1907.] Agricultural Research. 



393 



funds for local experiments under such terms as will permit 

 the station officers to make the experiments wherever they 

 can best be made, and for such periods as may be desirable 

 to secure definite results. It is better, for example, for the 

 station to be able to conduct experiments temporarily with 

 reference to the introduction of a new crop in 50 localities 

 under the direction of the station agronomist than to have 

 such an experiment made at five permanent sub-stations. 

 If the plant pathologist is called upon to devise a method for 

 the repression of a disease of melons, he should be able to go 

 into the regions where melons are largely grown and make his 

 experiments there. 



Since the agricultural experiment station is not only a 

 scientific institute for the discovery of new truth, but also an 

 agency to promote the advancement of practical agriculture, 

 its relations to the university should be such as will promote 

 rather than hinder its close communication with agriculture 

 and agricultural people. Its work should also be sufficiently 

 broad in scope and character to satisfy the requirements of 

 both science and practice. It should, therefore, have funds, 

 equipment, and workers sufficient to enable it to discharge 

 effectively both its scientific and practical functions. 



The work of the thorough \y organised station should include 

 (1) original research ; (2) verification experiments (with special 

 reference to the loca] application of results of researches made 

 at the station or elsewhere) ; (3) demonstrations of experi- 

 mental results on a practical scale in one or more localities 

 with a view to securing the application of such results to 

 general practice by farmers ; and (4) dissemination of informa- 

 tion regarding the station's work through publications, 

 addresses at farmers' meetings and agricultural schools, exhibits 

 at fairs, &c. 



The stations should have ample funds with which to publish, 

 systematically, accounts of their work in two ways — (1) more 

 or less detailed technical reports as permanent records for 

 scientific readers, and (2) popular and usually brief statements 

 of results and applications for farmers. The popular bulletins 

 should be distributed widely and freely, but publications 

 should not be solely relied on for the dissemination of informa- 

 tion from the stations. Unless these are supplemented by 



