There are almost innumerable gradations of soil types as regards chemical and physical characteristics, so that a satisfactory classifica tion on this basis is very difficult. The Bureau of Soils of the United States Department of Agriculture, therefore, maps and classifies the typical soils of the areas it surveys on the basis of the differences in agricultural value as de termined by field observations on "the character of the soil and its relation to crops and vegeta tion," supplemented, however, by physical and chemical examinations. Considering the soil mainly as a medium into which plants send their roots and from which they draw a part of their food, it is evident that its agricultural value will depend largely upon the character of the original rocks from which the soil was derived, the degree of fineness to which the ma terial has been reduced, the amount and char acter of the organic admixture, as well as upon the treatment to which the soil has been sub jected and the changes it undergoes under culti vation. The fertility of a soil is, therefore, de termined not only by the store of plant food which it contains, but depends to a large extent upon the chemical, physical and biological proc esses by which this plant food is rendered available to plants. The soil is not only a store house of plant food, but a workshop or labora tory in which this food is being constantly pre pared for the use of plants, and it is the object of cultivation to promote the processes by which this is brought about. Chemical, physical and biological properties and processes are so intimately associated in soil that it is not always possible clearly to distinguish between them. For convenience of discussion, however, it is desirable to consider them separately.
Chemical Composition and Properties of Soils.— Plants derive all of their ash or mineral constituents and a large part of their nitrogen from the soil, and in order that a soil may pro duce plants it must contain these constituents in proper proportion and in assimilable con dition. Nitrogen is one of the largest and most important constituents of plants. The ash con stituents are taken up in comparatively small amounts by plants, but they are none• the less essential to their growth. The more import ant mineral elements of plant food include sulphur, phosphorus, potassium, calcium, mag nesium and iron. Apparently sodium, silicon and manganese are also needed by plants. Since the rocks from which soils are derived contain more or less of all the mineral elements it is not likely that any soil will be entirely lacking in any of those required. Their proportion and availability for assimiliation by the plant may, however, vary so widely as to cause wide differ ences in productiveness. The constituents of soils may be divided into three classes: (1) Active, that is, soluble in water or the root secretions of plants and hence readily available; (2) latent, that is, not soluble in water or root secretions, and hence not readily available, but becoming so in time through natural agencies, by the application of fertilizers or soil amend ments, such as lime, marl, etc., or as a result of tillage; and (3) mechanical. The last is by far the largest class, constituting usually from 90 to 95 per cent of the entire mass of the soil, and chemical analysis shows it to be as a rule mainly silica or sand. Hilgard gives the aver
age composition of soils of arid, semiarid and humid regions as determined by ordinary methods of chemical analysis (treatment with strong acids) as follows: It will be observed that the proportions of actual fertilizing constituents — nitrogen, phos phoric acid, potash, lime, etc.— are relatively small. When, however, the amounts per acre are taken into consideration they assume greater importance. An acre of sandy soil to a depth of one foot may be assumed to weigh about 4,000,000 pounds, of loam soil 3,500,000 pounds, of clay soil 3,250,000 pounds. An acre-foot of loam soil of the humid region, having the aver age composition given above, would, therefore, contain 7,350 pounds of potash, 4,550 pounds of lime, 4,200 pounds of phosphoric acid and 4,200 pounds of nitrogen, amounts sufficient to sup ply the needs of many crops. It should be borne in mind, however, that these are average figures from which there may be wide depart ures even in soils of the same type, and that productiveness of a soil as determined by chem ical composition depends, as already stated, not so much upon the total amounts of the con stituents present as upon their availability for the uses of plants. Ordinarily chemical analysis may show the presence of large amounts of all the necessary fertilizing constituents and still the soil may be unproductive if these con stituents are not in a form in which they may be readily appropriated by the plant. Chemical analysis is capable of showing the presence, absence or marked deficiency of any necessary constituent or the presence of harmful sub stances, but it has not yet reached such a stage of perfection that it can show with absolute certainty the amount of available plant food which a soil contains.
The methods most commonly used for de termining the availability of plant food in soils involve the treatment of the with a weak solvent, such as 1 per cent citric acid, fifth normal nitric acid, carbonated water or am monia chloride solution, which it is assumed approximates the power of plants to utilize the soil constituents. Schloesing, among others, has called attention to the importance of studying the availability of plant food in the soil by means of water extracts, hut the Bureau of Soils of the United States Department of Agri culture applying a method based upon the solubility of the soil constituents in water to a large number of productive and unproductive soils of the United States, reached the conclu sion that *practically all cultivable soils con tain naturally a nutrient solution which varies within comparatively narrow limits with regard either to composition or concentration and which is usually sufficient for plant growth* and that, therefore, there is no obvious relation between the chemical composition of the soil and solution and the yield of crops. The work of others, however, indicate that the growth of plants in the water extracts of soil closely par allel their growth in the soil from which the extracts are made. The recent development of centrifugal, displacement, or pressure methods of obtaining unaltered or but little changed soil solutions in quantity permits of important investigations on this point not heretofore possible.