Abstract-Using an automatic food intake measuring apparatus ("food intakometer"), we recorded the approach behavior to food, eating behavior and food intake of morphine dependent rats and examined the relationship among these factors and morphine dependence.Morphine dependent rats were produced by the drug-admixed food (DAF) method. In the choice trial of free feeding group, morphine dependent rats showed only the approach behavior both to drug-free diet and to morphine-admixed food, then ate the morphine-admixed food and approached the drug-free diet at the same period. Eating behavior in the case of morphine-admixed food was observed not only at night but also during the day time in the morphine dependent rats. In the choice trial of the limited feeding group, preference for morphine rapidly increased in every choice trial of each session and the preference rate became stable at about 60%. Eating patterns of these rats were similar to these in the free feeding group. When these rats were given morphine prior to the choice trial, eating behavior of those on the morphine-admixed food was inhibited dose-dependently, while eating behavior of these on the drug-free diet was enhanced dose-dependently.When these rats were allowed free access to drug-free diet for 1 hour previously to the choice trial, eating behavior of the rats on the morphine-admixed food in the choice trial was markedly enhanced. Thus, the rats clearly showed drug-seeking behavior and seemed to be able to distinguish between the need for morphine and satisfaction without it. Morphine dependent rats apparently can control their required maintenance dose.Over the years a several techniques have been developed to examine drug-seeking behavior. A Y-maze with two goal boxes contrasting in physical as well as spatial charac teristics was introduced by Beach (1), self-administration by oral route in rats (2-4), intraperitoneal route in rats (5), intragastric route both in rats (6) and monkeys (7), intra venous route both in rats (8) and monkeys (9) and intraventricular route in rats (10) are the most common means for inducing drug dependence and investigating drug-seeking behavior in laboratory animals. However, when drugs are insoluble in water, these self-adminis trations cannot be employed to examine drug-seeking behavior, except for oral self-adminis tration, namely, the drug-admixed food (DAF) method. Using this method, preference for morphine, codeine, phenobarbital, diazepam and cocaine has been demonstrated in rats (4,(11)(12)(13).Continuous recordings during the course of the experiment can be made of the behavior of drug self-administration. Hill and Stellar (14) and Kuribara et al. (15) developed an electric drinkometer for the purpose of continuously measuring water intake. Stolerman and Kumar (16) studied the preference for morphine by animals, using this electric drinko