The study area is on the eastern edge of the Great Bend Prairie physiographic province of Moore (1930). The climate is characterized by moderate precipitation (ann. 30"), a wide range of temperature variations, moderately high wind velocities, and comparatively rapid evaporation. The summers are generally hot, and the winters are moderately cold but are free from excessive snowfall. The weather during the study period was unusually dry, and the summer temperatures were above normal. A drought had begun in 1952, following the cool and wet summer of 1951. The study area includes the zone of transition from bluestem or tall-grass prairie to the buffalo grass or short-grass prairie. The principal farm crop in the study area is wheat. Sorghum grain, oats, hay crops (especially alfalfa), and corn are also grown. The study area supported a small population of breeding crows; an estimate based on field observations mainly in eastern Harvey County, was not more than one pair per square mile. In winter a large population of crows migrates into the area from the northern Great Plains. Censuses showed that on parts of the area the feeding population might be as great as 180 birds per square mile. These wintering crows concentrate in the western part of the study area where the flat, fertile wheat fields of central Harvey County are replaced by sand dunes and the sandy Arkansas River Valley. Here much land is devoted to raising livestock, and sorghum grain is an important field crop. There is also more waste land there than elsewhere in the area. Data From Analysis of Pellets Data obtained from the analysis of pellets were grouped in biweekly collections, and percentages of various food residues in the pellets collected within each biweekly period were averaged. Also frequency of occurrence was computed, and maximum and minimum percentages were included to permit a broader interpretation. In determining the minimum percentage, only those pellets were considered in which the food residue was present. Pellets from roosts of resident crows were collected on a year round basis in eastern Harvey County near Newton (see tables 1 and 2). The data from these pellets were interpreted separately from data on collections made in the western part of the study area from under roosts of wintering crows (see tables 3 and 4). In studies of the food of owls analysis of materials in regurgitated pellets has been widely used, but with crows this method has been little used because the nature of their food makes identification of material more difficult. Analysis of pellets has certain merits, however, and, if closely correlated with field studies, can give valuable information concerning food habits. The availability of pellets and the ease of collecting them are obvious advantages. Under large roosts in winter the number that can be collected is almost unlimited. At other seasons, pellets are scarcer, but even so they usually are more available than stomachs.