The present research related the feeding ecology of seven rodent species to the reactions of laboratory-reared and prey-inexperienced members of each species both to live prey and to an artificial moving stimulus predicting food pellets. Feeding ecology was determined by the degree of carnivory, based on reported stomach contents and observations of feeding. Experiment 1 assessed predatory reactions to a live cricket placed in each animal's home cage. Killing and latency of eating the cricket were directly related to the degree of reported carnivory on moderately fast-moving arthropods. Experiment 2 examined behavior toward a rolling ball bearing that predicted delivery of food. Average percentages oftrials with approach or contact of the bearing, and the conditional probability of a mouth contact were all positively related to the degree of reported carnivory and to cricket predation in Experiment 1. In addition, the topography of ball bearing contact for a species often resembled its topography of cricket contact. We conclude that (1) rodent predatory behavior can be studied in the laboratory using appropriate artificial stimuli and prey-inexperienced subjects, and (2) the predatory behavior of a species is based on underlying appetitive organization related to carnivory, including differential sensitivity to stimulus movement, motor preorganization, and susceptibility to conditioning. This appetitive organization appears to influence responding to both live and artificial prey.Prey-naive laboratory rats (Rattus norvegicus) interact vigorously with a rolling ball bearing that predicts the delivery of food (Boakes & Jeffery, 1979;Timberlake, 1983a;Timberlake, Wahl, & King, 1982). The responses directed to the bearing resemble those shown by rats in predation on insects, including digging, chasing, seizing, carrying, retrieving, biting, and turning in the forepaws (Karli, 1956; Timberlake, personal observation, 1977). In short, rats appear to treat the moving bearing as a prey item, a conclusion also supported by the effect of previous poisoning in preventing eating but not predatory responses to the bearing (Timberlake & Melcer, 1988). Timberlake and Melcer (1988) argued that their results show that the determinants of predatory behavior can be studied in the laboratory using artificial rather than live prey. Timberlake (1983aTimberlake ( , 1983bTimberlake ( , 1984 and Timberlake and Lucas (in press) further argued that such organized predatory reactions to artificial stimuli by preyinexperienced rats reflect the contribution of an underlying species-typical, food-getting organization to the results of conditioning procedures. Such an appetitive organizaThis research was supported by National Science Foundation Grants 82-10139 and 84-11445. We thank Frank Adams and James Dougan for assistancein analysis, Moshe Yuchtman for his translation, andValeri Dougan and Gary Lucas for critical comments. We are especially indebted to David Gubemick, John King, Elise Steinberg, Miwako Tamura, and Connie Warner for assistan...