CHEMICAL NEWS, April 4, 1919, Production of Nitrogen Compounds. able temperature and pressure conditions, and used, at first, osmium as a catalyst. This process was developed in Germany by the Badische Anilin und Soda Fabrik, a corporation which was highly subsidised by the Government. As far as has been revealed this process is known to take place in a cylindrical reaction chamber surrounded by an electrical heating resistance. The catalyst used now is probably a cheaper one than osmium. This is put into the reaction chamber, which is maintained at a temperature of 500° C. Nitrogen, obtained by the fractional distillation of liquid air, and hydrogen, made by the action of steam on iron at 300° C., are mixed in the proportion of one volume of the former to three of the latter, compressed to 200 or 225 atmospheres, and passed through the reaction cylinder. About 8 per cent of the materials present are converted into ammonia. This must be constantly removed or the reaction will come to a halt by the establishment of equilibrium in the system N2+3H22NH2. The ammonia is removed by passing the mixed gases through a condenser where the ammonia is liquefied and removed, the remaining nitrogen and hydrogen being continually worked over with fresh supplies added from time to time. The development of this process is believed to have a very great bearing upon Germany's ability to supply herself with munitions of war. Ammonia made by this process is converted into nitric oxide and nitrogen per oxide, by a process to be described presently, and these in turn are made into nitric acid and nitrates. A synthetic process developed by the General Chemical Company, a process believed to be a modification of the Haber process, will be used at the Nitrate Plant No. 1 at Muscle Shoals. In this plant a part of the ammonia will be oxidised to nitric acid, and the remainder will be com bined with the acid in the formation of ammonium nitrate. While this is as yet an experimental plant, it is expected that the production will be large and that other plants of the same kind will be established at other points. Although this is a "war plant," it is expected that the product will be available for use as fertilising material. This process is expected to furnish ammonium nitrate at a lower cost than is possible with any other process, and we are awaiting the outcome of the experiment with great interest. Indirect Method First Tried. While the Haber process is of great theoretical and practical interest it must be remembered that the indirect or cyanamide process was the pioneer method for making ammonia from atmospheric nitrogen. Frank and Caro sought to make cyanide for gold extraction, obtaining the nitrogen from the air. Although not directly succeeding in this, they worked out a process the main feature of which was a method of fixing atmospheric nitrogen in a compound, calcium cyanamide, which in turn can be used in the manufacture of ammonia. In the cyanamide method calcium carbide, with the manufacture of which you are familiar, is treated in an electric oven with atmospheric nitrogen. About three-quarters of a ton of carbide is treated per oven, and the reaction is started by using the oven as a sort of resistance furnace, the charge of carbide being employed as the core offering resistance to the current. After a temperature of about 1000° C. is reached the reaction, being exothermic, proceeds to completion according to the equation CaC2+N2 CaCN2+C. After cooling, the cyanamide charge is crushed and any residual carbide is removed by careful hydration. The ammonia is then produced by treatment with steam in an autoclave. About four tons of cyanamide are charged into the autoclave containing a dilute caustic soda solution. Steam is admitted to the autoclave, and a pressure of 200 163 to 250 pounds maintained for some hours. The steam and cyanamide react according to the equation CaCN2+3H2O → CaCO3+2NH3. When the autoclave valves are opened the escaping gas is about one-fourth ammonia and three-fourths steam. The mixed gases are passed into condensers and the water and ammonia separated. The nitrogen for this process, as for the synthetic process, is made by the fractional distillation of liquid air. The Nitrate Plant No. 2 at Muscle Shoals, just been completed by the Air Nitrates Corporation, will employ this process. This plant consists of a series of factories for the manufacture of lime, carbide, liquid air, cyanamite, ammonia, and nitric acid, the latter being made by the oxidation of the ammonia. The building of this immense fixation plant has challenged the attention of the whole country because of the rapidity of the construction of such an excellent and complete series of factories, any one of which would reflect great credit upon those undertaking such a task. While this plant is a monument to American efficiency in engineering, it remains to be seen whether or not, with the country on a peace basis, it will be required or whether it can be made to pay. It should be said, however, that this is by no means an experimental plant. Every process has been well tried