10 FOEESTS, WOODS, AND TKEES 



Eussian river in the plain and the American rivers in the 

 mountains. Oppokov's conclusions are as follows : The 

 flow of the Dnieper is closely related to the amount of 

 rainfall in the whole basin. Eainfall and evaporation are 

 the most important factors influencing the level of the 

 river. The influence of woods and also of peat-bogs has 

 been greatly over-estimated. He even believes that a 

 considerable amount of vegetation in a river basin may 

 lower the level of a river owing to the amount of evapora- 

 tion set up, and says that the best conservers of water are 

 not peat-bogs or forests, but beds of sandstone. 



Mr. Cecil H. Eoberts, C.E., has made investigations on 

 the climate and physical features of the basin of the river 

 Dee, in connection with proposals for the extension of the 

 Aberdeen Waterworks. These investigations are described 

 in a paper as yet unpublished. Mr. Eoberts has not been 

 able to trace any influence either of forests or of the felling 

 of large areas of wood on the maximum or minimum flows 

 of the river (10). 



As the results of observation are capable of varied 

 interpretation, it is of interest to record that the influence 

 of forests on stream-flow is now being experimentally 

 studied (11). This important experiment, which will 

 probably settle the question, was inaugurated in 1910 in 

 the Eio Grande National Forest in the Eocky Mountains of 

 Colorado. This experiment involves the careful measure- 

 ment for a number of years of two streams flowing out of 

 two well-wooded watersheds ; and later a comparison of 

 the flow of these streams after the forest cover has been 

 removed from one of the watersheds. Dams, weirs, and 

 recording instruments for measuring the flow of the streams 

 have been installed, as well as instruments for measuring 

 temperature, rainfall, evaporation, humidity, and other 

 atmospheric factors that may affect the flow. All outside 

 . factors will be eliminated ; and the records at the end of 

 ten or twenty years are expected to throw much light on 

 the relation of the forests on mountain watersheds to the 

 flow of the mountain streams. 



