HYDROGEN AND OXYGEN IN CONTACT WITH HOT SURFACES. 
27 
After the above experiments were concluded, the combustion tube containing the 
catalysing material was transferred to the “ absorption apparatus,” where its power of 
occluding hydrogen at dull red heat (circa 650°) was tested. In one experiment the 
pressure fell from 764 to 648 millims., and in another from 765 to 657 mill inis., or by 
about one-seventh, during 30 minutes. Corresponding experiments with oxygen gave 
negative results. 
The catalysing power of the surface, as well as its susceptibility to any stimulus by 
oxygen, underwent a marked alteration as the result of the repeated occlusions of 
hydrogen, followed by prolonged exhaustion, at dull red heat. Its activity was now 
much reduced, the new “ normal ” value for “ l\ ” at 450° being only 0'033 as compared 
with the 0’063 obtained in previous experiments. It is, however, significant that the 
new value was almost identical with that obtained in Experiment XXXVII., namely, 
0 , 0382, after the surface had been exposed to hydrogen at 450°. 
In its new condition of reduced activity, the surface had completely lost its 
susceptibility to stimulus by oxygen; indeed, prolonged exposure to oxygen at 450° 
now actually diminished its catalysing power. On the other hand, it had acquired 
the power of responding, to a slight degree, to the stimulating influence of hydrogen 
at 450°. In short, its behaviour now approximated to that of surface B, a circum¬ 
stance which proves how profoundly the action of porcelain may be affected by its 
past history. 
Summary. 
Reviewing now the results obtained with the five surfaces as a whole, it is clear 
that the rate of steam formation is governed neither by the “ order” of the reaction, 
as Bodenstein asserted, nor by diffusion factors simply, as Nernst has suggested. 
Nor is the theory of a rapidly alternating series of oxidations and reductions of 
the catalysing surface admissible. The evidence is, on the contrary, altogether 
in favour of the view that the catalytic action of porcelain is primarily due to the 
condensation, or occlusion, of one, and possibly both, of the reacting gases on its 
surface. The fact that whenever one or other of the gases is present in excess, the 
rate of combination is proportional to the partial pressure of the hydrogen, and that 
the material absorbs hydrogen at red heat, indicates that occluded hydrogen is an all 
important factor in the process. It would also seem, from the results obtained with 
surfaces C and E, that occluded oxygen also may exercise an accelerating influence 
on the process; otherwise it is difficult to account for the fact that in several of the 
experiments with an excess of oxygen a surface proved more than normally active, 
and that surface E, before it had been subjected to the treatment with hydrogen at 
red heat, could be stimulated by previous exposure to oxygen at 450°. Another 
significant circumstance is that, whilst excess of hydrogen always produced a 
continuous acceleration throughout a given experiment, any acceleration due to an 
E 2 
