14 BULLETIN 791, IT. S. DEPARTMENT OF AGRICULTURE. 
to the plant 10 days earlier than in the two lower depths. During 
the driest period of the season, usually beginning about August 10 
in the 0-6 inch depth, and about August 20 in the deeper layers, the 
main root system often occupies soil whose moisture content is well 
below that at which vegetation can absorb moisture. The fluctua- 
tions in the water content observed to occur in the superficial soil 
layer and the practical absence of such fluctuations in the deeper 
layers, notably the 12-24 inch depth, are chiefly accounted for by 
the fact that the supply of moisture in the upper stratum, in which 
the greater portion of the feeding roots are located, is used up by 
the vegetation at a relatively rapid rate. Since the upper soil layer 
is especially rich in organic matter, hence is capable of absorbing 
a very high percentage of water (45-65 per cent) as it percolates 
through the matlike layer, the rapid desiccation of the superficial soil 
is all the more significant. Thus it will be seen that the reduction 
in the water content in the superficial layer to a point below the 
amount necessary to n>ake it available to the use of the plant was 
reached as early as August 10, so that any water absorbed by the 
plant later in the season had to be obtained at a depth greater than 
6 inches. The two lower depths of soil, it will be noted, likewise 
became desiccated after August 20. Therefore at the end of the 
growing season the 0-6 inch layer of soil was 5 per cent below the 
point of available moisture, while the 6-12 and 12-24 inch layers 
were 3.3 and 2.1 per cent below, respectively. In general the growth 
and seed production are completed in the case of small wheat grass 
by August 15, when the herbage dries up and remains dormant until 
the spring. 
Section B of figure 5 represents the moisture conditions on a blue 
bunch grass area during the growing season 1915, the soil samples 
from which the data were obtained being taken simultaneously with 
those represented in section A. Comparing first the general position 
of the respective curves, it will be seen that the water content was 
greater on the blue bunch grass area than on the turfed wheat grass 
area in each period, with the exception of September 10 to 20 in the 
12-24 inch depth, prior to which growth had been arrested. In con- 
trast with the condition on the turfed area, the moisture content in 
the 0-6 inch layer on the bunch grass type was appreciably in excess 
of that at greater depths during the first four periods. In the fifth 
period the moisture content decreased rapidly in the upper soil layer 
and dropped below that recorded at the two lower depths ; but in the 
sixth period, as the result of a fairly heavy rainstorm, the moisture 
content again exceeded that in the lower soil layers. Thus, instead 
of the 6-12 inch soil depth containing the highest percentage of 
moisture during the most active period of growth, as in the turfed 
wheat grass type, the 0-6 inch layer contained the maximum amount. 
