Email is considered as one of the most widely accepted computer-mediated communication tools among university students. Evidence from the present literature shows that students make a significant amount of their course-related communications (e.g. discuss a topic with peers) using this tool. This study explores the dynamics of an email communication network, which was evolved among 34 university students throughout a semester, using measures of social network analysis and network simulation. These 34 students were doing a masters-degree course. They made 621 course-related email communications throughout the semester which consisted of 15 weeks including 13 semester-weeks, 1 week for mid-semester vacation and 1 week vacation before the final examination. From the analysis of this email communication network, it is found that: (i) students make an increased number of email communications with their peers at the end of the semester compared to the beginning of the semester; (ii) students' communication network becomes sparse or decentralised over time during a semester; (iii) students have different levels of network participation at different times during a semester; and (iv) the reliabilities of the predictive power of "reciprocity" (i.e. an actor's tendency of making reciprocal relations with other actors of the network), "indegree-activity" (i.e. effects of an actor's present "indegree" on its future "outdegree") and "outdegree-activity" (i.e. effects of an actor's present "outdegree" on its future "outdegree") parameters of simulation models are changing significantly throughout the semester. Interpretations of these findings are also discussed in this paper. (Contains 6 tables and 1 figure.)