Chemical Composition of various kinds of Glass. 439 



In calculating the equivalent formulae, I have deducted from 

 the alkalies as much as is necessary to form a neutral salt with 

 the sulphuric acid found. The remaining alkali 1 have reckoned 

 in equivalents of that alkali which was present to the greatest 

 amount in the glass ; and in like manner I have reckoned mag- 

 nesia and oxide of lead as lime. The small quantities of arsenic 

 acid, oxide of iron, oxide of manganese, and alumina I have 

 neglected. Thus calculated, the equivalent formulas of the re- 

 spective kinds of glass will be as follows : — 



Glass No. 1. . . (Ki, Ca,), Sides' 



No. 2. . . (Na^a^/Si,^ 



No. 3. . . (Na 1 Ca . 7 ) 1 Si 2 ^ 



No. 4. . . (N^Cao.^Si^g 



No. 5. . . (tojCvJiBi,*, 

 According to these formulae, Nos. 1, 2, 3, which were found 

 to be good insulators, must be allowed to possess more chemical 

 durability than the inferior sorts Nos. 4 and 5. The glass No. 3, 

 with the same proportion between lime and alkali as in the last 

 named, contains in fact considerably more silica ; No. 1 contains 

 both more silica and more lime ; and as for No. 2, it is certainly 

 less saturated with silica, but it contains instead a far greater 

 quantity of lime in proportion to the alkalies than the other 

 kinds of glass. Here I would call attention to the observation 



of Pelouze, who found that glass of the formula (Na 2 Ca . 4 ), Si 2 . 3 

 was fretted by the action of water three times as rapidly as glass 



of the formula (Na } Ca 1 . 4 ) 1 Si r7 . 



In order to test experimentally the chemical durability of the 

 specimens of glass before us, I first applied Weber's test*. I 

 placed pieces of the glass, well cleaned, for a period of twenty- 

 four hours over fuming hydrochloric acid in a closed vessel, and 

 then for some time over lime. When the specimens were taken 

 out, it was evident that the glass No. 5 was the most affected; 

 it was only necessary to hold it up to the light to see that it had 

 been etched. By scraping the glass surface with a blunt knife 

 it appeared that even specimens Nos. 1, 2, and 3 were every- 

 where somewhat affected, but visibly less than No. 4, and far less 

 than No. 5. 



In order more directly to test the durability of the five kinds 

 of glass under the action of the atmosphere, I first exposed 

 weighed specimens, in the form of an impalpable powder, to the 



* R. Weber, "Ueber das Beschlagen und Blindwerden des Glases," 

 Dingl. Polyt. Journal, vol, clxxi. p. 129. 



