198 EXPERLMEXT STATION. [Jan. 



(1) differences in soil, (2) differences in aspect or exjx)siire, 

 (3) some inherent quality of the tree_, (4) the influence of 

 stock. ^ 



That the first two of these cause difference no one will dis- 

 pute, but there are nianj variations which can hardly be ex- 

 plained by differences in soil or site. It has been assumed by 

 numy that variations in productiveness arise from within the 

 tree, and arc transmissible. We know of no direct evidence to 

 support this view. Inheritable variation in color and form has 

 appeared in certain varieties. The Collamer, Banks and pos- 

 sibly (Jauo apples are instances of the former, and a probable 

 case of the latter has been reported by the writer.- Whether 

 the sliiiht differences in form and size reported here are trans- 

 uiissible by bud is by no means certain. We are of the opinion 

 that they are not, for it seems possible to explain these and the 

 other variations in productiveness, not attributed to soil and 

 site by reference to a different cause. 



Waugh has shown that in plums different stocks produce 

 marked modification in the trees grown on them.'"' Apple stocks 

 do not differ as widely as do the plum stocks, above referred to, 

 but the observed differences are also less marked. Every apple 

 tree of a named variety is growing on a stm*k of a different, 

 unnamed variety, i.e., a seedling. These seedlings differ to a 

 considerable degree. May not the slight differences observed 

 between individual trees of a variety, growing under apparently 

 similar conditions, be largely due to the influence of the seed- 

 ling root ? We know of no direct evidence to support this view, 

 but to us it seems a more promising theory than that of indi- 

 viduality of the different buds. 



If this supposition is true, it is probable that the production 

 of the most desirable trees of a given variety would be favored 

 by growing on a particular known root; thus the Baldwin 

 grown on ro(»ts of S]W. Wonltby or Siberian Cvnh might be an 

 especially desirable tree, while if grown on Tolman or Iving ^ it 

 might be less desirable. Different soils and localities might be 



1 There are, of course, large seasonal fluctuations in productivonoss due to conflition<i pec\iliar 

 to the different years. These are not considered in tliis (liseussioii. 



2 See Report Massachusetts Experiment Station, 22, Part IT., p. 1S7. 

 ' Report Massachusetts Experiment .Station. 21. Part II.. p. 174. 



'' The varieties mentioned are random select ionsfor illustration. There is no reason to believe 

 that they would influence the Baldwin as indicated. 



