BECAUSE OF LIMITATIONS DURING THE ACTUAL STUDY, THIS PROJECT WAS DELIMITED TO THE STUDY OF THE EFFECTIVENESS OF USING MATERIALS ADAPTED FROM STRUCTURAL AND GENERATIVE GRAMMARS UPON THE ABILITY OF SEVENTH-GRADE STUDENTS TO WRITE MORE MATURE COMPOSITIONS. FORTY-THREE STUDENTS WERE GIVEN EXPERIMENTAL LINGUISTICALLY-ORIENTED MATERIALS DESIGNED TO TEACH THEM TO ANALYZE ENGLISH SENTENCES BY A STRUCTURAL GRAMMAR APPROACH AND TO CREATE AND COMBINE NEW SENTENCES BY A GENERATIVE-TRANSFORMATIONAL APPROACH. FIFTY STUDENTS IN THE CONTROL GROUP WERE GIVEN TRADITIONAL ENGLISH GRAMMAR INSTRUCTION. AS PRE- AND POST-TESTS OF WRITING MATURITY, ALL STUDENTS TOOK THE SEQUENTIAL TESTS OF EDUCATIONAL PROGRESS (STEP) WRITING TEST AND WROTE SAMPLE COMPOSITIONS WHICH WERE APPRAISED BY EMPLOYING THE T-UNIT LENGTH. THE RESULTS INDICATED THAT BOTH GROUPS MADE SIGNIFICANT GAINS IN ACHIEVEMENT DURING THE STUDY AND THAT, ALTHOUGH THE LINGUISTIC GROUP MADE SIGNIFICANTLY HIGHER SCORES ON THE POST-TEST, THE DIFFERENCES IN THE GAIN BETWEEN THE EXPERIMENTAL AND CONTROL GROUPS WAS NOT STATISTICALLY SIGNIFICANT. THEREFORE, NO SIGNIFICANT DIFFERENCE IN STUDENT ABILITY TO WRITE COMPOSITIONS WAS SEEN BETWEEN STUDENTS WHO HAD MATERIALS BASED UPON STRUCTURAL AND GENERATIVE GRAMMARS AND STUDENTS WHO HAD TRADITIONAL LATINATE GRAMMAR INSTRUCTION. IT IS FELT, HOWEVER, THAT ANOTHER STUDY, WITH THE SAME THEORETICAL BASIS AND WITH MORE CAREFUL CONTROL OF SOME FACTORS, WOULD SHOW A SIGNIFICANT POSITIVE CORRELATION. (AN APPENDIX CONTAINS THE INSTRUCTIONAL MATERIAL AND EXERCISES GIVEN TO STUDENTS IN THE EXPERIMENTAL LINGUISTICALLY-ORIENTED GROUP.) (AUTHOR/DL)