SEEPAGE OR RETURN WATERS FROM IRRIGATION. 27 
sibility of separating the drainage of extensive areas of irrigated 
land from land which receives little or no irrigation. 
COMPLICATING CONDITIONS. 
§ 20. The bed is almost invariably of sand of unknown depth. 
In a few places the rock of the ridges cut through by the river 
shows at the surface. If there is any flow in the sand, it may 
be expected to be forced to the surface at such places and increase 
the volume of the stream. Such a place is just above the 
Bijou creek, above Fort Morgan. And again below the Hardin 
ditch and above the Corona ditch the bluffs on the south side of 
the river are prominent, and show evidence of rock outcropping 
across the river. The sand is porous and has the capacity to hold 
much water. The results which may be met with in the natural 
inflow are masked by the effects due to the varying distance to 
the rock. These may sometimes be more than sufficient to counter¬ 
balance the increase from the inflow. This may explain the unusual 
gains noticed in several places and the losses which are found in 
certain stretches, even where an area of irrigated land is tributary 
to that section. The most marked case is at the mouth of Bijou 
creek. In the measurement of 1894, which was made above and 
below the Bijou, a gain was looked for in the few miles between the 
two measurements. The Bijou drains some 1,400 square miles. 
Besides, there was some water evident on the south side seeping into 
the river. Nevertheless, a loss was found in 1894, and in 1895 on 
making another test the gain was so slight that it may be called a 
loss. In both cases the Platte & Beaver canal was measured sev¬ 
eral miles from its head, and the loss for the few miles if considered 
may make a slight gain. But with all allowance possible for this, 
the gain is slight, or an actual loss which the measurements show. 
Moreover, at the last point of measurement, there are practically no 
bottom lands. 
§ 21. The methods of irrigation on the lower Platte are some¬ 
what different from those on the Cache a la Poudre, and this may 
account for the difference in the relative magnitude of the result. 
The Poudre being a mountain stream, fed almost solely by melting 
snow, is low in the autumn and late summer. On the Platte the 
summer flow is small, being reduced both by the usage above and 
by the avidity of the sands and atmosphere. In the fall, however, 
the seepage from all the streams nearer the mountains—Clear creek, 
Boulder creek, St. Vrain, Big Thompson, and Cache a la Poudre— 
pour the seepage from these channels and the greater part of the 
flow received from the mountains into the Platte. As a con¬ 
sequence the Platte is higher in the fall and winter. This gives 
the settlers along the Platte opportunity to irrigate extensively in 
the fall, and as late as the ditches can well carry water; they thus 
