Seed yield of grass is determined by environmental and cultivation factors. The aim of this study was to determine the effect of sowing rate and row spacing on germination in establishment year and development of generative tillers and seed yield of redtop (Agrostis gigantea Roth = A. alba) in the two consecutive production years. Two replicate experiments were performed for 3 years in the same locality in Poland (51°54′N; 17°12′E). The first experiment was established in 2005, with seed production for two successive years (2006 and 2007). The entire experiment was repeated, being established again in 2007 and seed production measured in 2008 and 2009. In both experiments, redtop was sown with row spacings of 24 and 48 cm (factor 1), in sowing of 0.8, 1.6 and 2.4 kg ha −1 (factor 2). The study indicated that at the smallest sowing rate 0.8 kg ha −1 and row spacing 24 cm, field emergence index was the highest (26.3%). The lack of response of the total seed yield from two consecutive production years to sowing rate indicates that in redtop grown for seed, sowing in the amount of 0.8 kg ha −1 is sufficient. Increasing the row spacing from 24 to 48 cm resulted in a reduction in the density of generative tillers in the first and second production years. The row spacing of 48 cm was more favourable for seed yield in the first year, whereas 24 cm in the second year. In view of the increase in the total seed yield from two consecutive production years by 9.5% at the row spacing of 24 cm as compared with 48 cm, the row spacing of 24 cm should be regarded as recommended. Also, higher green matter of autumn regrowth can be obtained at a row spacing of 24 cm than at 48 cm.