10 EXPERIMENTAL WORK IN FUEL TESTING. 
These values show that on most of the coals tested drying in the 
drier brought the sample to approximately an air-dry condition. No 
effort had been made to do more than this, as the primar}^ object in 
the air drying is to get the sample into such a condition that the fine 
sample will not be subject to large moisture changes during subsequent 
handling in the laboratory. The air drying, therefore, is not to be 
understood as being a rigidly fixed determination; but it has been 
found that the values obtained as a rule are within a sufficiently defi- 
nite range to give this determination some importance as showing the 
effect of standing and exposure on the percentage of moisture in the 
coal. This matter is of considerable commercial importance, since, so 
far as the moisture content is concerned, coals having a large air- 
dr}dng loss are obviously much more affected than coals having a 
small air-drying loss. It further has appeared that the amount of 
residual moisture in the air-dried sample prepared under the described 
conditions usually lies within a range which is somewhat characteristic 
of different kinds of coal. 
The foregoing table shows this residual moisture to be about 1 per 
cent in the West Virginia coals, 3 to 6 per cent in the Illinois and 
Indiana coals, and from 10 to 12 per cent in the Wyoming samples. 
COMPARISON OF RESULTS FOR MOISTURE OBTAINED FROM SAMPLES 
PULVERIZED BY VARIOUS METHODS. 
The possibility of moisture loss during grinding on a bucking board 
has already been referred to. In order to obtain more data upon 
different coals and under observed conditions of temperature and 
humidity, duplicate portions of a number of samples were ground 
down upon the bucking board and the moisture determinations made 
upon these portions. At the same time portions of the coal which 
had only passed the 1 10-inch mesh sieve in the process of sampling 
were reserved as samples for determining moisture in the coal at this 
size. The moisture was determined in these samples by drying 5 
grams for one hour in the air bath, instead of 1 gram as in the regular 
determination. The moisture results on these samples and the results 
upon the regular samples ground down in the ball mill, together with 
the temperature and humidity conditions of the laboratory at the time 
of sampling, are given in the following table: 
