We examined the response of the widely used Folin-Denis assay to purified tannins from 16 woody plant species and to three commercial polyphenol preparations often used as standards. The reagent's response to these chemical mixtures differed significantly among sources (tree species, commercial preparations) and sampling dates, even though the mixtures contained the same total dry weight of tannins. Response to commercial standards usually did not resemble response to actual plant tannin and produced estimates that differed from actual concentrations by as much as twofold. Species-based and seasonal differences in polyphenol composition are evidently responsible for these variable results. Reagents that depend on redox reactions, such as the Folin-Denis, do not produce reliable absolute or relative quantification of phenolics when different species or samples from different dates are compared, and use of commercial standards does not resolve this problem.