
that the above-mentioned incisur ji (text fig. 63; PI. 21, figs. 1—3; PI. 22, fig. 3) is situated 
in the roof of the fossa (my) in question. 
The ventral half of the caudal part of the sphenoid seems to have been narrower 
than the cartilaginous part following behind in the labyrinth region. Now as the para¬ 
sphenoid has been situated close to the lateral surface of this cartilaginous parts and in 
its forward extension over the most posterior part of the orbitotemporal region has 
retained its position in the same plane as in the labyrinth region, there has arisen on either 
side between it and the basal half of the sphenoid an anteriorly open cavity which, as we 
shall see from my account given below, seems to include the homologues of the trige- 
minofacialis chamber and myodome. The fossa my described above now occupies such 
a position that it must have formed the posterior end of the myodome part in this 
cavity and, for reasons that will be given in more detail later in this description, ought 
to represent at least an antero-lateral part of the myodome of the Teleostei. In other 
words the myodome is in a very primitive phylogenetic stage of development. For here 
we seem to have a fossa merely for m. rectus externus, and this muscle would thus, in 
relation to the exit for the canalis transversus, have penetrated in to the basis cranii in 
a medial and posterior direction. The two fossae (my) formed in this way are, as we 
have seen, still widely separated, but, if extended more in a medial direction, would 
have met the fontanel in the cranial base, which I interpreted as the fenestra basicra¬ 
nialis posterior. The conditions present in the basal part of the orbitotemporal region 
in B. mougeoti thus seem well to support the view attained by comparative anatomy 
and embryology as to the phylogenetic origin of the myodome (cf. Gegenbaur 1872, 
