Prof. Graham on the Law of the Diffusion of Gases. 175 



The smoke which is produced in the former cases invariably 

 reddens the flame, and its perfect removal causes the gas to 

 approximate to the light of the sun, which is always bluer than 

 that of the whitest flames from wax, oil, or tallow. 



There is a very pretty experiment illustrative of some of 

 the preceding observations, which is easily made. Place two 

 candles at the distance of three or four feet from the eye, and 

 about one foot from each other, and having closed one eye, 

 fix the other intently upon either of the candles, as if it 

 were examining with attention some point of the wick. The 

 other candle will be seen by indirect vision, and after a little 

 time it becomes much brighter and bluer than the first, in 

 consequence of the part of the retina on which its light falls 

 being more susceptible than the more frequently used portion 

 in the axis of the eye, upon which the light of the second is 

 incident. The higher degree of excitation of the retina pro- 

 duced by the candle seen indirectly, renders that portion of 

 the membrane less sensible to the red rays ; and if the excita- 

 tion is continued, the image will become actually blue, and will 

 be surrounded with a halo of yellow nebulous light. The blue 

 image, indeed, will sometimes disappear, and leave nothing in 

 its place but a nebulous halo. 



Allerly, Jan. 30th, 1833. 



XXVII. On the Law of the Diffusion of Gases. By Thomas 

 GrahaxM, Esq. M.A. F.R.S. Ed. Professor of Chemistry in 

 the Andersonian University, Glasgow*. 



TT is the object of this paper to establish with numerical 

 -*- exactness the following law of the diffusion of gases : 



" The diffusion or spontaneous intermixture of two gases 

 in contact, is effected by an interchange in position of inde- 

 finitely minute volumes of the gases, which volumes are not 

 necessarily of equal magnitude, being, in the case of each gas, 

 inversely proportional to the square root of the density of that 



g as C 



These replacing volumes of the gases may be named equi- 

 valent volumes of diffusion, and are as follows : Air, I ; Hy- 

 drogen, 3*7947; Carburetted hydrogen, 1*3414; Water-va- 

 pour, 1-2649; Nitrogen, 1*0140; Oxygen, 0*9487; Carbonic 

 acid, 0*8091 ; Chlorine, 0*6325, &c; numbers which are in- 

 versely proportional to the square roots of the densities of 



* Read before the Royal Society of Edinburgh, December 19, 1831; and 

 now reprinted from the Edinb. Phil. Trans., with an Appendix. — Com- 

 municated by the Author. 



