470 Mr Lawrence Balls, Pre-Determination of Fluctuation. 
This remarkable parallelism lasts till mid-July at least, and | 
may be resumed after an interval of confusion. 
The cause can only be climatic, not edaphic, since differential 
soil treatment does not disturb it. Yet at the time when it is | 
manifest the plant has passed under edaphic control*. 
Statistical investigations by Mr H. E. Hurst and by the 
writer, comparing these fluctuations of flowering with those of 
wind, temperature, humidity, evaporation, cloudiness, &c., both at 
the time of flowering and during the few preceding days, gave no 
result. 
After some three years, as further data accumulated, the 
writer traced the cause to pre-existing fluctuations in the growth- 
curve of the central axis, nearly a month before. 
There is a similarity between the daily curves of growth and | 
flowering, while the smoothed curves from at least two separate | 
years—one of them very abnormal—are practically identical. 
The growth-curve of the central axis is the same as those of | 
the flowering branches until mid-June. The curve thus records 
the rate at which the scaffolding is laid down, upon which the 
flower buds are differentiated. The bud takes rather less than a 
month to open from its first differentiation, so that fluctuations in 
the rate of construction of the scaffolding are repeated in the rate 
of opening of the flowers. 
Under the conditions of ordinary cultivation of cotton in Kgypt, 
subsequent environmental influences exercise very little deforming | 
effect. 
The same applies to the fruiting-curve, which is closely similar | 
to the flowering-curve, though of less amplitude, so that the © 
arrivals of Egyptian cotton at Liverpool in October are almost | 
entirely determined by the night-temperatures, &c. in Egypt in 
May. 
Conclusions. 
1. That we are largely in error when we seek to ascertain 
the causes of Fluctuation in some feature of an organism by | 
integrating the effects of those environmental factors which were ~ 
operating at, or nearly before, the time when the feature im 
question was manifest. 
2. That such factors merely exercise a subsidiary deforma p 
influence on a scaffolding which was pre-determined—or pre-— 
destined—often at a very much earlier stage in the life of the © 
organism. 
3. <A conception of Discontinuity is thus introduced into the » 
study of Fluctuation, which has oy been regarded as typical | 
of Continuity. 
* W, L, B. ‘The Cotton Plant in Egypt.’ London, 1912, chap. 11, 
—s 
