June 28, 1888J 



NATURE 



213 



coming from the uppermost spectrum were reflected by means of 

 a small right-angled prism in a direction nearly at right angles 

 to the original direction on to another right-angled prism. Both 

 prisms were attached to the card. From this last prism the rays 

 fell on a lens, and formed on a white screen an image ofthe face 

 of the spectroscope prism in monochromatic light. The ray of 

 the same wave-length as that reflected from the upper spectrum 

 passed through the lower half of the slit, and falling on another 

 lens formed another image of the face of the prism, superposed 

 over the first image. A rod placed in front of the screen thus 

 cast two shadows, one illuminated by monochromatic rays from 

 the top spectrum, and the other by those from the bottom 

 spectrum. The illumination of the two shadows was equalized 

 by means of rotating sectors which could be closed and opened 

 at pleasure during the time of rotation. The angle to which the 

 sector required to be opened to establish equality of illumination 

 of the two shadows gave the ratio of the brightness of the two 

 spectra. When proper adjustment had been made, the relative 

 brightness was the same throughout the entire spectrum. 



To measure the intensity of any ray reflected from a pigment, 

 a paper was coated with it and placed adjacent to a white surface, 

 and it was so arranged that, one shadow of the rod fell on the 

 coloured surface and the other on the white surface. The illu- 

 minations were then equalized by the sectors, and the relative 

 intensities of the two reflected rays calculated. This was ^re- 

 peated throughout the spectrum. Vermilion, emerald-green, 

 French ultramarine were first measured by the above method, and 

 then sectors of these colours prepared, which when rotated gave 

 a gray matching a gray obtained by rotation of black and white. 

 The luminosity curves of these three colours were then calculated 

 and reduced proportionally to the angle that each sector occupied 

 in the disk. The luminosity curve of the white was then reduced 

 in a similar manner, and it was found that the sum of the lumin- 

 osities of the three colours almost exactly equalled that of the 

 white. The same measurements were gone through with pale- 

 yellow chrome and a French blue, which formed a gray on 

 rotation, with like results. It was further found that the sum oj 

 the intensities of vermilion, (blue, and green varied at different 

 parts of the spectrum, and the line joining them was not parallel 

 to the straight line which represented white for all colours of the 

 spectrum and which itself was parallel to the base. Since a 

 straight line parallel to the base indicated degraded white, it 

 followed that if the intensity of the rays of the spectrum were 

 reduced proportionally to the height of the ordinates above a line 

 tangential to the curved line (which represented the sum of the 

 intensities of the three colours at the different parts of the spec- 

 trum) and were recombined, a gray should result. A method 

 was devised of trying this, and the experiment proved that such 

 was the case. The same plan enabled the colour of any pigment 

 to be reproduced from the spectrum on the screen. The com- 

 bination of colours to form a gray on rotation by a colour-blind 

 person was also tried, and after the curve of luminosity of the 

 colours had been calculated and reduced according to the amount 

 required in the disk, it was found that the sum of the areas of 

 the curves was approximately equal to the white necessary to 

 be added to a black disk to form a gray of equal intensity as per- 

 ceived by him. The spectrum intensity of gas-light in comparison 

 with the electric light was also measured, and the amount of 

 the different colours necessary to form a gray in this light was 

 ascertained by experiment. 



As before, it was found that the calculated luminosity of the 

 colours was equal to the white which, combined with black, 

 formed a gray of equal luminosity. 



The question of the coloured light reflected from different 

 metals was next considered, and the method of measuring it 

 devised, as was also the method of measuring absorption spectra. 

 The luminosity curves obtained by the old method were compared 

 with those obtained by the present method, and so close an 

 agreement between them was found to exist as to give a further 

 confirmation that our former plan was accurate. A number of 

 pigments that can be used for forming grays by rotation were 

 measured, and the results tabulated in percentages of the 

 spectrum of white light and on a wave-length scale. 



