362 THE ANALYST. BACTERIOLOGICAL, PHYSIOLOGICAL, ETC. A New Method for Estimating Diastatic Ferments. J. Wohlgemuth. (Biochem. Zeits. 1908, 9, 1-9.)-In each of a series of test-tubes are placed 5 C.C. of a 1 per cent. starch solution and successively increasing quantities of the ferment under examination. The tubes are placed in ice-water until the last one has been prepared, and all are then inserted in a water-bath heated to a temperature of 40' C., allowed to remain there for from thirty to sixty minutes, and next transferred to the ice-water bath in order to stop further action. One drop of & iodine solution is now added to each test-tube, the contents are shaken, and the tube is selected which shows no trace of blue colour, but is next in the series to the tube exhibiting a purple or blue colour.Suppose the selected tube contained 0.02 C.C. of saliva, then 1 C.C. of saliva would have converted the starch in 250 C.C. of the solution in the flame time. The starch solution used should be prepared by rubbing down the required weight of soluble starch with a little water, adding more water to the paste, and heating the whole on a water-bath for about ten minutes. The solution is then diluted to the correct volume. w. P. s. Method for the Estimation of Fat and Unsaponifiable Substances in Animal Tissues, and a Critical Examination of the Methods now in Use. M. Kumagawa and K. Suto. (Biochem. Zeits., 1908, 8, 312-347.)-The methods at present employed for the estimation of fat in animal matters, such as dried meat powder, etc., give very varying results, much depending on the solvent used for extracting the fat.The authors, therefore, propose a method in which the substance is hydrolysed, the fatty acids formed being then extracted by means of petroleum spirit, and the true fat calculated from the weight of fatty acids found. From 2 to 5 grams of the dry tissue are heated on the water-bath for two hours with 25 C.C. of 20 per cent. sodium hydroxide solution. The mixture is then transferred t o 8 separating funnel, acidified with 30 C.C. of 20 per cent. hydrochloric acid, and the fatty acids shaken out with 100 C.C. of ether. The ethereal extract is filtered through asbestos, and the ether is evaporated at a temperature of 50" C. The residue obtained contains, besides fatty acids, colouring matters, lactic acid, and other substances, After being dried at the above-mentioned temperature for several hours, the residue is treated with about 30 C.C.of pbtroleum spirit; the impurities remain insoluble, and the petroleum spirit solution of the fatty acids is filteredTHE ANALYST. 363 through asbestos, the solvent is distilled 08 at a temperature of 50' C., and the residue is dried until constant in weight. The fatty acids thus obtained are practically free from phosphorus and nitrogen. The unsaponifiable matters present can be estimated by treating the fatty =ids in petroleum-spirit solution with alcoholic sodium hydroxide solution, adding water, and separating the substances remaining dissolved in the petroleum spirit. R. Inaba (ibid., pp.348-355), as the results of his experiments, considers that $he method proposed by Kumagawa and Suto is trustworthy for the estimation of fat in faeces and vegetable materials. w. P. s. Estimations of Fermentations with Yeast. A. Slator. (Journ. SOC. Chem. Ind., 1908, 27, 653-654.)-An apparatus is described for measuring the rate of alcohol fermentation by yeast. A given quantity of the yeast and sugar solution is introduced into a glms bottle of about 150 C.C. capacity, fitted with a rubber bung and connected by pressure tubing to a mercury manometer, which can register pressures from a vacuum to about 30 cm. of mercury. The bottle contains some glass beads, to assist in the thorough shaking of the solution; it rests in a thermo- stat, and is taken out at intervals and shaken to overcome supersaturation of gas i n the liquid.The apparatus can be exhausted and kept air-tight by means of a side tube provided with a stopcock. The change in pressure registered on the manometer is a measure of the amount of fermentation in a given time. Experiments made with the apparatus show that the rate of fermentation of dextrose is exactly pro- portional to the amount of yeast present in the solution. The rate in the case of .dextrose, liEvulose, galactose, and mannose is approximately independent of the concentration of the sugar solution when the concentrations vary from 1 to 10 grams per 100 C.C. The fermentative activity of yeasts can also be measured by means of the apparatus. Fifteen grams of the yeast are mixed with water, and diluted to a volume of 250 C.C.; 10 C.C. of this mixture are placed in the bottle, together with $0 C.C. of 5 per cent. sucrose solution, and the bottle is connected with the mano- nieter and placed in the thermostat heated to a temperature of 30" C. The apparatus is exhausted, and readings of the manometer are taken every five minutes. Some specimens of yeast show a constant activity for several hours after mixing; the activity of others changes gradually. Where yeast is used in such a manner that the initial activity is of importance, such estimations are evidently of value. Small quantities of fermentable sugar can also be estimated by the use of the apparatus, as it is possible to ferment completely 0.1 gram of dextrose in twenty minutes, and to measure the fall in pressure due to this fermentation. The meaeure- ment, after correction for the autofermentation of the yeast, is proportional to the amount of sugar introduced into the bottle of the apparatus. By carrying out two experiments, one with a sugar solution of known concentration, and the other with a solution of unknown strength, the amount of sugar in the latter can be ascertained.w. P. s.36,4 THE ANALYST. Estimation of the Total Sulphur in Urine. E. Osterberg and C. G. L, Wolf. (Biochem. Zeds., 1908, 9, 307-309.)-As the result of a number of experi- ments, the authors conclude that the total sulphur in urine cannot be accurately estimated by oxidising the latter with fuming nitric acid, and precipitating the sulphuric acid formed with barium chloride.The results obtained did not agree amongst themselves, and were always lower than those obtained by oxidation with sodium peroxide or by combustion in a calorimeter bomb. Other observers have also found too low results when using the nitric acid method. (See ANALYST, 1903, 28, 45.) w. P. s. Estimation of Creatinine in Urine by Means of Picric Acid. M. Barberio. (SOC. Chim. d i Roma, May 24, 1908 ; Chem. Zeit., 1908, 32, 722.)-In Jaffe's method (Zeit. physiol. Clzem., 1886, 10, 391) of precipitating creatinine from urine by means of potassium picrate, the double salt - C,H7N,0.C,H,(N0,),.0H -t C,H,(NO,),OK -is accompanied by appreciable quantities of uric acid and of a substance of a mercaptan nature. Since repeated crystallisation is required to eliminate these completely, it has not hitherto been found possible to make the method quantitative. For this purpose the following modification is recommended : The urine is de- colorised by treatment with blood charcoal in the proportion of 2 grams per 100 c.c., and an aliquot portion (20 c.c.) of the 6ltrate treated with an equal volume of 1 per cent. solution of potassium picrate. The precipitate is then collected, washed with alco.ho1 and ether, dried at l l O o C., and weighed. As a check, the author advocates the subsequent conversion of the precipitate into potassium sulphate. C. ,4. M.