The Lateral Sensory System in the Muraenidae. 159 



portion of this lateral canal of selachians, and also the supratemporal 



commissure, become, in bony fishes, enclosed in bones that are gene- 



j rally supposed to. become incorporated in the skull of higher verte- 



1 brates. To determine whether this be so or not, forms no serious part 



of this investigation, but it may be noted that Sagemehl (No. 24, 



I pp. 48 and 107) considered them as parts of the secondary shoulder 



* girdle. Tliis I have already once had occasion to refer to (No. 4, 



i p. 365), and I have also once had occasion to quite fully discuss the 



bones of this region (No. 3). I now wish to call attention to certain 



further irregularities and peculiarities of these bones. 



In Amia (No. 1) the supratemporal cross-commissure lies in the 

 extrascapular bone of Sagemehl's descriptions, that bone being the 

 supratemporal bone of English authors. The bone lodges four sense 

 organs, one of which lies in the main lateial canal directly opposite 

 the lateral end of the commissute. The four organs are innervated 

 by branches of the supratemporal branch of the nervus lineae lateralis 

 vagi, that nerve also innervating one sense organ that lies in the 

 suprascapular, or posttemporal bone of the fish. Anterior to the extra- 

 scapular there is, in the hind poition of the squamosal, a sense 

 oigan innervated by a lateral sensory nerve that issues from the 

 cianium with the nervus glossopharyngeus. The pieopercular lateral 

 canal joins the main horizontal canal at the anterior end of the 

 glossopharyngeal section of that canal, between it and the hind end 

 of the otic section of the canal. A transverse line of surface pit 

 organs, my middle head line, continues mesially the line of the dorsal 

 end of the preopercular canal. 



In Scomber (No. 9) the arrangement is the same as in Amia, 

 excepting that the posterior organ in the squamosal canal of the fish 

 is innervated by a branch of the supratemporal branch of the linea 

 lateralis vagi instead of by a branch of the glossopharyngeus. 



In Polypterus (No. 5) the arrangement is practically the same as 

 in Amia, if the innervation given by Pollard (No. 22) is correct, ex- 

 cepting that the preopercular canal does not join the main horizontal 

 canal and that the suprateraporal bone is represented by three bones 

 instead of by a single one. 



