METHODS OF FOREST POLICY. 241 



on account of the time element involved in most 

 forestry experiments. In agriculture the answer 

 to an inquiry may be often secured in inexpensive 

 ways, and may be given in one season ; while in 

 forestry, years of patient waiting and observation, 

 wholesale methods or measurements, large areas, 

 and a large number of cases, are required to 

 permit generalization. In both directions the 

 activity of the private investigator is at a disad- 

 vantage. To conduct investigations that must be 

 continued for decades, and in a large way, a sys- 

 tematic plan and organization is needed, such as 

 only a public institution usually has at command. 

 Moreover, comparability of results can be secured 

 only when uniformity of method has been assured, 

 and this again is more likely secured by coopera- 

 tion between state institutions, or even by the char- 

 acter and organization of a single state institution. 

 The advantage of connecting such experiment 

 stations with institutions of learning needs hardly 

 any argument ; the mutual increase of educational 

 facilities and opportunities is patent. These edu- 

 cational means can, of course, be extended by 

 proper methods of publication . of results', by or- 

 ganization of meetings for their discussion, . by 

 so-called university extension, and iinally, by the 

 -promotion of associations which have for their 

 object-the increase of appJication of knowledge in 

 the actual forestry practice. Such associations 

 give opportunity of impressing and driving home 



