Two experiments examined the role that initial feeding experiences with varying consequences have on the subsequent food selection behaviour of young chicks, Gallus gallus. In experiment 1, chicks were given a single feeding experience with food or sand having either beneficial (food for chicks aged 2·5 days), neutral (food for chicks aged 0·5 days, or sand for both age groups), or aversive (lithium chloride injection for both age groups) long-term consequences. Chicks aged 0·5 or 2·5 days did not discriminate, between the two substances during a later pecking session. If the long-term consequences of the initial pecking experience were beneficial, the chicks increased their subsequent pecking regardless of whether they received the same substance as initially presented or the other substance. Similarly, if the long-term consequences were aversive, the chicks decreased their subsequent pecking. Experiment 2 showed that following differential feeding experience with two foods, chicks aged 2·5 days did discriminate between these substances on the basis of the long-term consequences of ingesting each substance. Chicks received three sessions with one food prior to the pairing of aversive consequences with ingestion of a second food. The chicks decreased their subsequent pecking to the food that had been paired with illness while continuing to peck at the other food, as well as at a novel food. Initially, chicks associate the long-term consequences of ingestion with general feeding-related cues such as pecking (experiment 1). With additional feeding experiences, however, chicks associate cues that are food-specific with the long-term consequences of ingestion (experiment 2), thus allowing young chicks to select substances having beneficial consequences and to reject those having aversive consequences.