Assessments of biotic nutrient limitation in aquatic ecosystems typically rely on concentrations and ratios of potentially limiting nutrients. While successful in lakes, this approach has been less effective in streams, which often are dominated by benthic autotrophs (e.g., algae and vascular plants). We compared water-column nutrient concentration (C) to 3 alternative metrics and assessed their ability to predict autotrophic nutrient limitation in streams. One metric, nutrient flux (F) is the product of nutrient concentration and stream discharge. The other 2 metrics, autotrophic uptake length (S_w,a) and autotrophic uptake velocity (V_f,a), are derived from nutrient-spiraling theory and incorporate nutrient uptake. To evaluate the ability of each metric to predict nutrient limitation, we analyzed nutrient diffusing substrata (NDS) data from the Lotic Intersite Nitrogen eXperiment (LINX) (phases I and II; n = 80 streams across North America). We calculated NDS response ratios (RRs) to quantify the strength of autotrophic nutrient limitation in each stream, and we regressed RRs against each of the 4 metrics for N and P. For the LINX I analysis, S_w,a was the best predictor of autotrophic N limitation. F and V_f,a also performed well, and C was a poor predictor. In contrast, all 4 metrics were poor predictors of N limitation for the LINX II analysis when evaluated for individual streams, by land use, and within regions. None of the metrics were able to predict P limitation for either LINX study. Accuracy could be enhanced by developing new methods to quantify autotrophic nutrient uptake and limitation in stream reaches, but consistency with nutrient-spiraling theory and improved predictability of autotrophic N limitation make S_w,a a potentially useful metric for evaluating N limitation in streams.