6 ALLIS. [VOL. I. 



of each artery, and its relation to the clinoid bar, considered at 

 the time as unimportant, was said to be upward along the 

 anterior surface of the dorsal edge of the bar, internal to a 

 projecting, anterior, dorso-lateral corner of the bar, the artery 

 not again entering the cartilage at all. 



In Scomber, work being done in my laboratory here by Dr. 

 Dewitz shows the carotid artery on each side, running upward 

 along the anterior edge of the lateral wing of the so-called 

 basisphenoid of that fish. In Salmo it apparently has a similar 

 course (No. 20, PI. V, Fig. 7). The artery in Scomber and Salmo 

 thus has exactly the same relation to the basisphenoid bone 

 that it has in Amia to the cartilaginous, anterior clinoid bar. 



Why, then, should the cartilaginous, anterior clinoid bar of 

 Amia not be the unossified homologue of the basisphenoid 

 bone of Scomber and other teleosts ? It fulfills exactly all 

 the conditions required; the otherwise perplexing difference 

 in the relations of the internal carotids to the so-called basi- 

 sphenoids in Amia and Scomber are naturally explained; and 

 Amia would not differ from all other ganoids in possessing 

 osseous rudiments of a basisphenoid, as Bridge was forced to 

 assert (No. 7, p. 614). If such be the case the homologues 

 of the two bones that cap the bar in Amia must be looked for 

 elsewhere than in the teleostean basisphenoid. I consider 

 each of them as a part of the orbitosphenoid of its side of the 

 head, ossified from an independent center and not fused with 

 the rest of the bone because of the great development of the 

 optic fenestra. In support of this proposition it is to be noted 

 that, in their general position, and especially in their relations 

 to the internal carotid arteries, the two bones of Amia agree 

 closely with the anterior clinoid processes of the orbitosphenoid 

 part of the sphenoid bone of man; and that the orbito- 

 sphenoid in man must necessarily ossify from two different 

 centers in those cases in which the presphenoid bone develops, 

 as Thane says it sometimes does, independently of the pair of 

 nuclei that first appear on the inner sides of the optic foramina 

 (No. 24, vol. ii, pt. i, p. 76). 



In fishes I can find no recorded instance of such a develop- 

 ment of the orbitosphenoid from two centers, nor of a second- 