out, and its success or failure will depend entirely upon economic conditions of cost of materials and labour and selling price of the product. This plant will have a rated capacity of 2,000,000 pounds of nitric acid per day. The only question is, can we in piping times of peace use all of it? Ammonia produced by these methods is usually spoken of as being an artificial nitrogen product as distinguished from the ammonia from the "natural" source, coal. Until a year or two ago, when the artificial methods of producing ammonia overtook it, the production of ammonia and ammonium salts in connection with the by-product distillation of coal was the most important source of nitrogen with the exception of the sodium nitrate from Chile, which was, and still is, the most important source. During the period from 1914 to November, 1918, so far as obtainable figures show, the production of byproduct and gasworks ammonia increased only 16 per cent. Of this increase Germany is to be credited with the greatest contribution. The relatively small increase in America doubt due to the fact that a by-product coal distilling plant and the lack of any increase at all in Great Britain was no requires such a considerable expenditure of capital and such an outlay of structural material that developments along this line were not to be undertaken during the war. But the potentialities of ammonia from this natural source are very great. At present comparatively little of our American coke is being produced in by-product ovens. As more and more of the beehive ovens are replaced by the by-product plants this source of ammonia will attain greater relative prominence because of its cheapness. Byproduct distillation of coal is, as an industry in this country, still in its infancy, and even though we marvel at the present feats of our engineers in that line, another decade will see these very wonders eclipsed. With the great development along this line which is bound to come, stimulated, we trust, by the lusty growth of the American dye industry, we may expect to see coal rank as a highly important source of nitrogen and outstrip in pounds produced not only ammonia by fixation but the nitric acid from sodium nitrate. Even now it appears to some of us that if the same amount of constructive energy had been directed to the development of by-product coke plants in this country as has been devoted to the development of fixations plants we would be farther along the road of preparedness. We could have gone just as far in solving our nitrogen problems and would at the same time have saved much valuable material that has gone to waste in the beehive ovens. We might have learned a great deal by paying attention to the by-product situation in Germany. In 1910 Germany was producing 400,000 tons of ammonium sulphate, a little more than one-third of the entire world production. While accurate figures since the war began are not to be had, it is known that the by-product coke industry in Germany has greatly expanded, and there is reason to believe that they had reached an annual production of 1,000,000 tons of ammonium sulphate, or about one-half of the world's production, by the middle of the present year. Much of this material was used in the manufacture of munitions, and Germany did not have to depend too largely upon artificial nitrogen compounds. (To be continued). On soils where giant sainfoin (see Leaflet No. 280) is known to succeed, especially where it has not been grown for a number of years, this crop sown pure might usefully replace red clover on "sick" land, whilst the value of lucerne (see Leaflet No. 160) on soils that suit it must, in this connection, be strongly emphasised. The veitch (see Food Production Leaflet No. 51) is another valuable SEED MIXTURES FOR LAND AFFECTED BY leguminous plant which, as well as providing nitrogenous CLOVER SICKNESS.* OWING to the high cost of nitrogenous fertilisers, farmers should endeavour to conserve nitrogen by every means possible. Leguminous plants should therefore take a prominent place in every rotation. It is well known that in many districts red clover is liable to become "clover sick" if sown on the same field at too short intervals. It is probable that none of the leguminous forage plants are absolutely safe from attacks of "clover sickness," but it is evident that red clover is the most susceptible. Peas or beans, both valuable crops under existing conditions, could with advantage be introduced into the rotation on land that is known to be "sick" provided that it is at the same time suitable for the production of these crops, while in the formation of rotation leys, plants other than red clover should be used on "sick" land; or a mixture consisting only partially of red clover should replace large seedings of red clover by itself. Alsike clover, white clover, and trefoil do not suffer from "clover sickness" to nearly the same extent as red clover, and may therefore be used as substitutes for red clover. residues for the benefit of subsequent crops in the rotation, has a special value at present from the point of view of adding to the