CONCLUSION. 



37 



The writer is fully convinced that persons having hand-pollinations 

 to make with unfamiliar flowers, or with flowers with which their pre- 

 vious work has been unsatisfactory, will find it profitable to make 

 some small experiments in order to determine the method to which 

 the flowers respond with a large percentage of fruits. These experi- 

 ments can usually be made with the proper pollen and flowers so that 

 the fruits will contain seed of the desired kind. 



While attention is constantly being called to the many adaptations 

 of flowers for accomplishing various purposes, such as cross-fertiliza- 

 tion, it might be well to attempt to discover if some adaptations for 

 preventing premature pollination have not been brought about by 

 natural selection. That many flowers are admirably formed to success- 

 fully prevent premature pollination is certain, but that such device- 

 have been brought about for this purpose will perhaps never be cer- 

 tain. Proterogenous flowers that open or unfurl in blooming usually 



„ 6 c 



Fig. 1.— Flowers of Sabbalia angidaris: {a) Appearance while pollen is being shed; (6) Stigma and 

 anthers avoiding each other; (c) Receptive stigma and deciduous anthers. (From drawings by 



E. E. Lower.) 



do so when their pistils have reached a receptive condition. The pis- 

 tils of such flowers are perfectly protected from pollen from all sources 

 until they are receptive. PI. IV, fig, 6, shows a fully open, recep- 

 tive begonia flower, while tig. 7 shows the two calyx lobes securely 

 closed over the young pistil of a similar but younger flower. Proter- 

 androus flowers open while the pistils are immature, but even here 

 modifications are met with which insure the pistils against premature 

 pollination. The stigmatic surfaces are often folded together, thus 

 preventing pollination even though the immature pistil be exposed. 

 Fig. 1 shows the interesting behavior of flowers of Sabbatia angidaris; 

 a shows the condition of the flower when the anthers discharge their 

 pollen, two petals and one stamen having been cut away in order to 

 show the young pistil, which at this stage has no stigmatic surface 

 exposed. Instead of growing straight out and thus coming in contact 

 with the drying anthers, the pistil invariably grows out to one side, 

 passing between the filaments of the stamens. The stamens at this 

 stage begin to bend to the side of the flower opposite the one occupied 



