642 
chondrification of the trabeculae. The trabeculae, when they begin 
to chondrify, necessarily follow somewhat the line of the ventral 
surface of the overlying portion of the brain, that line being more or 
less curved according to the more or less pronounced cranial flexure 
at the time of the chondrification. The corresponding portion of the 
lateral dorsal aorta, that part that lies anterior to the parachordal 
plate, must, on the contrary, tend to keep in, or to be brought into, 
the line prolonged of the posterior portion of the lateral dorsal aorta, 
because of the well known principle that when a fluid flows through 
a flexible vessel that vessel tends to become straight. Where there 
is marked cranial flexure, the anterior end of the lateral dorsal aorta 
must accordingly be lifted up relatively to the forming trabecula, internal 
to it in these fishes, and so come to lie dorsal to the trabecula; the length 
of the section of the artery so affected depending upon the amount of 
the cranial flexure at the time of the chondrification of the trabeculae. 
An arteria orbito-nasalis, lying ventral to the chon- drocranium, would 
not accordingly be found in these fishes, and one would be led to seek 
the homologue of that artery in an intracranial branch of the arteria. 
cerebralis sive carotis interna running forward to the nasal sac, 
In selachians, the section of the lateral dorsal aorta thus affected 
by the cranial flexure extends posteriorly to a point that lies between 
the dorsal ends of the mandibular and hyoidean aortic arches, 
and the dorsal ends of the mandibular (efferent pseudobranchial) and 
premandibular (ophthalmica magna) aortic arches of these fishes 
accordingly perforate the cranial wall dorsal to the trabecula in order 
to reach the aorta (carotis interna); falling into that artery, however, 
while it still lies between the cartilage of the chondrocranium and 
the tough lining membrane of the cranial cavity (ALLIs). There is 
no arteria orbito-nasalis in these fishes (ALLIS, 1908a, p. 119), and 
I now find, in Mustelus, a small artery which arises from the carotis 
interna at the point where the arteria optica is given off and that 
runs forward between the cartilage and the lining membrane of the 
cranial cavity, and when it reaches the foramen olfactorium there falls 
into a large branch of the cerebralis anterior that runs forward, in the 
cranial cavity proper, to the nasal sac. This small artery would thus 
seem to represent the persisting anterior portion of the primary lateral 
dorsal aorta. . 
In Polyodon (Aruıs, 1911), the conditions seem to be much as 
they are in selachians, excepting that here the efferent pseudobranchial 