diminished supplies of hay. (Leaflet No. 184 on red, white, and Alsike clovers may also be consulted). London, S.W. 1, February, 1919. PROCEEDINGS OF SOCIETIES. ROYAL SOCIETY. Sir J. J. THOMSON, O.M., President, in the Chair. Magnetic Storms of March 7-8 and August 15-16, 1918, and their Discussion." CHARLES CHREE, SC.D., F.R.S. The paper contains a discussion of two magnetic storms -the first on March 7.8, the second on August 15-16, 1918. For the March storm curves were utilised from Moreover, there is reason to believe that English-grown Kew, Eskdalemuir, and Agincourt (Toronto) Observatories. late flowering red clover in particular and also broad red For the August storm Kew and Eskdalemuir curves were clover, from seed harvested in England, are less susceptible alone available. The storms were of the same general to "clover sickness" than imported clover. When, there- type as one which occurred on December 16-17, 1917, and fore, red clover is used at all in mixtures for "sick" land was discussed in a previous paper; but, unlike the previous it is desirable that every endeavour should be made to storm, they both had conspicuous Sc's. ("sudden comsecure home-grown stocks of seed for the purpose. On mencements"). The movements constituting the Sc. on (chalky) soils, or in regions of comparatively low rainfall, August 15-16 were unusually large and their oscillatory trefoil may be largely used, while in districts of high rain-character was very prominent at Agincourt and Eskdalefall more reliance should be placed on Alsike clover. The following mixtures are suggested : One Year Leys. Chalky soils or low rainfall Red clover Alsike clover Trefoil High rainfall Red clover Lbs. per acre. .. 4 6 or Trefoil {Alsike clover .. 4 It appears also from experiments conducted in Yorkshire and from experience elsewhere that red clover grown in conjunction with rye-grass is less susceptible to sickness than when grown pure; in districts adapted for growing mixed "seeds" the clovers might be decreased to half the above quantities and half a bushel of Italian rye-grass added to the mixture. Two Year Leys.-White clover should be more largely relied upon in mixtures for two year leys, in which mixtures both Italian and perennial rye-grasses would also take a prominent place. muir. In both cases, as in the storm of December, 1917, disturbance was much larger at Eskdalemuir than at Kew, especially in the vertical force. The declination changes at the two places showed rather a close resemblance, but the variations in the other elements differed at times not merely in amplitude, but in general character. The disturbance at Agincourt on March 7-8 was similar in intensity to that at Eskdalemuir, but conspicuously different in many details. There were some exceedingly large and rapid changes, especially of declination and horizontal force, at Agincourt, the range of the former element being about 2° 5', as compared with 51' at Kew. Some consideration is given to the resemblances and differences between these storms and some mean results for storms having Sc's. given in a recent paper by Dr. S. Chapman. "The Transparency of Biotite to Infra-red Radiations." By L. C. MARTIN. The paper describes a curious reversible variation with temperature in the infra-red transmission of biotite. The infra-red apparatus employed was designed by Prof. Callendar and Mr. A. Eagle. Its optical properties are examined. Tables and curves showing the variation in transmission of biotite with wave-length at various temperatures are given and certain possible explanations are examined. The general nature of the effect is a halving Board of Agriculture and Fisheries, Food Production Leaflet of the transmission for a rise in temperature of about IN opening the meeting the Chairman referred to the loss the Society had sustained by the death of Prof. G. Carey Foster, one of the founders, and mentioned that the President, Prof. C. H. Lees, was representing the Society at his funeral that afternoon. The Report of the Council was read by Prof. W. Eccles. The Report of the Treasurer was read by Mr. W. R. Cooper. Both reports were unanimously adopted. While the ballot papers for the election of Officers and Council were being collected and counted, votes of thanks were accorded to the Honorary Auditors, the Officers and Council, and the Governors of the Imperial College, the respective proposers and seconders being Mr. C. C. Paterson and Dr. Bryan, Prof. Fortescue and Dr. Makower, Prof. Howe and Dr. Owen. Mr. E. H. Rayner moved and Mr. W. R. Cooper seconded a vote of thanks to Prof. W. Eccles for his past services as joint honorary secretary. The scrutators announced the result of the election of Officers and Council for the ensuing year : President-Prof. C. H. Lees, D.Sc., F.R.S. Vice-Presidents (who have filled the office of President) -Prof. R. B. Clifton, M.A., F.R.S.; Prof. A. W. Reinold, C.B., M.A., F.R.S.; Sir W. de W. Abney, R.E., K.C.B., D.C.L., F.R.S.; Prin. Sir Oliver J. Lodge, D.Sc., F.R.S.; Sir Richard Glazebrook, C.B., D.Sc., F.R.S.; Prof. J. Perry, D.Sc., F.R.S.; C. Chree, Sc. D.. LL