28 



RATIO OF SPECIFIC HEATS. 



Conclusion. 



We li;ive already iiitiiuateil tliat tlii^ values for n obtained with the new bolo- 

 meter would be larger than the corre.'=;[)ondiug data obtained in 1887 by the use of 

 a silver wire .004 cm. thick for Ijolometric purposes. This anticipation has been 

 quite borne out by the experiments, showing that the discrepancy due to heat con- 

 ducted inward from the terminals had seriously affected our earlier j-esults. 



A summary of the diffeient experiments made for evaluating h is given by A. 

 Winkelmann in his Ilandhiuh tier Fhysilc, 1895, vol. ii., pp. 381, 382, covering the 

 ground up to 1895. Furthermore, the Joxirnal de Phynique, 1895 (3), vol. iv., pj). 

 463-465, contains an historical and critical review ' of the subject as to methods and 

 observations, by M. G. Manoevriei-. We do not therefore think it advisable to 

 reproduce these bibliographies here, in which, to our knowledge, only the experiments 

 of J. Webster Low ^ seem to have been overlooked. On the other hand, a compari- 

 son of our i-esults with those of eai'lier observers in the form of a table such as is 

 given at the end of this paper, will conduce to an intelligent survey of the present 

 aspect of the sul)ject. All correlative results wei-e obtained by methods different 

 from ours. 



The researches of Rontgen and of Paquet are based on the well-known method 

 of Clement and Desoiines. 



The values of Kayser, Wullnei-, and J. Webster Low are computed from data for 

 the velocity of sound. The first two of these observers made use of Kundt's method, 

 while Low i-eached his results through the interferential method for the velocity of 

 sound waves devised by Quincke. Manoevrier's method is the following : A given 

 mass of gas is compressed adiabaticaUy, and the increase of pressure produced 

 determined. The change of pressure due to an equal but isotliermal change of 

 volume is easily computed fi'om known laws. If both volume decrements be 

 small, the ratio of the pressure increments is very nearly equal to the i-atio of the 

 specific heats. Hence k may be found fi'om the data for the adiabatic volume 

 reduction. 



All data given in the follow inL' t;ibl<^ if fci' to ni-dinary atmospheric tenqieratures. 



' In this review, as well as in aiiutiicr paper [you nut I Jc P/iys., 1S95 (3), vol. iv. pp. 368-373), 

 M. Manoevricr devotes some space to our work of 1S87 and to the results of the present investiga- 

 tion, as far as they were given in the brief summary contained in the Reports of the Oxford meeting 

 of the British Association. We consider a direct reply superfluous, for the points raised by M. 

 Manoevrier are all touched upon in the present extended publication in so far as they are matter-of- 

 fact and not merely controversial. 



' J. Webster T,ow, /lYc,/. A/m., vol. Hi., pp. 6.|i-()6_|. iSc;.). 



