THE CCELACANTH1NI. 11 



especially in the caudal fin, the rays of which are supported by inter- 

 spinous bones ; an arrangement which, in other fishes, is found ordinarily 

 only in the anal and the dorsal [caudale in the text]. Lastly, the 

 vertebral column is prolonged, more or less distinctly, between the prin- 

 cipal lobes of the caudal fin, so as to form a median taper process." 



Professor Agassiz then adverts to the resemblances and differ- 

 ences between Coelacanthus and Macropoma, in a passage Avhich I 

 shall cite hereafter, and proceeds : 



" Taking into account the extraordinary development of the dentary 

 system in some genera of this family, and particularly in Holoptychius, 

 one is tempted to approximate the Ccelacauths to the Sauroids ; 

 whilst the dentition of the genus Undina, as it has been described by 

 Prof. Minister, would seem to establish a closer affinity with the Pycno- 

 donts. On the other hand, the scales exhibit peculiarities which are to 

 be met with in no other family, whence I have been disposed to arrange 

 the Coelacanths provisionally between the Sauroids and the Pycnodonts. 

 It may be, however, that their true place is near the Scleroderms, or the 

 Accipenseridce." 



Of the genus Ccelacanthus itself, Professor Agassiz remarks 

 (p. 170): 



" This genus, which I regard as the type of the family, was long known 

 to me only by fragments ; but these were so different from most other 

 ichthyolites, that I did not hesitate to form them into a distinct genus. 

 What especially struck me was the form and the structure of the fins, 

 their relation with the interspinous bones, and the manner in which the 

 apophyses [vertebral arches and spines] are united on the one hand with the 

 bodies of the vertebrae, and on the other with the interapophysial [inter- 

 spinous] bones. The apophyses divide at their bases into, two branches, 

 forming a fork, which embraces the body of the vertebra ; to this 

 apophysis succeeds an ossicle, which, instead of being interposed between 

 two apophyses, is fitted on to the end of one, so as to form its direct 

 prolongation. The ray properly so called, the longest of the three pieces, 

 Lb also forked at its base ; its extremity alone is jointed, but never bifur- 

 cated. These three pieces, the apophysis, the inter-apophysial bone, 



and the ray, are about equal in length, and are all three hollow 



This singular structure characterises most of the rays which lie at the 

 posterior part of the body ; now as, usually, only the anal and the 

 dorsal have inter-apophysial bones, I at first concluded that these two 

 fins must be excessively developed ; and what helped to strengthen me 

 in this idea, was the fact that the vertebral column appeared to be con- 

 tinued beyond the two azgyos fins, to form, further on, a bundle of very 

 small articulated rays, attached directly to the vertebra?. But Lord 

 Enniskillen's discovery of an entire specimen of this remarkable type 

 has completely modified my views. It now appears that besides the 

 fins of so exceptional a structure, which I regarded as anal and dorsal, 

 this fish has a very distinct normal anal and two dorsals. Now, unless 

 the existence of three dorsal and two anal fins of very different structure 

 — an arrangement which occurs in no known genus of fishes — be 

 admitted, it is necessary to regard the terminal fin of the body as a 

 caudal. For the rest, this is not the only known example of a caudal 

 supported by inter-apophysial bones, the caudal of Polypterus bichir 

 being supported by similar bones, at least its upper lobe. What is truly 

 exceptional is the prolongation of the tail beyond these rays, and the 

 little fascicle of articulated rays surrounding its extremity. In this 



