108 



should be authorized until it has been examined and approved by the 

 State irrigation authorities. It requires expert knowledge of the irri- 

 gation conditions of a stream to determine the exact influence of a 

 transfer. The full effects of return seepage do not manifest them- 

 seh'^es the first year, nor can they be foretold in all cases. Hence when 

 appropriations are shifted from year to year there can be no stability 

 or certainty with respect to the amount of return seepage or the vol- 

 ume of Avater which the water commissioner can count on at any 

 point in the stream as available for the use of appropriators. With- 

 out a knowledge of the manner in which the stream loses water by 

 evaporation or is augmented by seepage, it is inevitable that great 

 injustice in the division of the water sujjply between priorities must 

 result. 



Refusing to permit of sales of appropriations and restricting rights 

 to the use by which acquired has this inestimable advantage : It in- 

 sures stability so far as return seepage is concerned. Where an 

 appropriation is always used on the same area the return waters 

 always reinforce the stream in the same place and are always availa- 

 ble for the same users below. Under this interpretation of user's 

 rights, A, when he irrigates his quarter section, acquires only a right 

 to the water needed for that quarter section, and if he ceases to irri- 

 gate this land his right ceases and the appropriation reverts again to 

 the public. The statutes of many of the States seem to be based on 

 this idea, as they state that wherever a use ceases the right ceases. 



If appropriations are measured by some particular use, those for 

 irrigation will vary in the different months of the irrigation season 

 with the needs of the crop grown on the land, and they will vary 

 somewhat in different years. In a rainy year little water will be re- 

 quired ; in a dry year more. The right will be limited to the irriga- 

 tion season. This on the Platte will be for about one-third of the 

 year. Mr. Adams found the average length of the irrigation period 

 in 1903 was 114 days. The records of irrigation on the Arkansas in 

 190-1 shoAved the average for the season on that stream was 69 days. 

 Such rights differ Avidely from an appropriation of a certain number 

 .of cubic feet per second flowing continuously. These are not riglits 

 of use, but assumed equivalents of such rights. They give a surplus 

 in August and September, when water is most valuable, or, if based 

 on the quantity used in August, limit irrigators unjustly in June. 

 Hence a right to a uniform flow of a certain number of cubic feet per 

 second is never in exact accord with the actual requirements of irri- 

 gation. 



The canal companies, which have acquired large appropriations of 

 water for supplying the needs of farmers instead of for use them- 

 selves, have recognized the necessity of attaching the rights which 



