『Abstract
　Paired watersheds are used to develop a deciduous nutrient uptake stoichiometry. The watersheds are those of the House Rock Run and the Brubaker Run located in the Pennsylvania Appalachian Piedmont, USA. These two watersheds are nearly identical with respect to bedrock, regolith, climate, geomorphology, morphometry, baseflow hydrology, and type and successional stage of forest vegetation. They only differ by the percentage of deciduous forest cover, with House Rock Run having 59％ and Brubaker Run having 76％. From differences in their stream chemistries the biomass nutrient uptake stoichiometry of K1.0Mg1.0Ca1.4 was determined. This stoichiometry applies to an aggrading deciduous biomass and differs from those previously used which were derived from net primary production (NPP) data. The difference may reflect that macronutrients in plant tissue may also originate from atmospheric inputs and/or decomposing biomass. Although this stoichiometry may not be applied to all deciduous forest-covered watersheds, it is likely an improvement over a stoichiometry determined from NPP data.
　Mass-balance calculations of mineral weathering rates often suffer from the number of unknowns exceeding the number of equations. To add equations to the mass-balance matrices two methods are introduced. The first method employs Zr-normalized bulk chemical compositions of bedrock and soil to calculate a mass transfer coefficient for chemical weathering. The second approach uses chemical formulae and modal abundances of primary minerals undergoing complete dissolution during weathering. Both methods allow for calculation of weathering rate constraints without biomass and cation exchange influences. These constraints serve as additional equations in the mass-balance matrix. This study finds that the watershed with the higher percentage of recently abandoned agricultural fields previously used for growing row crops has the higher chemical weathering rate. The higher chemical weathering rate reflects greater runoff resulting from reduced evapotranspiration.』

1. Introduction
2. Background
　2.1. Biomass
　2.2. Adding matrix equations and use of solid-phase data in watershed mass-balance
　2.3. Site description
　2.4. Conceptual framework
3. Methods
　3.1. Stream water chemistry
　3.2. Long term representativeness of data set
　3.3. Influence of former agricultural land use on stream chemistry
　　3.3.1. Liming and agrichemical applications
　　3.3.2. Ground water flow paths
　3.4. Determination of the biomass nutrient uptake stoichiometry
　3.5. Biomass chemistry
　3.6. Stream discharge measurements
　3.7. Elemental input-output fluxes
　3.8. Adding more mass-balance matrix equations using solid-phase data
　　3.8.1. Zirconium-normalized bulk chemical concentrations of bedrock and soil
　　3.8.2. Chemical formulae and modal abundances of primary minerals
　3.9. Watershed mass-balance methods
4. Results
　4.1. Stream water base cation fluxes
　4.2. Biomass chemistry and nutrient uptake stoichiometry
　4.3. Cation exchange
　4.4. Watershed mass-balance matrices with more matrix equations using solid-phase data
　4.5. Mineral weathering rates
5. Discussion
6. Summary and conclusions
Acknowledgments
References