CHEMICAL NEWS, Jan. 5, 1912 Experimental Investigation of the Bleaching Process. 9 321. "Note on Methyl-n-tridecyl- and Methyl-n-penta- | hydrogen phthalate separates from light petroleum as a decyl-carbinols and the corresponding Ketones." By ROBERT mass of needles, which melt at 50-51°. The hydrogen HOWSON PICKARD and JOSEPH Kenyon. succinate separates from light petroleum in a microcrystalline condition, and melts at 64-65°. The brucine salts of each of the esters were prepared and re-crystallised several times, but no ester could be obtained which exhibited any trace of optical activity. In a recent communication (Trans., 1911, xcix., 45) the authors have described the resolution into their optically active components of the members of the series of carbinols from methylethylcarbinol to methyl-n-undecylcarbinol. It was intended to continue the series so as to include methyln-tridecyl- and methyl-n-pentadecylcarbinols, but it has been found that the method of bringing about the resolution of the various members of the series fails in the case of these carbinols, containing fifteen and seventeen carbon atoms respectively. This note contains a short description of the compounds isolated during the course of the experiments, and also shows that the catalytic method with thorium oxide (Senderens, Comptes Rendus, 1909, cxlv., 995) can be used for the formation of ketones, even when they contain many carbon atoms. Methyl-n-tridecyl Ketone, CH3CO C13H27.-Myristic acid (300 grms.) dissolved in glacial acetic acid (2400 grms.) was passed at an hourly rate of 50 cc. through a Jena-glass tube containing thorium oxide, and heated to 400°. The products of the reaction were condensed, dissolved in ether, and washed with a solution of sodium hydroxide until free from acid. Myristic acid (125 grms.) was recovered from the washings, mixed afresh with glacial acetic acid, and again treated. When the products from the two operations were fractionally distilled, 200 grms. of methyl-ntridecyl ketone were obtained. The residue in the flask when re-crystallised from ethyl alcohol yielded 30 grms. of myristone, which melted at 75-76°. Methyl-n-tridecyl ketone boils at 184°/20 mm., and on cooling sets to a soft mass of crystals, which melt at 37°. It readily forms the semicarbazone, which crystallises from ethyl alcohol in needles melting at 126'5°. (Found, N = 150. Calc., N = 14.8 per cent). Methyl n tridecylcarbinol, CH3 CH(OH)·C13H27, is readily obtained when the corresponding ketone is reduced by sodium in 96 per cent alcoholic solution, little, if any, pinacone being formed. It boils at 181-1830/19 mm., and on cooling sets to a crystalline mass, which melts at 38-39°. The hydrogen bhthalate crystallises from light petroleum in pearly leaflets, and melts at 61-62°. (The purity of each of the acid esters described in this note was ascer. tained by titration of their alcoholic solutions with a standard solution of sodium hydroxide). The brucine salt crystallises from acetone in needles, which, when dried in the air, melt at 80-85°, but after desiccation in a vacuum or after heating at 105° for a few minutes they melt at 172-175°. After two re-crystallisatious, on polarimetric examination : 10136, made up to 20 cc., with ethyl alcohol, gave a -6.25°, whence [a] - 61.66°. The strychnine salt crystallises from chloroform and acetone in hair-like needles, melts about 130°, and has [a]D-52.6° when dissolved in chloroform. The hydrogen phthalate obtained from the purest specimen of these salts was in each case found to be quite inactive when examined in the polarimerer. The hydrogen succinate crystallises from light petroleum in nacreous leaflets, which melt at 60-61°. The brucine salt is a wax, and unsuitable for fractional crystallisation. The corresponding derivatives of heptadecane were obtained by similar methods. Seventy-nine grms. of methyl-n-pentadecyl ketone, were obtained from 112 grms. of palmitic acid. It boils at The 322. "An Experimental Investigation of the Bleaching Process." By SYDNEY HERBERT HIGGINS. Solutions of the hypochlorites of lithium, sodium, and potassium prepared from bleaching powder solution by precipitation with solutions of the alkali sulphates or carbonates have identical bleaching properties; the bleaching efficiencies of these solutions are almost equal to that of the original bleaching powder solution. Curves showing (1) the evolution of oxygen from hypochlorite solutions and from solutions of sodium peroxide in contact with copper oxide; (2) the rate of bleaching of solutions of hypochlorites and of sodium peroxide; (3) the rate of oxidation of oxalic acid by potassium permanganate and the rate of liberation of iodine from potassium iodide by acidified sodium peroxide solution (Harcourt and Esson, Phil. Trans., 1866, clvi., 193), point