LAND DRAINAGE BY MEANS OF PUMPS. 23 
fall and winter the rains and snows will have completely saturated 
the ground and filled the ditches. Hence, those districts which have 
had the longest experience in pumping are unanimously of the 
opinion that even though at that time there is no danger of imme- 
diate damage to crops, still it is best to pump occasionally during 
the winter so as to maintain the ground water constantly at a low 
level, and thus secure the great advantage of early drying of the soil 
in the spring and the opportunity for early cultivation and planting. 
Moreover, the lower the normal ground-water level is kept below the 
surface, the less will be the flooding effect of a severe storm. 
It should be clear from the foregoing discussion that a district 
which gives especial care and attention to its pumping, running 
throughout the winter as needed and frequently during the growing 
season, so as to maintain a minimum ground-water level while there 
is any danger from the occurrence of heavy and sudden rains, will 
require a smaller pumping plant than would otherwise be needed. 
From a surface covered with a growing hay or grain crop the run- 
off is less rapid than from land in corn. For corn and hay the ground 
water must be kept at a lower level than is necessary for grain. Corn 
is considered the most profitable crop in the region under considera- 
tion, and where the soil can be kept sufficiently dry it is raised as 
frequently as is consistent with a proper crop rotation necessary for 
maintaining the fertility of the soil. Much land can be observed, 
however, which is not sufficiently drained for the most successful 
raising of corn and which therefore is devoted to wheat and oats 
instead. 
The amount and distribution of the rainfall are the chief factors 
determining the size of pumping plant required in a given location, 
and we shall proceed to a detailed discussion of all the evidence that 
it has been possible to gather bearing upon this point. From a con- 
sideration of all the data collected to date, the conclusion seems justi- 
fied that the run-off coefficient of 0.3 inch in depth over the whole 
drainage district in 24 hours is the correct one to use in designing 
pumping plants on the Illinois and upper Mississippi Rivers. Direct 
observations on numerous plants in actual operation for a series of 
years will be needed to settle the question finally. 
In the region just referred to, though some pumping on an inade- 
quate scale has been carried on for about 15 years, all the older dis- 
tricts had until comparatively recently experienced trouble with 
their ditches and levees which prevented their pumping experience 
from furnishing conclusive data; from no plant has it been possible to 
obtain definite records going back more than a few years. Now, 
though there are more than 20 large plants in operating condition, 
it is surprising to learn that the majority of these still neglect to keep 
even the simplest, most fundamental records, such as would deter- 
