469 
Influence of Carbon Dioxide. IV. 
to which they were securely fastened with silk thread (see Text-fig. %b). 
These glass tubes were inserted through corks in the bottom of a wooden 
trough. The joint between the testa and the glass tubes was sealed by 
mercury held in the trough. The glass vessel D, through which a current of 
carbon dioxide was slowly passed, was placed over the trough as shown in 
the diagram. For a few minutes a rapid current of oxygen was passed 
through the tubes A and B via the lateral connexions E and F. At a given 
moment the stop- cocks H and J were turned off, leaving an atmosphere of 
D 
a . b . 
Text-fig. 3. 
pure oxygen in the tubes A and B, the lower ends of which were allowed to dip 
into a concentrated solution of potassium hydrate contained in the vessel K. 
Now, since any carbon dioxide which penetrates the testa- membrane was 
immediately absorbed by the KOH-solution, the actual rate of the passage 
of oxygen from the tubes A and B respectively through the seed-coat 
membranes was shown by the rise of the solution in these tubes, which was 
readily measured on the scales L and M. In this way the permeability of 
any two testas was directly compared. 
