366 PLANT RESPONSE 



and is not shown in the figure. The stop-cock a' is opened, A, 

 B, and b' being closed. Water from the compensator c thus 

 passes into the plant-vessel, and the air-bubbles which have 

 been accumulating at the top of the vessel escape, with the 

 water which is driven out at the escape-tube. When all are 

 expelled then the stop-cock of the escape-tube, and A', are 

 closed. The potometer stop-cock A is now opened, and the 

 water-index is adjusted at any point desired, by manipulation 

 of the stop-cocks B and A'. A temporary opening of the 

 stop-cock A' of the compensator c causes the water-index 

 to move to the left ; whereas, when the stop-cock B, of the 

 outflow pipe, is opened, it moves to the right. After this 

 preliminary adjustment the stop-cocks A', B, and b' are closed, 

 the potometer tap A being kept open ; the movement of the 

 water-index per unit-time now gives us the normal rate of 

 suction of the specimen. 



W T e come next to the question of making a direct record 

 of the rate of movement. For this purpose, a writing pen is 

 fitted on the potometric tube, by means of a brass collar. 

 This brass collar has a rectangular opening, which enables 

 us to watch the water-index. It has also stretched across it 

 a fine wire, which is kept always coincident with the water- 

 index. This wire is parallel with, and placed vertically above, 

 the recording pen. The collar is attached to a thread which 

 passes round small pulleys. One end of this thread carries 

 a counterpoise and the other is wound round a wheel, W, 

 which can be so manipulated as to make the index-wire 

 follow the movements of the water-column. When the wheel 

 is wound, the index moves to the right ; when it is slightly 

 released, the weight of the counterpoise makes it move to 

 the left. The weighted recording pen rests with its point 

 on a revolving drum, D, covered with paper for the record ; 

 this drum is kept revolving by clockwork at a known and 

 adjustable speed. When the water-index is followed in the 

 way described, there is produced a direct record of water- 

 movement in the plant. A curve is thus traced, the ordinate 

 of which represents the quantity of water sucked up, and the 



