In order to investigate the effects of grain size distribution, cement content, and molding pressure on the strength and durability of soil-cement mixtures, a laboratory test of soil cement mixtures was performed at four levels of cement content, five levels of molding pressure, and four levels of normal curing periods. The results are summarized as follows: 1. Optimum moisture contents in loam soil and maximum dry density in sand soil increased with the increase of cement content, but in others, both optimum moisture contents and maximum dry density were changed ununiformly. 2. When the specimens were molded with molding pressure, 50kg/$ extrm{cm}^2$, strength of soil cement mixture with cement content, 2 and 4 per cent, was lower than the strength of soil cement mixture without cement content by more than 40 to 50 per cent. 3. The strength of soil-cement molded with molding pressure, 100kg/$ extrm{cm}^2$, was higher than the strength of soil-cement molded with M.D.D. obtained from standard compaction test more than 40 per cent in sand loam cement and 50 per cent in loamy cement. 4. There was highly significant positive correlation among molding pressure, cement content and unconfined compressive strentgh and so the following multiple regression equations were obtained. Loam: fc=1.9693C+0.197P-0.84 Sandy loam: fc=2.9065C+0.235P-0.77 5. When the specimens were molded with molding pressure, 20 to 100kg/$ extrm{cm}^2$, the regression equation between the 28-day and 7-day strenght was obtained as follows. Loam : q28=1.1050q7+7.59(r=0.9147) Sandy loam : q28=1.3905q7+3.17 (r=0.9801) 6. At the cement contents of above 50 per cent, the weight losses by freeeze-thaw test were negligible. At the cement content of below 8 per cent the weight losses were singnificantly high under low molding pressure and remarkably decreased with the increase of molding pressure up to 80kg/$ extrm{cm}^2$. 7. Resistance to damage from water and to absorption of water were not improved by molding pressure alone, but when the soil was mixtured with cement above 6 per cent, damage seldoms occurred and absorbed less than 5 per cent of water. 8. There was highly significant inverse-corelationship between the compressive strength of soil cement mixtures and their freeze-thaw loss as well as water absorption. By the regression equation methods, the relationships between them were expessed as followed fc=-7.3206Wa+115.6(r=0.9871) log fc=-0.0174L+1.59(r=0.7709) where fc=unconfined compressive stregth after 28-days curing. kg/$ extrm{cm}^2$ Wa=water absorption, % L : freeze-thaw loss rate, %