This study is an investigation of the effects of stabilizing 2 types of residual soils (sample A and B) with natural river sand (sample C) in Tete Province, Mozambique. Various combinations of soils and sand (sample A+C, B+C and A+B+C) are assessed for the purpose of improving their strength to conform as pavement materials. Laboratory classification tests (grading and Atterberg Limits) were conducted on each of the 3 natural materials and strength tests (CBR and compaction) were performed on each of the 2 soil samples (sample A and B) and on the mixtures of soils and sand (A+C, B+C and A+B+C) in different proportions of the blended materials. The results of strength tests indicated that the addition of sand to sample A reduced the maximum dry density (MDD) and the optimum moisture content (OMC) and the CBR values (for 1-day soaked) decreased. For sample B, addition of sand caused a decrease in OMC and an increase of the MDD and CBR (1-day soaked). The optimum proportions of the blended material were obtained with 20% sand (C), 20% of soil sample B and 60% of soil sample A which gave a CBR of 98% after 4 days of soaking. A trial section of base course with materials dumped in the ratio of 1 truck of soil sample B, 1 truck of sand sample C and 3 trucks of soil sample A gave a 4-day soaked CBR of 79%. The study indicates that sand may be mixed with several soils from different borrow pits to improve the properties for road construction materials. Laboratory and field tests are presented to compare the properties of the mechanically stabilized trial section and the optimum design parameters.