140 HAY. [VOL. I. 



the tardily developing remains of a more complete sheath of 

 this substance, which must have enveloped the notochord of 

 the ancestors of this interesting fish. 



This cartilage is far from occupying the position assigned by 

 Gadow and Abbott to the outgrowths of cartilage described by 

 them, and far from having the thickness of those outgrowths. 

 Nevertheless, in sections taken near the ends of the vertebrae, 

 where, on account of the flaring of the centrum, the cartilage 

 is struck very obliquely, it has the deceptive appearance of 

 being very thick. Furthermore, in such sections, it may have 

 the still deceptive appearance of being mixed up with the dense 

 connective tissue there found. That such sections will explain 

 the figures presented by Gadow and Abbott I am far from 

 affirming; but I can hardly resist the conclusion that the layer 

 of cartilage which I have described has something to do with 

 the outgrowths which they have described and figured. 



Figure I given on page 204 of their paper presents a fa- 

 miliar appearance, although the notochordal sheath appears un- 

 usually thin. The section has evidently been taken near the 

 anterior end of the centrum. The bone there represented 

 belongs to the early formed layer which I have mentioned. 

 Between this bone and the elastica externa is seen a layer of 

 cells which the authors have not referred to. These belong to 

 the layer of cartilage of which I have previously spoken. Below 

 the bone is a collection of cartilage cells marked C. C. and 

 said to belong to the cartilage which surrounds the notochord. 

 It is nothing, however, but the anterior ends of the "aortal 

 supports," the interventrals of Gadow and Abbott. They 

 appear to lie below the bone because the cartilage has grown 

 downward and forward under the first formed layer of bone. 

 In Fig. 2 the supports are still excluded from contact with the 

 notochord for the same reason ; but in a section very close 

 behind this the supports would touch the notochord and would 

 at the same time become confluent with the cartilage of the 

 lower arch. That is, neither the cartilage C. C. nor the " sup- 

 ports " /. V. are in reality separated by bone from contact 

 with the notochord. The relation of the "supports" to the 

 lower arches is shown in Fig. 9 of my former paper. 



