On certain features of the lateral canals and cranial bones of Polyodon folium. ß71 



the roof of the mouth cavity itself. It is thus evident either that the 

 mouth cavities in the two animals, or the two bones here under con- 

 sideration, can not be homologous; and this applies to the vomer 

 bones of the two animals as well as to the so-called anterior para- 

 sphenoids. In the region where the vomers and the anterior para- 

 sphenoids articulate by suture with each other there are, on the 

 ventral surface of the snout, two little tentacles, one on each side. 

 1'hese little tentacles, which have not been heretofore described so far 

 as I can find in the literature at my disposal, must correspond to the 

 larger tentacles of the sturgeon, and would mark the anterior limit 

 of the region included in the roof of the mouth of teleosts, if they 

 mark that limit in the sturgeon, as I was led, in an earlier work 

 (No. 3, p. 274), to suggest. 



The prenasal portion of the main infraorbital lateral canal, which 

 CoLLiNGE says is a continuous canal uniting around the anterior border 

 of the snout with its fellow of the opposite side, I could not satis- 

 factorily trace. Only two of the snouts of my specimens had been 

 preserved, and in the two dissections that I made the canal apparently 

 came to the surface and there ended "at about the place where Col- 

 LiNGE says that it "diverges laterally". Anterior to this point other 

 canals were found, but I could not be at all sure that any of them 

 were lateral sensory canals. I accordingly had a considerable portion 

 of the snout of my small specimen prepared for sectioning, and 

 sectionized by my assistant, Mr. G. E. Nicholls, but the material 

 was found to be in such an unsuitable condition that but a few 

 sections were cut. These few sections, however, established the fact 

 that there is a cross- commissural lateral canal near the anterior end 

 of the snout of the fish, and that it traverses the cartilage of the 

 end of the rostrum instead of passing around its anterior border as 

 CoLLiNGE states. In the section of canal here enclosed in cartilage 

 there are apparently two sense organs on each side, for two nerves, 

 on each side, run backward from the transverse canal in canals 

 in the cartilage. The two mesial ones of these four canals, one 

 on each side, first unite to form a median canal, and then sepa- 

 rate and open separately on the ventral surface of the cartilage. 

 Anterior to the commissure an anterior continuation of the canal 

 of either side, or the trunk of a dendritic system, whichever it may 

 be, continues forward in a deep slit in the lateral edge of the carti- 

 lage, the slit being several times deeper than the diameter of the 

 canal. Tliis canal or tube, and also the transverse commissural canal. 



