Neither does a mechanical analysis indicate the fertility of a par ticular soil; for while showing the amounts of material of various degrees of coarseness or fineness which bear upon the important questions of looseness and resistance to tillage, and of the power to hold moisture and withstand drouth, it does not show whether or not there may be an abundance or deficiency in all or of any one of the chemical elements necessary and in condition available for the full vigor of the plant.

To insure a maximum fertility under favorable climatic conditions a soil requires

1. A full supply of those chemical elements that constitute the food of the plant, and in a condition available for assimilation.

2. Looseness and porosity to a depth of several feet to insure easy tillage and aeration, as well as freedom to the movements of the plant roots. This is accomplished either by a sufficient percentage of sand, or if the soil be highly clayey, by the presence of a sufficient amount of humus or lime to cause flocculation or molecular aggregation of the clay particles. This implies, of course, a position or elevation favorable to drainage.

3. A degree of compactness to insure resistance to drouth by the presence of sufficient fine material (silt and clay) to favor capillary movement in the ground water and the absorption and retention of hygroscopic moisture.

The fulfilment of the first of these requirements is ascertained alone by chemical, and the second and third by mechanical analysis; therefore, to obtain a knowledge of the properties of a soil, and the conditions under which plants grow on such a soil, it is necessary that both forms of analyses should be resorted to.

MECHANICAL ANALYSIS.

The divisions into which the materials composing the soil are soparated comprise grits with diameters of 1 millimeter and upward:

The extreme minuteness of the finer particles will be appreciated when it is remembered that a millimeter is about 1-25 (.0394) of an inch, the diameter of the grains of silt thus measuring from 1-500 to

1-2500 of an inch, and that of clay still less. The effect of the extremes of these divisions upon the tilling quality of soils is well known, the sand tending to make a soil loose and porous, and the clay, as low as 25 per cent., giving compactness and a high resistance to the passage of a plow. The effect of the intermediate grades of sands and silt are not so clearly marked, and it is hard to designate the exact size of particles that would affect a soil neither one way or the other, standing as a mean between looseness and compactness, and making neither a clay soil lighter nor a sandy one more compact in tillage, though, of course, the plasticity of a clay is affected by any admixture of sand.

The silts, on drying from a wet state, tend to cake together, though being in no wise plastic, and they should, hence, be classed with the clays in rendering a soil compact. The particles of finest sand or dust, on the contrary, while slightly adhering when quickly dried after being wet, easily separate on being jarred, and would, therefore, properly be classed with the sands proper.

In the groups presented above the line between porosity and compactness is clearly drawn between the very fine sand or dust, on the one hand, and the silt on the other.

Prof. Hilgard,in his study of the Mississippi soils,* finds that the coarser silt (having diameters of .05 to .03 mm.) acts as a mean between the divisions, neither affecting one or the other. The proportion of this coarse silt in the soils of the Stations was not determined.

The power of a soil to absorb moisture from the atmosphere and hold it in times of heat and drouth is, of course, largely dependent upon the proportion of fine particles present in its composition. Clay is known to be extremely tenacious of water or moisture, and lies at one extreme, while the coarsest material, such as grits, holds only a very small fraction of one per cent. of moisture. The presence of humus, or of ferric oxide, increases the retentive power of clay, but to what extent has not been determined.

In order to ascertain how much hygroscopic moisture was absorbed by each sediment from an atmosphere saturated with moisture, tests were made on a soil from the Spartanburg Farm, which contained 11.2 per cent. of ferric oxide, all of which was held by the silt and clay.

The following results were obtained, after exposure at 73° F. and drying at 200° C.:

*Silt Analyses of Mississippi Soils; Portland Meeting Am. Association for the Advancement of Science.

A glance at the above results will show that the percentages increase with the lessening diameters, and that the proportion contained in the sands on the one hand, and the silt and clay on the other, bear the relation respectively of 1.2 to 16.6.

It would be interesting to know the real effect that humus and ferric oxide would respectively have upon the absorptive power of pure clay, but no extensive tests have been made. Prof. Hilgard, in his analysis of Mississippi soils, found that a pure white pipe clay, with its 75 per cent. of clay, held but 9 per cent of moisture; that a soil with 40 per cent of clay and containing 10.5 per cent. of ferric oxide held 18.6 per cent. of moisture. The buckshot soil of the Mississippi bottom, probably the richest soil in the South, has a still larger per cent. of clay, 44.3, but only 5.8 per cent. of ferric oxide, and there is a less absorption percentage (14.5). Ferric oxide, therefore, has clearly a large influence in giving soils a large absorption coefficient.

Relation of Sediment to Plant Food.-A chemical examination of each of the sediments made several years ago* showed that the clay and silt contained nearly all of the soluble elements of plant food, and that the clay especially was very rich in such elements. The large quantity of water used in the mechanical analysis usually dissolves out much of the lime, potash, and magnesia. The sediments coarser than silt are little else than silica.

