202 Ewart and Mason- J ones . — The Formation of 
latter appearing, according to him, on the side of an erect tree subjected to 
compression by the action of the wind. 
Even here, however, although the tree may still remain upright, every 
time it is bent to one side by the wind it is subjected to geotropic action, 
and a geotropic stimulus, if frequently and rapidly repeated, will produce 
a geotropic curvature in a root, however short the individual periods of 
exposure may be. 
To investigate these points more fully, a series of experiments were 
begun upon the main axes and lateral branches of plants of Cupressns 
nutkaensis , C. Lazvsoniana , Finns contorta, and P. Cembra , in May, 
1905. 
The stems were forcibly bent into circular or elliptical forms and 
fastened in such a fashion that the geotropic and pressure effects were not 
equally distributed, the former influencing the horizontal portions, whereas 
the compression was restricted to the inner surface of the ring. 
The stems were sawn off and examined in November and December, 
and the distribution of the red wood is shown by the dark line on the 
appended figures. 
The specimens represented by Figs. 3 and 4 were shaded from above, 
whereas the others were exposed to normal illumination ; and it will at 
once be seen, by comparing the figures, that they give absolutely no 
evidence of the existence of the ‘ heliotropic ’ action which Sonntag 
postulates. 
It is also evident that it is almost solely the action of gravity, and not 
that of pressure, which has in these specimens determined the formation of 
the red wood, for this appears mainly on the under sides, whether they are 
subjected to compression or tension. 
The change from one side to the other is especially well shown in 
Fig. 1, but can be traced in the others also, the side of the erect portion 
on which the thin connecting layer of red wood develops being probably 
determined by the loop being slightly inclined to one side. 
It will be noticed that in all cases there is a tendency for the formation 
of red wood to spread into the vertical regions, where it fades away. 
This is, however, the result of the gravitational stimulus spreading 
from the upper and under surface, where it is directly perceived, to the 
neighbouring vertical portions, and in this way preventing the weakening 
effect of an abrupt transition from one type of wood to the other. 
The formation of red wood begins first on the under surfaces and then 
spreads laterally, so that in the outermost vertical or nearly vertical regions 
to which it extends only a thin superficial layer of the year’s annual ring 
consists of the red tracheides, the inner part being white wood. 
Thus, on the under side of the uppermost point of the curvature of 
a branch the red tracheides averaged thirty-two deep, towards the side of 
