Previously, it was shown that oxygen is a factor in the magnetic annealing of Permalloy (Ni‐Fe alloys) and Perminvar (Ni‐Fe‐Co alloys). Electron‐diffraction studies indicated that oxygen caused impurity faults by condensing on the (111) planes of the crystal. An intimate connection between the presence of these impurity faults and the ability to respond to heat treatment in a magnetic field was established. However, the orientation of these faults was not observed to be directly related to the direction of the field heat treatment. In the present work on single crystals of 76% Ni—24% Fe, it was found that those crystals which were heat treated in a purified hydrogen atmosphere did not respond to the magnetic annealing treatments when cooled at normal rates, whereas those that were heat treated in a slightly oxidizing atmosphere did respond. Furthermore, after subsequent heat treatment, x‐ray investigations showed that the crystals which did not respond to magnetic annealing had substantially more long‐range ordering present (4 to 1) than the crystals which did respond. Long‐range ordering first appeared in both specimens after 475°C for 18 h. The specimens that did not respond to magnetic annealing also developed substantially higher values of crystal anisotropy than those that did respond. The uniaxial anisotropy which was developed by the slightly oxidized crystals could not be obliterated by ordering heat treatments. This work suggests that the oxygen inhibits the long‐range ordering mechanism, and, therefore, it may aid in the formation of like pairs of atoms.