to the conclusion that the actions in all these cases are similar, that is, unimolecular. Hypochlorites therefore bleach because of their readiness directly to produce nascent oxygen. Those substances which accelerate the evolution of oxygen from solutions of hypochlorites or of peroxides also assist the evolution of that gas when they are heated along with potassium chlorate; hence it is taken that the main action in the latter case is a unimolecular one. Chlorine water is a much weaker bleaching agent than solutions of hypochlorites; therefore the bleaching action of the latter cannot be due to the generation of free chlorine as stated by Taylor (Trans., 1910, xcvii., 2541). Chlorine water probably owes its bleaching properties to the presence of hypochlorous acid in solution. The proof that the production of chlorine from bleaching powder solution and the stimulation of the bleaching action of that solution have no connection with one another has led the author to consider the two questions separately. In pursuit of this line of argument he gave further experiments supporting previous conclusions (compare Taylor, Proc. Chem. Soc., xxvii., 243). 323. "The Direct Esterification of Saturated and Unsaturated Acids." By EBENEZER REES THOMAS and JOHN JOSEPH SUDBOROUGH. The esterification of a number of saturated and unsaturated organic acids has been studied by heating the acid with a large excess of pure ethyl alcohol at 100° in sealed tubes for given periods, and titrating the amount of free acid left. The results show that an aß-olefine linking has a retarding effect, except when the unsaturated acid is a very much stronger acid than its saturated analogue, for example, methyl hydrogen maleate, which is some hundred times as strong an acid as methyl hydrogen succinate. In such a case the olefinic acid is esterified more rapidly than the corresponding saturated acid. two A By-unsaturated acid is usually esterified somewhat more rapidly than its saturated analogue, and adolefinic linking appears to have but little effect on the rate of The cyclopropane derivatives were tanacetone and its derivatives, esters of carbonic acids, a new hydrocarbon, namely, 12-dimethylcyclopropane, and some others. The cyclobutane compounds were chiefly esters of various acids. It was shown that (1) the three- as well as the fourcarbon ring exerts an influence on the molecular refraction. The increment in the case of a three-carbon ring is for Ma + 0.67 and for MD +071; in the case of a four-carbon ring for M. +0:45 and for My + 0'48. (2) A small exaltation is noticeable when the three carbon ring is conjugated with a double bond (ED + 0.2). (3) The three- and the four-carbon rings have practically no influence on the molecular dispersion. (4) The conjugation of a three-carbon ring and a double bond increases the molecular dispersion about 10 per cent. The magnetic rotation for a saturated tricyclic terpene, 3-pinolene, has also been determined. 325. "Contributions to our Knowledge of Semicarbazones." (Preliminary Note). By ISIDOR MORRIS HEILBRON and FORSYTH JAMES WILSON. In the course of some recent experiments the authors observed that m-nitrobenzylidenedeoxybenzoin on treatment with semicarbazide acetate formed two distinct semicarbazones, a colourless and a yellow modification. The addition of sodium hydroxide solution to an alcoholic solution of either the colourless or the yellow form produced an intense yellow coloration, which was destroyed by acids As semicarbazones have not as yet been very fully studied, the authors are extending their observations to semicarbazones derived from both saturated and unsaturated ketones and aldehydes. Phenyl styryl ketone gives on direct treatment with semicarbazide acetate two distinct semicarbazones: a colourless form, melting at 168°, which is rapidly precipitated from a dilute alcoholic solution, and a mere soluble yellow form obtained by dilution of the mother-liquors. On attempting to re-crystallise this yellow modification from chloroform, the authors found that it passes into a third modification which crystallises in colourless silky needles melting at 180°. This semicarbazone in the dry state is unstable, being gradually re-converted on exposure to diffused sunlight into the yellow modification. No melting-point for this yellow form can be obtained, as it passes when heated above 160° into the colourless modification melting at 180°. Alcoholic solutions of both colourless semicarbazones give with sodium hydroxide intensely yellow products, which are still under investi gation. on the subject. When o-thiol- or o-dithio-benzoic acid is treated with malonic acid or its ester in the presence of warm concentrated sulphuric acid a vigorous reaction' occurs, and "thioindigo" is formed. From preliminary experiments the yield of the latter substance appears to be as high as about 85 per cent of the theoretical, exclusive of thioindigosulphonic acid which is simultaneously produced. Traces of hydroxybenzothiophen were also isolated. There is no doubt that the preliminary stage of the interaction depends on the formation of the sulphoxylic acid (Trans., 1911, xcix., 640), which then reacts as follows:CO2H C6H4SO+CH2(CO2Et)2= CO = C6H4<<_>C(CO2Et)2+2H2O. Condensation of these thio-acids with ethyl acetoacetate, substituted malonic acids, and similar substances may also be effected; these interactions are now being more closely investigated. 