[ 203 ]j 



grav. was 1,850, Dr. Black's with an acid whofe fp. grav. was 

 1,798, and mine with an acid whofe fp. grav. was 1,565, we may 

 be affured that to that extent (which includes 25 determinations) 

 no material error exifts. 



We muft^'not however imagine, that all mineral alkali contains 

 the fame proportion of ingredients as foda ; for the natural mineral 

 alkali found in Africa, and called trona, contains a fomewhat larger 

 proportion of fixed air and a much fmaller of water. 195,6 grains 

 of trona, which Dr. Black had the goodnefs to fend me, were fa- 

 turated by 260,5 of vitriolic acid 1,383, and gave out 66,^ grains 

 of fixed air, therefore 100 grains of trona would require 133,18 of 

 this acid for their faturation, and would lofe 34 grains of fixed 

 air. 



Now this acid differs infenfibly from the tabular, whofe fp. 

 grav. is 1,387, which contains 40,18 grains per cent, real acid, and 

 therefore 133,18 grains of it contains 53,5 of real acid. But we 

 have fhewn that 100 parts real vitriolic acid faturate 78,32 of mere 

 mineral alkaH, therefore ^^iiS grains of this acid faturate 41,9 ; this 

 therefore is the quantity contained in 100 grains of trona, then 

 41,9 + 34 of fixed air = 75,9 and 1,8 grains of reddifh earth, con- 

 fequently the remainder, that is 22,3 grains, Avere water. 



Here we fee the alkali takes up more fixed air than ufual, for 

 fince ufually 60 of the alkali take up 40 of fixed air, 41,9 of the 



C c 2 pure 



