48 CORRELATION AND GEOTROPISM [CH. 



cases, and since this occurs even with short lengths 

 cut off the shoot, we seem to be justified in speaking of 

 a root-pole and a shoot-pole respectively ; yet there are 

 a few cases where the bent-down tip of a long arched 

 shoot appears to give origin to roots, and in some cases roots 

 can give rise to shoots. 



But the most striking result in such an experiment is 

 the sequel : w^hatever the position of the hanging mother 

 axis, the shoots put forth always turn up at the tip and 

 grow out at an angle which is acute to the vertical above 

 the insertion : the roots at the same time growing down 

 towards the ground. 



That these phenomena are in some way due to the 

 action of gravitation, is proved by experiments into the 

 detailed nature of which we cannot here enter, but 

 regarding which it may suffice to say that if we so 

 rotate the plant or organ concerned — on a slowly re- 

 volving instrument driven by clockwork, and known to 

 plant-physiologists as a Klinostat — so that every side of the 

 growing organ is for equal small periods equally exposed 

 to the vertical action of gravitation, no such constancy 

 of angular divergence is assumed : the organ merely 

 grows out in the direction it happens to be in on its 

 first emergence from the bud. But although gravitation 

 is a principal factor in determining the direction of growth 

 of a shoot from an axis which is a rest, the phenomenon 

 cannot be wholly explained by geotropism — the term 

 applied to this reaction. 



Take the case of almost any tree growing in the 

 normal position. The buds on the leader in most cases 

 point so that their tips form less than a right angle with 

 the parts of the axis above them : we may neglect for the 

 moment a few trees and shrubs from which the buds 

 stand out at right angles, as in certain Honeysuckles, and 



