Cysteinyl-tRNA synthetase (CRS) from rabbit liver was purified 8300-fold to a constant specific activity. SDS-PAGE revealed the presence of two polypeptides of 86 kDa and 92 kDa, in the proportions of 60% and 40% respectively. The SDS-electrophoretic migration of the major 86 kDa component was indistinguishable from that of the single polypeptide previously found in CRS from S. cerevisiae. The two polypeptides from rabbit CRS were inaccessible to Edman degradation, but internal peptides generated from each by in-gel proteolysis after SDS-electrophoretic separation, yielded sequences found in the deduced protein sequence of human CRS. Moreover, subjecting the two polypeptides separated by SDS-PAGE to a renaturation treatment showed that CRS activity was associated with both. The structure of the native enzyme was probed by limited proteolysis with elastase. The strikingly simple degradation pattern observed supported a model according to which the two polypeptides derive from the same gene, differing only by a approximately 6 kDa extension located at the C-terminal extremity of the 92 kDa component. Moreover, the finding that notwithstanding the presence of the two polypeptides, the behaviour of rabbit CRS upon gel-filtration or chemical cross-linking was indistinguishable from that of homodimeric yeast CRS, indicated that the 6 kDa C-terminal extension on the 92 kDa polypeptide does not impede dimerisation. The origin of the two components of rabbit CRS is discussed in light of the deduced protein sequence of human CRS derived from the published cDNA sequence and the recently released genomic sequence of the human enzyme.