506 EXPERIMENT STATION RECORD. 



burying older erroneous interpretations. The business is seldom 

 accomplished by direct attack, for error perishes only in the light of 

 truth, as micro-organisms are said to perish suddenly when struck by 

 ultraviolet rays. Owing to the load of false theories, we work like 

 a mental chain gang and are never unfettered. The handicap im- 

 posed by wrong hypotheses has always impeded the growth of 

 science." 



Agriculture is rich in illustrations of false hypotheses and wrong 

 theory. Many of these trace back to beliefs in practice which have 

 become deep-seated traditions, and which are frequent stumbling- 

 blocks to the investigator in his work and to the teacher in applying 

 the results of scientific deductions. This is partly because the art 

 developed in advance of the science and partly because of the com- 

 plexity of phenomena as presented in agriculture. Farming by the 

 moon long held sway, and to test the possibility of any foundation for 

 these theories a series of experiments was started several years ago 

 by a French investigator upon the effect of the moon's phases in 

 relation to the time of planting seeds. In the science of agriculture 

 the problems are so intricate and complicated as to give little wonder 

 that misleading deductions have been made, and erroneous theories 

 have temporarily been given the stamp of scientific truth. But this 

 emphasizes the need of adequate proof, which, as Dr. Minot says, is 

 the pivot of all research. 



Judgment as to when proof is adequate varies widely, and stand- 

 ards of practice are not always apparent. 



"It would certainly aid science if some competent philosopher 

 should make a study of the practice of investigators in the various 

 branches of science sufficient to render clear the general principles 

 by which investigators decide when a new observation or a new in- 

 duction is sufficiently proven. If we follow the advance of research 

 in any particular direction, we soon realize that there is a more or 

 less definite standard of proof, which, though never clearlj^ formu- 

 lated, is none the less insisted upon, so that any paper which does not 

 come up to this standard is subject to unfavorable criticism. Two 

 elements of this standard we know, the first the elimination of the 

 recognized sources of error; second, the repetition of the observations 

 so that the constancy of the phenomenon is assured. 



"The investigator lives in an atmosphere of concentrated uncer- 

 tainty, for he is convinced that at any time new data may turn up 

 and that all generalizations are likely to require modification. We 

 might well adopt as our cry — Incredulity toward the known; open 

 credulity toward the unknown. 



"We think of science as a vast series of approximations, and our 

 task is constantly to render our approximations closer to absolute 



