THE ANALYST. 99 OFFICIAL METHODS FOR THE ANALYSIS OF FERTILIZERS, ISSUED BY THE GERMAN MANURE MANUFACTURERS' ASSOCIATION, HARZBURG, MAY 28, 1895. CONTlCII<I'TI<D HT H. H. B. SHEPHNRD, ANGLO-CONTINENTAL GUANO WOI~KS, LONDON. I. DETERMINATION OF WATER. FOB the determination of moisture in superphosphates, saw phosphates, sulphate of ammonia, saltpetre, bone-ineal, dried blood, horn-powder, and similar substances, 10 graiiiiiies to be dried at 100" C. until a constant weight is obtained; substances containing gypsuni to be dried for three hours, Should any volatile matter, such as volatile amiiioniacnl salts, be present, it 'is to be separately determined, and the amount deducted froin the loss first Obtained. I I. 1) E: T 13: 11 3 I I Xi\ T I 0 N 0 1" I K S 0 L U H LE MATTE 13.The determination of the insoluble matter, like that of the moisture, is required principally for such purposes as calculating results into the pure or dried state. I t is to be carried out as follows : Ten grarnnies to be taken, and ( u ) if the solution is niade in acid, to be evaporated to dryness, to render the silica insoluble, and the residue filtered ofY, washed, and ignited. ( b ) If the solution is made in water, the residue to be filtered off, washed, and dried at 100" C. until the weight remains constant. 111. DETl~~E3IINATIOK OF PHOSPHOItIC ACID. A. I;cti~ospcct.-Owing to differences frequently occurring between the results obtained by works' cheinists and the clieinists of the experimental stations, it was thoaght desirable in the year 1889 to institute an inquiry into the matter.A meeting of the cheiiiists of the German experimental stations was held at Bremen in 1890, in conjunction with representatives of the German Manure Nanufacturers' Association, and it was decided to send uniformly inised samples of various superphosphates to different experimental stations and works' laboratories for the purpose of comparing the molybdate and citrate methods for determination of phosphoric acid, and of testing the various processes in use for the extraction of soluble phosphate. This plan was duly carried out ; but the results, especially those obtained by the inolylsdate method, were not considered satisfactory, and it was decided at the next meeting, held at Halle in 1891, to pursue the inquiry further.I t was desired, in the first place, to make a study of the rnolybdate method, by employing a solution of phosphoric acid of known strength, with the view of issuing reliable working in- structions for this method, and at the same time to investigate the alternative citrate method. As regards the method of extraction, a complete agreement had already been arrived at. These further experiments mere carried out siiiiultaneously in May, 1892, by the chemists of thirteen experinientaJ stations and five works. The results by the molybdate method were again discordant ; but the citrate iiiethod gave sufficiently good results , even though the ammonium magnesium phosphate precipitate contained100 THE ANALYST. some calcium phosphate. I t was, therefore, decided to accept this method for the determination of the soluble phosphoric acid in superphosphates.The cause of the differences in the results obtained by the use of the molybdate method was the subject of discussion at the fifth general meeting of the chemists of the experimental stations, held at Berlin in 1892, in which representatives chosen by the German Manure Manufacturers’ Association took part, and was traced to differences in procedure between the different chemists. I t was then thought tha,t concordant results might be obtained by issuing exact directions for working the method, and by operating upon a solution of phosphoric acid of known strength and of a composition similar to a solution of superphosphate. I t transpired in the course of the discussion that three principal modifications of the molybdate method were followed-vix., that in use at the laboratory at Halle, that of Fresenius, arid that of Professor Wagner of Darmstadt. Detailed descriptions of these three processes were accordingly prepared and sent to the members, and in the spring of 1893 they were tried at five works and twenty-nine experipiental stations, the results being published at the sixth general meeting held at Wiirzburg in the autumn of the same year.The results, however, were again disappointing, and it seemed that the more accurately the details were carried out, the greater were the divergencies experienced. I t was, nevertheless, resolved to follow up the experience already gained by making still further experi- ments, and especially to give attention to the researches of Neubauer on the subject..