Low‐energy electron diffraction, with the diffracted electrons post‐accelerated and observed on a fluorescent screen, has been used to study the adsorption of oxygen on a (100) face of a nickel crystal. Upon admitting oxygen to a clean face at room temperature, or at elevated temperatures up to 350°C, the first adsorbed atoms are arranged in narrow bands parallel to [110] directions on the crystal surface. Within one of these bands, atoms lie on lines at right angles to the band; these lines have a uniform separation of 4.98 A, but the atoms along each line are somewhat randomly spaced in multiples of the nickel spacing of 2.49 A. With further oxygen exposure (30×10−6mm Hg sec) sufficient to produce half of a monolayer (one oxygen atom for every four surface nickel atoms), the arrangement is very chaotic unless the crystal has been heated, but after heating the arrangement is a simple square array with edge of 4.98 A. Further exposure to oxygen (100×10−6mm sec) will produce a full monolayer consisting of this same square array with an additional oxygen atom at the center of each square. With still further exposure, oxygen, at least in part as physisorbed molecules, is piled on top of this first monolayer of atoms, the added molecules forming an array with square symmetry and edges parallel to the edges of the square array of surface nickel atoms. At least several layers of oxygen molecules can be adsorbed and are not removed by pumping at room temperature. Heating at 350°C removes a great deal of the adsorbed oxygen, and the structure of the remainder agrees with that of NiO having the orientation of the nickel crystal. This oxide is removed at 830°C, and the first layer of oxygen atoms at 880°C.