WASHING TESTS. 11 
impossible to make so perfect a separation that the washing process 
will not remove portions of some constituents other than the impu- 
rities, and therefore the percentage of each constituent in the washe< I 
coal is affected by the reduction of each of the other constituents. 
This is clearly indicated in test 192, on Alabama No. 6, and test 198, 
on Virginia No. 6. A comparison of the raw-coal and washed-coal 
analyses in these two tests shows that in the test on Alabama No. 6 
the percentage of sulphur was the same in the washed coal as in the 
raw coal; and in the test on Virginia No. 6 the sulphur in the washed 
coal was higher than in the raw coal. It will therefore be seen that a 
simple comparison of the raw-coal and washed-coal analyses will not 
always show whether any of the sulphur in the raw coal was actually 
removed with the refuse in washing. 
Formulas. — In order that these percentages might be determined, 
the following formulas were compiled and used in making up this 
report. It will be noted by referring to the test data (p. 15) that 10 
per cent of the original sulphur in the raw coal was actually removed 
with the refuse in washing Alabama No. 6, and that 13 per cent was 
actually removed in washing Virginia No. 6: 
Let X = the percentage of reduction of any constituent. 
Y = the percentage of any constituent removed by washing. 
M = the percentage that the amount of the constituent in the 
washed coal is of the raw coal. 
a = the percentage that the washed coal is of the raw coal. 
b = the percentage of the constituent in the washed coal. 
c = the percentage of the constituent in the raw coal. 
Then X = — , M = ab, and Y = ^^. 
c ' c 
TESTS MADE. 
Sixty-three domestic samples of fuel from fourteen States and Ter- 
ritories and two samples from Argentina were washed during the 
period covered by this report. The detailed results of the tests arc 
given in the following tables. 
