232 
and others to a very dull brown. From these alcohol ex- 
tracts a brown colour, soluble in water, similar to that pro- 
duced artificially. The leaves of the oak and the Spanish 
chestnut contain the same chrysophyll as the elm mixed with 
the gallotannic acid found in various galls occurring on oaks. 
This kind of tannic acid differs from all others in giving rise 
to several well-marked products. When dissolved in alcohol 
and oxidized by means of nitrite of potash and hydrochloric 
acid, a substance is formed, which, when the solution is only 
slightly acid, is of splendid red colour. ‘This belongs to 
Group C, and is not at all changed by adding sulphite of 
soda. When made slightly alkaline, this red substance 
rapidly passes into a colour belonging to Group A, which is 
of a splendid deep blue colour in a solution containing con- 
siderable excess of citric acid, and is then immediately altered 
by sulphite of soda to a pale yellow, the well-marked ab- 
sorption-band in the orange being completely removed. No 
such substance is formed in the case of the tannic acid used 
by photographers, prepared from Chinese galls, which is 
undoubtedly quite a different modification. When oxidized 
under other conditions, gallotannic acid gives rise to a deep 
brown colour, similar to that found in the changed leaves, 
and this difference between it and quinotannic acid readily 
explains why the leaves of the oak and Spanish chestnut 
change from yellow to brown, and not to the brighter tints 
seen in those of the beech and a few other trees. The leaves 
of the Acuba japonica of our gardens contain a rather 
singular kind of yellow colour belonging to the chrysotannin 
group. When kept dry at the ordinary temperature it gra- 
dually changes into a very insoluble dark brown colour, pre- 
cisely like that seen in the faded leaves; and it differs 
from the more typical kinds of tannic acid in being oxi- 
dized by the oxygen of the atmosphere when in an acid 
solution, but agrees with them in giving a dark colour with 
ferric salts. The chrysophyll colours which do not turn 
dark with persalts of iron seems usually to pass by oxidi- 
zation into brown substances, and it is very probable that 
similar dull brown tints result from the oxidization of a 
great variety of compounds met with in plants. On the 
whole, so far as my present experience enables me to decide, 
the brighter and redder browns of autumn are mainly due to 
the previous existence of quinotannic acid, modified by the 
presence of other substances, but there seems reason to 
believe that in different plants and in varying conditions it 
gives rise to several distinct colouring matters, differing in 
solubility, but being all more or less red, orange, or brown. 
