THE INFLOEESCENCE. 



83 



only : thus the primary flower-stalk or axis ends in a flower ; 

 beneath this arise not two branches, one on either side, as in a 

 dichasium, but one only, this one being terminated by a flower like 

 the primary branch and giving off a tertiary branch as before 

 (fig. 153). The flower-stalks here appear to be opposite to the 



Fig. 152. 



Fig. 153. 



irm 



Fig. 154. 

 IT] "IT HFJff W 



o 



Figs. 152-154. Diagrams illustrating the centrifugal development of cymose inflorescences. 

 Fig. 152 a globose dichasial cyme ; fig. 153 a sympodial scorpioid cyme, the dotted lines 



indicate the suppressed branches ; fig, 154 a corymbose cyme. One of the lateral 



branches at III is abortive. 



bracts ; but the bract in this case belongs not to the flower-stalk 

 immediately opposite to it, which is a primary formation, but to the 

 secondary flower-stalk which springs from its axil. 



The subsidiary flower-stalks are sometimes developed all on the same 

 side when the inflorescence becomes curled from the greater growth on 

 one side than on the other. Such cymes are called scorpioid cymes (fig. 

 151). At other times the subsidiary pedicels or flower-stalks are de- 

 veloped alternately, first on one side and then on the other, when the in- 

 florescence has a zigzag shape. When the main rachis is a sympode (p. 39), 

 and the flowers, instead of being all on one or on two opposite sides, are dis- 

 posed spirally, the term helicoid cyme is given. In these forms of cyme one 

 of a pair of peduncles is generally systematically suppressed, and this 

 happens successively on one side of the main rachis of the inflorescence, or, 



