The primary purpose of this paper is to acquaint physical chemists with the sensitivity and accuracy of Baldes' (1, 2, 3) modification of the thermoelectric method of measuring vapor pressure or of comparing differences in the vapor pressure of solutions. The method is one in which the difference in temperature of a drop of a reference or known solution and a drop of the unknown solution suspended in an humidified chamber is compared, by means of a thermocouple and a sensitive galvanometer, with the difference in temperature of drops of two reference solutions. Other conditions being equal, the difference in temperature of the two drops suspended in the humidified chamber is proportional to the difference in the rate of evaporation from, or condensation on, the drops, which in "turn is proportional to the difference in vapor pressure of the drops (3).The apparatus consists essentially of two thermocouples (permitting duplicate determinations) placed at right angles to each other, the thermo junctions being in the form of horizontal loops. The thermocouples are mounted in a watertight chamber suspended in a water bath and are connected to a sensitive galvanometer through double-pole, double-throw switches. The procedure consists in placing a reference solution, in the form of a small drop, on one of the loops of each thermocouple and the unknown solution on the opposite loop, the filter paper on the wall and floor of the chamber being moistened also with the reference solution or a solution of approximately the same vapor pressure. The deflection of the galvanometer to which the thermocouples are connected is proportional to the difference in vapor pressures of the two solutions and is calibrated by using two reference solutions, that is, solutions of known vapor pressures.
SUMMARYData are presented to show the sensitivity which may be obtained with Baldes' modification of the thermoelectric method of measuring vapor pressures. Attention is called to a number of factors which must be taken into consideration in order to obtain an accuracy comparable to the sensitivity of the method. Evidence is presented to show that the temperature of a drop of solution placed on a thermocouple loop is, for practical purposes, independent of its size.