191 
The Mechanism of Root Pressure. 
We can then understand the passage inwards of the water 
from the surface of the root to cells in the position of L, Text-fig. 
1, but the further passage of the water into the cavity of the xylem 
vessel seems at first sight to present a difficulty. The cellulose 
wall of L is completely lined within by the usual semi permeable 
layer of protoplasm, whilst M has no such lining. Under these 
conditions, unless the sap within M contains a sufficient 
concentration of solutes to be in osmotic equilibrium with the cell 
sap of L, water will apparently be withdrawn from M to L by 
osmosis. But if the cell sap of L is under sufficient pressure, the 
sap may remain in equilibrium with a more dilute solution in M, 
which, protected by its rigid lignified wall, is not under the same 
pressure. Gelston Atkins (1, loc. cit. p. 201) has, in reality, 
invoked this explanation when suggesting that the ascent of water 
in the xylem of the root may be explained as an osmotic 
phenomenon, acting from the dilute sugar solution in the xylem to 
the soil solution outside, through a mass of living cells, treated as 
one semi-permeable membrane. 
The parenchyma within the endodermis, e.g ., the cells H, J, K, 
L, are confined within an endodermis, the extensibility of which is 
limited, the radial walls of the cells being clamped by lignified 
thickenings; and frequently the other walls of the endodermal cells 
are also greatly strengthened. 
No air spaces are present between the cells, and the walls of 
the xylem vessel are rigid. Between endodermis and xylem vessel, 
the cells then are practically unable to expand, and under these 
conditions it is physically possible to have the internal hydrostatic 
pressure, within cells in the position of L, so high that water may 
be forced into M, although the solution in M has a lower osmotic 
pressure than the cell sap of L. 
The function of the endodermis in this process will now be 
considered more closely, but it must first be pointed out that the 
hydrostatic pressure exhibited by the cells from H to L will not 
alone explain the continuous rise of water in the xylem vessel M, 
although it would explain the sap in L remaining in equilibrium 
with a less concentrated solution in M. M would have to retain 
a very appreciable osmotic pressure, and, as M contains no living 
protoplasmic contents, the only conditions under which this would 
seem possible, would be unde-r circumstances which would permit 
of a constant supply of solutes to M from the adjacent cells. The 
phenomena investigated by Lepeschkin in the excretion of water 
by hydathodes seem to throw light upon this question. 
