Chapter VII.. The Mineral Constituents of the Soil Solution.[40]

THE MINERAL CONSTITUENTS OF THE SOIL SOLUTION.

The mineral constituents of the soil are products of the disintegration, degradation and decomposition of rocks. The decomposition products are mainly silica in the form of quartz, ferruginous material consisting of more or less hydrated ferric oxide and alumina, and hydrated aluminum silicate. The ferruginous material, being deposited or formed in the soil in a very finely divided condition, frequently coats the soil fragments to such an extent as completely to mask their true character. But if a soil be thoroughly shaken with water, and especially in the presence of some deflocculating agent such as a slight excess of ammonia, as in the ordinary preparation of a soil sample for mechanical analysis the coating material is generally removed quite readily, and the mineral particles appear as fragments and splinters of the ordinary rock-forming minerals. Sometimes these fragments are more or less worn and rounded at the edges, showing mechanical abrasion or solvent action; sometimes they show evidences of partial alteration and decomposition; but surfaces of the unaltered mineral individuals always are found. These unaltered minerals occur as fragments of all sizes, and are to be found in the sands, silts, and presumably in the clays. As might be anticipated, the minerals other than quartz generally show a tendency to segregate in the finer mechanical separations of the soil. The presence of these unaltered mineral fragments in the clays has so far defied direct experimental proof because of the limitations of the microscope, but from chemical reasoning and a priori considerations there can be but little doubt that they exist in the clays as in the coarser separations.

For a more detailed discussion and citations of the literature, see The mineral constituents of the soil solution, by Frank K. Cameron and James M. Bell, Bull. No. =30=, Bureau of Soils, U. S. Dept. Agriculture, 1905.

Centrifugal methods of mechanical soil analysis, by L. J. Briggs, F. O. Martin and J. R. Pearce, Bull. No. =24=, Bureau of Soils, U. S. Dept. Agriculture, 1904.

See, The mineral composition of soil particles, by G. H. Failyer, J. G. Smith and H. R. Wade, Bull. No. =54=, Bureau of Soils, U. S. Dept. Agriculture, 1909. Recent improvements in microscope methods make it possible to identify without serious trouble the mineral content of silts with a diameter as low as 0.005 mm., and many even of the clay particles have recently been determined with satisfactory accuracy.

The minerals to be anticipated in the soil are those commonly occurring in the rocks; but as a result of the action of mixing and transporting agencies, a soil normally contains minerals from rocks other than those from which it is primarily derived.

It would hardly be fair to regard a beach sand, for instance, as a normal soil. Yet it is surprising how many minerals other than quartz can usually be found even in a beach sand. Opinions may differ as to just what are the common rock-forming minerals, and perhaps no two mineralogists or petrographers would give identical lists, but there are a number of minerals which would appear undoubtedly in every list, and these would be found generally in any soil. Again, it might happen that in any given sample of soil, no pyroxene, for instance, could be found; but experience shows that it would never happen in such a case that no amphibole, chlorite, serpentine, or other ferro-magnesian silicates would be present. However distinct these minerals cited may be from each other morphologically or optically, they are much the same in their chemical characteristics, their solubilities and their reactions with water and such dilute solutions as exist in the soil. Hence from the point of view of the soil chemist they may be considered for all practical purposes varieties of one and the same mineral species. Consequently an important result of researches on the minerals of the soil is the generalization that soils are far more heterogeneous than are rocks, and that practically every soil contains all the common rock-forming minerals.

See Bull. No. =30=, Bureau of Soils, U. S. Dept. Agriculture, 1905, p. 9.

It is not difficult to account for the heterogeneity of the mineral content of the soil. Many of our rocks are reconsolidated soils, and the alternating formation of rock and soil from the same materials is probably an agency, in some part at least, in the mixing of soil material. The action of water in carrying off and transporting surface material and in gullying and eroding sloping surfaces is probably a large factor. But this agency, like the first, is rather restricted and localized. Just as important as a mixing agency is the wind. This, unlike water, works uphill as well as down, and is more or less in action at all times, continually transporting soil material from place to place. Wind-borne dust on roofs of dwellings, on rocky mountain tops and similar places, where it could have been brought by no other agency than the wind, is sometimes found supporting vegetation. Many chemical and mineralogical analyses of wind-borne dust obtained from various locations show it to have generally the same essential characteristics as ordinary soils.

Aside from the quartz and ferruginous materials mentioned above, the major part of the soil minerals are silicates, ferro-silicates, alumino-silicates, or ferro-alumino-silicates, of the common bases, sodium, potassium, calcium, magnesium, and ferrous iron. Other bases, such as lithium, barium, or the heavy metals may occasionally be present in appreciable amounts as may other types of silicates, or other mineral salts, but these may be regarded as more or less incidental and rarely affecting in any essential way the general character of the soil mass. These silicates or silico minerals are all somewhat soluble in water, and being salts of weak acids with strong bases, are greatly hydrolyzed. A convenient illustration is afforded by the well-known rock and soil mineral, orthoclase. Assuming its type formula, the reaction with water may be represented,

K.AlSi₃O₈ + HOH ⇆ H.AlSi₃O₈ + KOH.

Under ordinary soil conditions, with a relatively large proportion of carbon dioxide in the soil atmosphere, the potash formed would be more or less completely transformed to the bicarbonate,

KOH + CO₂ + H₂O ⇆ KHCO₃ + H₂O.

Confirmation of this view is afforded by the natural associations and known alteration products of orthoclase.

The acid of the formula H.AlSi₃O₈ is not known and is probably entirely instable under ordinary conditions, and breaks down with the separation of silica, to form the minerals pyrophyllite, kaolinite or kaolin, and diaspore according to the following equations:

H.AlSi₃O₈ - SiO₂ = H.AlSi₂O₆ (Pyrophyllite) H.AlSi₃O₈ - 2SiO₂ = H.AlSiO₄ (Kaolinite) H.AlSi₃O₈ - 3SiO₂ = H.AlO₂ (Diaspore).

All three of these minerals and their corresponding salts have been found in nature as alteration products of orthoclase. It is probable that, under soil conditions, the principal metamorphic product of feldspar is kaolin (or kaolinite when it is crystalline), hydrated aluminum oxide being of much less importance and pyrophyllite of doubtful occurrence. A still more interesting case, perhaps, because of the well recognized tendency of magnesium salts to form basic compounds, is the alteration of pyroxene, amphibole and olivine with the formation of a chlorite or serpentine, common associations in nature, which may be represented

See Ueber die Bildung von Bauxit und verwandte Mineralien, von A. Liebrich, Zeit. prakt. Geol., =1897=, 212-214.

