[17] THE OSTEOLOGY OF AMIA CALVA. 763 



Without going auy further into an explanation of these cartilaginoua 

 formations, which could only be done by a careful comparison of the 

 vertebral column of Amia with that of other fishes, I feel called upon 

 to invite attention to the remarkable — even in details — similarity of 

 the posterior portion of the basioccipital to the centrum of a vertebra. 



To make a comprehensive statement, the occiput of Amia calva re- 

 veals the elements of three vertebrae, which are co-ossified with it, 

 and whose individual independence becomes less and less marked from 

 behind forwards. The centrum of the hindmost vertebra, as well as 

 the centra of the other two, is co-ossified with the basioccipital; it is, 

 however, only in the posterior portion of this bone that the evident 

 likeness to the centrum of a vertebra can be recognized. The neural 

 arch of this vertebra cannot be distinguished from the neural arch of 

 a trunk- vertebra, and it possesses also a well-formed spinous process ; 

 the corresponding nerve is stamped with all the characteristics of a 

 typical spinal nerve. The middle vertebra, absorbed as it is by the 

 cranium, is quite similarly formed, only that its neural arch has be- 

 come broader and intimately blended with the cranium. The trans- 

 formation and co-ossification of the anterior vertebra is the most com 

 plete. Both halves of its neural arch are blended with the occipitalif 

 lateralia, and the nerve corresponding to it arises simply as a feeble 

 anterior root [ventral]. This rudimentary nerve is really the only safe 

 indication of the existence of this anterior vertebra, which has in other 

 respects been completely appropriated by the skull; and should one 

 imagine that this nerve was formed through a retrogressive process, or 

 became blended with the occipital nerve, then nothing would remain 

 to give us the slightest hint as to the original existence of this anterior 

 vertebra. This is of importance in so far as it gives rise to the possi- 

 bility that beyond this vertebra, the existence of which is still to be seen 

 through its last faint traces, there existed other ones, which, however, 

 have become thoroughly appropriated by the cranium so as not to be 

 any longer distinguishable. 



The number which I have indicated, then— that of three vertebrae co- 

 ossified with the skull— can therefore only be the fewest of these seg- 

 ments to be recognized. The view that the original number of these 

 vertebrae was greater is by no means to be precluded. 



It is hardly worth while mentioning that the facts just discussed by 

 me have nothing whatever to do with the question of the composition 

 of the primoidal cranium out of like constituents — the so-called verte- 

 bral theory of the skull. The formation of the primoidal cranium in 

 the Selachii — and maybe, too, in the Cyclostomata — has already been 

 perfectly defined; and setting the question entirely aside as to whether 

 any or how many metameres were contained in those skulls, my only 

 aim was to establish that between the Selachian skull and that of the 

 higher fishes no complete homology exists. The cranium of the iiigher 

 fishes corresponds to the cranium of the Selachii, plus several (at least 

 three) of the anterior vertebrae of the column. 



