Phenotypic variation and correlation analysis of major agronomic traits ofwheat recombinant inbred lines population under different potassium levels

Potassium (K) is one of the most important nutrient elements for wheat growth. In China, K mineral is in acute short supply and generally deficient. Hence K is an important limiting factor for agricultural production in China. Deepening our under-standing of genetic correlation among agronomic traits under different K treatments was critical for genetic improvements in wheat under soil K deficiency. Main agronomic traits of plant height (PH), spikes number per plant (SN), spike length (PL), spikelets num-ber per spike (SPI), grains number per spike (GNS), 1000-grain weight (TGW), kernel length (KL), kernel width (KW) and grain weight per plant (GW) of 131 “Shannong 483 × Chuan 35050” recombined inbred lines (RIL) were investigated in a pot trial with three K treatments (0 g·kg^?1, 0.1 g·kg^?1 and 0.3 g·kg^?1 of K₂O) in 2008—2009. The traits were analyzed for variance and correlation.Results showed that the frequencies of PH, SN, PL, SPI, GNS, TGW, KL, KW and GW in RIL population were approximate in nor-mal distribution in the 2-year 3-K treatments. Broad-sense heredities (H_b²) of the nine agronomic traits were 80.6%, 44.7%, 69.9%, 71.4%, 76.0%, 74.8%, 64.9%, 40.0% and 33.2%, respectively. Phenotypic variations among PH, SPI, GNS and KL were mainly attributed by genotypes traits. TGW and KW were mainly influenced by interactive effects of genotype and year. PL was mainly driven by genotypes and interactive effects of genotype and year. SN was influenced by interactive effects of genotype and K dose. Then GW was controlled by interactive effects of genotype, year and K dose. Significant positive correlations existed among GW and SN, PL and GNS under different K treatments. However, the correlation between GW and TGW was not stable across the treat-ments. TGW was not related or negatively related with SN and GNS (r=?0.07~?0.28). Also the correlation between SN and GNS was unstable (r=?0.03~0.27). This suggested that TGW, SN and GNS were weakly correlated and independent of genetic processes. No firm trade-off existed among important traits during crop breeding. The correlations among GW and SPI, GSN, TGW, KL and KW were significantly influenced by time.