PROPAGATION BY BUDDING, GRAFTING, ETC. 19 



stamens and pistils are in the same flower, and the chances of acci- 

 dental mixture from other trees, become very small unless affected by 

 insects, which, becoming thickly dusted with powder from one flower, 

 plunge into the recesses of another, and affect a cross fertilization. 

 Where many varieties grow in one garden, in close proximity, cases 

 of promiscuous intermixture are constantly occurring, which can be 

 developed only by raising fruit from the seedlings. 



"In the annexed figure of the pear blossom (Fig. 1), the five cen- 

 tral organs a, are the pistils ; the upper extre- 

 mity of each is the stigma. The surrounding 

 T) thread-like organs, , are the stamens, surmount- 

 ed by the anthers. The anthers are little bags 

 or cases, filled with the pollen, or fertilizing 

 dust. When the flowers open, the anthers 

 burst, and discharge the pollen on the stigma, 

 which operates on the embryo fruit at its base. 

 " The production of new varieties is greatly facilitated by cross-im- 

 pregnation, or by fertilizing the pistil of one variety with the pollen 

 of another. This was performed with great success by Knight. 

 Selecting two varieties, while yet early in flower, and before the 

 anthers had burst and discharged the pollen, he 

 cutout with a fine pair of scissors all the stamens, 

 leaving the pistils untouched (fig. 2). When the 

 stigma became sufficiently mature, which was in- 

 dicated by its glutinous surface, he transferred the 

 pollen of the other sort on the point of a camel's- 

 Fig. 2. hair pencil. The fruit, thus yielded, was un- 



changed ; but its seeds partook variously of the nature of both 

 parents, and the trees growing from them bore new and intermediate 

 varieties. 



*' For the success of such experiments, several precautions are requi- 

 site. The flower must be deprived of its stamens before it has fully 

 expanded, or before the anthers have already burst and scattered their 

 dust ; the pollen must be procured from a bursting, or fully matured 

 anther, when it will be dry and powdery ; the stigma must be inocu- 

 lated as soon as it becomes adhesive or glutinous, otherwise it may be 

 fertilized from another source, and then the intended pollen cannot 

 possibly take effect. For a stigma once inoculated, cannot be inocu- 

 lated again. It is safest, where practicable, to force the trees by artifi- 

 cial heat into flowering a few days earlier than others, so as to 

 be secure from accidental inoculations of pollen floating in the air ; 

 and to prevent its spread by bees, to apply a temporary covering of 

 gauze. A want of attention to these minutiaB, has led some experi- 

 menters to fancy they had obtained crosses, when they had only 

 natural seedlings." 



Budding. This process of propagation, as well as that by grafts^ 



