tain in unequal quantities soluble and nourishing principles the water absorbed by the roots may be more or less charged with these elements. We can understand, then, that the quantity of water necessary to enable a plant to furnish a given result is not the same for all soils, and that the richest soils may produce a greater result with a proportionably smaller consumption of water. By increasing the richness of the soil in soluble substances that can be assimilated, we should succeed in economically reducing the quantity of water consumed by the crops. In any case we might at the same time ask ourselves if all the water absorbed by the roots and introduced into the plant is utilized by it and at what limit the richness of the water should be arrested so as to be really profitable to the plant. In this connection Marié-Davy cites the following fact, mentioned by Perret in the Journal of Practical Agriculture for 1873:

In Perret's experiments a meadow having been covered with a sufficient quantity of nitrate of soda for a nitrogenous manuring of four years, the grass was magnificent in the spring. This grass was given green to the horses, who before, long began to show strong diuretic symptoms accompanied by raging thirst. These animals seemed to be completely under the influence of the administration of a strong dose of nitrate. The following year there was a complete cessation of the beneficial effects of the nitrate on the meadow, which showed conclusively that the plants of the first year contained nitrate in a natural state and not decomposed by the assimilation.

When nutritive substances are given to plants in abundance they can absorb a quantity of these elements besides what is necessary for their nourishment. This is particularly true when in the series of minerals which compose a normal nourishment, one of these substances is in excess of the others. Besides, if we compare the chemical composition of a crop cut green with that of a similar crop after arriving at maturity, we find that in the latter there is a diminution in weight of several of the substances present in the former. It would, therefore, have been interesting to know if the trouble mentioned by Perret was continued with the same intensity in the dry hay.

RELATION OF PLANTS TO MOISTURE OF SOIL.

E. Wollny (1887, Vol. X, p. 320) gives some results as to the influence of plants and shade on the moisture of the soil, being a modification of a memoir published by him in 1877. His conclusions are as follows:

(1) The water contained in the soil under a covering of living plants is, during the growing season, always less than in a similar layer of fallow, naked soil.

(2) The cause of the drying up of the soil by the plants is to be found in the very considerable transpiration of aqueous vapor by their leaves.

(3) The plants deprive the soil of water in proportion as they stand closer together and have developed their tops more luxuriantly. (4) The influence of the vegetation on the moisture of the soil extends to the deeper layers of soil.

(5) The moisture of the soil under a layer of inert objects, such as dead plants, manure, straw, pieces of wood, windfalls, etc., is always greater than that of the uncovered soil.

(6) The retention of the moisture in the soil under a cover of dead matter is a consequence of the protection afforded by the latter against the influences that favor evaporation.

(7) The quantity of moisture in the soil is, within certain limits and to a depth of about 5 centimeters, or 2 inches, greater in proportion as the covering of dead matter is thicker.

(8) The soil shaded by living plants is, under otherwise similar conditions, driest during the growing period, but that covered by dead objects is the moistest, while that which is not cultivated, not covered with plants and naked, is midway between the two previous in reference to its relations to moisture.

Wollny has also studied the influence of plants and shade upon the drainage of water from the soil. His conclusions are:

(1) A notably smaller quantity of water drains through the soil supporting living plants from the same quantity of rainfall than through a naked soil during the growing season.

(2) The quantity of drainage in cultivated fields is less in proportion as the plants stand more closely together and in proportion as they have developed themselves more luxuriantly.

(3) The quantity of drain water from a soil covered by inert objects is increased in comparison with that from fallow land in proportion as the covering layer is thicker, up to a certain limit, up to about 5 centimeters, beyond which a further increase in the thickness of the covering steadily diminishes the quantity of drainage

water.

(4) For the same quantity of rain and under otherwise similar circumstances, the soil covered with dead leaves and similar objects furnishes the greatest quantity of drainage water up to a covering of about 5 centimeters thickness; the naked, fallow land furnishes the next smaller quantity of water; the soil covered with living plants furnishes the least quantity of drain water,

RELATION OF WATER TO CROPS.

E. Wollny has studied the relation of the irrigation and rainfall to the development and productive power of plants in cultivated fields, and the following summary is essentially as given by him in Volume X of his Forschungen for 1888, page 153.

An early investigation of this subject was made by Ilionkoff, who filled five large tubs with soil and sowed buckwheat in each on the 15th of May; each tub was then watered regularly with a definite quantity of water, the total quantity used being given in the second column of the table following. The relative quantities of buckwheat harvested at the end of the season are given in the third column and the straw is given in the fourth column. The weight of the buckwheat originally sown in each tub was the same, viz, 0.154 gram.

These figures show plainly that the plants in tub No. 2 were most favorably situated. Probably No. 1 had too much water and Nos. 3, 4, and 5 too little.

Haberlandt, in 1866, experimented on the quantity of water needed in the growth of plants in three plats of 14.41 square meters each; of these No. 1 received no artificial watering; No. 2 was watered once a week, except in great droughts twice a week; No. 3 received a double quantity once a week. These quantities corresponded to a rainfall of 6.46 millimeters for No. 2 and 13.92 millimeters for No. 3. The total quantities for the season were 96.96 and 193.92 millimeters. The natural rainfall was as follows:

The number of rainy days was large, but the rainfall was small, and the plants in bed No. 1 suffered for want of water. The relative harvests for the different beds and crops were as follows:

Beds Nos. 1 and 2 showed about the same rate of growth. No. 3 showed a retardation. The barley and the rye were harvested from this bed four days later than from the other two. The quantity of harvest increased with the quantity of water, and the harvest of grain, except in the case of the wheat, was more increased by watering than was the harvest of straw; the quality of the grain showed only slight differences.

Hellriegel experimented (1867-1883) on the influence of water upon the crops. He filled a number of vessels with quartz sand and maintained the earth at a different state of dryness. The experiments were repeated for several years on wheat, rye, and oats, the general results being that when the ground contained from 60 to 80 per cent of its full capacity of water the harvest was larger than when the ground was drier and about in the following proportions:

Hellriegel also varied the experiment by giving the tubs daily, each evening, as much water as they had lost during the day, thus

maintaining a very constant state of moisture in each. with the

The general result, therefore, was that the largest harvest is given by soil containing 40 per cent of its maximum capacity for water. The general appearance of the plants showed that those having too little water had a less intensive life and were suffering from lack of nourishment rather than from the want of pure water itself.

Fittbogen (1873) conducted a series of experiments on twenty tubs in groups of four. The relative weights of his harvests of oats were as follows:

These figures show that for moistures varying between 30 and 80 per cent there was very little difference in the harvest, while for drier soils the harvest was decidedly diminished; but it is notable that for the driest soil (No. 5) the grain ripened earliest of all.

Haberlandt, in 1875, reports the results of experiments on three tubs sown with summer wheat. The quantity of water allowed to tub No. 1 was just sufficient to keep the wheat alive; the other quantities, with the harvest, are given in the following table:

Whence it would seem that the limit.of useful water had not yet

been reached.