The genotypes were grown outdoors in a wire-covered shelter to prevent damage from NECA and animals at two sowing times: (1) the normal sowing time (NS); 30th March, when day/night temperatures during the large part of the reproductive phase are below 40/25 °C, and (2) late-sown (LS) on 15th April, when temperatures during the reproductive phase are above 40/25 °C during the large part of the reproductive phase. The experiment was conducted at Panjab University, Chandigarh (30.7353°N, 76.7911°E), India. Sandy-loam soil was mixed with sand in a 3:1 ratio; one part farmyard manure was then added to three parts of the soil-sand mixture along with 10 mg kg−1 tri–calcium phosphate fertiliser and used to fill 300-mm diameter earthen pots (8 kg soil capacity). Seeds were inoculated with Rhizobium culture. Five seeds were sown in each pot and thinned to three per pot 20 DAS. Plants were fully irrigated daily between 10.00–11.00 h and 16.00–17.00 h Indian Standard Time to avoid any drought stress effects. There were 30 pots per genotype with three replications (90 pots per genotype) and arranged in a completely randomised block design. The daily maximum/minimum temperatures, relative humidity and photoperiod were recorded from the sowing date to maturity for both sowing times (Fig. 1; year 2011). The experiments were conducted in the pots to manage the irrigation effectively to avoid any water deficit stress, which may occur to a larger extent in the field-grown crop.