NEWS ROYAL & CIVIL ENGINEERING, UNIVS., &c. Mr. J. HAWKSWORTH COLLINS, B.A. (Hons, Camb. and Lond.), late Army Form Master of Cranbrook School and Eltham College), PREPARES BOYS for above. Besides the usual Chemical, Physical, and Carpentering Laboratories, there is one for Instructive Amusement, containing Gas, Steam, and Electric Engines, Wireless Telegraph, X-ray Apparatus, Dynamos, &c. OERTLING'S M BALANCES IN PLATINUM Utensils. We supply all forms and sizes of Platinum Utensils and Apparatus for Chemical and Physical purposes. All our utensils are hammered to shape, tested, and finished in the best manner. All kinds of Platinum scrap bought for cash or taken in exchange for new. - DERBY and CO., Ltd., 44, Clerkenwell Road, London, E.C. EDUCATIONAL AND SCIENTIFIC BOOKS MATHEMATICAL, THEOLOGICAL, AND FOREIGN BOOKS. J. POOLE & CO. (Established 1854), (Late of 39, HOLYWELL STREET, STRAND). Write to us for Quotations for any Books required of above description. SULPHUROUS ACID and SULPHITES. Liquid SO2 in Syphons, for Lectures, &c. PHOSPHORIC ACID and PHOSPHATES. CARAMELS & COLORINGS for all purposes. A. BOAKE, ROBERTS, & CO. (LIMITED), Stratford, London, E, The amount of knowledge now required for a boy to be successful in the above professions is very great, and can only be acquired by his taking a regular course from an early age. At the age of sixteen he must be ready to start Differential Calculus, and in Science he must have had large facilities for doing practical work. Only six boys received, ages twelve to eighteen. 128, OLIVE ROAD, CRICKLEWOOD, N.W. Reasons: (1). The observed atomic weights of Au, Ag, and Zr are 1972, 107.88, and 906; and, since Au and Ag are perissads and Zr an artiad; also, since the atomic weights of elements in general are too large (see Chem. News, Oct. 11, 1907), therefore the values are 197, 107, and 90. (2). Ag is a monad, Zr a tetrad, and therefore Au is a triad. (3). Au can act as a non-metal, owing to the free -H-H3 - (see Deduction C, Chem. News, Oct. 11, 1907). (4). It is especially found with Ag, Zr, and Cu. (5). Au and Ag are isomorphous. (6). Au is monobasic and diacidic. (7). Besides the pair Au and Ag, there are nine other pairs (the two elements of each being specially connected with one another) whose difference is approximately go, and the average of these differences is 89'97. These pairs are-Ir, Rh; La, Ti; Nd, Mn; Os, Ru; Pt, Pd; Ra, Ba; Th, Ce; Sn, Si; W, Cb. (8). It is much more likely that the difference between the two ele ments of each pair is exactly the same in every case, instead of being nearly the same, since the two elements of each pair are always found together and in close connection with Zr. (9). The relative volume of an atom of Au is 10:25, the same as that of an atom of Ag. (10). It is evident that we are dealing with the same increment in other cases, for besides the atomic volumes of Au and Ag being the same, the following elements have the same relative atomic volume: Ir, Rh; Os, Ru; Pt, Pd; viz., 9'05-so that the increment, whatever it is, produces in these cases also no addition to the atomic volume. It follows from the above that Ir=ZrRh, La ZrTi, Nd=ZrMn ZrSi, and Os = ZrRu, Pt= ZrPd, Ra=ZrBa, Th = ZrCe, Sn = W=ZrCb. These constitutions can be supported by lines of reasoning similar to those given for Gold. THERE HAWKSWORTH COLLINS. ARE CHEMICAL NEWS, } May 13, 1910 Fruit of Cornacea stolonifera. 217 THE CHEMICAL NEWS. 38.5 per cent of the dried fruits. A gummy substance in The sugar as reduced by Fehling's solution indicated large quantity collected in chunks on the sides and bottom of the flask containing the alcoholic extract. The specific gravity of this substance must be nearly equal to water, for the masses settled to the bottom, whereas small particles remained floating in the water. The specific gravity of gums varies from 1350 to 1490 (Allen), and this would rather indicate that the substance is not a gum, but possibly dextrine. The specific gravity was not determined, as there was no certainty of its degree of purity. There was the amount of this gummy substance. The method finally some difficulty experienced in filtering and in determining employed was to filter through a fine cloth. It was found that hot alcohol could easily remove the substance from the sides of vessels. We found that this amounted to 11.5 per cent of the dried fruit. According to Allen's "Commercial Organic Analysis" dextrine is precipitated as a gummy mass in presence of maltose by alcohol, and that an important difference between gums and dextrine is that in the case of gums mucic acid is formed by treatment with concentrated nitric acid. This substance was so treated, but no crimson colour was obtained, and therefore no mucic acid was formed, and no gum could have been present. Gums are insoluble in alcohol, but the substance indicate that the substance could not have been a gum, in question was soluble. These facts would seem to although the amount obtained seems rather large for dextrine. Fehling's