It is difficult to assess self-incompatibility (SI) as a phenomenon or to distinguish it from other reproductive barriers. Our objective was to demonstrate a method of separating reproductive barriers and other factors affecting SI expression. The analysis consists of four parts: (1) calculating male (MCC) and female (FCC) coefficients ofcrossability using seed set from a diallel crossing design; (2) determining significant male and female differences using a 1 : 1 paired Chi-square test; (3) constructing scatter diagrams of FCC versus MCC; (4) performing linear regression analysis between FCC and MCC. Diallel data from eight genera with various reproductive barriers were used to illustrate and validate the usefulness of this method. SI can be distinguished from other reproductive barriers (incongruity, inbreeding depression, post-fertilization phenomena) using 1:1 Chisquare and regression analysis. In addition, the incorporation of ovule counts is crucial for demarcating the SI phenomenon in species where ovule numbers vary between self-and outcross pollinations. Male and female steriles can be identified with the MCC/FCC equations and subsequently removed from further analysis.