Components of grain yield and total nonstructural carbohydrates in wheat : a thesis submitted for the degree of Doctor of Philosophy in the University of Canterbury [Lincoln College]

Abstract

Responses of autumn-sown wheat crops to various agronomic inputs were investigated in three field experiments over two seasons, and the causes of yield variation were analysed by simultaneous carbohydrate measurements in various plant parts beginning from shortly after double ridges to maturity. In the first season (1979-80), Experiment 1 was conducted with the semidwarf cultivars Oroua and Rongotea and the New Zealand standard height cultivar Kopara, with different times and rates of N application. In the second season (1980-81), Kopara and Oroua were studied with three rates of N applied at tillering, two plant populations, and with (Experiment 2) or without (Experiment 3) irrigation. Rongotea produced the highest and Oroua the lowest grain yields in the first season. This superiority of Rongotea was attributable to its largest grain size. Kopara was similar yielding to Rongotea and this was achieved by superior grain set and large grains. The strength of Oroua in yielding equal to Kopara in all experiments mainly lay in its high tillering and ear producing capacity. Both Kopara and Rongotea were high dry matter (DM) producing cultivars, but the semidwarf cultivars were superior in directing more of their DM to the grains relative to Kopara. Also, the semidwarf cultivars contained higher TNC (total nonstructural carbohydrates) levels in the vegetative parts, mainly stem, than Kopara at all stages of growth, including at maximum stem weight after anthesis. Stem TNC levels reached a peak at maximum stem weight after anthesis, and subsequent decline in stem weight was concurrent with the decline in stem TNC concentrations. Calculations of vegetative reserve contribution to grain filling by TNC difference between maximum stem weight and maturity showed that the semidwarf cultivars contributed more to grain weight than the standard height cultivar. Fructosans were the predominant TNC fraction in stems during grain filling. Their amount in the stems and translocation seem to be important for grain size and yield. The two highest grain yielding sites of Experiments land 3 also produced the highest total dry matter, but the former site yielded less grains because of more vegetative growth relative to reproductive growth. These effects were reflected in low TNC concentrations in stems in the first experiment. N application did not influence DM production in any experiment, nor grain yields of Experiments 1 and 3, and it had a general dilution effect on stem TNC concentrations. This is indicative of an excessive N availability to the plants, which tended to disrupt internal carbohydrate balance, in presence of adequate soil moisture. In Experiment 2, which suffered from moisture stress, grain yield increased linearly with increasing rates of N application (up to 100 kg N ha⁻¹), which was associated with an increase in grain number per plant. Early N applied at the tillering stage in Experiment 1 (0 and 80 kg N ha⁻¹) increased ear number and grain number per plant, but these effects were offset by decreased weight per grain. A low plant population (250 plants m⁻²) improved various yield attributes at both plant and tiller levels, whereas a high population (380 plants m⁻²) tended to produce a larger yield. In a low population, tillers jointly contributed to 50% or more of the plant's yield, but in a high population, MS (mainshoot) was the dominant component. MS always had the highest TNC concentration corresponding to its superior yield attributes. Irrigation vastly improved grain yield (53%) through its favourable effects on tiller survival, ear production and grain number per ear, although weight per grain was depressed. Water stress increased stem TNC concentrations at various growth stages and relocation of vegetative reserves to the grains substantially. In this study, no particular associations between tiller production or grain set and plant TNC concentration at relevant growth stages were found. On the other hand, tiller survival appeared to be related to plant TNC concentrations at the active and maximum tillering stages. MS study in the second season revealed a strong correlation between grain set and TNC concentration, predominantly SS (simple sugars and starch, if any), in young ears before emergence. Stem TNC concentrations declined markedly between the boot stage and anthesis which is suggestive of a shortfall in current assimilation due to increased demand for carbohydrates by the rapidly growing ear and stem. It is suggested that conditions restricting the supply of carbohydrates to the ear both before and after anthesis should be avoided to secure maximum grain yields. The results are discussed for practical significance in relation to the influence of various treatments, sites and season.