Erlam & Ellis (2018) published, in Canadian Modern Language Review, an experimental study that investigated the effect of input-based tasks on the acquisition of vocabulary and markers of plurality by adolescent near-beginner learners of L2 (second language) French. The present paper reports an approximate replication of the original study with the aim of confirming or disconfirming the results.1 The research questions of both studies addressed the receptive acquisition of new vocabulary and the receptive and productive acquisition of markers of plurality resulting from instruction using input-based tasks. Both studies investigated near-beginner adolescent learners of French. The teacher, the students’ usual classroom teacher, was the same in both studies. In the replication study, a new, larger group of students were investigated, the length of the instruction was increased, involving the development of additional tasks, and productive as well as the receptive knowledge of the vocabulary items was assessed. The results of the replication study confirm and extend those of the original study. The teachers’ views about the role of input-based tasks with near-beginner learners remained constant in the two studies. The paper concludes with a discussion of the contribution that approximate replications can make to instructed second language acquisition research.