Physical Society, June 9.— Prof. Reinold, President, in 

 the chair. — The following papers were read :— On the analogy 

 between dilute solutions and gases as regards Gay-Lussac's and 

 Boyle's and Avogadro's laws, by Prof, van 't Hoff, presented 

 by Prof. Ramsay, F.R.S. If a dilute aqueous solution of 

 sugar (say 1 per cent.) be placed in a vessel, A (the walls of 



which are permeable to water, but not to sugar molecules), and 

 immersed in a large quantity of water, B, water will pass from 

 B to A until a certain difference of pressure exists between the 

 inside and outside of A, that difference depending on the 

 temperature and concentration of the solution. The pressure is 

 called osmotic pressure, and the walls of A are said to be semi- 

 permeable. Such a vessel may be artificially produced by 

 depositing ferrocyanide of copper on unglazed porcelain ; but 

 many of the experiments dealt with in the paper have been made 

 with the cells of plants, the walls of which form good semi- 

 permeable membranes. At constant temperature the osmotic 

 pressure is found to be proportional to the concentration of the 

 solution, and for a given concentration the pressure is propor- 

 tional to the absolute temperature. Similar results have been 

 obtained with solutions of KN0 3 , K 2 S0 4 , NaCl, &c, and Soret 

 has found that if a solution be heated unequally at different 

 parts, the warmer parts are less concentrated, just as in gases 

 under similar conditions the warmer parts are more rarefied. 

 The numerical results are in fair accordance with those deduced 

 from the laws above stated. Theoretical proofs of the laws are 

 given, in which reversible cycles and the second law of thermo- 

 dynamics are made use of. By similar reasoning the author 

 concludes that " under equal osmotic pressure, and at the same 

 temperature, equal volumes of all solutions contain the same 

 number of molecules, and moreover the same number of mole- 

 cules which would be contained in a gas under the same condi- 

 tions of temperature and pressure." These results are confirmed 

 by Pfeffer's direct determinations of osmotic pressure, and 

 Raoult's experiments on the " molecular lowering of vapour- 

 pressure," and the "molecular depression of the freezing-point 

 of the solvent." The latter part of the paper contains applica- 

 tions to chemical phenomena. Prof. Riicker regretted that the 

 names Boyle's law and Gay-Lussac's law had been so persist- 

 ently made use of in the paper, as he thought a wrong impression 

 would be spread as to the nature of the phenomena. He also 

 considered it probable that the proportionality observed was 

 merely the result of the smallness of the ranges over which the 

 experiments had been made. Mr. H. Crompton took exception 

 to the imaginative character of the reasoning, and thought much 

 more experimental proof was required before the results could 

 be accepted for any but very small ranges of concentration. In 

 answer to Prof. Reinold, Prof. Ramsay said the experimental 

 data were not obtained by van 't Hoff himself, but were taken 

 chiefly from Raoult's determinations. — On a method of compar- 

 ing very unequal capacities, by Dr. A. H. Fison. One coating 

 of each condenser is joined to earth, and to one end, A, of a 

 high resistance (20,000 or 30,000 ohms), through which a current 

 is flowing. The small condenser is charged to the P. D. existing 

 between the ends A, B, of the resistance, and discharged into 

 the large one. This is repeated a great number of times. If C 

 be a point between A and B, the resistance between A and C 

 may be varied until the P.D. between them is equal to that 

 between the coatings of the condensers after n operations. If 

 theinsulatedcoatings.be now joined to C through a galvano- 

 meter, no deflection will result. The relation between the 

 capacities Cj and C 2 of the large and small condensers is 

 given by 



/. , c 2 y _ Rab 



where Rab, Rbc are the resistances between AB and BC re- 

 spectively. Since time is required to perform the operation, the 

 instantaneous capacities cannot be compared, and accordingly 

 the measurements are taken after a definite time of electrification. 

 A special rotating key was shown for performing ten operations 

 per revolution, in which a trigger arrangement was provided for 

 stopping the rotation after a predetermined integral number of 

 revolutions. The method has been used for comparing a small 

 air-condenser with a microfarad. The capacity of the former 

 was also calculated electro-statically (correction being made for 

 the edges), and that of the latter measured electro-magnetically 

 by a ballistic galvanometer. The results give a value for v 

 equal to 2*965 x io 10 . In these experiments the capacity of the 

 rotating key was allowed for. Under favourable conditions, 

 capacities in the ratio of 1 to 1000 or 1 to 10,000 can be com- 

 pared with an accuracy of \ per cent. Prof. Ayrton thought the 

 novelty of the arrangement was in the rotating key, as the 

 method of comparing unequal capacities by charging the smaller 

 and discharging it into the larger a considerable number ot 

 times had been described and used by himself and Prof. Perry 



