sst« 



THE 



AMERICAN JOURNAL OF SCIENCE 



[THIRD SERIES.l 



Art. LY. — An Experimental Comparison of Formulce for 

 Total Radiation between 15° C. and 110° C. y by 

 W. LeConte Stevens. 



The earliest attempt to express the rate of cooling of a 

 heated body was that made by Newton and embodied in a 

 .simple formula, which states that this rate is directly propor- 

 tional to the difference of temperature between the radiating 

 body and the medium surrounding it. It has long been known 

 that this law is only approximately true when these tempera- 

 ture-differences are small, and that it is wholly inapplicable 

 when they are large. 



During the early part of the present century an elaborate 

 research on the measurement of temperature and the com- 

 munication of heat was made, by Dulong and Petit.* Such 

 instruments as the thermopile, the bolometer, and the galva- 

 nometer had not yet been invented. Their method was to 

 employ as radiating body a large thermometer whose bulb con- 

 tained more than a kilogram of mercury. This was placed with- 

 in an envelope which was kept at constant temperature and made 

 nearly vacuous as quickly as possible after the introduction of 

 the heated bulb. The rate of cooling was found to be a func- 

 tion of the temperature of the envelope, the latter being kept 

 constant during any given experiment, but varied for different 

 experiments. For a given difference of temperature between 

 the bulb and its enclosure it was found that the rate of radia- 

 tion increases very nearly in geometrical progression while 

 that of the enclosure increases in arithmetical progression. If 



* Anrtales de Ghimie et de Physique, vol. vii, 1817. 



Am. Jour. Sci. — Third Series, Yol. XLIV, No. 264.— December, 1892. 

 29 



