REPORT ON FORESTS. 139 



EFFECT OF FORESTS UPON EVAPORATION. 



A study of this question is of the first importance to the water- 

 works engineer, for it is clear that if evaporation from forested 

 is so much less than from deforested catchments, the amount of 

 run-ofF which may be collected and utilized for public water- 

 supply will be correspondingly increased. We usually estimate 

 that about 14 inches of rain-fall can be collected yearly in this 

 vicinity, and if forests cause an increase of from 5 to 6 inches in 

 this amount, as has been claimed, the difference in the amount 

 collectible will be very large. It is because we believe that this 

 is an error dangerous to water-supply calculations that we have 

 given special study to this phase of the subject. 



The data which throw light upon this subject vary widely in 

 character. First, we have the general facts as to distribution of 

 forests and rain-fall. In the United States it is found that forests 

 are almost invariably abundant where the rain-fall exceeds 32 

 inches per annum, unless there is some peculiarity of the soil 

 unfavorable to tree growth. When the annual rain-fall becomes 

 less than 26 inches, forests almost disappear, but farm crops are 

 raised without irrigation where the annual rain-fall is less than 20 

 inches. We may infer from this that forests require for their 

 support a more abundant supply of water than farm crops. We 

 have a verification of these facts in New Jersey experience. In 

 1 88 1 and 1894 we had protracted droughts, during which the 

 spring and summer rain-fall fell below the equivalent of 32 

 inches per annum. The actual figures were for northern New 

 Jersey in the spring and summer of 1881, 13.16 inches, and for 

 1894, 15.67 inches, whereas the spring and summer rain-fall in 

 Kansas, at points which have an annual rain-fall of 32 inches, 

 amounts to 20.2. During 1881 the drought caused the death* of 

 numbers of forest trees, while those which did not die had their 

 foliage browned as if by fire. The same conditions prevailed to 

 slightly less degree in 1894, and from these lessons we may con- 

 clude that if the rain-fall in New Jersey should fall below 32 

 inches per annum for a period of years, the forests would die. 

 We may conclude, therefore, that in our latitude forests reqiiire 

 a rain-fall of more than 16 inches during the growing season. 



