A 9-yr study was conducted on a Gray Luvisolic loam at Loon Lake, Saskatchewan to determine the effect of ammonium nitrate-N (34-0-0) and urea-N (46-0-0) on bromegrass (Bromus inermis Leyss.) production and nutrient quality, N recovery, and N use efficiency when the N was applied either annually or as a single application at the start of the experiment. The effects of phosphorus (P) and sulphur (S) were also assessed. Each N source was applied at single rates of 0, 100, 200, 400, and 800 kg ha1 N and annual rates of 0, 50, 100 and 200 kg ha-1 N. The ammonium nitrate was also applied as a single dose at rates of 600 and 1000 kg ha-1 N. Phosphorus and S were applied every second year at 224 kg ha-1 P205 and 45 kg ha-1 S. Forage dry matter yields were positively related to precipitation received in May and July, and April-May, but negatively related to June precipitation. Forage dry matter production was greater over the first 8-yr period for annual N applications than for a single application receiving an equivalent amount of N; but fertilizer N recovered in forage over the same period was similar (about 30%) for the two methods of application. Dry matter yields were 19% greater after 4 yr and 26% greater after 8 when ammonium nitrate was used compared to urea. Recovery of fertilizer N in forage also favored the ammonium nitrate source. The application of low to moderate N rates depressed forage N and P concentrations when yield response was high but concentrations were increased by the highest rates of N fertilizer and by repeated annual applications of N. Toxic concentrations of NO3-N occurred in bromergrass in the first year when N was applied at or above 400 kg ha-1 as one application and also in the second year when N was applied at 800 kg ha-1. When N was applied annually at 200 kg ha-1, NO3-toxicity became a potential problem in later years as soil mineral N accumulated. Phosphorus and S fertilizers increased dry matter yields of forage; P fertilizer increased and S decreased forage P concentration. Annual N applications resulted in more uniform year-to-year and N uptake response compared to a single application and provided greater dry matter production without loss of nutritive quality. It was concluded that, in contrast to reports in the literature, the response of forage grasses to N is not site specific.