The accurate prediction of time to flower is important for the efficient exchange of genetic material among environments. The objective of this research was to evaluate methods of calculating thermal time for the sowing-silking interval (SSI) in a diverse group of 18 maize (Zea mays L.) cultivars, which were sown in a multilocation field trial at 53 planting date-location combinations. These ranged in mean temperature from 13.2 to 29.4 degrees C during the SSI, and in photoperiod at tassel initiation from 11.7 to 19.5 h. Thermal time computed from daily maximum (TX) and minimum temperatures with a base of 6 degrees C, an optimum temperature (T,) of 30 degrees C, and a linear adjustment for supra optimal values of TX such that TX = 30 - (TX - T-degrees), resulted in a lower coefficient of variation (CV) for thermal time across sites and cultivars (10.9%) than did other methods, though several of these methods gave similar results. The CVs for thermal duration of the SSI for cultivars adapted to temperate, lowland tropical and highland tropical environments averaged 7.8, 11.5, and 15.4% respectively. For tropically-adapted cultivars, the increase in the sowing-anthesis interval in day lengths >13 h averaged 47 degrees Cd h(-1) (range 24-100 degrees Cd h(-1)), while temperate cultivars were virtually photoperiod insensitive. Photoperiod sensitivity was positively correlated with the anthesis-silking interval [r = 0.62 (P = 0.01), 16 df]. This study confirms the need to determine photoperiod sensitivity and optimum temperatures for development of tropical cultivars if precise estimates of thermal time to flowering and maturity at sites with photoperiods >13 h are required.
