【Objective】 Soil aggregate is a basic unit in soil structure and its stability is an important index describing soil’s resistance to breakdown in the process of water erosion. However, in splash erosion how raindrops function through mechanical impact and slaking effect on soil aggregates and what are the mechanisms and contribution rates of the two are still unclear. This study is oriented to investigate effects of mechanical impact and slaking effect of rain drops on breakdown of soil aggregates during splash erosion. 【Method】 A series of indoor splash erosion experiments were carried out in the State Key Laboratory of Soil Erosion and Dryland Farming on the Loess Plateau, Institute of Soil and Water Conservation, Chinese Academy of Sciences and Ministry of Water Resources, China. Samples of Loessal soil (Sandy loamy soil) collected from Yan’an and Lou soil (Loamy loam soil) collected from Yangling in Shaanxi province, two typical soils in the Loess Plateau, were tested in the experiments. A home-made needle type rainfall simulator, consisting of three parts: water supply apparatus, needle nozzles and support frame, was used to simulate rain drops of ultra-pure water and alcohol to determine effect of mechanical impact alone and joint effect of mechanical impact and slaking on soil aggregates, separately. The splash erosion experiments were designed to have two type of soils and 5 treatments in height for rain drops to fall, i.e. 0.5 m, 1 m, 1.5 m, 2 m and 2.5 m. 【Result】 Results show that in the two soils soil aggregate stability exhibited an order of MWD_fw < MWD_ws < MWD_sw. Slaking effect (Fast wetting) was the major mechanism of the breakdown of soil aggregates, and followed by mechanical disturbance (Wetting and Shaking), and then chemical slaking (Slow Wetting) in the end. The soil of loamy clay was higher than the soil of sandy loam in RSI (Relative Slaking Index), suggesting the former is more susceptible to slaking effect than sandy loam soil, while the latter is more to mechanical impact. In splash erosion, when rain drops fell from the same height, splash erosion rate was lower in loamy clay soil than in sandy loam soil, and splash erosion rate caused by rain drops of pure water through the joint effect of mechanical impact and slaking was higher than that caused by drops of alcohol through mechanical impact alone in both soils. Regardless of pathways of the rain drops affecting soil aggregates, splash erosion rate increased with rising kinetic energy of the rain drops, and power function could be used to well describe the relationship between splash erosion rate and rain drop kinetic energy. The splash erosion rates caused by slaking effect and/or mechanical impact of rain drops both increased with rising rain drop kinetic energy or rising height where rain drops fell from. The slaking effect contributed more than 50% to the splash erosion rate, indicting slaking effect was the main factor causing aggregate breakdown effect, but the slaking effect decreased in contribution to splash erosion rate with rising rain drop kinetic energy, while the mechanical impact acted reversely. In the cases the same in rain drop kinetic energy, the contribution of slaking effect was higher in loamy clay soil than in sandy loam soil, but that of mechanical impact was just the reverse. 【Conclusion】Contribution rates of slaking effect and mechanical impact vary with rain drop kinetic energy and soil type. All the findings in this study could be of great significance to evaluation of aggregate stability and to in-depth understanding of the mechanism of aggregate breakdown during splash erosion.