328. "Asymmetry in the supposed Absence of an Asymmetric Atom." By JAMES ERNEST MARSH. Everest's paper (Proc. Chem. Soc., xxvii., 285) disposes effectively of the claim that a new type of asymmetry exists which is not dependent on the presence of an asymmetric atom in the molecule. At the same time, the views of Everest involve him in a difficulty of another kind; in the conversion of the unsaturated acid into its dibromide, the asymmetric carbon atom (1) is represented as losing its asymmetry, while a new asymmetry carbon atom (7) is created in a different part of the molecule outside the ringCH3 CO2H 66 CH2 CH2 C=C CBr-C (4) (7) Br H (1) CH2 CH2 Perkin, Pope, and Wallach, in stating (Proc. 1909, xxv., 83) that the constitutional formula of the unsaturated acid comprises no atom which can be described as asym metric, no matter which definition of an asymmetric atom be adopted," have overlooked the definition given by the author (Phil. Mag., 1888, [5], xxvi., 426). He there defined an asymmetric carbon atom as one which, when replaced by its image, gives a formula different from the original formula. This idea involved the consideration of an atom isolated from the rest of the molecule, and was justified in view of van't Hoff's theory ("Dix Années," &c., 1887, pp. 27 and 28) that the tetrahedron the centre of which is occupied by the carbon atom acquires an asymmetric form owing to the effect of four different affinities on it; thus the asymmetrically deformed atom can be The absorption spectra of these various semicarbazones are being investigated in the hope of throwing further light It has also been found that semicarbazones of both saturated and unsaturated aldehydes and ketones are affected by ultra-violet light, stereoisomerides being apparently produced in some cases, decomposition products considered apart from the particular groups to which it is in others. Investigations in this direction are also in progress. 326. "Conductivity and Dissociation of Diacetyltartaric Acid." By STELLA DEAKIN and ALBERT CHERBURY DAVID RIVETT. Diacetyltartaric anhydride is hydrated too rapidly for the course of the action to be followed by the conductivity method (compare Rivett and Sidgwick, Trans., 1910, xciii., 732. 1677). Since some connection appears to exist in such a case between rate of hydration and first dissociation constant of the acid, a determination of the latter was desirable, and has been made, giving the value 2.5 × 10-2, a higher figure than is given by the acid corresponding with any anhydride the rate of hydration of which has been measured. The second dissociation constant ap proximates to O'II X 10-2. attached. The groups may be four or less, according as the atom is situated in open-chain or in one or more rings. The atom being asymmetric will have a non superposable image. The replacement of the atom by its image was represented in the author's paper as brought about by an exchange of position of two of the groups attached to the atom. The same inversion may be done by flyping the tetrahedron (that is, by turning it inside out), a process applicable to the stereocentric benzene formulæ, and one which affords a simple explanation of the phenomenon called the Walden inversion. In this method of treatment there is no need for a kind of dissection of a formula, as employed by Everest to prove four different groups, a process which must become difficult when there are several asymmetric atoms in the molecule. The views expressed in the author's paper lead to the conclusion that the methylcyclohexylideneacetic acid has three asymmetric carbon atoms in the molecule (1), (4), and (7), and that all three remain asymmetric in the bromine additive compound. CHEMICAL NEWS, } Latent Heats of Vaporisation of Mixed Metals. The exchange of position of any two groups or linkings | oxyacetates the allyl has slightly the higher rotation at attached to C (1) in either acid gives a formula different low temperatures and the propyl compound at higher from the original one, being, in fact, that of the optical temperatures. isomeride. The case is the same with C (4) and C (7). The unsaturated acid is certainly a compound of a novel type, in