22 ABSTRACTS OF CHEMICAL PAPERS APPARATUS, ETC. Apparatus for Fat Extraction. I. Selecter. (J. h d . and Eng. Chem., 1915, 7, 871.)-The condensing section E is 18 inches long and 1.5 inches in diameter, flanged slightly at the top, and drawn out at the bottom into a ground-glass end to fit the top of section B. Into E is placed a Soxhlet extraction-tube (D), fitted with a siphon- tube at the side.Above is the condenser proper (F), which is 14 inches in diameter, and hangs with a flange from the top of E. I t is fitted with a rubber stopper (G), having two holes to admit of a cooling water- supply and exit. B contains the thimble C, 5 inches long and 1 inch in diameter. The reservoir A consists of a 75 C.C. Erlenmeyer flask ground to fit B. The solvent liquid passes in a state of vapour through B, and condenses and falls into D, until in siphoning over it floods C and effects extraction, and when this is complete, C is removed, replaced by a glass capsule of similar size, and the solvent collected. Figures are cited showing that higher values are obtained for ten different materials than those yielded by the Soxhlet or (‘ straight flow-through ” methods. H.F. E. H. Determination of the Concentration of Hydroxyl Ions. F. Francis, F. H. Geake, and J. W. Roche. (J. Chm. Soc., 1915, 107, 1651-1673.)-The method for the determination of hydroxyl ion concentration based on measurements of the rate of decomposition of nitrosotriacetonamine (J. Chem. SOC., 1912, 101, 2358 ; 1913, 103, 1722) has been further developed. By measuring the pressure instead of the volume of nitrogen evolved, results for the unimolecular constant were obtained of a much higher degree of accuracy than previously, without serious diminution of the rapidity of the experiments.In a few experiments on the hydrolysis of sodium carbonate the new method gave data in good agreement with those deduced from conductivity and electro-motive force measurements. The catalysis of other nitrosoamines has been studied in the hope that some light would be thrown on the nature of the reaction, but no definite conclusions could be drawn.This part of the investigation, however, led to the adoption of nitrosovinyldiacetona- mine and nitrosoisobutyldiacetonsmine to bridge that region of ionic conoentration which, owing to the “drift” in the constants, cannot be measured through the agency of nitrosotriacetonamine.In small concentrations of the catalytic agent a, general similarity was found in the catalyses of all the nitrosoamines examined, but in concentrated solutions this was not the case, and, further, a well-marked divergence was observed in the behaviour of potassium and sodium hydroxides.Numerous results have been accumulated on the action of neutral salts on the course of the catalysis. A modified form of apparatus is described suitable for measurements either by the old method or by the new method of pressure determinations. J. F. B.APPARATUS, ETC. 23 Laboratory Circulating Pump. J. S. Morgan. (J. Chem. Soc., 1915, 107, 1710-1711.)-A simple appa- ratus is described for maintaining a circulation of water from a thermostat, through the jacket of any portion of an apparatus which must be kept at constant tem- perature, and back again to the thermostat.In the accompanying figure, T represents the thermostat, and J t h e jacket surrounding the portion of the apparatus under observation. To the top of J is fixed a bulb with three inlets, as shown.One of these inlets is connected to J, one to the water-pump, and the third to a long tube, L, which just dips into the thermostat. Over the end of L a somewhat wider tube, fitted with a side- branch, is fixed with a rubber connection. The side- branch, controlled by a stopcock, serves as a means of adjustment ; by completely closing the stopcock, water only is raised by the pump.On opening the cock a mixture of air and water is raised, which on reaching the bulb separates into water, which returns to the thermostat by way of the jacket J, and air which passes away through the pump. By increasing the number of intake-tubes, very large jackets can be kept supplied with sufficient water so that the temperature gradient becomes almost negligible.J. F. B. c Method of Mechanical Soil Analysis. 0. Swen. (Int. Nitt, fur Bodenkunde, 1915, 6,257-311 ; through Int. Inst. Agric., Bull. Agric. Intell. and Plant Diseases, 1915, 6,1307.)