OF ARTS AND SCIENCES. 149 



ing the theory of the subject, but cannot have any weight as accurate 

 determinations of the equivalent. 



From Friction Experiments. 



Experiments of this nature, that is, irreversible processes for con- 

 verting mechanical energy into heat, give by far the best methods for 

 the determination of the equivalent. 



Rumford's experiment of 171)8 is only valuable from an historical 

 point of view. Joule's results. since 1843 undoubtedly give the best 

 data we yet have for the determination of the equivalent. The mean 

 of all his friction experiments of 1847 and 1850 which are given in 

 the table is 425.8, though he prefers the smallest number, 423.9, of 

 1850. This last number is at present accepted throughout the civil- 

 ized world, though there is at present a tendency to consider the 

 number too small. But this value and his recent result of 1878 have 

 undoubtedly as much weight as all other results put together; 



As sources of error in these determinations I would suggest, first, 

 the use of the mercurial instead of the air thermometer. Joule com- 

 pared his thermometers with one made by Fastre. In the Appendix 

 to Thermometry I give the comparison of two thermometers made by 

 Fastre in 1850, with the air thermometer, as well as of a large number 

 of others. From this it seems that all thermometers as far as measured 

 stand above the air thermometer between 0° and 100°, and that the aver- 

 age for the Fastre at 40° is about 0°.l C. Using the formula given 

 in Thermometry this would produce an error of about 3 parts in 1,000 

 at 15° C., the temperature Joule used. 



The specific heat of copper which Joule uses, namely, .09515, is 

 undoubtedly too large. Using the value deduced from more recent 

 experiments in calculating the capacity of my calorimeter, .0922, 

 Joule's number would again be increased 13 parts in 10,000, so that 

 we have, — 



Joule's value 423.9, water at 15°.7 C. 



Reduction to air thermometer . . -(-1.3 



Correction for specific heat of copper -f- .5 



" to latitude of Baltimore -{- .5 



426.2 



It does not seem improbable that this should be still further in- 

 creased, seeing that the reduction to the air thermometer is the small- 

 est admissible, as most other thermometers which I have measured 

 give greater correction, and some even more than three times as great 



