It has been previously predicted theoretically that a substantial separation of counterions will occur during the pressure-driven transport of electrolyte mixtures across charged porous membranes. This presumption has been supported experimentally using negatively charged sulfonated polysulfone and ternary electrolyte solutions with a simple colon like Cl-. In dilute solutions the selectivity has approached values between 8 and 10 and sometimes even greater. The experimental findings of the relationship of selectivity to feed ionic strength and feed composition agree fairly well with the theory. Moreover, the theoretical prediction is also supported by the correlation of selectivity with mobilities of counterions and transmembrane volume flow. The less mobile counterions of symmetrical mixtures like KCl/LiCl have a rejection inferior to the more mobile ones when porous charged membranes are used. The reverse effect was observed when more dense membranes of the same polymer and degree of substitution were employed. However, some deviations from normality were noticed when the KCl/MgCl2 mixture was used such as changes in the sign of the selectivity logarithm and the nonmonotonic dependences of the selectivity on the feed ionic strength and transmembrane volume flow. This leads to the assumption that the mobility of the Mg2+ ion in the membrane phase is lower than that of the K+ ion, which is just opposite of their bulks. This relative decrease in the Mg2+ ion’s mobility has been interpreted in terms of stronger electrostatic interactions with membrane fixed charges. over, the lack of anomalies when the LiCl/MgCl2 solution is used leads to the assumption that the mobility of the Mg2+ ion in the membrane phase is between that of Li+ and K+ ions.