that with three asymmetric carbon atoms there are only two possible isomerides, and those enantiomorphous. A somewhat similar example of a limited isomerism is that of the trihydroxyglutaric acids, where also with three asymmetric carbon atoms there are only two isomerides which are here optically inactive (without including the active forms which have only two asymmetric atoms). Perkin and Pope (Trans., 1911, xcix., 1510) claim that in the methylcyclohexylideneacetic acid" the optical activity is due to the enantiomorphous configuration of the molecule as a whole." This is certainly true, not only of this acid, but of every optically active substance whatever, and the author stated it in the paper already mentioned. The converse has also proved true that unless the molecule as a whole is asymmetric it will be optically inactive, however many asymmetric atoms it may contain. The term "centroasymmetry" is therefore not only unnecessary, but misleading; and van't Hoff's original proposition with regard to carbon still holds good, that every optically active compound has an asymmetric atom in the molecule, although the presence of asymmetric carbon does not necessarily involve optical activity. 329. "Latent Heats of Vaporisation of Mixed Liquids." (Part II.). By DAN TYRER." In continuation of previous investigations (Trans., 1911, xcix., 1633) it was pointed out that a mixture of two liquids has really two latent heats, namely:-1. Heat absorbed when I grm. of the mixture vaporises at constant temperature and pressure from an infinitely large quantity of the mixture of known composition. It is supposed that no change in the composition occurs in this process. 2. Heat absorbed when I grm. of a mixture is completely vaporised at constant temperature. If x is the composition of the liquid in (1) and y the composition of the vapour which leaves it, and if y is also the composition of the I gim. of liquid in (2), then it can be easily shown that (Lp)x = (Lc)y+H, where Lp and Lc are the two latent heats respectively defined above, and H is the heat developed when I grm. of the mixture of composition y is added to an infinitely large quantity of the liquid of composition x. Values of Lp for several mixtures have been experimentally determined, and also the compositions y of the vapour boiling off from mixtures of known compositions, and from this data values of Lc have been calculated. It was shown that whilst the results for Lp do not appear to bear any simple relation to the compositions of the mixtures, those for Lc follow Trouton's equation In the case of the allyl- and propyl-amides of tartaric and malic acids previously investigated by one of the authors (Trans., 1906, lxxxix., 1854 and 1861), the differences in rotation were much more pronounced, and in both instances the propyl compound had the higher rotation, thus:In pyridine solution. [M] 20° (tartaric). Allylamide 1. +251° [M] D 20 (malic). These greater differences in rotation in the case of the tartaric and malic compounds are probably due to there only being three intervening atoms between the unsaturated and the asymmetric carbon atoms, whilst in the above menthyl compounds there are in all cases five intervening atoms between the unsaturated carbon atom of the allyl and the nearest asymmetric carbon of the menthyl group. 331. "The Action of Aliphatic Amines on s-Dibromosuccinic Acid." (Part I.). By EDWARD PERCY FRANKLAND and HENRY EDGAR SMITH. The authors have prepared the dipropylamino- and dibutylaminosuccinic acids, R-NH-CH-CO2H' by the action of n-propylamine and n-butylamine respectively on s-dibromosuccinic acid in alcoholic solution. The reaction appears to proceed in a manner analogous to that described by E. P. Frankland (Trans., 1911, xcix., 1775) in the case of benzylamine and s-dibromosuccinic acid. The alkylamino-acids yield copper salts, hydrochlorides, and dinitroso-derivatives. 332. "The Velocity of the Reaction between Iodic and Sulphurous Acids in various Media." By THOMAS STEWART PATTERSON and WILLIAM COLLINS FORSYTH. The influence of methyl alcohol, ethyl alcohol, n-propyl alcohol, and of acetone on the velocity of the reaction between iodic acid and sulphurous acid was described. NOTICES OF BOOKS. Laboratoriumsbuch fur die Anorganische Grossindustrie. ("Laboratory Book for the Inorganic Major Industries "). By Dr. C. v. HOHORST. Halle-a.-S.: Wilhelm Knapp. 