D., F.R.S.; Prof. H. L. Callendar, M.A., LL.D., F.R.S.; Prof. A. Schuster, Ph.D., Sc.D., F.R.S.; Sir J. J. Thomson, O.M., D.Sc., F.R.S.; Prof. C. Vernon Boys, F.R.S. 165 Dr. BRYAN asked how dissolved gas affected the results. In experiments by Worthington and others the water required to be very carefully boiled before it showed any tensile strength at all. Prof. LEES also commented on the fact that the liquid experimented on here contained dissolved gas. In Worthington's experiments with gas-free water the tensile stress was about 2 atmospheres, as far as he remembered. This seemed of a different order from that obtained by the authors. Mr. F. J. W. WHIPPLE asked where the tensile strength came into the formula. The diagram seemed to connect velocity and temperature only. Mr. BURFITT, in reply to Dr. Vincent, said the liquid was always allowed to stand until all cloudiness had disappeared, and any dissolved gas was in equilibrium. There was no doubt the presence of dissolved gas promoted rupture. Worthington's measure was static, and there was no danger of air getting in if the liquid was initially air free. In the present method it was impossible to prevent some air becoming dissolved. With reference to Mr. Whipple's comments, the tables are not primarily intended to give actual values of the tensile stress, but to give its variation with temperature. If desired, the value of C in the theoretical formula can be calculated from the figures given. A paper entitled "Vector Diagrams of Some Oscillatory Circuits Used with Thermionic Tubes," was read by Prof. W. H. ECCLES. The method of the crank or vector diagrams used commonly in the study of alternating-current circuits is applied in the paper to the assemblage made up of an oscillator, the thermionic relay maintaining it in oscillation, and the devices linking these two parts. The diagrams then serve as substitutes for the usual treatment of the problem by differential equation, and from them may be obtained all the formula. They have, besides, the advantage of exhibiting to the eye the phases of the currents and voltages in every part of the circuits. In oscillator is calculated by the usual rules of the alternatingcurrent diagram, and added geometrically to the potential drop across the tube. This total is made equal, in magnitude and phase, to the voltage applied at the instant to the grid multiplied by the voltage factor of the relay. In its turn the voltage applied to the relay depends upon and is obtained from the current running in a portion of the Oscillator. The fitting together of these lines gives all the conditions to be satisfied for the maintenance of steady Vice-Presidents-Prof. W. Eccles, D.Sc.; Prof. J. W. Secretaries-(Papers) H. S. Allen, M A., D.Sc., 5, Presturg Road, New Malden; (Business) F. E. Smith, O.B.E., F.R.S., National Physical Laboratory, Teddington. Foreign Secretary-Sir Richard Glazebrook, C.B., D.Sc., F.R.S. Treasurer-W. R. Cooper, M.A., B.Sc., 82, Victoria Street, S.W. 1. Librarian-S. W. J. Smith, M.A., D.Sc., F.R.S., Imperial College of Science and Technology. Other Members of Council Prof. E. H. Barton, D.Sc., F.R.S.; Prof. W. H. Bragg, C.B.E., M.A., F.R.S.; C. R. Darling, F.I.C.; Prof. A. S. Eddington, M.A., M.Sc., F.R.S.; D. Owen, D.Sc.; Major C. E. S. Phillips, F.R.S.E.; E. H. Rayner, M.A.; S. Russ, D.Sc.; T. Smith, B.A.; F. J. W. Whipple, M.A. After the conclusion of the general business, Prof. LEES took the Chair. A paper entitled " Temperature Coefficient of Tensile Strength of Water," by S. SKINNER M.A., and R. W. BURFITT, B.Sc., was read by the latter. The liquid is forced under pressure through a capillary constriction between two limbs of a U-tube. By trial the pressure is adjusted until the speed in the capillary is sufficient to produce rupture. This is judged by the sound and also the appearance. The whole U-tube is immersed in a bath, the temperature of which can be varied. Actual observations of rupture, velocity, and temperature are recorded up to about 100° C., from which it is deduced that the tensile strength becomes zero in the neighbourhoɔd of 245° C., which is in agreement with theory. DISCUSSION. Dr. VINCENT asked what arrangements were made to get the liquid back into the virgin state after it had been churned up by forcing through the capillary. oscillations. DISCUSSION. Prof. G. W. O. Howe agreed that you could not really understand the conditions in a circuit unless you were able to put down the currents and voltages in a vector diagram. He had attempted himself to simplify Vallauri's treatment for students, but without as much success as Dr. Eccles. He had usually looked at these circuits from this point of view: The oscillatory circuit apart from the valve has a certain equivalent resistance. To maintain oscillations the equivalent of a negative resistance must be introduced. There are two ways of varying the current in the bulbs; the P.D. on the terminals or on the grid may be varied. In the latter case the current may be increased even if the total P.D. on the bulb is diminished, and if this condition can be arrived at we have the equivalent