8 BOTANICAL GAZETTE [JANUARY 
especially fleshy ones which have a very large proportion of 
parenchyma tissue, contract by the absorption of water. Kohl 
agrees with Noll that at the beginning of the curvature there is 
no measurable difference in the thickness of the cell walls on the 
two sides, and that as the curve progresses the difference comes 
rather by thinning of the convex membranes than by a thicken- 
ing of the concave membranes. Hence Wortmann’s view can- 
not be correct. This view does not provide for a contraction 
of the concave side, a fact observed by former experimenters 
and confirmed by Kohl. He finds in Pisum that the cells of the 
concave side may shorten 19.4 per cent. of their normal length. 
Noll’s theory as first enunciated provides for the shortening of 
the concave side only by a passive compression, though, as modi- 
fied later, it allows a greater elasticity of the walls on that side. 
In further support of his theory Kohl cites the fact that 
grass nodes, in curving, often break on the convex side with 
transverse clefts. This could not take place if the curve was 
produced by a greater extensibility on that side. This counts 
against Noll’s theory, but is in perfect harmony with Kohl’s 
theory which makes the lengthening of the convex side wholly 
a passive stretching. He shows by measurements in both stems 
and roots that the cells of the concave side become shorter and 
broader, and that those of the convex side become longer and 
narrower than normal. 
Kohl thinks his theory accounts for the plus curve observed 
by Noll, though I cannot see the force of his reasoning. If 
increased turgor contracts the cells, decreased turgor ought to 
expand them. Noll showed (50, pp. 49-50) that this is not the case. 
In a series of experiments Kohl cut transverse notches 2™ 
apart in stems to the depth of the vascular bundles, and laid 
them in a horizontal position, some with the notched side upper- 
most, others with the notched side downward. The latter always 
curved more rapidly than the former, but Noll has shown that 
the difference was due to the traumatropic effect of the cutting. 
It is an interesting and perhaps a significant fact that when these 
organs curved water was forced out of the notches. 
