THE RECORD OF GROWTH-RESPONSE 415 



This outflow of water is roughly adjusted by opening the 

 stop-cock, to a greater or less extent. The finer adjustment 

 is then effected by the suitable variation of the difference of 

 level in the two limbs of the syphon. One end is con- 

 nected with a flexible india-rubber tube, which is attached to 

 a string passing over a pulley, and fixed to the adjustable 

 wheel B, attached to the observer's table. Rotation of the^ 

 wheel in one direction will depress this end of the syphon, so 

 increasing the flow, and in the other direction will raise it, so 

 diminishing the escape of water. It will be noticed that the 

 rate of outflow of the water is not in any way affected by the 

 variation of level of the water in the cylinder. It simply 

 depends on the difference of level between the two ends of 

 the syphon. The rate of descent of the float is thus regulated 

 with the utmost nicety, till an absolute balance is obtained. 

 The observer recognises the condition of balance when there 

 is no drifting of the spot of light on the recording drum. 

 The adjusting wheel is now fixed at this position of balance. 

 If any agent should induce an acceleration of growth, the 

 balance is disturbed, and the spot of light moves, say, to the 

 right, or in a positive direction. Any agent which induces 

 retardation will, on the other hand, cause a deflection of the 

 spot of light in a negative direction. Should there be any 

 natural fluctuations in the growth, the oscillation of the spot 

 of light will give an indication of the fact. 



The record is made in the usual manner on a revolving 

 drum. Fig. 168 illustrates the complete apparatus, which 

 enables us to obtain a record under balanced — or by closing 

 the stop-cock of the syphon, also under unbalanced — con- 

 ditions. 



The wheel is graduated, and the absolute value of the 

 compensatory movement, at any position of the circular scale, 

 can be previously calibrated, by fixing the plant-mirror, and 

 observing the extent of movement of the spot of light on the 

 drum, due to the subsidence of the float in a given time. 



For the exhibition of pure longitudinal growth-response, 

 the most perfect specimens are the growing radial styles and 



