18 
Having been prepared three days previously and darkened 
by exposure during that interval to light. The containing 
cylinder was closely invested with black cloth and kept in 
a dark closet. On the morning of the following Monday I 
thought that it appeared not quite so dark as at first, and on 
the evening of the same day I thought it a little lighter than 
in the morning. After keeping it in the dark for a week I 
found that it had become much lighter, and on June 4th, 
when I examined it again, it seemed nearly as light as a 
freshly prepared solution; there was, however, a minute 
quantity of precipitate. 
From these results it is evident that in some cases special 
provision must be made to avoid needless exposure to light 
in quantitative determinations by colorimetry, or in studying 
the laws of the absorption of light passing through coloured 
solutions. 
I also made some experiments with chromate of potash. 
This I thought a stable salt suitable for experiments. 
Nevertheless some of the results were not satisfactory when 
one cylinder contained a solution which was several times 
stronger than the other. For instance : a standard solution 
was made containing 0’8 grams in 500 cub. c. of water. 
Another solution compared with this gave the following 
results : — 
A B C 
6*4 3-7 4-5865 
A repetition of the experiment gave nearly the same result, 
namely, 3-6 for the length of the column. 
It occurred to me that possibly, when potassic chromate 
is diluted, there may be liberated a minute quantity of chromic 
acid, which would increase its absorbent power ; this might 
