Comparison of field and laboratory data on abrasion coefficients is important in ascertaining the importance of fluvial abrasion in downstream fining in gravel-bed rivers. There are two kinds of coefficients, for size reduction and for weight (mass) reduction. In the case of abrasion and chipping or simply wearing of single sediment particles, we should define reduction rates rather than reduction coefficients. Conversion from one to the other is simple. A problem arises with discretization of particle-size distribution, normally by sieve analysis. Two cases should be distinguished: the weight of the sediment mixture remains constant, or is reduced. What we usually measure is a size-reduction coefficient, weight reduction rates were developed. The equations were tested with experimental data on abrasion of sediment mixtures, and the agreement was satisfactory. The main conclusion is that we should not determine weight-reduction coefficients for sediment mixtures by simply multiplying size reduction coefficients determined in the field of three but by using the conversion equations presented here instead. Until now, field determination of weight reduction coefficients for sediment mixtures has been inaccurate, because grain-size discretization has not been taken into account. This may be one of the main reasons why fluvial abrasion in some field situations has been given only minor importance.