438 M. F. L. Ekrnan on the Electrical Insulating -power and 



beyond doubt that the same change takes place, though far more 

 slowly, on a smooth surface of glass. It is known also that there 

 are kinds of glass which in moist air speedily become covered with 

 a visible coating that dims both the brilliancy and transparency 

 of the glass. That such kinds of glass, which must be consi- 

 dered absolutely faulty in their composition, must be unfit not 

 only for optical but for electrical purposes is self-evident ; but it 

 appeared to me probable that similar changes in the surface of 

 glass even of unexceptionable composition might take place in the 

 air, sufficiently minute, indeed, to escape observation, but suffi- 

 ciently serious to affect the electrical insulating-power, and that 

 the differences between various specimens of glass as insulators 

 may accordingly depend chiefly on their different chemical 

 durability. 



The expression for the chemical durability of glass is to be 

 found in its equivalent formula. The composition of ordinary 

 glass varies on each side of the mean formula (MO + 2Si0 3 ), 

 where MO signifies alkali and lime. The kinds of glass in which 

 the amount of silica exceeds that expressed by the formula possess 

 usually a greater degree of chemical durability, and vice versa ; 

 but that durability depends also in a great measure on the mu- 

 tual relation between the alkali and lime, so that, at least to a 

 certain limit, it is increased the more lime is substituted for alkali. 



On subjecting the above-mentioned kinds of glass to analysis, 

 I found their composition to be as follows: — 





Good for electrical 



purposes. 



Unserviceable for elec 

 trical purposes. 





A 







A 



r~ 

 Glass 



Glass 



— i 

 Glass 



t 

 Glass 



Glass 



No. 1. 



No. 2. 



No. 3. 



No. 4. 



No. 5. 



Per cent. 



Per cent. 



Per cent. 



Per cent. 



Per cent. 



Silica . . . 75-7 



71-2 



76*1 



74-2 



728 



Sulphuric acid. 0'6 



1-0 



0-7 



0-8 



10 



Arsenic acid . . . 





0-2 



03 





Oxide of lead . . . 







0-3 





Oxide of iron." 













Proto- sesqui- 













oxide of man- 



> 0-4 



1-5 



0-4 



10 



0-9 



ganese 













Alumina . .^ 













Lime ... 8*9 



15-2 



8-3 



8-7 



9-3 



Magnesia . . Ol 



0-4 



01 



0-1 



0-3 



Soda ... 2*9 



9-7 



11-8 



13-8 



15-6 



Potash. . . 11-2 



10 



29 



0-4 



10 



99-8 1000 100-5 99-6 100-9* 

 * The amount of lime found is probably somewhat too great. 



