372 
AMERICAN JOURNAL OF BOTANY 
[Vol. 8, 
The parenchyma within the endodermis, being confined, is limited in ex- 
tensibility on account of the structure of the endodermal cells. A strong 
hydrostatic pressure will therefore develop in this core of cells, sufficient 
to cause an excretion of water and solutes into the xylem vessels so long 
as the osmotic gradient obtains. When the water and solutes enter the 
xylem they are free to move upward in the vessels. Water may leak back- 
ward as far as the endodermis but no further on account of the suberized 
walls of the endodermis, which structure has been shown to prevent such a 
backward leakage. Priestly shows that the apical region of the root does 
not permit a backward leakage. 
This explanation of root absorption and secretion offers some difficulties. 
Priestly points out that it is difficult to understand how the necessary 
solutes can be provided in sufficient quantities to permit a constant flow 
of water across the inner membranes of the cells next to the xylem vessels, 
where a considerable amount of water may pass upward. On the basis of 
Lepeschkin's results, Priestly suggests that such solutes might be either 
organic or inorganic, and that in the root they are organic, either sugars 
or, more probably, organic acids derived from sugars, and cites Atkins' 
results as to the presence of sugar in the ascending sap. Curtis, however, 
holds that the xylem does not serve for longitudinal translocation of carbo- 
hydrates. 
In harmony with Bayliss, Priestly suggests that normal semi-permeabil- 
ity to glucose is a function of the difference in concentration on the two 
sides of the membrane, so that an accumulation or a dilution of sugar within 
the cells bordering on the xylem vessels might lead to a change of permea- 
bility in these cells and result in rendering the plasma membrane on the 
inner side temporarily permeable to sugar, and thus in an intermittent 
excretion into the xylem vessels. As Priestly suggests, the process, al- 
though intermittent, would appear as a continuous one in root pressure, 
due to the combined activity of many cells. 
Flood finds that the exudation of water from Colocasia leaves does not 
depend upon any special secretion tissue in the leaves, but that the phenom- 
enon rests upon the action of cells lower down, probably in the root. Dur- 
ing its passage upward from the roots until its exudation, the water passes 
through no filtration membranes. The water exuded is almost free from 
solutes as has been shown by Atkins. If we assume that solutes are ex- 
creted into the xylem along with the water, the question arises as to what 
becomes of these solutes in the case of Colocasia. Priestly suggests that 
they are adsorbed during their passage upward by the protoplasts surround- 
ing the vessels. 
Transpiration has by some been looked upon as a function on the sup- 
position that it is useful in concentrating the salts brought to the leaves, 
a supposition to which certain workers object on the ground that this 
assumption carries with it the further assumption that water in the vessels 
