The influence of dissolved natural organic matter (NOM) source on copper toxicity was investigated with larval fathead minnows (Pimephales promelas) in reconstituted moderately hard water. Ninety-six-hour static renewal toxicity tests were conducted to investigate an assumption of the biotic ligand model (BLM) that NOM source does not need to be considered to adequately predict copper toxicity. The nine different NOM isolates used in these toxicity tests were chemically well-characterized substances that were obtained by reverse osmosis as part of an NOM typing project based in southern Norway. Three median lethal concentration (LC50) values were estimated for toxicity tests conducted with each NOM, at nominal dissolved organic carbon (DOC) concentrations of 2, 5, and 10 mg/L. Tests also were conducted in dilution waters in which no NOM was added. Regression analyses were conducted to compare NOM-specific (specific NOM source) LC50s versus DOC concentration relationships to each other, as well as to the overall LC50 versus DOC concentration relationship. Statistical differences were found regarding the effects of NOM source on copper toxicity. Similar analyses were conducted with humic acid (HA) concentrations and spectral absorbance, and differences in the effect of NOM source on copper toxicity were similarly concluded. These results do not support the assumption that copper toxicity can be adequately predicted by utilizing DOC concentration, regardless of NOM source. Evaluation of relationships between LC50 values and other NOM characteristics revealed that despite significant differences due to NOM source on copper toxicity, DOC and HA concentrations were the most effective parameters in explaining variability in LC50 values. When BLM-predicted LC50 values were compared to observed LC50 values, predicted values showed reasonable agreement with observed values, but some deviations occurred due to NOM source and DOC concentration.