METHODS OF SAMPLING AND TESTING. 21 
the requirements of said notices from the company, or said difference may be 
deducted from any money then due or thereafter to become due to the con- 
tractor under the contract to be entered into. 
METHODS OF SAMPLING AND TESTING. 
In connection with the Survey's study of the coal deposits of the 
country and the best methods to prevent waste in mining and utiliz- 
ing the coal supply, trained inspectors have visited nearly 300 mines 
in 23 States, taking two or more samples from each mine. A study 
of the analyses of these samples and of samples taken from cars 
shipped from 175 of the same mines shows that the mine sample 
is in most instances better than the average of the coal as shipped in 
cars. On the average the coal delivered contains about one-third 
more ash than the mine sample taken in accordance with the instruc- 
tions to the miners regarding the rejection of slate and impurities. 
This difference is due to the failure of the miners to follow these 
instructions in getting out coal for shipment. The samples collected 
by the Government inspectors from the mines almost invariably show 
a higher moisture content than is usually obtained in commercial sam- 
pling, because of the precaution taken to have the sample represent 
the coal in the mine. a 
Mine samples when properly taken indicate the general character 
of the coal and enable one to judge of its probable value for any def- 
inite purpose. 
Samples taken from the cars should not be limited to a few shovel- 
fuls of coal from the top of the car, because the heavier pieces gradu- 
ally work down toward the bottom. Some samples taken at the 
bottom of a car have shown as much as 8 per cent more ash than 
the coal at the top. The moisture also may vary from top to bot- 
tom, depending on the weather. The only way to get a fair sample 
is to take a number of shovelfuls of coal from various points in the 
car, so as to procure a representative portion of the coal from top to 
bottom and from end to end. 
Bituminous coal when exposed to the air gradually depreciates in 
heating value, owing to losses of volatile matter, but aside from this 
loss there should be the same total number of heat units in a car of 
coal when it reaches its destination as when it started. If rain falls 
on the coal it will become heavier and a greater number of pounds 
will be delivered, but each pound will have a correspondingly lower 
heat value. On the other hand, if the weather is fair and the coal 
dries out on the way, it will weigh less and the heating value of 
each pound will be correspondingly higher. In other words, under 
a specification such as is used by the Government, neither the dealer 
a A description of the method of mine sampling is given in Bull. U. S. Geol. Survey 
No. 290, 1906, pp. 17-18. See also Bull. No. 316, 1907, pp. 486-517. 
