224 
alkaline solution of the tannic acid, and not to a change in 
the original yed/ow colour, since the intensity of the red varied 
as the amount of the tannic acid, and not as that of the yellow 
colour. On the contrary, when prepared in very different 
ways, or from leaves in very different states, if we find that 
all the reactions are equal when the original depth of the 
yellow is made equal, it is probable, if not certain, that the 
substance which gives the dark compound with the ferric 
salt, is itself of ‘yellow colour. When any well-marked 
absorption-band occurs in the spectrum, such methods are 
much simplified, and I have found that by carefully carrying 
them out they yield such uniform and intelligible results as 
to inspire great confidence in the conclusions. Some cases 
no doubt present great difficulties, but by care and per- 
severance they may be generally tolerably well overcome. 
By adopting such processes I have been able to prove that 
there are several different yellow colours not belonging to the 
tannic acid sub-group, some of which give spectra with 
tolerably well-defined absorption-bands when dissolved in 
alcohol, and oxidized by means of nitrite of potash and hydro- 
chloric acid. They also differ in the extent to which they are 
changed by the addition of ammonia; but, in applying that 
test, they must always be dissolved in alcohol, since the pre- 
sence of tannic acid might at once give rise to an abnormal 
depth of colour by oxidization, rapidly increasing by exposure 
to the air, if the colour were dissolved in water—a change 
which takes place far more slowly when alcohol is used. One 
of the most common kinds of chrysophyll is that which may 
be obtained from the yellow leaves of the elm, poplar, and 
mauy other trees, and one which gives the most characteristic 
spectrum, when oxidized, is that found in the yellow leaves 
of some plane trees. Much, however, remains to be learned, 
and the reactions are interfered with by the presence of tannic 
acids. The different kinds of this latter sub-group may often 
be distinguished by exposing to the air their aqueous solutions 
treated with carbonate of soda, or, in other cases, with a small 
piece of magnesium wire. Some also give very well-marked 
differences when their solutions in alcohol are treated with 
nitrite of potash and citric acid, and kept for a few hours. 
We thus get fine, well-marked, red, yellow, green, blue, or 
brown, accordizig to the particular kind under examination 5 
but, though some of them are of much interest in other re- 
spects, yet only three or four are of much importance in con- 
nexion with my present subject, and these will be described 
in the sequel. 
5. The phaiophyll group comprises a number of more or 
