COMPARISON OF CANOLA, INDIAN MUSTARD AND LINOLA IN 2 CONTRASTING ENVIRONMENTS .1. EFFECTS OF NITROGEN-FERTILIZER ON DRY-MATTER PRODUCTION, SEED YIELD AND SEED QUALITY

The potential for extending the production of winter oilseed crops int o the drier region of the cereal belt of eastern Australia was examine d by comparing canola, Indian mustard and Linola in field experiments in the contrasting environments of Junee and Condobolin, NSW. Junee is in the region where canola production is concentrated, and Condodolin is in the drier part of the cereal belt currently considered marginal for canola. Different rates of nitrogen (N) fertilizer were applied i n all experiments, and there were two times of sowing at Condobolin. W heat was grown for comparison in all experiments. Crop performance was evaluated over two seasons by measuring dry-matter production, grain yield, water-use efficiency, and seed oil and protein contents. Maximu m grain yields of all crops at Junee were higher than at Condobolin, a nd yield responses to N fertilizer in both environments were greater i n the wetter 1992 season than in the drier 1991 season. Canola and whe at showed the largest yield responses to N fertilizer, but Linola gene rally had the lowest yield responses because of poor seedling emergenc e at high N rates, intolerance of drought during grain filling (1991) and lodging (1992). Based on the overall growth and grain yield respon ses to N fertilizer, and on N (protein) removal in grain, it is sugges ted that about 25% more N be applied to canola than to wheat, that Ind ian mustard receive about the same N fertilizer rate as wheat, and tha t Linola requires about 20% less N than wheat. Sowing late at Condobol in, when there was below-average rainfall during grain filling (1991) reduced the yield of canola and Linola, but not Indian mustard. There was no yield penalty for sowing the crops late in 1992 when conditions were favourable during grain filling. Grain harvest indices based eit her on dry matter or the biosynthetic cost of seed production were hig her for wheat and canola than for Indian mustard and Linola, and were largely unaffected by the N treatments. A potential transpiration effi ciency value of 12.5 kg seed ha(-1) mm(-1) was derived for the oilseed s based on the biosynthetic costs of seed production. Oil concentratio ns were reduced by dry conditions during seed filling, by late sowing and by N fertilizer. Oil concentrations were higher at Junee than Cond obolin, due mainly to differences in ambient temperature as there was a 2.7% decrease in oil concentration for each 1 degrees C rise in mean temperature during seed filling. Oilseed meal and wheat grain protein levels were more responsive to N fertilizer when conditions were dry during grain filling. Protein levels in oilseed meals were higher but wheat grain protein was lower when conditions were favourable during g rain filling. The study indicated that there are prospects for extendi ng the production of canola and Indian mustard into drier regions of t he cereal belt because, when sown early, they have similar water-use e fficiencies to wheat based on the biosynthetic costs of grain producti on. While the yield of Indian mustard is limited by a low harvest inde x, its yield stability under dry post-anthesis conditions indicates gr eater potential than current canola cultivars for more marginal areas. Linola showed the least potential as an oilseed for the drier parts o f the cereal belt because of shallow rooting and susceptibility to dro ught.