In order to better understand the factors influencing zinc (Zn) isotope composition in hydrological systems, we analyzed the δ 66 Zn of dissolved Zn in the streams and groundwater of the Upper and Middle Rio Grande watershed in Colorado and New Mexico, United States. The stream water samples have a wider variation of δ 66 Zn (-0.57 to +0.41 ‰ relative to the JMC 3-0749-Lyon standard) than groundwater samples (-0.13 to +0.12 ‰) and than samples from streams that are in close proximity to abandoned mining sites (+0.24 to +0.40 ‰). Regional changes of bedrock geology, from primarily igneous rocks to primarily sedimentary rocks, have no resolvable effect on the δ 66 Zn of aqueous samples. Instead, an increase in water pH from 7.5 to 8.5 corresponds to a considerable decrease in the δ 66 Zn of dissolved Zn (R 2 =-0.37, p=0.003, n=22). Consequently, we link the observed Zn isotope variations to the process of adsorption of Zn onto suspended sediment and bedrock minerals (average 66 Zn adsorbed-dissolved = +0.31‰). Our results are in good agreement with previous experimental and empirical studies suggesting that Zn adsorption leads to a residual dissolved pool enriched in light Zn isotopes. Given that anthropogenic Zn sources can also be responsible for lowering of δ 66 Zn, and may overlap with the pH/adsorption effect on δ 66 Zn, the latter needs to be carefully considered in future studies to differentiate between natural and anthropogenic factors influencing Zn isotopes in this and other aquatic systems.