The response of water repellency to added clays dominated by kaolin or smectite was investigated for 23 highly water repellent sandy soils from the West Midland Sandplains, Western Australia. The effects of soil properties, and wetting and drying cycles on this response were also examined. Two standard clays-Georgia kaolinite and Wyoming bentonite-were compared with two soil clays-Jarrahdale laterite pallid zone kaolin and Folly Flats soil smectite-in laboratory experiments. The clays were saturated with either sodium or calcium, and were applied at rates of 0%, 0.2%, 0.4%, 0.8% and 1.6% by weight. Water repellency was measured by recording water-drop penetration time (WDPT). Clay morphology and the distribution of clay on the surfaces of sand grains were examined using optical and electron microscopy.The response of WDPT to added clay could be described by the equation: WDPT = A (clay)(-b), where A is the water-drop penetration time of the untreated soil and b is a coefficient that characterises a particular soil/clay combination. Clay type was the most important factor determining the slope of this response function. Kaolinitic clays were more effective than smectitic clays in reducing water repellency. The relative effectiveness of the four clays remained the same for all soil samples tested whether WDPT was measured on the dry mix of soil and clay or after a wetting and drying cycle had been imposed. Following a wetting and drying cycle, the effectiveness of kaolinitic clays increased, whereas that of the smectitic clays decreased. The nature of the exchangeable cation (sodium or calcium) and clay dispersibility had little effect on the ability of the clays to reduce water repellency.For smectitic clays, a negative relationship was found between the response of soil water repellency to clay application (i.e. the b coefficient) and either the clay or clay + silt contents of the soils. The b coefficient decreased as the content of these fine fractions increased.The type of vegetation overlying the soils varied depending on land use, but did not influence clay effectiveness. That is, clay applied to soil under bush or cultivation had the same effectiveness.A single wetting and drying cycle could lower the water repellency of the unclayed control samples, but subsequent wetting and drying cycles only caused a gradual decline in water repellency. (C) 2002 Elsevier Science B.V. All rights reserved.