AGRICULTURAL EXPERIMENTS. £85 



otherwise the experiments cannot be compared. If I feed corn 

 meal in a particular way iu Michigan, and it is fed in a' different 

 way in Illinois, Missouri or Pennsylvania, we cannot compare re- 

 sults at all. You are not assisted in the one place by the investi- 

 gations made in the other. The experiments must be conducted 

 in all places in precisely the same manner ; then you can compare 

 results, bearing in mind that certain conditions which cannot be 

 controlled may influence the result. 



The next class of experiments which I will suggest would be to 

 determine the best methods of applying manure — the application 

 of manure on the surface, or the application so that it may be 

 plowed under — and it would be desirable, likewise, to make 

 experiments with reference to the application of these manures at 

 different seasons ; whether it would be best to apply manure in 

 the spring or in the fall. This will be found to be a very compli- 

 cated problem, and one that will require great caution. I would 

 endeavor, likewise, to ascertain the value of commercial fertilizers 

 as compared with barn-yard manures. You want to know the 

 constituents of which the manure is composed, and the fertilizer 

 should be analyzed carefully. Barnyard manure must be our 

 staple. If we purchase other manures we ought to know how it 

 compares with this. 



I would like to have experiments made showing the effects of 

 change on the growing crops. In a system which has been tried, 

 wheat has been grown for quite a number of years in succession, 

 and a yield of from thirty to forty bushels obtained without any 

 manure. The system is this : The field is divided into strips of 

 three feet in width. The wheat is sowed on the alternate strips. 

 The vacant strips are kept thoroughly pulverized during the 

 season, and the uext year the wheat is put on the strips left bare 

 the year before. 



It would be desirable, likewise, to ascertain the methods of 

 seeding corn — whether hills or drills would be desirable. These 

 experiments all require a great deal of care and accuracy, and 

 they will require a good deal of study in order to fix the condi- 

 tions upon which they shall be tried by a large number of persons, 

 so that the results may be compared. 



Prof. Hamilton. I scarcely know how to get at this matter. 

 It seems to me, that before we start out to experiment at all, we 

 should understand one thing, and that is, that in this art of agri- 

 culture, as it has been defined, aside from the science, no absolute- 



