A series of potential heterogeneous nucleants were added to metastable solutions of lysozyme and of four other model proteins, as well as to a few target proteins. Some of the substances markedly affected the nucleation of lysozyme crystals, but did not influence the other proteins. The results for other proteins were disappointing because these nucleants have not turned out to be the universal nucleants we were hoping for. However, having worked so well for lysozyme, they may yet prove useful in individual cases.