NtrC (nitrogen regulatory protein C) is a bacterial enhancer-binding protein that activates transcription by catalyzing isomerization of closed complexes between σ 5 4 -holoenzyme and a promoter to open complexes. To catalyze this reaction, NtrC must be phosphorylated and form an appropriate oligomer so that it can hydrolyze ATP. NtrC can also repress transcription by σ 7 0 -holoenzyme. In this paper we characterize ''repressor'' mutant forms of NtrC fromSalmonella typhimurium, forms that have lost the ability to activate transcription by σ 5 4 -holoenzyme (in vitroactivity at least 1000-fold lower than wild-type) but retain the ability to repress transcription by σ 7 0 -holoenzyme.The amino acid substitutions in NtrC r e p r e s s o r proteins that were obtained by classical genetic techniques alter residues in the central domain of the protein, the domain directly responsible for transcriptional activation. Commensurate with this, phosphorylation and the autophosphatase activities of NtrC r e p r e s s o r proteins, which are functions of the amino-terminal regulatory domain of NtrC, are normal. In addition, these proteins have essentially normal DNA-binding, which is a function of the C-terminal region of NtrC and bind cooperatively to enhancers. (The NtrC G 2 1 9 K protein has ''improved'' DNA-binding, which is discussed.) We previously presented evidence that several NtrC r e p r e s s o r proteins have impaired ATPase activity. We now show that two other repressor proteins, NtrC A 2 1 6 V and NtrC A 2 2 0 T , have as much ATPase activity as wild-type NtrC when they are phosphorylated and bound to an enhancer and that they have considerably more activity than an unphosphorylated NtrC c o n s t i t u t i v e protein, which is capable of activating transcription. These results demonstrate that NtrC A 2 1 6 V and NtrC A 2 2 0 T fail in a function of the central domain other than ATPase activity. Although they may fail in contact with σ 5 4 -holoenzymeper se, the fact that alanine is the amino acid normally found at these positions leads us to speculate that these proteins fail in coupling energy to a change in conformation of the polymerase.