60 
SEEPAGE OR RETURN WATERS FROM IRRIGATION. 
the upper portions of the stream. The earlier canals, some built by 
the Moors before 1000, A D., were taken out from the lower por¬ 
tions of the stream, the later ditches near the head. 
The lower canals desired to close the upper ones. The latter 
claimed that the water that was applied by them in irrigation 
returned to the river to a great extent, and thus had the effect of 
making the stream more constant in its flow, and, therefore, was as 
a whole advantageous for the lower users. During several years a 
system of measurements was carried on at different places on the 
stream and included all the water that came into the stream 
through the smaller tributaries. Measurements were made daily by 
the local officers. The valley is one the total length of which is 
something like fifty or sixty miles, and the total area irrigated is 
32,000 acres. The cultivation consists largely of wheat, beans, 
alfalfa, meadows, and gardens, with small quantities of potatoes and 
flax. A biennial rotation is practiced which dates from the Moors. 
Grain is usually watered three times, once at the time of sowing, in 
November. Irrigation is practiced throughout the whole season. 
In the upper valley wheat is not watered. Beans are watered from 
the middle of July to the middle of September. From the data 
obtained, M. Vigan reached the following conclusions ; The return 
waters are derived from all irrigated lands of the valley, varying 
according to the crops, amount of water used in each season, thick¬ 
ness of the soil, its composition, and the slope of the impermeable 
layer. He concludes that, in the bottom lands, whi(^i form a bed 
about a mile wide along the stream, and are abundantly watered, 
from the first of March, that the return waters from this source are 
sufficient to compensate for the losses caused by irrigation during 
the greatest part of the low water. He also concludes that, in the area 
forming a strip two or three miles wide, with a very deep layer of 
permeable soil, the return waters come to the surface only in some 
places; that the greatest part of the springs which are caused flow 
unused in the subsoil and return frequently to the sea. On these 
lands irrigation occasions considerable loss; hence he concludes 
that, in case of an application for water right in the stream for 
canals, or ditches, which are to be newly constructed, the concession 
should be refused, except conditionally. In case water is lacking in 
the other canals, then the new ones should be closed. In general, 
under the conditions existing in that valley of the Tet, irrigation at 
the upper portions of the stream with water taken at periods of high 
water, is beneficial to the lower portions of the stream. The water 
thus applied gradually returns to the stream in such quantity that the 
stream is not so low as if the irrigation had not been practiced. 
Some of the measurements of the Poudre river have been given 
in the Colorado Agricultural Experiment Station Report, 1891, 
p. 45-50. 
