331 



quadrate that always lie external and hence posterior to all of the 

 trigeminal nerves, between those nerves and the facialis. Those pro- 

 cesses are the orbital (proc. muscularis, Gaupp) and otic processes, 

 and the pedicle (proc. basalis, Gaupp). The pedicle can here probably 

 be left wholly out of account, for it should normally connect with the 

 skull ventral to the foramina of the nerves, and not dorsal to them, 

 as the cartilage in Bdellostoma does. 



A second supposition is that the posterior arm of the cartilage 

 is a part of the hyoid arch, and as such both Paeker, and Ayers 

 and Jackson consider it; Parker calling it the hyomandibulo-sym- 

 plectic, and Ayers and Jackson calling it the hyoid. The hyomandi- 

 bularis facialis lies wholly posterior to this cartilage. The cartilage 

 can not, accordingly, if the relation to this nerve is important, be the 

 homologue of any part of the cartilaginous hyoid arch of Elasmobranchs? 

 though it might be the homologue of the superior post-spiracular li- 

 gament (see 3). For the same reason it can only be the homologue 

 of the "prae-facialis" part of the hyomandibulo-symplectic of Teleosts. 

 The "post-facialis" part of the teleostean hyomandibulo-symplectic, 

 and the entire selachian hyomandibular, must evidently be sought for 

 either in the ceratohyal or in the first epibranchial of Parker's de- 

 scriptions of Bdellostoma; and it would seem as if it must be in the 

 first epibranchial. The facialis nerve of Bdellostoma issues, in fact, 

 through a large opening called by Parker the lateral fenestra 3 . This 

 fenestra is entirely closed by membrane (Müller), and if this mem- 

 brane were to chondrify, or if the cartilaginous bars surrounding the 

 fenestra were to approach and fuse, the hyomandibularis facialis would 

 be enclosed in a foramen that would traverse a cartilage resembling 

 exactly, in this, the teleostean hyomandibular. The ceratohyal (Parker), 

 by slipping downward along the hind edge of this cartilage, would then 

 acquire the teleostean position and relation to both nerve and cartilage. 



A third supposition is that the hyomandibulo-symplectic of Parker's 

 descriptions is, as under the second supposition, the „prae-facialis" part 

 of the hyomandibulo-symplectic, but that that cartilage belongs to the 

 mandibular and not to the hyoidean arch. That the entire teleostean 

 hyomandibulo-symplectic may belong to the mandibular arch has been 

 frequently asserted, Pollard (24), in particular, saying that it "must 

 be sought in the articular portion of the quadrate of Heptanchus, that 

 is, in the part proximal to the skull". 



The relations of the cartilages to the nerves seeming to give no 

 further indications as to the interpretation of the skull of Bdellostoma, 

 I have turned to the muscles of the adult, and to the branchial clefts 



