Determination of seed number in sea level quinoa (Chenopodium quinoa Willd.) cultivars

The identification of the critical period for seed determination is a central aspect for adequate crop management for higher yields. As important variability in the developmental stages associated to this critical period and its duration has been detected for several crops, results from one species cannot be extrapolated to other. The flowering period, from first anthesis to the end of flowering, appears as the most sensitive to the environment in quinoa; as results from experiments conducted under field conditions in the Argentinean pampas, using sea level quinoa cultivars adapted to temperate environments, suggest. Data from a combination of four cultivars, three densities and 2 years were used to estimate the association (adjusted R2) between crop biomass increment (g m−2) and seed number (seeds m−2) for four phases: vegetative, reproductive, flowering and seed filling. The strongest association was that detected for the flowering phase (R2 = 0.71), followed by seed filling (R2 = 0.64); associations for the two phases preceding first anthesis were low. Differences in biomass increment during flowering where strongly associated with crop growth rate (g m−2 day−1) during flowering (R2 = 0.86); crop growth rate was associated to differences in the amount of intercepted photosynthetically active radiation (MJ m−2 day−1) (R2 = 0.79) and radiation use efficiency (R2 = 0.71). Significant differences between cultivars were detected for reproductive efficiencies (seeds per gram of panicle biomass at the end of flowering or seeds per gram of panicle biomass increment during flowering) in 1 year, but differences in biomass partitioning to the inflorescence during flowering were not significant. A negative association between reproductive efficiencies and panicle biomass was observed, and possible determinants of this association are discussed in the light of available knowledge for the species. As in several other crop species, maximization of growth during the late reproductive period appears as the main target for crop management. As radiation interception efficiencies were usually below 0.95 this variable explained most variation in growth during flowering; but even crops managed for maximum radiation interception could exhibit differences in biomass increment associated to variability in radiation use efficiency and perhaps the duration of flowering.