“…These results were chosen primarily on the basis of their coverage of wide ranges of temperature, pressure, or molality. Thus, in addition to the results [11][12][13][30][31][32][33][34][35][36][37][38][39]41,42, used by Holmes et al, 2,3 the following high-temperature and/or high-molality results were included: ͑1͒ Newly published volumetric data of Obsil et al; 18 ͑2͒ vapor pressure data of Lindsay and Liu, 4,5 Urusova and Valyashko, [6][7][8] and Emons et al; 9 ͑3͒ new enthalpy of dilution data of Wang et al 10 and those of Gillespie et al 11 and Simonson et al 12 at temperatures greater than 523 K; and ͑4͒ heat capacity data of White et al 13 for TϾ499 K. Additionally, freezing-point depression data of Rodebush 14 and those from the International Critical Tables of Numerical Data ͑ICT͒ 15 were also included, together with those of Gibbard and Gossmann, 42 to extend the model down to a temperature of approximately 240 K. Additional data that were considered in the data analysis also included vapor pressure results of Sako et al 43 and Derby and Yngve, 16 isopiestic molalities of Kuschel and Seidel, 40 heat capacity results of Saluja et al, 57 and volumetric results of Romankiw and Chou, 65 Saluja et al 57 and Pepinov et al 17 For the correct application of the least-squares method, relative weights were assigned to each set of the experimental results to reflect their different variances. The assignment of the weight was based on the compatibility of one set of results with others, the thermodynamic consistency of the data with other properties, and...…”