c-Myc is helix–loop–helix–leucine zipper (HLH–ZIP) oncoprotein that is frequently deregulated in human cancers. In order to bind DNA, regulate target gene expression, and function in a biological context, c-Myc must dimerize with another HLH–ZIP protein, Max. A large number of c-Myc target genes have been identified, and many of the encoded proteins are transforming. Such functional redundancy, however, complicates therapeutic strategies aimed at inhibiting any single target gene product. Given this consideration, we have instead attempted to identify ways by which c-Myc itself could be effectively disabled. We have used a yeast two-hybrid approach to identify low-molecular-weight compounds that inhibit c-Myc–Max association. All of the compounds prevented transactivation by c-Myc–Max heterodimers, inhibited cell cycle progression, and prevented the in vitro growth of fibroblasts in a c-Myc-dependent manner. Several of the compounds also inhibited tumor growth in vivo. These results show that the yeast two-hybrid screen is useful for identifying compounds that can be exploited in mammalian cells. More specifically, they provide a means by which structural analogs, based upon these first-generation Myc–Max inhibitors, can be developed to enhance antitumor efficacy.