696 
Journal of Agricultural Research 
Vol. XXVIII, No. 7 
MATERIALS AND METHODS 
The data are derived from four experimental series of plants grown at the 
United States field station, Sacaton, Ariz. These may he numbered for conven¬ 
ience of reference as follows: 
A. —A comparison of Pima Egyptian and Acala Upland cotton in 1920. 
B. —A comparison of Pima Egyptian and Meade and Acala Upland cotton in 
1922. 
C. —A second comparison of Pima Egyptian and Meade Upland cotton in 
1922. 
D. —A comparison of Pima Egyptian and Lone Star Upland cotton in 1923. 
The two or three varieties of plants to be compared were distributed uni¬ 
formly over the plots available in order to avoid the influence of substratum 
heterogeneity (3, 4). In this regard the technique was essentially that of the 
experiments made in 1921 (5 ). 5 
The first comparison of Pima Egyptian and Meade and Acala Upland cotton 
in 1922 (Series B, above) was made by planting 24 rows each 180 feet in length 
and consisting of six cultures of 10 feet each of Acala, Pima, and Meade. 6 
In the collection of samples, the field was in effect divided into 216 subplots 
(each 10 feet long and two rows wdde, and planted to one of the three varieties) 
by combining rows 1 and 2, 3 and 4, etc. The samples of tissue were drawn as 
widely and as uniformly as possible from the theoretically 20 7 plants of each 
subplot. The three adjoining subplots of the same pair of rows serve as units 
in the making of comparisons. Each unit therefore comprises determinations 
on Pima, Acala, and Meade plants. The samples of each unit were taken sim¬ 
ultaneously, so that external conditions which vary with time can not influence 
the results. 
This comparison (B) involves a first and a second series of determinations, 
the first based on samples collected from July 25 to August 9, the second based 
on samples collected from the same cultures between August 28 and September 
4, inclusive. 
The second comparison (C) of Pima Egyptian and Meade Upland cotton 
made in 1922 was based on samples of tissue collected from individual plants 
of the two varieties. These plants were distributed uniformly over a plot 26.5 
by 400 feet in area. They were interspersed among plantings of five newly 
imported Egyptian varieties which do not require consideration here. The 
Pima and Meade plants were separated by only three hills of F 2 hybrid plants. 
Tissue samples were collected from August 21 to August 27, inclusive. 
The comparison of Pima Egyptian and Lone Star Upland cotton in 1923 
was made on an irrigation plot of the standard dimension, which was selected 
because of the high salt content of the soil at the north end. The south half 
of the plot had lain in alfalfa from 1920 to 1923. The north half was too saline 
to permit a growth of alfalfa, and but few cotton plants survived on this portion 
of the plot. The collections made from the plants of the 10-foot subrows at 
the south and the north half of the plot will, therefore, be discussed separately. 
6 This statement applies to the cultures made in 1922 and 1923. The few determinations made in 1920 
were secured incidental to other determinations. 
• The 24 rows were distributed over the north half of three irrigation plots, each 26.5 by 400 feet in area. 
Ten feet at each end were used for buffer plants in order that the end plots used might have the same cul¬ 
tural conditions as those in the center of the field. 
i Because of the cold spring, germination was poor and many plants were wanting. The number was 
still furtner reduced by the development of root rot late in the season. 
