Comparison of Formulas for Total Radiation. 438 



•S = mT C-l) (4) 



This formula has the advantage of great simplicity and ease of 

 application. 



Almost simultaneously with the publication of Stefan's 

 paper in Vienna appeared an experimental research on the 

 temperature of the sun, by M. F. Rosetti, of Padua.* Rosetti 

 employed the thermopile and galvanometer for measuring 

 the radiation from a Leslie cube covered with lamp-black 

 and filled with water, which was stirred in order to secure 

 uniformity. For temperatures between 100° and 300° mer- 

 cury was substituted for water. In the liquid were immersed 

 the bulbs of two thermometers whose readings gave the mean 

 temperature at any selected moment. For still higher temper- 

 atures a copper ball was heated to redness, then exposed at a 

 fixed distance in front of the thermopile long enough to pro- 

 duce a galvanometer deflection, and immediately afterward 

 thrown into a calorimeter. The specific heat of copper being 

 known, it becomes possible to compute the temperature of the 

 ball at the moment of its immersion in the water. From 

 these and other experiments Rosetti was led to adopt the 

 empirical formula, 



S = aT a (T-T )-ft(T-T ) (5) 



Here a and h are two constants whose values for the particu- 

 lar instruments he employed were determined to be a = 

 0-00000335131, and 5 = 0-0636853. Rosetti considered the 

 first term of his formula to represent the heating effect if the 

 body were radiating in a vacuum, and the second term to rep- 

 resent the radiation of the air in which the body is immersed. 

 Comparing this with Newton's formula, 



S=a(T-T ), 

 it is seen that while Newton regarded the emissive power, a, 

 as independent of temperature, Rosetti regarded'it as propor- 

 tional to the square of the absolute temperature. 



The accordance between the results of experiment and of 

 calculation by means of this formula, as published by Rosetti, 

 are quite remarkable, though the experiments seem liable to 

 some criticism. Any assumptions regarding the uniformity of 

 temperature in a Leslie cube are subject to challenge, if the 

 water be not kept boiling. The calculations of temperature 

 from calorimetric measurements is usually affected with a large 

 probable error. For temperatures higher than those employed 

 in calorimetry Rosetti. used small disks of metal which were 

 rendered incandescent in the hottest flames at his command, 



*Annales de Cliitnie et de Physique, V, vol. xvii, 1879. 



