INSECTS AISU FLOWERS, 79I 



plant, flowers of which the one kind are accessible to foreign pollination, the 

 others exclusively to self-pollination. This is the case, for example in Oxalis 

 acetosdla, where the small flowers concealed in the soil appear when the large 

 flowers are already ripening their fruits ; further in Impatiens noli me iangere, 

 Lamium amplexicaule, Specularia perfoliata and many species of Viola ( V. odorata, 

 (latior, canina, mirabilis, &c.), Ruellia clandestina, in several Papilionacese 

 \Amphicarpaa, Voandzeid), Commelina bengalensis, Sec. Where in these cases the 

 large typically developed flowers are fertile, crossings with other flowers of the 

 same species can and must occur at least occasionally in the course of the generation, 

 and then the small aborted self-fertilising flowers appear more as an accessory 

 arrangement, the purpose and importance of which is quite unknown ; but it is 

 remarkable and apparently contradictory to the general rule that the large typical 

 flowers have occasionally a tendency to be barren (species of Viola) or are 

 entirely infertile (Voandzeia), so that the reproduction in such cases depends 

 chiefly or alone on the self-fertilising abnormal flowers. 



In other cases, as in most Fumariacese, Canna indica, Salvia hirta, Linum 

 vsilatissimum, Draba, verna, Brassica Rapa, Oxalis micrantha and O. sensitiva, the 

 pollen, in virtue of the position of the sexual organs, comes (according to 

 Hildebrand) directly on to the stigma of the same flower, and is also effective ; 

 but in such cases, since the flowers are visited by insects, an occasional crossing 

 with other flowers is at least not avoided. Even among the Orchidese, where 

 the most astonishing mechanisms for the avoidance of self-fertilisation occur, 

 the case is found in Cephalanthera grandiflora, according to Darwin, that the 

 pollen-grains send their tubes out from the anther into the stigma; from 

 Darwin's experiments, however, the yield of good seeds is smaller when the 

 plants are left to this self-fertiUsation alone, than when they are exposed to 

 crossing, to pollination with foreign pollen, by the aid of insects. 



It comes out in the fertilisation of flowers more than in any other case, how 

 closely the structure of the organ is adapted to perfectly definite vital relations 

 of the plant, and to the fulfilment of perfectly definite functions. Each plant 

 has its own special mechanisms to ensure the transference of the pollen to the 

 Stigma of another flower. It is thus impossible to say much of a general nature ; 

 but the following points may be noticed. 



In the flrst place it is to be observed that the insects effect the transference 

 of the pollen involuntarily and unknowingly, as they seek the nectar of the -flowers 

 which is (for this purpose) secreted deep down in the base of the flower; flowers 

 which are not visited by insects, and the Cryptogams which do not need them, 

 excrete no nectar. 



The position of the nectaries, usually deep down in the base of the flower, 

 as well as the size, form, position, and often also the movements of the floral 

 organs during the period of pollination, are always so calculated that the insect, 

 and this often of a particular species, must assume a particular position and make 

 definite movements when seeking the nectar, so that the pollen-masses hang on 

 to its hairs, feet or proboscis, and it then, by assuming the same positions in 

 another flower is bound to wipe it off on to the stigma. In the case of dichoga- 

 mous flowers, the movements of the stamens and the style or stigmatic lobes 



