Productive, qualitative and seed image analysis traits of guar (Cyamopsis tetragonoloba L. Taub.)

Due to the high content of galactomannans extractable from the endosperm, guar or cluster bean (Cyamopsis tetragonoloba (L.) Taub), is always more frequently cultivated as a grain crop. Guar galactomannans are a source of hydrocolloid able to form a thick solution even in cold water in low concentrations. For this reason, in the last few years the demand for guar gum has greatly increased both for the food and the oil industry. Guar seed accessions identification is quite difficult, since American and Indian varieties and unlabelled genotypes show similar phenotypic seed traits, but very different agronomic value. Eight guar accessions with different origins have been cultivated in a Mediterranean environment and grain yield and galactomannans enzymatic determination have been carried out. The data of 137 morpho-colorimetric quantitative variables were measured, on each seed, by image analysis techniques and analysed applying a stepwise Linear Discriminant Analysis. This procedure allowed us to implement an identification grading system able to discriminate among varieties, with an overall performance of 92.2%. With this work it has been proved that an objective, repeatable and non-destructive method, based on seed phenotypic characterization can be effectively applied also to identify and discriminate C. tetragonoloba accessions. Moreover, for the first time, many strong and positive correlations were found between the galactomannans content and the size descriptive variables measured on seeds, allowing an easy procedure for the determination and comparison of seed galactomannans content. This procedure could become a useful tool for researchers and stakeholders.