166 A. DeCandolle on the Life and Writings of Vaucher, 



logical function of the nectar, has relied on detailed studies of 

 the progress and series of the phenomena of the inflorescence. 

 His conclusions are rather inductions than proofs. Let us 

 quote some of the phenomena of which he speaks. Accord- 

 ing to him, it is from them collectively that a well-established 

 conviction results. 



" Of the four petals of Corydalis tuberosa, the two exterior 

 petals, placed one above the other, exactly close the flower, to 

 which they serve as a calyx ; their superior margins are free 

 and reflexed ; the two interior ones, lateral and applied ex- 

 actly one against the other, inclose the anthers in a pouch or 

 quadrangular hood (capuchon) and do not separate, although 

 a narrow slit is left between them by which the air penetrates. 

 Fecundation takes place a considerable time before the deve- 

 lopment of the flower ; the anthers lie upon the stigma, which 

 is a vertical and fringed disc, entirely covered with the yellow 

 fecundating dust ; the nectary which grows from the torus is 

 a greenish body, filled with a honeyed liquor which proceeds 

 from a well-defined pore, and is diffused in the cavity of the 

 superior petal ; thence it insinuates itself by the slit which 

 separates the two interior petals and penetrates to the anthers 

 and the stigma, which it thoroughly moistens. This humour 

 is not designed to attract flies, since it is contained in a closed 

 sac." The grains of pollen fall on the stigma and burst, in 

 order that their elongated pollinic tubes may penetrate into 

 the interior. It seems, indeed, that the abundant humour de- 

 scribed in this particular case must bring about the pheno- 

 menon of the rupture of the pollen, if the facts are precisely 

 such as the author has indicated. The advanced season has 

 not enabled me to verify them. 



In another chapter he says, that " the anthers of the Helle- 

 borus fcetidus open outwardly and in such a w T ay that their 

 yellowish pollen falls into the melliferous tubes*, filled at this 

 epoch with the honeyed moisture, and gradually as they [the 

 anthers] are matured from the circumference towards the cen- 

 tre, they raise themselves one above the other, and thus convey 

 their pollen to the stigma, which can hardly be fecundated ex- 

 cept by the emanations of the nectaries : the little nectariferous 

 tubes are seen quite open, powdered with the granules which 

 absorb the limpid humour." Here, I confess that the in- 

 duction appears rather indirect. If the grains of pollen fall 

 to the bottom of the flower, on the torus, or into special cavi- 

 ties in certain plants, we can scarcely conclude that these 

 same grains contribute to the fecundation. Does not an im- 

 mense quantity of pollinic grains fall upon the earth around 

 * The bilobed and tubular petals, so remarkable in the tribe of Hellebores. 



