A. DeCandolle on the Life and Writings of Vaucher. 165 



moisture is the agent of fecundation, without the concurrence 

 of which fecundation could not take place. This humour ordi- 

 narily resides in the gland which supports the ovary, where it 

 is often very visible ; but sometimes it escapes our observa- 

 tion and we find it on the stigma, which it impregnates at the 

 period of flowering, or on the torus, as in all the Rosacea ; 

 sometimes also it resides in the glands situated either at the 

 base of the stamens, or on some parts of their surface, or in 

 short in the stigma from whence it proceeds ; in a word, there 

 is no floral organ where it may not be met with and give signs 

 of its existence*." 



In this theory there are two assertions : one long ago ad- 

 mitted, namely, that the viscous liquid of the surface of the 

 stigma is necessary to fecundation ; the other new, the de- 

 monstration of which will be asked for, viz. that the nectar 

 secreted in other parts of the flower, particularly on the torus, 

 is an agent of the same kind as the viscosity of the stigma. 

 Until the present time the nectar produced by the torus and 

 by the glands has been considered as an excretion, that is to 

 say, as a substance produced to disembarrass the plant of the 

 matters become useless in the series of vegetative functions. 

 Those who saw in this product a liquid subservient in some 

 manner to fecundation, did not go so far as M. Vaucher; 

 they supposed at most, with Conrad Sprengelf? that the 

 transportation of the pollen on to the stigma must be made by 

 insects, and that the nectar, as well as the stains of lively 

 colour of many corollas, served to attract into the flower these 

 living agents of vegetable reproduction. 



The use of the nectar in fecundation may be demonstrated 

 in various ways. It has been sometimes attempted to sup- 

 press the secreting organs, but the conclusions thence deduced 

 appear to me uncertain, and M. Vaucher probably thought so 

 too, for he has not made trial of this kind of experiment. In 

 fact, one of two things must happen ; either fecundation takes 

 place notwithstanding the mutilation, or it does not take 

 place. In the first case, it may always be feared that it has 

 been produced by the honeyed moisture from the general sur- 

 face of the torus, or of the floral organs, which no mutilation 

 could hinder, and of which there are many examples in the 

 ordinary course of vegetable life. If, on the contrary, fecun- 

 dation has not been effected, we may suspect that this is 

 owing to the wounds being too deep, and to the extraction of 

 the glands necessary rather to the life of the flower than to its 

 fecundation. In natural history, observation leads us further 

 than experiment. M. Vaucher, wishing to prove the physio- 



* Vol. it. p. 521. f Das entdeckte Geheimniss, &c. 



