68 Various Tints of Autumnal Foliage. [January, 



gentle evaporation is left in the state of a dry gum, which in 

 some cases remains almost unchanged for months or even 

 years. It must, however, be borne in mind that, as thus 

 prepared, the erythropbyll must necessarily contain a 

 variable amount of the colours described below, and it is no 

 doubt to their presence that some of the reactions are due, 

 which both myself and others have referred to the red colour 

 itself. Thus, for instance, when they are slightly oxidised, 

 the broad absorption-band is lowered towards the red end, 

 and, by further oxidation, the colour becomes more or less 

 orange-yellow, just in the manner that the colour of dark 

 grapes is changed into that of new wine, and this in time to 

 that of very old, as described by me in a paper already 

 cited, but I am now inclined to believe that this further oxi- 

 disation, which destroys the main absorption-band, extend- 

 ing over the yellow and green, does really completely destroy 

 the colour of the red substance, and that the more or less 

 orange-yellow is due to the oxidisation of a pale yellow 

 colour previously obscured by the deeper red. This fact is 

 of considerable importance in the subject before us, since it 

 explains why intensely red leaves fade to almost, if not 

 exactly, the -same tint as those of the like kind which were 

 previously not at all red ; that colour being so completely 

 destroyed as to produce no effect on the tints subsequently 

 developed. 



The fourth group of colours is composed of those soluble 

 in water and aqueous alcohol, but insoluble in bisulphide of 

 carbon, which have a sufficiently decided gold-yellow colour 

 to justify my distinguishing them by the term chrysophyll 

 group. They vary somewhat in tint from a little more 

 yellow, to a little more red, than yellow ochre. They are 

 made darker and more orange by oxidisation, and thus 

 are in an unoxidised condition as compared with the 

 colours of the next group. In order to prepare them, the 

 leaves should be boiled in alcohol, and after evaporation to 

 dryness at a gentle heat, the soluble portion re-dissolved in 

 water. I have so far met with at least four different 

 species, distinguished by the spectra which they yield on 

 partial oxidisation. The most satisfactory method is to 

 dissolve some of the colour in a small quantity of water, 

 dilute this with alcohol, and then to add a little nitrite of 

 potash and hydrochloric acid. In some cases this gives 

 rise to one or more well-marked absorption-bands, and 

 changes the colour from yellow to pink. In others no bands 

 are developed, but the colour is altered from pale yellow to 

 deep orange-red. On evaporating to dryness, we obtain 



