Parent material also played an important role in cation export (Table 3 and Fig. 3) since losses were 1.5 to 4 times higher on schist (BES) than on granite (BEG). As cations Neuropeptide Y adsorbed on the surface of negatively charged materials such as organic colloids, higher exports would be expected from the organically-poor schist soils than granite soils (Table 1) due to the existence of fewer cation exchange sites (Granged et al., 2011, Knoepp et al., 2005, Terefe et al., 2008 and Shakesby, 2011). On the other hand, the sandy texture of granite soils might have provided greater water infiltration capacity at micro-plot scale (Boix-Fayos et al., 2006 and Shakesby, 2011), thereby generating lower amounts of overland flow and consequently lower cation exports at the BEG site (Table 3). Differences between the relative order of total Na+ and Ca2 + losses on granite and schist seem to be linked to soil properties and hydrological processes. The high correlation coefficients between Na+ losses and overland flow at the BEG site (Table 4) suggest that the export of the highly-soluble Na+ ions was promoted by the lack of Na+-adsorption sites in granite soils (possibly due to the higher availability of divalent cations), especially since lower amounts of overland flow were required for Na+ mobilization at the BEG than at the BES site (Table 3). As a consequence, Na+ exports were also higher at the BEG site than at the BES site. In the case of the divalent cations which are less easily mobilized as monovalent ions (Cancelo-González et al., 2013, Soto and Diaz-Fierros, 1993 and úbeda et al., 2009), however, the amount of overland flow generated was a limiting factor in promoting Ca2 + export at the BEG site (r = 0.35) but not at the BES site (r = 0.81).