of a negative resistance to the oscillations. You have to arrange a coupling device so that the variations of current in the plate circuit produces suitable variations of potential of the grid. From the characteristic of the plate circuit you can determine the critical value of the mutual inductance between plate and grid circuits to give the equivalent negative resistance necessary for maintenance of steady oscillations. Prof. C. R. FORTESCUE said the method of approach appears to be a distinct advance, in that the idea of the tube as a generator of an E. M.F., gEg, enables a voltage diagram to be plotted instead of the more usual current diagram. This is an undoubted advance, and simplifies the final adjustment of the diagram to suit the conditions of the tube and circuit. At various times many vector diagrams have been drawn for various oscillatory circuits; and in the whole the results have hardly come up to expectations. There appears to be three reasons for this, viz.: (a) In order to draw the diagram at all a very clear insight into the conditions is required. In other words, it is necessary to know the final result before the diagram is completed. In the diagram of Fig. 3 of the paper it is necessary to know, firstly, that the angle e is a positive angle, and, secondly, that the effective E.M.F. of the tube is in phase with the voltage applied to the grid. The latter condition is, of course, obvious from the action of the tube, but the former is by no means obvious, and has to be discovered by trial and error. If, for example, the current I is considered to be flowing through the inductance, then must be taken as an angle of lag, as will be found if an attempt is made to plot out this diagram. (b) The quantities are such that it is impracticable to plot the diagrams to scale. For example, if the numerical values of the first example on page 3 are taken, the ratio of the length of the line OUg of Fig. 3, to the length DQ is of the order of 1 to 500,000. Actually, if Fig. 3 is plotted to scale, it becomes a diagram vertically up and down the board. (c) Finally, there is the common experience of the sign difficulty in drawing the diagrams. There are many possibilities of confusion with the ordinary current diagram, but it is possible that with the author's voltage diagram these troubles will be very much reduced. It would appear that the true function of these diagrams, when numerical values are applied or when the diagrams are drawn to scale, is to justify the practical method commonly used of regarding the valve as a power supply. The anode current is in phase with the voltage across the oscillatory circuit. Taking the alternating components only, the product of the anode current and the circuit voltage gives the power supplied to the circuit. The anode current depends upon the anode and grid voltages, i.e., i=haea+hgeg, as given on page 1 of the proof of this paper. To within a small percentage this power is absorbed by the circuit losses. If R+S is the effective resistance of the oscillatory circuit, then for the oscillations to be maintained or built up, the power supply from the valve must be equal to or greater than the Ja(R+S) loss. This method of dealing with the problem has many practical advantages, and has been in use for some years. Hazeltine has described, in the "Proceedings" of the American Institute of Radio Engineers, the application of the method to various circuits, notably the circuit in which a condenser is connected across the grid inductance h. It is very well known that the author of this paper has had experience of many other circuits and applications of three electrode relays, and it is to be hoped that he may be able to give to the Society further papers on this same subject. In particular, a vector treatment of the De Forest ultraudion circuit would be of great interest, as this circuit is one which has presented very great difficulties when any attempts have been made to calculate the condition for instability. Mr. J. NICOL asked if there was any lag between the voltage applied to the grid and the effect on the current in the valve. Dr. D. OWEN said it was assumed that the action of the valve was an amplification of voltage. Actually what happened was a variation in the resistance of the valve. It seemed rather remarkable that this should be regarded as a voltage effect. He did not like Prof. Howe's idea of a negative resistance, as such a conception had no physical significance. A negative value of dVjdC was by no means the same thing as a negative value of V/C, which would be required before we could talk of a negative resistance. Did the author find it satisfactory to treat the quantity Ra as a constant ? Mr. F. E. SMITH thought the paper of great value. He agreed with Prot. Fortescue that a great amount of interesting work was still to be done in connection with these problems. that Capt. TURNER referred to a sentence occurring in the section dealing with a particular case: "It is noteworthy no resistance." This seemed to imply that an infinitesimal change in E would produce a finite change in the steady