its allies, which are such excellent dyes for wool and silk, give only
fugitive colors on cotton.

The main point which arrests our attention in connection with the
natural dyes seems to me to be the comparatively limited number of
fast colors. Very remarkable is the total absence of any really fast
yellow vegetable dye, and it is probably on this account that gold
thread was formerly so much introduced into textile fabrics. Notice
further the decided fastness of Prussian blue, especially on wool and
silk; while we cannot but remark the comparatively fugitive character
of vat indigo blue on cotton, and even on silk, compared with the
fastness of the same color when fixed on wool.

Now, let us turn our attention to the _artificial coloring matters_,
derived with few exceptions from coal tar products.

Here again we have two classes, "mordant dyes" and "direct dyes." Both
classes are somewhat numerous, but whereas the former may be
conveniently shown on a single diagram sheet, it requires a
considerable number to display the latter.

First let us examine the wool patterns dyed with the "mordant dyes."

We find there a few yellow dyes quite equal in fastness to those of
natural origin, or even somewhat surpassing them, e.g., two of the
alizarin yellows, viz., those marked R and G G W. Except in point of
fastness and mode of application, I may say that these are not true
alizarin colors, neither are they analogous to the natural yellow
dyestuffs, for they are incapable of giving dark olives with iron
mordants. Truer representatives of the natural yellow dyes appear,