solution was not reduced by this substance, which is additional evidence in favour of dextrine. The silica resembled grains of sand. The carbon dioxide was determined by the Bunsen method with o'4 of a grm. of ash, but no difficulty was experienced. The other determinations were made with o'i grm. of ash. So the experimental error may be somewhat large. The attempt was first made to determine the amount of iron volumetrically. A one-tenth normal potassium permanganate solution was used, but it proved impossible to detect the exact turning point. The same difficulty was experienced with N/100 potassium bichromate solution. No trace of manganese or chromium was detected, although these have been observed in other fruits. We placed 200 grms. of the berries in a litre flask, and treated them on the water-bath with alcohol for 224 hours. The extraction was continued with water for 86 hours longer. The extract was removed each day, and a fresh portion of alcohol or water was added. The colour of the alcoholic extract was reddish brown, which became less pronounced as the extraction proceeded. The water extract was black in appearance. The odour of the berries while digesting in alcohol was similar to that of ice-cream, while in the case of the water extract it resembled maple sugar. Weight of sugar Gummy mass 23 13-100*17 This leaves only 0.63 grm. unaccounted for. The alcoholic filtrate was treated several times with bone-black, but the decolorisation was extremely slow. The sub-acetate of lead sulphurous acid method as described in Allen was then used with much better results. It was necessary to evaporate down, and repeat the process several times before the filtrate became clear. The filtrate was then tested for lead, but none was found. On evaporating to a small volume and allowing it to stand for several days, crystals separated out. Two kinds were observed: some were needle-shaped and others rhombohedrous. These crystals gave the test for glucose, left a black residue when heated on a platinum foil, and reacted neutral to litmus; hence it was concluded that they were sugar crystals. After about two weeks, the rhombohedrous began to turn white and to disintegrate, but the needle shapes remained the same. These are characteristics of maltose crystals (Allen). A portion of the purified sugar solution was condensed, and a white cloudy precipitate was formed. Alcohol was added, and the mixture was warmed and filtered. The precipitate slowly dissolved in water. The solution would not reduce Fehling's solution, and when it was heated with concentrated nitric acid oxalic acid was found to be present. These are characteristics of dextrine. It was therefore concluded that dextrine was present, but not in the erythro form, as no change of colour was noted when it was treated with a dilute solution of potassium iodide. The reduction of Fehling's solution showed dextrose to be present. Pavy's test, the dextrine precipitate, and the inability to get an osazone precipitate (Fourn. Am. Chem. Soc., 1906, p. 629), all indicate maltose. The Oils. The dried berries from the sugar extraction were pulverised in a coffee grinder, and the 99 grms. were digested in 185 hours in ether, using an inverted condenser. The extract was removed each day, and a fresh portion of or tannin, but tannin transformed, i.e., the tannic chromogen converted to visible pigment more or less completely. And inasmuch as different species of flowers belonging to the same genus are habitually either red or blue, although precisely the same chromogen is produced by all of them, it seemed certain that it is only a difference in development of the pigment that can account for the difference in tint. ether was added. After purifying with bone-black in an, plasm on deassimilation had eliminated not merely tannoid ether solution the oil weighed 3.5 grms. The specific gravity of the oil, after correcting for temperature, was found to be 965.85. According to Allen the specific gravity of castor-oil is 965.5. The odour of the oil is similar to that of castor-oil. The colour is light green, due no doubt to the presence of chlorophyll. We used 114 grms. for the saponification, according to the method of Koetlstorfer, substituting sulphuric for hydrochloric acid. The saponification equivalent was found to be 349. Castor-oil ranges between 309 and 319. We could not repeat the experiment owing to a lack of material. The Reichert number was found to be 16; the Reichert number of castor-oil, when 2'5 grms. of the oil is used, is given as 14. As the number increases slightly for smaller quantities, it would seem to indicate castor-oil. Oil Residue. The ash of the residue from the oil extraction was found to be 2.15 per cent. Since more than half of the original berries was removed in the form of sugar and oil, it might be expected that the percentage of ash would be more than doubled, whereas there was a decrease. Since there was 3 per cent of ash in the beginning, there must have been 6 grms. of ash in the original 