the Mechanical Equivalent of Heat. 



171 



corrected, taking into account each individual experiment, 

 the thermometers used in it, and the number of observations 

 made with each thermometer. The original values, and the 

 corrected values found in this way, are compared at eeviral 

 temperatures, in the following table: — 



Temp. 



Old. 





Corrected. 



Griffiths. 



Schuster and 

 Gannon. 



6° 



4209 x 



10 4 



4204 x 



10 4 







10 



4200 





4197 









15 



4189 





4189 





4199-7 xlO 4 





20 



4179 





4183 





4193-2 



4191 xlO 4 



25 



4173 





4177 





4187-4 





30 



4171 





4173 









35 



4173 





4174 









t 



VZ'O «/</ 

















<s> 



















o 



















u 





















4. loo >(4* 

















tt-ifO « 10* 







It 



's 









u 



y-t to tio u 



















-J c 



2 









\ 





f t 











5° 



id 



IS' 



xo' 



2 J 5 ""----— _^ 



3°^ r "~ 



35" 



Te m/je ratu re. on the "P a T.s hudroqer. sco/e 



These numbers are in the C.G.S. system, and hydrogen scale. 

 The results given above are also represented by the curves 

 shown in the diagram . The corrected value of the mechanical 

 equivalent found in this way is the same as Rowland's old value 

 at 14°-5 C, at 10° it is lower by one part in 1700, at 20° it is 

 higher by one part in 1200, and at 25° higher by one part in 

 1000. The heat-capacity of water, as indicated by the corrected 

 values of the mechanical equivalent, varies at about the same 

 rate as that given by the experiments of Griffiths. (Phil. Trans, 

 clxxxiv. A. p. 361, 1893. Phil. Mag. xl. pp. 437 and 447, 1895.) 

 If, therefore, curves were drawn representing the specific heat 

 of water between 15° and 25°, as given by Rowland's experi- 



N2 



