340 
Journal of Agricultural Research 
Vol. XXVI, No. 8 
suggested by Walker (46), it is evidently not responsible for the reaction 
of rancid fats in the Kreis test, nor identical with substance K. 
Dihydroxyacetone. —In most of the reactions leading to the forma¬ 
tion of glyceric aldehyde the isomeric dihydroxyacetone is also formed. 
It is said to yield brilliant colors with phloroglucin in presence of concen¬ 
trated sulphuric acid. It was prepared by the method of Fischer and 
Tafel (15), which was originally supposed to lead to the formation of 
glyceric aldehyde, but which has since been shown by Wohl and Neuberg 
(45) to lead to the formation of dihydroxyacetone only. 
Ten parts of glycerin were mixed with 60 parts of water and 35 parts 
of crystalline sodium bicarbonate. The mixture was brought to a tem¬ 
perature of io° C., after which 15 parts of bromin were added from a 
burette. At the end of 30 minutes the mixture was acidified and the 
liberated bromin reduced with sulphur dioxid gas. After standing over¬ 
night the solution was tested as follows : 
A few drops were treated with concentrated sulphuric acid and solid 
phloroglucin, an intense red color, indicative of the presence of dihy¬ 
droxyacetone, being obtained. 
A few cc. were treated with an aqueous solution of phloroglucin, but 
no white precipitate was obtained in the course of several hours. The 
absence of glyceric aldehyde was thus established. 
A few drops were tested with phloroglucin-hydrochloric acid in the 
usual manner. A white precipitate was formed immediately, the pre¬ 
cipitate changing to a dirty purple color in the course of the day. When 
the test was made after the addition of hydrogen peroxid the same 
result was obtained. 
Substance K is therefore not dihydroxyacetone. 
Methyegeyoxae. —According to the literature (30) methylglyoxal 
results from the distillation in vacuo of aqueous solutions of dihydroxy¬ 
acetone in presence of sulphuric acid. 
The solution of dihydroxyacetone remaining from the preceding experi¬ 
ment was brought to a volume of 500 cc., treated with 100 cc. of con¬ 
centrated sulphuric acid, and distilled at a pressure of about 20 mm. 
A colorless distillate having an odor of burned sugar was obtained. The 
solution was tested as follows: 
A few drops of the solution were treated with about 1 cc. of con¬ 
centrated sulphuric acid and a few cc. of a 1 per cent solution of phloro¬ 
glucin. An intense red color developed immediately. 
A few drops of the solution were treated with 5 cc. of concentrated 
hydrochloric acid and the Kreis test completed as usual. A straw color 
was obtained in the aqueous phase. The same result was obtained in 
another test performed after the previous addition of hydrogen peroxid. 
In a third test, in which a larger quantity of hydrogen peroxid was used, 
the same result was again obtained, except that the original straw color 
deepened very slowly to a brownish red. The spectrum of this latter 
color bore no resemblance to that of the color obtained in the Kreis 
test on rancid fats and on mixtures of acrolein and hydrogen peroxid. 
Whether or not methylglyoxal be formed in the reaction between 
acrolein and hydrogen peroxid, it is not the same as substance K in 
which we are now interested. 
Oxyacroeein and maeonicdiaedehyde. —According to the litera¬ 
ture (7) these substances are known only in aqueous solution and are 
tautomeric. An intensely red color is said to be obtained when their 
aqueous solution is treated with ferric ehlorid. No such reaction could 
