are such that they may be safely applied in the development of the principles of dry-farming.

There is a general belief among farmers that the roots of all cultivated crops are very near the surface and that few reach a greater depth than one or two feet. The first striking result of the American investigations was that every crop, without exception, penetrates the soil deeper than was thought possible in earlier days. For example, it was found that corn roots penetrated fully four feet into the ground and that they fully occupied all of the soil to that depth.

On deeper and somewhat drier soils, corn roots went down as far as eight feet. The roots of the small grains,--wheat, oats, barley,--penetrated the soil from four to eight or ten feet. Various perennial grasses rooted to a depth of four feet the first year; the next year, five and one half feet; no determinations were made of the depth of the roots in later years, though it had undoubtedly increased. Alfalfa was the deepest rooted of all the crops studied by the American stations. Potato roots filled the soil fully to a depth of three feet; sugar beets to a depth of nearly four feet.

Sugar Beet Roots

In every case, under conditions prevailing in the experiments, and which did not have in mind the forcing of the roots down to extraordinary depths, it seemed that the normal depth of the roots of ordinary field crops was from three to eight feet. Sub-soiling and deep plowing enable the roots to go deeper into the soil. This work has been confirmed in ordinary experience until there can be little question about the accuracy of the results.

Almost all of these results were obtained in humid climates on humid soils, somewhat shallow, and underlain by a more or less infertile subsoil. In fact, they were obtained under conditions really unfavorable to plant growth. It has been explained in Chapter V that soils formed under arid or semiarid conditions are uniformly deep and porous and that the fertility of the subsoil is, in most cases, practically as great as of the topsoil. There is, therefore, in arid soils, an excellent opportunity for a comparatively easy penetration of the roots to great depths and, because of the available fertility, a chance throughout the whole of the subsoil for ample root development. Moreover, the porous condition of the soil permits the entrance of air, which helps to purify the soil atmosphere and thereby to make the conditions more favorable for root development. Consequently it is to be expected that, in arid regions, roots will ordinarily go to a much greater depth than in humid regions.

It is further to be remembered that roots are in constant search of food and water and are likely to develop in the directions where there is the greatest abundance of these materials. Under systems of dry-farming the soil water is stored more or less uniformly to considerable depths--ten feet or more--and in most cases the percentage of moisture in the spring and summer is as large or larger some feet below the surface than in the upper two feet. The tendency of the root is, then, to move downward to depths where there is a larger supply of water. Especially is this tendency increased by the available soil fertility found throughout the whole depth of the soil mass.

It has been argued that in many of the irrigated sections the roots do not penetrate the soil to great depths. This is true, because by the present wasteful methods of irrigation the plant receives so much water at such untimely seasons that the roots acquire the habit of feeding very near the surface where the water is so lavishly applied. This means not only that the plant suffers more greatly in times of drouth, but that, since the feeding ground of the roots is smaller, the crop is likely to be small.

These deductions as to the depth to which plant roots will penetrate the soil in arid regions are fully corroborated by experiments and general observation. The workers of the Utah Station have repeatedly observed plant roots on dry-farms to a depth of ten feet. Lucerne roots from thirty to fifty feet in length are frequently exposed in the gullies formed by the mountain torrents. Roots of trees, similarly, go down to great depths. Hilgard observes that he has found roots of grapevines at a depth of twenty-two feet below the surface, and quotes Aughey as having found roots of the native Shepherdia in Nebraska to a depth of fifty feet. Hilgard further declares that in California fibrous-rooted plants, such as wheat and barley, may descend in sandy soils from four to seven feet. Orchard trees in the arid West, grown properly, are similarly observed to send their roots down to great depths. In fact, it has become a custom in many arid regions where the soils are easily penetrable to say that the root system of a tree corresponds in extent and branching to the part of the tree above ground.

Now, it is to be observed that, generally, plants grown in dry climates send their roots straight down into the soil; whereas in humid climates, where the topsoil is quite moist and the subsoil is hard, roots branch out laterally and fill the upper foot or two of the soil. A great deal has been said and written about the danger of deep cultivation, because it tends to injure the roots that feed near the surface. However true this may be in humid countries, it is not vital in the districts primarily interested in dry-farming; and it is doubtful if the objection is as valid in humid countries as is often declared. True, deep cultivation, especially when performed near the plant or tree, destroys the surface-feeding roots, but this only tends to compel the deeper lying roots to make better use of the subsoil.

