The Sendai virus RNA polymerase complex consists of two viral proteins, L and P, which must be coexpressed in order to form the active enzyme. Pulse-chase experiments show that the L protein is unstable when synthesized in the absence of the P protein, but is stable in the P-L complex. Using sequential deletions in the P protein (568 amino acids), we have mapped the site on the P protein where the L protein binds by co-immunoprecipitation and gradient sedimentation analyses. The L-binding site residues in the C-terminal half of the P protein, since deletion of up to amino acid 324 of P protein does not affect complex formation. The L-binding site was mapped to a region of P protein encompassing amino acids 412-478. This region lies between the previously mapped amino acid regions which form the nucleocapsid-binding domain (amino acids 345-411 and 479-568). The data suggest that the L and NP protein-binding domains on P protein do not overlap.