component of the anode current. Dr. ECCLES, in reply, said that he did not think vector diagrams were ever used quantitatively. They were mainly used to derive a formula, and the scale of the vectors was immaterial. As regards lag, he was not aware of any definite knowledge on this point, but the fact that nowadays waves of 30,000,000 per second were obtained showed that the lag must be very small. He thought it was justifiable to assume it zero in slow circuits. The conception of the valve as a voltage amplifier seemed a difficulty; but if we vary the resistance of one part of a circuit containing a fixed E.M.F it is clear that the P.D. on the remainder of the circuit will also vary; so that the varying resistance can be regarded as a source of varying E.M.F. applied to the remainder of the circuit. The quantity ha was not constant. As regards Capt. Turner's point, he was referring here to a case in which there was no resistance in the inductance circuit. A paper entitled " A Small Direct-current Motor Using Thermionic Tubes instead of Sliding Contacts," by Prof. W. H. ECCLES and Mr. F. W. JORDAN, was read by the latter. iron teeth on its periphery, and the stationary part comIn this motor the rotating part is an ebonite disc with prises two electromagnets with their poles close to two teeth. One electromagnet is connected to the grid of a thermionic relay, the other is included in the plate circuit. When during rotation a tooth passing the grid magnet induces a voltage in its winding the consequent transient increase of current through the other magnet causes this magnet to exert a pull on the tooth approaching it. We thus have a small motor without commutator or spark which may under no-load be driven up to a speed of 4000 to 6000 revs. per min. from the lighting supply. Ordinary Meeting, February 28, 1919. Prof. C. H. LEES, President, in the Chair. A PAPER "On Simplified Inductance Calculations, with Special Reference to Thick Coils," by Mr. P. R. COURSEY, was read, in the absence of the author, by Mr. J. NICOL. The method of calculation advocated in the paper is layer coils, to include as well all ordinary forms of thick based on an extension of Nagaoka's formula for single coils. Rosa's formula for thick coils is put into the same form as Nagaoka's, and its use enables a series of correction factors to be calculated for various coil thicknesses. By the aid of a single sheet of curves giving values of these correction factors the inductance of any form of coil likely to be met with in practice may be readily calculated, using only one simple standard formula for all cases. Reasonable accuracy is obtained even in the limiting example of a single turn of wire. The results arrived at agree well with other published charts, which are usually of more limited application, while the use of a single formula for all cases lessens the liability to error. It is also shown that Raleigh and Niven's formula enables the calculation of the correction factors for very short coils to be carried out without having recourse to Nagaoka's more complicated expressions. DISCUSSION. Dr. ECCLES said that as the author had mentioned an abac given in the speaker's book on "Wireless Telegraphy," it was well to remark, first, that this abac was CHEMICAL NEWS, Demonstration of a New Polariser. deliberately made of its present range so as to provide an open scale for coils of the proportions used in practice. For the rare cases of coils outside the range of the abac very simple formulæ are available. Dr. Russell's formulæ were used in making the abac, because they seemed more convenient for the purpose than the results of Nagaoka, to which they were, of course, equivalent. By constructing new curves for dealing with coils of several layers the author has rendered a great service, and bas opened easy paths though a forest of laborious theoretical calculation. Now that this has been done it will be much easier to compare the measured values obtained on actual coils with the calculated values, and so we shall be aided to accumu late experience of the effects cf high frequency eddy currents on the inductance of coils. Prof. Howe agreed with Prof. Eccles' remark about the general utility of Mr. Coursey's investigation. The chief differences between the author's curves and those of the Bureau of Standards was that in the latter the ratio D/l and k, which was a function of D/l, were combined into a single function. 