116 THE FLOWERING PROCESS 



effectiveness as the middle is approached, is nearly a mirror image of 

 the curve on the right, which shows a decreasing effectiveness after 

 the middle has been passed. It seems reasonable to assume that a 

 light interruption near the end of the dark period (the right side of 

 the curve) is like returning the plants to light and is, therefore, very 

 much like studying the effect of different night-lengths (see Chapter 9). 

 A few years ago when the experiment was done, we went on to deduce 

 that an interruption during the first half of the dark period stopped 

 what had gone on up to that time, essentially reversing it so that the 

 processes of the dark period had to start over again. Thus if the 

 plants were interrupted 4 hr after the beginning of a dark period and 

 12 hr remained following the interruption, it was reasonable to 

 expect that they would flower to about the same degree as plants 

 which either had a 12-hr uninterrupted dark period or were inter- 

 rupted 12 hr after the beginning. The curves in Fig. 7-6 do seem to 

 fit the interpretation, but there are problems with the interpretation 

 which are discussed in the next section. 



3. Light Interruptions at Various Times Using Cohaltous Ion 

 Treated Plants 



Some time later the same sort of light interruption experiments 

 were done with plants which had been dipped in solutions of cobaltous 

 chloride just before the dark period. Plants treated with cobaltous 

 ion require a longer period of darkness for the first perceptible signs 

 of flowering than untreated control plants (see Chapters 8 and 9). 

 When the light interruption experiment was done, the right hand 

 curve (Fig. 7-7) shifted to the right, as was to be expected from this 

 effect of cobaltous ion on the critical dark period. Reasoning by the 

 above simple theory, it was easy to decide that the left-hand part of 

 the curve should be shifted to the left. A plant interrupted 6 hr after 

 the beginning of a 16-hr dark period should flower about the same 

 as one interrupted 10 hr after the beginning of a 16-hr dark period. 

 As can be seen from Fig. 7-7, however, our expectations were not 

 borne out. Rather than the curve shifting to the left, it stretches 

 towards the right so that it actually crosses the control curve. 



To reconcile these facts with the rather simple picture outlined 

 above, one must make some very elaborate assumptions about the 

 action of cobaltous ion, which at this stage of the game do not seem 

 to be justified. An alternative theory comes to mind: the effectiveness 



