312 PLANT RESPONSE 



determines the persistence of this after-effect, namely, the 

 larger or smaller capacity of the tissue itself to hold energy 

 latent within it. 



In the attempt to investigate the cause of automatic 

 movements in Desmodium, there were three points of inquiry 

 which had to be determined. First, there was the question 

 of the seat of excitation ; second, that of the nature of the 

 stimulus which maintained the rhythmic pulsation ; and third, 

 the determination of the process by which constant stimulus 

 found periodic expression. 



I have shown that the seat of excitation in Desmodium is 

 neither central nor peripheral, but is in, or in immediate con- 

 tiguity with, the motile tissue itself, which resembles in this 

 respect the animal heart, the seat of excitation there also 

 being in the cardiac tissue itself (Chapter XXVI.). 



I have also shown that as the cause of excitation it is 

 not necessary to have any sudden variation. A constant 

 stimulus, of whatever nature, is found efficient to produce 

 excitation. 



I have also demonstrated that periodic pulsation is pro- 

 duced in Desmodium at standstill, by a constant thermal stimu- 

 lus ; also that periodic pulsations are produced in BiopJiytum 

 and in Desmodium by the constant action of a chemical 

 stimulant. I have shown further that rhythmic excitation 

 was produced in Averrhoa carambola by the passage of a 

 constant electric current. And, finally, I have shown that, 

 just as in the retina, under constant stimulus of light we have 

 periodic visual excitations, so also in BiopJiytum, under con- 

 stant stimulus of light we obtain periodic excitations which 

 give rise to rhythmic movements. In the animal tissue, 

 similar multiple rhythmic responses are met with under 

 constant stimulus. 



Since we have found it to be a fundamental characteristic 

 of the tissue of a rhythmically-responding plant like Bio- 

 pJiytum or Desmodium to give a response which cannot be 

 mcreased by any excess in the stimulus-intensity (maximal 

 response or none), we should expect that the excess over the 



