The mechanism of the hydrogen-oxygen reaction; the influence of salts.

salts the autocatalysis is not very marked, and the results are fairly reproducible Silica and similar surfaces appear to be very variable: water has a marked, and often unpredictable, effect on them. The m atter is further complicated by the fact th a t water actually formed in the reaction seems to have a more marked effect than th a t added from without. Another phenomenon, probably depending on the way the surface changes with time in the initial stages of the reaction is the following. At temperatures and pressures just below the third explosion limit the rate rises rapidly with increasing pressure of hydrogen, or of inert gases, until explosion occurs a t the limit. When, however, the pressure of oxygen is raised, there is quite a sharp transition to ex­ plosion before the rate has attained nearly so great a value as th a t attainable with increased hydrogen pressure. I t seems th a t an increased $utocatalysis sets in a t higher oxygen pressures, and renders impossible the observation of the steady initial rate.