When maize (Zea mays L.) is stressed at flowering because of deficits of water, light, nutrients, and sometimes by long photoperiods and lack of thermal adaptation, ear growth slows in relation to tassel growth and the anthesis-silking interval (ASI) increases. This appears to be a general response by the plant to a reduction in photosynthate formed during this growth stage. Grain yield (GY) and its component, ears per plant, show a dependence on ASI of the general form GY = exp(a+b*ASI). The genetic correlation between GY and ASI in a diverse array of genotypes grown under drought at flowering is about -0.6, suggesting that ASI is a visual indicator of underlying processes affecting reproductive success. Recurrent selection in several diverse tropical maize populations for tolerance to drought at flowering over 2 to 10 cycles has increased GY under stress by about 100 kg ha−1 y−1 and reduced ASI by 0.6 d yr−1. Reductions in ASI were associated with a decline in spike lets eac1, increases in rates of ear, spikelet and silk growth, spikelet size at anthesis and harvest index, and an increased tolerance to low fertility and shade. Broad-sense heritability of ASI is typically 0.5 to 0.7, and several quantitative trait loci for the trait have been identified. Yield and grain number per plant show a dependence on dry weight increase per plant during the flowering period. Genetic variation for ASI may indicate differences in this relationship, and hence differences in partitioning of currently formed assimilates to the ear at flowering. Crop models that seek to predict GY under stress should, therefore, be capable of simulating effects of stress on ASI, grain number, GY, and crop growth rate, and should accommodate genetic differences in relationships among these traits.