134 Edward Phelps Allis jr., 



point of which it anastomoses with the lateral end of the supra- 

 temporal commissure. The canal then traverses another similar ossicle, 

 near the middle point of which a blind primary pouch has its orig:in, 

 and then traverses in succession two other tubular structures similar 

 to the preceeding ones but less completely ossified. At the hind end 

 of each of these two latter tubular ossicles there is a primary tube 

 opening on the outer surface by a single pore, and the canal ends at 

 the posterior one of these two pores. Whether it recommences farther 

 posteriorly, or not, could not be determined, as all of my specimens had 

 been cut off shortly posterioi- to this point. The supratemporal com- 

 missure arises, as stated above, from the main infraorbital canal as that 

 canal traverses a tubular ossicle that lies wholly posterior to the point 

 where the main infraorbital canal anastomoses with the preoperculo- 

 mandibular canal. Eunning mesially, it traverses in succession two deli- 

 cate tubular structures, partly ossified, and then unites in the middle line 

 with its fellow of the opposite side. Between the two incipient ossicles 

 of either side, and also in the middle line, there is a slight swelling of 

 the canal which represents an aborted primary tube. A small blind 

 pouch, which represents another and less completely aborted primary 

 tube, arises at the point where the commissure has its origin from 

 the main infraorbital. The two incipient ossicles that enclose the 

 supratemporal commissure on each side, and the one that encloses that 

 short section of the main infraorbital canal from which the commissure 

 has its origin, thus correspond, in their relations to the lateral canals, 

 to the single extrascapular bone of Amia. Posterior to this bone, 

 thus represented, the posterior continuation of the main infraorbital 

 canal is enclosed in three separate tubular structures, more or less 

 ossified, each one of which probably lodges a single sense organ of 

 the line, though this could not be established in the dissections. This 

 post-temporal part of the canal of Muraena thus apparently corresponds 

 exactly with that part of the main infraorbital canal of my des- 

 criptions of i\-mia that is enclosed in the suprascapular and supra- 

 clavicular bones; there thus apparently being no canal in Muraena that 

 corresponds to the lateral line canal of the body of Amia. The tubular 

 ossicles that enclose, in Muraena, the supratemporal commissure and 

 this postcranial part of the main infraorbital canal are all more or less 



