A magnetic body changes its thermal state when subjected to a changing magnetic field. In particular, if done under adiabatic conditions, its temperature changes. For the past 15 years the magnetocaloric effect has been the focus of significant research due to its possible application for efficient refrigeration near room temperature. At the same time, it has become common knowledge within the magnetic refrigeration research community that the magnetocaloric effect was discovered by the German physicist E. Warburg in 1881. We re-examine the original literature and show that this is a misleading reading of what Warburg did, and we argue that the discovery of the effect should instead be attributed to P. Weiss and A. Piccard in 1917.