October 9, 1896.] 



SCIENCE. 



515 



they result from the work done upon the 

 molecules in dissociation is not known. 

 Some progress has been made towards the 

 solution of the question by Ostwald, who 

 succeeded in measuring the heat energy of 

 ionization in a few cases. This problem is 

 one that should be carefully studied. 



An objection to the theory of the existence 

 of free ions in a solution has been urged 

 from the chemical side, namely, that the 

 ions possess different properties from the 

 atoms, or atomic groups. It seemed remark- 

 able that a potassium ion should be capable 

 of existing in water without combining 

 with the oxygen, as would be the case in 

 the ordinary atomic or molecular condition. 

 If we consider, however, that the amount 

 of associated energy in the two conditions 

 is different, it is not difficult to imagine dif- 

 ferent properties. We know, for instance, 

 that negatively charged zinc will not act 

 -on hydrochloric acid ; that several elements 

 exist in well-known allotropic conditions, 

 showing quite different properties. We ex- 

 plain this by different amounts of associated 

 energy, which, in some cases, is quite meas- 

 urable. 



The difficulty of applying Ohm's law in 

 the case of Grotthus' and Faraday's theories 

 disappears in case of the dissociation theory; 

 it rather becomes a necessary tjonsequence 

 -of it. 



Considering now a few phenomena not 

 directly involved in electrolysis, evidence 

 in favor of the dissociation theory may be 

 found. 



Substances form solutions when a homo- 

 geneous mixture results, the constituents of 

 which can not be separated by mechanical 

 means, the proportion between the parts 

 being continuously variable between certain 

 limits, with a corresponding continuous 

 variation in properties. 



According to the state of aggregation of 

 the dissolved substance before solution, en- 

 ergy changes usually become apparent, 



either in temperature changes, contraction 

 of the volume, or the like, when solution is 

 affected. As a rule, such energy changes 

 occur in the same sense when solutions of 

 different concentrations are mixed, until a 

 point is reached, with very dilute solutions, 

 when they no longer are observable. The 

 substance in the solution is then very small 

 in amount as compared with the solvent. 



It is a well-known fact that when solu- 

 tions of different concentration are carefully 

 superposed, the molecules of the dissolved 

 substance pass from the more concentrated 

 to the more dilute solution, until finally a 

 uniform degree of concentration is attained, 

 when a condition of kinetic equilibrium is 

 maintained. This diffusion phenomenon in 

 liquids is similar to that in gases, only it 

 progresses much slower. In the case of 

 gases the dynamics of the process is pretty 

 well understood and satisfactorily explained 

 by the kinetic theory, the mixture of the 

 gases resulting from the projectile energy of 

 the molecules. In the case of liquids it has 

 been variously explained ; in general, how- 

 ever, the molecular attraction between the 

 solvent and the dissolved substance has 

 been assumed as the cause. Van't Hoff has 

 recently offered an explanation along the 

 same kinetic lines so satisfactorily applied 

 in gases. The force tending to produce dif- 

 fusion must be measurable as a pressure, 

 if it exist ; if then, the two solutions are 

 separated by a semi-permeable membrane 

 which will allow but one of the two con- 

 stituents to pass, this pressure will become 

 measurable upon the membrane. The pro- 

 duction of such semi-permeable septa is a 

 matter of very great difficulty, but has been 

 accomplished to a very perfect degree for 

 some substances. The general method of 

 making such measurements is familiar to 

 all physicists. Traube, Pfeffer, De Vries, 

 Tammen and Pringsheim, from 1867 to 

 1885, have succeeded in producing semi- 

 permeable membranes of great perfection, 



