4 (>2 PHYSIOLOGY 



10 minutes and 48 minutes. The longer the stimulation, other things 

 being equal, the more marked the curvature ; from which it is evident 

 that there is an increase of the excitation with continued stimulation, 

 and thereby the end reaction becomes more marked. 



Summation. — Contrariwise, it should be expected that stimulation 

 too short to result in curvature would not be without effect. That it 

 does produce excitation is shown by the fact that if a plant be placed alter- 

 nately horizontal and erect, each period of stimulation being shorter 

 than the presentation time for that particular plant, and the interval of 

 rest shorter than is needed for recovery, curvature will finally occur. 

 Evidently this is a cumulative effect; yet it is not a summation of the 

 total successive excitations that occur during the times of horizontality, 

 a j- but only of the re- 



sidual excitation. 

 For, if a suitable 

 plant be placed 

 horizontal for 30 

 minutes continu- 

 ously, the reaction 

 curvature is more 

 C pronounced than 



Diagram: for explanation, see text. .? .. , 1 j 



5 ^ if it be so placed 



for ten 3-minute periods at 10-minute intervals. Clearly, while erect, 

 the preceding excitation is slowly disappearing, and if the interval before 

 the next stimulation is too long, recovery will be complete and no 

 evidence of the excitation will appear in the form of curvature. 1 In 

 such experiments, therefore, it is necessary to apportion properly the 

 intervals of rest and stimulation. 



Rotation. — From the above considerations it will be evident that when 

 a plant is rotated in the horizontal plane on a clinostat, its failure to exe- 

 cute any curvature is not at all due to a lack of excitation, for while the 

 side a of the stem is passing through quadrant A of its rotation (fig. 694), 

 quadrants a and c are under stimulation almost as though for a corre- 

 sponding time the stem were at rest. But these sides remain under stimu- 

 lation for less than the presentation time and so the excitation does not 

 suffice for the end reaction. When side a has passed into quadrant C 



1 It has been suggested that during the periods of no stimulation a counter-excitation is 

 set up; but simple recovery from excitation seems sufficient to account for all the facts 

 known. The process is apparently analogous to recovery from fatigue. 



