360 Proceedings of the Royal Physical Society. 



lower margin, extends higher up, and comes to be opposite a 

 considerable portion of the origin of the pectoral fin. There 

 is still another plate (Fig. 3, p.cl.), though it is a very small 

 one, appended to the posterior margin of the clavicle, and 

 this we may recognise as corresponding to the post-clavicidar 

 of Polyodon and the Palseoniscidse. 



I have seen no evidence of ossification in the scapulo- 

 coracoid cartilage, nor does any specimen afford a view of the 

 base of the pectoral fin, though in some a few dislocated 

 radials may be seen in this position. It is probable that the 

 arrangements here resembled those in Aciijenser, and I own 

 I am somewhat at a loss to understand the " two bones " to 

 which Mr Davis alludes as "apparently representing the 

 radius and ulna of Owen (coracoid and scapula of Parker)." 



The limits of the present paper hardly permit my entering 

 into detail as to the rest of the structure of Chondrosteus. 

 So far as the internal skeleton is concerned, its remains indi- 

 cate a structure very similar to that in Acipenser, while the 

 fins in shape and arrangement much resemble those of 

 Polyodon. 



Conclusion. — Although there is no evidence of any long 

 snout, Chondrosteus resembled Polyodon in the general shape 

 of the body, in the form and arrangement of the fins, and, 

 above all, in the absence of sccdes on any part, save the pro- 

 longation of the body-axis along the upper lobe of the very 

 heterocercal tail. In other respects its affinities are more 

 with Acipenser. 



How like the corresponding parts in Acipenser are the 

 suborbital bone, the edentulous jaws, the jugal bone, and 

 indeed the palatal apparatus, though that has also its own 

 peculiarities ; while it seems highly probable that the mouth 

 was protrusible as in the living Sturgeon. 



But where the resemblances to Acipenser become weaker, 

 they come to point in another direction, namely, that of 

 Palceoniscus, and of course through the Palyeoniscidse to the 

 truly " teleosteoid " Ganoids. This is in the first place well 

 seen in the cranial shield (Pig. 1), where the parietals and 

 frontals are mesially in contact with each other for their 

 whole length, where there is a well-marked supra-temporal 



