616 EDWARD PHELPS ALLIS, -JR. 
investigation. The rami hyoideus and mandibularis facialis here 
run outward on opposite sides of the hyomandibula, the one pos- 
terior and the other anterior to it, this being exactly the relations 
that these two nerves have to the little process on the lateral sur- 
face of the hyomandibula of Acipenser. If that little process 
were to undergo marked development, and the actual articular 
head of the hyomandibula to undergo a corresponding reduction, 
the hyomandibula of Polypterus would apparently arise. 
In vertebrates higher than fishes I have made but little attempt 
to trace the hyomandibula, for it is evident that, as the descent 
of these higher vertebrates is not known, any one, or even all, of 
the forms of hyomandibula developed in fishes might also be 
there found; and to determine which one of them is represented 
in any particular case requires an intimate knowledge of the 
anatomy and development of the region which I do not possess, 
and which, as in the case of fishes, the literature alone, of the 
subject would probably not give. 
In Kingsbury and Reed’s (’09) very careful descriptions of 
this region in the Urodela, the columella is said to develop froma 
group of cells which is shown, in a transverse sectional view of a 
13-14 mm. embryo of Amblystoma punctata, lying close 
against the lateral wall of the vena petroso-lateralis (jugularis) 
and extending both dorsal and ventral to that vein, the ventral 
cells lying close against the lateral wall of the auditory capsule 
in the interval between the vena petroso-lateralis dorsally and 
the arteria carotis interna (lateral dorsal aogta) ventrally. Kings- 
bury and Reed say (I. ¢., p. 555) that ‘‘The derivation of these 
cells was not definitely determined. Younger embryos lend 
Some support to the view that they migrate down around the 
vein.” The basal, or fenestral plate of the columella is said to 
develop in that part of the mass of cells which lies ventral to the 
vena petroso-lateralis and is there in contact with the membrane 
that fills the fenestra vestibuli, these cells thus having the rela- 
tion to the auditory capsule, to the vena petroso-lateralis and 
also to the arteria carotis interna, that the pharyngohyal has in 
Ceratodus and the Batoidei. The stylus of the columella de- 
velops in that part of the mass of cells that lies lateral to the vena 
