654 
Journal of Agricultural Research 
Vol. XXXI, No. 7 
MATERIALS AND METHODS 
The experimental plants are in part the same as those considered 
in the earlier investigations. All were grown at the Cooperative 
Testing Station of the Bureau of Plant Industry in the Gila River 
Valley at Sacaton, Ariz. The study included: 
A. A comparison of Pima Egyptian and Meade Upland cotton 
grown in 1922. Determinations were based on samples of tissue 
collected from individual plants. 
B. A series of Pima Egyptian and Lone Star Upland plants grown 
in 1923. Determinations were based on samples taken from groups 
of plants. This is the series on which the chloride determinations 
for Pima and Lone Star cotton have been based (5 ). 
Collection of samples and methods of preparation of tissue fluids 
are described elsewhere (4, 5). The sulphate content was deter¬ 
mined ( 1) as follows: 
Five to ten cubic centimeters of the expressed tissue fluid are 
placed in an 11.5 cm. porcelain evaporating dish, and 10 cubic centi¬ 
meters of Benedict-Denis oxidizing reagent 3 are added. This mixture 
is evaporated to approximate dryness on a water bath and then care¬ 
fully ignited, at first over a small flame and finally to dull redness 
for a few minutes. After cooling, the ignition residue is dissolved in 
dilute hydrochloric acid, the solution filtered, and the sulphate 
precipitated from the filtrate by the addition of barium chloride and 
weighed as barium sulphate. 4 
Analyses were based on carefully pipetted samples of 5 to 10 cubic 
centimeters, preserved in sealed tubes after the addition of a drop of 
formaldehyde. Concentrations are expressed in terms of grams of 
sulphate (S0 4 ) per liter of tissue fluid. 
RESULTS 
The distribution of sulphate content in class units of 0.5 gm. of 
sulphate per liter in the single series of determinations made m the 
cultures of 1922, and in the two series of determinations made on the 
cultures of 1923, appears in Table I. 5 
3 This reagent is made by dissolving 25 gms. of crystalline copper nitrate, 25 gms. of sodium chloride, and 
10 gms. of ammonium nitrate in enough water to make 100 cubic centimeters of solution. Ten cubic centi¬ 
meters of the reagent are ample for 10 cubic centimeters of tissue fluid. 
4 It is necessary to make “blank” determination of the reagent and to subtract this from the total weight 
of barium sulphate. By using high-grade chemicals, a blank of 0.0010 to 0.0035 gm. of barium sulphate is 
obtained. 
5 In Table II of the paper presenting the results for chloride content in the 1923 cultures, the second 
series of determinations was divided into Part I and Part II. The sulphate analyses were made for samples 
taken from the south half of the plot only. Thus the sulphate determinations correspond to Part I only of 
the chloride series. 