1911. (5.60 Mk.). THIS laboratory manual is intended more particularly for the guidance of chemists who are entering upon their pronology, during which they have gained a certain amount fessional life after a course of training in chemical techof proficiency in the usual analytical operations, and have in addition made a special study of some particular branch. It not infrequently happens that they find the equipment of the works laboratory very different from that of the laboratory belonging to the technical college, and a considerable amount of time and energy is often wasted in getting used to the altered conditions of work. The book is not to be regarded as a concise survey of all the methods which have been suggested for analysing the substances dealt with, but rather as a guide to those which, having the greatest claims to accuracy, rapidity, and simplicity, are practically universally adopted. It gives succinct directions without any superfluous detail for the carrying out of the analyses of fuels, boiler waters, and the raw material and finished products of the acid, salt, chlorine, and ammonium industries. All the tables used in these processes are included, and a chapter on the analysis of artificial manures by Dr. M. Rosenberg has been added. Who's Who Year Book by the same publishers, is intended chiefly as a supplement to the larger volume, and is, in fact, made up of the tables formerly in that book itself. In addition will be found some original tables not to be met with elsewhere. such as Race Meetings, London Theatres with their Lessees and Managers, Professors of Universities, &c. Price Is. net. The Writer's and Artist's Year Book.-This book has been devised to meet the needs of authors and contributors to journals to enable them to decide where to place their stories and articles. The book is written for those who wish to make money by their writing, and in most cases some indication is given of the fees paid by publishers for contributions. The Englishwoman's Year Book and Directory.-The work is divided into two parts. The first treats of Education, Professions, and Social Life; and the second Philanthropic effort. It is edited by G. E. Mitton, with the assistance of an Honorary Committee of prominent women of the day, and contains a mass of valuable information relative to women's professions, trades, sports, pastimes, and philanthropies. Special attention has been given to the position of women in Municipal, Industrial, and Social Life. The book can be studied with profit by every woman. Red cloth, 400 pages, price 2s. 6d. net. and does not melt until a much higher temperature is reached. It is the most stable of all the bismuth-sodium compounds. It is soluble in excess of liquid bismuth, and causes a lowering of the melting point when it dissolves; the maximum is reached with a melt of composition Na3 Bi+5Bi. The author has applied the same method to many other elements which have a melting-point below the boiling-point of paraffin. These include K, Na, Rb, Cs, Li, Cd, Hg. Ga, In, TI, Sn, Pb, Bi, and the metalloids P, Se, Te. The bismuthides of the alkali metals are readily oxidised in air, being converted into a black powder which consists of the alkali and bismuth suboxide. They are decomposed by water. With hydrogen they yield metallic hydrides. Zeitschrift für Anorganische Chemie. Tantalum and Niobium Pentafluorides. - Otto Ruff and Emil Schiller.-Large quantities of the pentafluorides of tantalum and niobium can be prepared by the action of anhydrous hydrofluoric acid on the pentachlorides, which may be obtained by heating the oxides in a current of chlorine and carbon tetrachloride. Tantalum pentafluoride forms colourless double refracting crystals of melting-point 96.8°. It is, generally speaking, stable towards non-metals, but is attacked by metals. With water it yields the pentoxide. The melting-point of niobium pentafluoride is 75'5°. In its reactions it resembles the tantalum salt, but it does not react so readily. Niobic and tantalic acids can be separated by dissolving in hydrofluoric acid, adding excess of KI, and subjecting the mixture to fractional crystallisation. To separate tantalic and niobic acids from titanic acid Knop's method is the best; i.e., the conversion of the oxides into chlorides and the fractional sublimation of the latter in vacuo. Decomposition of Alkali Double Sulphates of CHEMICAL NOTICES FROM FOREIGN Cerium Earths by Fusion with Carbon.-Philip E. SOURCES. Norr. All degrees of temperature are Centigrade unless otherwise expressed Browning and Philip L. Blumenthal.-When the double sulphates of the alkalis and the cerium earths are fused with wood charcoal, they are readily decomposed, and are converted into a form which dissolves in hydrochloric acid with the formation of soluble chlorides. The reduction of the double sulphates to sulphides by carbon is not com chloric acid. Comptes Rendus Hebdomadaires des Séances de l'Académie plete although the product is readily attacked by hydro- Condensation of Acid Amines in presence of Glycerin.-L. C. Maillard.