The Latero-Sensory Canals and Related Bones in Fishes. 479 



above referi'ed to, one on each side of the head, and the third a 

 median siipratemporal or supraoccipital. 



In 1894 Collinge [29] redescribed the canals in Acipenser, his 

 intention evidently being to give a complete and detailed description 

 of the latero-sensory system of the fish. But even this second descrip- 

 tion of this fish is incomplete in many important particulars, and as 

 it seemed to me also inaccurate in certain other particulars, I have 

 had two heads of Acipenser carefully dissected in my laboratory here. 

 In the first dissection the canals were successfully traced, and drawings 

 made of them, but the nerves innervating the canals were missed. 

 These were found in the second dissection, and have been incorporated 

 in the drawings of the first one. The figures of this fish (figs. 36 — 39) 

 are thus, to this extent, compilations. 



The latero-sensory canals of Acipenser are relatively small, and 

 they apparently belong to the plagiostomian rather than to the teleostean 

 type. Numerous tubes, many of them branching repeatedly, in large 

 specimens, lead to the outer surface and there open by very small 

 pores. These pores are not easily distinguished, in surface examinations, 

 and while the accompanying figures show the larger part of them, 

 certain of them are certainly lacking. The dendritic systems shown 

 in the figures are also certainly incomplete; for although the canals 

 were injected, the injection fluid certainly did not enter all of the 

 branches of the dendritic systems, and, furthermore, as the bones had 

 to be filed considerably to properly disclose the canals and branches, 

 many of the latter were filed off and so all record of them lost. This, 

 while it should be noted, is not important for the purposes of this 

 investigation. 



The main infraorbital canal apparently begins in the mid-dorsal 

 line of the anterior end of the snout, there being in communication 

 with its fellow of the opposite side of the head. The canal here lies 

 in dermal tissue only, but, running laterally, it soon enters and then 

 turns downward in a triangular or three-armed bone (fig. 36) that lies 

 on and embraces the dorsal corner of the lateral edge of the cartilaginous 

 rostrum. Two nerves entered this bone, in both of my two specimens, 

 and a third nerve entered the dermal part of the canal slightly beyond 



