122 



supplying the muscles of the region. Another short branch is given 

 off at the dorsal end of the ceratohyal, the afferent mandibular artery 

 then traversing the suspensorial apparatus, between the preopercular 

 and the posterior process of the quadrate on one side and the body 

 of the quadrate and the symplectic on the other, exactly as it does 

 in Scomber (Allis, 1903) and in the Loricati (Allis, 1910), It then 

 crosses the external surface of the symplectic, giving off while doing 

 so a large mandibular branch and a small opercular one. It then 

 again traverses the suspensorial apparatus, between the hyomandibular 

 and the metapterygoid, exactly as in Scomber and the Loricati, and 

 having reached the internal surface of the hyomandibular enters and 

 supplies the pseudobranch. It does not perforate the hyomandibular 

 at any point. 



The ophthalmica magna, or efferent pseudobranchial artery has 

 its usual origin from the pseudobranch, but the full course of this 

 artery, and that of the internal carotid, could not be determined, my 

 only specimen having been killed by a blow on the head that had 

 whoUy destroyed the arteries in the region of the myodome. Dohrn 

 says that, in a 11 mm. embryo of the trout, the efferent pseudo- 

 branchial artery — that is, the efferent artery of the mandibular 

 arch — still empties into the lateral dorsal aorta, and that at this 

 stage the ophthalmica magna artery arises from the lateral dorsal 

 aorta, as a branch of that vessel, slightly anterior to the point where 

 the aorta is joined by the efferent pseudobranchial artery. In an 

 18 mm. larva of the same fish Dohrn says that the ophthalmica 

 magna has become a branch of the efferent, pseudobranchial artery, 

 and that the latter artery has lost its connection with the lateral 

 dorsal aorta and become united with its fellow of the opposite side 

 by transverse commissure. These conditions in these two larvae very 

 markedly suggest that the ophthalmica magna is the efferent artery 

 of the premandibular arch and not a dorsal commissure between that 

 arch and the mandibular arch, as I have heretofore considered it, 

 and this probability will be again later referred to. Dohen shows 

 neither external carotid nor orbito-nasal artery in his figures of either 

 of these embryos, nor does he show any posterior cerebral artery, 

 the internal carotid being shown running directly forward as a single 

 vessel after giving off the optic artery (A. centralis retinae); a con- 

 dition of these vessels that certainly needs control. 



The external carotid, in ray adult salmo, arises from the lateral 



