In order to study the effects of planting pattern and plant density on yield of soybean genotypes, an experiment was carried out in Safiabad Agricultural Research Center, Dezful, Iran in 2011. The experimental design was a factorial strip plot based on completely randomized block design with 3 replications. Planting pattern consisted of 3 levels (60 cm with 1 planted row, 75 cm and 90 cm with 2 planted rows) as vertical factor, a factorial of plant density (30, 40 and 50 plant/m²) and genotypes (504, L14 and DPX) was considered as horizontal factors. The means of planting pattern × density showed that the highest number of pods/plant (52.85) belonged to 60 cm with single row of plants and 40 plant m^-2. The lowest (28.35) number of pods/plant was observed in 90 cm with double rows planted and 50 plant m^-2. The genotypes responded to planting density differently. The highest (4936.11 kg/ha) grain yield belonged to DPX genotype with 50 plants m^-2. The lowest (4936.11 kg/ha) grain yield was achieved at L14 genotype with 30 plant m^-2. The results of this expriment suggested that planting pattern of 60 cm with single row of plants, 50 plant m^-2, DPX and 504 genotypes were probably suitable for this area.