Greenhouse experiments have been carried out to study the genotypic variation among 35 bean ( Phaseolus vulgaris L.) genotypes with regards to tolerance to zinc (Zn) deficiency ( Zn efficiency). Plants were grown for 45 days in Zn deficient soil supplemented with 0 or 5 mug Zn g(-1) soil) and analyzed for Zn efficiency, plant Zn concentration and content, and the distribution of Zn between old and young parts of the shoot. Zn efficiency (ZE) was defined as the ratio of dry matter production at low and high Zn supply and was calculated for the whole shoot as well as for young and old parts of the shoot. There were marked differences in ZE among the bean genotypes. Genotypes G4449 and G11360 were about 2-fold and 10-fold more Zn-efficient than G11229 and G3871 in whole shoot and young-part based ZE, respectively. Interestingly, the older portions of the shoot for most genotypes had higher dry matter production under Zn deficiency than under sufficient Zn supply, suggesting that there was a significant inhibition of new shoot growth and transport of photosynthates from source to sink organs under low-Zn conditions. Zinc concentrations of both old and young portions of the shoot did not correlate with ZE, but shoot Zn content was found to be significantly correlated with ZE. Furthermore, Zn-efficient genotypes distributed more Zn into young parts of the shoot under Zn-deficient conditions than did the inefficient lines. Variation in seed Zn content did not significantly influence the determination of ZE. We concluded that there is a substantial variation in Zn efficiency in the bean genome, and ZE based on analysis of the young shoot tissues represents a suitable screening technique for the evaluation of ZE in low-Zn soils.