356 A. H. Clark — Pentamerous Symmetry of Crinoidea. 



animals which possess two similar pairs of appendages like our 

 theoretical ancestral crinoid. I may mention in this connec- 

 tion three-legged (five-limbed) chickens among the birds, and 

 five-winged butterflies and moths (in cases where the additional 

 wing does not replace a leg). In one case of the latter, a 

 beautiful female of Platysamia cecrqpia, which has recently 

 come under my notice, the similarity to the theoretical phylo- 

 genetic condition among the crinoids is very striking, for the 

 additional wing (which unfortunately failed to spread properly) 



Specimen of Platysamia cecropia with an additional primary inserted 

 just anterior to the normal right primary. 



is inserted anterior to one of the primaries, where, were the 

 animal sessile and incapable of motion, it might, through fixa- 

 tion in subsequent generations, very well give rise to as truly 

 a pentamerous condition as is exhibited by the crinoids. 



Although I have treated in detail only the evidence brought 

 out by the crinoids, I have no doubt that the pentamerous 

 symmetry in all the other groups has arisen in the same way, 

 by the interpolation of an extra ray between the two original 

 anterior rays ; and that this accounts for the excess of plasti- 

 city of this ray in certain groups, and its more or less frequent 

 suppression in others. In six-rayed specimens almost invari- 

 ably the sixth ray is obviously formed by a more or less com- 

 plete division of one of those already existing — a sort of 

 echinodermal polydactylism — and these six-rayed examples, 

 therefore, are very suggestive in indicating that echinoderm 

 symmetry is not by any means radial, but is a composite of 

 five equivalent parts, two paired and one, the anterior, odd. 



The echinoderms as known to us are extremely ancient 

 animals, zoologically speaking, so ancient that they were 



