Chemisorption of dihydrogen, dioxygen, and dinitrogen on metal surfaces during reactions of hydrogenation, oxidation, and dinitrogen fixation appears generally to take place after crossing an activation barrier. The nature of this barrier is unknown and the state of the chemisorbed intermediate is, in all cases, a matter of conjecture based on ingenious but indirect observations. In the case of metals, the state of the catalytic surface appears to be very different from that of a clean surface. The definition of a working catalytic surface must include all the surface species of the catalyst in its working state.