24 
He also discovered that in perfectly pure air-free water, metals like 
zinc and iron retain their luster undiminished, whereas if air be 
admitted they soon rust. Evidently, therefore, two factors are 
concerned in the rusting of metals, viz, water and oxygen. 
It was also recognized about this time that water is essential to 
the oxidation of combustible gases. Dixon ( 140 ) had observed that 
a mixture of perfectly dry carbon monoxide and oxygen does not 
explode under the influence of the electric spark, nor do they combine 
when passed over red-hot platinum gauze, and Traube ( 434 ) himself 
made the observation that a flame of burning carbon monoxide is 
extinguished nn being brought into perfectly dry air. He therefore 
reached the conclusion that “no substance can act upon dry oxygen 
at ordinary temperature.” (Traube ( 433 ), p. 1881.) Other chemists, 
among them Nef ( 3n ) and Armstrong ( 13 ), have arrived at essentially 
similar conclusions respecting the necessity for water in many such 
processes of oxidation, and recently Dunstan, Jo wet t, and Gould- 
ing ( 154 ) have found that both water and oxygen, the former in the 
liquid condition, are necessary for the rusting of iron. Cushman ( 128 ) 
also, in a recent communication on the “Corrosion of Iron,’ 7 states 
that while not agreed as to the mechanism of the process, all investi- 
gators are agreed that both oxygen and water are essential to the 
rusting of iron. 
That hydrogen peroxide is formed during the oxidation of oxidizable 
substances by molecular oxygen was known even to Schoenbein 
(see Traube ( 431 ), supra), and since his time instances of its formation 
during oxidation had greatly accumulated in chemical literature. 
When, for example, finely divided zinc is shaken with air and water, 
the metal is gradually oxidized and hydrogen peroxide is found in 
considerable quantities in the solution. These observations on the 
production of hydrogen peroxide as the result of oxidation by 
molecular oxygen acting in the presence of water were greatly 
extended by Traube. Thus he ( 432 ) observed its production in large 
amounts at the cathode during electrolysis when oxygen or air is 
passed through the solution surrounding the negative pole, whereas 
it is never produced by the action of molecular hydrogen on the 
oxygen liberated at the anode. Schuller ( 391 ) had shown that 
hydrogen peroxide results from the burning of hydrogen, and on 
repeating these experiments Traube ( 435 ) succeeded in obtaining as 
much as 0.0108 gram of hydrogen peroxide by the burning of 1 
liter of hydrogen. Traube ( 434 ) also made the interesting observation 
that hydrogen peroxide is produced by allowing a flame of burning 
carbon monoxide to impinge on the surface of water. These and 
many facts of similar import led him to believe that in all oxidations 
it is the molecule of oxygen and not the atom which first enters into 
combination with the autoxidizable substance. 
