Genotypic variation in grain phosphorus concentration, and opportunities to improve P-use efficiency in rice

Continual removal of phosphorus (P) from fields in rice grains at harvest results in lower soil fertility in low-input farming systems and drives the need for fertiliser inputs in high-input farming systems. High-P content in rice grains (the majority as phytate) contributes little to human nutrition and agronomic practices such as growing seedlings in high-P media or seed P coating at sowing (in direct-sown crops) may overcome the reported need for high-P seed for seedling establishment. Thus, reducing the amount of P in rice grains at maturity through breeding may represent a novel means to reduce ‘mining’ of soil P. We investigated the uptake and partitioning of P in rice plants and examined the scope for breeding rice with lower grain P by assessing genotypic variation for phosphorus harvest index (PHI) and seed P concentrations among a set of 38 rice genotypes in the field. At maturity approximately 75% of total plant P was found in grains and translocation of P from stems and leaves contributed substantially to grain P. However, unlike other cereals such as wheat, rice plants continued dry matter and P accumulation until maturity with approximately 40% of total plant P taken up post anthesis. In the field study, PHI differed significantly among genotypes (from 57 to 87) but was highly correlated to HI (P ≤ 0.001), suggesting that exploiting genotypic variation for this trait may be counterproductive. Grain P concentrations varied from 1.96 to 3.18 mg P g−1, and were neither associated with reductions in grain yield or seed size, nor significantly correlated to HI. Grain P concentration therefore appears to be a suitable screening criterion and the 50% variation observed among genotypes suggests that scope exists for breeding rice with lower grain P concentration to reduce the off-take of P from rice fields at harvest.