434 DICKINSON: combustion calorimetry 



to the rounding off of the edges and corners of the fragments 

 formed at the inception of bending. Evidently different values 

 depending upon the amount of the strain and the nature of the 

 substance strained might be obtained, but the present experi- 

 ments indicate that 6.73 per cent is the limitirg value for sulphur 

 subjected to shear under ordinary hydrostatic pressure. Were 

 the bending to be carried further, or to be repeated, it seems 

 possible that a diminution of interstitial space would result. 



The author wishes to acknowledge his indebtedness to Mr. 

 C. E. Van Orstrand for helpful suggestions made during the 

 progress of the experiments. 



PHYSICS. — Combustion calorimetry and the heats of combustion 

 of cane sugar, benzoic acid, and naphthalene.'^ Hobart C. 

 Dickinson, Bureau of Standards. 



The uniform standardization of combustion calorimeters of the 

 Berthelot bomb type can best be accomplished by the combustion 

 of substances having a known heat of combustion. A critical 

 study of the stirred water calorimeter as used for bomb combus- 

 tion shows that the sources of error are mainly of four kinds, 

 viz., (1) temperature measurement, (2) evaporation, (3) lag effect, 

 i.e., failure of some parts of the system to reach a steady con- 

 dition with sufficient speed, and (4) uncertainties as to the 

 boundaries of the calorimeter, i.e., as to what portions should 

 be included in its heat capacity. It is shown that most of the 

 errors may be avoided by the use of resistance thermometers 

 and by the proper construction of the calorimeter and its jacket. 

 The cooling corrections for a calorimeter designed in accordance 

 with the conclusions reached can be made by a very simple pro- 

 cedure and with an accuracy corresponding to perhaps 1 part in 

 10,000 of the total amount of heat measured. 



A method of electrical calibration was used, which enables the 

 results of combustion observations to be expressed directly in 

 calories almost independently of the electric units, or, if the heat 

 capacity of the electric heating element used in the calibration 



1 The complete paper under the above title will be published in the Bulletin of 

 the Bureau of Standards. 



