633 
The ramus palatinus facialis of Ceratodus, after issuing from the 
trigemino-facialis chamber through its posterior opening, runs forward 
along that portion of the palatoquadrate that forms its processus basalis 
and at the same time the floor of the trigemino-facialis chamber, lying 
at first slightly lateral to the carotis interna and then slightly lateral 
to the arteria palatina. Its basal portion must accordingly lie, in the 
adult, in a canal that lies between the chondrocranium and the under- 
lying parasphenoid (basale,GüntHEr), and that corresponds to the palatine 
canal of my descriptions of Lepidosteus (Arrıs 1909); the course of 
the nerve, in Ceratodus, differing from that in Lepidosteus only in that 
it issues by the posterior opening of the trigemino-facialis chamber 
instead of perforating the floor of that chamber. In Amia, Scomber 
and Scorpaena the course of this nerve is somewhat different, this 
being apparently due to the transformation, in these latter fishes, of 
the pituitary fossa into a myodomic chamber, which chamber is not 
developed in either Ceratodus or Lepidosteus. In Amia the mesial 
wall of the trigemino-facialis chamber is the mesial one of the 
three walls potentially present in this region of the chondro- 
cranium of fishes (Arrıs, 1914b), and the chamber is continuous 
ventrally with the myodomic chamber. The ramus palatinus arises 
from the nervus facialis after that nerve has entered the trigemino- 
facialis chamber, and running downward through the opening that 
puts that chamber in communication with the myodomic chamber it 
perforates that part of the floor of the latter chamber that leads to 
its orbital opening. It therefore does not perforate the cartilaginous 
floor of the trigemino-facialis chamber, as it does in Lepidosteus. In 
Scomber and Scorpaena the inner wall of the trigemino-facialis chamber 
is apparently the middle one of the three walls here potentially present, 
and not the mesial one, and apparently because of this, or related to 
it, the chamber does not communicate directly with the myodomic 
chamber. The ramus palatinus here arises from the nervus facialis 
either before that nerve has perforated the mesial wall of the trigemino- 
facialis chamber (Scorpaena) or immediately after it has perforated 
that wall (Scomber), and running downward in a relatively long canal 
in the lateral wall of the myodome enters that chamber and then 
issues from it by its orbital opening. It does not here at any time 
lie ventral to any part of the chondrocranium, this being due to the 
fact that a cartilaginous floor to the orbital opening of the myodome 
is not developed; this apparently being related to the resorption of 
the hind ends of the trabeculae in early larval stages. 
