152 Scientific Intelligence. 



the conclusions legitimately drawn from it, may be considered to 

 have the same accuracy as the thermodynamic basis upon which 

 they rest. — Zeitschr. Physikal. Che?n., vi, 187, Aug. 1890. 



G. F. B. 



2. On an Osmotic Experiment. — The following osmotic experi- 

 ment has been suggested by Nebnst. If two solutions of benzene 

 in ether are separated by a water-layer, this layer must evidently 

 act as a semi-transmitting partition, for the reason that the ether 

 only is markedly dissolved by the water and can therefore diffuse 

 through it, the benzene not having this power. A diffusion-cur- 

 rent must necessarily be set up in consequence, provided the con- 

 centration of the two benzene solutions is different. Evidently if 

 the two solutions are isosmotic, the ether will dissolve into the 

 water-layer equally from both sides, according to the law of its 

 solubility, and hence no transference of the ether will take place 

 through the separating layer. But if the number of dissolved 

 molecules in the two solutions is different, the solution with less 

 osmotic pressure will have a greater solubility in the water than 

 the other ; the direct consequence of which will be a fall of con- 

 centration of the ether in the intermediate layer and a consequent 

 diffusion, transferring ether from the more dilute to the more con- 

 centrated solution. At the same time, the separating partition 

 will be subjected to an excess of pressure in one direction, which 

 excess is the difference of the osmotic pressures of the two solu- 

 tions. The author has experimentally verified these conclusions, 

 fixing the water-layer by means of an animal membrane, tied over 

 the mouth of a fuunel. With an 8 per cent solution of benzene 

 in ether within the funnel and ether alone upon the outside, a 

 rise of from 5 to 10 cm. in an hour was observed, the flow con- 

 tinuing until the height of the column was over a meter. Nernst 

 believes that osmotic phenomena in plants are accomplished in a 

 similar way by the selective solubility of liquid separating layers. 

 — Zeitschr. Physikal. Chem., vi, 37, Aug. 1890; Ber. Berl. Chem. 

 Ges. y xxiii, Ref. 620, Nov. 1890. g. f. b. 



3. On Determinations of Molecular Mass by means of Solid 

 Solutions. — Raoult's law of the reduction of the freezing point of 

 solutions by dissolved substances has now been satisfactorily 

 established for eighteen different solvents ; and in consequence 

 the law of Avogadro must also hold true for these substances. 

 Certain exceptions, however, have been noticed ; and vax't Hoff 

 supposes these to be due to the fact that the crystals which sepa- 

 rate are not the pure solvent but a solid solution of the dissolved 

 substance in the solvent. Examples of solid solutions are isomor- 

 phous mixtures, amorphous solutions such as glass, hydrogenized 

 palladium, etc. In these solutions, diffusion goes on as in liquids; 

 double decomposition taking place in solids under pressure, glass 

 acting as an electrolyte, carbon diffusing through iron and 

 porcelain. So that there is osmotic pressure in solids; and if 

 this is kinetic, the laws of gaseous and liquid diffusion will apply 

 here, and the osmotic pressure should be proportional to the concen- 



