The Latero-Sensory Canals and Related Bones in Fishes. 435 



Tlie preoperculo - mandibular canal has the course shown in the 

 fig'ures and needs no special description here. 



Hydrocyon hr-evis. The main infraorbital canal begins, in this 

 fish (Figs. 13 and 14), at a pore that lies immediately ventral to the 

 closely adjoining- anterior and posterior nasal apertures. Starting* from 

 this pore the canal encircles the anterior, ventral and posterior 

 aspects of the eye, traversing six circumorbital bones. The anterior 

 one of these six bones lodges three sense organs, the postero -dorsal 

 one two organs, and the other four bones one organ each. The 

 postero-dorsal bone is the strict homologue of the bone that I have 

 identified, in Macrodon, as the postfrontal bone of that fish. Leaving 

 this bone the canal traverses, in succession, the squamosal, extra- 

 scapular, suprascapular and supraclavicular bones, the three latter 

 bones lodging each a single sense organ of the line, and the squamosal 

 lodging two organs. As the canal passes from the postfrontal to the 

 squamosal, it anastomoses with the penultimate tube of the supra- 

 orbital line; and as it traverses the squamosal it anastomoses with 

 the dorsal end of the preopercular canal, this latter anastomosis 

 taking place between the two squamosal organs. 



The postfrontal bone of this fish, as in Macrodon, overlaps slightly 

 and is loosely bound to the dorsal postorbital bone, but this is much 

 less marked than in Macrodon. There is, in Hydrocyon, no bone ly- 

 ing posterior to, and continuing, as it were, the postfrontal, as in 

 Macrodon, but there is a dermal bone lying anterior to, and closely 

 connected with it. This latter bone would seem to be the homologue 

 of the prefrontal bone of the Stegocephali, and also of the so-called 

 supraorbital bone of Esox [49]. It is not found in Macrodon, and 

 must not be confounded with the "praefrontale" of Sagemehl's 

 descriptions. Whether it is the "Knochenlamelle" which Sagemehl 

 says [64^ p. 71] extends backward from the dorsal surface of 

 the "praefrontale", or not, I can not definitely tell, but it would 

 seem not. 



The supraorbital canal of Hydrocyon begins at a pore that lies 

 antero-mesial to the anterior nasal aperture, and from there runs al- 

 most directly backward through the nasal and frontal bones, and 



28* 



