640 
to the tissues from which the chondrocranium would later be devel- 
oped, and in that position it must have extended forward to the point 
where it was joined by the most anterior of the visceral aortic arches. 
This most anterior aortic vessel was certainly related to a preman- 
dibular visceral arch, and one such premandibular aortic vessel is 
quite certainly represented in the arteria ophthalmica magna of fishes 
(Aruıs, 1912). Whether or not there were primarily other and more 
anterior aortic vessels there is nothing that I know of in the arterial 
system to indicate, but it would seem as if there might have been 
at least one more such vessel, for there are certain indications that 
the palatoquadrate of existing selachians is formed by the fusion of 
three visceral arches (ALLIS, 1914a). From this primary lateral dorsal 
aorta branches were probably sent to the acoustic, latero-sensory, 
optic and olfactory organs, and one or more branches were certainly 
sent to the related portion of the neural tube. As the brain devel- 
oped and required a larger supply of blood, one of these cerebral 
branches, the so-called arteria carotis interna, cerebralis or encephalica 
of current descriptions, underwent a corresponding development and 
finally became a more important vessel than that portion of the lateral 
dorsal aorta that lay anterior to it. This artery probably at first arose 
at practically a right angle to the main aorta, but the strong blood- 
current running forward in it would tend to lessen this angle, and 
the branch (arteria cerebralis sive carotis interna) would finally come 
to appear, in the adult, as the direct anterior prolongation of the aorta, 
When the tissues around the neural tube began to chondrify, 
which was certainly after this portion of the arterial system was well 
developed, the arteria cerebralis sive carotis interna would necessarily 
be enclosed in the forming cartilage at the point where it passed 
upward into the cranial cavity, but the lateral dorsal aorta, lying 
ventral to those tissues, would not primarily be so enclosed. This 
condition of these vessels apparently persists in the adult teleostean 
and in Polypterus, for there is, in these fishes, an arterial vessel, 
the arteria orbito-nasalis of current descriptions, that lies anterior to 
the arteria cerebralis and external to the chondrocranium, in the line 
prolonged of the lateral dorsal aorta, and hence in the position that 
the anterior portion of that aorta must primarily have occupied. This 
is well shown in Amieurus and Polypterus (ALLIS, 1908 c, and 1908b), 
for in each of these fishes the efferent pseudobranchial artery (man- 
dibular aortic arch) falls into the internal carotid artery (lateral dorsal 
