resistance, corrosion properties, and antiwear properties. These deficiencies can be corrected by the introduction of appropriate additives. The additives must be compatible with the base stock, with the other additives, and with the hydraulic system components. Here we are presenting results obtained through the use of mathematically designed experiments to select the optimal combination of additives and the optimal concentrations of each additive in the development of a working fluid for hydraulic systems. The base stock used in this study was a highly treated mineral oil thickened with a polymethacrylate additive. The additive package used in this polymer-compound ed base stock consisted of antfoxidant, anticorrosion, and antiwear additives. On the assumption that the properties of the final prod. zt will depend to a greater degree on the combination of additives (i.e., on the compatibility) than on the concentrations of each additive, the work was divided into two stages. In the first stage, the best combination of additives was selected, on the basis of concentrations commonly used (according to information reported in the literature); in the second stage, optimal concentrations of each additive were established for this best additive package. In evaluating the various additive packages, the finished blends were oxidized under the following conditions: oxidation temperature 125~ test period 70 h, air bubbling rate through oil 5 liters/h. The criteria used to rate the oxidation and corrosion resistance of the oils were the change in acid number, change in viscosity, weight changes of steel and copper specimens due to corrosion, and appearance of the oil after oxidation. The antiwear properties were evaluated in a four-ball friction tester (C_~ST 9490-60) on the basis of the wear scar diameter on the steel bails after a 4-h test at 18-20~ In the first stage of this test program, 1 5 experiments were conducted in accordance with a partially balanced incomplete block design. The results were worked up by an analysis of variance in order to distinguishthebest combination of antioxidant, anticorrosion, and antiwear additives. This first stage was described in detail in [2]; the additive package selected for further study consisted of Ionol [2,6-di-tert-but yl-4-methylphenol], an alkenylsuccinic acid, and DF-11 [a zinc dialkylphenyldithiophosphate]. The first stage was concerned with qualitative factors, i.e., with types of additives; in contrast, the second stage was concerned with quantitative factors, specifically, the concentrations of additives in the package. Naturally, there were changes in the experiment design and in the equations use to model the output parameters of the working fluid. In resolving the problem of selecting optimal concentrations of three additives in the test package, the most economical approach is the use of a fractional factorial experiment 2 3-1 with a generating relationship x 3 = xax 1, requiring in this case only four experiments. It is assumed that each of the chemical indices of the working fluid is linearly dependent on the concentrations of the additives used in the package: