SOME NOTES ON THE DIRECT DETERMINATION 
OF THE HYGROSCOPIC COEFFICIENT 1 
By Frederick J. Alway, Chief of Division of Soils, Agricultural Experiment 
Station, University of Minnesota, Millard A. Kline, formerly Assistant in Chem¬ 
istry , Nebraska Agricultural Experiment Station, and Guy R. McDolE, Assistant in 
Soils, Agricultural Experiment Station, University of Minnesota 
INTRODUCTION 
The hygroscopic coefficient expresses the percentage of moisture con¬ 
tained in a soil which, in a dry condition, has been brought into a satur¬ 
ated atmosphere, kept at a constant temperature, and allowed to remain 
until approximate equilibrium with this atmosphere has been attained 
(i2, p. x; 14, p. 76; 15, p. 196). 2 It has a twofold significance, both 
serving as a single-valued expression of the relative fineness of texture 
(12, p. xi), and, in soil-moisture studies, permitting the approximate 
estimation of the maximum amount of water available for growth and 
for the maintenance of life in the case of ordinary crop plants—the 
difference between the total amount of water and the hygroscopic 
coefficient (3, p. 121). 
The error is sometimes made (20, p. 209) of confusing Mitscherlich’s 
“ Hygroskopizitat” with the hygroscopic coefficient as above defined. 
The former is determined by allowing the exposed soils to come into 
equilibrium with an atmosphere in contact with a 10 per cent sulphuric- 
acid solution (24, p. 56) instead of with water, Mitscherlich holding that 
the determination by Hilgard’s method gives results much too high on 
account of the condensation of moisture on the exposed samples (24, 
p. 156). The values obtained by his method are much lower than the 
hygroscopic coefficients, although either will serve to indicate the rela¬ 
tive hygroscopicity and fineness of texture of the samples, as may also 
the much simpler determination of the hygroscopic moisture contained 
in air-dried soils exposed freely side by side to the air of an ordinary 
room for some time (4, p. 351), or that of the moisture equivalent 
(7, p. 140). 
The term “hygroscopic coefficient” was introduced by Hilgard in 
1874 (13, p. 9), but as early as 1859 he had developed the method for its 
determination which he continued to use for more than 50 years, referring 
to the value, previous to the date mentioned, as the hygroscopic moist- 
1 The work reported in this paper was carried out in 1910 to 1913 at the Nebraska Agricultural Experiment 
Station, where the authors were, respectively, Chemist, Assistant in Chemistry, and Research Assistant 
in Chemistry. 
2 Reference is made by number to “ Literature cited,” p. 165-166. 
(147) 
Journal of Agricultural Research, 
Washington, D. C. 
ki 
Vol. XI, No. 4 
Oct. 22, 1917 
Key No. Minn.—20 
