TEMPERATURE AND AUTONOMOUS RESPONSES S3S 



at 35° C. was allowed to descend to the thenno-tonic 

 minimum. It was in that case found that there was a regular 

 diminution of amplitude, and the pulsation afterwards dis- 

 appeared below 29° C. (fig. 124). 



I was next desirous of determining whether this theoretical 

 inference could be verified in the case of Desmodiiim. For 

 this purpose I rapidly cooled the plant, by means of the 

 cooling coil, through which ice-cold brine was passed, a 



Fig. 140. Effect of Cooling to Thermo-tonic Minimum on Pulsation 

 of Desinoditini 



The first two pulses to the left were taken at the normal temperature oi 

 29"^ C. ; those to the right during continuous cooling. The first 

 response of the latter series occurred at 22° C. ; the second at 20° C. ; 

 and the third at 17° C. Note that while in normal responses there 

 are no sub-pulses, these are seen with increasing distinctness as the 

 pulsation becomes continuously slower. They first make their appear- 

 ance during the up movement of the first pulsation under cooling. In 

 the two successive pulsations they are seen more and more clearly in 

 the down as well as the up movements. 



photographic record being taken of the pulsations all the 

 time. It will be noticed that the first effect of cooling was 

 the normal increase of amplitude and prolongation of period. 

 This latter— which at the temperature of the room had had 

 a value of three minutes — was now prolonged to nearl}- six 

 minutes. But the most interesting fact was that, as the 

 thermo-tonic minimum was approached, the amplitude was 

 reduced (fig. 140). 



