NO. 10 CRUSTACEAN METAMORl'llOSES — SNODGRASS 67 



Walcott (1931), for example, in discussing his Burgess Shale fossils 

 seems to accept this theory without question when he says: "The 

 biramous limb of Marella, like that of the trilobite, undoubtedly 

 passed through the foliaceous or multiramous limb stage in its evolu- 

 tion, probably in pre-Cambrian time." There is no disproving this 

 idea, which should apply to the other arthropods as well, but such 

 implicit faith in a theory is hard to understand. 



On the other hand, Raymond (1920) says the theory of the phyllo- 

 pod origin of the arthropod limb "has been completely upset" by the 

 finding of such "undoubted branchiopods" as Burgessia in the Middle 

 Cambrian with trilobitelike legs. The same idea has been expressed 

 by Heegaard (1947) in his statement that the "undoubted branchio- 

 pods" found by Walcott in the Middle Cambrian having trilobite legs 

 show that "it can no longer be held that the phyllopodial limbs are 

 primitive." The writer fully agrees with this conclusion, but for 

 different reasons than those given by Raymond and Heegaard. Such 

 fossils as Burgessia and Marella are certainly not "undoubted" 

 branchiopods. Walcott (1931) says of Marella that it is a less primi- 

 tive form than the Apodidae and more primitive than the trilobites, 

 but is nearer to the latter than to the former. Among the Middle 

 Cambrian fossils, however, is a form, Opabina regalis Walcott, par- 

 ticularly studied by Hutchinson ( 1930) , which evidently is an anostra- 

 can branchiopod with foliaceous appendages. 



Another popular belief concerning the derivation of the arthropod 

 limb, taken to support the theory of its biramous phyllopodial origin, 

 is that the limb was evolved from the polychaete parapodium. Reasons 

 have already been given in section I of this paper for believing that 

 the annelids have only a remote connection with the arthropod pro- 

 genitors. Certainly the arthropods can have no relation to modern 

 polychaetes, which are highly specialized annelids and could give rise 

 only to more polychaetes. The appendages of the worm, though they 

 are bilobed flaps, have a lateral position on the body (fig. 26 A), and 

 there is nothing in tiieir structure having any likeness to an arthropod 

 limb at any stage of its development. The parapodium bears two 

 bundles of bristles supported on a pair of long internal rods giving 

 attachment to muscles. Its only common feature with an arthropod 

 limb is that, being a locomotor organ, it is movable forward and back- 

 ward by body muscles. In short, the idea that the arthropod append- 

 ages were derived from annelid parapodia appears to be just another 

 case of excessive zeal for generalization. 



Among modern wormlike animals those closest to the arthropods are 

 the Onychophora; some zoologists have even included the ony- 



