272 BOTANICAL GAZETTE [OCTOBER 
ber of instances which they investigated that anatomical changes 
resulting in greater strength go hand in hand with the develop- 
ment of the fruit. These changes have been attributed both to 
the strain which is caused by the weight of the developing fruit, 
and to greater growth activity which is due to the carrying of the 
greater amount of building material which the fruit demands. 
KELLER (16) affirmed, however, that the fruit-bearing axis is not 
strengthened by strain or fruit-bearing in itself, but that the 
bending of the stem from the orthotropic to the plagiotropic 
position causes the anatomical changes. It is evident that these 
observations cannot throw any light upon the stimulatory effect — 
of a tension or a compression acting alone. An experiment by 
VécutING (34), however, does have a more direct bearing upon 
the question. He found that by removing the flower buds he could 
greatly retard the development of the mechanical tissue; the plants 
were sunflowers and fruiting stems of the Savoy cabbage. He also 
found that by stress brought about by weighting he could not cause - 
the mechanical elements to reappear (the discussion of this stress 
is taken up in another place), but when he grafted a scion upon @ 
decapitated stem, the cambium at once began to form the mechani- 
cal elements again, although the weight of the scion was equal to 
only a fraction of the weights which he had placed upon similar 
plants without result. He sums up the results of his experiments 
in the following words: 
Aus dem mitgetheilten folgt, dass der ontogenetische Gang der Gewebe- 
differenzierung von inneren, correlativen Verhaltnissen beherrscht, dass die 
Bildung der einzelnen Gewebeformen nicht einfach durch das Bediirinss 
bestimmt wird. Die Auslésungs-Theorie geniigt hier nicht. 
This experiment shows very clearly that correlation may have much 
more to do with mechanical tissues than mere stress or strain. 
Correlation is also seen in many trees in which one-sided growth 
in the wood is correlated with a greater development of the root 
system on that side, or with the greater number of branches and 
the correspondingly greater leaf surface of that side. KNY (17) 
asserted that one-sided leafing of a tree must be associated with 
one-sided increase in the growth of the trunk. HARTIG (no) Te 
ported that this is the most common cause of eccentric growth in 
