274 BovieE: NON-AVAILABLE WATER IN SOILS 
plants wilted in a moist atmosphere. The soils used were a mix- 
ture of black beach humus and sand, a loam, and a pure sand 
(ground quartz). He found different amounts of non-available 
water in each of the soils; the beach humus having the greatest 
amount and the sand the least. He pointed out the importance 
of determining the amount of water available to plants in various 
soils, and showed that it bears no fixed relation to the amount of 
water which the dry soil will take up and hold (saturation capac- 
ity). He attributed the greater amount of non-available water in 
the humus soil to the greater adhesion between the water and the 
soil. 
From the time of Sachs’ paper, until Bogdanhoff! published his 
paper in 1893, but little attention was given to the subject of 
non-available water, though soil moisture received the attention 
of a great many investigators. In his paper, Bogdanhoff pointed 
out the errors introduced through chemical changes, when hygro- 
scopic water is measured by determining the loss of weight in 
soils dried in an oven at 100° C. He found percentages of hygro- 
scopic water, as measured by germinating seeds used as an indi- 
cator. Seeds were soaked until they had absorbed amounts of 
water nearly sufficient for germination. They were then mixed 
with the soil to be tested, using an excess of seeds, and stirring 
from time to time, so that all particles of the soil came in contact 
with the seeds. Seeds in this condition have strong absorptive 
powers and ble to take water from a relatively dry soil and begin 
growth. Varying percentages of water were mixed with a par- 
ticular soil, and the lowest percentage which would permit growth 
determined. 
In a sand containing 0.15 per cent. (by dry weight) of water, 
seeds germinated after 19 days; while in a sample containing 0.09 
per cent., the seeds did not germinate, but lost water to the sand. 
He made measurements with samples of clay, but did not compare 
the clay with the sand, since the amount of germination could not 
be accurately determined, and the temperatures in the two sets of 
experiments were not held alike. 
Heinrich,’ in 1894, determined the amount of water contained 
in soils when plants begin to wilt, and also determined the hygro- 
scopic water content. The hygroscopic water was determined by 
» nee Eas 
