The Mechanism of Root Pressure. 
193 
the walls can he followed by its colour or by subsequent staining 
reactions, then this substance will be traced diffusing outwards as 
far as the endodermis and no further. These experiments of De 
Rufz de Lavison seem very conclusive, and quite similar results 
have been obtained by the writer, but are not described in detail as 
De Lavison’s papers are readily accessible. 
De Lavison (3, loc. cit. 1910), likewise Haberlandt (6, loc. cit. 
pp. 371, 372) refers to experimental work of de Vries (10, 1886) not 
accessible in the original to the present writer. In these 
experiments de Vries apparently found that if he forced water into 
the root from the surface of the cut stem it was impossible to 
obtain a leakage of this water from the root unless the endodermis 
was injured. Presumably a pressure was never used which 
exceeded in magnitude the opposing osmotic pressure retaining the 
water within the protoplasts of the endodermal cells. 
The above facts taken in conjunction lead then to the 
conclusion that water, in passing through the endodermal cylinder, 
must pass through the protoplasts. As the considerations studied 
previously show that these protoplasts find themselves linked in a 
series of cells in which the osmotic gradient rises towards the 
inside, this is tantamount to saying that under normal conditions 
the passage of water through the endodermis is completely 
controlled by osmotic phenomena, and is always taking place in an 
inward direction. Before considering the further bearing of these 
facts we will discuss the significance of some modern work on the 
excretion by hydathodes. 
(2) Excretion by hydathodes. The hydathode is one special 
case, dealt with by Lepeschkin (8, 1906) in his study of the 
excretion of water by plants. 
For details reference must be made to the original papers ; for 
present purposes, it is sufficient to cite the facts as to the excretion 
of water by the multicellular hair of Phaseolus multiftorus. (Fig. 2.) 
On the basis of his experimental results with the young 
sporangiophore of Pilobolus, Lepeschkin explains the continuous 
excretion of water from the two top cells of this hair by the 
assumption of a different semi-permeability of the protoplasmic 
layer on the upper, as compared with lower, side of the cell. 
If this protoplasmic layer is more permeable on the upper side, 
then before the internal hydrostatic pressure arrives at the level 
that it is possible to obtain through the entry of water at the lower 
side of the cell, water and certain solutes will be passing out of the 
cell at the top on to the upper surface. This excretion of water 
