Discussion
In flooded soils, air space is displaced by water, and the oxygen remaining in the soil is quickly depleted (Fig. 1) by respiration of plant roots and soil microorganisms (Kawase 1981). When plants in this experiment were removed from water, soilO2 recovery was slow (Fig. 1). A similar slow recovery period was found in citrus plants flooded for 8 days (Ruiz-Sánchez et al. 1996). Flooding had no influence on the leaf water potential of apricot plants during the first three days of the flooding period (Fig. 2), while leaf conductance and net photosynthesis were more affected, decreasing from the beginning of flooding (Figs. 3 and 4). No osmotic adjustment was found in flooded plants. Nor were we able to detect differences in leaf turgor potential between the leaves of flooded and control plants (data not shown) at a time when leaf conductance was obviously affected (Fig.3). For these reasons, stomatal regulation, which has been recognised as one of the earliest responses to flooding in fruit trees (Kozlowski 1982, Syvertsen et al. 1983, Davies and Flore 1986, Smith and Ager 1988, Schaffer and Ploetz 1989, Larson et al. 1991), appears to have had a positive impact on the water balance of both apricot rootstocks during the early stages of flooding (McNamara and Mitchell 1989, Ruiz-Sánchez et al. 1996). Consequently, it and may be considered as an adaptive mechanism to prevent leaf dehydration (Bradford and Hsiao 1982). This stomatal regulation did not occur as a result of a water deficit in the leaf, but as a hormonal imbalance caused; it has been proposed by an increase in abscisic acid as well as a deficiency in such factors from the roots as cytokinins and gibberellins, which promote stomatal opening (Wright 1972, Bradford and Yang 1981).