『Abstract
　The dissolved silica concentrations in groundwater, springwater, and streamwater were measured on an unchanneled hillslope in the Tanakami Mountains of central Japan. The effects of preferential flowpaths, including lateral and vertical flow in the soil layer and flow through bedrock fractures, on the variation in the dissolved silica concentrations in runoff and groundwater were examined, as were the effects of the mixing of water from geochemically diverse water sources on the dissolved silica concentrations. The mean dissolved silica concentrations in water sampled from 40 cm below the soil surface and in transiently formed groundwater above the soil-bedrock interface during rainfall events were relatively constant, independent of the variation in the mixing ratio of pre-event water and incoming throughfall. These waters were mostly supplied by the vertical infiltration of water in soil. The mean dissolved silica concentrations were similar, regardless of sampling depth, although the mean residence time of the water increased with depth. These results indicated that the dissolved silica concentrations in soil water and transient groundwater were independent of contact time between the water and minerals. The mean dissolved silica concentration in perennially saturated groundwater above the soil-bedrock interface, which was recharged by water infiltrating through soil, and water emerging from bedrock in an area near the spring was more than twice that of transient groundwater, and the variation was relatively large. The mean dissolved silica concentration increased significantly downslope, from perennial groundwater to spring from soil matrix to stream, and the spring and stream concentrations also showed large variations. The dissolved silica concentration was highest in the spring from a bedrock fracture and was relatively constant. The mixing of water from two geochemically diverse water sources, soil and bedrock, controlled the dissolved silica concentrations of the perennial groundwater, the spring from soil matrix, and the stream. Our results demonstrated that in most areas of this headwater catchment, the preferential flowpaths have only a small effect on the dissolved silica concentrations. In a small area, which was ＜2％ of the total catchment area near the spring, the dissolved silica concentration was controlled by the mixing of water from geochemically diverse water sources.』

1. Introduction
2. Study site
3. Terminology and background hydrologic conditions
4. Material and methods
　4.1. Field and laboratory method
　4.2. Evaluation of new water ratio
　4.3. Endmember mixing analysis
　4.4. Diseqruilibrium indices
5. Results
　5.1. Dissolved silica concentration
　5.2. Streamflow
　5.3. Temporal variation in δD
　5.4. Spatial distribution of endmembers
　5.5. Temperature
　5.6. Disequilibrium index
6. Discussion
　6.1. Small variation in dissolved silica concentration in transient groundwater
　6.2. Large variation in dissolved silica concentration in perennial grounwater and spring and stream water
　6.3. Differences in dissolved silica concentration in soil and bedrock
7. Implication for chemical weathering reactions and suggestions for future studies
8. Summary and conclusions
Acknowledgments
References
Appendix 1
Appendix 2