508 PLANT RESPONSE 



Curvature of grass haulms under gravity. — We shall 

 now take up the consideration of the curvatures induced in 

 grass haulms, laid horizontally, when growth had originally 

 been at standstill. It will be remembered that it was the 

 appearance of curvature under such circumstances in the 

 pulvinus of the grass haulm that gave the strongest support 

 to the theory, that negative gravitational curvature in general 

 was due to the increase in the rate of growth on the convex 

 side, rather than to its retardation from active contraction on 

 the concave. In the present case it was argued that since, 

 at the beginning of the experiment, the upper side of the 

 pulvinus was not undergoing growth, it was clear that growth 

 there could not be retarded. The curvature, therefore, must 

 be due to the induction of growth on the convex side, under 

 the stimulation of gravity. 



This misconception has arisen from the supposition that 

 all curvatures must be induced by differential growth. I 

 have shown, however, (i) that contraction takes place in a 

 stationary organ in response to stimulus ; (2) that the 

 unilateral stimulation of such an organ induces concavity ; 

 and (3) that retardation of growth in a growing organ is 

 itself the result of the contractile effect of stimulus. Now, in 

 a horizontally laid grass haulm in which growth has ceased, 

 the upper side — which we have found to be relatively the 

 more effective — will contract under stimulus of gravity 

 That this is the case is seen from the fact that this upper 

 surface is found to become actually shorter than it was 

 before. But, as regards the convexity of the lower surface, 

 the water expelled from the actively contracting upper 

 side will reach the lower, and the increased turgidity thus 

 produced is sufficient, as we have already seen, to re- 

 initiate growth in a dormant tissue. This explains the 

 renewed growth and convexity of the lower side. Thus the 

 curvature of the grass haulm cannot be held to support the 



at T)i will be greater than the corresponding sum of effects at P/ and D^. As 

 the result of this difference we shall have a right-handed or positive heliotropic 



