1034 REPORT OF COMMISSIONER OF FISH AND FISHERIES. [54] 



axis. The preservation of only a part of the marginal portion of the 

 median skeletogenous tract from which the hindmost iuterspinous sup- 

 ports for the caudal are developed is without a parallel, except Fieras- 

 fer, so far as the author is aware, amongst fishes. These supports 

 sometimes fail to develop, but the layer whence they are derived is pres- 

 ent, as in Chimara, for example. 



How far is the foregoing theory of the structure of the caudal flu of 

 Mola justified by what is known of the development of this fish? Since 

 we have satisfied ourselves that Ostracion hoops is merely a stage of the 

 development of Mola, it is clear that that form has already advanced 

 far beyond the condition of the larva which mnst be assumed to have 

 had some traces of a complete caudal axis, for it already exhibits a 

 condition approaching that of the Molaca7ithus form, with its two devel- 

 oped dorsal and anal fins, which have a slight interval between them. 



The youngest Molacantlms stage hitherto figured has the dorsal and 

 anal more approximated than the specimen figured by Putnam, or the 

 one in the United States National Museum, which would indicate that 

 a metamorphic process was in progress in the interval between 0. hoops 

 and Molacantlms. This view is supported by the fact that the fin-fold, 

 which appears in still older Molacanthij is very narrow or low, and the 

 fin rays which it includes are very short and feebly developed. In the 

 next stage. Fig. 5, this fin-fold between the dorsal and anal has be- 

 come higher, and the first indications of the production of the middle 

 lobe, such as is found in the next older stage, begin to be apparent. 

 It is thus made very evident that the development of the caudal in 

 these fishes is a progressive process, and that it is remarkable and even 

 unique in this one respect, viz, that all of the unpaired fins do not de- 

 velop simultaneously as is usually the case with young Teleosts. This 

 can only be explained on the theory already suggested, that the poste- 

 rior i^art of the primitive axis has been suppressed, so that in order to 

 develop a caudal at all a secondary and indirect mode of development 

 has been necessary, in the course of which the remnants of the anterior 

 median ventral and dorsal parts of the urosome still preserved have 

 supplied the materials lor the growth of the interspinous bones which 

 support the caudal rays, but these i)ortions of the primitive urosome 

 have been drawn forward and included between the anal and dorsal. 

 Then as the caudal grows out it manifests another singular tendency, 

 viz, to develop a certain number of soft rays in its central prolongation 

 which are subsequently suppressed in the adult fish, which is simply 

 additional evidence in favor of the belief that these rays belonged to a 

 dorsal and ventral series, at one time developed along the upper and 

 lower margins of the urosome of some ancestral form, but on account 

 of the suppression of the latter, have been carried toward the end of the 

 persistent part of the axis of the existing derivative form in excess of 

 the number which, are permitted to persist as permanent rays. 



