Notes on the Corolla in the Composites. 31 
Helianthus annuus or Taraxacum officinale with the scape of 
Digitalis purpurea or Verbascum Thapsus. The economy then 
becomes obvious. Species in lower orders, such as the Ranun- 
culaceae and Caryophyllaceas may develop more seeds with a 
smaller output of corolla material, but they have not the advantages 
of developing one or two seeds in each fruit which are possessed by 
the higher orders. 
Colour of the Corolla. 
Apart from descriptions in the systematic literature, the colour 
of the corolla in the Compositse has not been the subject of 
research, except when species have been taken as examples in 
researches dealing with the relation of insects to flowers and a few 
scattered enquiries into isolated species. The distribution of colour 
in the various tribes, however, follows the lines of development 
already suggested to such an extent that it is worthy of consideration. 
Practically all authorities are agreed that blue is the highest 
colour, and Avebury (1, p. 308) states his opinion “that all blue 
flowers have descended from ancestors in which the flowers were 
green-; and that they have passed through stages of white 
or yellow, and generally red, before becoming blue.” Muller and 
Avebury disagree on the relative attractive power of white and 
yellow, but it seems probable that while moths are more attracted 
by yellow than by white, as Muller’s results show, bees may be 
attracted more by white than by yellow, as Avebury’s results prove. 
1 
Muller, however, includes bees in his placing of yellow above white 
in the power of attraction. Willis and Burkill (35) confirm in 
general Avebury’s scheme of colour values in regard to desirable 
insect visitors to flowers of the various colours. 
An interesting experiment is recorded by Daniel (10) who 
propagated a pure white variety of Chrysanthemum by cuttings for 
about seventeen years and, among other evidence of degeneration, 
he found that the colour of the corolla became a more or less 
greenish yellow. Muller and Hildebrand (24) have pointed out 
that blue flowers frequently show atavism and revert to lower 
colours and that the reverse is rare. De Vries (33) gives many 
examples of this among the Compositae. 
It is not clear whether purple, which from the Mendelian work 
on pigmentation seems to be compounded of red and blue, is of a 
higher type than blue and, as the entomological results seem to 
indicate the contrary, the point requires experimental enquiry. 
