REPORT OF THE DIRECTOR I916 I 63 



graphic stereograms though details may be lost in the process of 

 reproducing as a halftone. 



Covering both column, infrabasals and basals is a coating of 

 epidermal structures which measures about 0.08 mm in thickness. 

 The numerous white dots shown are sections of spines between 

 0.02 and 0.03 mm in diameter. Near the arms are cross-sections 

 of biserial pinnules with cover-plates. One of these cross-sections 

 is less than 0.03 mm in diameter. 



Specimen collected by Dr C. D. Walcott from the Trenton near 

 Trenton Falls, N. Y., and now in the Museum of Comparative 

 Zoolog}^, Cambridge, Mass. It was loaned the author through the 

 courtesy of Dr P. E. Raymond and Hon. Samuel Henshaw. 



If the pinnule fragments belong to the specimen, which is 

 extremely probable, they serve to increase the difficulty of inter- 

 pretation. In the most primitive cystids brachioles were no doubt 

 associated directly with short food grooves. The extension of the 

 food groove possibly never involved the turning of a brachiole 

 into an arm. Epithecal extensions like those found in Blastoi- 

 docrinus and the blastids were never developed from arms which 

 were once exothecal ; the recumbent arms of Malocystites may have 

 had their origin in free arms like those of Canadacystis. If 

 brachioles developed before supporting arms and if the arm 

 structure was afterward developed to form a common food con- 

 duit, enabling the bracholar structures to function at some dis- 

 tance from the mouth and also allowing a large increase in their 

 number, then one of our difficulties disappears. 