200 grms. of fruit; but 2:15 per cent of 99 grms. is 2.13 grms., which signifies that 3.87 grms. of the ash material was removed by the alcohol and water. Tests were made for tartaric, oxalic, and citric acids, but no trace of them was discovered. I desire to express hearty thanks to Dr. N. Knight for his advice and assistance in this work. Cornell College, April 12, 1910. RESEARCHES ON ANTHOCYAN. By P. Q. KEEGAN, LL.D. Further light was thrown upon the question when in a series of analyses of several whole plants performed by myself (see The Naturalist, 1902 till 1910) it was clearly demonstrated that the depth of tint of the floral organs by no means corresponded to the quantity of tannic chromogen contained in the plant as a whole; that is to say, the corolla may be of a deep blue tint, while the amount of tannoid or tannin in the rest of the plant may be hardly detectable. Hence it was concluded that the formation and development of the anthocyan were strictly and absolutely confined to the corolla or perianth. In this way a close and intimate connection between the physiological activity of the inflorescence and the production of pigment was traced. Processes of particular and special deassimilation brought about in certain cells to the profit of other neighbouring cells, is a phenomenon of general order freThe vital activity quently recognisable in plant-life. incident to the process of fecundation, the formation of the ovules, and the development of the fruit suffice to induce a drain on the proteids of the corolla, where they are not so much required. There was a violent disruption of the proteid molecule of the corolla cell, a separation of its nitrogenous groups, and a relinquishment of its aromatic groups, which were left behind as pigments more or less complete and pure in the epidermal cells of the organ. PROCEEDINGS OF SOCIETIES. ROYAL SOCIETY. Ordinary Meeting, April 28th, 1910. Sir ARCHIBALD GEIKIE, K.C.B., President, in the Chair. "On the Rotatory Character of some Terrestrial Magnetic Disturbances at Greenwich and on their Diurnal Distribution." By ROBERT B. Sangster. NUMEROUS microscopical and chemical researches have been conducted with the view of discovering the origin of the colouring matter of red and blue flowers. It has been successfully traced to the aromatic substances, which are the results of a vigorous process of deassimilation undergone by the protoplasm of the vegetable cell under certain circumstances. The principal of these substances thuswise produced or eliminated is what is somewhat vaguely known as tannin. This tannin contains certain phenolic groups from two up to six in number, and their further The paper commences with an investigation of the oxidation into ketones or ketone-like compounds would changes in direction of the line of total magnetic force at give rise to the blue or red pigment (anthocyan). As the Greenwich on 1903 October 12d. 18h. to 23h. when a conresult of numerous experiments performed by myself (for siderable magnetic disturbance was in evidence. Measuresome of which see Nature, lxi., 105), it would appear ments of the published registers of all three force components certain that the conversion of tannin into pigment varies in were made at equivalent time intervals of about five intensity and completeness very considerably according to minutes, whence is obtained a diagram showing the variathe petal examined. The test of this conclusion has been tion of the force component perpendicular to the line of taken to be the capacity or otherwise of the pigment of total force. The diagram shows there was an almost any flower to form blue compounds with certain bases, wholly rotatory motion of the transverse disturbance vector, manganese especially. It was found that by mixing the the trace consisting of six distinct convolutions varying aqueous solution of the pigment with a few drops of a greatly in size but consistent in anticlockwise progression. saturated solution of succinate or acetate of manganese, Several other disturbances during epoch 1900-07 are exand leaving the liquid to dry in a shallow basin, the amined in detail, and it is shown that a right or leftresulting coloration was deep blue, dark red, or green handed rotatory character in the motion of the disturbance ccording to the anthocyan of the flower experimented vector was of fairly frequent occurrence, while change upon. For instance, those of cranesbill, tufted vetch, and from left to right not uncommonly occurred about midnight. sweet pea dried up blue; poppy, burnet, cineraria, and It was also found that the same direction of rotation often pæony remained red; while clover became green, and fox-persisted for several hours, and tables of the diurnal disglove and carnation showed blue only on the edges of the tribution of right- and left-hand rotatory disturbance are basin. These results, considered in conjunction with those furnished to show that those of right-hand character were obtained by the application of various other tests, led