than 2 hours should be consumed only when it has been shown that the particular sample does not require the full 2 hours. (2) That the Kjeldahl method be specifically limited as an official method to substances to which it is particularly applicable. (3) That the Gunning method and its modifications be specifically limited as an official method to substances to which it is particularly applicable. (4) That the next associate referee be instructed to collect data and, if necessary, conduct collaborative work to determine for what common substances the Kjeldahl and Gunning methods can be used. (5) That some statement, indicating the necessity for the careful heating of protein-containing substances at the beginning of the digestion, be incorporated into the official method, such statement to advise mixing the sample and acid mixture and to warn against heating portions of the sample not in intimate contact with the sulphuric acid. REPORT ON POTASH1. By T. E. KEITT2 (Agricultural Experiment Station, Experiment, Ga.), Referee. Samples were prepared to send out to those who had signified their willingness to cooperate, but as there was nothing in the way of improvement of method to add to the instructions sent out by the former referee, it was decided, on account of the wide range of results shown by the former cooperators, to work on the outline of the method and endeavor to prepare a very definite procedure, especially with reference to mixed fertilizers. The referee has serious misgivings about a method requiring burning below redness and subsequent lixiviation ever becoming satisfactory in the hands of the average analyst. He, therefore, undertook a combination of the moist combustion method first proposed by de Roode3 and subsequently reported by Keitt and Shiver1. The outline as finally formulated and used follows: MOIST COMBUSTION PERCHLORIC ACID METHOD. Place 2.5 grams of the sample upon a 12.5 cm. filter paper and wash successively with portions of boiling water into a 250 cc. flask, until the washings amount to 200 cc. Acidify the solution with 5 cc. of concentrated hydrochloric acid. While hot, precipitate the sulphates by adding drop by drop, in slight excess, normal barium chlorid 1 Presented by C. C. McDonnell. Present address, Keitt and Caldwell, Newberry, S. C. J. Am. Chem. Soc., 1895, 17: 85. J. Ind. Eng. Chem., 1918, 10: 219; 1919, 11: 1049. solution acidified with hydrochloric acid (10 cc. usually suffice). Cool, make to the mark and shake. Allow the precipitate to settle. Transfer a 25 cc. aliquot, corresponding to 0.25 gram, to a porcelain evaporating dish, add 30 cc. of aqua regia and evaporate to dryness on a hot plate; add a second 30 cc. of aqua regia and evaporate to dryness; then add about 10 cc. of concentrated hydrochloric acid and 20 cc. of water and evaporate to dryness. Dissolve in 20 cc. of hot water and add 5 cc. of perchloric acid (sp. gr. 1.12); evaporate on a hot plate or steam bath until copious fumes evolve. Remove and run the liquid around the bottom of the dish. If solidification does not occur upon cooling, continue the evaporation. It can be cooled rapidly by floating the dish on water. Take up the residue with 5 cc. of water and add a second portion of 5 cc. of perchloric acid; evaporate the solution until dense fumes of perchloric acid appear. Remove so that the residue does not appear baked. In case it does, repeat the last operation. After cooling add 20 cc. of 95% alcohol, stir and allow to stand 30 minutes. Decant the alcohol through a Gooch crucible, carrying an asbestos pad about inch thick, then wash the precipitate twice by decantation with 95% alcohol containing 0.2% perchloric acid; transfer the precipitate to a Gooch using the same wash, and wash until the filtrate and washings amount to about 75 cc. Finally wash twice with alcohol ether (1 to 1), using 3-5 cc. each time, to remove all of the perchloric acid. Dry for 30 minutes at 120°C. Weigh, dissolve the potassium perchlorate from the Gooch with about 200 cc. of hot water. Use another receiver and wash the pad with alcohol or ether to facilitate drying, dry 30 minutes and weigh. Calculate the loss as potassium perchlorate and potassium oxid, using the factor 0.34. Comparative results with the Lindo-Gladding, de Roode, and de Roode-perchloric acid method for determining potash. (1) That the work on the availability of potash be continued. (2) That the study of the perchlorate method be continued to include a study of the moist combustion method outlined in this report. THE EFFECT OF MANURE-SULPHUR COMPOSTS UPON THE SOLUBILITY OF THE POTASSIUM OF GREENSAND. By A. G. MCCALL (Agricultural Experiment Station, College Park, Md.). Recent work at several of the agricultural experiment stations upon the effect of manure-sulphur composts upon the availability of the phosphorus in floats has suggested the desirability of a similar study with respect to the effect of sulphur composts upon the solubility of the potassium of greensand. Accordingly, two series of composts were prepared, using as the basis for one series a greensand from New Jersey containing 5.9 per cent of potassium, and for the other series a Maryland greensand containing 1.4 per cent of total potassium. The materials added were the same for each series and were as follows: Compost No. 1.-Nothing. Compost No. 2.-Sulphur, 500 grams. Compost No. 3.-Sulphur, 500 grams; manure, 500 grams. Compost No. 4.—Sulphur, 500 grams; manure, 250 grams; soil, 250 grams. Compost No. 6.-Sulphur, 500 grams; soil, 500 grams; aluminium sulphate, 0.02 per cent; ferrous sulphate, 0.02 per cent. Compost No. 7.