Seventy-two grade 8 students were randomly assigned to one of four experimental groups. One group studied five programed lessons in structural grammar, written without use of symbols or diagrams. A second group studied the same content presented with a symbolic notation to represent the grammar concepts. A third group studied the same content except that the symbolic notation of the second group was used in figural sentence-tree diagrams. The fourth group, a control group, studied five programed lessons in how to read poetry. The students studied the lessons in their regular English classrooms over a period of 1 week. Post-tests were given 2 days and 2 weeks after the last lesson. Test results showed that all three experimental groups learned and retained more knowledge of grammar than the control group. The lessons using symbolic notation and diagrams were each superior to the verbal presentation. Low ability students did not benefit from the diagrams and only high ability students mastered the strictly verbal presentation. The symbolic notation produced significant learning at all ability levels. The experiment seemed to show that the teaching of verbal concepts is facilitated by appropriate symbols and diagrams, provided students understand these nonverbal materials. (Author/LH)