1887.] PAIRED FINS OF CERATODUS. 9 



described by Huxley (19, p. 49), to which Schneider likens that of 

 the pelvic fin, is constant in its relationships and invariably post- 

 axial. I emphatically deny that structural similarity of the second 

 mesomere of the fore and hind fins suggested by him, while I desire 

 to lodge a protest against the unqualified assertion that (23, p. 523) 

 " das Problem der Entwickeiung von Arm und Bein, welches gegen- 

 wartig so vielfach behandelt worden ist, wird dadurch .... seiner 

 Losuug einen Schritt naher gefiihrt." 



The great variation here demonstrated in the relative number 

 and calibre of the parameres of opposite sides of the normal pelvic 

 fin at least shows that the numerical differences existing between 

 them and those of the so-called irregular fin described at the outset 

 are insignificant. What now of the " branching," to which attention 

 was originally directed by Haswell (15, p. 7) ? In the fin furnished by 

 him all the rays not indicated in the drawing (fig. 1) are simple and 

 unbranched, though somewhat unusually elongated. Many of them 

 are transversely segmented. The question resolves itself into this — 

 Can the irregularities represented in fig. 1 as it stands be shown to 

 exist in a more normal fin ? Bifurcation of the terminal portion of 

 one or more parameres is no exceptional feature. Giinther (14) 

 and Davidoff (7) have both described it for the pelvic fin, and I 

 figure an example (fig. 7) in which it had attained a marked develop- 

 ment. Fig. 5 shows that it is no new peculiarity for the pectoral 

 fin also'. I have seen a dichotoiny of the pectoral paramere in one 

 other case, and that in a fin in all other respects normal. The trans- 

 verse segmentation of the axis of Haswell's fin (fig. 1) is not a whit 

 more remarkable tliau that of fig. 7 ; while in the fin there repre- 

 sented, as in the pectoral one of fig. 5, irregularities of the preaxial 

 parameres existed which far exceed in abnormality (if such it may 

 be termed) anything forthcoming in the first-named specimen. 

 Briefly stated, Haswell's fin differs most conspicuously from that 

 of the more constant type in respect to the longitudinal cleavage 

 of the axis. This phenomenon has already been recorded by 

 Haswell, and that in a fin which recalls the one here described 

 (15, pi. 1. fig. 6). Albrecht has figured and described (Sitzungsb. 

 d. koaig. preuss. Akad. Berlin, vol. xxxii. p. 545, 1886) a specimen 

 ol Protopterus (P. annectens) in which the distal half of the axis of 

 the left ])ectoral fin had similarly bifurcated ". 



Haswell (15, p. 8), commenting upon the " branching" process 

 which he first described, asserts the belief that " it is reasonable to 



^ I found, ou examiniug this specimen minutely, that many of the parameres 

 terminated in small nodules such as are represented at *. On comparison with 

 the other specimens dissected by me, I am convinced that similar terminal seg- 

 ments existed in two cases, but that, owing to their delicate nature, they had 

 been for the most part torn away in the process of dissection. The free ends of 

 the rays from which they had been thus removed presented a characteristic 

 truncated appearance, identical with that represented in some of the rays so 

 carefully dravyn by Davidoff (1). Putting all together, I incline to the belief 

 that the terminal nodules in question are of tairly general occurrence. 



^ The deductions which he has drawn from the study of this fin appear to me 

 no less unwan-antable than those of Schneider alluded to above. 



