218 
ADULTERATION OF POTASSA BY SODA. 
quantity of pure carbonate of potassa which is found in the 
potassa examined, we seek for the difference which exists 
between the parts of the test acid employed in the alka- 
limeter, and the hundredths of carbonate of potassa found 
in the measuring tube, and this number sought in the Table 
(No. III.) then indicates, without any other calculation, the 
proportion of hundredths of soda with which the potassa 
examined was adulterated. 
Let us suppose, for example, that 100 grains of a sample 
of potassa had employed for their neutralisation 60 degrees 
of test acid, the quantity of carbonate of potassa which 
corresponds to them must be sought for in the table, No. I. 
The number in question is 87.5. Now, if 100 grains of the 
same potassa gave likewise 87.5 of crystalline powder in 
the measuring tube, there would be no difference between 
the two numbers found, and it must be concluded that the 
potassa in question was not adulterated with soda. 
If, in another experiment, the quantity of carbonate of po- 
tassa found by the alkalimeter was 52, and if my measuring 
tube indicated only 32, it would be necessary to subtract this 
latter number (which represents the hundreths of pure carbo- 
nate of potassa really contained in the potassa) from the first, 
namely, 52, and to seek the difference (20 in this case) in 
Table III. in order to find the number of hundredths of soda 
with which the potassa examined is adulterated : this would 
be 22.5 hundredths in the present case. The 25.5 hundredths 
which are still wanting should then be attributed to the or- 
dinary and accidental impurities of potassa. 
If a saline mass sold for potassa, gave by the alkalimeter 
80 degrees, and if there were not formed, on the contrary, 
any crystalline powder in the measuring tube, the salt in 
question would be only carbonate of soda, with 10 per cent, 
of other impurities; for the number found of alkalimetrical 
degrees would immediately indicate the difference sought 
for, since, in the latter case, there would be nothing to be 
subtracted. 
