Selectivities of an enzyme catalyzed chemical reaction and of controlled potential amperometric analysis have been combined in a novel approach to clinical chemistry which is amenable to self-testing of glucose concentrations in blood by diabetic patients in home-care situations. Advantage was taken of the capability of the enzyme glucose oxidase (from Aspergillus Niger) to catalyze reactions of the type,(UNFORMATTED TABLE OR EQUATION FOLLOWS)usdusd\eqalignno{{\rm \beta-D-Glucose + 2Ox\sb1}& = {\rm Gluconic Acid + 2Red\sb1}& {\rm(Ia)}\cr {\rm \beta-D-Glucose + Ox\sb2}& = {\rm Gluconic Acid + Red\sb2}&{\rm(Ib)}\cr}usdusd(TABLE/EQUATION ENDS)where Ox and Ox2 denote appropriate one-electron or two-electron acceptors. The enzyme was uniquely specific for the substrate usd\betausd-D-Glucose, while it was capable of promoting Reaction I with a variety of oxidizing agents as cosubstrates. When Reaction I proceeded to virtual completion, glucose was determined by quantitating the product Red via measurement of faradaic currents engendered by the electrooxidation process,(UNFORMATTED TABLE OR EQUATION FOLLOWS)usdusd{\rm Red = Ox + ne}\eqno\rm(II)usdusd(TABLE/EQUATION ENDS)occurring under judiciously controlled conditions of 100% current efficiency. "Lay-use" implementation involved a disposable microelectroanalytical cell, equipped with thin film palladium working electrodes (1,000 A in thickness) sputter-deposited on a polymer matrix and similar silver/silver chloride reference electrodes. Solid reagents (oxidant, buffer, enzyme, and supporting electrolyte) prepositioned in the cell were dissolved in situ in 25 microliter glucose samples (equivalent to one drop of blood). Currents corresponding to Reaction II were recorded at preset potentials and times. Oxidant choice was predicated on the thermodynamic requirement that the equilibrium for Reaction I be 99%+ to the right. Accordingly, several redox couples were investigated whose standard potentials were at least 0.2 volt positive compared to the glucose/gluconic acid couple. With one-electron oxidants, primary salt effects suggested the involvement of charged reactants in the rate determining step. When benzoquinone was used as oxidant, kinetic data were consistent with a double displacement ("ping-pong") enzymatic mechanism. Ferricyanide was selected as the preferred oxidant for glucose analysis. Capabilities and limitations of glucose determination by enzymatic amperometry were evaluated by analyzing clinical serum samples. The dynamic range of the method extended from glucose concentrations of 2 millimolar (36 mg/dl) to 30 millimolar (540 mg/dl).