562 REPOKT — 1881. 



in the sodium vapour placed in the Bunsen flame is strong for a certain set of 

 yellow vibrations, and weak for a certain set of green vibrations. If the Bunsen 

 flame is now removed into an enclosui'e tlie temperature of which is equal to its- 

 self, the internal radiation in the gi-een will suddenly be raised, while that in the 

 yellow will not be (appreciably) affected. As tlieremust be some relation between 

 the internal radiation and the energy of vibration of the molecules, the gi-een 

 vibrations will be raised in intensity and be made more nearly equal to the yellow 

 vibrations. 



Tlie following may show a little more clearly that the relative intensities of 

 the vibrations need not be the same when the body is placed in an enclosure of 

 uniform temperature, as when it is allowed to radiate into space. 



Imao'ine a set of molecides not losing any heat by radiation, and assiune that 

 the molecules are, like the sodium molecules, capable of vibrating in two distinct 

 periods, one corresponding to the yellow, and one corresponding to the green rays. 

 There is a constant interchange between energy of translatory motion and energy 

 of vibratory motion, and we imagine that the energy of translation is more easily 

 transformed into yellow vibrations than into gi-een vibrations. If the average 

 energy for each set remains constant, it is because the energy of vibration is at the 

 same time for other molecules transformed into energy of translation, and the 

 yellow vibrations more easily than the green vibrations. The mathematical cal- 

 culations of Maxwell and Boltzmann have shown that when the final equilibrium 

 is reached, the final energy of vibration for the two periods must be the same. 

 Imagine now the gas to be taken out of the euclosiu-e and allowed to radiate into 

 . space. During the first instant, the yellow and the green vibrations will Ije of the 

 same intensity, but very soon the molecular encounters will show their influence. 

 We have assumed that yellow vibrations are more easily produced than green vi- 

 brations. While, therefore, the gi'een vibrations are allowed to decrease in inten- 

 sity, that of the yellow rays is constantly renewed by the encounters. It follows 

 that the yellow rays ■\\ill now appear the brighter, though in the enclosm-e of uni- 

 form temperature the energy of radiation for the green light was as large. The 

 mere fact therefore that we observe lines of different intensities in the spectra of 

 bodies is not in itself sufficient to invalidate Boltzmann's theoretical conclusion. 

 But other considerations show that these conclusions are not supported by experi- 

 ment. Thus, from the ratio of the two speciffc heats of mercury vapour, Boltzmann 

 concludes that the molecule of mercury is a rigid particle, incapable of rotating or 

 vibrating. It is suggested that forces due to the vibrations themselves, which have 

 been neglected by Boltzmann, may be the cause of the discrepancy between theory 

 and experiment. 



8. On a New Electrometer and some preliminary Experiments on Voltaic 



Action. By J. Brown. 



The author exhibited and described an instrument for measuring the difference 

 of potential of two substances in contact while immersed in any desired gas. The 

 metals to be examined are formed into a pair of quadrants fixed on an ebonite sup- 

 port in the apparatus, and their potential-difference is examined by means of a 

 needle similar to half that in Sir William Thomson's electrometer, swinging over 

 the slit between them, and suspended by a platinum wu-e, -OOl-inch diameter from 

 a torsion head at the top of an ebonite tute rising from the cover of the airtight 

 metal case enclosing the quadrants, and into which any required gas can be intro- 

 duced. The needle is electrified by connecting its suspension wire to either pole of 

 a 100-cell Daniell battery, and its deflections are observed by mirror and scale. 

 One quadrant is kept permanently connected with the earth, while a wire, joined to 

 the other and passing airtight through the case, provides means for connecting the 

 two together metallically, or one to each pole of any cell with theE.M.F. of which 

 it is desired to compare their potential-difference either by Kohlrausche's method or 

 that of Su- William Thomson (' B.A. Report,' 1880). 



As the observed potential-difference of copper and zinc continually decreases 

 with the oxidation of their .surfaces in the air, a series of observations was made to 



I 



