CHAP. V. FUNCTION OP REPRODUCTION. 26'l 



the vague conjectures of the old philosophers; and it 

 is now clearly established that the two innermost floral 

 whorls, the stamens and pistils, are the organs essen- 

 tial to the fertility of the seed. In the case of double 

 flowers where all the stamens have assumed the condi- 

 tion of petals, seed is never produced ; but if the pistil 

 be perfect, it may be supplied with pollen from another 

 plant of the same species, and will then ripen its ovules. 

 Some apparent anomalies are recorded among the various 

 experiments which have been made to prove the necessity 

 of the action of the pollen in securing the fertility of the 

 seed. The females of certain dioecious plants have ma- 

 tured their seeds although they were carefully excluded 

 from the action of the stameniferous individuals ; but 

 in some of these cases, this was probably owing to the 

 fact that dioecious plants are frequently partially mo- 

 noecious, and that a stameniferous flower is here and 

 there developed on the fertile plants, which may have 

 furnished sufficient pollen to set the fruit. Accord- 

 ing to some recent experiments, however, the universality 

 of a law which establishes the necessity of the pollen's 

 action has been rather shaken, unless there be some 

 error which it is difficult to account for. If they are 

 correct, it seems to have been proved that hemp and 

 a few other annual dioecious species are capable of ri- 

 pening their seed without the action of the pollen having 

 taken place. Even if the fact should be satisfactorily 

 established it will in no way disprove the general neces- 

 sity of the pollen's action, or the sexual distinctions of all 

 phanerogamous plants. But such isolated exceptions 

 may possibly be considered analogous to the case of the 

 Aphides, in which insects a single impregnation is suf- 

 ficient to enable several generations to become fertile. 

 But after all we have such marvellous accounts of the 

 distance to which the pollen may be carried and yet 

 preserve its proper influence, that it seems hardly pos- 

 sible to feel quite certain that the plants in question 

 may not have been fertilized from others growing in 

 the neighbourhood. It is stated that in the year 1505 

 s 3 



