The effectiveness of two courses designed to train 96 regular classroom teachers in rural areas to teach mildly handicapped children in their classrooms was examined. The courses covered teaching methods and materials and included self instructional materials. A posttest only control group design was employed. Data were collected on four measures: a criterion referenced test covering each course, a classroom observation and recording form, and a teacher inventory. Results indicated that teachers in the experimental group made significantly fewer errors on the criterion referenced tests and used more of the recommended techniques than teachers in the control group. No significant differences were found in perceived job satisfaction or teaching effectiveness. (Among 13 appendixes are sample criterion referenced tests and rating forms.) (CL)