342 RECORD OF SCIENCE FOR 1886. 



surface of the glass. It must absorb water until the vapor-pressure 

 above the solution of alkali is equal to that at the place of the experi- 

 ment. (3) That this deposit on alkaline glass, which can be weighed, 

 is the cause of the electrical surface conductivity which such glass shows 

 in moist air when tried by the electroscope. Any glass which showed 

 no wcighable deposit of water with the balance in question was found 

 to be a good insulator when tested by the electroscope. (4) That rock- 

 salt showed a deposit of moisture some millionths of a millimeter in 

 thickness, at temperatures for which the vapor-pressure was greater 

 over a saturated solution of sodium chloride th;m the pressure in the 

 place of experiment. There was, however, in the rock-salt ustnl a small 

 quantity of magnesium chloride, which would attract moisture, also, 

 until the vapor-pressure over its solution was equal to the vapor press- 

 ure in the place of experiment. (Wied. Ann., 1886, jSTo. 4; Phil. Mag., 

 May, 188(3, V, xxi, 452.) 



Bunsen has analyzed the glass fibers used in his apparatus for the 

 condensation of carbon dioxide gas upon the surface of glass, and 

 which he had already proved to contain moisture. The analysis showed 

 the concentrated carbonic-acid solution forn)ing these capillary layers 

 to have attacked the glass very decidedly ; 49.5-13 grams of glass fiber 

 yielding to cold water sufQcient sodiuiu carbonate to give 0.8G45 grain 

 sodium chloride when evaporated with hydrochloric acid. It further 

 appeared that during the course of the experiment one hundred and nine 

 days, 5.83 per cent, of the glass had been decomposed. However, oidy 

 two thirds of the observed absorption can thus be accounted for, but 

 it is evident, notwithstanding the residuum, that glass is not a suitable 

 material with which to try experiments on capillary absorption. (Wied. 

 Ann., 1880, xxix, 161; Phil. Mag., December, 1886, V, xxii, 530.) 



The subject of capillarity and surface tensions has received consid- 

 erable attention. Sir William Thomson has discussed the phenomena of 

 capillarity ex[)erimentally in a lecture delivered at the Royal Institu- 

 tion (iSTature, xxxiv, 270, 290, 366). Magie has determined the capil- 

 lary constants of several liquids by means of a method suggested by 

 von llelmholtz, which consists in deducing the radius of curvature of 

 the capillary surface, in a tube of small diameter, from observing the 

 image of a small linear object produced by this surface. For mercury, 

 he obtained the mean value 45.82 ; distilled water, 7.226 ; carbon disul- 

 phide, 3.240; olive oil, 3.235 ; turpentine, 2.726 ; chloroform, 2.(i38; pe- 

 troleum, 2.441 ; alcohol, 2.214 {Wied. Ann., xxv, 421). Reinold and 

 Riicker have made a careful comparison of the surface tension of black 

 films {i. e., those thin enough to show the black of the first order of 

 Newton's colors) with that of colored films whose thickness was from 

 ten to one hundred times greater. They conclude that when the black 

 part of a soap-film forms in the normal way, sprea<ling slowly over the 

 surface, no evidence of any change in surface tension dependent on the 

 thickness of the film is furnished by a direct comparison of -the tensions 



