355 
vitality of the seed, due to unknown causes, and which we have no 
means of measuring except by just such experiments as these. The 
elaborate measurements made by Lawes and Gilbert at Rothamsted, 
England, since 1850, furnish innumerable illustrations of this same 
principle; so, also, do those of W. R. Lazenby, at Columbus, Ohio, 
and many others. 
We shall therefore hope to derive more reliable results from the 
study of farming operations on a large scale, taking the averages by 
counties and States where the crops have been carefully measured. 
We may possibly eliminate irregularities in many disturbing ele- 
ments, and be able to clearly set forth that small percentage by which 
the crops of the United States as a whole are influenced by purely 
climatic conditions. Such influences may in extreme cases be very 
large, but, on the average, they are not so large as those which depend 
upon seed, cultivation, rotation, and fertilizers. 
EFFECT OF VARIATIONS IN METHOD OF CULTIVATION AND IN 
QUALITY OF SEED FOR DIFFERENT REGIONS AND YEARS. 
Among the modes of cultivation that materially affect the devel- 
opment of the plant and the quantity of the harvest must be consid- 
ered the practice of sowing seed broadcast with the hand as con- 
trasted with that of putting it in with the drilling machine. The 
drilling requires less seed, the saving being about one-half bushel 
per acre; the grain is buried more evenly, starts more uniformly, and 
stands the droughts better. Moreover, the drilled wheat fields are 
considered to yield more per acre, although it is difficult to state how 
much is due to the drilling independent of the character of the soil, 
because in general the fields that are drilled are most apt to be those 
free from stumps, stones, and steep slopes, while the broadcast sow- 
ing is especially adapted to this latter character of field. The census 
of 1879 shows that the drilled fields of winter wheat in Ohio yielded 
50 per cent more than the broadcast fields of summer wheat in the 
Northwest; but it is not plain what proportion of this is respectively 
due to the drilling and to the soil. 
In the report for 1875 of the Department of Agriculture (p. 42) 
the following statistics are given as to the percentage of area drilled, 
the quantity of seed per acre, and the increase of harvest in drilled 
fields over that in broadcasted fields: 
The following table omits the New England States, which produce 
little wheat, nearly all of which is sown broadcast. The wheat area 
of New York is divided equally between the two methods. In New 
Jersey, Pennsylvania, Delaware, and Maryland the drill greatly 
predominates. In the Southern States the area is small, particu- 
larly in the cotton States, and the drill is comparatively unknown. 
North of the Ohio River, in the winter-wheat States, the drill is very 