-The ultimate object of mechanical soil analysis is the construction of a curve with abscissae representing values having a mathematical relation to the diameter of the particles, and ordinates representing values from which the weight or number of the corresponding particles may be calculated.The construction of such a curve entails a large amount of work, and the author has endeavoured to obtain an id- of the distribution of the components, not by the isolation of the various groups, but by studying a particular character in direct relation to the size (with reference to the occurrence) of the particles.This character is the rapidity of settling of the soil particles in water. The soil sample being uniformly distributed throughout the water, the weight, P, of the particles sinking to the bottom is determined as a, function of the time, t, and the curve so obtained is characteristic for each sample. The apparatus consists principally of a cylinder 112 inches high, filled with water.About 2 mm. above the bottom, a copper plate of about the same diameter as the cylinder is suspended by thin silver wires to one arm of an accurate compensating balance. To the other arm is attached a compensating weight, which can be increased or decreased automatically. A sample of soil is poured into the cylinder ; the soil partioles fall through the water and settle on the plate, As soon as the24 ABSTRACTS OF CHEMICAL PAPERS sedimentation has begun the plate falls; shortly after it rises again, owing to the automatic increase of the compensating weight on the other arm.This process is repeated until all the soil particles are precipitated ; the increase in the compensating weight represents the weight of the sample of soil.Synchronously with this opera- tion, the time required by the soil to sink is determined by a chronograph. Thus, the speed of the sedimentation may be recorded in the form of curves; each type of soil has some special characteristic. J. F. B. Comparative Method for Determining Vapour Densities. P. Black- man. (J. Chem. Soc., 1915,107, 1500-1503).-A simple method for the determina- V tion of vapour densities by comparison with a piece of ordinary glass tubing is drawn out to a straight capillary, about 50 cm.long, to act This must be sufficiently fine to t t 4 ' pass inside a straight piece of capillary tubing y2- , 9; ~ 100 cm. long, and not too wide in bore. One 8 c - V ~ end, B, of this tube is sealed, and with the tube in a vertical position, sealed end down- wards, a small bead of mercury, N, is placed by means of the filler about halfway down the tube.Similarly, a short thread (less than 10 mm.) of the liquid to be tested is placed between the mercury and the open end A of the tube, well away from A , which is then sealed up. The end B is next opened, and when the end A has cooled the tube is placed horizontally, and the length V of the air-space between A and M is measured, also the length I, of the thread of liquid.A similar short thread of a liquid, the vapour density of which is known (d2), is next inserted through the end B, which is then resealed. After cooling, the tube is placed horizontally, and the lengths Vl of the air thread between A and N, V , of the air thread between B and N, and I, of the second liquid, are measured. The sp.gr. s1 and s, of the two liquids must be determined, also the temperature of the room, t,, and the atmospheric pressure, p. The tube is next heated in a, horizontal position to a, temperature sufficient to volatilise both liquids, and when the bead of mercury has become stable, the lengths v1 and wa of the vapour threads on either side of it are measured.The vapour density (d,) can then be calculated by a complex formula, which may be simplified down t o G = G, with an error of less than 1 per cent., suitable for all ordinary purposes, and reducing the number of measurements required. J. F. B. I e - - 3A liquid of known vapour density is described. A " I r v2 M 6 @ = - . A 3 ,q as a filler. dIV1 4 v2 Direct-Reading Viscosimeter.R. F. MacMichael. (J. Ind. and Eng. Chzem., 1915, 7, 961-963.)-1n this apparatus the oil is placed in a revolving cup within which is suspended a disc by a torsion wire about 10 inches long, which runs through the stem of a plunger and is fastened near the bottom. The cup has an oil jacket, within which is an electric heating coil, so that the temperature may be kept constant throughoutREVIEW 25 the test.At the top of the plunger is a graduated dial, secured to the stem by means of a friction disc, so that the zero mark may be adjusted. About 100 C.C. of oil &re placed in the cup, which is then covered and the thermometer inserted. The liquid is heated to the desired temperature, and the excess above the level is removed by means of a pipette. Within a few seconds after starting the motor the dial will be steady, and the reading is taken in degrees of angular deflection (300" to the circle), and recorded 8s OM. Since water at 20" C. has i+c of the absolute unit of viscosity, water at this temperature would read 10" M. By the um of lighter or heavier torsion wires readings for water of 100' M. and 1" M. are obtained. Calibration of the viscosimeter is best done by means of a standard solution of sugar syrup. Colloidal solutions may be tested with this instrument, and the readings are not affected by small particles of ampended foreign matter. C. A. M.