Field evaluation of a model of photothermal flowering responses in a world lentil collection.

A model to predict flowering time in diverse lentil genotypes grown under widely different photothermal conditions was developed in controlled environments. The present study evaluated that model with a world germ plasm collection of 369 accessions using two field environments in Syria and two in Pakistan. Photoperiod alone accounted for 69% of the variance in 1/f, the reciprocal of time (d) from sowing to flower. In contrast, temperature alone did not account for a significant proportion of variation in flowering time due to the exposure of plants to supra-optimal temperatures in the late-sown Syrian trial. With the model mean pre-flowering values of photoperiod and temperature combined additively to account for 90.3% of the variance of 1/f over accessions. The correlation of field-derived estimates of temperature sensitivity of accessions to glass-house-derived estimates was significant at P = 0.05, but the equivalent correlation for estimates of photoperiodic sensitivity was higher at P < 0.01. Flowering in the field was better measured as time from sowing to 50% plants in flower rather than time to first bloom or its node number. Dissemination of the lentil crop following domestication in West Asia to the lower latitudes such as Ethiopia and India has depended on selection for intrinsic earliness and reduced sensitivity to photoperiod. Movement from West Asia to the higher latitudes accompanied by spring sowing has resulted in a modest reduction in photoperiod sensitivity and an increase in temperature sensitivity.