MgSiO₃ + HOH ⇆ MgSiO₃.nMg(OH)₂ + SiO₂.

It is tacitly assumed in the foregoing statements that the reaction between a silicate mineral and water is a reversible reaction. This is not definitely known to be the case, for the formation of the ordinary silicate rock-forming minerals in the wet way at ordinary temperatures has as yet been realized in only a few cases. The assumption has, however, some experimental support. Minerals have been often made in the wet way at somewhat elevated temperatures, especially interesting cases in this connection being the formation of orthoclase by Friedel and Sarasin at slightly elevated temperatures, and the formation of zeolites by Gonnard and by Doroshevskii and Bardt, and the formation of apatite by Weinschenk. Feldspars and zeolites are common natural associations, it being generally conceded that zeolites are alteration products of the feldspars through the action of water; but Van Hise has pointed out that under conditions of weathering such as would obtain in the soil, the tendency is for the zeolites to alter to feldspars. Wöhler’s classical experiment of recrystallizing apophyllite from hot water is significant, for only the products of hydrolysis should be obtained if there is an irreversible reaction between the mineral and water. Lemberg found that leucite (KAlSi₂O₆) when treated with an aqueous solution containing 10 per cent. or more of sodium chloride, was partially transformed to analcite (NaAlSi₂O₆.nH₂O), potassium chloride being formed at the same time. The reverse reaction was also realized, that is, the partial conversion of analcite to leucite by treatment with a solution of potassium chloride, and similar transformations were carried out with the feldspars. Lemberg’s experiments are of especial value as they were carried out at ordinary as well as at high temperatures. It appears probable, therefore, that the hydrolysis of a silicate of the alkalis or alkaline earths is a reversible reaction. It should be noted, however, that Kahlenberg and Lincoln have shown that probably, in very dilute solutions of alkali silicates, the hydrolysis is practically complete and the silica is nearly all present as colloidal silica and not as silicic acid. Nevertheless at higher concentrations silicates are formed, and there is abundant evidence in nature that the alumino- or ferro-silicates are reacting with bases to form salts, for example such as the micas. If the hydrolysis were quite complete, it would appear to follow that the reaction between water and the silicate is irreversible. In that case it is difficult to see how any silicate mineral could persist in the soil for any length of time, and all soils should soon become sterile wastes composed essentially of quartz, kaolin and ferruginous oxides. It has been suggested that the original mineral particles are protected from decomposition by the formation of a coating “gel.” That is, that silica, alumina, ferruginous or other materials result from the decomposition of the minerals in a jelly-like form on the surface of the soil grains, protecting them from further action of the soil solution. If diffusion can take place through the gel, solution and hydrolysis of the mineral would proceed, although the presence of the gel would probably retard the rate of the reaction. If it be postulated, however, that diffusion through the gel does not take place, the minerals of the soil can have no influence on the composition of the soil solution, which is an unthinkable alternative. The presence of such gels in the soil has frequently been assumed, but satisfactory proof is generally wanting.

Sur la reproduction par voie aqueuse du feldspath orthose, par Friedel et Sarasin, Comptes rendus, =92=, 1374, (1881).

Note sur une observation de Fournet, concernant la production des zéolites a froid, par F. Gonnard, Bull. Soc. min. France, =5=, 267-269, (1882); Jahrb. Min., =1884=. I, Ref. 28.

Metathetical reactions with artificial zeolites, by A. Doroshevskii and A. Bardt, Jour. Russ. Phys. Chem. Soc., =42=, 435-42 (1910). Chem. Zentr., 1910, II, 68.

Beiträge zur Mineralsynthesis, von E. Weinschenk, Zeit. Kryst., =17=, 489-504, (1890).

U. S. Geol. Surv. Monograph, =47=, A treatise on metamorphism, by Charles R. Van Hise, 1904, p. 333.

Jahresb. Fortschr. Chemie Liebig and Kopp, =1847-48=, 1262; note.

Ueber Silicatumwandlungen, von J. Lemberg, Zeit. deutsch. geol. Ges., =28=, 519-621, (1876); Inaug. diss. Dorpat, =1877=; Bied. Centbl., =8=, 567-577, (1879).

Solutions of silicates of the alkalis, by L. Khlenberg and A. T. Lincoln, Jour. Phys. Chem., =2=, 77-90, (1898).

Van Hise, loc. cit., p. 693.

A gel is a jelly-like substance, apparently continuous, which forms either by the settling from suspension in a liquid of very fine particles which then become aggregated; or, is formed by the evaporation of a liquid containing fine particles in suspension until the quantity of liquid remaining is just sufficient to serve as a cementation medium holding the suspended particles together in a semi-rigid mass. For an experimental demonstration of the formation of such a gel, see, The effect of water on rock powders, by Allerton S. Cushman, Bull. No. =92=, Bureau of Chemistry, U. S. Dept. Agriculture, 1905.

In general, the same kind of considerations developed for orthoclase hold for the other soil minerals. If minerals of this character be pulverized or ground reasonably fine and then be shaken with distilled water which has been previously boiled to eliminate the dissolved carbon dioxide, the resulting solution will give an alkaline reaction with such indicators as phenolphthalein or litmus. If a soil be shaken up thoroughly with water, the resulting solution filtered free of suspended matter, as by passing through a Pasteur-Chamberland bougie, and then boiled to eliminate the carbon dioxide, in the vast majority of cases the solution will also give an alkaline reaction with phenolphthalein or litmus. The waters of most of our springs, ponds, creeks or rivers being natural soil solutions, give an alkaline reaction after boiling.

In making such experiments in the laboratory or in lecture demonstrations, it is well to have the mass of water large in comparison with the mass of powdered mineral or rock; otherwise secondary adsorption effects may occur and obscure the results of the hydrolysis.

But the mineral content of these natural waters varies greatly. These waters are composed in part of the “run-off,” in part of a portion of the “cut-off” waters, described above. This portion of the cut-off, normally, in passing through the soil goes mainly through the larger interstices. It is not long in contact with the individual soil particles and floccules, and because diffusion of dissolved mineral substances is quite slow, especially in dilute solutions, it takes up but little mineral matter from such aqueous films as it may intercept.

A different state of things exists with that portion of the cut-off water which returns towards the surface by reason of capillary forces, to form the great natural nutrient medium for plants. This water is moving over the soil particles in films, and with slowness. It is long in contact with successive fragments of any particular mineral and all the different minerals making up the soil. Consequently, it tends towards a saturated solution with respect to the mineral mass; and it follows that if every soil contains all the common rock-forming minerals, every soil should give the same saturated solution, barring the presence of disturbing factors. Disturbing factors, however, enter into all cases under field conditions, such for instance as the presence of some uncommon or unusual mineral in appreciable amounts, differences in temperature, surface effects, or extraneous substances. These will be considered later, but another disturbing factor requires immediate consideration.

