111.] VARIATIONS AMONGST BRANCHES. 91 



themselves, for he instructs his budders to cut buds only 

 from the topmost shoots of the nursery rows, in order 

 that he may grow straight, vigorous trees; and every 

 farmer's boy knows that the reddest and earliest apples 

 grow on the uppermost branches, and his father al- 

 ways tells him that he should never select cions from 

 the center or lower part of a tree. Every skillful horti- 

 culturist will tell you that the character of the orchard 

 is determined very largely by the judgment of the prop- 

 agator in selecting cions. To select out the extreme 

 forms of these variations and to attempt to explain bud- 

 variation by them is exactly like selecting the extreme 

 types of seminal variations, and, by ignoring the lesser 

 ones and the intermediates, to attempt to build thereon 

 a theory of the variation of plants. If you ask me why 

 it is that the nectarine was produced upon the branch of 

 a peach tree, I will ask you why it is that -nectarines have 

 also been produced from peach seeds. The answer to 

 one answers the other. It is true that bud -variations, 

 if we use that term, as we logically must, to denote all 

 variations between phytons, are commonly less marked 

 than seed variations, but this is only because the con- 

 ditions of origin and environment of the phyton are less 

 varied than those of the seedling. The phytons origi- 

 nate from one parent, not from two; and they all grow 

 in very like conditions. But I am convinced that, when 

 we consider the plant individual in the light of evolu- 

 tion, the bugbear of bud -variation vanishes. 



A good proof that bud -variation and seed -variation 

 are one in kind is afforded by the fact that selection can 

 be practiced for the improvement of forms originating 

 by either means. Darwin was surprised, as he says, to 

 "hear from Mr. Salter that he brings the principle of 



