292 S. O. MAST AND K. S. LASHLEY 



duced. If the cone contains the stimulating agent the organism 

 again contracts after which it very soon expands again, but now 

 it bends much more than it did previously so that the anterior 

 end is usually no longer directed toward the region containing 

 the stimulating agent (fig. 6). The production of the cone in 

 these reactions seems to be due in part to the inertia of the body 

 and in part to the additional resistance due to the curvature 

 in it. 



Free-swimming rotifers regularly produce a cone which, how- 

 ever, is rarely more than one-fourth of the length of the body. 

 When they enter an injurious solution they contract the disk 

 and settle to the bottom. Then they usually expand the ciliary 

 lobe but not the entire disk, draw in a small cone of water, turn 

 through a certain angle and contract again. Thus they continue 

 contracting and turning after each stimulation until they are no 

 longer stimulated by the cone, when they expand the disk and 

 proceed. 



DISCUSSION 



We have thus demonstrated that the feeding-cone in Para- 

 mecium, Spirostomum and Stentor is produced only under 

 certain conditions, that during locomotion it is produced only 

 (1) when there is an acceleration in rate without any change in 

 the relative activity of different cilia, (2) when there is a reversal 

 of the ciUa, except those in the oral groove and (3) when the 

 animals are mechanically retarded by lateral contact with the 

 substratum, etc. We have also demonstrated that in these or- 

 ganisms the avoiding reaction is not due to the production of a 

 feeding cone. The feeding cone can not, therefore, be of any 

 considerable value in ascertaining the condition of the environ- 

 ment ahead as maintained by Jennings. 



These facts do not, however, overthrow the contention that 

 Paramecium continuously tests the condition of the environment 

 ahead. The anterior end of these creatures is undoubtedly sen- 

 sitive to practically all agents to which they respond, and they 

 usually respond as soon as this end reaches the stimulating agents. 

 Thus they detect favorable and unfavorable regions before they 



