Sucrose was determined by utilizing sucrose phosphorylase which is specific to sucrose, in contrast to common methods using invertase and requiring the treatment of concomitant glucose. Sucrose phosphorylase was immobilized on aminopropyl glass with two other enzymes (phosphoglucomutase and glucose 6-phosphate dehydrogenase). Sucrose was determined using a flow-injection system containing an enzyme reactor with the three enzymes. The shape and area of the NADPH peak, one of the last enzyme reaction products, varied with the kind of buffer used as a carrier PIPES buffer was selected as the most favorable. The phosphate concentration, pH and PIPES concentration in the carrier were also optimized, as were the temperature of the water bath for the enzyme reactor, and the flow rate of the carrier Under the optimum condition, a linear calibration (r=0.9999) was obtained for 0.1-200 mu M sucrose. The lowest concentration, 0.1 mu M, corresponded to 5 pmol (50 mu l injection). Real samples were analyzed using a carrier containing phosphate at a high concentration (100 mM) to decrease the inhibiting effect of other sugars, especially fructose. The detection Limit (S/N=3) under this condition was 0.2 mu M, that is 10 pmol. A 50 mu M sucrose sample could be analyzed 16 times in 1 h. The sucrose concentrations measured in five soft drinks were in good agreement with those obtained by another method.