Reagents.

Nitric acid.-50 cc of nitric acid (1.40 sp. gr.) in 1,000 cc of water.

Sodium nitrate wash.—10 grams of sodium nitrate per 1,000 cc of water.

Phosphomolybdic acid solution.-100 grams of phosphomolybdic acid (Kahlbaum's preferred) in 750 cc of water and 250 cc of nitric acid (1.40 sp. gr.). This solution must be freshly prepared-not over three or four days old before using. If properly made the evaporated residue from a portion of this solution is never white and readily redissolves in the dilute nitric acid solution in the cold.

Standard solutions.-Standard caustic potash and nitric acid prepared for volumetric phosphoric acid diluted to 2 volumes. One cubic centimeter of this potassium hydroxid solution is equal to 0.812 mg of potassium oxid.

Determination.

Transfer 10 cc of solution to a platinum dish, add 0.25 cc of sulphuric acid (1 to 1). Evaporate to dryness and ignite to whiteness. Dissolve residue in hot water plus a few drops of hydrochloric acid and transfer to a tall 200 cc beaker, add 30 cc phosphomolybdic acid solution and slowly evaporate to complete dryness on top of a steam bath.

It requires approximately 22 mg of phosphomolybdic acid, in order to have an excess, for each milligram of potassium oxid present.

Add 30 cc of nitric acid wash to the dried residue and stir thoroughly in the cold, with a grinding motion with a policeman, allow to settle a moment and decant supernatant liquid at once through a gooch crucible packed with moist filter paper pulp, approximately one-sixteenth inch in thickness. Wash twice by decantation with sodium nitrate wash, transfer precipitate to a gooch and wash with sodium nitrate wash until acid free. Transfer gooch to casserole, run in excess standard alkali solution and add phenolphthalein. Heat to boiling and titrate excess alkali with standard acid.

Some samples of asbestos seem to hold or "fix" some of the excess acid, making the gooch filter very hard to wash acid free. Hence it is suggested to use a paper pulp filter. It is also desirable to make comparative tests, employing the usual asbestos filter.

If excess of phosphomolybdic acid has been used, the dried residue has a reddish hue. If excess has not been added the residue is bright yellow. Residue should not appear white.

In each case, run blanks to ascertain corrections to be made for impurities.

It is also desired that in sample No. 1 determinations of potash be made, not only by the official method (which provides for direct evaporation of the solution without addition of ammonia and ammonium oxalate), but also by the method applicable to mixed fertilizers, adding ammonia and ammonium oxalate, followed by evaporation and subsequent ignition with sulphuric acid.

Several chemists have urged that this latter method of procedure be tried, as it is claimed that the official method for potash salts gives too high results owing to impure precipitates.

The reports of results of cooperating chemists are as follows:

E. L. Baker, Geneva, N. Y.: Moist filter paper pulp was used in one of each set of duplicates and a thick pad of asbestos in the other, with no appreciable variation in results. In some cases the precipitate showed a tendency to run through the filter paper pulp. It was easier, however, to wash the filter paper free from acid. Corrections were made for a blank of 0.3 cc of potassium hydroxid. Corrections were also made for blanks in the official method. You will notice that in the case of the mixed fertilizer the two methods differ by about 0.4 of a per cent. During a series of determinations I was unable to obtain any closer agreement.

E. C. Carlyle, College Station, Tex.: The use of pulped filter paper for filtering the phosphomolybdate is found satisfactory and it reduces the bumping when the liquid is heated for the purpose of dissolving the potash salt.

G. S. Farnham, Cincinnati, Ohio: I regret to report that I failed to get checks for the volumetric method.

P. Rudnick, Chicago, Ill.: It seems from the results by the official method that there is some truth in the claim that the method for mixed fertilizers when applied to sulphate of potash gives somewhat lower results. The proposed volumetric method

was given as thorough a trial as time and opportunity permitted, and although the results obtained were not very satisfactory, the method itself certainly looks very promising. The difficulties are, first, the very small amount of sample taken; second, the extreme proneness of the precipitate to go through the filter; third, the great difficulty of removing the precipitate from the sides of the beaker or casserole; fourth, the difficulty in washing all the nitric acid out of the precipitate. Asbestos is much inferior to paper pulp for filtering.

W. D. Richardson, Chicago, Ill.: With the volumetric method, following your directions, we did not have very good success.

0. M. Shedd, Lexington, Ky.: From my work, I would suggest on samples similar to No. 1 that ammonium hydroxid and ammonium oxalate be added as in the case of mixed fertilizers, and that an aliquot be used of 0.10 to 0.20 gram and not over 0.25 gram, instead of 0.50 gram, for the smaller potassium platinic chlorid precipitate can be worked better; besides it is my experience that very large precipitates carry down a greater proportion of impurities.

In addition to the above results Mr. Shedd made determinations in sample No. 1 by evaporating with sulphuric acid, igniting, and then evaporating with platinic chlorid. The results secured were 49.75 and 49.88 per cent.

M. G. Donk, Washington, D. C.: Could get no satisfactory results on sample No. 1 by the volumetric method.

E. L. Baker, associate referee, made additional comparative determinations of potash in several samples of potash salts, both with and without the use of ammonia and ammonium oxalate, in making up the solution, the results being as follows: Comparison of results obtained with and without the use of ammonia and ammonium oxalate.

