64 BOTANICAL GAZETTE [yuLy 
measured, but this is doubtless owing to the fact that the products 
of the reaction must diffuse from the spongy platinum into the 
solution before they can be detected by titration. 
It seems evident, therefore, that in this reaction, as in the oxi- 
dation of formaldehyde, the platinum is capable of carrying oxygen 
into the solution, and it appears from fig. 3 that if the oxygen were 
supplied at sufficiently frequent intervals the reaction. could be 
made to follow the curve OE, which represents the action of colloidal 
platinum in the oxidation of potassium iodide by hydrogen peroxide. 
From these two reactions it may be concluded that when 
colloidal platinum is introduced into a mixture of hydrogen 
peroxide and an oxidizable substance the platinum takes up oxygen 
from the peroxide, thereby forming a more efficient oxidizing’ 
agent than the original hydrogen peroxide. The catalytic action 
of the platinum in this case, that is, its peroxidase action, therefore, 
depends upon its aptitude for forming unstable oxygen compounds 
when it is in contact with hydrogen peroxide. 
Similar experiments with plant material will be reported on 
in a subsequent paper. 
LABORATORY OF PLANT PHYSIOLOGY 
HARVARD UNIVERSITY 
