Effective use of water (EUW) was evaluated along with grain yield in the same environment over different years and locations. The root system sizes of sets of winter wheat varieties (20, 20, 20 and 18 varieties in 2007, 2008, 2009 and 2010, respectively) were evaluated by measuring their electric capacity. The RSS results were compared with the yield and grain quality in official variety trials at 20, 22, 23 and 23 stations in 2007, 2008, 2009 and 2010, respectively. All varieties were tested under two conditions with low and high inputs, and in some years they were also tested after late sowing and after unsuitable forecrops at a few stations. The RSS was mostly controlled by the location; the varieties differed significantly, explaining 2.6–15.6% of the variation in RSS, while 1.7–3.9% of the total variation was unexplained. In 2007, a dry year, the RSS was significantly correlated with yield in both variants (r2=0.285* and 0.284*, respectively) and with yield of starch (r2=0.248*) at all stations. In 2008, the highest-yielding year, the correlation of RSS with yield at the three stations at the lowest altitude and with the highest temperature was negative (r=−0.459*). In 2009, the yield was highly significantly correlated at only five locations in medium altitudes, where drought occurred, only in the high input treatment (r2=0.391**). No relationship between RSS and yield was found in the wet year 2010. The third of the varieties with the greatest RSS had an RSS that was 21% greater than those of the one-third of varieties with the smallest RSS in 2007 and a grain yield 420kgha−1 higher. The difference responded to additional use of about 7mm of subsoil water. The most contrasting varieties showed a yield difference of 860kg, i.e., approximately 15mm of subsoil water. We conclude that differences in the EUW expressed as different yields under the same conditions can be partly attributed to different RSSs (probably due to deeper rooting) and can be improved by breeding.
