170 
Suppose we alter the lengths of the columns and in a second 
experiment; we find, 
by cross multiplication and elimination of the common 
factor W 1 W 2 we obtain 
or as we may write it 
T’hen taking the logarithms of both sides and dividing by 
log k we get finally 
= i^2 "f" h 
In this way I proposed to test the law. 
I took, as standard tints, a smooth surface of BaS 04 
and another of BaS 04 and carbon, in the proportion of 10 
grms. of BaS 04 to 0006 grms. of carbon; these materials 
were intimately mixed and the powder reduced by pressure 
to a flat surface. The colorimeters used were glass cylinders 
covered externally with black cloth, the circular apertures 
at the bottom admitting light being 8 mm. in diameter. 
One experiment gave the following results. Length of 
column 2 2 ’2 c., standard tint Wi (BaS 04 ). I now attempted 
to get the same intensity when looking through the second 
cylinder at tint W 2 (BaS 04 + carbon). The mean of two 
trials gave 14’8 c. as the equivalent column. Hence we 
have the following results : 
I now made the length of the column over Wi 13 c. The 
equivalent column over W 2 was as the mean of two trials 
7'5. Hence 
From these experiments we get as the sum of h and U 29 ’7 
and the sum of U and U, 27*8. 
A second experiment gave 
Here h + = 28 ’45 and + A = 28-76. 
A third experiment gave 
WiP®-S= W2F0-65 
WiP'-» = W2F*‘' 
Here Lx + = 25-3 and q' + ^2 = 24*93. 
