It is obvious that adequate nutrition is essential for growth and development of school children, and many elementary schools in this country have already practiced it. Therefore, it would seem apparent that the school feeding program would have a significant effect on the growth and development of school children. This paper presents a two-year experimental school-feeding program from 1973 to 1974, and attempts to evaluate its effects by before-and-after nutrition surveys conducted in two elementary schools, one experimental and the other as a control. The two schools are both located in the same county (Yongin-Kun, Kyunggi-Do), and the families of their students are presumed to share the same socio-economic level. To assess the effect of school-feeding, we measured height, weight, chest circumference and grasping power. Physical examination was done foresigns of nutritional deficiency. A stool examination for parasites and blood examinations for hemoglobin, hematocrit and serum protein were included. Analyses were done for 150 students selected randomly at the beginning of the program. These students attended the school throughout the program period. Results are as follows: 1. The amount of increase of height, weight, chest cirumference and grasping power were greater in the experimental school than in the control school, but the differences are not statistically significant. 2. Signs of vitamin deficiency decreased in both experimental and control schools during the two-year program period. 3. At the time of the 1974 post-survey, values of Hb. & Hct. revealed no significant differences between the two schools, but serum protein level was a little higher than that of general Korean rural children of the same age. 4. Infestation rate of parasites had increased in both schools during the two-year program period. 5. Each student of the two schools was classified into three major classes, according to the level of economic condition of his or her parents, namely higher, middle and lower. The results of each class of the experimental school was compared with that of the corresponding class of the control school, expecting the relative magnitude of change largest in the lower economic class of the experimental school. However, change was greatest in the middle class, still not being statistically significant. Finally, the authors concluded that the two-year period for such a program is not sufficiently long for its beneficial effects to be demonstrated and measured. As long as the growth and development of children are concerned, planning with a more distant perspective is required, as well as the development of new methods of evaluation.