310 A NATIONAL PLAN FOR AMERICAN FORESTRY 



and they would not apply to the normally litter-covered soil of a forest. 

 European evidence, as quoted by Zon, 13 shows wide variations, but 

 indicate that evaporation from bare soil in the open, under average 

 conditions, amounts to about 50 percent of precipitation; and that a 

 forest, even without leaf litter, may reduce this to 15 to 25 percent. 



CONSUMPTION OF WATER BY FOREST VEGETATION 



The water which all plants rooted in the soil withdraw from it in 

 maintaining growth and life is transpired, or given out into the air, 

 chiefly from the leaves. It is very difficult to measure accurately the 

 transpiration from a single tree beyond the seedling stage, and in- 

 finitely more so to measure the transpiration from a forest. Blaney 

 et al. 14: employed observations of stream flow to determine the water 

 evaporated from the soil or consumed by canyon-bottom vegetation 

 " willows, tules, and kindred moist land growths" in southern 

 California. 



The evapo-transpiration losses from Temescal Canyon during only 

 30 spring days they found to equal 12.9 inches of rainfall. The same 

 author 15 estimated from stream-flow measurements in Coldwater 

 Canyon that the transpiration losses from " alders, sycamores, bay, 

 oak, and herbaceous growth" during the 6-month summer season 

 of 1931 was 45 inches per acre. Evaporation was judged to be small 

 as the water in the canyon bed is constantly cooler than the air. 

 Inasmuch as the precipitation for the entire year is normally only 

 about 30 inches, it is fortunate that the area of canyon-bottom vege- 

 tation is very small, and that the loss per acre of entire watershed 

 is only 0.10 inch per mile of canyon. The transpiration losses just 

 described are probably at or near the maximum for any forest type 

 in the United States, and fully warrant the expedient, already adopted 

 by such cities as San Bernardino, of piping water out of the stream 

 channel before it can be consumed by the canyon-bottom forest. 

 That the forest cover of the slopes and ridges in this region does not 

 begin to make the same demands on soil moisture is very clear from 

 its dwarfed development. 



Data on transpiration rates for other American forest types are 

 entirely lacking and these rates may only be inferred from general 

 knowledge. Interception of precipitation, evaporation from the soil, 

 and transpiration account for a very large part of the difference be- 

 tween the total precipitation over a watershed and the flow of the 

 stream draining it. These differences have been earlier described 

 for various parts of the country. Transpiration probably fully equals 

 the other two factors combined in the hardw^ood forests of the humid 

 eastern United States, but in the evergreen and chaparral forests 

 East or West, may be subordinate to either. 



INFLUENCE OF FOREST LITTER 



Probably more important than any of the previously listed influ- 

 ences of the forest on run-off and stream flow is that exerted by litter. 



1 3 Zon, R. " Forests and water in the light of scientific investigation." Final report, Nat. Waterways 

 Com., Sen. Doc. no. 469, 62d Cong., 2d sess., 1912. 



14 Blaney, H. F., Taylor, C. A., and Young, A. A. " Rainfall penetration and consumptive use of water 

 in the Santa Ana River valley and coastal plain." Calif. State Bui. no. 33. (Calif. Dept. Public Works, 

 Div. Water Resources, in coop. U.S. Dept. Agr., Bu. Agri. Engineering) 1930. 



Blaney, H. F., Discussion of " Forests and Stream Flow. " Proc. American Soc. of Civil Eng., 

 December 1932. 



