108 



the daily rate is 0-0003 grms. per sq. cm., and the rate appears to be 

 the same either in a water- saturated atmosphere or when immersed 

 in water. The absorption will continue without whiteness appearing 

 until the wood is impregnated and the emulsion can form. 



Similarly, gelatine under sheUac wiU ]pull water through the film. 

 If the film is in glass, cloudiness will appear at once in the absence 

 of the absorbing undercoat, or if the layer is apphed on an impervious 

 surface. The mUkiness disa]3pears generally on drjong the film. On 

 continued immersion sweUing ensues with the formation of blisters 

 and detachment of the fUm. 



Often the surface is ridged and shows numerous perforations as 

 if the surface had been scratched and punctured, so that the water 

 absorption on a glass plate becomes steady owing to complete 

 saturation. The swelling must be due to the osmotic pressure of the 

 colloid solution under the protecting layer of ImoxjTi compeUing 

 the compensating migration of water which forms the disperse phase 

 of the emulsion. 



It has been shown that normal solutions of NaCl, MgClg and CaCgl 

 prevent the whiteness of an ordinary varnish film and reduce largely 

 but do not prevent the passage of water through the fUm. 



N/2 solutions of the above salts have nearly the same effect, and 

 this is true for solutions of K2SO4 and KCNS. There are shght 

 difi'erences in behaviour due to the nature of the metal, so that although 

 sodium and potassium salts show the same behaviour yet magnesium 

 is slightly different to Calcium, and that again different to aluminium 

 in the form of chlorides in normal and half normal solutions. In 

 .the case of calcium there is an mdication of surface adsorption with 

 the production of a surface bloom which can be rubbed off leaving a 

 clear film. From the figures given by the Earl of Berkeley and 

 Hartley {Roy. Soc. Proc. A, 92, 477, 1916), it would appear that an 

 osmotic pressure approaching 13-5 atmospheres is necessary to 

 prevent the jjassage of water into a high-class ordinary outside varnish. 



With the solutions of N/20 and N/200 of the above salts the water 

 absorption mcreases largely, and attams its maximum in distilled 

 water. The concentration of the linoxjTi surface fiilm together with 

 polymerisation of the drymg oil present appear to be factors deciding 

 the impermeability; whereas the formation of the emulsion with the 

 absorbed water depends on the nature of the emulsifjang agent in the 

 oil. (:Morrell, Jour. Oil and Col. Chemists' Assoc, 111, 30, 1920.) 



Sufficient has been given to show that in varnish films similar 

 problems a\;'ait solution as in ordmary emulsions, and the experience 

 gained in researches on colloids in a water medium are of great value 

 although the presence of non-aqueous solvents render many of the 

 generalisations inapplicable. 



Reference may be made here to some instances of application of 

 knowledge gained by investigation of other coUoid systems. 



Bancroft {Jour. Phys. Chem., 19, 275, 1915) gives a number of 

 instances of emulsions involving the use of varnish materials, e.g., 

 bronzing liquids in which the metal goes into the dineric surface. 



Gelatine can be jjiecipitated from a solution of glue by shaking 

 with benzole, and rosin dissolved in dilute caustic alkali can be removed 



