68 BULLETIN 1017, IT. S. DEPARTMENT OF AGRICULTURE. 
of the mean year. The maximum storage under these conditions 
is seen to be 5.9 billion cubic feet, at water elevation 979. 1. The 
reservoir in this case would be empty by October 1. 
In diagram C, figure 20, the run-off curve was made up of the 
spring run-off of the mean year (January- June) and the maximum 
summer and fall run-off (July-December). Maximum storage here 
is shown to be 7.2 billion cubic feet, at elevation 980.5. The reservoir 
would be empty by the middle of December. 
In diagram D, figure 20, the most extreme conditions that may 
be expected are shown. The run-off plotted here is for the hypo- 
thetical maximum year as stated on page 64. This diagram shows 
that the maximum storage required in the most extreme case (that of 
closing the reservoir during the last 15 days of August) would be 
12.1 billion cubic feet, if the rate of outflow were 1,000 second- feet 
continuously after May 1. However, before the water surface 
reached elevation 985.1 required for this amount of storage, the spill- 
way alone would be discharging at a much greater rate than 1,000 
second-feet, so that this stage would not be reached under these con- 
ditions and there would be a minimum freeboard on the clam of more 
than 3 feet. By regulating the outflow during September at 2,000 
second-feet, which could be taken care of by the outlet channel if 
necessary, the storage could be reduced to 7.8 billion feet by October 
1, and with a flow of 1,000 second-feet from then on the reservoir 
would be emptied by early in January. If the heavy July run-off 
were to occur in June, with that of June transferred to July, the 
maximum storage would be 10.6 billion cubic feet. If in addition to 
closing: the reservoir for 15 days at one time it should become neces- 
sary. on account of some great storm, to close it at another time for a 
short period, say 4 days, an increase in the rate of outflow to 1.400 
second- feet would in 10 days bring the storage back to normal. In 
all cases the reservoir would be empty before the heavy spring run- 
off of the following year begins. These conditions are extreme and in 
all probability never will occur; but should such a combination of 
run-off come, it can be controlled without damage to the reservoir 
and with less damage from flooding in the lower valley than would 
result from the unhampered flow of such run-off. 
Should even worse conditions prevail than have here been con- 
sidered, the reservoir would still be safe from destruction. The spill- 
way and culvert together would be capable of discharging more 
than 4,000 second-feet when the water reaches elevation 985.0. in the 
reservoir. Before the water could rise high enough to overtop the 
reservoir the discharge through sluiceway and spillway would be 
more than doubled. To raise the reservoir to a height that would 
involve this rate of discharge would require a rate of run-off very 
much higher than has ever been recorded in the Red River watershed 
