THB INFLORESCENCE. 



113 



remains tolei-ably fixed in the same species, so long as it retains its wild condition, it is 

 apt to vary greatly when the same plant is cultivated. Thus, if we take the rose as an 

 illustration, we find that its normal number of petals is five, as in the hedge rose ; but, 

 when cultivated, the number vastly increases, until a " perfect " rose, in horticultural 



Fig. 207. A perfect E,ose, having nearly the whole of its stamens converted into petnls. 



language, should present to view nothing but petals (Fig. 207). Whence, then, has 

 the rose obtained its additional petals ? Not from new formations, since that would be 

 in opposition to the established law of development, but from a modification of other 

 organs which were originally formed for another purpose. This applies not only to the 

 corolla, but to every part of the flower ; and, as a further rule, it may be remarked, 

 that the parts so modified are usually, if not invariably, those which are naturally 

 placed higher on the stem than those into which they become transformed. Therefore 

 the petals are not produced from sepals, and sepals from bracts ; but, on the contrary, 

 the bract may assume the place of calyx, and the calyx that of corolla. The newly- 

 formed petals are thence the product of transformed stamens, or the parts of- fructifica' 



VOL. II. 7 



J 



