This study aims to describe the gesture use and multimodal behavior of future French teachers in Lyon, France during videoconferenced pedagogical interaction with learners of French in Dublin, Ireland. These multimodal conversations were recorded, transcribed and annotated using an annotation scheme that was designed for this multimodal corpus. The use of gesture was measured and compared quantitatively during sequences of incomprehension and sequences in which there was no manifest incomprehension between participants to measure the extent to which gesture was used to repair incomprehension. Results show that contrary to expectations, teacher trainees did not use gestures to repair incomprehension.