Feb. a, 1924 
Tissue Fluids in Cotton 
273 
The cultural technique was that generally employed for the growth 
of Egyptian and Upland cotton at the Cooperative Testing Station at 
Sacaton. Since all the materials considered were grown under as nearly 
as possible identical conditions, it seems unnecessary to detail these 
methods here. 6 
collection of samples 
Cotton is not in its morphological features a desert plant. It wilts 
readily and must be copiously supplied with irrigation water in order to 
produce a crop under the intense heat and scanty precipitation of the 
Southwest. During August the plants and the condition of the soil in 
which they are growing are continually varying, not merely from day to 
day but literally from minute to minute. Immediately after irrigation, 
in which the whole surface of the “border’' is flooded with a sheet of 
water, the soil is fully saturated with water, but this is lost very rapidly. 
Water loss from the plants through transpiration increases very rapidly 
after sunrise, and great care must be taken not to obtain samples of leaves 
which have already begun to wilt. When there is a heavy dew in the 
morning, as occasionally occurs after the summer rains have begun, it is 
difficult to secure samples in the limited period between the time when 
the leaves have scattered droplets of external moisture and the time 
when they begin to wilt under the influence of the intense insolation. 
Because of the continual changes in the condition of the plants, the 
taking of the samples is one of the most exacting features of the work. 
In addition to the care necessary in selecting the brief period most suit¬ 
able for taking the samples, two possible sources of error had to be care¬ 
fully avoided. 
There must be introduced no artificial source of differentiation between 
the two species under consideration, or between the species and their 
hybrid, through the taking of the samples from different portions of the 
field or at different times. 
The leaves taken from the plants to be compared must be compara¬ 
ble in maturity. 
Artificial differentiation between the species or between the species 
and their hybrid may be avoided in cases in which the varieties to be 
compared belong to the same duplet (rows 2, 4, and 6) or the same 
triplet (rows 1 and 5 and rows 3 and 7) by making sure that collections 
are taken simultaneously from the two or three members of the duplet 
or triplet. 
In the first series of collections (1920) it was possible to avoid the 
influence of heterogeneity in the soil conditions 7 because of the fact 
that in a number of breeding experiments under way at Sacaton, Pima 
cotton and Upland cotton varieties were grown side by side in the same 
plot or “border.” All collections were taken from adjacent rows or 
from closely associated plants in the same row. In the investigations 
of 1921 the possible influence of this factor was precluded by a special 
planting scheme, as described above. 
To avoid the possible influence of time variations, the samples from 
the two or three members of the duplet or triplet were taken at as 
6 We are greatly indebted to Mr. Walter F. Gilpin, who carefully made the plantings, took some of the 
records and facilitated our work in other ways. . . _ , , 
7 Readings made by Mr. Kearney with the soil bridge on samples of soil taken in the neighborhood 
of the various sets of tissue samples, indicated differences in the salt content from place to place m the 
field. 
