This article begins by synthesizing findings from observational classroom research on corrective feedback and then presents an observational study of patterns of error treatment in an adult ESL classroom. The study examines the range and types of feedback used by the teacher and their relationship to learner uptake and immediate repair of error. The database consists of 10 hours of transcribed interaction, comprising 1,716 student turns and 1,641 teacher turns, coded in accordance with the categories identified in Lyster and Ranta's (1997) model of corrective discourse. The results reveal a clear preference for implicit types of reformulative feedback, namely, recasts and translation, leaving little opportunity for other feedback types that encourage learner‐generated repair. Consequently, rates of learner uptake and immediate repair of error are low in this classroom. These results are discussed in relation to the hypothesis that L2 learners may benefit more from retrieval and production processes than from only hearing target forms in the input.