Four food-deprived rats barpressed with food reinforeement on a I-min fixed-interval sehedule in 7-h sessions with water either present or absent. They all became polydipsic, drinking after many pellets early in sessions but after fewer pellets later in sessions. Pellets were seeured at near maximum rates for about 3 h. Thereafter, the three heaviest drinkers obtained more food with water than without it. The results indieate that schedule-indueed drinking does not eontinue indefinitely and that it ean be adaptive with regard to feeding.Rats switch between eating and drinking apparently at random when food and water are freely available together Premack, 1965), but if water is free and food pellets are scheduled intermittently, then occasions and durations of drinks are predictable . At interpellet intervals between about 15 sec and 2 min, a bout of drinking begins after almost every pellet delivery and lasts for a time that varies directly with the interpellet interval (Colotla, Keehn, & Gardner, 1970;Hawkins, Schrot, Githens, & Everett, 1972). Such eat-drink sequenees of behavior do not terminate when normal water intakes are attained; Falk (1961) showed that in 3.17-h sessions, with food pellets scheduled once per minute on the average, water consumption reaehes polydipsie levels. Falk (1961) and most subsequent investigators have stressed the excessiveness of schedule-induced drinking, but have not described the distribution of eating and drinking over long periods of time. Such information was gathered in the present study, whieh also compared rates of barpressing and eating with and without water available.
METHOD SubjectsFour 3-month-old experimentally naive male albino rats of the Wistar strain supplied by Woodlyn Farrns, Guelph, Ontario, were maintained at 80070 of their free-feeding weights in individual horne cages. Water was available at all times.
ApparatusTwo standard 11.5 x 9.5 x 7.5 in. (29 x 23.5 x 19 cm) Grason-Stadler two-bar rat chambers (Type E3125 B) with pellet rnagazines that delivered 45-mg Noyes rat pellets were the experimental spaces. In each box, the lever beside the door was removed and its housing covered with a metal plate mounted flush with the wall. A force of 0.19 N (20 g) on the remaining bar activated relay scheduling and recording equipment in an adjoining room.When water was to be available, a water bottle was attached to each ehamber door such that the outlet tube was 2 in.
ProcedureThe animals were trained to barpress for food pellets and then given two l-h sessions in which every barpress was reinf'orced, five I-h sessions with reinforcement scheduled at fixed 30-sec intervals, with water available on the last two of these, and two y,-h sessions with water available and reinforcement scheduled at fixed 6O-sec intervals. Then, subjects received seven "wet" and seven "dry" sessions, each 7 h long. On wet sessions, the water bottle was in place; on dry sessions, the bottle was removed, For Subjects land 3, wet and dry sessions occurred in the order WDWDDDWDWWDWDW; the oppo...