We have identified the N terminus of the beta subunit as an essential domain for G-protein betagamma assembly. A C-terminal fragment, beta1-(130-340), fails to bind gamma unless coexpressed with the complementary N-terminal fragment, beta1-(1-129). Deletion of the N-terminal 33 residues of beta1, a region identified by computer algorithm to favor coiled-coil formation, abolishes gamma2 association. On the basis of these findings, we propose a coiled-coil model of betagamma interaction and refine this by computer-assisted molecular modeling. The model is tested by further mutagenesis: reversing the charge of residues in beta1 that are hypothesized to be involved in interhelical salt bridges precludes gamma association. Insertions in the coiled-coil region, which disrupt the proposed hydrophobic interface, prevent gamma association. This structural basis for betagamma dimerization provides a starting point for the design of beta and gamma mutants that can be used to map regions in betagamma critical for interactions with the alpha subunit, receptors, and effectors.