84 THE COLOURS OF FLOWERS. 



there we still find purple or white central florets, as in 

 Petasites vulgaris, the butter-bur ; but even then we 

 get closely related forms, like Titssilago farfara, colts- 

 foot, which have declined to yellow. The smallest and 

 most debased species, such as Solidago virga-aiLvea, 

 golden rod, Tanacetitm vtilgare^ tansy, and Senecio 

 vulgaris^ groundsel, have all their florets yellow and 

 similar ; unless, indeed, like GnapJialiwn and Filago, 

 cud-weed, Artemisia absinthium, wormwood, or Xa7i- 

 thiimi stnimaritim, burweed, they have declined as 

 far as colourless or green florets, in which case they 

 must be considered under our next head, that of 

 Degeneration. On the other hand, the larger and 

 better types of Corymbifers began a fresh progressive 

 development of their own. In many Senecios, Imilas, 

 Chrysanthemums, they produced yellow ray florets, 

 similar in colour to those of the disk. In Chrysau- 

 themum leucanthejnum, AntJiemis cotula, Matricaria 

 inodora, &c., these rays, under the influence of a 

 different type of insect selection, became while. In 

 the daisy they begin to show signs of pink, and 

 in \}iiQ Astei's, Cinerarias, &c., they have become lilac, 

 purple, and blue. Complicated as these changes 

 seem, they must yet have taken place two or three 

 times separately in various groups of Corymbifers, 

 for example in the Asteroidece, the AntJiemidece, and 

 the Scnecionido!. 



The Ligulates were again developed from yellow- 

 rayed Corymbifers by the conversion of all the disk 

 florets into rays. Appealing for the most part to 

 very large and varied classes of miscellaneous insects, 

 they have usually kept their yellow colour ; but in a 

 few cases a fresh progressive development has been 



