The simple and efficient manner in which this machine does the work is obvious, while with an unskilled hand it will do more and better work than half-a-dozen or more expert men could do without machinery. PROCEEDINGS AND NOTICES OF SOCIETIES. ROYAL INSTITUTION. A general meeting of the Members of the Royal Institution was held on the afternoon of December 6, Sir Arthur Keith, M.D., LL.D., F.R.S., Treasurer and Vice-President, in the chair. The Chairman announced that the Managers had elected Professor Julian Huxley Fullerian Professor of Physiology. The special thanks of the Members were returned to Mr. Frank Scudder for his gift of a crystal model, nine lantern slides used by Sir Henry Roscoe in his lectures on Faraday, and a Franklin Medal in aluminium which belonged to Faraday; and to the Department of Scientific and Industrial Research for the gift of the whole of the apparatus and equipment used by Sir William Bragg in his researches on the X ray Analysis of Crystals. Mr. W. H. Ballantyne, Mr. A. C. Eborall, Mrs. Eccles, Mr. W. C. Edwards, Dr. C. V. Drysdale, Commander J. O. Gawne, Mr. H. C. Holder, Miss McClean, Mr. A. J. Mundella, Dr. C. C. Murdock, Mr. E. A. Notcutt, Dr. Mata Prasad, Dr. . M. Robertson, Mr. E. W. Rowntree, Mr. H. G. Spicer, Mr. W. A. S. Vardy, and Mr. C. P. Williamson were elected Members. PHYSICAL SOCIETY. At the meeting on November 26, 1926, at the Imperial College of Science, Professor O. W. Richardson, M.A., D.Sc., F.R.S., in the chair, the following papers were read: Electro-Endosmosis and Electrolytic Water Transport. By H. C. HEPBURN, B.Sc., Research Student, Birkbeck College. ABSTRACT. This paper gives the results of determinations of liquid transport produced by passing an electric current through aqueous solutions of copper sulphate divided perpendicular to the flow of electricity by a diaphragm of powdered glass. The probable factors that determine the liquid transport are investigated, over a wide range of concentrations, by an examination of the dependence of the flow at constant applied voltage and of the electric charge of the diaphragm on the electrolyte concentration, and also by a study of the relation between the flow per faraday and the dilution of the electrolyte. The Input Impedances of Thermionic Valves at Low Frequencies. By L. HARTSHORN, A.R.C.S., B.Sc., D.I.C., Electrical Department, National Physical Laboratory, ABSTRACT. It is shown that accurate measurements of input admittance (or of input impedance) under various conditions can be made by means of the Schering Capacity Bridge. The input circuit is regarded as being equivalent to a condensed with a definite phase angle, , or "loss angle," = 90°, and 4, the results are expressed by stating the effective capacity and the value of tan d for each set of experimental conditions. A series of measurements made on an R valve is recorded, and it is shown that the results are in good agreement with the theoretical investigations of Miller and Nichols. The theoretical investigation has been extended to allow for the effect of dielectric losses in the valve, since these were found by experiment to be rather large, and to have an appreciable effect on the capacity and phase angle of the input circuit. It is shown that the effective capacity may vary from about 10 μuF. to 100 μμF. for an R valve, and the phase angle may vary from about 80° leading to 126° leading, depending mainly on the load in the anode circuit. Values of phase angle greater than 90° correspond to a negative resistance, or negative power factor, and occur when the load in the anode circuit is inductive. The variations of input capacity and phase angle with filament voltage, anode voltage, input voltage, and frequency are also investigated. SOCIETY OF PUBLIC ANALYSTS. An ordinary meeting of the Society was held at the Chemical Society's Rooms, Burlington House, on Wednesday, December 1st, 1926, Mr. E. Richards Bolton, President, being in the chair. Certificates were read for the first time in favour of Messrs. Solomon Greenberg, F.I.C., Frank Crafer Ray, M.A., F.I.C., and Geoffrey Charles Matthews, B.Sc., A.I.C. A certificate was read for the second time in favour of Mr. Stanley Grove Burgess, B.Sc., A.I.C. The following were elected Members of The four tests used to differentiate the high and low types of coliform organisms in water and their value as indicators of excremental contamination are discussed. As regards milk, the value of examining centrifugalised deposits for the tubercle bacillus is considered, and new tests for distinguishing pathogenic streptococci from those of bovine origin not injurious to man are described. An outline is given of the types of bacter aicausing food poisoning, together with details of laboratory methods, and the examination of canned foods is discussed. On the Presence and Detection of Furfural in Vinegar. By L. H. LAMPITT, D.Sc., F.I.C., E. B. HUGHES, M.Sc., F.I.C., AND L. H. TRACE, B.Sc., A.I.C. In the distillation of malt vinegar or wine vinegar furfural is produced by the acid hydrolysis of the pentosans contained in the vinegar; this explains the presence of furfural in commercial distilled vinegar. Added caramel in vinegar will only give the furfural it originally contained