1900 | MECHANISM OF ROOT CURVATURE 47 
Another series of control experiments was used in which 
straight roots were split into halves, and the water content of 
the separate halves ascertained as before. The stimulated roots 
were all branded on the side of the tip facing the hilum of the 
seed, so that side was always the convex side of the curve. The 
straight unstimulated roots were always split in the correspond- 
ing plane and a record kept as to which was the side next to the 
hilum. So the two series were exactly comparable. The lots 
were nearly all treated at the same time and under exactly the 
same conditions as the curved ones. A total of fifty-nine 
straight roots were split and weighed in six different lots. The 
variation on the side next the hilum, corresponding to the con- 
vex side of the curved roots, was from 87.72 per cent. to 92.51 
per cent., and on the opposite side from 87.63 per cent. to 92.49 
per cent. In three lots the excess was on the hilum side and in 
three lots on the opposite side. 
Nothing has been proved as to the absolute increase or 
decrease of the water content in stimulated roots, but it is shown 
beyond a reasonable doubt that there is a relative increase of 
water in the convex side of curving roots. This does not prove, 
however, that the greater amount of water on the convex side 
increases the size of the cells. It may simply mean that the 
increased size of the cells enables them to hold more water with 
the same turgor pressure. Indeed, the increased amount of 
water must necessarily reduce the concentration of the cell sap 
and therefore the osmotic pressure, unless there is at the same 
time an increase in the production of osmotic substances in those 
cells. I know of no evidence for such an increased production 
of osmotic substances, and Kraus has shown that there is a rela- 
tive and absolute decrease in osmotic substances on the convex 
side during the later stages of the curve. My results agree -with 
his as to the increase of water on the convex side. 
SUMMARY OF EXPERIMENTS. 
The experiments submitted justify the following conclu- 
sions: 
