Decembee 31j 1915] 



SCIENCE 



949 



certainty of results heretofore obtained in experi- 

 ments on the dissociation pressure of FeaOs. 

 The Water Correction in Conductivity Determina- 

 tions: James Kendall. 



Conductivity water, however carefully prepared, 

 can not be kept for more than a short period in 

 contact with air without its specific conductivity 

 rising to about 0.9 X 10"° reciprocal ohms (at 

 25° C). This value represents also the specific 

 conductivity of pure water saturated with CO2 

 under the ordinary atmospheric partial pressures 

 (3.69 parts in 10,000). It is therefore possible to 

 obtain accurate conductivity values for very di- 

 lute solutions of any electrolyte by applying a 

 correction for carbonic acid. This has been done 

 for strong electrolytes (Arrhenius), transition 

 electrolytes (Kendall), and weak electrolytes 

 (Walker and Kendall). The results obtained indi- 

 cate conclusively the accuracy of the corrections so 

 applied. 



Conductance Data and Empirical Equations: 



Stuaet J. Bates. 



The application to experimental data of em- 

 pirical equations or of the similar but less sensi- 

 tive method of plotting the results, is of value 

 chiefly as a means of interpolation and of judging 

 the accuracy of data. However, in the latter case 

 great caution must be employed. This is illus- 

 trated by the fact that as a result of the appli- 

 cation of the Kraus equation to the conductance 

 data for KCl, the data below 0.001 N were re- 

 jected as inaccurate. It is found, however, by 

 the application of the equation 



\os^^ = K + TX'-, 



where X = Ci, Cu or C, to these data, that they 

 are consistent. The data for KCl between 1.0 N 

 and 0.0001 N agree with this equation with an 

 average deviation of but 0.03 per cent, when 

 X^Ci or Cu and of 0.07 per cent, when X^C. 

 The above equation is applicable to other aqueous 

 solutions and to non-aqueoua solutions. It is ap- 

 parently as generally applicable as that of Kraus. 

 However, in the case of salts such as KNO3 which 

 give a minimum value for n (the exponent in 

 Storch's equation), neither the equation of Kraus 

 nor that given above is capable of representing 

 the data throughout the entire concentration 



A Quantitative Measure of the Deviation from 

 the Law of Mass Action: Stuakt J. Bates. 

 The strict obedience of an electrolyte to the 



law of mass action may be readily tested by ob- 



serving the constancy of the equilibrium expres- 

 sion dlCu- However, this does not afford a 

 quantitative means of judging the magnitude of 

 the deviations at different concentrations of two 

 or more electrolytes. For example, which deviates 

 the more, dichloracetie acid at 0.1 N or KQ at 

 0.001 Af? The equilibrium expression for a strong 

 electrolyte corresponding to any concentration 

 may be calculated, but there is nothing with which 

 to compare it, for in this case the uncertainty in 

 the ' ' ionization constant at infinite dilution ' ' 

 which is a true constant in the case of weak 

 electrolytes, is often as great as 1,000 per cent. 

 However, if the law of mass action is obeyed by a 

 diionic electrolyte at a certain concentration, then 

 the exponent n in Storch 's equation C'^jCu = ^; 

 has the definite and theoretical value 2. By com- 

 paring the value which n does have at a given con- 

 centration with this value (2), a quantitative meas- 

 ure of the deviation is given. Let d= (2 — n)/n, 

 then it d^=Q the mass law is obeyed. The greater 

 d is, the greater the deviation; d may be either 

 positive or negative. Since in general n changes 

 with the concentration, d changes with the con- 

 centration, becoming smaller with decreasing con- 

 centration. The undissoeiated molecules are in 

 general largely responsible for the deviation from 

 the law of mass action. Since their behavior may 

 be readily expressed in terms of osmotic pres- 

 sure, d has been defined so that it is an approxi- 

 mate measure of the deviation of these molecules 

 from van't Hoff's law. 



Ion Concentration and the Law of Mass Action: 



Stuaet J. Bates. 



The deviation of solutions of strong electrolytes 

 from the law of mass action is usually considered 

 to be due to the presence of the charged ions in 

 the solutions. This view has an a priori possibil- 

 ity and is considered by many to be supported by 

 two facts. The first is that the conductance of 

 solutions containing two salts may be readily cal- 

 culated upon the isohydric principle. Bray and 

 Hunt have pointed out, however, that the conduc- 

 tance may be calculated upon the basis of the 

 total concentration also. The second is that the 

 equilibrium expressions of strong electrolytes may 

 be readily expressed as functions of the total ion 

 concentration, as for example, by the equations of 

 Storch and of Kraus. But the expression may be 

 equally well represented as a function of the total 

 concentration or of the concentration of the un- 

 dissoeiated molecules. (See preceding abstracts.) 

 Direct proof that the ion concentration does not 

 control the thermodynamic environment of the so- 



