4 EXPERIMENT STATIOX RECORD. 



conviction locally the facts must be presented through experiments 

 by the local station, and this has been the excuse for a kind of experi- 

 mental work which in reality lies in the field of demonstration. Ex- 

 periments which merely demonstrate locall}^ facts already developed, 

 without adding anything to the understanding of the factors that 

 limit or aflPect them, aim to teach and are now in the province of the 

 extension service rather than of the experiment station. 



In the past there has been no comprehensive general plan beyond 

 the individual stations. As a result they have not supplemented each 

 other in their work and extended their findings in the most effective 

 way. Because the work has been so independent and individualistic, 

 it has frequently not connected up in a way to fully cover the ground 

 or be entirely conclusive as far as it went. Sometimes it has not 

 seemed to take full account of the work and the results elsewhere. 

 This is apparent to any one who examines the combined product of 

 the past, or attempts to sninniari/e or digest it. As time goes on the 

 weak spots are strengthened and the omissions filled in. but in the 

 meantime it remains patchy. 



The extent to which unwarranted duplication and repetition are 

 carried on has been a fiequent subject of criticism of our stations, 

 and has sometimes led outsiders to infer that the proportion of new 

 or original work was relatively small. The necessity for repetition, 

 especially under a suflicicnt variety of conditions, is not to bo (lues- 

 tioned, but at the present stage its defense lies in a well-directed 

 attempt to verify doubtful points or add to the information. Tho 

 economic results will hardly be twice the same in tho same locality, 

 because the economic relations are subject to constant change: and 

 strictly local features and conditions will often affect the general 

 results of other classes of experiments to a considerable degree. 

 These facts impose a practical limitation on the extent to which 

 repetition to ascertain within narrow limits the local or current 

 application of an experiment is feasible. 



Manifestly every new experiment ought to be undertaken with 

 full knowledge of the status of the subject, and so planned as to form 

 a definite contribution as far as it goes. In order that it may ad<l 

 its mite to the sum of definite information, it should be conducte<l. 

 as Director Kilgore has stated, on a basis which will make the result- 

 comparable with those of other workers. This is especially true, of 

 course, of the more elementary experiments. These should be on a 

 basis or plan that will enable direct comparison and summation. 

 Then new experiments will at least mean additional evidence which 

 can be applied in strengthening the suggestions or conclusions from 

 previous experiments. But where the work is done from a too local 

 standpoint new factors are often introduced or limiting ones left 

 unchecked which in the end make the experiment stand practically 



