MIS CELL A NEC US. 1 1 1 



brilliantly scarlet. In RicJiardia africana, the so- 

 called Ethiopian lily, the spathe, surrounding a 

 spike of very degraded achlamydcous flowers, is 

 pure white, and very attractive. AmJierstia nobilis, 

 Brornelia piiigiiin, several species of Salvia, and 

 many other exotics, have handsome bracts, which 

 add greatly to their beauty. 



The fact that in such cases flowers do not develop 

 new petals from bracts or leaves, but acquire instead 

 coloured calyxes or involucres, goes to prove the 

 validity of the view with which we set out, that 

 petals are really altered stamens, not altered leaves 

 or sepals. For if they could once be developed from 

 leaves, there would be no reason why they should 

 not be developed from them here again. But if they 

 were developed from stamens, and then lost in these 

 instances, we could easily understand why the plant 

 could not afford to waste any more of its now 

 diminished number of stamens for purely attractive 

 purposes, and so was forced to pour the necessary 

 pigment for alluring insects into the other surround- 

 ing organs. In other words, on the Wolfian principle, 

 there would be no reason why flowers with petals 

 should not appear sporadically among monochlamy- 

 deous families ; on the principle here advocated, it is 

 quite clear why stray entomophilous species, de- 

 veloped from these degraded types, should have 

 coloured calyxes, instead of coloured petals. 



Among highly-developed, or succulent plants, the 

 calyx and bracts often tend to assume colours like 

 those of the petals, as do also the peduncles and the 

 stem. Cases occur in Ajitga rcptans, Echiuvi, Scdiim, 

 and Rinncx^ among British plants ; and more notice- 



