Stabilization with additives such as lime and cement has been practiced for many years and has proved effective in improving the behavior of expansive soils. Stabilizer design based on plasticity index (PI) and soil gradation is also a widely accepted method. However, some subsoils treated by this methodology have exhibited premature failures. Clay mineralogy plays an active and important role in the chemical reactions between soil and additives. A research study was conducted on six natural expansive soils with known clay mineralogy to evaluate the effectiveness of stabilization by using existing PI-based design charts. Engineering performance of treated soils was studied for long-term durability by exposing these soils to wetting and drying cycles and measuring both the volume change and the unconfined compressive strength during and at select wetting and drying cycles. Treated soil specimens containing high percentages of montmorillonite failed in durability tests when the treatments used PI-based design methods. A few treated soils did not survive the 21 wetting and drying cycles or experienced higher volumetric strains of more than the limiting value. Additional durability studies showed that soils with higher montmorillonite contents could be effectively stabilized with higher dosages of lime and cement additives. A new design chart incorporates the percentage montmorillonite content of a soil. This design chart provides an approach in selecting the optimum dosage of stabilizer to stabilize montmorillonite-dominant soils.