The present investigation was a partial replication and extension of Saks' (1977) study on the effects of group size on mock jury decision making. Mock juries of size 4, 6, and 12 were formed by randomly assigning 232 student volunteer subjects to one of 30 groups, or to the condition where subjects worked alone (N=12 individuals). The case used was a written transcript adapted from Saks (1977). After reading the transcript, groups deliberated until they reached a verdict. Overall, it was found that groups of 12 came to the correct verdict more often and deliberated the longest, groups of 6, unexpectedly, produced the most hung juries, and the individuals recalled the least amount of the testimony. Groups of 4 perceived their group as being the most fair and were also the most satisfied with their group's decision. They also rated their influence on the decisions made by other members of the group as the highest. In all groups, there was a shift in the pre- and post-deliberation guilt ratings toward the group verdict, indicating an effect of group polarisation, and discussion also increased individuals' confidence in their rating of guilt, providing some support for the model of group influence proposed by Myers and Lamm (1976). In general, the results support previous findings and add to the growing literature which suggests that the effect of jury size is a complex phenomenon requiring more investigation.