in- entirely absent during the hours 4 p.m. to 9 p.m., while, evitably to the conclusion that the blue compounds were meantime, the left-handed rotations were very prevalent the products of pigments which had been, as it were, more and reached a notable maximum at 8 p.m. Other points completely evolved than the others. They were more in the diurnal distribution are noted, including the more perfectly evolved because the process of deassimilation in decided effect resulting from a seasonal grouping of the the corolla cell had been more consummate. The proto-seventy disturbed days dealt with. "The Liberation of Helium from Minerals by the Action of Heat." By D. ORSON WOOD, B.Sc., A.R.C.Sc. Experiments were made to determine how the volume of helium liberated from radio-active minerals by the action of heat depends on the temperature, and on the time for which that temperature is maintained; in particular with a view to the future use of heat to release all the helium contained in minerals not easily treated by chemical methods. The minerals experimented on were monazite and thorianite-the one comparatively poor and the other very rich in helium. The ground minerals were heated in vacuo in tubes of Jena glass or quartz, by an electric heater consisting of a single coil of nickel wire, to temperatures up to 1200° C., which were measured by a Pt resistance thermometer or a Pt Pt-Rh thermocouple. The gas released was purified by drawing it through KOH and P2O5 tubes and finally by Na-K electrodes. The volume was measured in a modified McLeod gauge (described by Prof. Strutt, Proc., vol. lxxx.) specially constructed for the measurement of volumes over a large range-I cc. to 1 cmm. Curves are given to show the volume of helium liberated with time at constant temperatures (250-1000° C.), and also the percentage of the total content obtainable after prolonged heating at the different temperatures. The way in which the gas must be supposed to be retained within the mineral to accord with the results obtained is discussed, and it is concluded:-(1) That heat may be used for the complete liberation of the gas if a sufficiently high temperature (about 900° C.) is reached; and (2) that the results are in agreement with the supposition that small proportion of the gas is diffused through the mineral, and that the remainder is concentrated in very minute cavities within it. 219 some time ago, employing the electrostatic analogy, were in substantial agreement with the conclusions of the authors, but he differed in some important points of detail, and considered that it was unnecessary to make any assumptions as to the particular mode in which absorption of the sugar takes place within the cell. The phenomena of steady diffusion induced by an absorbing sphere in a diffusive field are closely analogous to those presented by an insulated conductor of the same dimensions which has received an electrical charge. In the former case, when the permanent state has been reached, there are produced around the sphere a series of concentric shells of the diffusate which in distribution, although with opposite sign, correspond with the shells of equipotential in the dielectric surrounding the charged sphere. The variations in the gradient of concentration of the diffusate, as measured along the lines of flux, correspond exactly with the variations in the gradient of potential around the charged sphere, measured along the lines of force. In the special case of a spherical yeast cell immersed in a solution of sugar, if the concentration of the diffusate at a remote point is represented by p, and that at the immediate surface of the cell by pr, then the gradient of concentration at the surface, on which the rate of absorption depends, will be represented by P-01, being the radius of the cell. If F be taken as the total absorption in unit time, then : being the coefficient of diffusivity of the sugar. When actual values in C.G.S. units are substituted, and F has been experimentally determined, all the problems given by the authors can be solved by simple mathematical "The Chromophil Tissues and the Adrenal Medulla." treatment, and without the necessity of making any Prof. SWALE Vincent. CHEMICAL SOCIETY. Prof. HAROLD B. DIXON, M.A., F.R.S., President, in the Chair. (Concluded from p. 214). *go. "Studies in Fermentation. Part III. The Rôle of Diffusion in Fermentation by Yeast Cells." By ARTHUR SLATOR and HENRY JULIUS SALOMON SAND. An Facts of a directly experimental nature lead to the belief that during the fermentation of a sugar by yeast, diffusion usually supplies the latter with material so rapidly that convection currents in the solution do not play any part in determining the apparent velocity of the reaction. investigation has now been carried out by means of the apparatus described previously, in order to determine the limiting conditions under which convection currents would begin to become a controlling factor of the rate of the process. These have been deduced from the following