242 ANNUAL RECORD OF SCIENCE AND INDUSTRY. 



tion of the polarization of the sky is to assume that it is the 

 result of specular reflection from molecules of air or aqueous 

 vapor. According to this, the polarization should be com- 

 plete at a distance of ninety degrees from the sun, while in 

 reality it is only about seventy per cent. The explanation 

 of this and other deviations is possibly found in the reflection 

 of the sun from the sea lying to the east of Boston. When 

 the polarimeter is directed toward a polished colored plane 

 surface, the two images assume different tints, one having 

 the color of the reflecting body itself, the other having the 

 color of the original light. On examining the light of the 

 sky under proper circumstances, Professor Pickering con- 

 cludes, from a consideration of this principle, that the true 

 color of the sky particles is blue. Pickering, Amplications of 

 FresneVs Formula. 



THE DIFFUSION OF GASES. 



In a recent number of the Archives of Physical Sciences, 

 Dufour pives a detailed account of his researches on the va- 

 nation of temperature which accompanies diffusion of gases 

 through a porous partition. The author first studied the in- 

 fluence of the dry or humid state of gases coming in contact 

 with the porous wall without diffusion; next, the variation 

 of temperature, where there is no change of pressure ; and, 

 third, where there is such a change. With constant pressure, 

 there is a fall of temperature on the side where the denser 

 eras is, and a rise on the other side. Each current seems to 

 have a heating effect where it enters the porous wall, and a 

 cooling one where it issues. With change of pressure through 

 endosmose of a lighter gas, the temperature slightly increases, 

 sinking again as the pressure tends to equilibrium. Where 

 exosmose of a lighter gas causes diminution of pressure, the 

 temperature slightly falls, subsequently rising again. 12 A, 

 1873, IX., 454. 



NEW NORMAL FLAME FOE FHOTOMETEIC PUEPOSES. 



As a substitute for the inexact and inconvenient normal 

 candle, Carcel lamp, etc., Professor Wartha suggests a simple 

 apparatus for photometric observations, constructed on the 

 following plan, by which a flame that is easily controlled, 

 and perfectly constant for hours, may be readily produced by 



