Nocturnal activity and foraging of 39 radio-equipped raccoons (Procyon lotor) in eastern North Dakota were studied from April-July in 1974-1976. Sixteen of the raccoons were collected after foraging bouts for stomach content analysis. Raccoon activity consisted of running (13 %), walking (49%) and local movement in confined areas (38%). Local movement was foraging on large or locally abundant food items. Adult males traveled farther in a night, ran twice as often, and moved locally only half as often as adult females and yearlings. Differences in activity patterns between adult females and yearlings were not detected. There was no difference among age-sex groups in use of foraging habitats. All raccoons foraged extensively in farmyards and wetlands. Stomach content analysis substantiated foraging determinations obtained by radiotelemetry. Principal foods were grain, aquatic animals, rodents, birds and bird eggs. INTRODUCTION Nocturnal habits of raccoons (Procyon lotor) in relation to home range, habitat use and food habits have been studied intensively in recent years (Johnson, 1970; Urban, 1970; Schneider et al., 1971; Fritzell, 1978b). Few studies, however, have examined raccoon foraging in relation to movements or habitat use. In the Prairie Pothole Region, raccoon movement patterns vary among age-sex groups and individual home ranges can exceed 4500 ha (Fritzell, 1978b). Raccoon foods are diverse, derived largely from wetlands, and show extensive use of bird eggs (Cowan, 1973; Greenwood, 1981). Fritzell (1978b) suggested that raccoon age and sex status may influence movement and impact on prey populations. The present study investigated raccoon movements and foraging activity by age, sex and habitat during the waterfowl nesting season in order to better understand the role of different age-sex goups in egg predation. A second purpose was to define foraging sites and foods used by raccoons for a better interpretation of fecal analysis (Greenwood, 1981). STUDY AREA AND METHODS The study was conducted from 15 April15 July in 1974-1976 on a 65-kM2, intensively farmed area in E-central North Dakota. The area was 76% cultivated, 12% wetlands, 10 % upland grassland, 1 % trees, <1 % farmyards and <1 % roads and trails (Greenwood, 1981). Raccoons were captured in live traps or nets and were equipped with continuous signal radio collars (Fritzell, 1978b). An individual animal was monitored continuously from a vehicle after it left its daytime bed in the evening until it arrived at a bed site again the next morning. Location was determined at 15-min intervals and recorded on a field map. Tracking was conducted at least 4 nights each week and was distributed among adults and yearlings; the same individual was seldom monitored on successive nights. When logistical problems were encountered during tracking, another raccoon was monitored for the rest of the night. Selected raccoons were shot at weekly intervals in 1976 after they had been tracked throughout the night; stomach contents were examined in order to relate foods consumed to foraging activities. Raccoon activity and habitat occupied were recorded at the onset of each 15-min sample period. Activity was interpreted from signal modulation (Urban, 1970; Sunquist, 1974) and change in animal location. All raccoon activity away from the bed site was apportioned among local movement (locomotor activity without detectable change