formula, which were shown to represent with sufficient accuracy the changes of concentration in a stationary solution in which uniformly distributed yeast cells are operative: -C-Co=3F/8TкR and C1-Co=F/8πKR. In this expression, Crepresents the concentration of the sugar at any point of the solution not in the immediate vicinity of a yeast cell, C, the concentration on the surface of a cell, Co the concentration at its centre, R its radius, and x the diffusion coefficient of the sugar. F is the amount of sugar fermented per cell per unit of time, and was determined by experiment to be approximately 3 × 10-14 g/sec. at 30°. DISCUSSION. Dr. HORACE BROWN fully recognised the importance of knowing the particular conditions under which diffusivity becomes a factor in limiting the specific activity of the yeast cell. The results of an enquiry which he had made assumptions as to the particular mode in which the sugar is dealt with in the interior of the cell or the amount of resistance introduced by the cell-membrane. His (the speaker's) determinations of the amount of sugar fermented by a single yeast cell in a second of time at a temperature of 17-20° led to a value of F of o'92 X 10-14 grm. The authors of the paper had found 3'0 x 10-14 for a temperature of 30°. Taking into account the temperature-coefficient, which had been previously determined by Dr. Slator, these two results were almost identical. One of the problems presented by the authors was to ascertain the minimal concentration, p, of a sugar solution which would be just sufficient to supply the yeast cell with its requirements by diffusion only. From the above formula, and taking the value of F as 30 × 10-14, it follows that : that is to say, the minimal gradient of concentration at the surface of the cell must be 0.85 mgrm. per litre if these conditions are to be fulfilled. If the absorption of the sugar were complete at the exterior surface of the cell, then or would be zero, and the required minimal concentration of the sugar solution, p, would be 0.85 mgrm. per litre. We know, however, that the metabolism of the sugar takes place within the cell, and therefore that pi must have a positive value, which, however, cannot exceed 0.85 mgrm. per litre if the gradient is to remain constant and p is to be at a minimum. The values p and pi can be evaluated by looking at the problem somewhat differently. Let it be supposed that all the fermentative processes are reversed, the activity of the cell as measured by F remaining the same. The cell will now emit sugar into the surrounding medium, and when p is at a minimum, that is, zero, the concentration or at the exterior surface of the cell must equal the gradient, that is to say, must be equal to 0.85 mgrm. per litre if the outward flow is maintained at F. This must also represent the minimal surface concentration when the cell is absorbing sugar and p is also at a minimum; hence, under | is also taking place during the titration. As, however, it these latter conditions, p=2p1=1'70 mgrms. per litre, instead of 1.29 mgrms. as found by the authors. By the same reasoning it could be shown that with maximal stirring a concentration of 0.85 mgrm. per litre ought to be sufficient to supply the cell with its full requirements. In this way it seems possible to avoid all complications introduced by having to take into account internal diffusion, the rate of metabolism and diffusion within the cell being included in the value of p1, which can be evaluated by simple means. Apart from all calculations of this kind, it is evident that diffusion can play but a very small part in limiting the activity of the cell when we consider the very large surface area represented by a relatively small amount of yeast. o'i grm. of pressed yeast, corresponding with about 0.025 grm. of dry substance, contains about 4X 10 cells, representing a surface area of 804 sq. cm. When this yeast is actively fermenting at 30°, this large area is called upon to deal with not more than about 43 mgrms. of sugar per hour. Dr. Slator, in reply, expressed the opinion that it was not possible to calculate these limiting conditions unless diffusion in the yeast cell itself was taken into consideration. *91. "Synthesis of p-Hydroxyphenylethylalkylamines." By GEORGE STANLEY WALPOLE. The methylamino- and ethylamino-homologues of hydroxyphenylethylamine have been prepared in order that their physiological properties might be examined. The starting point was p-methoxyphenylethylamine, and the series of reactions employed may be represented by the following scheme : MeO CoH4'CH,CH,NH, -> 2 2 MeO CoH4CH,CH,NHẠC MeO C6H4 CH2·CH2·NAc(Alk) → HO C6H4 CH2 CH2 NH (Alk). Another series of compounds was prepared similar to the above, but in which the benzenesulphonyl group replaced the acetyl group whenever it occurred. In chemical and physical properties, p-hydroxyphenylethylmethylamine m. p. 130°) and p-hydroxyphenylethylethylamine (m. p. 157-158°) resemble the parent substance, p-hydroxyphenylethylamine, very closely. The hydrochlorides melt at 148.5° and 184-185° respectively. A dibenzoyl derivative, melting at 99°, and a platinichloride, melting at 205°, were prepared from the hydrochloride of p-hydroxyphenylethylmethylamine. The platinichloride corresponds exactly with the data given by Blau for the platinichloride prepared from the base obtained by the destructive distillation of surinamine. There is no longer room for doubt, therefore, that the latter substance is, as supposed, methyltyrosine. 