as such. A modified aniline acetate test has been devised whejein the rose-coloured furfuraniline is extracted by an immiscible solvent, amyl alcohol. Thus dark-coloured vinegar may be tested for furfural without distillation. The test is satisfactorily quantitative. The Rapid Determination of Opium in Stomach Contents. By E. R. DOVEY, A.R.S.Sc., F.I.C. A standard solution is made from ordinary prepared opium which has itself been standardised to contain 10 per cent. of morphine and 5 per cent. of meconic acid. A measured quantity of the stomach washings (which have given the reactions for morphine and meconic acid) are treated successively in a Nessler cylinder with dilute hydrochloric acid, mercuric chloride solution, and ferric chloride solution, and the resulting colour is matched with that given by the standard under the same conditions. The addition of the mercuric chloride inhibits the formation of any coloration with thiocyanic acid in the opium. A Rapid Method for the Sorting of Butters and Margarines. By C. H. MANLEY, M.A., F.I.C. The method depends upon the butyric acid content of butter fat. The filtered fat is saponified, the soap solution acidified with sulphuric acid, and the insoluble fatty acids filtered off under specified conditions. The excess of sulphuric acid in the filtrate is neutralised (methyl orange as indicator) and the titration with standard alkali continued to neutrality, with phenolphthalein as indicator. The difference between the result and that obtained in a blank test affords a measure of the soluble fatty acids. The neutral solution may be used for determining the Kirschner value. THE INSTITUTION OF ELECTRICAL ENGINEERS. The Design of City Distribution Systems, and the Problem of Standardisation. By J. R. BEARD, M.Sc., Member, AND T. G. N. HALDANE, B.A., Graduate. SUMMARY. It is believed that the future will bring a rapid growth of load and that this makes the present a suitable moment to analyse city distribution methods, settle a policy of development, and consider the possibility of standardisation. The main object of the paper is, therefore, to design a system suitable for adoption as a standard and to consider how far existing systems may gradually be brought into line with it. The various sections of a city distribution system are reviewed in sequence from the consumer to the main transmission system (or local power station) under the following headings, particular attention being given to the design of a system lay-out which facilitates expansion to deal with growing load : (a) The type and voltage of current to be delivered to the consumers. (b) The general lay-out of the low-voltage distribution system. (c) The general lay-out of the high-voltage distribution system for feeding the low-voltage system through transforming centres. (d) The voltage to be adopted for supplying the transforming centres from the main substations on the transmission system. (e) The size and spacing of transforming centres. 66 The desirable characteristics are then summarised so as to outline the proposed standard system. Suggested standards are also laid down for the capacity of the individual parts in areas of different type and load density. The above is followed by three short studies, based on the detailed calculations necessitated by the main investigation. These deal with (i) the relative costs of the different elements of the distribution system, (ii) the effect of load density on the cost of distribution, and (iii) the relative costs of direct-current and alternating-current distribution. In conclusion, some consideration is given to the general question of bringing existing city distribution systems into line with the proposed standard system with the object of attaining, ultimately complete standardisation of distribution throughout the country. CONCLUSION. In conclusion the authors feel that their investigations indicate that not only is the adoption of a standard system for all future extensions a matter of vital importance to eletericity supply undertakings, but that a great deal moer can be done than is commonly supposed in modifying and adapting existing systems so as to achieve some of the main benefits of a standard system. These conclusions are all based solely on a consideration of the self-interest of the electircity supply industry, but similar ocnclusions would undoubtedly apply if the subject were approached from the standpoint of the manufacturer. Indeed, the advantage of standardisation in facilitating mass production must be so great that the authors venture to suggest to manufacturers that it might be worth their while to cooperate with the electricity supply industry in a vigorous effort to establish standadrisation. If, for example, manufacturers could offer some immediate advantage to the consumer as regards the purchase of standard apparatus, there can be no doubt that this would greatly influence supply undertakings in making such changes as are necessary to bring existing systems into line with the standard system. As it is impossible in a paper of this type to make personal acknowledgment to those who may recognise ideas peculiarly their o.vn, the authors wish to express their general obligation to all those, both in