946 
Journal of Agricultural Research 
Vol. XXIII, No. xa 
CONCLUSIONS 
Recognizing summer irrigation as a distinct problem, experiments 
were conducted at Sacaton, Ariz., to determine the effects of different 
frequencies of irrigation during the months of July and August on Pima 
Egyptian cotton that had reached a normal early fruiting stage. 
Nine irrigated plots or “borders” were arranged in three series, 
affording three independent tests of contrasted irrigation treatmemts. 
Each series included a “normal” border irrigated according to the usual 
practice of applying water as indicated by the behavior of the plants, 
a “heavy” border receiving irrigations at frequent intervals, and a 
“medium-heavy” border receiving an intermediate irrigation treatment. 
The behavior of the plants in each of the borders was observed and 
compared, and the data of the principal features, including growth, 
flowering, shedding of young bolls, and yields, were recorded through 
the season. 
As a general result of the experiment it was determined that the 
different frequencies of irrigation, after the plants had reached a normal 
fruiting stage, did not cause any consistent significant difference in the 
growth of the plants, or in the yields. 
No consistently different rates of flowering or of shedding of young 
bolls resulted from any of the different kinds of treatment. 
A wide range in the daily flowering production occurred, there being 
days when all borders had large numbers of flowers and days when all 
borders had a few flowers. The rate of shedding of young bolls also 
varied from day to day, though on only a few days was there a consistent 
behavior throughout all the borders. No correlation between these 
fluctuations and factors which might be supposed to influence this 
phenomenon could be definitely established. 
No consistent increase or decrease at a regular interval after an irri¬ 
gation was found, either in the flowering or in the shedding of young 
bolls. 
Shedding of young bolls was greatest from 4 to 14 days after flowering, 
the interval varying only slightly between borders. The shedding of 
flower buds, or “squares,” was almost negligible throughout the ex¬ 
periment. 
Plant development was checked and other symptoms of distress ap¬ 
peared in one of the three heavily irrigated borders, apparently from an 
excess of water. 
The yields varied in the differently treated borders, but from hetero¬ 
geneity of the soil rather than from the different frequencies of irrigation 
or in the loss of bolls by shedding. 
The experiment indicates the importance of giving more attention to 
the spring treatment of cotton, so as to have the plants in a normal 
fruiting condition when summer irrigations begin. When this normal 
fruiting condition is attained, the summer irrigation problems are 
simplified, since the plants are not so easily forced into rank growth by 
the application of water in excess of the actual requirements. 
