『Abstract
　The study of biogeochemical and hydrological cycles in a small experimental watersheds on silicate rocks, common for the Temperate Zone, has not yet been widely applied to the tropics, especially humid areas. This paper presents an updated database for a six-year period for the small experimental watershed of the Mengong brook in the humid tropics (Nsimi, South Cameroon). This watershed is developed on Precambrian granitoids (North Congo shield) and consists of two convexo-concave lateritic hills surrounding a large flat swamp covered by hydromorphic soils rich in upward organic matter. Mineralogical and geochemical investigations were carried out in the protolith, the saprolite, the hillside lateritic soils, and the swamp hydromorphic soils. Biomass chemical analyses were done for the representative species of the swamp vegetation. The groundwater was analysed from the parent rock/saprolite weathering front to the upper fringe in the hillside and swamp system. The chemistry of the wet atmospheric and throughfall deposits and the Mengong waters was monitored.
　In the Nsimi watershed the carbon transfer occurs primarily in an organic form and essentially as colloids produced by the slow biodegradation of the swamp organic matter. These organic colloids contribute significantly to the mobilization and transfer of Fe, Al, Zr, Ti, and Th in the uppermost first meter of the swamp regolith. When the organic colloid content is low (i.e., i the hillside groundwater), Th and Zr concentrations are extremely low (＜3 pmol/L, ICP-MS detection limits). Strongly insoluble secondary thorianite (ThO2) and primary zircon (ZrSiO4) crystals control their mobilization, respectively. This finding thus justifies the potential use of both these elements as inert elements for isoelement mass balance calculations pertaining to the hillside regolith.
　Chloride can not be used as a conservative tracer of hydrological processes and chemical weathering in this watershed. Biogenic recycling significantly influences the low-Cl input fluxes. Sodium is a good tracer of chemical weathering in the watershed. The sodium solute flux corrected from cyclic salt input was used to assess the chemical weathering rate. Even though low (2.8 mm/kyr), the chemical weathering rate predominates over the mechanical weathering rate (1.9 mm/kyr). Compared to the Rio Icacos watershed, the most studied tropical site, the chemical weathering fluxes of silica and sodium i the Mengong are 16 and 40 times lower, respectively. This is not only related to the protective role of the regolith, thick in both cases, but also to differences in the hydrological functioning. This is to be taken into account in the calculations of the carbon cycle balance for large surfaces like that of the tropical forest ecosystems on a stable shield at the global level.』

1. Introduction
2. Field setting
　2.1. Protolith
　2.2. Hillside regolith and groundwater
　2.3. Swamp regolith and groundwater
3. Methodology
　3.1. The watershed equipment
　3.2. Parent rock and soil sampling and analyses
　3.3. Atmospheric input, surface and groundwater collection and analyses
　　3.3.1. Sampling
　　3.3.2. Field treatment
　　3.3.3. Laboratory techniques
　　3.3.4. Data handling
　3.4. Biomass sampling and analyses
4. Results
　4.1. Precipitation and runoff
　4.2. Atmospheric deposit chemistry
　4.3. Protolith and saprolitization process
　4.4. Regolith and groundwater
　　4.4.1. Chloride, nitrate and sulfate
　　4.4.2. Silica and base cations
　　4.4.3. Iron and aluminum
　　4.4.4. Zirconium, titanium and thorium
　4.5. Mengong Brook chemistry and input-output balance
　　4.5.1. Carbon
　　4.5.2. Chloride, nitrate and sulfate
　　4.5.3. Silica and base cations
　　4.5.4. Iron, aluminum, titanium, zirconium and thorium
　4.6. Biomass chemistry
5. Discussion
　5.1. Mobilization and transfer of chemical elements in the Nsimi watershed: How does the Mengong Brook acquire its biogeochemical signature?
　　5.1.1. Source of uncertainties in the calculation of output fluxes
　　5.1.2. Silica behaviour
　　5.1.3. Sodium behaviour
　　5.1.4. Calcium, magnesium and potassium behaviour
　　5.1.5. Aluminum and iron behaviour
　5.2. Contrasted behaviour of zirconium, thorium, and titanium in the Swamp and Hillside systems: Implications for their use as inert elements i mass balance calculations
　5.3. Short-term weathering rates
　　5.3.1. Proportion of granodiorite and granite within the composite protolith
　　5.3.2. Estimation of the current chemical and physical weathering rates
　5.4. Comparison of the Nsimi site with other watersheds developed on silicate rocks: Influence of the hydrological functioning
6. Conclusion
Acknowledgments
References