330 



its normal position, which must be the one shown in Necturus; and 

 this might be due to the fact that the dorsal trabecular bar, Gaupp's 

 "dorsale Randspange", is of relatively much later development in Rana 

 than in Necturus. In Amia this little pillar of Rana has apparently 

 entirely disappeared, the nervus oculomotorius and radix profundi 

 issuing, in that fish, by a single large opening that is separated from 

 the two trigeminal foramina by the pedicle of the alisphenoid. 



Turning now to Bdellostoma, I find, in a very badly preserved 

 embryo that is marked as 12 mm in length, that the still procarti- 

 laginous pharyngeal basket is nowhere connected with the trabecula, 

 the dorsal edge of the basket forming a short bar parallel with the 

 trabecula. And this separation of the trabecula from the pharyngeal 

 basket in embryos of Bdellostoma, is definitely indicated in Myxine, 

 even in the adult; for Parker (22) there shows the "hard cartilage" 

 of the trabecula separated from the similar cartilage of the pharyngeal 

 basket by three processes of so-called "soft cartilage". In my 12 mm 

 Bdellostoma the maxillo-mandibularis trigemini and hyomandibularis 

 facialis nerves run downward between the two parallel procartilaginous 

 bars, these bars thus, in this embryo of Bdellostoma, corresponding 

 exactly, in their relations to these nerves, with the ventral and dorsal 

 trabecular bars of the embryos of Necturus. The two bars in Bdello- 

 stoma must, accordingly, form part of the skull of that fish, as the 

 corresponding bars unquestionably do in Necturus; for the relations 

 of the bars, in Bdellostoma, to the trigeminus and facialis nerves is 

 certainly of much greater morphological significance than the fact that 

 one of the bars, in Bdellostoma, lies lateral or even ventro-lateral to 

 the trabecula, while in Necturus it lies dorsal to that cartilage. 



The bar of cartilage that forms, in Bdellostoma, the lateral 

 boundary of fenestra 1 and fenestra 2 thus being probably a portion of 

 the skull, there remains of the pharyngeal basket only the cartilage 

 that forms the anterior, ventral, and posterior boundaries to fenestra s , 

 and the superior and inferior lateral cartilages of Ayers' and Jackson's 

 descriptions. Leaving these latter cartilages out of consideration, the 

 remainder of the cartilage might be, in so far as its relations to the 

 nerves are concerned, either a quadrate, or a quadrato-hyomandibulo- 

 symplectic: and there are here three suppositions tobe made. Under 

 either supposition a part of the cartilage may be simply a crista 

 parotica. 



The first supposition is that the cartilage is wholly a quadrate. 

 In that case it can only find its homologue in the articular part of 

 the quadrate of other fishes plus certain of those processes of the 



