BACTERIOLOGICAL, PHYSIOLOGICAL, ETC. 313 ORGANIC ANALYSIS. Estimation of Benzol in Coke-Oven Gas by Burrell’s Vapour Apparatus. L. C. Whiton. ( J . Ind. and Eng. Chem., 1916,8,733-734.)-Burrell and Robertson’s apparatus (ANALYST, 1915,40,162) is not suitable for estimating benzol in coke-oven gas. The manometer readings range from 10 mm. for raw gas to 1 mm. for de- benzolised gas, and with an ordinary manometer it is difficult to read closer than 1 mm.In a vertical position it is impossible to surround the apparatus with CO,-alcohol slush with economy of this material. Again, the phosphoric anhydride used requires frequent renewal, which is impossible with the apparatus of Burrel and Robertson. The principle of their method, however, is excellent, and, with modifications of the apparatus, has proved of practicaJ service.The author gives314 ABSTRACTS OF CHEMICAL PAPERS the manometer an inclination of 1 in 10 from the horizontal, and thus gets large readings. The bulb is laid horizontally in a trough, with a notch a t the end through which the stem of the apparatus could pass. In this position it could be more economically surrounded with carbon dioxide slush.The trough, 8 x 24 x 24 inches, has a hinged cover and double walls with the air evacuated. As it is difficult to make this vacuum chamber absolutely tight, it is equipped with a cock by means of which it can be put in connection with a pump from time to time. The phosphoric anhy- dride is contained in a separate tube connected to the rest of the apparatus by stout rubber tubing.The method is substantially exact for benzol, but underestimates toluol, xylol, etc. The total of these substances may be underestimated by almost 25 per cent., so that the method cannot properly be spoken of as a means of estiniat- ' ing benzol and its homologues. As a means of controlling benzol scrubber efficiency, however, it is of some value. Finally, it is stated that the method is more difficult of operation than it appears and requires an operator skilled in handling such apparatus.G. C. J. Some Properties of the Water in Coal. H. C. Porter and 0. @. Ralston. ( U.X. Bureau of Mines, Technical Paper 113, p. 28.) - Air-dried coal may contain from 1 to 25 per cent. of moisture, according to the source, origin, and size of the sample.It can be substantially deprived of this moisture by heating for a short time in a current of dry air at 100" C. or by prolonged exposure to dry air a t lower temperatures. The use of reduced pressure haafens the process, but does not carry it farther. The dry coal on exposure to ordinary atmospheric conditions reabsorbs moisture, in amount depending on the character of the coal and the humidity of the atmosphere. If the atmospheric conditions are identical with those under which the coal was air-dried, it will reabsorb moisture until it has substantially the same water-content as the original air-dried coal.I n the experiments described in the paper, the final moisture content was always slightly less than the initial moisture content, but this may well be due to the facts that the attainment of final equilibrium is very slow and that the experiments were not indefinitely prolonged.It is shown that the vapour-pressure of the water in air-dried coal is below normal and that the degree of deficiency in this respect varies widely among different coals. Heat is produced by wetting dry coal or partly dry coal containing less than its normal percentage of inherent water.The relative quantity of heat generated depends on the kind of coal and its relative deficiency in moisture as referred to its maximum normal content. I n other words, the thermal effect of wetting varies directly as some function of the relative vapour-pressure deficiency in the coal. The above facts are discussed from the standpoint of colloidal theory, which in the authors' opinion affords a ready explanation of them.G . C. J. Preparation of Diphenylamine-Sulphuric Acid Reagent. W. Tonius. (Zeitsch. Unters. Nahr. Genussm., 1916, 31, 322.)-T he diphenylamine-sulphuric acid reagent used in Tillmans' method (ANALYST, 1911,36,67) for the detection and estimation of nitrates may be prepared with ordinary sulphuric acid.If any blueORGANIC ANALYSIS 315 coloration appears, due to the presence of nitric acid in the sulphuric acid, it may be discharged by heating the reagent a t 110" C.; this heating expels the nitric acid and does not injure the reagent. w. P. s. Relationship between the Physical and Chemical Constants of Fats. H. J. Backer. (Chem. Weekblad, 1916, 35, 954-967.)--The relationship between the refractive index, n, specific gravity, d, and the saponification value, V , and iodine value, I , of ordinary fats, may be expressed by the following equation: n2t - 1 100 h-2 X d; = 33.07 + 0*000751 - 0.01375V + 0.002 ( t - 15).In the presence of oxy-acids the first figure of the equation is lower. For example, the calculated and observed figures for linseed oil were 30.56 and 30.66 respectively, while in the case of castor oil the calculated index was 30.63 and the observed value 29.51 at 15" C.The relationship also holds good in the case of hydrogenated oils, C. A. H. Sensitive Indicator for Acidimetry and Alkalimetry and for Determina- tion of Hydrogen Ion Concentration. M. T. Bogert and G. Scatchard. ( J . Amar. Chem. Xoc., 1916, 38, 1606-1615.)