WATER-STRESS BEHAVIOR OF PIMA COTTON. 
15 
THE PERIOD BETWEEN FLOWERING AND SHEDDING. 
The interval between the opening of the flower and the completion 
of the abscission process is of some importance in identifying the 
stimulus which produces shedding. From Table VI it is apparent 
that there is little or no correlation between the length of this period 
and the moisture content of the soil at the time of flowering. 
Table VI. — Period of retention of shed bolls developing from flowers of Pima 
cotton blooming on successive weeks on four plats compared with the avail- 
able moisture present in the soil during the season of 1919. 
Date. 
Plat 1. 
Plat 2. 
Plat 3. 
Plat 4. 
July 16 
da vs.. 
9.0 
8.2 
7.7 
8.1 
22 
do.... 
10.8 
10.6 
6.0 
10.7 
30 
do.... 
7.8 
13.8 
12.5 
8.5 
Aug. 6 
do.... 
7.5 
8.1 
7.7 
5.6 
13 
do.... 
9.2 
8.0 
12.0 
10.1 
20 
do.... 
7.1 
9.2 
8.9 
10.7 
27 
do.... 
9.4 
9.0 
10.4 
11.8 
Sept. 3 
do.... 
16.1 
16.5 
15.1 
12.0 
10 
do.... 
18.3 
13.8 
8.8 
10.4 
16 
do.... 
10.6 
9.5 
6.1 
10.0 
23 
do.... 
16.2 
10.2 
11.3 
21.2 
Mean 
11.1 
5.0 
10.6 
3.0 
8.8 
2.5 
10.8 
Mean available moisture in top 4 feet of soil 
percent.. 
2.8 
Ewing (12) states that " the period between flowering and shed- 
ding in general is longest when shedding first begins and shortest 
at the close of the season." Lloyd (18) shows that " there is little 
evidence that the mean of age frequencies of shed bolls was reduced 
•during the season, but that there is definite evidence that the mean 
may recede or advance as a result of the additive or subtractive 
effects of stimuli, such as the additive effect of rain upon that of the 
soil moisture which resulted in a recession of the mean of three 
days." It appears from the results shown in Table VI that in Pima 
cotton there is a tendency for the period between flowering and 
abscission to increase as the season advances, under the conditions 
which obtained under this experiment, but the fluctuations were great 
even near the end of the season. 
A study of the results in Table VI will show that there is only one 
weekly mean which exhibits clearly the effect of some stimulus oper- 
ating generally on all the plats. This is the mean for the week 
ended August 6 when there was a recession of the mean on all the 
plats. It so happens that the only rain of any importance during the 
fruiting season fell on August 1 (fig. 2), so that there seems to be 
little doubt that rainfall was the responsible factor for the relatively 
large shedding of bolls during the week specified. 
VARIATIONS. OF PERIODICITY OF DEVELOPMENT. 
The cotton plant, like many other plants, exhibits pronounced peri- 
odic variations in its developmental activity. Among these is the 
