Abstract
Variation between genotypes in the ability to accumulate abscisic acid (ABA) in response to water stress was investigated in pearl millet (Pennisetum americanum (L.) Leeke). Using a detached leaf test a more than four-fold variation in accumulation capacity was observed amongst a set of 16 genotypes grown in a controlled environment. Two genotypes which contrasted in their accumulation capacity, BJ 104 and Serere 39 (the latter accumulating most ABA), maintained the difference over a range of leaf water contents and potentials. Some of the genotypes were grown in the field in the semi-arid tropics with and without irrigation, and sampled for ABA content. In two experiments, substantial genotypic variation in ABA accumulation was observed, which could not be attributed to differences in leaf water potential (ψ). In a field experiment comparing three genotypes (Serere 39, BJ 104, and B282), differences in ABA accumulation were also shown to be largely independent of genotypic differences in turgor potential (ψp). For a set of six of the genotypes, the amounts of ABA accumulating in leaves of intact, droughted plants, in the field, when adjusted for differences in ψ, were found to be significantly (P<0.05) correlated with amounts of ABA accumulated in detached, water-stressed leaves. It is concluded that the detached leaf test adequately reflects the ability of pearl millet genotypes to accumulate ABA under field conditions.