The purpose of this paper was to report on the findings of a study examining students' online interaction patterns. The context of the study was a graduate online class delivered via Blackboard[R]. The primary data for the study came from students' discussion board postings, online learning journals, and course grades. Various data analysis methods such as descriptive and regression analyses were utilized to examine students' evolving interaction patterns and different interaction patterns among students in the same class. Results of the study indicated that there was considerable variability in students' postings. Students' postings were found mostly heterogeneous across students and across modules. The study suggested no correlation between the number of posts and students' success. (Contains 3 tables and 3 figures.)