-\s was made known through the medium of the official report of the Dresden meeting of September, 1894,* this resolution was acted upon by the chemists of the experimental stations, but without the co-operation of the committee of the Manure Manufacturers’ Association. A solution of absolutely pure potassium phosphate was prepared by neutralizing pure potassium carbonate with phosphoric acid obtained by the coiiibustion of pure phosphorus, and with this solution fresh experiments were made both with the niolybdate and the citrate methods. No conditions, however, were laid down for working the niolybdate method, it being left to each cheinist to use any recognised mode of procedure according to his judgment.The results obtained were most satisfactory, as, indeed, would be expected when dealing with a cheinically pure substance. Operating upon 0.150 graniine phosphoric acid for each experiment, nineteen out of twenty-three of the niolybdate determinations came within 0.001 grainiiie, corresponding to -1 per cent. in a superphosphate containing 15 per cent. phosphoric acid. The determinations by the citrate method showed also a remarkable agreement, the results of ten chemists not deviating more than 0.001 to 0.002 grarnme from the mean. As a consequence, it was decided to recommend that official analyses should continue to be carried out by some approved niodification of the rnolybdate method. The matter, therefore, stands at present thus, that for ordinary superphosphate testing the citrate method may be used, but for official analyses the molybdate method is to be employed.As, however, complete agreement on matters of detail was not arrived at, it was left to each chemist to adopt such plan as according to his judgment afforded the most accurate result. The analytical commission of the * La~dwi~khticli Versuc?&sstat., xlv., 350.THE ANALYST. 101 German Manure Manufacturers’ Association, on the strength of the experience obtained during the foregoing investigation, reconirnends the so-called Halle method, a description of which follows, as the one which furnished, comparatively speaking, the most concordant results. B. diznlysis of Swperphosphatcs and lZnzo Xatcrinls. -- Agreement has been arrived at upon all points here referred to.1. Prepnrutioib of tlzc Sa?i~pZes.*-(a) Dry samples of phosphates or other artificial fertilizers may be sifted and then mixed. ( b ) In the case of daiiip materials, where this is not possible, the preparation is to be confined to a careful mixing by hand. (c) In raw phosphates and animal charcoal, for reasons before mentioned, the moisture is first to be determined. ( d ) In Zealing with substances which lose water in the process of reducing to powder, the moisture is to be deterniined in the coarse aswell as in the fine, and the results of the analysis calculated into the original state as received. 2. Extyaction of Sii~~ci~lLosi)Ii(~tc.-The extraction of superphosphate is to be carried out as follows : Twenty graiiiiiies superphosphate to be introduced into a litre flask with SO0 C.C.water and shaken continuously for half an hour, then made up to the iiiark with water, mixed, and filtered. The filtration is ho be taken in hand promptly, as if left long standing, phosphoric acid separates out. They can be worked either by hand or by motor, and should make about 150 revolutions per minute. Solutions of concentrated superphosphate (so-called double superphosphate j should be boiled with nitric acid before precipitating the phosphoric acid, to convert any pyrophosphoric acid present into orthophosphoric acid. Ten C.C. of nitric acid of 1.4 spocific gravihy are sufficient for 25 C.C. solution. Vor ordinary work the citrate method may be used, but for official analyses the inolybdate method is to be employed. For the determination of the phosphoric acid in bone-meal, fish guano, meat- meal and raw phosphates,+ and the total phosphoric acid in superphosphates, 5 grarnmes are to be dissolved in 50 C.C. nitrohydrochloric acid, consisting of 3 parts hydrochloric acid of 1.2 specific gravity, and 1 part nitric acid of 1.25 specific gravity ; or the substance may be boiled for half an hour in a mixture of 20 C.C. nitric acid of 1.42 specific gravity and 50 C.C. sulphuric acid of 1.8 specific gravity. After cooling, the solution is to be made up to 250 C.C. For carrying out the shaking operation, shaking machines are desirable. (To he continued.) * I n accordance with the resolution of the third general meeting of the chemists of the experimental stations held at Bremen in 1890. Thornas-meal is referred to in another section.