Spread of Liquids on Solid Bodies. 423 



exhibited by aqueous solutions of carbonates of soda and 

 potash on the glass plate under which the bubbles of air 



But even in those cases where the film is so thin that in- 

 terference-colours fail, there are formed, on the surface of the 

 solid substance, near the sharply-defined flat drop of a saline 

 solution, crystals of the salt employed, or, as the phenomenon 

 has been well named, the salt creeps. 



The creeping, or efflorescence, of the salts is usually ex- 

 plained f by the liquid which is drawn up between the wall of 

 the vessel and the salt which has crystallized out. This expla- 

 nation is legitimate when once the first crystals have formed. 

 The formation of these first crystals, however, which often 

 form upon the solid surface at a great distance from the liquid, 

 is determined by the thin film of saline solution which over- 

 spreads all solid bodies (metals, glass, quartz, &c.) in imper- 

 ceptible thickness. If a portion of the water evaporates, a 

 fresh quantity of the saline solution streams into the interior 

 of the thin film. The thickness of the film and the propor- 

 tion of the inflowing saline solution are the greater as the 

 surface of the solid body is the cleaner. 



Temperature also appears to have an essential influence 

 upon the rapidity of the liquid mixture flowing into the thin 

 film. 



The crystals of the salt form where the water evaporates 

 most rapidly, at the outermost edge of the thin liquid film. 



Frequently the surface of the solid body is unequally clean 

 at different points ; and then the crystals form first on the 

 cleanest places, and at a greater distance from the flat drops 

 of liquid than on the places that are less clean. 



The creeping is the more striking as the solid surface is 

 cleaner, or as the edge-angle at the boundary of the flat liquid 

 drop is less (since this latter depends closely, as I have shown 

 above, upon the former). 



Creeping does not occur if the solid surface be covered with 

 a thin film of oil. Since glass surfaces remain clean longer 

 in the open air than metal surfaces, salts usually creep more 

 readily upon the former than upon metal surfaces. Besides, 

 it has long been known that the creeping of salts may be 

 hindered by smearing a glass surface with grease. 



Moreover those salts whose solutions in the requisite con- 

 centration are most mobile must exhibit creeping the most 



* Pogg. Ann. clx. p. 369 (1877). 



t Compare Barentin. Handworterbuch der Chemie und Physik, p. 636 

 (1842). 



