LA IV OF PROGRESSIVE COLOUR A TION. 2 1 



reason to believe that the modification runs from any- 

 one colour toward any other ? Apparently there is. 

 The general conclusion to be set forth in this work is 

 tlie statement of such a tendency. All flowers, it 

 would seem, were in their earliest form yellow ; then 

 some of them became white ; after that, a few of them 

 grew to be red or purple ; and finally, a comparatively 

 small number acquired various shades of lilac, mauve, 

 violet, or blue. So that, if this principle be true, such 

 I flower as the harebell will represent one of the 

 most highly-developed lines of descent ; and its 

 ancestors will have passed successively through all the 

 intermediate stages. Let us see what grounds can be 

 given for such a belief. 



Some hints of a progressive law in the direction of 

 a colour-change from yellow to blue are sometimes 

 afforded us even by the successive stages of a single 

 flower. For example, one of our common little 

 English forget-me-nots, Myosotis versicolor, is pale 

 yellow when it first opens ; but as it grows older, it 

 becomes faintly pinkish, and ends by being blue like 

 the others of its race. Now, this sort of colour-change 

 is by no means uncommon ; and in almost all known 

 cases it is always in the same direction, from j'cllow 

 or white, through pink, orange, or red, to purple or 

 blue. For example, one of the wall-flowers, CJiciran- 

 tJiiis cliaincelco, has at first a whitish flower, then a 

 citron-yellow, and finally emerges into red or violet. 

 The petals of Stylidiuni fruticosuDi are pale yellow to 

 begin with, and afterwards become light rose-coloured. 

 An evening primrose, Oenothera tctraptera, has white 

 flowers in its first stage and red ones at a later period 

 of development. Cobcca scandcns goes from white to 



