208 STATE BOAED OF AGEICULTUEE. 



'Some experiments now in progress were begun ut your College five years ago 

 next spring. They were intended to decide whether it was better to cultivate 

 large trees or leave them in grass Avhere no fertilizer was used or no grass taken 

 off; also to discover Avhether it was of much account to cultivate small circles 

 about trees ; also whether a small circle of grass about a tree affected it for 

 good or bad, or whether it was of little importance ; also whether we could dis- 

 cover any difference in the effect on the trees or fruit if we applied manure close 



. about the tree or in a large circle, or spread it broad-cast over the ground. 



.Experiments are also going on in which a heavy mulch is kept on the ground ; 



-also experiments in the use of unleached ashes. Time will not permit a full 



;account of these experiments at this time. Although not yet completed and 

 but few results published, yet they have attracted a good deal of attention from 

 J. J. Thomas and others. For a full account see page 119 of my report in this 

 volume. 



Improving Wheat and Other Grains. 



This seems well worthy of trial everywhere by every man Avho raises wheat. 

 Aithough this is not a horticultural experiment, I will venture a few remarks 

 upon this interesting and important subject. "We should not only give good 

 <3ulture on ground well adapted to wheat or well prepared for wheat, but we 

 :should take more care in the selection of good seed. 



Touching on this subject, I quote from an article by George Geddes in the 

 Michigan Keport of the Board of Agriculture, p. 182, 1869: ''At or before the 

 time of threshing, set some sheaves on a floor, heads upward, and then draw 

 out the most perfect heads, — those of the greatest length and the best filled. 

 Put this selected seed on land in the best condition in every respect, weed the 

 wheat the next spring. Out of this again the next year draw the best heads, 

 and so on for several years, — the longer the better, — by and bye he will have 

 eonae fieed wheat to sell that he may be Avilling to have bear his name, and he 

 will be a public benefactor." By this means we will get seed wheat which has 

 had geod ancestors for a series of years. Every breeder of stock knows the 

 importance of a long line of good ancestors. A native cow may be a good 

 milker, but her calves may make good milkers or poor. The calves may take 

 after some of their poor relations not very far back. If well-bred for milk, the 

 calves are much surer to be good milkers. The same principles apply with 

 equal force in regard to plants and seeds. I will enforce this idea by referring 

 to some experiments made by Frederic F. Hallett, in England, as reported in 

 the Journal of the Royal Agricultural Society, 1861, page 371. He began by 

 selecting the best head of wheat he could find, paying especial attention to the 

 quality. 



" TABLE allowing the importance of each additional generation of selection. 



LENGTH. 



1857, original -ear 43.8 iuches 



1858, finest ear GJ^ inches 



1859, finest ear 1% inches , 



18G0, ears imperfect from wet season. 

 1861, finest ear S% inches 



Grains. 



123 



No. of ears 

 on finest 



stool. 



10 

 22 



39 



52 



