The radioactively labelled puromycin analogue, 5′‐O‐(N‐bromoacetyl‐p‐aminophenyl‐[32P]‐phosphoryl)3′‐N‐l‐phenylalanyl puromycin aminonucleoside, has been used as an affinity label to identify components of Escherichia coli ribosomes that are in the active centre for the peptidyl transferase function of the ribosome.The affinity label as an acceptor substrate at 0°C in the “fragment reaction” assay of peptidyl transferase was only about 10% as efficient as puromycin, although both compounds had the same Km (approximately 0.1 mM). Preincubation of 70‐S ribosomes with the affinity label at 27°C led to irreversible inactivation of the peptidyl transferase activity (pH optimum 8.8) and extensive covalent attachment of radioactivity to the ribosomes (up to 25 molecules of affinity label per ribosome). All the inactivation of the enzyme, but only a small proportion of the covalent labelling (up to 2 molecules per ribosome), could be prevented by the presence of chloramphenicol, an A′‐site‐specific inhibitor of peptidyl transferase. The chloramphenicol‐insensitive affinity labelling, designated nonspecific, was shown to be distributed over many ribosomal proteins in both the 50‐S and 30‐S ribosomal subunits.The specific, chloramphenicol‐sensitive, affinity labelling was shown to be attached to the 23‐S component of ribosomal RNA by polyacrylamide gel electrophoresis. Comparison of the stoichiometry of the specific RNA affinity labelling with the percentage inactivation of peptidyl transferase revealed a linear correlation between the two quantities, such that 2.1 ± 0.2 moles of affinity label per 23‐S RNA corresponded to complete inactivation.As previously reported [Harris, R. J., Greenwell, P. & Symons, R. H. (1973) Biochem. Biophys. Res. Commun. 55, 117–124], incubation of the covalent ribosome‐affinity label complex with the donor substrate, acetyl‐l‐[3H]leucylpentanucleotide, caused acetyl‐[3H]leucine to become covalently attached to ribosomal RNA. It has now been shown that this attachment was also on the 23‐S RNA and that it was linearly correlated both with the extent of 32P‐affinity labelling of the RNA, and with the percentage inactivation of peptidyl transferase activity. However, only approximately 0.5 % of the ribosome‐affinity label complexes were able to effect acetyl‐[3H]leucine transfer.The above results are consistent with a two‐site model for the ribosome, if affinity labelling of the peptidyl transferase acceptor (A′) and donor (P′) sites follows a hypothetical mechanism in which the affinity label binds to (and subsequently reacts with) first the A′ site, and then the P′ site, in an ordered and strongly positive cooperative manner. The hypothesis explains the linear 2:1 correlation between specific affinity labelling and loss of peptidyl transferase activity, and also the small proportion of ribosome‐affinity label complexes able to effect peptidyl transfer. The latter is considered to be the population of ribosomes labelled only in the A′ site, because P′ site labelling lags slightly behind that of the A′ site. The results also suggest that 23‐S ribosomal RNA has a direct functional role in the binding of the 3′‐CCA terminal triplets of tRNA to peptidyl transferase.