484 Edward Phelps Allis jr., 



border of the rostrum, and near the anterior end of the rostrum ana- 

 stomoses with the suborbital branch. This certainly is not true of 

 either of my two specimens. 



What is apparently the preopercular canal of the fish is represen- 

 ted by a short canal that traverses four flat bones that lie in a dorso- 

 ventral line on the cheek and the ventral surface of the head, about 

 one third of the distance backward between the postoibital and oper- 

 cular bones. This canal has never heretofore been described, and, as 

 just stated, is would seem to be the preopercular canal, but it is to be 

 noted that its position corresponds more to that of the vertical cheek 

 line of pit organ, in Amia, than to that of the preopercular canal. If 

 it be the preopercular canal, and as such it seems best to consider it, 

 the bones it traverses must represent the preoperculum of the fish. Its 

 innervation could not be determined in either of the two dis- 

 sections made. 



Still one more canal is found on each side of the head of the 

 fish, and although of somewhat smaller caliber than the canals above 

 described, it would nevertheless seem certainly to be a latero-sensory 

 canal. It is a short transverse canal that lies wholly in the parietal 

 bone, and occupies only about two fifths of the widtli of that bone. 

 Its innervation could not be positively determined, but it is apparently 

 innervated by a delicate branch of the nerve that innervates the 

 extrascapular sense organs. Its only possible homologue in Amia and 

 teleosts is, accordingly, the middle head line of pit organs; this pit 

 line of these latter fishes thus being, in Acipenser, enclosed in a canal, 

 and its relations to the parietal well established. 



In addition to the latero-sensory lines above described, there are, 

 in Acipenser, as is well known, numerous patches or groups of what 

 are called "nerve-sacs". The more evident of these groups are shown 

 in the figures, but no attempt was made to represent, or even determine 

 all of them. The groups on the ventral and lateral surfaces of the head 

 are small, but on the dorsal surface of the head there are four large groups 

 on each side, and, along the dorso-lateral border of the snout, one long 

 series of groups; all of which form prominent lines that occupy marked 

 depressions in the dermal bones. A certain part of most of these de- 