—Sulphur, 500 grams; manure, 250 grams; soil, 250 grams; calcium carbonate, 10 grams. Commercial flowers of sulphur, partially rotted yard manure, Collington sandy loam soil and precipitated calcium carbonate were the materials used. After being thoroughly mixed, each compost was placed in a glazed pot and water added to one-half of the water-holding capacity, after which each compost was inoculated with sulphofying organisms. The pots were kept in the greenhouse, covered with two thicknesses of muslin, and once each week the water lost by evaporation was replaced and the composts removed from the pots and mixed to provide aeration. For the water extractions, a 75-gram sample from each compost was weighed, air-dried, and 50 grams of the air-dried material shaken every 30 minutes for 8 hours with 500 cc. of distilled water. After standing overnight, the contents of the flasks were again shaken and filtered rapidly through folded No. 3 Whatman filter papers. The acidity of the water extracts was determined by first boiling an aliquot, cooling and titrating with N/10 sodium hydroxid. Sulphur was determined by acidifying aliquots with 2 cc. of concentrated hydrochloric acid and precipitating at boiling temperature with barium chlorid. The potassium determinations were made gravimetrically by the platinic chlorid method from aliquots of the water extracts, first eliminating the soluble organic matter, silicates, iron, aluminium and phosphorus by evaporation with sulphuric acid, ignition and precipitation. The duration of the experiment was 23 weeks, water extracts being made at the beginning, and at the end of each week for the first nine weeks and thereafter at the end of the twelfth, fifteenth, seventeenth, twentieth and twenty-third weeks. In the composts consisting of greensand, sulphur and manure, there was a slight but gradual accumulation of water-soluble acidity up to the end of the fifth week, after which there was a very rapid increase for three weeks. For the remainder of the period, the acidity remained at a high and practically constant level. When half of the manure was replaced with an equal quantity of soil, the acidity was greatly reduced, the maximum in one case being reached in 12 weeks and in another only after 15 weeks. When the manure was entirely replaced by soil, the acidity increased gradually to the end of the twenty-third-week period, but the amount developed was only about one-third as much as when equal weights of manure and soil were used. The accumulation of soluble sulphates closely paralleled the development of acidity, and in every case, with the increase in acidity and the accumulation of sulphates, there was a corresponding increase in the amount of potassium in the water extract. The potassium, however, continued to increase for some weeks after the acidity and sulphates had attained their maximum. The outstanding results of the experiment may be summarized as follows: SUMMARY. 1. In composts consisting of greensand, manure and soil in different proportions, an appreciable amount of the potassium of the greensand was made water-soluble through sulphofication. The most effective compost contained sulphur and manure in equal amounts. 2. The composts in which a part of the manure was replaced by soil showed a marked decrease in the amount of acidity and sulphur oxidized, as well as in the amount of potassium made soluble. When all of the manure was replaced by soil the rate of sulphofication was so slow that only a very small amount of potassium was found in the water extract. 3. The addition of small amounts of ferrous and aluminium sulphates failed to stimulate sulphofication. 4. Calcium carbonate, added to the sulphur-manure-soil compost, produced a stimulating effect during the early part of the period, but failed to increase the acidity or the soluble potassium above the maximum reached by the corresponding composts in which no calcium carbonate was used. 5. A greater total amount of water-soluble potassium was recovered in the composts containing the high potassium greensand, but a larger percentage of the total potassium was liberated from the low potassium greensand. 6. In the composts containing manure, the total amounts of potassium recovered in the water extracts varied from 9.1 per cent to a maximum of 41.3 per cent of the total initial amount present. A COMPARISON OF RESULTS OBTAINED BY THE DE ROODE, OFFICIAL LINDO-GLADDING AND FORMER OFFICIAL LINDO-GLADDING METHODS FOR THE DETERMINA TION OF POTASH IN MIXED FERTILIZERS1. By E. R. TOBEY (Agricultural Experiment Station, Orono, Me.). For some time the official Lindo-Gladding method of determining potash has not proved satisfactory to many fertilizer manufacturers and chemists. They contend that this method does not account for all of the water-soluble potash, the source of error being due to occlusion. A second source of error of an opposite nature is caused by the diminished volume of the solute, due to the volume occupied by the precipitate formed on the addition of ammonium hydroxid and ammonium oxalate. To overcome these difficulties, several different methods have been suggested by chemists but none have shown sufficient worth to be adopted as official. A method suggested by T. E. Keitt and H. E. Shiver, both of the Agricultural Experiment Station, Experiment, Ga., using the de Roode method as a skeleton, promised, from determinations made by them, to overcome the objections of the manufacturer and chemist. The method is claimed to be applicable to all commercial fertilizers, including concentrated salts, and is as follows: Place 10 grams of the sample in a 500 cc. flask and add 300 cc. of water. Keep the contents of the flask at the boiling temperature for approximately 30 minutes, cool and dilute to volume. After allowing to stand until the material has settled, filter and draw out 50 cc., an aliquot representing 1 gram. Place the aliquot in a porcelain dish and add 3-5 cc. of nitric acid to destroy any organic matter that may be present. Evaporate to dryness over a water bath, take up with hot water and an excess of hydrochloric acid. Evaporate again to dryness, take up with hot water, adding several drops of hydrochloric acid, and enough platinic chlorid to precipitate all of the potash present. (Thus all of the details through the precipitation are carried out on one bath, in almost one operation, and in a very short time.) Cover the precipitate with the acidulated alcohol. Allow to stand 15-20 minutes, in which time all iron, aluminium and magnesium will dissolve; filter, and wash with the acidulated alcohol solution until the washings are colorless, washing free of the excess of platinic chlorid. Next wash well with ammonium chlorid (saturated with potassium chlorplatinate). This washing should be thorough, for the accuracy of the method is largely dependent upon this 1 Presented by J. M. Bartlett. |