Recently Berry (in press) showed that maximum recall for verbal pairedassociates was associated with a narrow range of palrnar skin conductance levels. This range was the same for both the beginning of the learning session and the beginning of the recall session 17 min. later. The best recall was associated with a palrnar skin conductance, in log pnhos, of 1.40. The unusual treatment of the data from the present experiments was motivated by this earlier finding. It was conjectured that one of the general effects of an external reinforcer might be a change in the organism toward a more optimal physiological level, more specifically toward the optimal skin conductance level already established for the verbal learning situation. METHOD Since absolute values of conductance are dependent on measurement procedures, the essentials of the technique used in the present experiments, as well as in the previous experiment, will be described. The electrodes on S were plastic cups, l/z in. in internal diameter, with a zinc disc at the botrom The disc was covered with manganese dioxide paste, and the rest of the cup was filled with saline-soaked cotton. One lead was attached to the left palm and the other to the upper left zrm. The arm lead was placed over a 7-in. X l%-in. X %-in. foam pad dampened with saline solution. The direct current impressed between the two leads was held constant at 30 pamp. The electronic circuits for the continuous ink-recording apparatus, with some minor modifications, are those described by Davis, Siddons, and Stout (1954). To be described are three consecutive and very similar experiments which used as Ss a total of 76 male volunteers from an elementary psychology course. The number of Ss in each experiment is indicated in row N of Table 1. Although the complete procedures of the three experiments were not identical, the routine of the first 16 min., which is the only interval considered here, was essentially the same. All Ss were presented with either a 400-cycle or a 650-cycle tone at either 56 or 70 db. In a:] three experimencs, each frequency and decibel combination was presented to an equal number of Ss in each of the two reinforcement conditions. 'The author is greatly indebted to the late Prof. R. C. Davis for invaluable assistance in the conduct of both the earlier experiment and the ones described in this report. Financial assistance has been provided by Contract Nonr 908-15 and by a research grant (M-4987) from he National Institute of Mental Health, United States Public Health Service.