A possible role of abscisic acid (ABA) in the regulation of grain set in water-stressed wheat (Triticum aestivum L.) was investigated using a split root system to dry half the roots while the remainder were kept watered. Water uptake by the wet roots maintained the leaf water potential at the normal level, whereas the ABA produced in the dry roots was transported to the spike. This caused the spikelet ABA level to increase to the same extent as when the entire root system was stressed to permit a drop in the leaf water potential. In spite of this, the former treatment did not induce a reduction in grain set, whereas the latter did. Thus, contrary to previous reports, water stress-induced changes in spikelet ABA level alone do not appear to regulate grain set.The male sporogenesis in wheat is highly sensitive to water stress. Stress during meiosis causes male sterility and, hence, a serious reduction in grain set (3,10 (8,9,11,14,15). The greatest sensitivity toward water deficit or exogenous ABA occurs at the same developmental stage, i.e. meiosis (10,11,15). The anthers and pollen grains resulting from the two treatments look morphologically similar at maturity (8, 10, 11). In well-watered plants, the increased inherent likelihood of sterility in the distal florets within a spikelet is associated with an elevated ABA content (6). Finally, the cornerstone nuclear-male-sterile mu- ' Supported by an operating grant from the Natural Sciences and Engineering Research Council of Canada to H.S.S.2 Abbreviation: At, water potential.tant of wheat has greater ABA levels in spike and leaves and a lower rate of ABA metabolism than the corresponding fertile plants (16). These observations, taken together, suggest that ABA may act as an endogenous male sterilant in waterstressed wheat. However, these observations are entirely correlative, and a causal relationship between endogenous ABA and water stress-induced male sterility has never been established. One obstacle to the study of this problem has been the lack of a specific inhibitor of ABA biosynthesis or action. We took an alternative approach and employed a split root system to dry half of the roots while keeping the other half well-watered. This enabled the spikelet ABA level to increase via the transport of the hormone produced in the drying roots (4), whereas the shoot 4t was maintained at a high level through adequate water uptake by the wet roots. We could thus separate the effects of ABA from other metabolic consequences of shoot water deficit, and investigate whether increased ABA concentration in the spikelets could indeed influence grain set.
MATERIALS AND METHODS
Plant Growth ConditionsWheat (Triticum aestivum L. cv Katepwa) was grown in 9.5 cm x 15 cm plastic pots (two plants per pot) containing a 50:50 mixture of sand and Pro-Mix (Premier Brands Inc., Stanford, CA). For experiments with the split-root system, 2-week-old plants were transplanted into modified pots (Fig. 1A)