THE 

 LONDON, EDINBURGH, and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE 



[FOURTH SERIES.] 



OCTOBER 1868. 



XXXII. On Some Effects of a Chemically Clean Surface. 

 By Charles Tomltnson, F.R.S.* 



THE supersaturated solutions of a number of salts contained 

 in chemically clean vessels can be kept during a long time 

 without crystallizing, and even be reduced to temperatures much 

 below the freezing-point of water, provided they are protected 

 from the motes and dust of the air, and other chemically unclean 

 bodies, which act as nuclei. For this purpose it is sufficient to close 

 the mouths of the tubes and flasks with cotton-wool, which filters 

 the air as it enters the vessel during the period of a rising baro- 

 meter. When, however, the barometer falls, a portion of the 

 air escapes from the tube, and carries with it some of the aqueous 

 molecules of the solution; and the effect of this action often repeated 

 is to lower the surface of the liquid and to deposit all along the 

 width of the depression a crystalline crust of the salt. This 

 crust is formed with singular facility in supersaturated solutions 

 of magnesic sulphate; it may often be noticed in similar solu- 

 tions of sodic sulphate, ammonia phosphate, and some others. 



Many persons will think it a remarkable fact that this crys- 

 talline crust does not act as a nucleus to the solution itself when 

 this is brought into contact with it. Nor does the solution dis- 

 solve it, even though the tube be kept inclined for days together 

 so as to immerse a portion of the crust. It cannot do so, since 

 its high state of saturation has more than satisfied its solvent 

 powers. I say more than satisfied, because the water contains 

 a much larger quantity of salt than it can take up at the tem- 



rature of the air. 



* Communicated by the Author. 



Phil. Mag. S. 4. Vol. 36. No. 243. Oct. 1868. R 



