Abstract: The double winter wheat and summer maize rotation cropping system in the North China Plain (NCP) is mainly limited by water shortage. To solve this problem, a land fallow policy has been introduced in NCP, especially in the groundwater-over-exploited region. A three cropping system in two years which includes winter wheat, summer maize and spring maize was introduced in place of the double cropping system in a year. Meanwhile, grain yield of spring maize is mainly limited by environmental (e.g., precipitation and temperature), biological (e.g., released cultivar) factors. Improving spring maize yield and water use efficiency using optimal sowing date that is adapted to maize cultivar characteristics is critical in solving this problem. A field experiment was conducted in Nanpi Eco-agricultural Station of Chinese Academy of Science in 2016, consisting of 5 maize cultivars (HN887, HX338, ZD958, HN866 and LC1) and 3 sowing dates (May 1, May 15 and May 30) treatments. The phenological stages were observed, and soil water content in the 2 m soil profile, grain yield and yield components were measured and weather factors recorded for the period of study. The results showed that the days from sowing to maturity of the 5 cultivars all significantly reduced with sowing date delay, which mainly reduced the days from sowing to tasselling stage (P < 0.05). For different cultivars, the days from sowing to maturity were significantly different and the differences were similar (5-7 d distance) for all three sowing date treatments. There was no significant difference between averaged grain yield for 5 cultivars for May 1 and May 15 sowing date treatments. May 30 sowing date had the highest grain yield, which was mainly caused by the higher 100-grain weight (3.3 g higher than others). For water consumption, there was no significant difference among 3 sowing dates because there was no irrigation during the growth stage for all treatments and precipitation was 405.2-416.0 mm. There was also no significant difference in water consumption among different maize cultivars. The trend in water use efficiency (WUE) was also similar to that in grain yield, which increased with sowing date delay. There were significant differences in WUE among different sowing dates and maize cultivars. Meanwhile, the correlation was significant between 100-grain weight and precipitation before and after tasselling and then between 100-grain weight and accumulated temperature before and after tasselling. The comprehensive analysis of the grain yield, water uptake and water use efficiency showed that the optimal sowing date was May 30, with 'HN866' and 'HN887' as the cultivars with the best performance in the study area.