In every soil, varying proportions of the soluble mineral constituents are present otherwise than as definite mineral species; that is, they are present as solid solutions, or absorbed on the soil grains or perhaps absorbed in some other manner. The concentration of the liquid solution in contact with a solid solution or complex of absorbent and absorbed material is dependent upon the relative masses of solution and solid. Thus, the concentration of a solution with respect to phosphoric acid, when brought into contact with so-called basic phosphates of lime or iron, is dependent in a marked way upon the proportion of solution to solid. Consequently it is to be expected that an aqueous extract of a soil will vary in concentration with the proportion of water used; and that with the same proportion of water, different soils or different samples of the same soil will yield different concentrations.

Feldspars certainly, and phosphorites possibly, are mineral components of the soil; and these substances when ground sufficiently fine have been added to soils with sometimes an increased production of crop. Other minerals, such as leucite, have given similar results. But also apparently pure quartz sand sometimes accomplishes the same results, as for example, in the experiments of Hilgard cited above. It has not been shown, however, that the addition of any of these substances produces an appreciable change in the concentration of the soil solution.

The action of water and aqueous solutions upon soil phosphates, by Frank K. Cameron and James M. Bell, Bull. No. =41=, Bureau of Soils, U. S. Dept. of Agriculture, 1907.

How far absorbed mineral constituents affect the solubility of the definite minerals in the soil or influence the concentration of the soil solution, it is not possible to predict with any approach to certainty. Those soils which hold the most moisture are generally the best absorbers. Moreover, the soluble mineral constituents of the soil, for instance potassium or phosphoric acid, are absorbed to a very high degree from dilute solutions. Consequently it is to be expected that variations in the concentration of the natural soil solution would be less than in aqueous extracts, when there is employed a constant and relatively large proportion of water to soil. These considerations are of great theoretical importance since they appear to negative the possibility of getting, with present experimental resources, any exact knowledge of the concentrations of the mineral constituents in the soil solution when the soil is in condition to grow the common crop plants. Moreover, they furnish a guide to the limitations which must be recognized in attempting to postulate what these concentrations may be on the basis of analytical data obtained from aqueous soil extracts.

Many attempts have been made to extract the solution naturally existing in the soil and to analyze it. The results obtained have not been very satisfactory, owing mainly to the mechanical difficulties involved. As pointed out above, the solution in a soil under suitable conditions for crop growth is held by a force of great magnitude. Nevertheless, by using powerful centrifuges, with saturated soil, it has been possible to throw out the excess of solution over the critical water content of the soil. In this way small quantities, generally a very few cubic centimeters at a time, have been obtained. The analysis of a few cubic centimeters of a very dilute solution is in itself difficult, involving necessarily more or less uncertainty as to the absolute value of the results. Nevertheless, the concentration of the soil solutions thus obtained, with respect to phosphoric acid and potash, varied but little for soils of various textures from sands to clays, and the variations observed could not be correlated with the known crop-producing power of the soils. The average concentrations of the soil solutions thus obtained lies in the neighborhood of 6-8 parts per million (p.p.m.) of solution for phosphoric acid (P₂O₅) and 25-30 parts per million for potash (K₂O). In the following table are given the results obtained by analyzing solutions extracted from different samples of loams and sands by means of a centrifuge. The crop growing on these soils and the crop condition at the time the samples were collected are given in the table, and the percentages of water in the samples when placed in the centrifuge are also given.

In this connection it is interesting to note that recent investigations on the proportions of phosphoric acid, potassium and nitrates in cultural solutions best adapted to the growth of wheat, give the same ratio of phosphoric acid to potassium as the figures just cited show to exist normally in the soil solution.

ANALYSIS OF SOIL SOLUTION REMOVED FROM FRESH SOILS BY THE CENTRIFUGE.

==================+=======+==========+=========+================== | | | |Parts per million | | | | of solution Soil | Crop | Condition|Per cent +--------+----+---- | | of crop |moisture.| PO₄ | Ca | K ------------------+-------+----------+---------+--------+----+---- Leonardtown loam | Wheat | Good | 22.0 | 6 | 17 | 22 Leonardtown loam | Wheat | Poor | 25.2 | 10 | 9 | 19 Leonardtown loam | Wheat | Good | 17.6 | 8 | 22 | 38 Sassafras loam | Clover| Good | 19.7 | 5 | 18 | 19 Sassafras loam | Corn | Medium | 17.5 | 8 | 13 | 36 Sassafras loam | Corn | Medium | 18.3 | 8 | 83 | 25 Sassafras loam | Wheat | Good | 18.8 | 7 | 44 | 34 Sassafras loam | Wheat | Poor | 20.0 | 7 | 27 | 24 Sassafras loam | Corn | Good | 17.3 | 8 | 24 | 25 Norfolk sand | Forest| Poor | 10.0 | 5 | 18 | 31 Norfolk sand | Corn | Good | 11.9 | 11 | 36 | 31 Norfolk sand | Wheat | Good | 10.7 | 18 | 45 | 31 Norfolk sand | Wheat | Poor | 11.2 | 8 | 38 | 24 Norfolk sand | Corn | Medium | 10.6 | 9 | 65 | 35 ------------------+-------+----------+---------+--------+----+----

The concentrations of the solutions obtained from the samples do not justify any correlation with the crop-producing power of the soils, nor with the texture of the soils. The wide variation in the concentrations with respect to calcium is probably due to the fact that all of the samples came from fields which had been limed, some quite recently, and that the content of carbon dioxide in the different samples varied. It is of special interest to note that the content of calcium in the solutions does not show any obvious relation to the content of phosphoric acid.

For the literature of the earlier work on the composition of aqueous extracts of soils, see: How crops feed, by Samuel W. Johnson, 1890, p. 309 et seq.; see also. On the analytical determination of probably available “mineral” plant-food in soils, by Bernard Dyer, Jour. Chem. Soc. =65=, 115-167, (1894); and Soils, by E. W. Hilgard, 1906, p. 327 et seq.

An effort has been made to ascertain the mineral concentration of soil solutions as they occur naturally in the field. Because of the practical impossibility of extracting the actual soil solution, an empirical method was employed. Areas were selected where good and poor crops were growing near each other on the same soil types, and preferably in the same field. Samples of soil from under these crops were taken at several intervals during the growing season, quickly removed to a nearby laboratory, shaken thoroughly with distilled water in the proportion of one part of soil to five parts of water, allowed to stand twenty minutes and the supernatant solution passed through a Pasteur-Chamberland filter.

