The Latero-Sensory Canals and Related Bones in Fishes. 415 



tiires, there anastomosing- with what is probably the sixth primary 

 tube of the ethmoidal part of the main infraorbital, as already fully 

 described. This anastomosis of the recurrent anterior end of the 

 supraorbital canal with the main infraorbital, is a selachian Characte- 

 ristic, not found in any other ganoid, or in any teleost that I know 

 of. Starting from this point, the supraorbital canal first runs for- 

 ward, then curves mesially and backward and leaves the nasal at its 

 hind edge, immediately mesial to the posterior nasal aperture. Up to this 

 point the canal does not, in any part of its course, enter the dermal 

 ethmoid bone, the prenasal of Collinge, nor communicate, at any place, 

 excepting only at its anterior end, with the main infraorbital; the 

 anastomoses shown by CoUinge being in marked disaccord with what 

 I find. Having left the nasal bone the canal immediately anastomoses, 

 by the primary tube that arises between the nasal and the so-called 

 ethmo-nasal, with the anterior tube of the suborbital part of the 

 main infraorbital canal. It then runs directly backward through the 

 ethmo-nasal and frontal of Parker's descriptions, until it reaches the 

 level of the anterior edge of the orbit, where it turns laterally and 

 backward, traverses the remainder of the frontal, and, leaving that 

 bone, enters the anterior end of the squamosal and there soon anasto- 

 moses, by its posterior terminal tube, with the main infraorbital canal. 

 No indication of an anterior head line of pit organs, continuing the 

 supraorbital sensory line beyond this point, could be found in my 

 skeletons. At the level of the anterior edge of the eye, the canal 

 gives oft" what Collinge has called the preorbitai commissure, which 

 connects it with its fellow of the opposite side. This commisure is 

 an apparently complicated series of tubes, ''six distinct commissures" 

 according to Collinge, but they all certainly result simply from the 

 repeated subdivision and inter- anastomosis of two primary tubes, one 

 belonging to the canal of either side of the head. The commissure, 

 while certainly a commissural connection between the two canals, is 

 thus not a commissural canal, a distinction that should be borne 

 in mind. 



The ethmo-nasal bone of Parker's descriptions, both of larvae and 

 of the adult, is described as a splint bone that belongs to the dorsal surface 



