Abstract: Abstract: This study aimed to analyze water repellency and infiltration performance of soils with octadecylamine addition. Air-dried remolded loam was mixed with octadecylamine to obtain water repellent loam. A total 5 kinds of loam with octadecylamine content of 0.2%, 0.4%, 0.6%, 0.8% and 1.0% were prepared. For each loam with a certain octadecylamine content, 16 samples with different initial moisture content were then prepared. The water repelling of water repellent loam was determined by water drop penetration time (WDPT), and the critical moisture content of water repellent loam was proposed. Then the permeability tests of water repellent loam under different water head differences were carried out by using a triaxial permeameter. The relationship between octadecylamine content, seepage discharge and water head differences was analyzed, and the permeability coefficient of water repellent loam was obtained. The octadecylamine and soil moisture content were important factors affecting soil water repellency and the higher octadecylamine mass fractions had the stronger water repellency of loam, the lower moisture content of upper limit, and the higher moisture content of lower limit. When the moisture content was less than 12%, the drop penetration time of loam with octadecylamine content of 0.2% and 0.4% was less than 5 s, which showed hydrophilicity. When the moisture content of loam increased to 21%, the drop penetration time of loam increased rapidly to 3 611 and 4 012 s, respectively. That means the corresponding water repelling of loam changed from hydrophilicity to extreme level. The drop penetration time of loam soil with moisture content between 21% and 25% was more than 3 600s, which showed extreme level. When the moisture content increased from 25% to 32% (saturation), the water repellency of loam decreased rapidly from extreme to hydrophilicity. In loam with natural air-dried moisture content, the drop penetration time of loam with octadecylamine content of 0.6% and 0.8% were 731 s and 2 584 s, which showed severe level. With the increasing of moisture content, the drop penetration time increased rapidly to 9 345 s (moisture content of 23.1%) and 12 364 s (moisture content of 21.6%), which showed extreme levels. Then the drop penetration time decreased rapidly and the water repellency disappeared completely at saturation. The drop penetration time of natural air-dried loam was 4 886 s with octadecylamine content of 1.0%, which showed extreme level. With the increasing of moisture content, the drop penetration time also increased rapidly. The maximums of drop penetration time were more than 18 000s (moisture content of 19.3%, 20.0% and 23.7%) which showed extreme levels. When the moisture content increased to 26.8%, the drop penetration time decreased rapidly to 9 346 s, and the water repellency disappeared completely at saturation. For loam with octadecylamine content of 0.2% and 0.4%, the upper limits of moisture content were 11.1% and 10.8%, and the lower limits of moisture content were 23.9% and 24.1%, respectively. For loam with octadecylamine contents of 0.6%, 0.8% and 1.0%, the upper limits of moisture content were 3.5%, 2.8% and 2.0%, and the lower limits of moisture content were 26.3%, 27.1% and 27.3%, respectively. When the octadecylamine content were the same, the initial infiltration rates was less affected by water head differences. After a period of time, the infiltration rates were decreased suddenly and kept stable. The larger water head differences had the higher stable infiltration rates. When the water repellency of soils changed from wettable to extreme level, the stable infiltration rate gradually decreased. The mutation time of infiltration was shortened with the increasing of water head differences. When the octadecylamine content were low (0%, 0.2% and 0.4%), the permeability coefficients were greatly affected by the water head differences: the greater the water head differences had the smaller permeability coefficients. When the octadecylamine content were high (0.6%, 0.8% and 1.0%), the permeability coefficients were basically not affected by the water head differences. When the octadecylamine content was 0.8% and the water head difference was 20 kPa, the water hardly infiltrated into loam which showed complete impermeability. The results can provide theoretical support to analysis on hydrophobized soil and its application in civil and hydraulic engineering.