LATENT PERIOD AND REFRACTORY PERIOD 269 



The rate of responsive movement is also seen to have under- 

 gone considerable diminution. In the first, or normal, case, 

 during half a second after the commencement of response, the 

 rate of movement was, as said before, 71 mm. per second. 

 In the second case, however, after slight cooling, it is seen to 

 have been reduced to 26 mm. per second, or almost one-third 

 of the original rate. 



Record by means of electric sparks : prolongation of 

 latent period by fatigue. — In order to overcome the difficulty 

 of the insensitiveness caused by keeping the plant in a photo- 

 graphic dark room, I have recently devised a method of 



Fu;. III. Electric Spark Record, Showing Increase of Latent Period by 

 Fatigue, in Successive Responses of a Leaf of Mimosa 



Latent period in normal topmost record seen to be ~ second ; this increased 

 in next — taken i minute before full recovery — to ■-— second ; latent 

 period further increased in last case — taken 3 minutes before full 

 recovery — to ~; second. Note also progressive change in slope of 



record by means of a series of punctures produced by electric 

 sparks on a recording paper surface. The sparks occur at 

 the short gap between the end of the long arm of the record- 

 ing aluminium lever, and the drum, these being connected 

 respectively with the two electrodes of a Ruhmkorff's coil. 

 The electrical disturbance does not affect the plant, as the 

 leaf is attached to the lever by a long silk thread. One 

 great advantage of this method lies in the fact that the time- 

 intervals, which may be as short as desired, are indicated by 

 the distance between successive punctures, which are deter- 

 mined by the frequency of the vibrating interrupter of the 

 coil. In the case of which the record is given (fig. i 1 1) the 



