230 
converted by oxidization into a number of coloured products, 
which, though differing from one another in detail, have all 
certain optical characters in common. ‘The properties of the 
resulting oxide depend very much on the conditions under 
which oxidization takes place, and no doubt many of them 
are mixed colours. ‘Thus, for example, when oxidized by 
means of nitrite of potash and some acid, they often differ 
entirely according as the colour is dissolved in water or in 
alcohol, and in both cases they may also differ according as a 
weak acid like citric, or a strong one like'hydrochloric, is used. 
They also sometimes vary further when oxidized by means 
of hyponitric ether, and still more so when the solution in 
water or alcohol is exposed to the air for a few days, after 
having been made alkaline or acid in various degrees. Much 
remains to be learned before we can understand the cause of 
these differences and relationships, but they appear to me to 
be of much interest in connection with the formation of the 
different colours of the petals in varieties of the same plant, 
for they show that variations in the intensity of the oxidizing 
process may give rise to a considerable number of very 
different colours ; and, since the chemical changes taking 
place in leaves are manifestly related to their vital state, it 
seems by no means improbable that the production by culti- 
vation of very different colours in flowers naturally of nearly 
uniform character may be due to some kind of abnormal 
change in their vital powers, thus modifying chemical affini- 
ties. In connection with this I might refer to the difference 
in the colours of certain flowers, according to the time of the 
year at which they happen to bloom, but must leave the 
consideration of such questions, and return to the more im- 
mediate subject of this communication. Though there isa 
material difference in different cases, yet, on the whole, the 
various substances of the chrysotannin group changed by 
oxidization from more or less pale yellow to orange or red 
compounds, and some of these readily pass into blue or 
green, or into brown modifications of greater stability, appa- 
rently by undergoing some molecular change independent of 
further oxidization. These facts may be expressed by saying 
that, by combining with a certain amount of oxygen, the 
absorption of light is much lowered from the blue towards 
the red end of the spectrum, and that there is often a ten- 
dency to pass into a further modification, characterised by 
absorption at the red end. Such results of experiments 
made with the different colours in a separate state are com- 
pletely in accordance with what occurs in natural foliage, 
not only agreeing in general character with fading leaves as 
