The suitability of Passive Capillary Samplers (PCAPS) for collection of representative colloid samples under partially saturated conditions was evaluated in two studies. Different colloids were used for PCAPS wick testing: silica microspheres, (330 nm in diameter), feldspathoids, ferrihydrite, montmorillonite, kaolinite, and a mixture of mineral colloids extracted from a coarse calcareous sediment. The colloids were dispersed in one of the following: a buffered sodium carbonate solution, a synthetic pore water, and/or de-ionized water. Breakthrough curves were collected for several combinations of colloid, ionic strength and pH. A moisture characteristic curve, relating tensiometer measurements of matric potential to moisture content along the length of the wick, was developed for the fiberglass wick. Results indicate that retention of the colloids on the wick depended upon the type of colloid, the pH, and in some cases, flow rate (which can be correlated to moisture content). For the silica microspheres at pH 6.5, colloid concentration recoveries were 90% or better. The mixture of mineral colloids extracted from sediment moved almost conservatively through the wicks; the colloid recoveries ranged from 88 to 100% for pH 7 and 10 respectively. For ferrihydrite and the other colloids, recoveries were strongly dependent on pH. The results suggest that for certain conditions and colloid types, fiberglass wicks can be an acceptable tool for colloid sampling in the vadose zone. However, under many conditions studied here, colloids were significantly retained inside the wicks, and therefore we conclude that the use of wicks for colloid sampling in the vadose zone must be considered with caution.