Mr. Baker also reported results of determinations of potash in a number of samples by the sodium cobalti-nitrite method first proposed as a quantitative process by Adie and Wood. This method involved the use of sodium cobalti-nitrite as a precipitant, and in the original process precipitation was effected in the presence of acetic acid in a solution which should contain from 0.5 to 1 per cent of potash. Drushel b has modified this method as follows:

The solution of a potassium salt, containing not more than 0.2 gram of potassium oxid and free from ammonium salt, was treated with a rather large excess of sodium cobalti-nitrite solution, acidified with acetic acid, and evaporated to a pasty condition over the steam bath. It was then cooled and treated with from 50 to 100 cc of cold water, and stirred until the excess of sodium cobalti-nitrite was dissolved. It was allowed to settle and was decanted through a perforated crucible fitted with an asbestos felt. The precipitate was washed two or three times by decantation, after which it was transferred to the crucible and thoroughly washed with cold water. In the meantime a measured excess of standard potassium permanganate was diluted to ten times its volume and heated nearly to boiling. Into this the precipitate and felt were transferred and stirred, after which the crucible was also put into the solution, since particles of the precipitate stick persistently to its sides. After the oxidation had proceeded five or six minutes manganese hydroxid separated out and the

a J. Chem. Soc., 77: 1076.

b Amer. J. Sci., 24: 433; Chem. News, 97: 124.

color of the solution darkened. At this point from 5 to 25 cc of sulphuric acid (1:2) were added, and the solution, after stirring, was allowed to stand a few minutes. Then a measured amount of standard oxalic acid, containing 50 cc of strong sulphuric acid per liter, was run in from a burette, taking care to add an excess. The temperature was maintained a little below the boiling point until the solution became colorless and the manganese hydroxid had completely dissolved. It was then titrated to color by permanganate in the usual manner. From the whole amount of permanganate employed, the permanganate equivalent of the oxalic acid used was subtracted and the remainder multiplied by the factor calculated for the strength of permanganate used, 0.000856 being the factor for strictly tenth-normal potassium permanganate.

While work was in progress in the referee's laboratory with a view to testing the adaptability of the cobalti-nitrite methods, letters were received from several cooperating chemists commending this process quite strongly, as a result of some preliminary work which had been done with it. Mr. A. M. Peter, of the Kentucky station, reported that by the cobalti-nitrite method results of 49.89 and 49.92 per cent were obtained by Mr. G. Edgar for sample No. 1, as against 49.82 by the official method, while for sample No. 2, 4.50 and 4.48 per cent of potash was found, as against 4.41 by the official method. Mr. Baker obtained the following results on the official samples by Drushel's modification: No. 1, 50.24 and 50.85 per cent; No. 2, 4.46 and 4.31 per cent. Following are results reported by Mr. Baker, using the original Adie and Wood method, in which precipitation is effected without evaporation:

Comparison of potash determinations by the original Adie and Wood cobalti-nitrite method (volumetric) and the official method.

a The kainits and mixed fertilizers evidently did not entirely precipitate owing, probably, to improper concentration.

CONCLUSIONS.

It appears from the result of this year's work, that while some good results are obtained by the volumetric method, there are difficulties connected with the working of the process which affect the reliability and rapidity of its execution. Among these may be mentioned the trouble experienced in washing the precipitate free of acid and the tendency of the precipitate to run through, while the smallness of the aliquot used in the determinations would, of course, tend to affect the accuracy of the results. On this account it would seem desirable that work with this method be held in abeyance for the present and that a trial be made of the cobalti-nitrite method with a view to determining its adaptability to fertilizer work.

The results of tests of the employment of ammonia and ammonium oxalate in potash determinations in potash salts indicated that lower figures are secured in this way, so that from this partial investigation the contention of those who claim that the usual method gives impure precipitates would seem to be sustained. However, no positive conclusion can be reached from the limited data at hand and hence this question should be investigated further.

REPORT of committee C (FOOD ADULTERATION).

It is recommended

By L. M. TOLMAN, Chairman.

WINES.

(1) That a committee of five be appointed to cooperate with the Bureau of Standards in drawing up a standard alcohol table.

Adopted.

(2) That the question of a standard temperature of 20° for specific gravity and alcohol determinations be also referred to the committee of five.

Adopted.

(3) That the following subjects be given further study:

(a) Methods for determining glycerol.

(b) Methods for determining total, fixed, and volatile acids.

(c) Methods for determining coloring matter in genuine wines. Adopted.

It is recommended

FLAVORING EXTRACTS.

That the colorimetric method for the determination of citral in lemon extract be adopted as provisional. (See page 32.)

That the Baier and Neumann method for the detection of sucrate of lime in milk and cream be studied. (See page 53.)

Adopted.

DISTILLED LIQUORS.

It is recommended

(1) That the modified Allen-Marquardt method for the determination of fusel oil be made a provisional method.

Adopted.

(2) That in the present method (Bul. 107, Rev., p. 98) a second washing with sodium sulphate be prescribed.

Adopted.

(3) That the method for the determination of water-insoluble color in whiskies be made provisional. (See page 207.)

Adopted.

(4) That the modified Marsh test for the quantitative determination of the color insoluble in amyl alcohol be adopted as a provisional method. (See page 206.) Adopted.

(5) That the provisional Roese method for determining fusel oil (Bul. 107, Rev., p. 97) be dropped.

That methods for the detection of added oil in paprika be further studied.
Adopted.

It is recommended

MEAT AND FISH.

(1) That the study be continued in an attempt to apply, improve, or devise methods for the most accurate separation possible of the various protein bodies in meat.

(2) That the method for determining ammoniacal nitrogen by distillation under reduced pressure be compared with the magnesium oxid method now generally used. (Bul. 107, p. 9.)