Sensitivity of enzymes to inhibition by thiol-reactive reagents is often presented as evidence for the possible involvement of cysteine residues in substrate binding and catalysis or to highlight possible important differences in structure and mechanism between closely related enzymes. The primary phenotypic distinction between the enterobacterial type II chloramphenicol acetyltransferase (CATII; typified by the enzyme encoded by the incW transmissible plasmid pSa) and the CATI and CATIII variants is the greatly enhanced susceptibility of CATII to inactivation by thiol-specific modifying reagents. Determination of the nucleotide sequence of the gene, catII, present on pSa and that of a related determinant, catIIH, isolated from Haemophilus influenzae indicates that sensitivity to such reagents cannot be due to the presence of additional reactive cysteine residues in CATII. Comparative analysis of the inactivation of CATII and CATIII by 5,5'-dithiobis-(2-nitrobenzoic acid) (DTNB), 4,4'-dithiodipyridine (DTDP) and methyl methanethiosulphonate (MMTS) suggests that (i) inactivation occurs as a result of chemical modification of the same residue (Cys-31) in each enzyme, (ii) reagents that inactivate via a pseudo-first-order process (DTNB and DTDP) appear to bind with a greater affinity to CATII, and (iii) the intrinsic reactivity of Cys-31 in CATII greatly exceeds that of the corresponding residue in CATIII. The results lead to the conclusion that a striking difference in chemical reactivity of a unique and conserved thiol group between closely related enzyme variants may not be easily explained even when a high-resolution tertiary structure is available for one of them. Plausible explanations include more favourable access of reagents to Cys-31 in CATII or an enhanced reactivity of its thiol group imposed by the side chains of residues that are not in immediate contact with it.