Methods of Analysis.-There are a number of methods by which the mechanical ingredients of a soil may be separated from each other, but all have proven to be in some way defective, except two, viz. Hilgard's Elutriator and Osburn's Beaker method. The elutriator by Prof. E. W. Hilgard, of University of California, is an ingenious apparatus (described in Am. Journal of Science and Arts, Oct. and Nov., 1873), well adapted to its purpose. It is based on the buoyancy of the particles in currents of water of certain regulated *See Pro. of Am. Association for Advancement of Science.

velocities, which are caused to flow through the apparatus, while at the same time flocculation of the fine particles is prevented by a rapid churning given to the lower part of the column of water. By it the number of divisions of the ingredients can be made almost indefinite and with great accuracy, requiring, however, care and a large quantity of pure water.

The Beaker method, by Prof. Osborn of the Connecticut Station, is also based on the buoyancy of the particles; but instead of allowing currents of water to bear off the various sizes, sedimentation is resorted to. This method was used in the analysis of the Station soils chiefly because of the lack of a sufficient supply of good clear water necessary for the elutriator.

The method, in brief, is as follows: A quantity, say twenty grains, of the sample dried at 212° F. is placed in a porcelain dish with a little water, and severely rubbed with a soft rubber in order to separate the clay from the surface of the grains of sand. This water, holding the clay and silt, is poured into a beaker, and the attrition of the sand in more water is renewed. This is continued until after thorough rubbing no cloudiness appears in the water.

The sand is dried in a steam bath and by means of sieves having meshes of 1.0, 0.5, 0.25, and 0.1 millimeters in diameter, is separated into those divisions, and each weighed. The water holding the clay, silt, and some very fine sand, is stirred up and allowed to stand for a length of time sufficient for the fine sand to settle. This time is determined by experiment, the microscope being used to examine the sediment. The clayey water is poured carefully into a narrow and tall beaker, water added to the sediment of sand and again poured off, the operation being continued until all of the clay and silt have been removed. This sand, added to that that had passed through the finest sieve, comprises the group of particles whose diameters are from 0.1 to 0.05 millimeters.

The clayey water, diluted in a tall beaker to a height of about 200 mm., is allowed to stand for twenty-four hours, which allows practically all of the silt to sink to the bottom, and the two are separated by decantation, the silt dried and weighed.

The clay is precipitated with about 50 cc.m. of a saturated solution of common salt (e. g., 1.5 per cent. of salt) or brine, and is thrown on a weighed filter, "washed with weak brine, dried at 100° and weighed. It is again placed in a funnel and washed with a weak solution of sal-ammoniac until all of the chloride of sodium is removed. The filtrate is evaporated, the residue ignited and weighed; its weight plus that of the filter deducted from the total weight gives that of the clay itself." (Hilgard.)

The silt is not divided by the Beaker method into the various grades given in that of Hilgard's elutriator, but embraces all diameters from .05 mm. down to .01 mm.

CHEMICAL ANALYSIS.

There is in all soils a very large percentage of inert matter, comprising sand, clay, and minerals, not soluble in acids, varying from 60 to as much as 98 per cent. This "insoluble matter" not only has its influence upon the mechanical condition of the soil, but, in some cases where minerals are present, it serves as a storehouse from which new supplies of plant-food are slowly drawn, through atmospheric and other agencies.

The material soluble in acids, according to the method given below, may be considered as immediately available to plants, and comprises the following:

Potash, probably as silicate.

Phosphoric acid, as a phosphate.
Soda, probably as silicate.
Lime, as sulphate or silicate.
Magnesia, as sulphate or silicate.
Ferric oxide.

Brown oxide of manganese.

Alumina, as silicate.

Humus, in organic matter.

Plant-food.

In part as direct, but chiefly as indirect plant-food.

Indirect plant food.

Sulphuric acid, as sulphate of lime or magnesia.

Soluble silica, combined with potash, alumina, or some other one of the above.

Water and organic matter from decayed vegetation.

Carbonic acid, with lime. Very seldom present in appreciable

Nitric acid, as a nitrate.

Chlorine, with soda.

quantities.

With the determination of the amounts of available plant food in a soil there naturally arises the question of how we are to interpret the results; or, in other words, how much of each element should there be present to give to a soil a maximum fertility. The answer can only be had by a comparison of soils of known composition, physical conditions, and average yields; and as Prof. Hilgard has made it his study to a greater extent than perhaps any one else, we cannot do better than present here, mostly in his own words, some of his conclusions as given in the Report on Cotton Culture, Vol. V., Tenth U. S. Census.

The lime percentage should not fall much below 0.1 per cent. in the lightest sandy soils; in clay loams not below a fourth of 1 per cent., 0.25, and in heavy clay soils not below 0.5, and may advantageously