COAL 29 



them preserved them from decay till the mineral struc- 

 ture entirely replaced the cells, molecule by molecule. 

 Evidence is not wanting that this process went on with- 

 out disturbance, for in fig. 16 is shown a mass of coal 

 in which lie several coal balls, two of which enclose 

 parts of the same plant. This means that round dif- 

 ferent centres in the same stem two of the concretions 

 were forming and preserving the tissues; the two stone 

 masses, however, did not enlarge enough to unite, but 

 left a part of the tissue unmineralized, which is now 

 seen as a streak of coal. We have here the most im- 

 portant proof that the coal balls are actually formed in 

 the coal and of the plants making the coal, for had those 

 coal balls come in as pebbles, or in any way from the 

 outside into the coal, they could not have remained in 

 such a position as to lie side by side enclosing part of 

 the same stem. There are many other details which 

 may be used in this proof, but this one illustration serves 

 to show the importance of coal balls when dealing with 

 the theories of the origin of coal, for they are perfectly 

 preserved samples of what the whole coal mass was at 

 one time. 



There are but few seams, however, which contain 

 coal balls, and about those in which they do not occur 

 our knowledge is very scanty. It is often assumed that 

 the plant impressions in the shales above the coal seams 

 can be taken as fair samples of those which formed the 

 coal itself; but this has been recenriy shown to be a 

 fallacious argument in some cases, so that it is impossible 

 to rely on it in general. The truth is, that though coal 

 is one of the most studied of all the geological deposits, 

 we are still profoundly ignorant of the details of its for- 

 mation except in a few cases. 



The way in which coal seams were formed has been 

 described often and variously, and for many years there 

 were heated discussions between the upholders of the 

 different views as to the merits of their various theories. 

 It is now certain that there must have been at least four 



