TABLE V.-Determination of total potassium by modified method.

(1) The modified J. L. Smith method for total potassium be further tested. (2) The sodium peroxid fusion method for total phosphorus be adopted as a provisional method of this association and be further tested.

(3) The Determination of Volatile Matter" (U. S. Dept. Agr., Bureau of Chemistry, Bul. 46, Revised, p. 72; Bul. 107, p. 14) be replaced by the Determination of Total Organic Carbon" (J. Amer. Chem. Soc., 1904, 26: 1640).

REPORT ON INORGANIC PLANT CONSTITUENTS.

By W. W. SKINNER, Referee.

The work upon inorganic plant constituents has been a continuation of the scheme outlined for the preceding year. The chemists of six experiment stations signified their willingness to undertake the cooperative work, and samples were sent to each, but only a partial report from one station has been received. The lack of interest shown in the work has been rather discouraging to the referee, and it would seem that the objects of the association are defeated unless more cooperative work on proposed methods can be obtained.

The table given below is a tabulated statement of the sulphur content of two samples of cotton-seed meal and two samples of mustard-seed meal determined by three methods. The results are expressed on a moisture-free basis, and the sulphur calculated to sulphanion (SO4). For purposes of comparison the sulphur content of the samples used the previous year are included.

TABLE I.-Determination of sulphur as SO. in moisture-free cottonseed meal.

TABLE I.-Determination of sulphur as SO, in moisture-free cottonseed meal—

TABLE II.-Determination of sulphur as SO, in moisture-free mustardseed meal.

The results for total sulphur by the peroxid method are fairly satisfactory and the referee recommends that this method, which is now provisional, and which was recommended last year for adoption, be made an official method.

Two of the analysts reported considerable difficulty with the peroxid method, especially with mustard-seed meal. This was probably due to one of two causes. Either the charge was not thoroughly moistened before the addition of the sodium carbonate and sodium peroxid, or it was not allowed to stand a sufficient length of time after the addition of the peroxid for the preliminary oxidation to occur. If this oxidation is not completed before the charge is placed over the flame, almost invariably the contents will be blown from the crucible by the violent action when it does begin.

The results by the combustion method are interesting and significant. The results for total sulphur, calculated to sulphanion (SO4) agree within 0.1 per cent with the results by the peroxid method, while the results upon nonvolatile sulphur by this method are in striking agreement with the sulphur obtained in the ash by the ordinary method of ashing.

The combustion method is probably too slow and the technique too intricate for it to become the routine method for ash analyses. As an exact method for determining separately the volatile and nonvolatile mineral constituents of organic materials, it is, however, of great importance, and is therefore recommended for provisional adoption. The percentage loss of sulphur by the ordinary ashing method is shown in the table. On sample A, cotton-seed meal, the loss found by two analysts was, respectively, 89 per cent and 91 per cent, and on sample D, 89 per cent and 90 per cent. On samples C and D, mustardseed meal, the loss was 92 per cent and 92 per cent, and 89 per cent and 89 per cent by two analysts.

The loss of sulphur by volatization is very important in the interpretation of those proximate analyses in which ash is one of the factors. Thus in the analysis of a cattle food the determination of ash, by the ordinary method of incineration in air, constitutes one of the separations, the percentage of moisture, ash, fat, protein, and crude fiber being added together and subtracted from 100 to obtain the percentage of nitrogen-free extract. It is evident, therefore, that in a product with a sulphur content similar to mustard-seed meal, in which the loss of sulphur is about 4 per cent, the nitrogen-free extract would be just so much too high, provided the sulphur was not accounted for in some other form. The matter is further complicated by the fact that the sulphur is partially accounted for in the protein, as the protein figure is obtained by multiplying the nitrogen content by a factor which includes sulphur in the formula for the protein molecule.

When the association decided at the meeting in 1901 to adopt the name “inorganic plant constituents" to replace "ash," by inference, at least, the older methods for ash were abandoned. The referee believes this was a mistake which has resulted in considerable confusion and to which attention was specifically called in his report for last year. At present the official method is by ignition with calcium acetate, which in many cases is impracticable. For instance, in the analysis of a forage crop one of the determinations is ash, and it is frequently important to determine in the ash the mineral constituents. If. however, the mineral constituents in the material are determined by the calcium acetate method it is impossible to translate them in terms of the ash as determined by the official method for ash in feeds and feeding stuffs. The result has been that workers along this line have continued to use the methods for ash given in Bulletin 46, Revised, although now not official. It seems advisable, therefore, that this method of ashing by charring, extracting, etc., be

again inserted in our methods as an official method, and the referee recommends this.

There has never been an official method for the determination of iron and aluminum in ash. It is therefore recommended that the development of methods for this purpose be especially referred to the referee for next year.

It is recommended that

RECOMMENDATIONS.

(1) The peroxid method be adopted as the official method.

(2) The combustion method be adopted as provisional.

(3) The official method for ash given in Bulletin 46, page 77, be again adopted.

(4) The development of methods for determining iron and aluminum in ash be referred to the succeeding referee.

THE PHOSPHORIC ACID OF THE SOIL.

