The tolerance of individuals to sugar has been determinedpreviously by testing how much could be given by mouth or how much could be injected at one time subcutaneously or intravenously before sugar appeared in the urine. It was recognized that these methods were open to objections. In order to overcome these objections Woodyatt1 constructed an apparatus permitting a continuous intravenous injection of solutions at any desired rate. The apparatus consists of a motor driven syringe filling and discharging automatically. In this way it was possible for Sansum and Wilder2 to determine the tolerance of adults to dextrose much more accurately than had been possible before. They found that the appearance of the sugar in the urine depends on the amount of sugar injected in a unit of time, independent of the concentration of the sugar solution. The tolerance is expressed in grams of glucose which can be injected