92. "The Condensation of Anisaldehyde with Resorcinol." By FRANK GEORGE POPE and HUBERT HOWARD. In continuation of their previous work (Trans., 1910, xcvii., 78), the authors have condensed anisaldehyde with resorcinol, and have prepared 2: 4-dihydroxy-4-methoxybenzhydrol and its derivatives. 10-Hydroxy-7-phenyldihydro-aß-phenonaphthacridine was also described. 93. "The Influence of Persulphates on the Estimation of Hydrogen Peroxide with Permanganate." BY JOHN ALBERT NEWTON FRIEND. The author showed a few years ago (Trans., 1904, Ixxxv., 547, 1533; 1905, lxxxvii., 738, 1367; and 1906, lxxxix., 1092) that, in ordinary circumstances, a correct estimate of hydrogen peroxide in the presence of potassium persulphate is not obtained by titration with permanganate, for the amount of the latter used always falls short of that required theoretically. It was further shown that for every molecule of peroxide not accounted for by the permanganate, a molecule of persulphate disappears. This suggests that the reaction H2O2 + K2S2O8 = K2SO4 + H2SO4 +Ō2 | proceeds with extreme slowness in ordinary circumstances, the author assumed that it is here "catalytically accelerated by some oxide of manganese formed during titration." In a recent communication (Oesterr. Chem. Zeit., 1910, No. 3), Dr. A. Skrabal, whilst acknowledging the correctness of the experimental work, has suggested another interpretation. He says:-"Nach unserer Meinung dürfte die von Friend beobachtete Erscheinung dadurch besser charakterisiert werden, dass man dieselbe in die Klasse der induzierten Reaktionen verweist. Wir nehmen also an, das die Reaktion 5H2S2O8+2KMnO4+2H20= = = K2SO4+2MnSO4 + 7H2SO4+5O2 welche für sich allein nicht oder nur ausserordentlich langsam verläuft, durch die freiwillige und rasche Reaktion zwischen Hydroperoxyd und Permanganat induziert wird.” If, however, such were the case, it is clear that more permanganate would be required in the titrations than theory demands for the decomposition of the peroxide, the excess being measured by the extent to which the reaction between permanganate and persulphate is induced. Since less permanganate is actually required, such an explanation is seen to be untenable. In a private communication, Dr. Skrabal has kindly given the author permission to state that he acknowledges the error, which arose through his consulting abstracts of the author's papers instead of the originals. He prefers, however, to regard the reaction which takes place between the persulphate and peroxide as “induced " rather than as catalytically accelerated by an oxide of manganese. This is merely a question of choice of terms, however, for induction is very frequently merely a special case of catalysis, neither of which is at present understood. 94. "Amido-oximes and Thioamides." By WILLIAM FRASER RUSsell. The conversion of thioamides into amido-oximes by the action of hydroxylamine is restricted to those thioamides of the type R-CS NHR, those of the formula RCS NRR scarcely being acted on by hydroxylamine under the same conditions. The author has prepared a number of thioamides of the second type, and the effect of hydroxylamine on them was described. 95. "Preparation of the Acyl Derivatives of the Aldehydecyanohydrins." Part II. By OLIVER CHARLES MINTY DAVIS. The reaction between acyl chlorides, aldehydes, and aqueous potassium cyanide (Trans., 1909, xcv., 1403) has been further investigated, and found to be quite general for all types of aldehydes. Benzoyl lactonitrile, a- benzoyloxyisohexonitrile, and benzoyloxyoctonitrile have been prepared; they are oils of high boiling-points. With the exception of acetone, it was found impossible to obtain compounds from ketones analogous to the aldehyde-cyanohydrin derivatives. From acetone, benzoyloxyisobutyronitrile, m - nitrobenzoyloxyisobutyronitrile, and carbethoxyisooxybutyronitrile were obtained. The reactions of the acylaldehhyde-cyanohydrins have also been investigated. 96. "Some Derivatives of Tetramethyl Ferrocyanide." By ERNALD GEORGE JUSTINIAN HARTLEY. Dry potassium ferrocyanide reacts readily with methyl sulphate at 80-90° with the formation of potassium methyl sulphate and a substance which could not be obtained pure, but probably has the composition Me, FeC6N6,2MeSO4, 2MeHSO4. This, on further heating or on keeping in a vacuum, gives off methyl sulphate, and is converted into Me, FeC6N6, Me2SO4, H2SO4, a very stable substance soluble in water and methyl alcohol, and crystallising well from the latter solvent. It gives none of the usual |