118 



FISHES. 



destined to remain permanently. The superior of these two 

 divisions, when both have become fully marked out, sur- 

 rounds the end of the up-turned chorda (fig. 492, a), and it 

 must be regarded as an embryonic structure, since it finally 

 disappears. The inferior of the two divisions, on the other 

 hand, is placed below the embryonic tail, and is ultimately 



Fig. 492. — Tail of young Flounder (Pteifroreecte) in its lieterocereal stage of development, 

 n, Embryonic caudal fin ; 6, Permanent caudal fln, occupying an inferior position ; c, Bent-up 

 end of the notocliord. (After A. Agassiz.) 



developed into the permanent tail. At first the permanent 

 caudal fin has the appearance of a distinct lobe, w^hich looks 

 like a second anal fin. In process of growth, however, the 

 embryonic caudal becomes thrown more and more upwards, 

 and the rays of the permanent caudal acquire a fan-like arrange- 

 ment. At the stage figured above (fig. 492), the tail is truly 

 " lieterocereal," and is wonderfully similar in appearance to 

 the tail of many Palaeozoic fishes. Finally, however, the 

 turned-up end of the notochord becomes replaced by the 

 long " urostyle ; " the embryonic caudal diminishes in size 

 and disappears ; and the permanent caudal increases in size, 

 and is gradually transformed from a ventral into a terminal 

 appendage, the tail-fin thus assuming its ultimate " homo- 

 cereal " constitution. 



The above-mentioned facts have an important bearing 

 upon certain paleeontological theories. As is well known. 

 Professor Louis Agassiz, one of the greatest of authorities 

 upon all su])jects relating to this class, maintained that the 

 Palaeozoic fishes in general were of an " embryonic " char- 



