The establishment of critical tissue nitrogen (N) levels for greenhouse rose production has been primarily based on visual symptoms of N deficiency, with relatively less consideration to yield parameters. This work examined the relationship between rose leaf N concentration and biomass-flower yields and quality. Container-grown 'Royalty' rose plants were irrigated for 13 months with complete nutrient solutions containing N concentrations of 30, 60, 90, 120, 150 and 220 mg·L-1. Over the course of the experiment optimum flower and dry biomass yields were consistently observed in plants receiving 90 mg·L-1 N. Leaf N concentrations increased asymptotically with N application rate, stabilizing for treatments ≥ 90 mg·L-1. Time of the year had an effect on overall leaf N concentrations across treatments, with higher values observed in the winter and lower values in the summer. Leaf N concentrations were linearly, and significantly, correlated with leaf chlorophyll content and color attributes (i.e. hue, chroma and value). Quadratic relationships between leaf N concentration and rose plant yields were observed only for the second half of the experimental period, and depicted an apparent, and striking, plant control over tissue N status. In addition, these relationships indicated that optimum plant yields were possible during spring and summer with leaf N concentrations below the recommended critical level of 3% (as low as 2.4-2.5%). These results suggest that, as currently utilized, leaf N concentration per se is not a dependable indicator of rose productivity.