Leaves in Autumn, 275 



ties of the chromule. This would easily explain the phenomena 

 presented by certain leaves, as those of Arum bicolor, which exw 

 hihit the three colours, red, yellow, and green, at once ; tho£(e 

 of the Tradescantia discolor^ which present a beautiful red co- 

 lour at their under surface, while the upper is green; and 

 there may, in fact, be obtained from these different parts, chro- 

 mules of different colours, the yellow and red passing ta green 

 by the action of potash, &c. .! 



Having found that the colouring part of the leaves might, 

 with very slight modifications, present the varied tints of green, 

 red, yellow, and their mixtures, it became interesting to find 

 out whether, agreeably to the analogy which the observations of 

 botanists have demonstrated between the various organs of 

 plants, such as the leaves, the calyces, the corollas, and their 

 appendages, the same colouring principle that was met with in 

 the leaves might also be found in the flowers. 



It was at first easy to find in the calyces the green chro- 

 mule, such as it presented itself in the leaves ; and, on taking 

 as a medium the coloured calyces of Salvia splendens, I ob- 

 tained from them, by means of alcohol, a substance of a beauti- 

 ful red colour and resinous nature, having all the characters 

 which the chromule of reddened leaves presented to me. Like 

 these leaves, it was rendered green again by the alkalies, became 

 red a second time by the addition of an acid, was insoluble in 

 oils, &c. Passing to the petals of the flowers of Salvia splen- 

 dens, and the portion of the stem which supports the flowers, 

 and which is red like them, I found the same product. The 

 petals of red geraniums, Bengal roses, asters, &c. treated by the 

 same means, all yielded, as their colouring principle, the red 

 chromule, and the flowers remained semitransparent and colour- 

 less. All the yellow flowers which I examined also presented a 

 chromule of that colour, which was rendered green again by al- 

 kalies, &c. 



The white flowers, the small number at least which the ad- 

 vanced season permitted me to procure, appeared to contain a 

 slightly yellowish chromule, modified in its colour by some na- 

 tural process, which it would be necessary to examine after- 

 wards. The reddish-blue flowers, such as those of the st«ck« 

 gilliflower, yielded a tint, at first rosy, then purplish, and 

 leaving a residuum of a fine violet colour. The flowers of a 



