Knowledge in general, and tacit knowledge in particular, has been hailed as an important factor for successful performance in knowledge-worker teams. Despite claims of the importance of tacit knowledge, few researchers have studied the concept empirically, due in part to the confusion surrounding its conceptualisation. The present study examined the acquisition and sharing o f tacit knowledge and the consequent effect on team performance, through social interaction and the development of a transactive memory system (TMS). TMSs are important for the acquisition and sharing of tacit knowledge, since they enact ‘collective minds’ of teams, and are also a factor in successful team performance. In order to conduct this research, a team-level operational definition of tacit knowledge was forwarded and a direct measure of tacit knowledge for software development teams, called the Team Tacit Knowledge Measure (TTKM ) was developed and validated. To investigate the main premise of this research an empirical survey study was conducted which involved 48 software development teams (n = 181 individuals), from Ireland and the UK. Software developers were chosen as the example of knowledge-worker teams because they work with intangible cognitive processes. It was concluded that tacit knowledge was acquired and shared directly through good quality social interactions and through the development of a TMS. Quality of social interaction was found to be a more important route through which teams can learn and share tacit knowledge, than was transactive memory. However, transactive memory was not a mediator between social interaction and team tacit knowledge, indicating that both provided separate contributions. Team tacit knowledge was found to predict team performance above and beyond transactive memory, though both were significant. Based on these findings recommendations were made for the management of software development teams and for future research directions.