27 
When first prepared, the alkaline solution of phenolphthalin is 
perfectly colorless. On long standing, however, it gradually acquires 
a faint color, due to the oxidation of traces of the compound by 
atmospheric air. In no case, however, has this color proven of 
sufficient intensity to interfere with the use of the reagent in testing 
for blood, and if much work of thia kind has to be done the solution 
would be used up long before it became unfit for use on account of 
this change. Utz (191) also found solutions of phenolphthalin in 
sodium carbonate to be stable. 
(2s) Alkaline 'phenolphthalin containing hydrogen peroxide . — The 
principal difficulty which has been encountered in the preparation 
of alkaline solutions of phenolphthalin containing small amounts 
of hydrogen peroxide is that it sometimes happens that the hydrogen 
peroxide alone oxidizes the alkaline solution of phenolphthalin, and, 
as already pointed out, it was this difficult}^ which caused me to 
undertake a reinvestigation of this subject in relation to the use 
of phenolphthalin in testing for blood. In our earlier work on the 
peroxidase activity of the blood in health and disease, however, 
Amoss and myself (80) had no special difficulty in preparing a color- 
less peroxidase reagent, but when I came to repeat some of this 
work later considerable difficulty was experienced in the prepara- 
tion of a colorless peroxidase reagent. The cause of the trouble was 
found to be in the distilled water employed in the preparation of the 
reagent. With certain specimens of distilled water a pink to red 
coloration was produced in the reagent just as soon as the h 3 ^drogen 
peroxide was added. With other specimens of distilled water, how- 
ever, the reagent remained colorless after the addition of the hydro- 
gen peroxide. The spontaneous oxidation of certain solutions of 
alkaline phenolphthalin containing h}^drogen peroxide is probably 
due to the presence of very small amounts of copper. Bourquelot 
and Bougault (23) have pointed out, for example, that it is not 
rare to encounter distilled water which blues guaiacum at 50° C. 
Similarly, Carnelly (35) found distilled water to dissolve small amounts 
of metallic copper, and in different samples of distilled water Barth 
(Nessler, 119) found amounts of copper varying from none to 0.015 
gram of copper acetate per liter of water. Ascarelli (7) found copper 
sulphate to react with benzidin at a dilution of 1 to 1,000,000. 
It therefore occurred to me that this difficulty could be overcome 
by redistilling the water used in the preparation of the reagent, in 
glass vessels. This was tried and found to give good results. As a 
matter of fact, since I began using water which had been once redis- 
tilled in glass vessels no difficulty has been experienced in the prep- 
aration of colorless alkaline solutions of phenolphthalin containing 
hydrogen peroxide in suflficient amount to accomplish its oxidation 
under proper conditions. 
