New YorJP Agricultural Experiment Station. 393 
All of the data given indicate that this section is one where drain- 
age (percolation) is very small, and where the evaporation is nearly- 
equal to the water which actually enters the soil, the water discharged 
by the streams being mostly derived from the flood waters of heavy 
showers in summer and from melting snows in spring, which have 
never entered the soil. It is impossible from any data now at hand to 
determine the flood water, but that it is considerable is shown by the 
volume of water discharged by streams during freshets. It probably 
amounts in this locality to four or five inches per annum. 
In our lysimeters, where all of the water falling upon the surface is 
compelled to enter the soil, and where exceptionally good drainage is 
provided, the drainage water cannot be much in excess of the flood 
water, and in lysimeter No. 1 is no doubt less. Owing to the facility 
with which surplus water drains from the lysimeter soil, this soil is 
liable to become abnormally dry, and being cut off from the natural 
water-table, redistribution of moisture by capillarity does not occur to 
the same extent as in natural soils. This tends to reduce evaporation 
from the lysimeters below the average, but its effect is to some extent 
offset by the larger amount of water which the lysimeter soils absorb, 
so that the total evaporation from the lysimeters may approximate to 
that from soils outside. It is probably less in quantity, and when 
expressed in per cent of the water which enters the soil must be very 
much less. The drains which pass through the experimental plats 
deliver very little water during the summer months, and the soil is 
apparently dryer at the end of the season than in the spring, indicat- 
ing what I believe to be true, not only for this, but for most localities 
devoted to agriculture, that during the warm months evaporation con- 
siderably exceeds the rainfall which enters the soil. I believe, more- 
over, that throughout the whole season the evaporation from agri- 
cultural lands nearly equals the rainfall which directly enters the soil. 
Most of the water which penetrates the soil to the level of the water- 
table is precipitated during the winter months, when the surface of the 
soil is frozen, and much of it enters the ground through crevices 
without leaching. In consequence of this the water which reaches the 
lower strata of the soil carries with it very little of the soluble 
elements of plant food contained in the surface soil. This assumption 
is confirmed by the small quantity of those elements found in well and 
spring waters in most localities. On the other hand, there is a gradual 
and almost constant movement of water by capillarity towards the 
surface to supply that lost by evaporation, which tends to concentrate 
the soluble matters near the surface, where they are most available for 
plants. 
50 
