38 
hydrogen peroxide reagent. The residue gave much the stronger 
test, however, indicating some adsorption of blood. Some of the 
mixture was then boiled and filtered. One c. c. of the filtrate with 
2 c. c. of the alkaline phenolphthalin-hydrogen peroxide reagent gave 
only a faint pink color after standing, whereas a very small amount 
of the residue gave with 2 c. c. of the reagent a deep purplish red color 
after standing a few minutes. Hence the adsorptive power of soil for 
blood is greatly augmented by boiling. A control test was carried out 
using 25 c. c. of water and 1 gram of the same sample of soil, but no 
blood. Neither the filtrate nor the residue, before or after boiling, 
gave a color with the alkaline phenolphthalin-hydrogen peroxide 
reagent. 
Hence by means of the alkahne phenolphthalin-hydrogen peroxide 
reagent we were able to detect 1 part of blood in 9,600 parts of water 
containing 1 gram of soil, and from such a solution the blood is com- 
pletely adsorbed by the soil on boiling. In testing soil for blood, 
therefore, it is advantageous to boil the soil with twenty to thirty 
times its volum.e of water, hlter, and test the conduct of the residue 
toward the alkahne phenolphthalin-hydrogen peroxide reagent. 
Should such a test turn out to be negative, it is reasonably certain 
that no blood is present in the soil, or if any, not more than traces. 
Should a positive test be obtained under these conditions, it is 
reasonably certain that blood is present, and this should be con- 
firmed by other tests. 
BLOOD IN MILK. 
In certain diseased conditions it becomes a matter of some im- 
portance to be able to recognize small amounts of blood in milk. 
(See ^^The Significance of Leucocytes and Streptococci in Milk,” by 
William Whitfield Miller, Article No. 13, Bulletin No. 41, Hygienic 
Laboratory; ^Alilk and Its Relation to the Public Health”.) 
Some preliminary observations were made on cow’s milk contain- 
ing moderately large amounts of mouse’s blood. Such milk, raw and 
after boiling, gave excellent tests for blood with Adler’s benzidin 
test (2); indeed, this reagent seems to be admirabty suited to the 
recognition of blood in milk. On the other hand, with phenolphthahn 
the color developed much more slowly than with aqueous solutions 
of blood, indicating the presence of substances which, like animal 
tissues generally, greatly retard the oxidation. 
With milk the effect of acid coagulation was studied with the view 
of determining the distribution of the blood between the curd and the 
whey and the possibility of the adsorption of the blood pigment by 
casein. In order to determine the effect of acid coagulation, 2 c. c. 
of the fresh milk containing mouse’s blood was coagulated by the 
addition of two or three drops of dilute acetic acid. The curdled 
