The minimum proportion of chlorine in this case lay between 1 in 518,587,360 and 550,513,022 parts of water. Table XX.-Zn + Cd + Chlorine in 1550 grains of Water at 11° C. Eleven other solutions of different strengths, all weaker than 0.02027, each gave the same potential as water. The minimum proportion of chlorine required to disturb the balance lay between 1 part in 55,436 and 76,467 parts of water. In this case, the addition of chlorine decreased the electromotive force; a similar effect occurred with a zinc-platinum couple in a solution of potassic iodide ("Influence of the Chemical Energy of Electrolytes, &c.," Roy. Soc. Proc.,' June 7, 1888). The following are the minimum proportions of chlorine which were required to change the potential : Mg + Pt+ Cl. Zn + Au+ Cl. Zn + Cd + Cl. Table XXI. With an Astatic Galvanometer. Between 1 in 17,000,000,000 and 17,612,000,000 Mg+Pt+Cl. Between 1 in 27,062 millions and 32,291 millions. The examples contained in this paper are sufficient to show, that the proportion of the same exciting liquid, necessary to disturb the potential of a voltaic couple in water, and the order of variation of potential caused by change of strength of liquid, vary with each different positive or negative* metal. The numbers in Tables IV, VIII, XII and XXI, show that the more positive or more easily corroded the positive metal, or the more negative and less easily corroded the negative one, the smaller usually was the proportion of dissolved substance required to change the potential. In the case of chlorine, *If the negative metal is not at all corroded, the order of change of potential by change of negative metal is not much alected. as well as in that of bromine, the magnitudes of the minimum proportions of substance necessary to change the potential of magnesiumplatinum, zinc-platinum, and cadmium-platinum couples, varied directly as the atomic weights of the positive metals. The experiments also show that the degree of sensitiveness of the arrangement for detecting the minimum-point of change of potential depends largely upon the kind of galvanometer employed. As a more sensitive galvanometer enables us to detect a change of potential caused by a much smaller proportion of material; and as the proportion of substance capable of detection is smaller the greater the free chemical energy of each of the uniting bodies, it is probable that the electromotive force really begins to increase with the very smallest addition of the substance, and might be detected if our means of detection were sufficiently sensitive or the free chemical energy was sufficiently strong. VII. “Magnetic Qualities of Nickel (Supplementary Paper).” By J. A. EWING, F.R.S., Professor of Engineering in University College, Dundee. Received June 14, 1888. (Abstract.) The paper is a supplement to one with the same title by Professor Ewing and Mr. G. C. Cowan, which was read at a recent meeting of the Society. It describes experiments, conducted under the author's direction by two of his students, Mr. W. Low and Mr. D. Low, on the effects of longitudinal compression on the magnetic permeability and retentiveness of nickel. The results are exhibited by means of curves, showing the relation which was determined between the intensity of magnetisation of the metal and the magnetising force, when a nickel bar, reduced to approximate endlessness by a massive iron yoke which formed a magnetic connexion between its ends, was magnetised under more or less stress of longitudinal compression. Corresponding curves show the relation of residual magnetism to magnetising force, for various amounts of stress; and others are drawn to show the relation of magnetic permeability to magnetic induction. Initial values of the permeability, under very feeble magnetising forces, were also determined. The experiments were concluded by an examination of the behaviour of nickel in magnetic fields of great strength. Magnetising forces ranging from 3000 to 13,000 c.g.s. units were applied by placing a short bobbin with a narrow neck made of nickel between the poles of a large electromagnet, and it was found that these produced a practically constant intensity of magnetisation which is to be accepted as the saturation value. VIII. "Evaporation and Dissociation. Part VIII. A Study of the Thermal Properties of Propyl Alcohol." BY WILLIAM RAMSAY, Ph.D., F.R.S., and SYDNEY YOUNG, D.Sc. Received June 14, 1888. (Abstract.) In continuation of our investigations of the thermal properties of pure liquids, we have now determined the vapour-pressures, vapourdensities, and expansion in the liquid and gaseous states of propyl alcohol, and from these results we have calculated the heats of vaporisation at definite temperatures. The compressibility of the liquid has also been measured. The range of temperature is from 5° to 280° C., and the range of pressure from 5 mm. to 56,000 mm. The memoir contains an account of the purification of the propyl alcohol; determinations of its specific gravity at 0°, and at 10°-72; and of the constants mentioned above. The approximate critical temperature of propyl alcohol is 263°7; the approximate critical pressure is 38,120 mm., and the approximate volume of one gram is 3.6 c.c. The first two of these constants must be very nearly correct; the third cannot be determined with the same degree of precision. The memoir is accompanied by plates, showing the relations of volume, temperature, and pressure in a graphic form. IX. "Contributions to the Chemistry of Chlorophyll. No. III." By EDWARD SCHUNCK, F.R.S. Received June 19, 1888. (Abstract.) This paper is a continuation of the previous ones on the same subject. In it the author gives an account of the action of alkalis on phyllocyanin so far as regards the first stage of the process, and of the products thereby formed. Phyllocyanin when acted upon by alkalis yields in the first instance a well-crystallised substance of a peacock- or steel-blue colour, to which he gives the name of Phyllotaonin. He describes its properties and those of some of its compounds. When hydrochloric acid gas in excess is passed through a solution of chlorophyll in alcoholic soda, a compound crystallising in lustrous purple needles is formed, which seems to be the ethyl ether of phyllotaonin. By substituting methylic for ethylic alcohol a very similar compound is obtained, which the author considers to be the corresponding methyl ether. Though these compounds readily yield phyllotaonin by saponification with alcoholic potash or soda, the author did not succeed in reproducing them by the combined action of alcohol and hydrochloric acid on phyllotaonin. X. "On the Specific Resistance of Mercury." By R. T. GLAZEBROOK, M.A., F.R.S., Fellow of Trinity College, and T. C. FITZPATRICK, B.A., Fellow of Christ's College, Demonstrators in the Cavendish Laboratory, Cambridge. Received June 19, 1888. (Abstract.) The paper contains an account of experiments made to determine the value of the resistance of a column of mercury, 1 metre long and 1 sq. mm. in cross section, in terms of the B.A. unit. The method employed differed very slightly from that of Lord Rayleigh and Mrs. Sidgwick ('Phil. Trans.,' 1883). Tubes of about 1, 2, and 3 sq. mm. in cross section were calibrated and filled with mercury. They were then immersed in melting ice, and their resistance compared with that of the B.A. standards, using Carey Foster's method and the B.A. bridge. The length of the mercury column, occupying nearly the whole of the tube, was measured, and the mass of the same determined. From this the average cross section is obtained, and hence the value of r, the resistance of a column 1 metre long, 1 sq. mm. in cross section. The mercury used to find the cross section was with few exceptions that which had been employed in finding the resistance. The results of the measurements are given in Table I. In the table, Column 1 gives the number of the tube, Column 2 the number of the observation. L is the length of the tube, and a the mean radius of the cross section, R the observed resistance in B.A. units. The mean value of r found from the three 1 mm. tubes is 0.95354 B.A. units. The other four tubes of one-half and one-third units respectively lead to the value r 095344 B.A. units. The difference between the two is considerable, and reasons are given for assigning more weight to the first value. For an account of the experiments and of the small precautions necessary to secure accuracy, reference must be made to the paper. Table II gives a list of the various values which have been found for r with the lengths of the column of mercury which, according to the different observers, has a resistance of 1 ohm (109 C.G.S. units of resistance). In combining our own observations we have assigned. weights to the various tubes inversely proportional to their diameters, and we find as our final value r = 0.95352. VOL. XLIV. 2 F |