250 ANNUAL REPORTS OF DEPARTMENT OF AGRICULTURE. 



ment stations, where continuous records and observations can be 

 maintained indefinitely. 



In the numerous studies in tree phintin<^, wliich have occupied 

 much attention for the last 5 or 10 years, much progress has been 

 made. These studies have been directed hu'f^ely toward determining 

 the best methods of nursei-y practice and of iiehl planting and sow- 

 ing. They have aided largely in developing the science of tree 

 planting in the National Forest regions, and their results are being 

 constantly utilized in the large-scale operations carried on by the 

 Service. The planting investigations Avhich now remain are largely 

 observations over long periods in order to establish the final con- 

 clusions definitely and the improvement of methods whose general 

 features have been developed. Meantime greater emphasis is being 

 given, in research, to the best methods of cutting in order to secure 

 satisfactory natural reproduction and rapid growth of desirable 

 species. This will be a factor of importance in the national move- 

 ment for the prevention of forest devastation. 



Studies of existing forests to determine the rate of growth and 

 yield of forest products which may be expected at different ages 

 when the forests are properly managed have already furnished the 

 basis for regulating the cut on a number of National Forests with 

 a view to a perpetual yield of timber. 



Investigations in Arizona and New Mexico which have been under 

 way for a number of years have yielded conclusive evidence that 

 satisfactory natural reproduction of western yellow pine, the species 

 of chief commercial importance in that region, can be secured only 

 at wide intervals, averaging probably about 20 years. The high 

 value of such natural reproduction, when it does occur, requires that 

 every reasonable safeguard be provided to protect it from injury. 

 These facts have led to unavoidable reductions in the live stock 

 grazed on certain National Forests in the Southwest, in order to 

 reforest cut-over areas. 



The investigation of the effect of forest cover on stream flow, which 

 is being conducted at the Wagon Wheel Gap Experiment Station in 

 Colorado, in cooperation with the Weather Bureau, has entered its 

 most interesting phase as a result of the removal of the forest cover 

 on one of two adjacent watersheds under observation. During the 

 preceding eight years the behavior of the two streams at all times of 

 the year and under all conditions was definitely established by sys- 

 tematic and continuous records. 



As a result, it will be jxissible to predict what the flow of the stream 

 from the denuded watershed would luive been under practically any 

 conditions if the forest cover had remained intact. During the 

 second phase of the experiment the same observations will be con- 

 tinued, to determine what effect the removal of the forest cover will 

 have on the flow of the stream coming from that Avatershed. Already 

 a marked difference has been observed in the behavior of this stream 

 in that the spring flood occurred much earlier and the water ran off 

 more rapidly than Avould have been expected under undisturbed con- 

 ditions. 



In southern California, where the regulation of stream flow and 

 prevention of erosion by forest cover are of the greatest importance 

 to large areas of valuable irrigated lands, observations on water- 



