The complete amino acid sequence of Japanese chestnut (Castanea crenata Sieb. et Zucc.) agglutinin (CCA) was determined. Analysis by SIMS of the acidic peptide obtained by pepsin digestion revealed that the N-terminal amino acid sequence should be Acetyl-Met-Glu-Glu. Prior to sequence analysis, redetermination of cysteine residues indicated the presence of one cysteine residue per subunit. The complete sequence was determined by endoproteinase Arg-C and Achromobacter protease I digestion, and CNBr cleavage. CCA consists of 309 amino acid residues with a high content of glycine (16.5 mol%) and one cysteine residue. The calculated molecular mass was 33, 387 Da including the N-terminal acetyl group. C-terminal sequence analysis of intact CCA gave only one sequence, HMEYF, indicating that no heterogeneous CCA formed by posttranslational cleavage at the C-terminal region, as occurs in some legume lectins. Analysis of the sequence of CCA itself revealed that CCA could be divided into two structural domains, the N-domain and the C-domain, almost at the center. These domains share about 35% identical residues, so CCA has a repeat sequence. Also, both domains show a homology to jacalin-related lectins with 27-38% identity. These results suggest that the structure of CCA resembles two molecules of jacalin-related lectin.