The effect of a meaningful instructional context and practice variety on rule retention was investigated. One hundred high school students participated. The design was a 2 by 3 completely randomized factorial design. The instructional context factor had two fixed levels: context and rule. The practice variety factor had three fixed levels: broad-, narrow-, and no-practice. Students learned computer programming rules either from a context or a rule booklet on Day 1 of the experiment. Students in the broad and narrow practice groups received either a broad or a narrow practice booklet consisting of ten practice instances with feedback on Day 2. All students took a posttest 11 days after initial learning. An analysis of variance on retention scores found both the broad and narrow practice groups scoring significantly higher (p < .01) than the no-practice groups. No significant main effect for instructional context was found. It was concluded that practice instances facilitates rule retention. (Author/SD)