Seeds of Ricinus communis contain two types of lectins; the toxin ricin (approximately 60 kDa) and R. communis agglutinin (approximately 120 kDa). The toxin is a heterodimer composed of a toxic A subunit and a lectin B subunit, while R. communis agglutinin is a tetramer, constituted of two ricin-like dimers held together by non-covalent forces. The lactamyl Sepharose affinity-purified ricin consists of two major groups of variants designated ricin II and III [Hegde, R. & Podder, S. K. (1992) Eur. J. Biochem. 204, 155-164]. The purified A subunits of all the variants of ricins and R. communis agglutinin show heterogeneity in the molecular mass as shown for ricin before [Fulton, J. R., Blakey, C. D., Knowles, P. P., Uhr, J. W., Thorpe, P. E. & Vitetta, E. S. (1986) J. Biol. Chem. 261, 5314-5319]. Since the isoelectric points of the R. communis agglutinin variants fall between the isoelectric points of ricin II and III, we investigated the possibility that this lectin is made up of ricin II and III. The isoelectric points of the purified B subunits of R. communis agglutinin matched well with those of ricin II and III on urea-polyacrylamide isoelectric focussing gel. Further, two-dimensional electrophoretic analysis of the ricin constituants of R. communis agglutinin in the presence of 9 M urea, showed two protein bands, differing by nearly pH 2 in their isoelectric points, the more alkaline one corresponding to that of ricin III analyzed under the same conditions, while the other, although a higher molecular mass variant, corresponding well with ricin II in its isoelectric point. Based on these results and those obtained from adenine binding to A chains of both ricin and R. communis agglutinin, we provide a plausible evolutionary relationship between R. communis agglutinin and two groups of ricin variants; ricin II and III. The model predicts that one half of R. communis agglutinin is derived from ricin I and II, and the other half from ricin III. The results presented, contrary to the existing notion, unequivocally show that the two halves of R. communis agglutinin are not identical protein units, but differ both in surface charge and molecular mass.