When, as in arid regions, the subsoil is fertile and furnishes a sufficient amount of water, destroying the surface roots is no handicap whatever. On the contrary, in times of drouth, the deep-lying roots feed and drink at their leisure far from the hot sun or withering winds, and the plants survive and arrive at rich maturity, while the plants with shallow roots wither and die or are so seriously injured as to produce an inferior crop. Therefore, in the system of dry-farming as developed in this volume, it must be understood that so far as the farmer has power, the roots must be driven downward into the soil, and that no injury needs to be apprehended from deep and vigorous cultivation.

One of the chief attempts of the dry-farmer must be to see to it that the plants root deeply. This can be done only by preparing the right kind of seed-bed and by having the soil in its lower depths well-stored with moisture, so that the plants may be invited to descend. For that reason, an excess of moisture in the upper soil when the young plants are rooting is really an injury to them.

CHAPTER VII

STORING WATER IN THE SOIL

The large amount of water required for the production of plant substance is taken from the soil by the roots. Leaves and stems do not absorb appreciable quantities of water. The scanty rainfall of dry-farm districts or the more abundant precipitation of humid regions must, therefore, be made to enter the soil in such a manner as to be readily available as soil-moisture to the roots at the right periods of plant growth.

In humid countries, the rain that falls during the growing season is looked upon, and very properly, as the really effective factor in the production of large crops. The root systems of plants grown under such humid conditions are near the surface, ready to absorb immediately the rains that fall, even if they do not soak deeply into the soil. As has been shown in Chapter IV, it is only over a small portion of the dry-farm territory that the bulk of the scanty precipitation occurs during the growing season. Over a large portion of the arid and semiarid region the summers are almost rainless and the bulk of the precipitation comes in the winter, late fall, or early spring when plants are not growing. If the rains that fall during the growing season are indispensable in crop production, the possible area to be reclaimed by dry-farming will be greatly limited. Even when much of the total precipitation comes in summer, the amount in dry-farm districts is seldom sufficient for the proper maturing of crops. In fact, successful dry-farming depends chiefly upon the success with which the rains that fall during any season of the year may be stored and kept in the soil until needed by plants in their growth. The fundamental operations of dry-farming include a soil treatment which enables the largest possible proportion of the annual precipitation to be stored in the soil. For this purpose, the deep, somewhat porous soils, characteristic of arid regions, are unusually well adapted.

Alway's demonstration

An important and unique demonstration of the possibility of bringing crops to maturity on the moisture stored in the soil at the time of planting has been made by Alway. Cylinders of galvanized iron, 6 feet long, were filled with soil as nearly as possible in its natural position and condition Water was added until seepage began, after which the excess was allowed to drain away. When the seepage had closed, the cylinders were entirely closed except at the surface. Sprouted grains of spring wheat were placed in the moist surface soil, and 1 inch of dry soil added to the surface to prevent evaporation. No more water was added; the air of the greenhouse was kept as dry as possible. The wheat developed normally. The first ear was ripe in 132 days after planting and the last in 143 days. The three cylinders of soil from semiarid western Nebraska produced 37.8 grams of straw and 29 ears, containing 415 kernels weighing 11.188 grams. The three cylinders of soil from humid eastern Nebraska produced only 11.2 grams of straw and 13 ears containing 114 kernels, weighing 3 grams. This experiment shows conclusively that rains are not needed during the growing season, if the soil is well filled with moisture at seedtime, to bring crops to maturity.

What becomes of the rainfall?

The water that falls on the land is disposed of in three ways: First, under ordinary conditions, a large portion runs off without entering the soil; secondly, a portion enters the soil, but remains near the surface, and is rapidly evaporated back into the air; and, thirdly, a portion enters the lower soil layers, from which it is removed at later periods by several distinct processes. The run-off is usually large and is a serious loss, especially in dry-farming regions, where the absence of luxuriant vegetation, the somewhat hard, sun-baked soils, and the numerous drainage channels, formed by successive torrents, combine to furnish the rains with an easy escape into the torrential rivers. Persons familiar with arid