236 J. K. Thacher — Ventral Fins of Ganoids. 



sends oif this process is thick and heavy. It is, moreover, a real pro- 

 cess, a continuation of the cartilage of the ray, and not even articu- 

 lated with it. The processes are highest in the middle of the fin and 

 diminish in size aborad until they hardly appear on the last two or 

 three rays. Toward the orad exti-emity of the fin they remain very 

 well developed, though not quite so long as they are in the middle of 

 the fin. One rises at the point marked a fig. 3, PI. I, and others from 

 the plate b d at points which would have been the orad sides of the 

 distal ends of the proximal segments 2, 3, 4 and 5 had the proximal 

 segments of those rays not united more or less completely to form the 

 large plate h d. 



If now we compare this fin of Polyodon with that of Acijyenser and 

 Scaphirhynchus, we find that in each there is a process growing 

 dorsad from the place marked a in Polyodon^ PI. I, fig. 3, and the 

 corresponding points in Acipenser and Scaphirhynchus, PI. 1, figs. 1 

 and 2, PI. II, fig. 3, a. There is no difference between the two pro- 

 cesses except that of size. The correspondence between the two is so 

 exact that had the species been very different from one another there 

 could have been no doubt of the homology of the process in the one 

 with that in the other, but this is made doubly sure by the fact of 

 the close similarity of Polyodon and the Acipenserids. 



Again, there can be no doubt of the correspondence, homodynamism, 

 of the first process in Polyodon and those that belong to the other 

 rods. They all spring from exactly the same place and differ in size 

 only. We have in Polyodon a foot with thirteen toes lohere each toe 

 has its own separate ilium. 



It has thus become apparent that the iliac process in the Acipen- 

 serids is not a vestige of anything more complex and important, but 

 simply what it was stated to be, a process from the ray, and that the 

 orad part of the fin does not include anything essentially distinct 

 from what appears in the hinder pait. We are thus brought to a 

 choice between a series of simple and of forked rays as that from which 

 the Gnathostome limb has been derived. There does not, however, 

 seem to be any room for doubt as to which of these the original form 

 was. That the simpler structure should have been the first produced 

 and that the more complex should have been developed out of this is 

 certainly more consonant w^ith our general experience of organic 

 change than the reverse sequence would have been. Moreover, I 

 have proved that in the similar median fins the original skeletal ele- 

 ments were perfectly simple rods developed in the freely flapping 

 membrane of the fin, and the analogical inference from these to the 



