872 Wisconsin Academy of Sciences , Arts, and Letters. 
Liebermann and Kostanecki 1 undertook a study of a large num¬ 
ber of hydroxy anthraquinones in order to ascertain the re¬ 
lation between the number and position of the hydroxy groups 
and the dyeing properties of the compound. The result of their 
investigations may be summarized as follows: At least two 
hydroxy groups are necessary in order that the anthraquinones 
may become dye stuffs. This is shown by the fact that no mono¬ 
hydroxy anthraquinones have dyeing properties. Of the known 
dihydroxy anthraquinones, only alizarin with the hydroxy 
groups in positions 1 and 2 has strong dyeing properties. 
Hystazarin, which was not known at this time, 2, 3, dihydroxy 
anthraquinone, does, it is true, combine with mordants but its 
dyeing properties are weak and it is not satisfactory as a dye 
stuff. That the dyeing property of alizarin is not dependent 
on only one of the two hydroxy groups is shown by the fact that 
the monomethyl or mono ethyl ether of alizarin does not dye mor¬ 
danted fabrics. From these facts Liebermann draws the conclusion 
that in order to have dyeing properties the polyhydroxy anthra¬ 
quinones must have two of their hydroxy groups in positions 1 
and 2. 
All of the known trihydroxy anthraquinones which have the 
property of dyeing mordanted fabrics have two of their hy¬ 
droxy groups in positions 1 and 2 or in similar positions. The 
same holds true for the tetrahydroxy derivatives. Those with 
hydroxy groups in positions 1, 4, 1', 4', have no dyeing proper¬ 
ties whatever, and those with hydroxy groups in 1, 3, 2', 4', 
possess very weak dyeing properties. It is theoretically im¬ 
possible to have pent- and hexhydroxy derivatives in which two 
of the hydroxy groups are not connected to carbon atoms in 
position 1 and 2. All of the known pent- and hexhydroxy an¬ 
thraquinones are good dye stuffs. 
An exception to Liebermann’s rule seems to be found in chry- 
sophanic acid. The formula for chrysophanic most favored 
at present represents the compound as possessing two hydroxy 
groups in positions 1' -4', while none of the formulas consid¬ 
ered have hydroxy groups in positions 1 - 2. 
Although Liebermann has shown that all the hydroxyanthra- 
quinones which have dyeing properties, with possibly a few 
exceptions, have hydroxy groups in positions 1 and 2, he has 
* Ann., 240, p. 245. 
