Physiological Studies in Plant Anatomy 125 
(these results are in accordance with those recently published by 
Ziegenspeck ( 36 )). 
Experiments designed to test the resistance of the apical meri- 
stem of the root to positive hydrostatic pressures (Priestley and 
Tupper-Carey ( 23 )) have shown at the same time that no leakage takes 
place across the endodermis under such pressure. 
As the result of experiments with acid and basic dyes, in which 
these dyes were driven under pressure through the xylem, the 
following point emerges which has probably considerable significance. 
In the slightly tuberous roots of Ranunculus Ficaria and the basal 
intemodes of the stems of Vicia Faba L. basic dyes, methyl green, 
neutral red and malachite green were strictly retained within the 
endodermal cylinder, only escaping into the cortex where gaps 
occurred in the endodermis, as occasionally in the basal internodes 
of the stem of Vicia Faba L., or after prolonged experiment when it 
was clear that the dye was penetrating dead cells. On the other hand, 
acid dyes (acid green and eosin) penetrated the endodermal proto¬ 
plasts relatively easily and thus escaped into the cortex. 
These results, supported by a number of other observations, lead 
to the following conclusions. The primary endodermal cylinder 
possesses in the Casparian strip a network of wall intervening between 
the protoplasts which renders the radial passage of water or 
solutes by way of the wall uniformly difficult. There are no grounds 
for assuming, as is done by the whole Marburg school of investigators, 
that whilst impermeable to solutes the wall is readily permeable to 
water, and the generalisations of Ziegenspeck ( 36 ) are largely based 
upon experimental studies of membranes that do not possess 
Casparian strips. The Casparian strip therefore places the exchange 
of water and solute through the primary endodermal cylinder under 
the control of the living protoplasts. These protoplasts seem to be 
relatively easily penetrated by complex ions carrying an electro¬ 
negative charge, but refuse passage to complex ions with an electro¬ 
positive charge so long as they retain their living semi-permeable 
plasma membranes. The consequences of this remarkable behaviour 
of the primary endodermis will require much further elucidation, and 
some possible indications of its significance are submitted in later 
papers (Priestley and Tupper-Carey ( 23 ), Priestley and Woffenden( 24 )). 
The Secondary Endodermis 
(1) The Suberin Lamella. 
The essential structural difference between the primary and 
secondary stages of the endodermis is that, in the secondary stage 
