The molar absorption coefficient, epsilon, of a protein is usually based on concentrations measured by dry weight, nitrogen, or amino acid analysis. The studies reported here suggest that the Edelhoch method is the best method for measuring epsilon for a protein. (This method is described by Gill and von Hippel [1989, Anal Biochem 182:319-326] and is based on data from Edelhoch [1967, Biochemistry 6:1948-1954]). The absorbance of a protein at 280 nm depends on the content of Trp, Tyr, and cystine (disulfide bonds). The average epsilon values for these chromophores in a sample of 18 well-characterized proteins have been estimated, and the epsilon values in water, propanol, 6 M guanidine hydrochloride (GdnHCl), and 8 M urea have been measured. For Trp, the average epsilon values for the proteins are less than the epsilon values measured in any of the solvents. For Tyr, the average epsilon values for the proteins are intermediate between those measured in 6 M GdnHCl and those measured in propanol. Based on a sample of 116 measured epsilon values for 80 proteins, the epsilon at 280 nm of a folded protein in water, epsilon (280), can best be predicted with this equation: epsilon (280) (M-1 cm-1) = (#Trp)(5,500) + (#Tyr)(1,490) + (#cystine)(125) These epsilon (280) values are quite reliable for proteins containing Trp residues, and less reliable for proteins that do not. However, the Edelhoch method is convenient and accurate, and the best approach is to measure rather than predict epsilon.