STEVENSON, COAL BASIN OF DECAZEVILLE, FRANCE 277 



the Boiirran, when they cooperated with the Valzergues and possibly with 

 the Lugan in filling up that side of the basin. 



The studies by Bergeron, Jardel and Picandet have shown that each of 

 these streams was short, especially those entering along the southern 

 border. Those students make no reference to this matter, but the facts 

 recorded by them leave room for no other conclusion. The Longuefort- 

 Auzits stream as well as the Haute Serre could have cut back their valleys 

 barely three miles into the ancient rocks, even so late as the end of the 

 Auzits system. The same is equally clear respecting the Lugan ; for the 

 inassive conglomerate above the Soulier coal bed consists almost wholly 

 of microgranulite pebbles, so that by that time the stream headed in the 

 microgranulite area, the Auzits brook had cut back across the schists to 

 the other area of microgranulite, for its deposit shows mingling of tlie 

 two rocks. In the case of each one of the three streams, the source of 

 the fragments is open to no doubt. The streams were longer during 

 deposition of the Campagnac beds, though even then the character of the 

 materials shows little change ; possibly there was a rehandling of the 

 older beds. Studies of deposits made b}^ the other streams lead to similar 

 conclusions, so that up to the close of the Campagnac, one has to do only 

 with streams of insignificant extent outside of the basin and which in no 

 case could have had an additional length of more than four miles within 

 the bapin. 



The matter is somewhat less clear on the western side, where, except 

 in the southwest, no deposit was made prior to the Bourran and the con- 

 ditions already described indicate dry lands during the Auzits and Cam- 

 pagnac. The streams recognized by Bergeron. Jardel and Picandet may 

 have been in existence throughout the whole period, carrying their load 

 over to the eastern side of the basin, comparatively gentle streams con- 

 tributing largely to the finer deposits separating the Campagnac con- 

 glomerates. But certainly at the beginning of the Bourran, these streams 

 became rapid, bringing in large pebbles, which were dropped abruptly on 

 the border of the basin, while the finer materials were carried farther 

 eastward. 



The suggestion that the Viviez and Moulin du Faux streams were 

 already old at the beginning of the Bourran seems all the more probable, 

 when one compares the conditions above the Campagnac coal bed with 

 those above the Bourran. At times, the Campagnac bed passes upward 

 through the ordinary changes of coal and shale to the coarser rocks above, 

 while at others, the passage to coarse conglomerate is abrupt, with evi- 

 dence of erosion during deposit of the first layers of the overlying rock. 

 Everywhere, the interval from coal to the mass of conglomerate is short, 



