l6,2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 57 



very superficial manner in those parts of North America where they 

 are well developed and finely exposed for the collection of fossils. 

 This leads me to think that it is only a question of time and detailed 

 work to bring to light a large and varied crustacean fauna. This 

 almost certainly existed, as proven by the occurrence of Beltina in 

 the pre-Cambrian of the Rocky Mountains.^ 



Bernard's very interesting and valuable study of Apus^ is replete 

 with suggestions and inferences bearing on the evolution of the Crus- 

 tacea from a browsing carnivorous annelid with its first 5 segments 

 (head) bent so that its mouth faced ventrally and posteriorly, and 

 using its parapodia for pushing food into its mouth. He concludes 

 that the modern representative of this crustacean-annelid is Apus. 

 With Bernard's theory in mind I have examined the Burgess shale 

 annelidan and crustacean fauna to ascertain if there was an annelid 

 that could be considered as representing his hypothetical crustacean- 

 annelid, and nearer to it in structure than Apus. I found specimens 

 of Canadia spinosa Walcott laterally flattened in the shale with the 

 head bent down, so that the mouth faces posteriorly,* also that 14 

 out of 24 specimens have the head bent under and out of sight 

 beneath the flattened body. Possibly these annelids and the crusta- 

 ceans were derived from the same general type of animal. 



Among the crustaceans Marrella splendens (pis. 25 and 26) has 

 an Apus-like form, but it is evidently a more highly developed form 

 than Apus. This is shown among other characters by its carapace, 

 long jointed legs, and fewer segments.* Burgessia hella (pi. 27, figs. 

 1-3) has a simple carapace, few thoracic segments, and many 

 abdominal segments, if those of the telson-like extension of the body 

 are considered as belonging to the abdominal region. The eight 

 thoracic segments serve to separate Burgessia from Apus and other 

 genera of the Apodidse and at the same time bring it near to the 

 Phyllocarida as represented by Nebalia. On the other hand, the 

 simple Lepidurus-Yike carapace, sessile eyes, and hepatic glands in the 



^ C. D. Walcott, 1899, Pre-Cambrian Fossiliferous Formations ; Biill. Geol. 

 Soc. America, Vol. 10, pp. 238-239, pis. 25 and 26. 

 ' The Apodidse. Nature Series, London, 1892. 



* Smithsonian Misc. Coll., Vol. 57, No. 5, 1911, pi. 23, fig. 4. 



* Dr. Austin H. Clark considers that a comparison should be made between 

 Marrella and the Trilobita. He suggests that the cephalon is comparable 

 with that of Acidaspis, the two anterior spines being the genal spines and the 

 posterior spines the same as the occipital spines or processes of Acidaspis. 

 The terminal plate he takes to be the pygidium, and the feathery organ (m) 

 the last cephalic appendage. 



