The terbium-dipicolinic acid complex (Tb(III):DPA) fluorescence quenching method of Ma, Hwang, and Lee [Pharm. Res. (1993) 10, 204-207] was used at a lower initial concentration of Tb(III):DPA (0.2 μM compared to 1.0 μM) to minimize the concentration of chelate-conjugated macromolecules required for estimating concentrations of metal-binding sites. Fluorescence quenching by either EDTA, diethylenetriaminepentaacetic acid (DTPA), poly (L-lysine)-DTPA, or poly (L-lysine)-DTPA linked to mouse IgG (IgG-PL-DTPA) was nonlinear, suggesting initial binding of chelator results in minimal change in Tb(III):DPA complex fluorescence. On the basis of the law of mass action and nonlinear curve fitting, the data suggests a three-site binding model for Tb(III):DPA complex formation in which the binding of the first and second DPA produce the majority of complex fluorescence. Initial binding of chelators to the complex results in only slight decreases in fluorescence with the majority of fluorescence quenching resulting from dissociation of more than one DPA from the complex. Although fluorescence quenching was nonlinear, binding capacity of each chelator relative to either EDTA or DTPA (as suggested by Ma et al.) could be used to estimate the concentration of metal-binding sites consistent with concentrations reported in the literature using either radionuclides or nuclear magnetic resonance. Estimation of chelating groups on chelate-conjugated macromolecules at lower concentrations would minimize the amount of sample required for analysis during synthesis and isolation of chelate-conjugated macromolecules. The suggested model for the stoichiometry of Tb(III):DPA complex fluorescence may be useful in further refinement of methods for estimating chelator concentrations and for design of fluorescent probes for chelate-conjugated macromolecules.