167 The PRESIDENT asked if there was much variation in the ratio D/d, and said he thought the formula was important if it could be verified over a wide range. He referred to Lord Rayleigh's experiments on flasks as closely analogous to those of Dr. Dunstan. In reply, Dr. DUNSTAN stated that the range of D/d In the course of experiments with polarisers built of piles of glass plates disadvantages due to bulkiness of the apparatus and loss of light had to be overcome. The idea occurred to him to place the pile of plates between two prisms of the same glass in such a manner as to (a) Reduce the length of the polariser by one-half; Results obtained with experimental prisms he showed were so good that they could be considered a very fair Mr. NICOL, in reply, referred to formulæ published in substitute for Nicol prisms of corresponding size, and the very small amount of light escaping through crossed the Electrician by Mr. Coursey a few years ago, and considered that the step now taken, which resulted in restrict-prisms (which could be reduced further by additional plates) is for most purposes negligible. ing the values of k to between o and 1, combined with the general inclination of the curves to about 45 deg. with the axis, gave much greater convenience in reading. A Demonstration of Some Acoustic Experiments in Connection with Whistles and Flutes was given by Dr. R. DUNSTAN. Experiments were made with hollow spheres, cylinders, and cones with holes of various sizes and in various positions. Bernouli's theorem, which gives the wave-length of the sound produced by a cylindrical pipe in terms of the length of the pipe and an end-correction depending on the diameter only, was shown to be quite inadequate for practical purposes, the pitch depending on many other factors, such as the wind pressure, the size and shape of the blow-hole, &c. Cylindrical flutes appear to require an end-correction which-within certain limits-is equal to D2/d, where D is the diameter of the pipe and d the mean diameter of the mouth hole (which is often oval in shape). In the shortest flute experimented with, which was only half an inch long, Bernouli's theorem would give the wave-length as two inches, whereas it was actually fourteen inches. The conclusions drawn from the experiments are that in blowing across a hole in a hollow body a force existed on an elastic substance. The result is a "spring back," which produces an aerial throb, puff, or pulsation. The frequency of the pulsation is determined by relations between the dimensions of the instrument, the size of the hole, the wind pressure, &c. Any resulting sound has its wave-length determined by the frequency and not primarily by the dimensions of the instrument, as in the usual textbook treatment. DISCUSSION. Prof. BRAGG referred to Lord Rayleigh's work on open pipes and flasks, and said that since the internal pressure of the air would vary with the size of the aperture it was easily conceivable that a difference of pitch would result when a change was made in the size of the aperture. Mr. F. J. W. WHIPPLE outlined Lord Rayleigh's explanation of the long wave-length of the notes obtained from a hollow sphere. Mr. T. SMITH asked if the variation in pitch with the position of the hole along the cylinder had been fitted to a formula-for instance, was the change in pitch proportional to the square of the distance of the hole from the centre of the cylinder ? Mr. NICOL spoke of the end correction, and stated that students often applied the correction to the closed end of pipes, which was incorrect. Mr. F. E. SMITH said the end correction was useful in electrical problems, and gave an illustration of the variation of D2/d in an electrical case. There would be no difficulty in building such polarisers to any required size, as all the material consisted entirely of glass in unlimited quantities and at reasonable price, and it was hoped that this invention (Brit. Patent 121,906) would be used for many purposes. Polarisers for directly transmitted light were hitherto very scarce and costly, so that many uses they could be put to remained undeveloped (such as stereoscopic kinematography). The new polariser could also be adapted to advantage in microscopes, saccharimeters, optical pyrometers, &c., and used for optical benches in the lecture room. Experiments with piles of glass plates showed a very large discrepancy between the calculated and observed angle for best extinction. Whatever the glass used, and whatever the quality of the surface, this discrepancy came consistently to some 10 deg., whereas thin microscope cover plates were found to be useless. There seemed to be still an interesting field for investigation as to the conditions affecting the surface of glass plates used in polarisers. DISCUSSION. Mr. T. SMITH said it was difficult to discuss the merits of various forms of polariser, as many forms had been proposed and much was of a confidential character. He had been struck by the presence of two opposite tendencies in constructing such apparatus. In one group the optical portions of the apparatus were made large, and efforts were directed to increasing the size; in the other group, where the attainment of the same ultimate object was in view, these parts were made as small as possible. The PRESIDENT asked if it was not possible to get equally good results without prisms by reflection only. He thought prisms made the arrangement complicated. Mr. BRODSKY, in reply, explained that he agreed with the remarks about reflection, but that his efforts were confined to transmission effects for a particular purpose. SOCIETY OF PUBLIC ANALYSTS AND Dr. SAMUEL RIDEAL, President, in the Chair. A CERTIFICATE was read for the first time in favour of Certificates were read for the second time in favour of |