SynopsisFew years ago, the two broad standard method of molecular weight (MW) calibration was proposed' and the presence of negative uz (peak dispersion coefficient) was found to be somewhat disturbing. The limitation of the method was not specified. Since large negative values of u2 cannot be tolerated, herein is reported another type of instrumental spreading function for which the method may seem to apply, and an evaluation of this method. In its evaluation, plots of log, (intercept of a linear molecular weight calibration curve), that is, log,( 4) vs. the corresponding slope of the molecular weight calibration curve, D,, which were found to be linear, were used. The systems employed were Dextran/Corning controlled porous glass (CPG-10) packing in well-chosen mobile phase.