JOURNAL 
OF THE 
WASHINGTON ACADEMY OF SCIENCES 
VoL. 18 JANUARY 19, 1923 No. 2 
PHYSICS.—Notes on the electric heating of calorimeters.1 WALTER 
P. Wuire. Geophysical Laboratory, Carnegie Institution of 
Washington. 
Various systematic errors in calorimetry, especially those in the 
difficult heat capacity determination, are diminished or removed if 
the calorimeter is used to compare two thermal quantities, one of 
which is a standard. It now seems that such comparison methods, 
with rare exceptions, should be regarded as the normal thing in ac- 
curate calorimetry. . 
A fundamental standard should be above suspicion, hence the stand- 
ard adopted should possess to the full the precision of the comparative 
calorimetric measurements. This is much higher than may generally 
be recognized. More than ten years ago calorimeters in two different 
laboratories showed a precision, i.e., an agreement, of 20 to 30 per 
million. It will hardly do to assign such a value for the probable final 
accuracy of a calorimetric experiment at present, but it is possible, 
under favorable conditions often attainable, and with no great amount 
of adjustment or experimenting, to secure, as far as the calorimeter 
itself is concerned, a precision of 100 per million, which would usually 
mean a like final accuracy if a comparison method is used. The stand- 
ard for comparison should be safely more accurate than this, say to 30 
per million. 
Except in certain specific heat determinations, this standard of 
comparison must be a quantity of heat. Of the two available stand- 
ards, mechanical and electric, the electric is unquestionably of prepon- 
derant practical value. It is really the standard in most cases at the 
1 Received December 20, 1922. 
17 
