STEVENSON, COAL BASIN OF DECAZEVILLE, FRANCE 293 



During this long period the water-covered area expanded, and. very 

 possibly some of the deposits in the western portion, assigned by Berge- 

 ron, Jardel and Picandet to the Bourran system, may antedate slightly 

 the Grande Couche de Bonrran. It is wholly probable that many of the 

 later deposits there are contemporaneous with earlier parts of the Bour- 

 ran bed. Until the closing stages of the Campagnac system, the western 

 third of the basin had been above water and its rocks had been exposed 

 to subaerial agencies, while the streams draining the western border had 

 carried their water and their loads to the eastern side. But toward the 

 close of Campagnac deposition, the relative conditions were changed; the 

 unstable east side became comparatively stable, while the hitherto stable 

 west side subsided rapidly. While the eastern portion of the basin was 

 becoming an area of mnd flats, the western portion became a pond into 

 which streams from that side dropped their coarse materials and the silts 

 were carried eastward to be spread over the emerging land. This condi- 

 tion continued into the Bourran. 



When the area between the Enne and Eiou Mort, midway in the basin, 

 had been converted into an ill-drained plain, the swamp took possession 

 and advanced thence in all directions, rapidly toward the east, but less 

 rapidly toward the north and the south. The period of very gentle sub- 

 sidence, following the abrupt change on the west side, was very long, 

 sufficing for the accumulation of at least 100 feet of coal within the 

 original area. The western portion of the basin was filled slowly, and it 

 seems probable that there was little addition to the swamp land in that 

 direction, until late in the formation of the Bourran bed. The irregu- 

 larities of the swamp border at the west are proved by the clay partings 

 in that direction, especially in the southern prolongation of the bed, 

 where the stream deposits were confluent. 



Slow subsidence of the whole block continued long after the accumula- 

 tion of the Bourran bed. The streams had cut down their channels, and 

 as they flowed over the marsh, they deposited only fine material until a 

 thickness of at least 150 feet had been reached, in marked contrast with 

 conditions succeeding the formation of the Campagnac bed. More rapid 

 adjustment followed, the speed of the streams was increased and the Car- 

 boniferous was closed by a deposit of moderately coarse sandstones. 

 There was no return of coal-making conditions. After the deposition of 

 several hundred feet of rock, the land and freshwater features disap- 

 peared, and the Permian sea invaded the basin from the south and laid 

 down its beds in practical conformity with the older series below. 



The date of andesitic eruption is somewhat uncertain, Bergeron, 

 Jardel and Picandet, who devoted much time to the investigation, seem 



