242 A. DeCandolle on the Life and Writings of Voucher, 



generally described in the true Ranunculacea. He adds that 

 the C. integrifolia has introrse anthers, which is perfectly 

 accurate. " The exterior stamens open first, and spread them- 

 selves backwards (s'ecartent), after having scattered their 

 whitish powder, which falls abundantly to the bottom of the 

 flower. The others take their place and elongate in their 

 turn. The stigma is always an enlarged ligula (languette), 

 papillose and turned outwards. I am not aware that I per- 

 ceived any nectariferous organ ; but I have often remarked, 

 especially in the C. Vitalba, cirrhosa, &c, honeyed drops ad- 

 hering to the filaments of the stamens In those spe- 

 cies whose styles are hairy, the hairs without doubt serve to 

 retain and to preserve the pollen for a longer time ; but in the 

 Viticellce with drooping flowers, where the styles, almost al- 

 ways smooth, are placed in close juxtaposition, the anthers, 

 which are exactly lateral, turn back the edges of their walls 

 (parois) in order to cast their pollen on the stigmas ; and gra- 

 dually as they shed their pollen they fall back to make way 

 for others situated more interiorly, which also turn themselves 

 backwards in a similar manner. In the Flammulce with erect 

 flowers, whose extrorse lateral anthers do not elevate them- 

 selves above the stigmas ordinarily arranged in close juxta- 

 position, the fecundation appeared for a long while inexpli- 

 cable, or, if I may venture so to speak, badly contrived, until 

 I discovered the singular manner in which it was effected. 

 The sepals are felt-like, and have their interior covered with 

 moist and glandular hairs, which are the agents in effecting 

 it ; for this purpose they receive the pollen immediately from 

 the anthers, which with their filaments fail after shedding 

 their pollen; at the same time the stigmas, wonderfully papil- 

 lose, cast themselves down, and turn about in different direc- 

 tions for the better reception of the molecules of the pollen ; 

 as may be seen in the Flammula, the Vitalba, and especially 

 in the orientalist 



The flowering of the Garidella Nigellastrum is described in 

 the following manner : — " At the epoch of fecundation the sta- 

 mens are curved over the centre of the flower, and their anthers 

 cover the as yet undeveloped stigmas (stigmates non encore 

 conformes). In this situation they open their walls (parois) 

 from bottom to top, so that their pollen, always adherent, is 

 situated exteriorly and out of all contact with the stigmas. It 

 can then only scatter itself on the swollen and glutinous hairs 

 at the summit which cover the bilabiate petals or nectaries, 

 and by which it is retained. It is then afterwards, and only 

 when the anthers in withering have uncovered it, that the bi- 