-When a mixture of glycocoll with four or five times its weight of glycerin is kept for some hours at 170°, most of the glycocoll is transformed into cycloglycylglycine. As an intermediate product the tetrapeptide triglycylglycine is formed. By a secondary reaction triglycylglycine yields the hexapeptide, pentaglycylglycine. The reaction is general, and can be applied in the synthesis of mixed cycloglycylglycines. Probably a glyceric ether is first formed, and is then destroyed with regeneration of the alcohol; thus the reaction is a generalisation of that of Th. Curtius. Berichte der Deutschen Chemischen Gesellschaft. Bismuthides and Compounds of Metals with One Another.-A. C. Vournasos.-When bismuth and a metal are melted separately under paraffin, and one liquid is poured into the other, a crystalline precipitate is obtained. If sodium is used the precipitate is Na3Bi; it is insoluble in excess of sodium at the temperature of boiling paraffin, MEETINGS FOR THE WEEK. MONDAY, 8th.-Society of Chemical Industry, 8. "Production of Formic and Acetic Acids by the Atmospheric Oxidation of Turpentine," by C. T. Kingzett and R. C. Woodcock. "Rapid Volumetric Method for the Determination of Free Sulphur," by C. Davis and J. L. Foucar. "Relative Adsorption of Dyes by Sand and Natural Fibres," by W. P. Dreaper and W. A. Davis. "Ingrain Dyeing-Influence of certain Groups on the Re-solution Factor," by W. P. Dreaper. TUESDAY, 9th.-Royal Institution, 3. (Christmas Lectures, adapted to a Juvenile Auditory). "The Childhood of Animals," by P. Chalmers Mitchell, LL.D., D.Sc., F.R.S. Society of Dyers and Colourists, 8. "Some Problems in Garment Dyeing," by F. G. Newbury. "Aluminium in the Service of Chemical Industry," by R. Seligman. WEDNESDAY, 10th.-Royal Society of Arts, 5. (Juvenile Lectures). "Soap Bubbles," by Prof. C. V. Boys, F.R.S. THURDAY, 11th.-Royal Society. "Propagation of Waves through a Stratified Medium, with Special Reference to the Question of Reflection," by Lord Rayleigh. "Variation of the Specific Heat of Water, with Experiments by a New Method," by H. L. Callendar. "Mechanism of the Semi-permeable Membrane, and a New Method of Determining Osmotic Pressure," by F. T. Trouton. "Mobility of the Positive and Negative Ions in Gases at High Pressures," by A. L. Kovarik. "A New Method of Determining the Radiation Constant," by G. A. Shakespear. "Mechanics of the Water Molecule," by R. A. Houston. I CH ON T By W Fig. repr CHEMICAL NEWS,} Jan. 12, 1912 Thermo-regulators. THE CHEMICAL NEWS. radiator, worked direct from the power mains carrying an VOL. CV., No. 2720. NOTE ON TWO THERMO - REGULATORS.* By W. R. BOUSFIELD, M.A.. K.C. THE first Fig. represents a regulator specially designed for an air thermostat, in which a battery of standard cadmium cells was kept. The principle is equally available for a A 13 The source of heat to be regulated was an electric alternating current at 240 volts. A B are the upper and lower surfaces of a column of mercury contained in a barometer tube. Part of this mercury column is in series with the radiators and the current from the main, which passes into and out of the mercury by platinum wires, I mm. thick, sealed in the glass walls of the tube. The current enters through the wire c, which is placed sufficiently far below the surface A to be always immersed. The circuit is closed by contact between the lower platinum wire D and the lower surface B of the mercury. The barometer tube communicates with the bulb E, which is filled with hydrogen through a cock, F. A cock, С, serves to regulate the quantity of mercury for the desired temperature, and enables the bulb E to be thoroughly washed out with hydrogen gas. Overheating of the thermostat expands the gas, and causes the mercury surface, B, to be FIG. 2. forced away from the platinum wire, thus cutting off the current. The reverse process takes place when the temperature falls below the mean. The tube is furnished with two bulbs at A and H. The mean pressure in E must be so chosen that the mercury in A normally occupies about half the bulb A. The purpose of the bulb H is to supply sufficient mercury to fill the bulb A in case of overheating, without allowing the lower level of the mercury to be depressed to the bottom of the tube, which would permit air-bubbles to pass and spoil the vacuum in A. It is best to instal the regulator with the mean pressure in E about equal to mean atmospheric pressure. The length of the tube will then be that of an ordinary barometer. In this way the possible leakage of the mobile hydrogen through defective taps is prevented. In such an apparatus, where the full current through the water thermostat, though slight changes of detail would radiators from a pressure of 240 volts in the mains passes be necessary. * Exhibited before the Faraday Society, December 6, 1911. through the mercury, and gives rise to a spark at в on breaking the contact, it was a matter of vital importance to |