『Abstract
　This study presents uranium and thorium concentrations and activity ratios for all riverine phases (bedload, suspended load, dissolved load and colloids) from basaltic terrains in Iceland and the Azores. Small basaltic islands, such as these, are thought to account for 〜25％ of CO2 consumed by global silicate weathering, and for 〜45％ of the flux of suspended material to the oceans. These data indicate that [U] and [Th] in the dissolved and colloidal fractions are strongly controlled by pH, and to a much lesser extent by levels of dissolved organic carbon (which are low in these environments). At high pH, basalt glass dissolution is enhanced, and secondary mineral formation (e.g. Fe-oxyhydroxides and allophane) is suppressed, resulting in high dissolved [U], and low colloidal [U] and [Th], indicating a direct chemical weathering control on elemental abundances. When the dissolved (²³⁴U/²³⁸U) activity ratio is ＞〜1.s (i.e. when physical weathering, groundwater contribution or soil formation are high), there is little isotope exchange between dissolved and colloidal fractions. At lower activity ratios, the dissolved load and colloids have indistinguishable activity ratios, suggesting that when chemical weathering rates are high, secondary clay formation is also high, and colloids rapidly adsorb dissolved U. Many of the suspended sediment samples have (²³⁴U/²³⁸U) activity ratios of ＞1, which suggests that uptake of U onto the suspended load is important. Identical (²³⁰Th/²³²Th) in suspended, dissolved and colloidal samples suggests that Th, like U, is exchanged or sorbed rapidly between all riverine phases. This particle-reactivity, combined with poorly constrained contributions from groundwater and hydrothermal water, and short-term variations in input to soils (volcanic and glacial), suggests that U-series nuclides in riverine material from such basaltic terrains are unlikely to reflect steady state erosion processes.

Keywords: uranium; thorium; U-series; weathering; erosion; clay minerals』

1. Introduction
2. Field areas
3. Methods
　3.1. Sample collection and filtration
　3.2. Cation analyses
　3.3. Isotope analyses
4. Results
　4.1. Ultrafiltration mass balance
　4.2. Elemental concentrations
　　4.2.1. Organic carbon
　　4.2.2. Elemental ratios
　　4.2.3. U-series
5. Discussion
　5.1. Colloid composition
　5.2. Uranium and Thorium behaviour - secondary mineral control
　5.3. Colloid transport of U-series nuclides
　5.4. Thorium activity ratios - rapid exchange between all phases
　5.5. The ²³⁸U decay chain and erosion rates
6. Conclusions
Acknowledgements
References