ox THERMAL RAHIATIOX IN ABSOLUTE MEASURE. 
GIG 
Taking the common logarithms of the numbers in column 4 we have the fifth column 
of the table on p. 615, 
The differences between the successive logarithms of deflections gives col. 6. To 
obtain the proper ernissivities. these numbers must be multiplied b}^ (M X c)/(300 X S); 
where M is the modulus for reducing Neperian logarithms to common logarithms, 
c the capacity for heat of the globe, S the cooling surface, and 300 the interval of 
time in seconds. (See p. G04 of the paper.) The value of Mc/300 S is 4*323 X 10 “ ^ 
The multiplication I perform by means of Crelle’s table. 
Lastl}^, to find the difference of temperatures of globe and bath correspondmg to 
the ernissivities, I use the logarithms of the deflections in col. 5, and to each of these 
add the logarithm of the multiplier corresponding to the first term of the thermo¬ 
junction formula. In the formula of the note of p. 603 of the paper, this multiplier 
would be 3.^ and the additive logarithm 0'4701 — 1. In the present case 
April 2, 1890, the multiplier is 0*27976 ; and the additive logarithm 0*4468 — 1. 
Taking the anti-logarithms of the numbers so found, I obtain the difference of tem¬ 
peratures in Centigrade degrees between cooling body and surroundings. 
