158 Edward Phelps Allis jr., 



extension being longer than the posterior one. At the hind end of 

 this pouch a round and median pore opens on the outer surface, the 

 pore lying almost directly dorsal to the hind end of the short spina 

 occipitalis of the skull. 



The lateral end of the commissure is directly continuous with 

 the dorsal end of the preoperculo-mandibular canal, and at the point 

 where the two canals unite the lateral canal of the body begins, and 

 from there runs backward in a straight line as far as it was traced, 

 which was only to the level of the end of the pectoral fin. In this 

 anterior part of the length of the canal it communicates with the 

 outer surface, at regular intervals, by short primary tubes directed 

 ventro-laterally and each ending in a small round pore. The anterior 

 tube and pore of the line have fused with the terminal tubes and 

 pores of the infraorbital, supratemporal, and preopercular canals, and 

 are represented in the large blind pouch arising near the dorsal end 

 of the preopercular canal, as already several times stated. Between 

 this pouch and the next succeeding primary tube of the line, and also 

 between every two succeeding primary tubes of the line, so far as it 

 was examined, there is a sense organ, wholly or partly enclosed in a short, 

 cylindrical tube or scale of bone. These organs are all innervated by 

 relatively long branches of the linea lateralis vagi, the two organs of 

 the supratemporal commissure being innervated by an independent nerve, 

 probably a branch of the linea lateralis which issues from the skull by 

 a foramen distinct from that by which the latter nerve has its exit 



The lateral canal of the body of Conger, as thus above described, 

 includes all that part of the main horizontal sensory canal of the fish 

 that is innervated by the nervus lineae lateralis vagi. This canal, in 

 Conger, is not a direct posterior continuation of the main infraorbital 

 canal, as has been already stated in describing the latter canal. It 

 seems to be, much more, a continuation, at an angle, of the supra- 

 temporal commissure. Conger agreeing in this with Chimaera (No. 10), 

 and Mustelus (No. 7). In Laemargus (No. 14), on the contrary, the 

 main lateral canal of the body extends a short distance anterior to 

 the supratemporal commissure, as it does also in the diagram that 

 Ewart gives (No. 14, fig. 2) for selachians in general. The anterior 



