100 James Small on 



" action current," l)ecause it is at an angle to the vertical. This 

 action current flows ui'jwards along the under side of the rootlet 

 and back along the upper side. At the point where the rootlet, 

 B, joins the main axis. A, we have therefore the possibility of a 

 leakage of the current from the main root into the rootlet, 

 towards the apex of the latter. This leaking current would tend 

 to flow against the action current in the under side of the rootlet, 

 and in the same direction as the action current in the upper side 

 of the rootlet. The leaking current would, therefore, tend to 

 augment the effects of the action current in the upper side and to 

 diminish the effects in the under side of the rootlet, thus 

 neutralising the unequal effects of the action current and prevent 

 ing curvature. 



The strength of the " action current " is known to vary 

 directly as the sitie of the angle with the vertical ; its strength at 

 the point of junction would obviously decrease the longer the 

 rootlet (i.e., the more distant the battery or apical meristem). 

 The strength of the leak from the main root would vary similarly 

 with the distance of the junction from the tip of the main root. 

 The angle at which the leak just balances the unequal effects of 

 the action current will be the only stable position for the lateral 

 rootlet. This "angle of balance" will increase with any increase 

 in length of the rootlet, because a stronger " action current " 

 from the meristem, involving a larger angle will be required to 

 give the same strength of current at the junctioyi of root and 

 rootlet. This angle will in the same way decrease with any 

 increase in the distance of the junction from the main root apex, 

 because the leaking current will then be weaker and the vmequal 

 effects of the action current in the rootlet will cause downward 

 growth until the " leak " is again strong enough to neutralise the 

 effect of the action current. In other words, the angle varies 

 directly as L ( = length of lateral rootlet) and inversely as D 

 (= distance of junction from main apex). Since the strength of 

 the action current varies directly as the sine of the angle, the 



fraction -p.- should vary as the sine of the angle, unless other 



