shrublands (Elkins et al., 1986;Lyford and Qashu, 1969;Wainwright et al., 2000), mesquite rangelands (Wood In semiarid environments, vegetation affects surface runoff either and Blackburn, 1981), and piñ on-juniper rangelands in by altering surface characteristics (e.g., surface roughness, litter absorption) or subsurface characteristics (e.g., hydraulic conductivity).the USA (Roundy et al., 1978). Similar findings have Previous observations of runoff within a piñ on-juniper [Pinus edulis been reported from other parts of the world. Examples Englem. and Juniperus monosperma (Englem.) Sarg.] woodland led are Australia, where studies were performed in both us to hypothesize that hydraulic conductivity differs between vegetamulga woodlands (Greene, 1992) and arid shrublands tion types. Using ponded and tension infiltrometers, we measured (Dunkerley, 2000a); Niger, in tiger bush (Bromley et saturated (K s ) and unsaturated [K(h )] hydraulic conductivity at three al., 1997); and Spain, in semiarid shrublands (Cerda et levels of a nested hierarchy: the patch (canopy and intercanopy), the al., 1998). In other studies, differences in infiltrability unit (juniper canopy, piñ on canopy, vegetated intercanopy, and bare have been found within the intercanopy, between areas intercanopy), and the intercanopy locus (grass, biological soil crust, exhibiting differing degrees of herbaceous cover (Wilbare spot). Differences were smaller than expected and generally not cox et al., 1988). Similarly, Wood and Blackburn (1981) significant. Canopy and intercanopy K s values were comparable with the exception of a small number of exceedingly high readings under found higher infiltration rates for mid-grass than for the juniper canopy-a difference we attribute to higher surface macro-short-grass areas. And in Spain, Cerda (1997) reported porosity beneath juniper canopies. The unsaturated hydraulic conducthat infiltration rates under the grass species Stipa tenativity, K(h ), values were higher for canopy soils than for intercanopy cissima were almost double those for adjacent bare soils, although differences were small. At the unit level, the only ground. significant differences were for K(h ) between juniper or piñ on cano-Enhanced infiltrability under vegetation canopies pies vs. bare interspaces. Median K values for vegetated intercanopy may be due to a number of factors, including textural areas were intermediate between but not significantly different from differences resulting from rain splash or trapping of those for canopies and bare areas. There were no significant differeolian sands by vegetation (Parsons et al., 1992); higher ences between grass, biological soil crust, and bare spots within the organic-matter content of the soil under vegetation; proherbaceous intercanopy area. Overall, the observed differences in K between canopy and intercanopy patches do not account for differ-tection of the soil surface by leaf litter; enhanced aggreences in runoff observed previously.