COAL: ITS STRUCTURE AND FORMATION. 441 



suggested, or in other cases we may find a satisfactory- 

 analogy in the extensive sheets of vegetable mud spread 

 out on the floor of a lagoon or lake after transport by the 

 raft-covered waters of a river. 



The peat bog method of formation, or the accumulation 

 and subsequent sealing up of vast thickness of forest dSrzs, 

 present serious difficulties to be overcome, and require cer- 

 tain conditions which we are not warranted in assuming for 

 upper carboniferous times. A certain proportion of the 

 material, which has played a part in coal formation, was, in 

 all probability, derived from floating vegetation, and from 

 the ddbris of plants growing on submerged ground. The 

 description by African travellers of the dense masses of 

 floating weeds and soft pulpy vegetable material which prove 

 an efficient barrier to navigation, affords a picture of one 

 possible source of vegetable sediment which might, under 

 certain conditions, be converted into coal. 



Mr. Graham Kerr has been good enough to draw up the 

 following brief account of the floating vegetation in some 

 South American waters ; such conditions as he describes 

 may well have occurred during the coal-producing age. "A 

 very large area of the more central parts of the ' Gran 

 Chaco ' is covered by wide-spreading swamps and extensive 

 lagunas. The latter are normally of great extent, and are in 

 many cases completely covered with a floating carpet con- 

 sisting mostly of Azolla and Pistia, often accompanied by 

 Pontederia. As the periodic dry seasons come round, the 

 lagunas shrink in volume, and the floating carpet is gradually 

 let down until it finally rests on the ground. The Azolla 

 develops its sporangia in great numbers, but the plants 

 themselves mostly rot away, and in the black deposit of the 

 laguna floor one finds their remains, including the numerous 

 sporangia. The rivers of the Paraguay- Parana system are 

 subject to periodic ' crecientes,' in which the level of their 

 waters often rises some thirty feet. During these periods 

 of high water one meets with enormous floating islands 

 made up of tree trunks, masses of Pontederia and other plants, 

 the latter of which have been floated out of the lagunas of 

 the Chaco, etc. These floating islands carry down with 



