This article reports a classroom-based study of the effects of two types of corrective feedback (CF) on the acquisition of 3rd person "-s." One hundred and nine Chinese university students completed three communicative tasks: One group received implicit CF consisting of a single corrective move (implicit recasts), a second group received explicit CF consisting of a dual corrective move (a prompt followed by a more explicit recast), a third group performed the tasks without feedback, and a control group just completed two tests. The groups completed two pre-tests, immediate post-tests and delayed post-tests. Gains in accuracy in a test of procedural knowledge (an elicited imitation test) and in a test of declarative knowledge (an untimed grammaticality judgement test) were evident in the first three groups but with few differences among the test scores of these groups. No gains occurred in the test control group. Explicit feedback involving a dual feedback move elicited more uptake-with-repair but this was unrelated to gains in accuracy and overall the implicit CF involving a single feedback move was as effective as the explicit. This finding, which was unexpected, may reflect the Chinese learners' general orientation to form in this particular classroom context.