128 JOUENAL OF THE ROYAL HORTICULTURAL SOCIETY. 



petals ; the five stamens have their anthers, c, pressed down upon the stigma, 

 and pour out their pollen-tubes while still within the cells as in h, 

 penetrating the stigma d, concealed beneath them. As the process of 

 pollination and fertilisation is thus concealed, such fiower-buds are called 

 cleistogainous {i.e. " concealed union "). 



By comparing the details in fig. 61 with those in fig. 62 the degra- 

 dations of the former will be readily seen. Thus, the calyx in 62 a 

 is much larger than in 61 a. The corolla has five petals, one being 

 spurred for receiving honey, 62 h ; 61 has no corolla. Two stamens (62 c) 

 have honey-secreting appendages, three having none (62 d). In 61 c the 

 connectives of all five anthers grow into broad spoon-like processes ; no 

 stamens secrete honey. In 62 e the style is prolonged into a beak-like 

 stigma, to pick up the pollen from the insect. In 61 d the style is very 

 short, having the blunt stigma tucked under the anthers. 



In common Plantains [Plantacjo species) we find that degradation 

 has passed from some ancestral insect-visiting forms (since they retain 

 the corolla) to wind-fertilising conditions. The anthers are on long 



filaments, easily dispersing the pollen, while the stigmas are elongated 

 so as to readily catch it. 



Grasses are in the same condition. They probably descended from 

 some form of Monocotyledon which possessed a perianth, for they still 

 have a rudiment in two lodicules, generally; but sometimes three are 

 present, as in bamboos ; but grasses are now mostly wind-fertilised. 



The question may now be asked — ^Yhat are the causes of degradation 

 in, or the unmaking of flowers ? ^Ye may suggest the following as 

 perhaps worthy of consideration. 



AYhen a flower passes from an insect-fertilising condition to a self- 

 fertihsing one, it is presumably due to the neglect of the insects, and the 

 loss of the stimulus which kept up the flower to its proper standard. In 

 their absence it reverts, more or less, to the hud-state of the flower, 

 sometimes expanding, but sometimes not, as we have seen in the 

 Shepherd's Purse in the one case, and cleistogamous flower-buds of the 

 Violet in the other. When it reverts to wind-fertilisation it is reassumin» 

 what appears to have been the ancestral condition of flowers, so far, at 

 least, as can be gathered from Gymnosperms. 



^ 



Fig. 62. — Details of flower of Violet. 

 a, a sepal ; h, the anterior petal with 

 a spur ; c, one of the 2 front stamens, 

 with a honey-secreting tail, which is 

 included within the nectary ; d, one 

 of the 3 posterior stamens without 

 nectaries ; e, pistil. 



Fig. 63. — Vertical section of flower of 

 Mare's-tail, showing 1 carpel with 

 inferior ovary containing 1 seed, with 

 1 stamen. 



