J. K. Timelier — Ventral Fins of Ganoids. 237 



paired fins is a strong one, for it is something more than a mere 

 analogy, being supported by causal considerations. 



Ao-ain, the wide-spread absence of this dorsal process is inconsistent 

 with its constituting a part of each ray of the Frotognathostomi or 

 any of their ancestors. Although Chima^ra has a large iliac process 

 from the orad part of the fin, the Sharks proper, where we might ex- 

 pect to find the skeleton in no very specialized condition, have noth- 

 ing- which seems to rejiresent a dorsal process of any of the rays. 

 Neither do the Holostean Ganoids, nor most, not to say all, of the 

 Teleosts. Among the Batoidei we do have dorsal processes from 

 the pelvic cartilage. Their position renders their strict correspond- 

 ence with the iliac jDrocess of the Acipenser fin somewhat improbable, 

 while it suggests the possibility that they may represent some of the 

 processes of the aborad rays of the Polyodon fin. Even if this 

 should be the case the absence of these processes in the Sharks would 

 force us to regard the conformation in the Rays and in the Holo- 

 cephali as having arisen independently of that in Polydon. With 

 this possible exception and the exception of the Polydon ventrals, 

 no ventral fin or limb anywhere shows signs of any forking aborad 

 of the first ray. Nor is there any forking observed in pectoral and 

 median fins except so far as this may be represented in the scapula. 

 It is easier to believe that these comparatively few instances were 

 late, and some of them independent, developments, than to believe 

 that there has been an annihilation, so nearly complete, of dorsal 

 processes which were universal in the beginning. 



Thus it seems to me that the order of events, as far as regards the 

 Chondrostean Ganoids, was like this : The simple ray with the three 

 segments was the early form. An upgrowth of cartilage dorsad from 

 the proximal segment of the first ray took place in the Acipenserids. 

 In Polyodon this was imitated by the hinder rays. This is an 

 instance of the development of likeness out of unlikeness in liomo- 

 dynamous parts, a kind of change which must have frequently 

 occurred in the history of stocks. It is the opposite of diiferentiation 

 and it does not consist in mere reversion to primitive similaritv, but 

 in the attainment of similarity in any way. It is exemplified in a 

 department of phenomena resembling somewhat those of biology in 

 the formation of French plurals "by the addition of s, though the Latin 

 had formed the plurals of the same words in very diverse ways. 

 Thus, too, the close similarity between the fore and hind limb in 

 various forms now living, has doubtless in many cases been preceded 

 by a greater dissimilarity in the direct ancestral lines. 



