222 
of their solutions in water treated with hydrochloric or sul- 
phuric acid. The relation of these two colours to each other, 
and to the development of the leaves, will be described in the 
sequel. 
4. The chrysotannin group. When I wrote my paper on 
the various tints of autumnal foliage, published in the 
‘ Quarterly Journal of Science,’ I hesitated to decide whether 
certain changes were due to the alteration of coloured sub- 
stances, or to the development of new colouring matters from 
compounds previously almost colourless. I had observed 
facts indicating that probably the tannic acid contained in 
some leaves might have considerable influence in giving rise 
to the autumnal tints, but had not had time and opportunity 
for clearing up that question in a satisfactory manner. Since 
then I have paid much attention to this point, in order to 
ascertain what connexion there is between tannic acid and the 
substances described by me as the chrysophyll group. If we 
compared some of the yellow colours found in leaves with the 
tannic acid used by photographers, prepared from Chinese 
galls, we might be led to conclude that they could not be 
looked upon as members of one group of substances; but 
further inquiry has shown that there are at least six entirely 
different kinds of tannic acid, some almost colourless, and 
others so yellow that I had included them in the chrysophyll 
group. Not only is there this gradual passage from one to 
the other, but the character of the different changes produced 
by the action of various reagents shows that they are so in- 
timately connected, that for the purposes of the present subject 
they must be classed together as one great group, the various 
members of which have more or less claim to be called tannic 
acid or chrysophyll. On the whole, then, it seems to me that 
it would be better to call it the chrysotannin group, and to 
confine the name chrysophyl/ to those members which do not 
strike a dark colour with persalts of iron. Those having this 
latter character may be considered to be more or less con- 
nected with the typical kinds of tannic acid. Some of these 
are very nearly, if not quite, colourless, whereas others are 
very decidedly yellow, even when their solution is acid, and 
still more so when it is alkaline. They are soluble in water, 
in alcohol, and in ether, but not in bisulphide of carbon. 
The spectra show a variable amount of absorption at the blue 
end, but usually no well-marked absorption-bands in their 
natural state. When oxidized and otherwise changed they, 
however, sometimes give bands in different situations, or they 
are distinguished by other characters. The intensity of their 
colour is usually much increased, and the absorption lowered 
