16 DOCUMENT 6 FEBRUARY 1903 and additions. The first slip probably constitutes what is presented here as the first three para- graphs of the document, set off by an asterisk. The other two slips constitute what are presented here as enclosures to the document. [1]The preceding document. [2]Einstein's request that Besso not pay Maja Einstein's salary is discussed in the preceding document. [3]Bernardo Ansbacher (1845-1914), a lawyer. The Ansbachers were close friends of the Einstein family in Milan. [4]Angelo Battelli (1862-1916) Annibale Stefanini (1855-?) Battelli and Stefanini 1899. [5]The parameter i was introduced by Jacobus van 't Hoff (1852-1911) in his theory of os- motic pressure to account for the difference between the actual osmotic pressure of a solution and the ideal gas law. Thus, the osmotic pressure p for n moles of solution in a volume V fol- lows from pV = inRT (see Van 't Hoff 1885). In 1887 Arrhenius made a connection between the parameter i and the number of ions in which a molecule of an electrolyte is dissociated (see Arrhenius 1887). [6]I.e., through the determination of the freezing point. [7]Frangois Marie Raoult (1830-1901) Svante Arrhenius (1859-1927). Raoult 1888a, Raoult 1888b, Arrhenius 1887, Arrhenius 1888a, Arrhenius 1888b. [8]Gustav Tammann (1861-1938) Tammann 1887. [9]I.e., through the determination of the boiling point. [10]Ernst Beckmann (1853-1923) Beckmann 1890. [11]Wilhelm Ostwald (1853-1932) Ostwald 1887. [12]Friedrich Kohlrausch (1840-1910) Kohlrausch 1879. [13]Max Roloff Roloff 1902. [14]William Sutherland (1859-1911) proposed an explanation for osmotic pressure in which a semipermeable barrier was represented as a mesh that holds back the molecules of the solute because of their size, while the (smaller) solvent molecules can pass freely (see Sutherland 1897). [15]Einstein's interest in semipermeable membranes shows in Einstein 1902a (Vol. 2, Doc. 2), in which he mimics their effects with external conservative forces. [16]See the two enclosures. [17]Jacobus van 't Hoff postulated that in solid solutions the solute exerted an osmotic pres- sure, analogous to the osmotic pressure in fluid solutions (see Van 't Hoff 1890). The observed phenomenon of the diffusion of the solute into the solid seemed to support this idea. See Nernst 1898, pp. 167-170 for a contemporary discussion of solid solutions. [18]A possible relation between gravitational and molecular forces was also a topic of spec- ulation in Einstein's first published paper (see Einstein 1901 (Vol. 2, Doc. 1), p. 523). See Vol. 2, the editorial note, "Einstein on the Nature of Molecular Forces," p. 6 for more details. [19]Ostwald's dilution law, which predicts a simple relation between the degree of ionization of an electrolyte and the molar density of the solute, proved to be valid only for weak electro- lytes. As a possible explanation of this failure it had been postulated that the ions of the solute combined with solvent molecules to form various kinds of aggregates ("hydration"). This hy- pothesis was not undisputed at the time see, e.g., Bousfield 1905 for a contemporary discus- sion. When in 1905 Einstein wrote his dissertation on the determination of molecular dimensions (Einstein 1905j (Vol. 2, Doc. 15)), the method he chose (a determination of the effect of the size of the solute molecules on the viscosity of a solution) allowed him to conclude that hy- dration did indeed take place. He later claimed that a solution of the hydration question had been a motivating factor in his choice of viscosity as a tool to study the size of molecules (see Doc. 186). For more details, see Vol. 2, the editorial note, "Einstein's Dissertation on the De- termination of Molecular Dimensions," pp. 170-182.