Capillary studies and filtration of clays from soil solutions, by Lyman J. Briggs and Macy H. Lapham, Bull. No. =19=, Bureau of Soils. U. S. Dept. Agriculture, 1902; Colorimetric, turbidity and titration methods used in soil investigations, by Oswald Schreiner and George H. Failyer, Bull. No. =31=, Bureau of Soils, U. S. Dept. Agriculture, 1906.

As has been pointed out above, the aqueous extract of a soil thus arbitrarily prepared has no definite or causal relation to the soil solution in the field. It is certain that the solutions would not generally be the same. It should also be emphasized that such a procedure can not, as some investigators have assumed, afford a criterion between soluble and insoluble salts in the soil, else the proportion of water to soil used above some minimum would be immaterial as far as the amounts which go into solution are concerned. The proportion of water to soil is not immaterial, however, considering the chemical nature of the soil components and the results of experiment. Consequently, it is clear that the concentration of the soil solution is not simply the ratio of the amounts found in the aqueous extract, to the percentage of moisture in the soil, but something quite different.

Artificial solutions prepared in the manner described above should, however, furnish evidence as to whether or not there are recognizable differences in the soluble mineral constituents of good and poor soils respectively; and if such differences exist, whether they are consistent. That is to say, if the more productive soils also uniformly yield aqueous extracts of a higher concentration, then it would be a fair inference that their natural soil solutions are maintained at a higher concentration than in the less productive soils.

Results obtained for several localities and several crops, taken from the original records, are given in the following tables.

The chemistry of the soil as related to crop production, by Milton Whitney and F. K. Cameron, Bull. No. =22=, Bureau of Soils, U. S. Dept. Agriculture, 1903.

WATER SOLUBLE CONSTITUENTS OF SOIL.

Locality, Salem, N. J. Soil type, Norfolk sand. Crop, wheat. Yield, good. =========+=======+==========+======================================= | | | Parts per million of oven-dried soil | Depth | Moisture +-------------+---------+--------------- Date | inches| content | Phosphoric | Calcium | Potassium | | Per cent.| acid (PO₄) | (Ca) | (K) ---------+-------+----------+-------------+---------+--------------- March 10 | 0-12 | 13.2 | 12 | 5 | 12 | 12-24 | 11.5 | 7 | 5 | 16 June 8 | 1-24 | 4.3 | 4 | 14 | 13 June 13 | 1-24 | 4.6 | 5 | 13 | 17 June 19 | 1-24 | 9.6 | 2 | 14 | 24 ---------+-------+----------+-------------+---------+---------------

Locality, Salem, N. J. Soil type, Norfolk sand. Crop, wheat. Yield, poor. =========+=======+==========+======================================= | | | Parts per million of oven-dried soil | Depth | Moisture +-------------+---------+--------------- Date | inches| content | Phosphoric | Calcium | Potassium | | Per cent.| acid (PO₄) | (Ca) | (K) ---------+-------+----------+-------------+---------+--------------- April 3 | 0-12 | 12.0 | 11 | 5 | 32 | 12-24 | 12.0 | 10 | 3 | 22 June 16 | 1-24 | 9.3 | 4 | 29 | 20 ---------+-------+----------+-------------+---------+---------------

Locality, Salem, N. J. Soil type, Sassafras loam. Crop, wheat. Yield, medium. =========+=======+==========+======================================= | | | Parts per million of oven-dried soil | Depth | Moisture +-------------+---------+--------------- Date | inches| content | Phosphoric | Calcium | Potassium | | per cent.| acid (PO₄) | (Ca) | (K) ---------+-------+----------+-------------+---------+--------------- March 10 | 0-12 | 23.2 | 19 | 10 | 8 | 12-24 | 21.6 | 11 | 10 | 14 March 14 | 0-12 | 22.3 | 18 | 8 | 18 | 12-24 | 20.2 | 15 | 12 | 21 | 24-36 | 20.3 | 18 | 17 | 16 March 20 | 0-12 | 19.3 | 7 | 10 | 21 | 12-24 | 18.6 | 4 | 11 | 21 | 24-36 | 12.6 | 5 | 12 | 21 June 16 | 1-24 | 22.5 | 4 | 14 | 23 ---------+-------+----------+-------------+---------+---------------

Locality, Salem, N. J. Soil type, Sassafras loam. Crop, grass. Yield, fair. =========+=======+==========+======================================= | | | Parts per million of oven-dried soil | Depth | Moisture +-------------+---------+--------------- Date | inches| content | Phosphoric | Calcium | Potassium | | Per cent.| acid (PO₄) | (Ca) | (K) ---------+-------+----------+-------------+---------+--------------- March 10 | 0-12 | 25.0 | 13 | 28 | 18 | 12-24 | 23.8 | 7 | 26 | 13 | 24-36 | 19.9 | 16 | 8 | 15 March 14 | 0-12 | 25.8 | 21 | 12 | 21 | 12-24 | 23.1 | 8 | 12 | 15 | 24-36 | 21.8 | 9 | 15 | 21 March 31 | 0-12 | 23.0 | 11 | 23 | 43 | 12-24 | 21.6 | 8 | 20 | 34 April 2 | 0-12 | 24.8 | 8 | 16 | 41 | 12-24 | 24.0 | 6 | 21 | 38 | 24-36 | 21.4 | 3 | 11 | 25 ---------+-------+----------+-------------+---------+---------------

Locality, Salem, N. J. Soil type, Sassafras loam. Crop, wheat. Yield, good. =========+=======+==========+======================================= | | | Parts per million of oven-dried soil | Depth | Moisture +-------------+---------+--------------- Date | inches| content | Phosphoric | Calcium | Potassium | | per cent.| acid (PO₄ | (Ca) | (K) ---------+-------+----------+-------------+---------+--------------- March 17 | 0-12 | 22.0 | 8 | 6 | 10 | 12-24 | 18.1 | 8 | 15 | 14 March 17 | 0-12 | 18.3 | 10 | 15 | Lost | 12-24 | 18.1 | 9 | 24 | 25 March 24 | 0-12 | 24.7 | 14 | 12 | 30 | 12-24 | 22.3 | 8 | 11 | 38 March 26 | 0-12 | 23.4 | 4 | 16 | 16 | 12-24 | 23.9 | 12 | 16 | 20 | 24-36 | 22.4 | 8 | 3 | 21 April 2 | 0-12 | 25.6 | 8 | 16 | 30 | 12-24 | 24.4 | 8 | 17 | 47 | 24-36 | 21.6 | 8 | 11 | 38 June 5 | 0-12 | 5.2 | 14 | 51 | 23 | 12-24 | 8.0 | 15 | 55 | 32 June 8 | 1-24 | 10.6 | 2 | 20 | 13 June 11 | 1-24 | 15.5 | 6 | 26 | 14 June 13 | 1-24 | 8.2 | 6 | 19 | 22 June 16 | 1-24 | 15.0 | 5 | 21 | 19 June 17 | 1-24 | 10.6 | 7 | 63 | 17 ---------+-------+----------+-------------+---------+---------------

