TABLE II. Opium assays by Method II, United States Pharmacopaia, Lamar's modification. Analyst. Ash in morphine. Per ct. Per ct. Per ct. Per ct. Per ct. Per ct. Per ct. Per ct. Per ct. Per et. Per ct. Do.. 12.47 12. 32 12.54 12.39 The results, as recorded in the tables and as graphically represented, are full of interest for the alkaloidal chemist. The average difference between weighing the morphine on filter papers direct and on watch glasses after removal is, for Methods I and II, 0.06 per cent; for Method III, 0.15 per cent, and for Method IV, 0.23 per cent. These results certainly show, as in last year's work on opium, that there is nothing gained by removing the morphine from the filter paper to the watch glass before weighing in Methods I and II The time and trouble involved by such a procedure may therefore be saved. The variations in the amounts of impurities contained in the morphine obtained by the several workers amounted in Method I, by limewater, to 17.75 per cent; by potassium hydroxid, 8.72 per cent; by titration, 5.24 per cent. For Method II, with limewater, 4.20 per cent; for potassium hydroxid, 0.52 per cent, and by acid titration, 4.10 per cent. For Method III, with limewater, 2.20 per cent; for potassium hydroxid, 0.57 per cent, and by acid titration, 3.28 per cent. For Method IV, with limewater, 3.70 per cent; with potassium hydroxid, 1 per cent, and by acid titration, 3.30 per cent. These results clearly show that the morphiometric assay method of the 1890 Pharmacopoeia is unreliable. The limewater correction, as directed by the eighth revision of the United States Pharmacopoeia certainly enables the several workers to obtain more uniform results, as is shown by the average differences obtained, namely, 2.26 per cent for limewater; 1.17 per cent for potassium hydroxid, and 1.66 per cent for acid titration. While the average differences obtained by these various correction factors are markedly in excess of those usually allowed in commercial transactions, there is certainly a decided improvement over the method as originally outlined in the 1890 Pharmacopoeia. It has long been known that morphine crystallized from the various grades of opium, in the regular course of assay, carries with it more or less foreign matter, but why such varying amounts of impurity should be obtained from the same opium in the hands of different analysts, supposedly under the same conditions, is difficult of explanation. Some of the factors which undoubtedly influence the amount of impurity are, (1) the difference in time allowed by the various workers for the precipitation of the morphine; (2) the temperature at which the morphine is crystallized out, and (3) the different degrees of agitation employed by the various operators. It has been observed in the drug laboratory of the Bureau of Chemistry that different degrees of agitation employed in shaking out the morphine have a decided influence on the purity of the resulting morphine. Vigorous agitation produces purer morphine than simply shaking the flask well. The longer the time allowed for crystallizing out the morphine, the greater the amount of impurity associated with the morphine. The great variation in the amount of ash obtained by the various workers is remarkable. The high per cent obtained by Parker is undoubtedly due to excessive length of time allowed for precipitation. There is a marked relation between the amount of morphine indicated by acid titration and the amount of ash present, which clearly shows that the impurity represented by the ash possesses some basic properties. The results obtained by acid titration of morphine during the past two years of cooperative work, coupled with past experience, clearly show that pure morphine can be satisfactorily titrated with acid solutions, but such titrations are unreliable with impure morphine, such as is commonly obtained by the 1890 pharmacopœial method. The potassium hydroxid method has also been found wanting. These two methods of correction will therefore not be accorded additional trial under the present conditions of opium assaying. The average per cent of morphine obtained by the four methods employed, using limewater as correcting agent, are as follows: Method I, 12.58; Method II, 12.33; Method III, 11.62, and Method IV, with correction factor 1.12, 11.82; without this correction factor, 10.70. These results clearly show that Methods I and II are the most satisfactory, so far as per cent of pure morphine is concerned. Method III requires practically as much time for execution as Methods I and II, with a lower yield of morphine: this method must be considered inferior to either of the other two methods. Method IV can be executed in much less time than either one of the other three methods, which is an important desideratum, but the results, even with the correction factor 1.12, are considerably below those of Methods I and II. It possesses an additional advantage over Method I in that it gives more uniform results, but in this respect it is not equal to Method II, which gives the most uniform results of all the methods considered thus far. Method II gives not only the most uniform results, but after excluding the lowest result for limewater and acid titration the greatest differences are 0.6 and 0.7. per cent, respectively-differences which very closely approach the permissible variation between workers. In addition to giving closely agreeing results, Method II yields on the average only 0.25 per cent less of morphine than Method I, both based on limewater correction. It is recommended: (1) That the method recognized by the present Pharmacopoeia, eighth revision, for the assaying of opium be adopted as a provisional method by the Association of Official Agricultural Chemists. (2) That the cooperative work on the assaying of plant products be extended to other drugs. REPORT OF COMMITTEE C ON RECOMMENDATIONS OF REFEREES. It is recommended By A. L. WINTON, Chairman. (Food adulteration.) (1) SACCHARINE PRODUCTS. 1. That the methods of Circular 23, Bureau of Chemistry (Methods for the examination of maple products, Hortvet), be adopted as provisional. 1. That the following words in Bulletin 65, page 68, be eliminated: "The presence of malic acid distinguishes cider vinegars, though the quantity is often small." * * * "If a precipitate be obtained, parallel tests with silver nitrate and barium chlorid to determine the absence of chlorids and sulphates should be made before the presence of malic acid be considered proved." (Referred to committee, on motion by Mr. Tolman.) 1. That Leach's test for coumarin be adopted as a provisional method. (See Leach, Food Inspection and Analysis, p. 738.) Adopted. 2. That the colorimetric ferrous sulphate method for determination of vanillin be adopted as a provisional method. (Ibid., p. 735.) Adopted. 3. That Winton and Bailey's modification of the Hess and Prescott method for the determination of vanillin, coumarin, and acetamid be substituted for the present gravimetric methods. (J. Amer. Chem. Soc., 1899, 21: 256; 1902, 24: 1128; 1905, 27: 719.) Adopted. 4. That Leach and Lythgoe's refractometric method for the detection of methyl alcohol be adopted as a provisional method in the analysis of extracts. (J. Amer. Chem. Soc., 1905, 27: 964.) Adopted. 5. That the recommendations of the associate referee on flavoring extracts in regard to work for the following year be adopted: (a) A thorough test of the present methods for the determination of citral in oil of lemon. |