We hypothesised that mixing of rice lines with different maturity dates can improve the productivity per unit area while improving lodging resistance on account of the heterogeneous canopy. To test this hypothesis, we grew two lines, wild-type (WT) and early-maturing (EML), of each of two cultivars (Koshihikari and Hitomebore) in monoculture plantings and within-cultivar mixed plantings, with WT and EML plants alternating every one or two rows, over 2 years in northern Japan. The mixtures formed a canopy with different heading dates (5–7 days earlier than WT in Hitomebore and 14–19 days earlier in Koshihikari). For Koshihikari, the mixture increased grain yield per plant in the WT by 18–33%, but decreased it in the EML by 19–22%; however, for Hitomebore, there was no significant difference. Consequently, grain yield per unit area did not change relative to the mean of monocultures of WT and EML in either cultivar. Lodging tolerance improved significantly in the mixtures, especially for Koshihikari. Thus, mixing rice lines with different maturity dates improved lodging tolerance without decreasing productivity.