no MR. E. PHKLPS ALIilS, JUX,, OX 



trabecul;e and lying anterior to the sac, and two derived either 

 from the trabecular, ethmoidal, or sphenolateral cartilages and 

 lying posterior to the sac ; and the nasal capsule, when it 

 develops, also lies between and is more or less fused with these 

 processes. The dorsal trabecular process and the ethmoidal con- 

 stituent of the dorso-lateral antorbital process alwa^^s fuse with 

 each other above the nervus olfaetorius, so forming the dorsal 

 boundary of the fenestra olfactoria, and they also form, together 

 with the sphenolateral cartilage, the lateral wall of the cavum 

 pr8ecerebrale of my descriptions (AUis, 1913). The two ventral 

 processes may also fuse with each other at their outer ends, so 

 forming a perforated plate which lies beneath the nasal sac, the 

 perforation of this solum nasi being closed by membrane and 

 representing an unchoudrified portion of the nasal capsule. 



The ventro-lateral antorbital process and the two prenasal 

 processes of the Selachii are found in embryos of Ceratodus, the 

 former being called by Greil (1913) the processus antorbitalis and 

 the two latter the doi\so-lateral and the anterior or lateral trabe- 

 cular cornua. No dorso-lateml antorbital process is shown in 

 (ireil's figures of embryos of this fish, but it is shown by 

 Fiirbringer (1904) in a figure of the adult Ceratodus, the process 

 there apparentl}^ arising as a lateral process from the dorsal end 

 of the posterior wall of the nasal capsule, and being called by 

 Fiirbringer the preorbitai process. The septum nasi of Ceratodus 

 apparently lies dorsal to the trabeculpe, as it does in the Teleostei, 

 Holostei, and Orossopterygii, the septum in these several orders 

 of fishes accordingly not being the homologue of the septum in 

 the Plagiostomi, unless it be that the internasal prolongation of 

 the trabecular of embryos of the former fishes corresponds to the 

 ventral edge of the subethmoidal keel of embryos of the Plagio- 

 stomi and not to the rostral stalk and rostral plate, wliich seems 

 quite probable, 



C ALLORH YXCH us . 



In Callorhyncltus, ns in A canthias and Pristmnif!, the trabeculse 

 are fii'st laid down practically at right angles to the parachordals, 

 as consideration of Schauinsland's (1903) figures of a 60 mm. 

 embryo of this fish makes plainly evident. This is due, as in the 

 Selachii, to the marked cranial flexure at this period ; but in 

 C allorhijnchus the anterior portion of the brain, instead of 

 recurving upward and forward as if seeking to reacquire a posi- 

 tion in the line of the axis of the body, projects ventro-anteriorly 

 in a straight line. Related to this retention, in this position, of 

 this anterior portion of the brain, the eyeballs of the adult fish 

 lie dorso-anterior to it; but Avhether it is this position of the 

 brain that has forced the eyeballs to pass antero-dorsal to it, or 

 the precocious development of the eyeballs that has prevented the 

 brain from reeuin^ing upwai-d, is not apparent. 



The trabecul* fuse to form a wide trabecular plate which. 



