AGRICULTURAL EXPERIMENTS. 277 



art of agriculture. The rules of practice have been derived from 

 experience and observation. The world of science has stepped in 

 and explained these rules. The rule is of no more force in 

 practice than it was before. It simply serves to suggest new 

 lines of inquiry for future experiments. 



I will now speak briefly of some of the difficulties in the way of 

 successful experimentation ; and in this discussion I wish to be 

 understood as limiting my remarks entirely to those experiments 

 which we inaugurate for the purpose of improving- our practice — 

 experiments for the improvement of the art. 



In the first place, we have great variations in soils. Unfortu- 

 nately, they are variations we are not able to detect except by ex- 

 perimental trials ; soils apparently similar so far as their compo- 

 sitions are concerned, and so far as their physical characters are 

 concerned, give very different results. In the experiments at the 

 Michigan Agricultural College, this was one of the most striking 

 points brought out by our first experiment. Ordinarily, field 

 experiments have been conducted by taking a single manured 

 plat, and then comparing with it different plats to which has been 

 applied different varieties of manure. The experiment was sup- 

 posed to be complete. The comparison of the unmanured plat 

 with the manured plat would apparently give an indication of the 

 result. But such experiments misled us. We found on quite a 

 number of unmanured plats, on soil precisely alike, so far as we 

 could judge of the character and composition, a very great dif- 

 ference in yield. Peculiarities of climate and seasons will have 

 much to do with varying results, and this seems to be one 

 reason why it would be desirable to try experiments at quite a 

 number of different points, having all the conditions precisely 

 alike so far as we can, making the conditions that vary simply 

 those of climate, soil, etc. 



There is another difficulty in the way of conducting field ex- 

 periments which is exceedingly difficult to obviate — that is, the 

 difference arising from variations in the cultivation. The time of 

 the cultivation of each plat should be the same. Not only the 

 same amount of labor should be expended on each plat, but that 

 labor should be performed at the same time. From this you will 

 see the difficulty of experiments on very large plats. If your 

 plats are very extensive, and you have a large number of them, 

 you cannot harvest them all at the same time. You cannot put in 

 all the seed at the same time. Some two years ago, at the Agri- 



