Sucrose consumption and preference were examined in rats subjected to a 6-week regimen of unpredictable mild stressors, after Willner et al. (11). These subjects were compared with groups exposed to: 1. only the food deprivation element of the stress protocol; or 2. the stress protocol without the food deprivation element. A control group was not exposed to stressors. Body weight and sucrose consumption were significantly reduced in stressed and food-deprived animals compared to the other 2 groups. These variables therefore appeared dependent on food deprivation and independent of other elements of the stress protocol. Neither sucrose consumption per gram body weight nor sucrose preference differed significantly among the 4 groups. These results indicate that food deprivation is not only necessary, but sufficient, to produce sucrose consumption deficits in rats. It is, therefore, likely that reduced sucrose consumption in stressed rats results solely from diminished body weight rather than exposure to the series of stressors. We conclude that sucrose consumption is not a valid index of reward responsiveness. Other measures (such as place-preference conditioning or intracranial self-stimulation threshold) should be evaluated also with respect to body weight change when considering the validity of stressor-based models of depressive disorder.