this country and abroad, whose work has made the present paper possible. DETERMINATION OF ZINC OXIDE By B. S. EVANS, M.C., M.B.E., Ph.D., INSTITUTE OF METALS, MARCH, 1926. The method described in this paper was worked out in an attempt to account for the presence of certain inclusions in samples of brass. The only method hitherto available appears to be that of West, in which the oxide is reduced by heating the sample in a current of carbon monoxide, the resulting carbon dioxide being absorbed and weighed; this method has undesirable features, notable among which is the use of large volumes of the highly poisonous carbon monoxide; in addition to this, however, it seemed to be impossible to bring the blank to manageable proportions-in fact, there was a certain amount of evidence to show that the carbon monoxide underwent a slight amount of autoxidation, carbon being deposited on the walls of the furnace tube; moreover, the CO + ZnO reaction being reversible, West's method requires a high rate of flow of gas, and also is useless in the presence of tin or nickel. In consequence a new method was sought. T.Turner has shown that the ordinary method of determining oxygen in copper, by passing hydrogen over the heated sample, is inapplicable in the presence of zinc, owing to the volatilisation of the latter and its reduction (in the cooler parts of the furnace) of the water formed; it was, however, found by the authors that this back reaction could be suppressed by the addition of certain metals, notably copper and nickel. That the action of the metal thus introduced is simply one of absorbing, and thus lowering the vapour pressure of, the zinc is evidenced by the fact that the metal need not be in contact with the sample at all, and that, to be successful, sufficient metallic surface must be present to prevent entirely any passage of zinc beyond it; in other words, if zinc be removed from the system, thus disturbing the equilibrium, the reaction will go in the direction which produces zinc--i.e., zinc oxide will be reduced by hydrogen. The following are the details of the process. copper gauze. containing (i) Electrip tube furnace containing platinum asbestos. (j) Soda lime towers. (k) Concentrated sulphuric acid. (1) Phosphorus pentoxide tower. (m) Experimental electric tube furnace. (n) Glass-stoppered U-tube-glass beads + phosphorus pentoxide. (0) Guard tube of concentrated sulphuric acid. (p) Graduated aspirator containing water for measuring rate of flow of gas. Two rolls of copper gauze (strips of fine gauze 30 in. x 5 in. are suitable for this) are prepared, 5 in. in length, tightly rolled, and of such a diameter that they will just slide into the furnace tube; they are each fastened with a strand of thin copper wire and provided with a loop of iron wire at one end, so that they may easily be withdrawn from the furnace by a hooked rod; copper wire is also rolled into balls which will just fit the furnace tube. The two rolls and two of the balls are deoxidised by keeping for some hours at 1,000° C. in a current of hydrogen. BLANK. The silica boat which is to be used to hold the sample is now placed in the middle of the furnace tube, on either side of it one of the deoxidised copper rolls, and on either side of these again one of the two balls of Weight of ZnO wire; the stopper at each end of the tube is protected from the heat by a plug of asbestos. The apparatus is now connected up and the hydrogen turned on; when the air has been swept out the current is switched on and the furnace taken up to 1,000° C., and maintained at that temperature ;after 2 or 3 hours the phosphorus pentoxide tube is weighed, and this weighing is repeated at intervals of 2 to 3 hours until the gain in weight per litre of gas passing has reached a constant value. The rate of flow of the gas, which should be about 1 litre per hour., is measured from time to time. The gain in weight per litre of hydrogen should not exceed 0.4 mg. if the copper has been properly deoxidised. ESTIMATION. The blank having been satisfactorily determined, the furnace is allowed to cool, and when quite cold, the wire ball and roll at one end having been withdrawn, the boat is removed and the weighed sample (this should be a lump of about 80 grm.) placed in it; the boat, gauze roll, and ball are then replaced, the appratus connected up, and the hydrogen turned on. It is desirable to expose the deoxidised roll, etc., as little as possible to the air; these operations should therefore be conducted quickly. Hydrogen is passed through the cold apparatus, the U-tube being weighed from time to time, until the weight remains constant. It is advisable to take a whole day for this, so as to be able to have a clear day for the estimation. The temperature is finally raised to 980° C. (the time and rate of flow of the hydrogen being noted), and maintained at this point until the gain in weight of the U-tube again reaches a steady figure, which should be the same as that found for the blank. The total gain in weight of the U-tube-(blank × number of hours run) represents the weight of water formed from the zinc oxide. The following test estimations were carried out with different weights of zinc oxide : taken of H2O Theoretical weight of H2O |