Locality, Salem, N. J. Soil type, Sassafras loam. Crop, clover. Yield, fair. =========+=======+==========+======================================= | | | Parts per million of oven-dried soil | Depth | Moisture +-------------+---------+--------------- Date | inches| content | Phosphoric | Calcium | Potassium | | per cent.| acid (PO₄) | (Ca) | (K) ---------+-------+----------+-------------+---------+--------------- March 20 | 0-12 | 20.8 | 5 | 15 | 32 | 12-24 | 20.2 | 5 | 15 | 27 | 24-36 | 18.6 | 5 | 12 | 36 March 26 | 0-12 | 26.8 | 9 | 31 | 20 | 12-24 | 22.9 | 8 | 20 | 18 | 24-36 | 22.5 | 4 | 14 | 20 June 6 | 0-12 | 8.1 | 8 | 16 | 17 | 12-24 | 12.7 | 9 | 18 | 20 ---------+-------+----------+-------------+---------+---------------

Locality, St. Marys, Md. Soil type, Leonardtown loam. Crop, wheat. Yield, good. =========+=======+==========+======================================= | | | Parts per million of oven-dried soil | Depth | Moisture +-------------+---------+--------------- Date | inches| content | Phosphoric | Calcium | Potassium | | per cent.| acid (PO₄) | (Ca) | (K) ---------+-------+----------+-------------+---------+--------------- April 27 | 0-12 | 21.8 | 5 | 10 | 12 | 12-24 | 21.3 | 4 | 7 | 10 April 29 | 0-12 | 22.2 | 8 | 15 | 52 | 12-24 | 21.8 | 4 | 11 | 38 May 1 | 0-12 | 22.4 | 7 | 14 | 23 | 12-24 | 21.8 | 7 | 8 | 30 May 1 | 0-12 | 17.0 | 5 | 16 | 25 | 12-24 | 21.0 | 5 | 7 | 19 May 9 | 0-12 | 15.0 | 13 | 34 | 28 | 12-24 | 15.9 | 9 | 17 | 26 May 15 | 0-12 | 14.2 | 3 | 14 | 24 | 12-24 | 19.9 | 4 | 13 | 25 August 14| 0-24 | 15.0 | 6 | 11 | 13 August 15| 0-24 | 15.7 | 5 | 3 | 17 August 15| 0-24 | 16.4 | 8 | 15 | 15 ---------+-------+----------+-------------+---------+---------------

Locality, St. Marys, Md. Soil type, Leonardtown loam. Crop, wheat. Yield, poor. =========+=======+==========+======================================= | | | Parts per million of oven-dried soil | Depth | Moisture +-------------+---------+--------------- Date | inches| content | Phosphoric | Calcium | Potassium | | per cent.| acid (PO₄) | (Ca) | (K) ---------+-------+----------+-------------+---------+--------------- May 14 | 0-12 | 14.7 | 5 | 8 | 35 | 12-24 | 19.9 | 4 | 4 | 30 May 23 | 0-12 | 7.8 | 4 | 7 | 22 | 12-24 | 14.9 | 4 | 11 | 23 August 14| 0-24 | 16.0 | 4 | 4 | 16 August 15| 0-24 | 19.5 | 6 | 4 | 13 ---------+-------+----------+-------------+---------+---------------

Locality, St. Marys, Md. Soil type, Leonardtown loam. Crop, corn. Yield, good. =========+=======+==========+======================================= | | | Parts per million of oven-dried soil | Depth | Moisture +-------------+---------+--------------- Date | inches| content | Phosphoric | Calcium | Potassium | | per cent.| acid (PO₄) | (Ca) | (K) ---------+-------+----------+-------------+---------+--------------- May 8 | 0-12 | 18.2 | 9 | 12 | 29 | 12-24 | 18.9 | 10 | 7 | 26 May 18 | 0-12 | 18.2 | 3 | 24 | 38 | 12-24 | 18.8 | 6 | 19 | 28 August 8 | 0-24 | 17.5 | 7 | 30 | 18 ---------+-------+----------+-------------+---------+---------------

Locality, St. Marys, Md. Soil type, Leonardtown loam. Crop, corn. Yield, poor. =========+=======+==========+======================================= | | | Parts per million of oven-dried soil | Depth | Moisture +-------------+---------+--------------- Date | inches| content | Phosphoric | Calcium | Potassium | | per cent.| acid (PO₄) | (Ca) | (K) ---------+-------+----------+-------------+---------+--------------- May 23 | 0-12 | 16.6 | 5 | 12 | 22 | 12-24 | 17.4 | 6 | 8 | 22 August 8 | 0-24 | 19.9 | 9 | 25 | 20 August 15| 0-24 | 21.6 | 7 | 15 | 13 ---------+-------+----------+-------------+---------+---------------

It will be observed that the results given in the above tables are expressed in parts per million of oven-dried soils, in order to have some definite basis of comparison, and because it was anticipated at the time the investigation was made that larger quantities of dissolved minerals would be found under the better crops, and vice versa. An inspection of the results, however, shows that no such correlation can be made, nor in fact can any consistent correlation be made between the dissolved material and crop, soil type, water content, depth of soil or part of the growing season. It appears, therefore, that in so far as the field method of analyzing an arbitrarily prepared aqueous extract is competent, there is no evidence that there are important characteristic differences in the concentration of the mineral constituents in different soil solutions in the field.

King, however, claims that the concentration of the soil solution with respect to mineral plant nutrients, is higher in the soils of the northern states than in the soils of the South Atlantic states. See: Some results of investigations in soil management, by F. H. King, Yearbook, U. S. Dept. Agriculture, 1903, p. 159-174. Bailey E. Brown has obtained some preliminary results which suggest that there may be seasonal variations with respect to some of the dissolved mineral constituents. See, Annual Report of the Pennsylvania State Experiment Station, 1908-9, pp. 31 et seq.

