ANALYSIS OF COAL. 113 



large pieces as possible, this moisture would all be accounted 

 for. 



In spite of all precautions, samples do not always agree in 

 mineral content with the mass. The difference seems to be due 

 not only to the unequal distribution of the foreign mineral 

 matter throughout the coal, but principally to the difference 

 in specific gravity between the coal and this mineral, so that 

 the purer the coal the more satisfactory the sampling. 



Sometimes a coal is rich in foreign matter, and is contained 

 in a tube open at one end. From this samples may be drawn 

 showing differences of several per cents; as for example, 12.49 

 and 16.74 per cent obtained in two successive cases. The 

 following experiment shows how this happens and how to 

 prevent it : 30 grams of coal, finely pulverized, and contain- 

 ing 20 per cent of mineral, was put into a glass tube, which 

 was closed with a cork and placed vertically, giving it slight 

 taps to settle it down. In a short time most of the foreign 

 material was at the bottom of the tube, the upper portion 

 being nearly free. To avoid such an error the sample must 

 be drawn only after thorough mixing, and without any shaking 

 or jarring of the tube. It is well to use pastilles made up 

 immediately after thorough mixing. A sample containing 

 only 13 to 14 per cent of foreign matter has given from a 

 tube, 12. 20, 12. 81, 13.12, 13.50, 14.42 per cent. 



ANALYSIS OF THE COAL. 



No attempt will be made to treat the methods of ana- 

 lyzing coal ; still, as this usually accompanies a calorimetric 

 determination, some hints may be useful. Scheurer-Kestner 

 usually burns the coal in tubes of white glass placed on an 

 iron gutter. The same tube may thus serve several times if 

 asbestos cloth be placed between the tube and the iron and 

 the cooling be properly regulated. His tubes are 70 to 75 

 centimetres (27 to 29 inches) long and 15 to 20 millimetres 