-A dinitrobenzoylene urea has been dis- covered whose monosodium salt is a very sensitive indicator for hydrogen ion con- centrations between and changing from colourless to greenish-yellow.In its behaviour as an indicator i t resembles p-nitrophenol more closely than any other well-known indicator. Like p-nitrophenol, its yellow colour makes it unsuitable for work by artificial light.It is but slightly affected by neutral salts, not at all by chloroform or toluene, and proteins have no more influence upon it than upon p-nitrophenol. Its colour is unchanged by nitrous acid. It can be used in the cold or a t 100" C. It gives a sharp end-point with ammonia and hydrochloric acid, but cannot be used to titrate carbonates. It should be superior to rosolic acid for many purposes for which that indicator is used.It may be prepared as follows: Twenty grms. anthranilic acid are dissolved in 700 C.C. water and 15 C.C. hydro- chloric acid by warming. To the filtered and cooled solution, a solution of 15 grms. potassium cyanate in 50 C.C. water is added slowly with continuous stirring, this stirring being continued for twenty minutes after all the cyanate solution has been added.The uraminobenzoic acid precipitates as a pasty white mass of microscopic needles. Sodium hydroxide (300 grms.) is added, the solution being kept cool. The uramino acid dissolves and the sodium salt of benzoylene urea soon separates. The latter is filtered off, dissolved in 1,000 C.C. of boiling water, precipitated with acetic acid and the free benzoylene urea filtered off, washed with water and dried at 120" C.Of this product, 10 grms. are heated on the water-bath with 100 C.C. sulphuric acid and 12 C.C. nitric acid are added. Heat is evolved and the mixture becomes at first bright red, but soon changes to bright yellow. After digesting for an hour a t 100" C., the mixture is cooled and poured into 1,000 C.C.of ice and water, the precipitate filtered off, washed with water and recrystallised from 1,000 C.C. of 50 per cent. acetic acid. The yield of About 22 grms. are obtained.316 ABSTRACT8 OF CHEMICAL PAPERS dinitrobenzoylene urea is about 14.3 grms. For the preparation of the sodium salt, 25 grms. are dissolved in 115 C.C. ;: sodium hydroxide and 500 C.C. boil- ing water and the solution is filtered and cooled. A mass of long, bright yellow needles crystallises out.These are dried between filter paper and finally in a vacuum desiccator. For use as an indicator, a 0.3 per cent. aqueous solution is made. G. C. J. Determination of Very Small Quantities of Water in Alcohol by Means Of the Critical Solution Temperature. V. Rodt. (Mitt. K .'Naterialpriif ., 1915,33, 426-433; through J . SOC. Chem. I d . , 1916, =,938.)-The method devised by Crismer has been rcinvestigated and proved to be capable of giving extremely accurate results. A suitable petroleum oil giving sharp indications without artificial cooling may be prepared by mixing transformer oil with commercial paraffin oil. For standardising the oil, perfectly anhydrous alcohol must be prepared.About 2 litres of 99 per cent. alcohol are placed in a large flask with 300 grms. of quicklime, and the flask is suspended in a water-bath which can be quickly removed. The aleohol is gentlly boiled for some days under a reflux condenser protected by a phosphorous pentoxide guard-tube. The Gondenser is then reversed and arranged to deliver the distillate into a, wide, corked burette, also guarded witch moisture-absorption tubes.Successive portions of alcohol are distilled over into the burette and tested with the petroleum until a constant mini- mum value is attained. For the test, all the apparatus must be dried in an oven a t 150" C. and every precaution taken against access of atmospheric moisture. An Erlenmeyer flask of about 50 C.C.capacity fitted with rubber stopper is accurately tared, a few drops of water are introduced rapidly from a pipette and the weight noted, then 15-25 C.C. of alcohol are drawn from the distillation burette and the flask is weighed again, so that the gravimetric composition of the aqueous alcohol is accurately known. The rubber stopper is replaced by another, also dried in the oven, having two holes, through one of which passes a standardised thermometer, reading to 0.2" C., and through the other a short glass rod.The glass rod is with- drawn, the point of a burette being inserted through the hole, and a measured volume of petroleum is run in, calculated from its sp. gr. to amount to 48/52 by weight of the alcohol employed; the measurement should be accurate to 0.2 C.C.The rod is replaced, the contents of the flask are warmed with gentle rotation until the liquid is perfectly clear, and rotation is continued while cooling, until a sudden sharp clouding of the solution appears. The critical temperature may be noted with an accuracy of 0.2" C., corresponding t o 0.01 per cent. of water in the aloohol. A standardisation curve is plotted for a given petroleum oil which may be used subsequently for rapid and simple estimations of water in strong spirits; the rise of critical temperature is almost a straight line function of the percentage of water within the limits of 0-2.5 per cent.Rinck's Method for the Detection of Methyl Alcohol. G. Fendler. (Zeitsch. Unters. Nahr. Genussm., 1916, 30, 228-23O.)