tALATOQUADRATE WITH NEUROCRANIUM IN CCELACA^THIDS. 55 



Axelia and Bijdocercides . The vena jugulaiis would then con- 

 tinue posteriorly and, having joined the nervus facialis, issue 

 witii it through the facialis foramen at the hind edge of the 

 basisphenoid. In this part of its course it must have lain in a 

 canal in the cranial wall, for that it perforated the dura mater and 

 entered the central cranial cavity seems wholly improbable. This 

 canal would thus, in this, exactly resemble the trigemino-facialis 

 chamber of recent fishes, and it seems wholly impossible that the 

 truncus maxillo-iiiandibularis trigemini did not run forward in 

 it, and issue from it between the process e and the alisphenoid. 

 If the truncus issued posterior to the process e, as suggested by 

 Stensio, it would be separated from its ophthalmicus lateralis 

 branch by the entire width of that process, which would be 

 wholly exceptional, as compared with recent fishes, whether the 

 process be a postorbital one or a basipterygoideus. The notch 

 at the base of the hind edge of the process e would then probably 

 transmit the ramus paliitinus facialis, which is not otherwise 

 accounted for. 



If the trigeminus and profundus nerves of Wimania and Axelia 

 had the courses that I have above ascribed to them, they must have 

 fii^st run upward along the posterior surface of the basisphenoid, 

 then have passed, on either side, lateral to that posteriorly project- 

 ing dorsal end of the basisphenoid that resembles a dorsum sellse, 

 and then have run forward along the dorsal surface of the body of 

 the basisphenoid, there either lying free in the cranial cavity or, 

 much more probably, being enclosed in the canal traversed by 

 the vena jugularis. In recent fishes these nerves run forward 

 across the ventral end of the hollow of the cephalic flexure, instead 

 of upward and forward over it. That part of the basisphenoid 

 bone of Wimania and Axelia that occupies the hollow of this 

 flexure could nob then have arisen from the conditions found in 

 recent fishes by the simple chondrification or ossification of tissues 

 that filled the hollow of the flexure, for the nerves would then 

 have been enclosed, in situ, in the cartilage or bone so formed 

 and have traversed canals or foramina in it. And it seems ei^ually 

 improbable that the bone could have grown upward in the hollow 

 of the flexure, pushing the nerves before it. It must then be 

 that, at a certain, stage in the development of embryos of 

 Wirnania and Axelia, the anterior end of the notochord was 

 bent upward and slightly backward, as it is in 33 mm. embryos 

 of the recent Acanthias (Goodrich, 1918, fig. 14, pi. 2), and 

 that, in the former fishes, this curvature was not later reduced. 

 The cartilago acrochordaiis, which, in recent fishes, develops in 

 tissues related to the anterior end of the notochord, was there- 

 fore, in Wimania and Axelia, directed dorso-posteriorly, and its 

 morphologically ventral surface, presented dorso-anteriorly, lay 

 directly beneath the laembrane that, in recent fishes, extends 

 from the anterior edge of the cartilago acrochordaiis to the top 

 of the preclinoid wail. The hind ends of the trabeculse, and the 

 bases of the alisphenoid cartilages, would accordingly be carried 