By G. S. FRAPS.

It is the purpose of this paper to present briefly some of the more important results which have been obtained at the Texas experiment station in the study of the phosphoric acid of the soil. The work has been confined chiefly to a study of the phosphoric acid soluble in fifth-normal nitric acid, and in ammonia under the conditions of the humus extraction.

AMMONIA-SOLUBLE PHOSPHORIC ACID.

The solubility of the phosphoric acid of mineral phosphates in ammonia was first studied, and it was found that ferric phosphate, vivianite, and wavellite give up a large part of their phosphoric acid to ammonia, while the calcium phosphates do not. It thus appears probable that some of the phosphoric acid soluble with humus in ammonia is of inorganic origin.

Twelve soils were then treated with potassium phosphate and a comparison made of the phosphoric acid soluble with humus before and after the treatment. We found an increase in so-called humic phosphoric acid in some of the soils, which proves that the absorbed phosphoric acid is soluble in ammonia, and some of the phosphoric acid soluble with humus must be of mineral origin. One of these soils, which is a subsoil and contains practically no humus, was nearly as rich as some black Minnesota soils in ammonia-soluble phosphoric acid, showing 0.06 per cent before and 0.228 per cent after treatment.

The distribution of the phosphoric acid in the ammonia extract as prepared for the estimation of humus was also studied. About one-ninth of the phosphoric acid is precipitated with the clay and is probably held in mechanical suspension. About one-third is precipitated with the organic matter and is probably in combination with it. About four-ninths remains in solution and is probably of mineral origin. These figures represent average results from seven Minnesota soils rich in humus.

The phosphoric acid associated with the organic matter does not diffuse and is not increased when the organic matter is precipitated in the presence of phosphoric acid, nor decreased when the organic matter is dissolved in ammonia and reprecipitated. It is concluded that the phosphoric acid is in organic

combination.

The organic phosphoric acid is present in soils in much smaller quantity than has generally been assumed, even in soils rich in humus. On account of the smallness of the quantity in the soil and the slowness of decomposition of the

humus, the organic phosphoric acid can not be considered as of great importance, except possibly in some soils rich in humus, where it may serve as a reserve store of phosphoric acid.

An increase in the phosphoric acid soluble in ammonia during processes of decay does not necessarily mean that phosphoric acid has entered into organic combination.

PHOSPHORIC ACID SOLUBLE IN FIFTH-NORMAL ACID.

Efforts were also made to ascertain what mineral phosphates are affected by the fifth-normal nitric acid. Two thousand cc of solvent and 0.2 gram of phosphoric acid mineral were brought together for five hours at 40° C. and phosphoric acid was determined in the filtrate. It was found that practically all of the phosphoric acid of iron, aluminum, and calcium phosphates, and the minerals phosphorite, vivianite, triplite, and apatite was dissolved. Wavellite, dufrenite, and variscite were only slightly affected. Experiments with other solvents gave similar results.

The conclusion drawn from this work is that the solvents can not differentiate phosphates of different agricultural value in the soil. To take an extreme case, equal quantities of phosphoric acid in acid phosphate and apatite would yield the same results with fifth-normal acid and other solvents, but when applied to the soil the crops produced would be decidedly different. We are therefore justified in comparing the phosphoric acid extracted by weak solvents only with soils which probably contain the same phosphates.

The next point considered was whether the phosphoric acid dissolved from the mineral phosphates was removed from solution by the soil; that is, if any fixation from the fifth-normal nitric acid occurred. This question was studied by ascertaining the amount of phosphoric acid dissolved, first, from the soil alone, and second, from the soil plus a known quantity of phosphate. It appeared that a number of soils have a high absorptive power for phosphoric acid from solution in fifth-normal nitric acid. There were at least three soils in which over 70 per cent of the added phosphoric acid (210 parts per million) were absorbed. Similar results were observed with other solvents tested. It is evident that the phosphoric acid extracted from some soils may represent less than 30 per cent of the phosphoric acid which actually went into solution. Two soils with the same amount of phosphoric acid extracted by fifth-normal nitric acid may really contain quite different quantities of soluble phosphates.

This question of absorption must be seriously considered in connection with soil analysis. Its significance and the correction that should be made for it are being studied at the Texas Station. We are satisfied that the fifth-normal nitric acid method will yield results of great practical value. A large number of pot experiments have been conducted, but chemical analyses can now take the place of such investigations to some extent, and we expect to discontinue pot experiments upon certain types of soils containing less than 20 parts per million of phosphoric acid, as all of these soils appear to be seriously deficient in phosphoric acid. Full details of this work will be published by the Texas station at a later date.

REPORT OF COMMITTEE ON NOMINATIONS.

Mr. B. L. Hartwell, chairman of the committee on nominations, submitted the following report: For president, Mr. Harry Snyder, of Minnesota; vice-president, Mr. W. D. Bigelow, Washington, D. C.; secretary, Mr. H. W. Wiley, Washington, D. C.

For additional members of the executive committee, Mr. B. B. Ross, Alabama, and Mr. G. F. Fraps, Texas.