Background: In both general education and physical education literature, considerable attention is paid to identifying the effectiveness of communities of practice (CoPs). This literature tends to be more qualitative in nature with less research examining the quantitative nature of interactions in CoPs through direct observation. Purpose: This study examined the ways teachers interact within a newly-developed community (LC) /CoP to understand how conversations evolve; how teachers question, challenge, and/or encourage one another to grow professionally in a learning community. Theoretical Framework: Situated learning perspectives provide a powerful framework for examining teacher learning and facilitation of teacher development. Method: Bales's Interaction Process Analysis Tool (IPA) was selected for analyzing the socio-emotional and task-oriented communication between members of the LC/CoP. Six physical education teachers (four female, two male) and a facilitator, representing the university, participated in LC/CoP meetings designed in collaboration between teachers and the facilitator. The data source was audio-taped transcripts taken of teacher interactions during six weeks of LC/CoP meetings. Findings: Results revealed three main interactive patterns that emerged from analysis of participant interactions during the six LC/CoP meetings. These patterns were; 1) Variable member interactions; 2) Member roles; and 3) Personal interaction pattern development. By coding participant interactions as they occur allowed us to understand and analyze directly what and how LC/CoP members actually talked at the initial stages of community development. Conclusion: The contribution of this study is to use an interaction analysis tool to conceptualize the communicative actions that emerge through a LC/CoP and provide practical insight into the gap in the literature relative to how conversations evolve in a CoP; how teachers question, challenge, and/or encourage one another to grow professionally. Ultimately linking an interactive analysis tool with a well-structured qualitative research component to provide participant voice, feelings and behaviours would provide more robust rusults.