The Development of the Cartilaginous Skull ete. in Necturus. 429 
chondrifies before the posterior extremity of the bar, so that the 
cartilage of this bar is at first separated from the cartilage of the 
cerato-branchial bars of the third visceral arch by the unchondrified 
procartilage of the posterior extremity of the basibranchial bar. When 
eartilage arises in the posterior part of the first basibranchial, this 
cartilage is divided into right and left halves separated in the axial 
plane by procartilage. Each half, however is directly continuous 
anteriorly with the undivided cartilage of the body of the bar, and 
posteriorly each half adjoins the corresponding cerato-branchial of 
the third arch. This relation seemed of interest as tending to justify 
the name given to the first axial bar, by showing it to be, at least 
in its posterior part, composed of basal elements from the third 
visceral arch. There is an early stage of development when the 
cartilage of each cerato-branchial of the third arch is directly conti- 
nuous with the cartilage of the corresponding half of the posterior 
extremity of the first basibranchial. From the fusion of adjacent 
cartilages, however, little may be inferred regarding their primitive 
relations to one another. 
The second basibranchial is not formed in the same kind of 
tissue, mesectoderm, as the remaining branchial bars, but arises from 
cells on the anterior wall of the pericardium which are mesodermie, 
although it is probable that a few cells at the proximal end of the 
bar come from the mesectoderm. The bar arises later than the 
remaining branchial cartilages, and I am inclined to doubt the serial 
homology implied in its name, »second basibranchial«, and to regard the 
bar as a foreign element associated secondarily with the branchial arches. 
The dorsal, relatively lateral, margins of the cartilaginous bars 
of the three posterior arches, are connected with one another above 
the gill clefts by procartilage, indicated in fig. 15 by dark bands. 
No fusion has yet taken place between the quadrate cartilage 
and the auditory capsule. Gaupp says (93, page 434) »some false 
statements of GOETTE, such as the participation of the ascending 
process in the formation of the lateral wall of the skull have already 
been set right by Sréur«, and Srdur does indeed say ('80, page 492), 
that in the Urodela studied by him no such participation in the 
formation of the wall of the skull can be ascribed to the quadrate. 
In Necturus, nevertheless, the ascending process actually does bound 
the brain laterally, thus forming part of the skull-wall, as maintained 
by GoETTE (75). 
In the embryo 19 mm long, the inner wall of the auditory 
