133 



sends a long branch downward in the second arch, this branch lying 

 in the deeper part of the arch, along its posterior surface, and hence 

 in what would seem to be the position of the posterior one of the 

 two efferent arteries found in each branchial arch in selachians. A 

 second and similar branch is then sent downward in the first branchial 

 arch, the artery then continuing onward and ending in the muscles 

 related to the dorsal end of the latter arch. The posterior portion of the 

 artery sends branches to the dorsal transverse muscles of the arches, sends 

 a branch dow nward in the third branchial arch in the same position as the 

 branches in the anterior arches, sends another branch downward in a 

 similar position in the fourth arch, and then continues backward a certain 

 distance but was not further ti-aced. The branch sent downward in 

 each of the four branchial arches extends more than half the length 

 of the arch and approaches but, so far as could be determined, does 

 not connect with a short branch sent upward in the posterior portion 

 of the ventral end of the arch; these latter branches arising either 

 from the ventral longitudinal commissure or from the ventral portion 

 of the efferent artery. 



The longitudinal portion of this dorsal branch of the second 

 efferent branchial artery of Amia lies dorsal, and hence morphologically 

 external to the dorsal ends of the branchial arches, and thus corres- 

 ponds, in position, to the dorsal commissures that connect the efferent 

 branchial arteries in selachians, while its ventral branches correspond, 

 in position, to the posterior efferent arteries in the related arches of 

 those fishes; and this correspondence is so striking in appearance that 

 it seems proper to consider that the artery in Amia is developed from 

 such a system of arteries. In Ophiodon, this single artery of Amia 

 is represented by two arteries called by Allen (1905) the opercular 

 and dorsal branchial muscle arteries, neither of which arteries, however, 

 sends a posterior efferent branch downward in either of the arches. 

 In Lopholatilus, Silvestrr (1904) here shows but a single artery 

 called by him a "small muscular branch". In both Esox and Gadus, 

 as already described, the corresponding artery has its origin from 

 the lateral dorsal aorta, instead of from the second efferent branchial 

 artery, and no branches were found running downward in the branchial 

 arches. 



The hyo-opercularis, in the one adult Amia here under con- 

 sideration, has its origin from the lateral dorsal aorta opposite or 

 shghtly posterior to the point where that aorta is joined by the dorsal 



