219 



sition of the thymus, somewhat posterior to, but otherwise corre- 

 sponding somewhat to that of the mandibular pseudobranch in Amia, 

 would suggest that the mandibular and hyoidean gills, one or both, 

 bad contributed largely to the formation of the organ. The organ is 

 penetrated, and apparently supplied by branches of a small artery 

 that arises from the efferent hyoidean artery at or immediately ventral 

 to the point where that artery joins the first efferent branchial artery; 

 this still further suggesting a hyoidean, or mandibulo-hyoidean origin. 

 Pollard does not show the thymus in any of the cross-sectional views 

 he gives, and in his Fig. 21, where it is shown, it is relatively much 

 shorter, antero-posteriorly, than in my specimen. 



From the point where the efferent hyoidean and first branchial 

 arteries unite, the carotid artery runs forward, as shown in Budgett's 

 figures, this artery in reality forming the anterior prolongation of the 

 lateral dorsal aorta of its side. The aorta, posterior to this point, 

 curves inward and enters a canal in the base of the skull, as stated 

 by the several authors. But, on both sides of my specimen, the aorta, 

 in this short section of its course, seemed quite definitely separated, 

 by a more or less complete horizontal partition , into dorsal and 

 ventral portions; the ventral portion seeming to communicate more 

 particularly with the efferent artery of the first branchial arch, and 

 the dorsal portion with the efferent hyoidean artery. And, although 

 there is nothing at all definite in these determinations, and the ar- 

 rangement may be without importance, it would nevertheless seem to 

 suggest that the two efferent arteries originally joined the dorsal 

 aorta independently; that they became connected by dorsal commissure, 

 as they actually do in the adult Ceratodus and Elasmobranch ; and 

 that this commissure was then pressed upward against, and finally 

 coalesced with the dorsal aorta ; the existing partition representing the 

 vanishing, coalesced walls of the two vessels. This assumed condition 

 of this part of the aorta is shown in the accompanying figure, which 

 is constructed on the same purely diagrammatic principles as those 

 illustrating my earlier work. 



The carotid artery, starting from its point of origin, runs forward 

 in a canal that lies partly in the substance of the greatly enlarged 

 basal portion of the ascending process of the parasphenoid and partly 

 between that process and the cartilaginous side wall of the skull, the 

 canal leading anteriorly into the hind end of the orbit. This is well 

 shown in Pollard's Figures 27 to 31, the carotid artery being called 

 by him the ophthalmic artery. Well within the large funnel-shaped 

 posterior portion of this canal of the adult, the median canal in the 



