526 



SCIENCE 



[N. S. Vol. XXV. No. 640 



glass witli regularly spaced strips of tin- 

 foil, and this led the present writer to 

 investigate the possibility of making a thin 

 film of transparent substance more trans- 

 parent by covering it with a much thinner 

 film of metal. 



A film of dye was deposited upon one 

 half of a plate of glass, and silver was de- 

 posited on half the glass but in a direction 

 at right angles to the dye film. One could 

 thus compare silvered dye with silvered 

 glass. The film of silver was so thin that 

 it was impossible to see any difference be- 

 tween the part of the glass that was silvered 

 and the part that was not, but where the 

 glass was covered with both silver and dye, 

 less light was transmitted than when the 

 glass was covered with dye alone. Similar 

 effects were obtained with combined films 

 of silver and selenium. 



The explanation is that the phase change 

 at the top surface of the silvered dye film 

 is different from that at the top surface of 

 the bare dye film. The rays reflected from 

 the back surface of the dye will therefore 

 interfere in a different way with the ray 

 reflected from the top surface, in the two 

 cases, and for certain thicknesses of the 

 dye film, the intensity of the light reflected 

 from the film will be diminished by silver- 

 ing, and there will be a correspondingly 

 increased transmission. 



A Preliminary Communication concerning 

 a New Fundamental Principle of the 

 Kinetic Theory of Gases: LuiGi d'Aueia, 

 Philadelphia. 



In this paper the author proves that the 

 square of the speed of an elastic sphere 

 forced to oscillate between two elastic 

 plates perpendicular to the direction of its 

 motion while one of the plates approaches 

 the other with speed which is very small 

 compared with that of the sphere, varies 

 inversely with the length of its free path. 

 The author observes that in a medium com- 



posed of numerous elastic spheres moving 

 in all directions in a bounded space, each 

 sphere can be considered as if moving with 

 the mean square speed in a path equal to 

 the mean free path of all the spheres ; and, 

 therefore, in accordance with his new 

 theorem, in such a medium the mean square 

 of the speed would vary inversely with the 

 mean free path. 



Combining this result with the axpres- 

 sion for the mean free path, which, accord- 

 ing to Clausius, varies inversely with the 

 density of the medium and the square of 

 the diameter of one of the spheres, and 

 observing that the pressure of the medium 

 upon unit area varies with its density and 

 the mean square of the speed of the 

 spheres, the author arrives at the equation 

 pv- = constant, independent of thermo- 

 dynamic considerations, in which p is the 

 pressure per unit area and v the volume of 

 unit quantity of the medium. Thus in a 

 medium composed of elastic spheres in mo- 

 tion in a bounded space, the pressure per 

 unit area of the bounding surface would 

 vary inversely with the square of the 

 volume. 



If the above medium represents a gas, 

 we must have also pv^ = constant, in 

 which y is the ratio of the specific heats, 

 and therefore, y = 2. According to Clau- 

 sius and Maxwell, for such a gas the ratio 

 of the specific heats would be 5/3, and this 

 is too small to account for the ratio of the 

 specific heats of mercury vapor found by 

 Kundt and Warburg which ranges from 

 1.631 to 1.695, since the theory requires 

 that this ratio should be considerably less 

 than the theoretical value. 



The result y = 2 leads to the new funda- 

 mental equation pv = E in which E repre- 

 sents the energy of agitation of the gas, 

 and the author shows that this equation 

 accords with the principle of the conserva- 

 tion of energy, and that, therefore, the 

 equation pv = 2/3E, which results from 



