COLOURS OF FLOWERS. 179 



an interesting little book on The Colours of Flowers,'^ in 

 which he expresses his belief that the first colour on depart- 

 ing from the primitive green was yellow. When Ave remem- 

 ber that the spore-cases and spores of Lycopodium, tbe 

 anther-cells of Cupressus, and the whole anther-scale of 

 Pinas and all the pollens of Gymnosperms are yellow, — again, 

 when we come to Dicotyledons and find the prevailing tint 

 of stamens is the same, — we gather probabilities in support 

 of that view. That Nature next introduced reds, and only 

 lately, so to say, succeeded in manufacturing blues, seems 

 probable from the comparative rarity of the last colour. 

 Moreover, when flowers individually change from one colour 

 to another as they develop from the bud to maturity, it is 

 always in that order — i.e. from reds to mauves or purples, as 

 in Echium, Fuhnonaria, etc., or even from yellows through 

 reds to purples, as in Myosotis versicolor, so that we still seem 

 to gather additional support to the theory. 



If, however, we ask what has caused these changes, we 

 are as yet in the dark. A few hints are attainable, and that 

 is all. Yellows and reds seem to be due to substances allied 

 to the oxidized products of chlorophyll in autumn leaves. 

 Again, chlorophyll grains on turning yellow in fruits (Lyciiim) 

 become angular, two or three pointed, and finally granular. 

 Tn the same way tbe yellow granules of petals {Cucurhiia) 

 resemble " amyloplasts," or starch-forming corpuscles. f 



The general conclusion one arrives at from various 

 observations is that the original change from the ancestral 

 green to, probably, yellow is correlated to the change of 

 function; but why the first colour was yellow, and why it 

 ever gave place to red or blue, is unknown. 



Supposing the yellow-green colour to have spread to the 

 adjacent parts which then attracted insects, as it does in 

 * la Nature series. f Sachs' Veg. Phys., p. 320. 



