Abstract
Recent neutron-diffraction experiments by Mook, Lynn, and Nicklow, as well as previous photoemission experiments by Pierce and Spicer, provide evidence that the Stoner band splitting in nickel might not go to zero at Tc, as it does in conventional band theory of magnetism. It will be shown that the persistence of such splitting in the paramagnetic regime can be explained quite naturally within the framework of the random-phase-approximation theory of spin fluctuations.