The order of concentration of the soil solution can be approximated from the given data, if the assumption be made that in the preparation of the aqueous extract, soluble mineral constituents are of minor importance, other than the constituents already dissolved in the soil solution. The calculation is very laborious, is not exact, and on account of the assumptions made the actual figures obtained are of no especial value in any particular case. Remembering the method of making up the solutions from which these results were obtained, it would be sufficiently near the truth to assume an average moisture content of 20 per cent., when the figures given here for the soil approximate those which would be obtained for the soil solution. More exact calculations have been made for a large number of such cases, and it has been found from this method of estimation that the average composition with respect to phosphoric acid would be about 6-8 parts per million, and for potash about 25 parts per million, figures which agree with the results obtained for the examination of solutions extracted from saturated soils by means of the centrifuge.

The results given in the foregoing tables were obtained under great difficulties, and in some part the variations they show are undoubtedly due to inevitable inaccuracies of analytical work done under such circumstances. Some of the variations may also be due to the disturbing influences in the soil referred to above. Experience has shown, however, that the preparation of an aqueous extract of the soil of any particular field is by no means a simple matter. Extracts made from samples taken within a few feet of one another frequently show variations of the same order as with samples from entirely different fields, or even soil types. Differences in the preliminary drying out of the sample before the addition of the water, seems to result in the same order of differences as obtained between different soils. In consequence of these facts, and of the further fact that an arbitrary aqueous extract of a soil cannot be assumed to represent in any definite way the natural soil solution, the results of the field examination are inconclusive as to the concentration of the soil solution in situ. It is more necessary, therefore, that other lines of evidence should be sought as to the mineral characteristics and concentration of the soil solution. Such a line of evidence is found in certain percolation experiments.

The absorption of phosphates and potassium by soils, by Oswald Schreiner and George H. Failyer, Bull. No. =32=, Bureau of Soils, U. S. Dept. Agriculture, 1906.

If a solution of a soluble phosphate be percolated through a soil, a part of the phosphate will be removed from the solution and absorbed by the soil; that is, there will be a redistribution of the phosphate between the soil and the water. As the process continues, however, relatively less and less phosphate is absorbed by the soil and the concentration of the percolate becomes more and more nearly that of the added solution. This absorption takes place more or less closely in accordance with the simple law that the absorption of phosphates by the soil, per unit of solution which is percolating, is proportional to the total amount of phosphate which the soil may yet take from that solution if percolated indefinitely. This law is expressed by the equation

dy —————— = K(A - y) dx

where y is the amount absorbed, x amount of solution that has passed, and A is the total amount which can ultimately be absorbed by that particular soil from that particular solution. K is also a characteristic constant. If the percolation be maintained at constant rate, then t, time, can be substituted for x and the equation becomes

dy ———— = K(A - y), dt

the ordinary rate equation for a mono-molecular reaction of the first order, whether chemical or physical.

With such absorptions as are involved in soils, a clay exposes a greater amount of absorbing surface than does a loam or sand, and it will show the greatest absorption towards any particular solution, other things being equal. The curve showing the concentration of percolate would lie lower for a clay than for a loam, or for a sand. This is illustrated in the accompanying sketch diagram, where y represents concentration of percolate and t represents time.

If after percolation has proceeded for some time (in some experiments for several weeks and until the soil contained 1 or 2 per cent. of phosphoric acid) pure water be passed through the soil, then, as soon as the previously used phosphate solution has been displaced, the concentration of the percolate drops and continues practically constant for an indefinite period. Moreover, no matter what the soil may be as to texture or composition, the same concentration of percolate is obtained, namely, 6-8 parts per million, the concentration which the soils yielded prior to treatment with the phosphate solution. Similar experiments when the soils were treated with salts of potassium have given like results, although the curves obtained from passing pure water through the soils do not lie quite so close together; but the concentration of the percolate with respect to potassium generally lies somewhere between 25 and 30 parts per million.

The removal of a soluble constituent from the soil by percolating water appears to be described by a rate equation similar to that given above for absorption. If the rate of percolation be maintained constant this formula is

dx ————— = K(B - x) dt

where x is the amount removed by the percolation, with time t, K is a constant characteristic for the particular system under consideration, and B is the total amount of the constituent which may ultimately be leached out. In other words, the rate in any particular soil will depend upon the amount of the constituent still absorbed in that soil but has no necessary connection with the rate which would hold for the same amount of the constituent in any other soil.

Theoretically, two consequences follow from this law which require consideration here. The rate at which a constituent is removed gradually becomes less as percolation proceeds. If the soil contains an amount of the constituent approaching the total amount which it can absorb, as for instance is probably the case sometimes when large applications of lime have been made to the soil, the concentration of the percolating solution might be expected to change noticeably. Generally, however, a soil contains nowhere near as much phosphoric acid or potassium as it is capable of absorbing, so that the concentration of the percolating water changes but very little with respect to these constituents. It follows from the equation that if percolation continues uninterrupted, the concentration of the percolate, so far as it is determined by an absorbed constituent, must get less and less until it becomes a vanishing quantity. This state of affairs does not exist in the soil, however, for percolation by pure water does not continue uninterrupted for any length of time. The rise of the capillary water in the soil will, under normal conditions, enable the soil to reabsorb more of the ordinary mineral constituents than is removed by percolating waters. Further attention will be given the matter in another chapter.

