G proteins play a central role in transmitting signals from cell surface receptors to effector proteins inside the cell. Signaling can only occur, however, if all these protein components are properly assembled and localized at the plasma membrane. Past studies have shown that certain segments within the N-terminal region of the G protein alpha subunit are necessary for membrane attachment. Here we identify a region within the yeast G alpha (Gpa1) that is sufficient for membrane attachment, as well as for specific targeting to the plasma membrane. Initially, we constructed chimeric proteins that replace the N terminus of mammalian Gsalpha with the corresponding sequence from Gpa1. Gsalpha is inefficiently targeted to the yeast plasma membrane and therefore cannot fully complement the loss of Gpa1. Gpa1-Gsaplha chimeras were assayed for proper membrane localization by functional complementation of a gpa1Delta ;) mutant, and by sucrose density gradient fractionation of cell membranes. Most of the chimeras tested, including one with only the N-terminal 7 amino acids from Gpa1, exhibited normal membrane targeting and complementing activity. We also fused various lengths of N-terminal Gpa1 sequence to glutathione-S-transferase (GST), a heterologous protein normally expressed in the cytoplasm. The first 67- 36- or 9-amino acids of Gpa1 were all sufficient to direct GST specifically to the plasma membrane in yeast. This analysis defines the extreme N terminus of Gpa1 as the primary determinant of proper membrane targeting, and represents an essential step towards isolating and identifying G protein-targeting proteins within the plasma membrane.