Increasing the harvest index of wheat in the high rainfall zones of southern Australia

The harvest index (HI) of wheat in the high rainfall zones (HRZ) of Australia is low compared with values in similar rainfall environments in Europe. From 2005 to 2007, we examined dry matter (DM) partitioning, yield formation, post-anthesis dry matter accumulation, and remobilization of water soluble carbohydrates (WSC) of a number of wheat genotypes (5 spring wheat and 1 winter wheat) to determine the factors influencing the HI in the HRZ of south-western Australia. Grain yield was strongly correlated with grain number (GN) in all three years and with the sum of post-anthesis DM accumulation and WSC in stems at anthesis when the amount of assimilates available to fill grains was less than 600 g m−2. However, when the amount of assimilates exceeded 600 g m−2, grain yield reached a plateau and did not increase with the increased availability of assimilates because of sink limitations. Genotypes with more GN tended to extract more WSC to fill grains than ones with fewer GN in 2005 and 2007. GN increased with increased spike DM at anthesis, the number of grains g−1 of spike DM, and the duration of spike growth, but was not correlated to the fraction of DM partitioned to the spike at anthesis for individual genotypes and across the 3 years. Compared to winter wheat in the UK, Australian spring and winter wheat had a lower number of grains g−1 spike DM and g−1 chaff DM. HI was positively related to WSC at anthesis and post-anthesis DM accumulation in the drought year of 2006, but not in average (2007) and above-average rainfall (2005) years. HI was negatively related to the proportion of stem DM to above-ground dry matter (AGDM), but positively related to the proportion of spike DM to AGDM. Dry matter partitioning indicated that the proportion of stem DM to AGDM and chaff DM to AGDM was 20% and 50% higher than that of winter wheat grown in the UK, respectively. This was related to a lack of remobilization of WSC and a net increase between spike DM at anthesis and chaff DM at maturity. Therefore, we suggest that increasing the number of grains m−2 by breeding for more grains g−1 spike DM, reducing dry matter partitioning into stem and chaff, and remobilizing WSC can lead to an improved HI. We conclude that the yield of wheat can be increased without increasing AGDM and the potential HI can reach 0.50 in the HRZ of south-western Australia if the number of grains m−2 can be increased.

► We examined yield, harvest index (HI) and dry matter partitioning of wheat. ► Grain yield was strongly correlated with grain number and increased with increased assimilates (source) up to 600 g m−2 and reached a plateau when the amount of assimilate exceeded 600 g m−2. ► Grain number increased with increased spike dry matter at anthesis, the number of grains g−1 of spike dry matter, and duration of spike growth, but was not correlated to the fraction of dry matter partitioned to the spike at anthesis or crop growth rate. ► Dry matter partitioning indicated that a high proportion of dry matter was allocated to stem and chaff, and that water soluble carbohydrates (WSC) left in stems at maturity. ► Increasing sink strength to remobilize WSC in stem and reducing dry matter partitioning into structural stem and chaff may lead to an improved HI.