Another but quite different line of evidence as to the probable concentration of the soil solution is furnished by the investigation of the solubility of certain phosphates. It is popularly supposed that when superphosphate containing mono-calcium phosphate, CaH₄(PO₄)₂.H₂O, is added to a soil there is a more or less permanent increase of readily soluble phosphoric acid in the soil, although a part “inverts” to the somewhat less soluble dicalcium phosphate, Ca₂H₂(PO₄)₂·2H₂O. Such probably is far from a correct view of what actually takes place. The results obtained by studying the solubility of the different lime phosphates in water at ordinary temperature (25° C.) can be expressed in a diagram similar to the accompanying sketch, which is much distorted for convenience in lettering. As the diagram indicates, when the concentration of the solution increases with respect to phosphoric acid, the lime is at first less and less soluble until the point represented by B is reached, then becomes more and more soluble until the point D is reached, from then on becoming less and less soluble, until the solution reaches a syrupy consistency. In contact with all solutions represented by points on the line DE the stable solid substance which can exist is mono-calcium phosphate, CaH₄(PO₄)₂.H₂O. Along the line CD the only solid which is stable and can continue to persist is the dicalcium phosphate. From the point C the composition of the stable solid varies continuously with the concentration of the liquid solution. Therefore, these solids form a series varying in composition from pure dicalcium phosphate to pure calcium hydroxide. One of these basic phosphates, as they would ordinarily be called, has a less solubility than any other, as indicated by the point B. All solutions to the right of the point B have an acid reaction, while all solutions to the left possess an alkaline reaction. It follows from these facts that if we start with any lime phosphate corresponding to some point to the right of B and dilute it, or what amounts to the same thing in case it has been added to the soil, if we leach it, phosphoric acid will go into solution more rapidly than will lime until the composition of the residue is that of the basic phosphate stable at B. Similarly, if we start with a phosphate more basic, lime will be removed more rapidly than phosphoric acid, until the residue has the composition of the phosphate of lowest solubility. From this point, with continued leaching, the lime and phosphoric acid will dissolve in a definite ratio, which ratio is obviously that of the phosphate of least solubility. That is to say, if the leaching process is slow, as would be the case under soil conditions, the solution would have a perfectly definite concentration with respect to lime and phosphoric acid. What the ratio of lime to phosphoric acid may be, is of no particular interest in this connection, but the order of concentration of phosphoric acid is of interest. Owing to serious analytical difficulties, this has not yet been determined with any great precision, but by interpolating on the experimentally determined curve AC, this concentration is found to be somewhere in the neighborhood of 5-10 parts per million, figures close to those obtained for the concentration of the soil solution with respect to phosphoric acid by the previously described investigations.

For reference to the literature and detailed discussion see: The action of water and aqueous solutions upon soil phosphates, by F. K. Cameron and J. M. Bell, Bull. No. =41=, Bureau of Soils, U. S. Dept. Agriculture, 1907.

Under ordinary circumstances, however, it is not probable that lime is the dominant base controlling the concentration of phosphoric acid in the soil solution, since the great majority of agricultural soils contain vastly more ferric oxide (more or less hydrated) than is equivalent to any amount of phosphoric acid that will ever be brought into the soil; and ferric phosphates are less soluble relatively than lime phosphates. Investigation of the relation of ferric oxide to solutions of phosphoric acid shows that the system is quite similar in many respects to the basic lime phosphates and water just described. When the ratio of iron to phosphoric acid in the solid is greater than that required by the formula of the normal phosphate, FePO₄, the aqueous solution will have an acid reaction and contain a mere trace of iron and an amount of phosphoric acid determinedly the composition of the solid and by the proportion of solid to water. The basic ferric phosphates seem to be solid solutions which yield a very dilute aqueous solution when brought into contact with water. What the concentration will be under soil conditions is shown by the percolation experiments cited above.

The addition of other substances will in many cases affect more or less the solubility of the soil minerals. If these substances be electrolytes, they will generally, but not always, affect the solubility of the minerals as would be anticipated from the hypothesis of electrolytic dissociation. Thus, the addition of potassium sulphate lessens the solubility and hydrolysis of a potash feldspar or a potash mica. Contrary, however, to the indications of the hypothesis, sodium nitrate decreases the solubility of a ferric phosphate. While appreciable solubility effects take place with sufficiently high concentrations, laboratory experiments indicate that the addition of such substances, even in a liberal application of fertilizers, is not sufficient to produce any great effect on the concentration of the soil solution. Similarly, it has often been supposed that the ammonia, and nitrous and nitric oxides of the atmosphere carried into the soil by rain, or formed in the soil by bacterial action, affect the solubility of the soil minerals, but it is highly improbable that the concentration with respect to these agents ever becomes sufficiently high, as laboratory investigations show to be necessary to affect appreciably the solubility of the ordinary rock- or soil-forming minerals.

Rain brings from the atmosphere into the soil two agents, however, which do markedly affect the solubility of the soil minerals, namely, oxygen and carbon dioxide. The atmosphere within the soil contains normally a somewhat smaller proportion of oxygen than does the air above the soil. Rain in falling through the air absorbs or dissolves relatively more oxygen than nitrogen. Therefore when the rain water has penetrated the soil to any considerable depth there should be, and probably is, a liberation of dissolved oxygen into the atmosphere of the soil interstices. This dissolved oxygen in becoming liberated or when dissolved in the film water appears to be especially active towards the ferrous or ferro-magnesian silicates. These minerals are, moreover, as a class probably the most soluble of the rock-forming silicates. Consequently oxygen brought into the soil in this manner is one of the most important agencies in breaking down and decomposing such minerals as the amphiboles, pyroxenes, chlorites, certain serpentines, phlogopites and biotites; at the same time there is formed ferric oxide (more or less hydrated) and silica (probably as quartz) and magnesium, potassium, calcium or sodium pass into solution, probably as bicarbonates. That the concentration of the soil moisture may thus be made temporarily abnormal is not impossible, though scarcely probable.

The soil atmosphere has normally a decidedly higher content of carbon dioxide than the atmosphere above the soil. Consequently the soil water is always more or less “charged” with carbon dioxide, and the presence of the carbon dioxide decidedly augments the solvent powers of the water towards a great many and different kinds of rock-forming or soil minerals.

For references to the literature see Bull. No. =30=, Bureau of Soils, U. S. Dept. of Agriculture; also, The action of carbon dioxide under pressure upon a few metal hydroxides at 0° C., by F. K. Cameron and W. O. Robinson, Jour. phys. chem., =12=, 561-573, (1908); The influence of colloids and fine suspensions on the solubility of gases in water, Part I. Solubility of carbon dioxide and nitrous oxide, by Alexander Findlay and Henry Jermain Maude Creighton, Trans. Chem. Soc., =97=, 536-561, (1910).

What the mechanism of the reaction may be is far from clear. The obvious explanation, at least in the case of the ordinary silicates of the alkalis or alkaline earths, is that by forming bicarbonates of the hydrolyzed bases, the active mass of the reaction product with water is decreased and hydrolysis thereby increased. But this explanation is apparently insufficient to account for the effects sometimes observed. It has been shown that the passage of carbon dioxide through solutions of the silicates, will produce more or less slowly a precipitation of silica, and there seems little reason to doubt that it does induce to some degree a decomposition and consequent greater solubility of the silicates of the alkalis and alkaline earths. It also increases to an appreciable extent the solubility of the phosphates of iron, alumina, and lime. Therefore, the variation in the content of carbon dioxide in different soils, and its continual variation from time to time in any one soil, must be expected to produce corresponding changes in the soil solution with respect to such bases as potassium and lime, and also with respect to phosphoric acid. This has been verified experimentally with aqueous extracts of soils, the solutions being charged with carbon dioxide while in contact with the soils. It is not conceivable, however, that any great difference can exist in the partial pressures of carbon dioxide in different soils which are in a condition to support crops, and therefore great absolute differences in the mineral content of the soil solution are not to be anticipated, nor are they actually observed.

