8 IF. Ftrrel — Law of Thermal Radiation. 



the observations through a considerable range of temperature, 

 and for short ranges it becomes almost entirely indeterminate. 

 The value of m, therefore, as determined from (5) or (9) with 

 values of A thus determined, cannot be relied upon as being 

 any more than a very rough approximation to the heat radia- 

 tion into empty space from a unit of surface at the temperature 

 of 0° C. 



7. With the value of A= 1*592, as given in the preceding- 

 table, and the values of i\ c and a in §5, we get from (5') with 

 the temperature r =0, m= 0-7188 of a calorie as the rate per 

 minute with which heat is radiated from each square centi- 

 meter of the surface of glass at the temperature of 0°. Now 

 with this value of m we get from (2), putting r =0, 



(12) H 100 — H = 0-7188(1-O082 ,on — 1) = 0-9092 



for the difference between the values of H in (1) at 100° and 

 at 0°. 



Again, in the other form of expression of the law of radia- 

 tion, with the value of A = 0'730 from the last column of the 

 preceding table, and the values of r, c, a above, we get from 

 (9 ; ) for the temperature T = 273, in which case q =l, m— 

 - 3296. And with this value of m we get from (7), putting 



2o = l, 



(13) H 100 -H = 0'3296(^) 42 -l) = ()-8926. 



The value of H 100 — H for glass has been obtained experi- 

 mentally by Lehnebach by the method of ice calorimetry with 

 apparently great accuracy.* His value is 0'0152 where the 

 second is the unit of time, or in our notation, the same as that 

 used by Dulong and Petit and Stefan, it is 0'912. This value 

 does not differ much from either of the values above, which 

 are also for glass. In obtaining the values above it is seen that 

 the value of r=S cm enters into the computation in the expres- 

 sions of (5') and (9'), and it is doubtful whether Dulong and 

 Petit's glass bulb was exactly a sphere with a radius of 3 cm , and 

 so there is some uncertainty with regard to these values. 



Lehnebach obtained the same value of H 100 — H for both a 

 bare and blackened glass bulb, and so it would seem that the 

 radiativity of glass at 100° is equal to that of lampblack. This 

 does not accord with some other experiments, and so this is a 

 matter which perhaps needs still further research. If the 

 radiativities are the same, then this value of H 10o — H applies 

 to both a lampblack and bare glass surface, at least at high 

 temperatures. Stefan reduced these values obtained from bare 

 glass to a lampblack surface by dividing by 0*88 the assumed 

 relative radiativity of glass with reference to lampblack. 



* Pogg. Ann., cxlvi, 497, 1875. 



