There are two key categories of moves that drive interaction in science classrooms; initiating moves and rejoinders. Closed-ended initiating questions have always been associated with low student engagement and open-ended initiating questions with prolonged interaction in science classroom. In this paper, we report on a qualitative case study conducted in South Africa with three physical sciences teachers. The purpose of the study was to understand the nature of initiating and rejoinder moves and their effect in the science classrooms. We draw on a hybridised framework combining Analysing Teacher Moves (ATM) with teacher discursive moves. We present data that contradicts the claim that closed-ended initiating questions always lead to minimised interaction and that only open-ended initiating questions lead to prolonged interaction depending on the nature of rejoinders. We discuss the value of understanding this intricacy of teacher move patterns and its implications for teacher education and professional development of science teachers.