The weighted residual procedure was used to expand the fast reactor space-energy synthesis approach to include multiple (discontinuous) weighting functions with continuous trial functions. In the past (except for discontinuous trial function applications) the number of weighting functions was chosen equal to the number of trial functions, and all region and current residuals were weighted with the same set of weights. In this article, each region residual and each boundary residual is separately weighted. The region residuals are weighted with region reaction rates, and the current boundary conditions are weighted with a boundary “reaction rate.” Numerical results are presented for a typical two-region (core and blanket) fast reactor in which multiple reaction rate weighting and a special use of reaction rate weighting are compared with previously used fast reactor space-energy synthesis weighting forms. The results, based on using realistic trial functions, show that multiple reaction rate weighting is generally better than Galerkin or reaction rate weighting, and approaches or exceeds the accuracy of adjoint weighting for the cases examined. Although the group balance or weighted group balance weighting is improved with an application of reaction rate weighting, preliminary results based on an extension of the two techniques to multiple reaction rate weighted group balance were not encouraging. Applications of various weighting functions using poor trial functions show the necessity of realistic trial spectra if weighting functions are to be improved. Multiple reaction rate weighting maintains the “easy to use” feature of Galerkin weighting with considerable potential for multiregion error reduction.