See, for instance, the results obtained by Peter, Proceedings of the 19th Annual Convention of the Association of American Agricultural Colleges and Experiment Stations, Bull. No. =164=, Office of Experiment Stations, U. S. Dept. Agriculture, 1906, p. 151 et seq.

It has long been held that the organic substances in the soil have an important solvent effect on the minerals. This assumption seems quite unwarranted in the light of our present knowledge, although it is not to be denied that occasionally there may be present in the soil some soluble organic substance which influences the mineral content. Generally it has been assumed that the effective organic substances influencing the solubility of the minerals are organic acids, of which a number have found their way into past and even current literature, and which have been designated as humic, ulmic, crenic, apocrenic, azohumic acids, etc. Their existence has been predicated upon two facts: First, humus is soluble in alkaline solutions but is more or less completely reprecipitated on the addition of an excess of a strong mineral acid, a phenomenon also characteristic of many organic acids. But many other organic substances than acids are also soluble in the presence of alkalis and insoluble in the presence of an excess of strong mineral acids. Second, organic-copper complexes have been obtained from humus constituents, and supposed to be copper salts of various humus acids. The descriptions of these complexes so far given do not show that they met the usual criteria for definite compounds, but indicate on the contrary that they were the results of absorption or possibly adsorption phenomena. Consequently the existence of “humic” acids is purely hypothetical and without experimental or other scientific verification, and calls for no further consideration here.

It is a widespread and popular notion that substances with a slight solubility also dissolve slowly, and that consequently the solubility of the minerals in the soil water must necessarily be a very slow process. This is, however, a misapprehension. It has been shown with a number of the common rock-forming minerals, that if they be powdered and then stirred into a relatively small volume of water, they dissolve very rapidly at first, and in a very short time, generally a few minutes, the solution is nearly saturated with respect to the mineral. Complete saturation, however, may require many days. The general shape of curve expressing the rate of solubility is shown in the accompanying figure. For soils, this fact has been verified repeatedly, in the following way: A cell fitted with parallel electrodes is placed in circuit with a slide-wire or Wheatstone bridge in such a manner that the resistance of the cell contents can be quickly determined. Distilled water is then placed in the cell and its resistance found. Generally this will be upwards of 100,000 ohms. The soil or rock powder under examination is then added to the cell, being rapidly stirred into the water contained therein. The resistance drops to about 5,000 ohms within a short space of time, usually three or four minutes. A further slight drop in the resistance generally takes place, but it requires days, and sometimes even months to become more than barely appreciable. In this manner it has been shown that the soil and many of the common soil minerals dissolve quite rapidly if they are sufficiently fine to offer a large surface to the action of the water. It would seem to follow, therefore, that in the case of the soil solution the concentration with respect to these constituents derived from the soil minerals, will be rapidly restored whenever disturbed through absorption by plants, leaching, or otherwise.

See, for example, Umwandlung des Feldspars in Sericit (Kaliglimmer) von Carl Benedick, Bull. Geol. Inst. Upsala, =7=, 278-286, (1904).

See Electrical instruments for determining the moisture, temperature and soluble salt content of soils, by L. J. Briggs, Bull. No. 15, and the electric bridge for the determination of soluble salts in soils, by R. O. E. Davis and H. Bryan, Bull. No. =61=, Bureau of Soils, U. S. Dept. Agriculture.

That the minerals of the soil, or a powdered mineral or rock-powder, will dissolve continually as the concentration of the solution in contact with it is disturbed by abstraction of a dissolved mineral substance, has been shown by numerous experimenters. An apparently obvious way to test this point would be to treat the soil sample with successive portions of water, and to analyze the successive portions for the dissolved mineral substances. This method, however, involves serious experimental difficulties, owing to the smaller sized mineral particles being suspended in the mother liquor, thus precluding satisfactory decantation and clogging filters. Moreover, such a process in no case simulates field conditions. To meet these difficulties, the soil or mineral powder has been placed between two porous media, as in the space between two concentric cylinders of unglazed porcelain, the space being closed by a rubber stopper. To the interior cylinder is fitted a stopper carrying a tube of insoluble metal, such as platinum or tin. This tube is bent into a goose-neck form, and just below the stopper the tube is perforated with a small opening. The whole apparatus is filled with water and set in a beaker, also filled with water. The metal tube is made the cathode in an electric circuit, a platinum or other suitable anode being introduced into the beaker. In a few minutes the dissolved and hydrolyzed bases pass into the cathode chamber, and as the water also accumulates in the chamber by electrolytic endosmosis, a solution of the bases dissolved from the soil minerals drops from the end of the metal goose-neck. By adding water to the outer beaker from time to time, a steady stream of alkaline solution has been obtained for months, and in no case yet has a soil thus treated failed to continue to yield up the bases it contains in its mineral particles. The acids, such as phosphoric acid for example, are of course found in the water outside the porous cells, and in the case of the phosphoric acid it also appears to continue indefinitely to be withdrawn from the soil. It thus appears that as the products of solution and hydrolysis are removed, by such an endosmotic device as that just described or by the roots of growing plants, by leaching or otherwise, the soil minerals will continue to dissolve.

For detailed description of the apparatus and experimental data, see Bull. No. =30=, p. 27, et seq., Bureau of Soils, U. S. Dept. Agriculture.

The foregoing arguments as to the concentration of the soil solution with respect to those constituents derived from the soil minerals, are based on the generally recognized principle that a material system left to itself tends towards a condition of stable equilibrium or final rest, that is, a condition where such changes as are taking place are so balanced that no change occurs in the system as a whole. But the soil is a system continually subject to outside forces and influences, and as pointed out above, is of necessity a dynamic system. It is doubtful in the extreme if any soil in place is ever in a state of final stable equilibrium. It would be natural, therefore, to expect and to find that even if the solution in the soil were dependent on the solubility of the soil minerals alone and were continually tending towards a definite normal concentration, actually this concentration would seldom if ever be realized. Most important in this connection is the fact that the concentration of the soil solution is always dependent in some degree upon the concentration of the soluble constituents in the solid phases in other than definite chemical combinations. Other factors affecting the concentration of the mineral constituents in the soil solution are always existent, and theoretically at least, can not be ignored. Nevertheless a priori reasoning as well as the experimental evidence at hand indicates that the various processes taking place in the soil as a whole continually tend to form and maintain a normal concentration of mineral constituents in the soil solution.