70 BOARD OF AGRICULTURE. 



yielded 27.35 bushels; while another yielded 63.85 bushels, or 

 more than twice as much as the first. The yield of stalks on 

 different plats varied one-half. It is not strange that Professor 

 Miles closes his account by saying : " From the wide range of 

 variation of these plats, all treated in the same way, it will be 

 seen that the results of a single field experiment in the appli- 

 cation of manure cannot be relied upon to establish any rule of 

 practice." " Improvements in agriculture can only be made by 

 means of a systematic series of experiments, so conducted as to 

 guard against all sources of fallacy, and then carefully repeated 

 under a variety of circumstances." 



I might add that similar results have been reached on our 

 experimental farm connected with the University of Wisconsin. 

 In experiments on potatoes cultivated exactly alike, results ap- 

 parently widely at variance with each other have been reached. 

 Like conflicting results have been reached at Lansing in experi- 

 ments in the feeding of animals, especially swine and sheep. 

 The same feeding would cause one animal to gain and another 

 to lose, when no assignable cause for the difference could be dis- 

 covered. Among the sheep there was a loss of weight in all 

 the pens one week, for which the Professor adds, " The only 

 cause that could be assigned was the change of management of 

 other sheep in the same building." The sheep not under ex- 

 periment were turned out by day and returned to the barn at 

 night, and this so disturbed the experimental sheep that there 

 was a general loss of weight among them, although the ration 

 of food was increased. 



We give the Professor's conclusion in his own language, which 

 we consider well established by all these experiments as well as 

 by the many others that have misled those that trusted to them. 

 " It is exceedingly difficult to conduct any kind of experiment 

 in practical agriculture in a satisfactory manner, from the great 

 variety of circumstances that tend to modify results ; but when 

 the subtle principle of life as exhibited in animated beings is 

 involved in the line of investigation, the difficulties in the way 

 of exact determination seem almost insuperable." Is experi- 

 menting, then, in agriculture, a hopeless undertaking ? By no 

 means. The facts already cited only show the great difficulty 

 of the work, or rather the careful training the experimenter 

 must have to make his work of any value. Time and care will 



