1885.] OF THE HUMAN SPHENOID BONK. 579 



After the trabeculEe have coalesced, growth occurs in tliem in 

 three directions: — 



1. Laterally, to form the side walls of the skull anterior to the 

 periotic capsules. 



2. Mesially, to fill up the floor of the pituitary fossa. 



3. Forwards, to form the ethmo-vomeriue and fronto-nasal 

 plates. 



Although the trabeculce at a very early period form a floor to the 

 pituitary fossa, yet this fossa is never completely shut off from the 

 pharynx in the chondro-cranium ; a small opening persists for a 

 very long time, the consideration of which leads to some very inter- 

 esting conclusions, and which renders necessary the study of the 

 early stages of the formation of the mouth and pharynx. It has 

 been very satisfactorily proved that the buccal mucous membrane 

 is derived from the epiblast ; the process by which this derivation 

 occurs is usually described as a tucking-in of the epiblast ; but 

 in reality it is a necessary outcome of the primary cranial flexuie. 

 Passing between the open arms of the trabeculse is a narrow 

 tubular portion of the anterior primary encephalic vesicle, know^i as 

 the infundibulum ; this diverticulum from the primitive brain comes 

 into contact wilh the buccal epiblast ; the meeting point of the two 

 structures is represented by the pituitary body. Tliis disposition of 

 the parts has long been known. 



Whilst engaged working over the development of this complex 

 region, I found that even at the mid period of intra-uterine life of 

 the human embryo a narrow cavity may be detected passing frum the 

 pharynx througii the basisphenoid, so as to come into close relation 

 with the infundibulum. 



The point of communication with the pharynx is in the middle 

 line in contact with the basisphenoid, the spot being indicated at 

 birth by a recess in the mucous membrane known as the bursa pha- 

 ryngea (see fig. 2, p. 580). After birth the canal suffers obliteration ; 

 but a band of fibrous tissue, passing from the pituitary body to the 

 pharynx, represents the original position of the canal. When the 

 bone is macerated the fibrous tissue disappears, leaving a hole in the 

 basisphenoid, termed by Landzert the canalis cranio-pbaryngeus 

 (Petersburger med. Zeitschrift, Bd. xiv.). This cranio-pharyngeal 

 canal may be detected in the floor of the sella turcica in very many 

 mammals at birth. 



Up to the present time I have been unable to assure myself that the 

 lower end of the infundibulum ever opens into the pharynx, but 

 there is every probability that such is the case. However, the exist- 

 ence of this diverticulum of the first encephalic vesicle raises an 

 exceedingly interesting question. It will be remembered that the 

 central canal of the spinal cord at its caudal end is brought into 

 relation in the early embryo of very many of the Vertebrata with 

 the hind gut, by a narrow and, in most cases, very temporary passage 

 known as the neurenteric canal, which passes around the caudal end 

 of the notochord, but afterwards becomes obliterated. 



Turning now to the cephalic end of the notochord, we find the 