-This method (ANALYST, 1914, 39, 407) was found to be unreliable owing to the fact that methyl alcohol yields traces of formaldehyde tinder the conditions of the test.w. P. s.ORGANIC AEALYSIS 317 Titrations with Oxalic Acid, using Methyl Orange as Indicator. G. Brwhns. (Zeitsch. anal. Chem., 1916, 55, 321-340.)-Calcium, barium, strontium, and magnesium bicarbonate solutions may be titrated directly with oxalic acid solution, using methylorange as the indicator.Alkali solutions containing car- bonate or borate can be titrated in t'he same way provided that calcium chloride, strontium chloride, barium chloride, zinc sulphate, cadmium sulphate, manganous sulphate, silver nitrate, or lead nitrate be added to tlhe solution previous to the titration.In the titration of ammonia solution, the addition of calcium chloride or barium chloride is necessary. It is mentioned that the addition of 35 grms. of crystallised boric acid per litre of i% ammonia solution renders the latter SO stable that itl may be kept in an open vessel for several days without losing in sfrength as determined by titrat,ion with oxalic acid solution. The standard oxalic acid solution used in the titrations should be prepared with water free from carbon dioxide.w. P. s. Precipitation of Cholesterol and Phytosterol by Digitonin. 0. Bfeff er. (Zeitsch. Unters. Nahr. Genussm., 1916, 31, 38-40.)-The following procedure is recommended for the rapid collection (by fihation) of the precipitate obtained when fatty acids from a fat or oil are treated with digitonin: The fatty acids are melted, digitonin solution is added, and the hot mixture poured on a filter which is placed in a water-oven until the fatty acids have passed through; chloroform should not be added, as its presence retards the rate of filtration by destroying the granular state of the precipitate.The precipitate and filter are then washed with hot chloroform, next with ether, and the filter is dried; the film of precipitatc can now be removed readily from the filter and treated with acetic anhydride, etc.A specimen of hydrogenised fish oil (m.-pt. 52" C.) examined by the author yielded a cholesteryl acetate which melted at 112" C. w. P. s. Detection of Phytosterol by Precipitation with Digitonin. A. olig. (Zeitsch. Unters.Nahr. Genussm., 1914, 28, 129-138.)-The method described by Marcusson and Schilling (AKAL'S'ST, 1913, 38, 458) was applied to numerous mixtures of animal and vegetable fats and oils, and it was found that the melting-point, of the acetate, after it had been recrystallised once or twice, almost always indicated whether phytosteryl acetate was present with the cholesteryl acetate. The pre- sence of paraffin and of waxy or resinous constituents of shea and mowrah fat,s did not interfere with the test.The author agrees that the cholesterol and phyfo- sterol can be precipitated with digitonin directly from the fats, and it is only when an inconclusive result is obtained, which is seldom, that i t is necessary to saponify the fat and make the precipitation from the fatty acids.w. P. s. Estimation of Phytosterol in Vegetable Oils. M. Klostermann and H. Opitz. (Zeitsch. Unters. Nahr. Genussm., 1914, 28, 138-145.)-The quantity of phytosterol present in a vegetable oil depends to a certain extent on the method by which the oil has been obtained from the seeds, etc. Usually, the quantity varies from 0.133 per cent. in olive oil to 0.549 per cent.in sesame oil. The greater318 ABSTRACTS OF CHEMICAL PAPERS part of the phytosterol occurs in the oil in the form of its ester (cf. ANALYST, 1914, 39, 310), and consequently, in the detection of vegetable oils in animal fats, it is advisable to saponify the fat before the cholesterol and phytosterol are precipitated with digitonin. The latter precipitates the alcohols only and does not react with the esters.w. P. s. Estimation of Volatile “Thinner” in Oil Varnish. E. W. Boughton. (Chem. News, 1916, 114, 75-77.)-The following methods were found to yield satisfactory results, the maximum error being about 1 per cent. in the case of a varnish contain- ing 50 per cent. of thinner. The evaporation of the latter (turpentine or petroleum spirit) from drying films of freshly prepared varnishes is practically complete in two hours at 30” C.Hethod 1 : One hundred C.C. of the varnish are steam-distilled, the vessel containing the varnish being maintained a t 115”-120” C. until from 550 to 700 C.C. of aqueous distillate has been collected; this requires about ninety minutes. The water is then separated and the oily portion of the distillate weighed. To the weight is added 0.3 grm. for each 100 C.C. of water collected in the distillate. Method 2 : About 1.5 grms. of the varnish is poured over the bottom of a, flat- bottomed metal basin S em. in diameter, and the basin is then heated in an air- bath a t 110”-115” C., until its contents cease to decrease in weight. Method 3: Brown’s method (see ANALYST, 1914, 39, 560). Method 4: A small stoppered weighing bottle, containing about 1 grm. of the varnish and a small camel’s-hair brush, is weighed. The varnish is then poured on to a glass plate, 10 by 15 em., and brushed over the surface in a thin film, and the bottle and brush are re-weighed. The plate is now placed in a cabinet with free access of air until no further loss of weight takes place; about two hours’ drying is required. w. P. s.