352 THE SURVIVAL OF THE UNLIKE. [xXI. 



in clusters of from five to ten, and yet, except in the 

 crabs, apples are usuallj- borne singly; that is, most of 

 the flowers fail. And trees which bloom full rarely 

 average even one fruit to the cluster. Small wild 

 apples are frequently' borne in clusters, and there 

 is every reason to believe that originally all the 

 flowers normally set fruit. With the enormous 

 development in size and other qualities of fruits, 

 the plant is unable to use all its flowers. I am in- 

 clined to think, however, that these extra flowers 

 serve a very useful purpose in supplying pollen to 

 those which chance to set, for not only is the sup- 

 ply of pollen in the individual flowers probably 

 becoming less with the improvement of the apple, 

 but it is also probable that more is needed to incite 

 the enormous increase in size over that of the infe- 

 rior aboriginal apple. What is true of the apple is 

 true in various degrees of all orchard fruits, even 

 of the cherry ; and it is most graphically shown in 

 the tomato. It is here worthy of remark, also, that 

 probably the chief reason why the bush fruits, as 

 blackberries, raspberries and currants, do not more 

 rapidly improve in size is because all the flowers 

 upon the clusters still set fruit. All these instances 

 show that cultivation or improvement seriously inter- 

 feres with the mutual relationships of the sexes, 

 and this disturbance or unbalan(?e is likely to increase 

 rather than diminish. 



But it now transpires that not only are some 

 plants impotent or infertile with themselves, but in 

 some cases all the plants of a given variety are infertile 

 among themselves. Thus it has long been known that 

 the Wild Goose plum is usually unproductive when 



