THE 



LONDON, EDINBURGH, and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE. 



0* 



I 



[FIFTH SERIES.] 



JULY 1899. 



I. A Comparison of Rowland's Mercury Thermometers 

 a Callendar- Griffiths' Platinum Thermometer; a Compa- 

 rison of the Platinum Thermometer ivith a Tonnelot Ther- 

 mometer standardized at the Bureau International; and a 

 Reduction of Rowland's Value of the Mechanical Equivalent 

 of Heat to the Paris Nitrogen Scale. By Chaiiles W. 

 Waidner, Jr., and Francis Mallory*. 



THE recent determinations of the mechanical equivalent of 

 heat by electrical methods, while in fair agreement among 

 themselves, seem to give results considerably higher than those 

 obtained by experimenters using the more direct mechanical 

 methods ; these differences, which are probably greater than 

 the errors of experiment, must be due to a difference in the 

 standards of thermometry employed or to a still undiscovered 

 error in the system of electric units. The object of the 

 present investigation was to obtain a connexion between 

 Rowland's standard of thermometry and that employed in the 

 electrical determinations of the equivalent, in order to render 

 the different determinations more easily comparable. 



Joule compared his thermometer with Rowland's Bauclin 

 6166 (Proc. Amer. Acad. xvi. p. 38, 1880), and in this way 

 Joule's determinations of the mechanical equivalent have 

 been reduced to the Rowland " air-scale.'' By a further 

 elaborate comparison by Prof. Schuster (Phil. Mag. xxxix. 1895) 



* Communicated by Prof. H. A. Rowland. An abstract of tbis investi- 

 gation was read at tbe Toronto meeting of tbe British Association, and a 

 preliminary account appeared in tbe Jobns Hopkins University Circulars, 

 June 1897, 1898. 



Phil. Mag. S. 5. Yol. 48. No. 290. July 1899. B 



