generally used, the proportion rising to 76 per cent in Illinois. In the spring-wheat region there are several reasons for prominence of broadcasting. One comes from a prevalent practice of sowing wheat on the irregular surface of a cornfield without plowing; another is found in the use of the combined cultivator and broadcast seeder, which destroys many of the weeds that would otherwise be left between the drills. * * * The result of the investigation shows that 47 per cent of the winter wheat and 30 of the spring, or 37 of both, represent the proportion seeded by the drill. The improvement by drilling is made to average 10 per cent. The average quantity of seed used for seeding winter wheat is 1.35 bushels per acre; 1.24 for drilled, 1.44 for the sown. The details are as follows:

The following table, from the Agricultural Report for 1882 (p. 636), gives the proportion of winter wheat that was drilled and broadcasted in the autumn and winter of 1881 and 1882 for each State:

As it has not been practicable to obtain data that will accurately present the effect on the crop of the diverse features of cultivation that are independent of climate, I give, in addition to the preceding, the following general statements bearing on the annual crop statistics kindly communicated by Mr. J. R. Dodge, Statistician to the Department of Agriculture. Relative to the seeding and the stand of the crop and other matters, he says:

The practice varies with the kinds of corn. The small northern corn is planted closer than the larger more southern varieties. In the South corn is given greater distances than in the West. It grows larger there and makes more stalk growth and fewer ears. Only one or two stalks are planted in the hill there, while two or three in the middle, and three and even four in the extreme northern latitudes, are sometimes left in the hill. We have allowed one-third of a bushel per acre.

The individual differences in yield per acre in the States of highest, as well as of the lowest yield, are far greater than the differences in these State averages, as produced by differences in soil, in the effects of the various vicissitudes on different soils, in fertility or lack of it, in thoroughness of cultivation.

In the extreme West, beyond the Mississippi, where land is plenty and labor scarce, the cultivation is reduced to the minimum. Satisfactory results are now produced in southern Iowa in winter-wheat growing by simply "cultivating" between corn rows and sowing wheat at a labor expense of 60 cents per acre. The rough surface is favorable for exemption from winter killing, and some records of experiment show an increase of 25 per cent in yield over planting. after clover on a smooth surface. This is so notwithstanding the clover soil might be expected to have something like as great an advantage in real fertility over the soil that had grown a crop of maize. The corn exhausts, the clover enriches, and still the yield is

the greater after the corn, because the plants are not much injured by frost.

EFFECT OF VARIATIONS IN DATES OF SEEDING AND
HARVESTING.

The injurious effects of late frosts on early vegetables and on grains sown in the spring is generally annulled in part by a second sowing, so that the crop reports for the year do not show the full extent of the injury done to the plant by the climate.

In a general comparison between the climate and the crops accuracy would require that we know the date of last planting, but in the absence of this fundamental datum we are obliged to use the average dates between which the planting is done in any given State, and such dates are given in the following table and are assumed to refer to the dates of planting the seed which actually brought forth the subsequent harvest, whose yield per acre is given in the tables published by the statistician of the Department of Agriculture.

These tables are also necessary in order to compute the thermal constants and to anticipate the dates of bloom and harvest. The following tables, for 1882 and 1889, as published in the Annual Reports of the Department of Agriculture (pp. 409 and 636, respectively), give the dates of seeding for wheat:

The following table gives dates of sowing and ripening, especially in America additional to those given by Lippincott (1863), and in many cases will give useful indications of the progressive change that has gone on since 1860 in methods of cultivation and in the habits of the wheat itself: