MUSTELUS LiEVlS. 199 



overcuine if one accepts Hoffmann's conclusion, already stated, 

 that the ramus buccalis facialis is a ramus ventralis and not 

 a ramus dorsalis. While I can express no opinion as to this, 

 it seems to me highly probable that the ramus buccalis may 

 be a serial homologue of the ramus mandibularis externus 

 facialis, and if it be accepted as such it would seem to lead 

 legitimately to the conclusion that the separation of the 

 buccalis lateral canal of fishes, by its innervation, into three 

 somewhat separate sections, is due to the fact that each 

 one of these sections has a descending course, one in relation 

 to the mandibular arch and the other two in similar relations 

 to the next two anterior arches. The preopercular canal of 

 ganoids and teleosts would then be a similar descending line 

 related to the hyoid arch, and the descending line of epithelial 

 pits described by Alcock (1, p. 145) in the glossopharyngeal 

 arch of one specimen of Annnoca^tes would be a similar line 

 related to that arch. The several ventral pit lines of Ammo- 

 coetes would then represent lines corresponding, in the several 

 arches to which they are related, to the mandibular part of 

 the canal line of the hyoid arch of Amia and teleosts, these 

 lines not existing in the premandibular arches. 



One of the branches of the ramus pretrematicus facialis of 

 Mustelus runs backward and then outward posterior to the 

 dorso-mesial diverticulum of the spiracular cleft, and there 

 turns downward posterior to the latter cleft. The other two 

 branches, and also the so-called chorda tympani, lie anatomi- 

 cally anterior to the dorso-mesial diverticulum of the sjnracu- 

 lar cleft. If, then, that diverticulum represents a remnant of 

 the mandibular cleft, the nerves here concerned must actually 

 lie anterior to that cleft. How they could have acquired that 

 position, if they are regular pretrematic branches of the nerve 

 of the hyoid arch, it seems difficult to imagine, unless it be 

 assumed that the persistent remnants of the mandibular and 

 more anterior clefts represent only the outer, external portions 

 of the clefts concerned, the fusion of the clefts with the hyoid 

 cleft, and with each other, taking place at their external ends 

 only, the internal ends wholly aborting. Hoffmann's descri])- 



VOL. 45, PART 2, — NKW SERIES. V 



