NATURAL HISTORY OF PLANTS. 



Aquilegia vulgans. 

 Fig. 6. Fig. 7. 



Flower. Longitudinal section 



of flower. 



stated above, an oval flattened anther, whose cells are attached to 

 the two sides of a vertical linear connective ; but this connective is 



sometimes laterally expanded into 

 a hooked spur, resembling the 

 spurs of the petals : and still more 

 frequently each anther assumes 

 the appearance of a small green 

 leaf, or coloured petal. In many 

 flowers, too, the stamens are me- 

 tamorphosed into spurred petals 

 fitting one within another. The 

 carpels are equally susceptible of 

 all these modifications. 



These carpels are normally 

 closed (fig. 5).' Each consists of 

 an elongated unilocular ovari/, surmounted by a narrow sti/Ic, bearing 

 along the whole length of the inner angle a longitudinal groove. 

 At the slightly dilated summit of the style, the thickened edges of 

 this groove are covered with a large number of projecting stigmaiic 



papilla3. On opening the back of 

 the ovary, we find in the inner 

 angle of the cavity a double pro- 

 jecting cord, or placenta, which 

 supports on each half a vertical 

 row of nearly horizontal ovules, 

 those of one row touching those 

 of the other by their inner 

 borders. Here each ovule has a 

 projecting ridge, or raphe. The 

 micropyle or hole, through which it is impregnated is placed near 



Aquilegia vulgaris. 

 Fig. 9. Fig. 10. 



Transverse section Fruit, 



of ovary. 



FORT {Insiit., 428) has enumerated as so 

 many species all these varieties, due almost con- 

 stantly to the influence of cultivation, and which, 

 besides, may be seen in flowers of every tint — 

 pink, white, or bluish. He further describes 

 Columbines that are variegated, punctated, with 

 pendulous flowers, with erect flowers. 1)e Can- 

 DOLLE has studied the structure of many of these 

 monstrous forms {Mem. de la Soc. d'Arcueil, 

 iii. 393-396). ISIany other anomalies have been 

 noted, amon^ others by (^los (Bull. Soc. Bot., 

 iv. 160), A. Tassi (ibid. viii. 394), and by 

 ourtielves (Adnnsonia, iv. 17-lR;, &c. 



' It may happen accidentally, especially in 

 gardens, that the carpels remain 

 open, and spread out like leaves. 

 We tlien usually see bodies on each 

 margin that represent ovules, some- 

 times nornial,8ometimes transformed 

 into tongues, or leaflets, of very 

 variable size and form (fig. 8). This 

 anomaly has often been pointed 

 out in Aconite, Lariispiir, and 

 many otlier genera, especially in 

 those which have multiovulate 

 carpels. j,^^_ g 



