AGRICULTURAL EXPERIMENTS. 283 



I will refer to it again. It is the difficulty of getting over a large 

 area in a given time. If your plats are so large that it takes a 

 whole day to go over them and perform the labor, you will be 

 very liable to error from the variations mentioned. 



Wires may be used for separating the plats, and that is a very 

 convenient plan, and one we have practiced at the agricultural 

 college, but it does not obviate the objection that the roots pass 

 from one to another. It enables you to make a fair division 

 between two adjoining plats, which is exceedingly difficult — more 

 so than a person would think who has had no experience in the 

 matter of making exact divisions through a crop of growing grain. 



In the feeding of animals, I would be particular, as I mentioned, 

 to have but a single animal in a pen. If I was going to feed 

 twenty animals on a given feed, I would place each one in a pen 

 by itself, and then would confine them to a single article of food. 

 It would be an important matter to test the nature of corn meal, 

 and corn prepared in different ways, as food for swine. The 

 natural mode that would be suggested, for conducting the experi- 

 ment, would be to put a number of swine in a pen and give four 

 or five of them corn unground, four or five more of them cooked 

 corn meal, and four or five more uncooked corn meal. But I ap- 

 prehend the results of the experiment would be very unsatis- 

 factory, to say the least. I would not like to place any very 

 great amount of reliance on it. I would prefer to take the same 

 number of hogs and put them in pens separately and feed them 

 with unground corn, and follow that up until I got the range of 

 variation between the animals. Then I would take the same num- 

 ber and feed them corn meal ; and if your arrangements are of 

 sufficient extent, you may have these experiments going on at the 

 same time. We have been feeding swine for a number of years, 

 and have from ten to fifteen pens. We have fed nothing but raw 

 meal thus far. The question is often asked, " Why don't you feed 

 cooked meal ?" I have not yet got the standard of comparison 

 with raw meal, by which I can compare results with the cooked 

 meal. There is a great range of variation in the animals, and 

 there would be made a serious error in the experiment. This 

 error has arisen very much from the force of circumstances. We 

 could not get animals of uniform size and uniform ages to fill our 

 pens, and for that reason, the attempt to do the two things which 

 I have mentioned is objectionable. There has been so much 

 variation in this matter that we need to have more experiments 



