1000 REPORT OF COMMISSIONER OF FISH AND FISHERIES. [20] 



meut.* An investigatiou of the structure of the caudal flu of Anguilla, 

 Fig. 4, PI. IV, very young individuals being used for the purpose, sbows 

 that this view of the structure of the tail of fishes is probably the true 

 one, ibr reasons which we will now present. 



In the first i)lace, the Eel seemed to us to be especially well suited to 

 clear up any doubts upon this question, for the reason that, if there was 

 any actual serial homology to be found throughout the extent of the 

 unpaired fins, such a condition ought to be apparent in this type, in 

 which the dorsal, anal, and caudal are serially confluent, and the mode 

 of transition from the one to the other ought consequently to be appar- 

 ent. This, I am glad to say, was found to be the case, as had been an- 

 ticipated on morphological grounds. 



An exact drawing of the relation of the last interspinous pieces of 

 the dorsal and anal fins shows that these pieces are not confluent with 

 the neural and hiemal spines, but the first true hypural process is 

 sharply bent backward at a point which is serially on the same level 

 with the break in the continuity between the hasmal arches and every 

 alternate interhiemal piece of the anal in front of it, as seen in Fig, 

 4, PI. IV. A slight flexure of the same kind is visible in the next hy- 

 pural piece. From this circumstance I infer that the hypural pieces 

 are not simply hiemal, as held by Balfour, but are, as held by Huxley^ 

 comj)osed of at least both hjcmal and interhtemal elemeuts, and that 

 the point where the abrupt flexure of the first hypural piece is found 

 marks the point where there was formerly a separation between the 

 hjBmal spines and the interluiimal pieces. That the distal i)arts of the 

 hypural pieces of the heterocercal part of the tail of the Eel are not 

 haemal is proved by their exact serial relation to the interspinous i)ieces 

 pertaining to the anal in front of them, the haimal arches being abbre- 

 viated so as not to extend as far out or so near to the margin of the 

 muscular mass, dorsally and ventrally, as does the terminal portion of 

 the hypural pieces. The coinplete separation of the interspinous carti- 

 lages from the haimal arches shows that the former cannot be medially 

 fused pleurapophj'ses, but must be hismapophyses and interspinous 

 elements together. Moreover, at no time can it be shown that the distal 

 parts of these hypural pieces in the Eel arise from the fusion in the 

 middle line of distinct cartilaginous rudiments homologous with ribs, 

 any more than it could be demonstrated that the interspinous pieces 

 were so formed. Of other young fishes and embryos the same may be 

 said ; yet I admit that, if the truth of the theory of caudal degeneration 

 and specialization which I have sought to establish is well founded, 

 it may be that the lateral halves of the median, proximal, neural, and 

 ha3mal elements fail to develop, for the reason that the process of de- 

 generation has affected their primitive mode of formation. Yet even 

 this, for evident reasons, would not necessarily vitiate the conclusions 



* Quar. Journ. Mic. Sci., 1859. 



