A hypothesis that artificial reforming of corn canopy could improve solar light penetration and dry matter production was tested in corn fields (var. Suwon 19) with three planting densities; low (60 ${ imes}$ 40cm), medium (60 ${ imes}$ 24cm) and high (60 ${ imes}$ 16cm). Natural canopy was found that leaf orientations were even over all azimuth but somewhat inclined toward north-south direction and leaf angle ranged 38$^{circ}$ to 71$^{circ}$ from horizontal surface. Reforming corn canopy included following treatments: 1) natural canopy planted in north-south rows (natural canopy), 2)east-west plane canopy planted in north-south rows (E-W canopy), 3)east-west plane canopy and upright leaves in north-south rows, 4)north-south plane canopy (N-S canopy) in east-west rows. After corn plots were installed with training system by supporting poles and connecting wires, corn leaves were induced to a reforming direction and tied on wire. Average light intensity at the mid-point of plant height showed 5-10% increases in E-W canopy and in E-W canopy plus upright leaves, but a 2-10% decrease in N-S canopy from natural canopy. At yellow ripe stage, total dry wt. was increased in E-W canopy but not in N-S canopy. The E-W canopy produced 3-10% more grain yield than natural canopy. Though E-W canopy plus upright leaves yielded less at low density, it yielded up to 10% more at higher density. The N-S canopy yielded similar to low compared with natural canopy. These results suggests that reforming canopy toward solar incident direction increases light penetration into lower canopy, photosynthetic efficiency and grain yield, especially at high planting density in corn.