Effect of drought stress on lipid peroxidation, chlorophyll stability and several antioxidant enzyme activities was evaluated under controlled environmental conditions, for possible use as drought tolerance selection criteria, in four tobacco (Nicotiana tabacum L.) cultivars of different, but known, drought tolerance. A progressive highly significant (p <0.01), but differential (in correlation with their individual drought tolerance), increase in glutathione reductase (EC 1.6.4.2) activity was detected in all four cultivars as their leaf water potential (Psi(L)) decreased. In spite of an initial lag (up to a Psi(L) of -1.5 MPa), on average, the superoxide dismutase (SOD, EC 1.15.1.1) activity of the drought-tolerant cultivars increased by 244 % while that of the drought-sensitive cultivars only increased by 161 %. Contrary to the other enzyme activities monitored, only a moderate increase in catalase (EC 1.11.1.6) activity (p <0.05) relative to their respective controls was observed in all four cultivars. Increased ascorbate peroxidase (EC 1.11.1.7) activity was not only observed to be ca. 300-400% higher (p <0.01 already at a Psi(L) of -1.5 MPa) in the drought-tolerant cultivars under stress, but was also more pronounced than the increase in catalase activity. This indicated that ascorbate peroxidase rather than catalase might be mainly responsible for scavenging drought-stress produced H2O2. On reaching a Psi(L) of -2.5 MPa, the glutathione reductase activity of the drought-sensitive cultivars increased by only 159 % (TL33) and 187 % (CDL28), opposed to the 233 % (GS46) and 250 % (ELSOMA) in the drought-tolerant cultivars. A differential (significantly higher in the drought-tolerant cultivars at a Psi(L) of -2.5 MPa) drought stress-induced increase in the level of lipid peroxidation occurred in all cultivars which decreased faster in the drought-tolerant cultivars upon rehydration. The latter was also observed for catalase and glutathione reductase activity, but of which the activity returned to levels comparable with that of their respective controls. The levels of SOD and ascorbate peroxidase activity on the other hand remained higher (p <0.01) than their respective controls after rehydration, but did not differ statistically significantly between the respective cultivars. No destruction of chlorophyll, due to photoperoxidation was detected in any cultivar, as reflected by a stable chlorophyll a/b ratio. These results are discussed in relation to the potential drought tolerance adaptive advantage of an effective antioxidant system. In this regard the use of monitoring the capacity to increase ascorbate peroxidase activity and/or glutathione reductase activity as possible drought tolerance selection criteria in tobacco is advocated.