226 



AMERICAN FORESTRY 



phenomenon at a time must be varied. 

 If the physician wishes to determine 

 whether mince pie is a cause of indiges- 

 tion, he will not give his patient, at the 

 same time, mince pie, lobster salad and 

 Welsh rarebit, for, if he does and indi- 

 gestion results, he cannot fairly attrib- 

 ute it to the mince pie. If a metallur- 

 gist wishes to determine the effect of 

 various percentages of carbon upon the 

 strength of steel, he must be very care- 

 ful to procure a number of samples of 

 steel precisely alike in every respect, ex- 

 cept that the percentage of carbon va- 

 ries ; and it will not do, in such a case, 

 to say that because there is only a slight 

 difference in the percentage of some 

 other element, such as nickel, the expe- 

 riments prove the effect of the carbon, 

 because it may be that an extremely 

 small variation in nickel would have 

 more effect than a large variation in 

 carbon. Only one element must be va- 

 ried at a time, %nd this is an extremely 

 important consideration. 



A second difficulty arises with the in- 

 ductive method if the observations van- 

 very greatly from each other. If, for in- 

 stance, it is desired to find the effect of 

 Phenomenon A upon Phenomenon B, 

 and if this effect in any case is small, 

 while Phenomenon B, is exceedingly 

 variable, it may be difficult to show by 

 statistics that Phenomenon A has any 

 effect at all, though that effect may be 

 unquestionable. There are other mat- 

 ters affecting the general methods of 

 scientific investigation which are sug- 

 gested by this paper, but they will be 

 taken up in connection with the specific 

 points in Professor Moore's paper 

 which illustrates them. 



EFFECTS OF FORESTS ON RAINFALL 



Fourteen pages out of thirty-six in 

 Professor Moore's paper are devoted to 

 a discussion of the relation between 

 forests and rainfall. 



This relation is of no consequence so 

 far as concerns the Weeks bill, or the 

 claims made by those who believe that 

 the government should acquire forest 

 reserves. No one has insisted, so far as 

 I know, that forests have any consider- 



able effect upon rainfall. Mr. Pinchot in 

 his "Primer of Forestry" states that no 

 generally accepted result has yet been 

 reached in the matter ; that the observa- 

 tions are not all in agreement, and con- 

 cludes by saying: "The truth probably 

 is that more rain falls over the forests 

 than over open country similarly placed, 

 but how much more it is impossible to 

 say." Professor Moore deals with this 

 subject as though it were one of the 

 most important involved. He quotes 

 from several authors with reference to 

 desiccation in other countries, where the 

 rainfall, once abundant, has become 

 much less or disappeared entirely, and 

 says that "this decrease or precipitation 

 might better be regarded as the came 

 rather than as the result of the barren 

 condition of the soil" * "that 



the forests ceased to exist when the 

 rainfall became deficient ;" and he seems 

 to think that this proves that forests do 

 not increase the rainfall. He compares 

 the rainfall records in New England 

 and in the Ohio Valley for the past 

 seventy years and concludes that it is 

 not decreasing. His conclusion on this 

 point is stated in the following para- 

 graph : 



"In New England, where deforestation 

 began early in our history and has been 

 extensive, the mean of the fluctuations of 

 the rain-curve is a steady rise since 1836 

 up to a few years ago, and in the Ohio 

 Valley, where the forest area has been 

 greatly diminished, there has been no 

 decrease of rainfall shown by the aver- 

 age of the fluctuations of the curve. 

 These facts are important and cannot be 

 successfully disputed." 



A careful examination of Professor 

 Moore's statements will show that they 

 offer no proof whatever that forests do 

 not affect precipitation. It is admitted 

 that climatic and geologic changes are 

 taking place very slowly at many places 

 on the earth ; mountain ranges are being 

 elevated, other districts are subsiding, 

 ocean currents are changing, and in the 

 course of thousands of years variation.- 

 in climate, including temperature and 

 rainfall, result. Undoubtedly forests 

 have disappeared in many places be- 

 cause the rainfall has been diminished : 



