STEVEN SOX, COAL BASIX OF DECAZEVILLE, FRANCE 285 



trine in its original form conceivos of the coal basin as occupied by a lake 

 of great depth ; this no doubt to avoid the difficulty of accounting for the 

 2500 feet of rock by continental subsidence. But the great original 

 depth of the depression, as shown by Delaunay, is by no means essential 

 to the hypothesis, which without that deserves all the praise which has 

 been lavished upon it for its ingenuity. Lemiere has expended much 

 labor and great skill in the effort to determine the forms which the de- 

 posits would assume in deep water ; but that question does not concern 

 the matter in hand, for there is no evidence whatever to support the sug- 

 gestion that the basin held a deep lake; such evidence as does exist all 

 indicates that the water area was often small and that it was never dee]) ; 

 that until the end of Campagnac deposition a considerable part of the 

 basin was exposed to subaerial action ; even that the Haute Serre deposits 

 may have been begun as a subaerial dejection cone. 



The Decazeville basin embraces about 30 square miles; the entering 

 streams, at first short and abrupt, cut back their valleys until in the 

 Bourran time they may have attained a length of 10 miles, sufficient to 

 account for the whole deposit of inorganic detritus. Adding to the sur- 

 face of the basin itself a strip 10 miles wide, surrounding the depression, 

 the extreme area whose waters found outlet at Firmy would not be more 

 than 350 square miles at the time of greatest extent. In all probability 

 it was much less. The vegetation to supply detritus for the coal beds 

 must have been derived from that area, and the detritus could reach the 

 basin only by w^ of streams, torrential during a great part of their exist- 

 ence and flowing over tough, resistant rock. 



The type of valley through which these streams flowed is familiar in 

 France. The gorge between Eygurande and Bort and that between 

 Allassac and Vigeois on the way to Limoges suffice for illustrations of 

 cutting in the schists. One has only to var}^ the scale in order to have 

 all the features shown by the Decazeville streams in the several stages of 

 their development. 



The work around the basin was confined at first necessarily to cor- 

 rasion, and erosion made slow progress. But it may be that disintegra- 

 tion of the upland rocks was rapid and afforded material in which 

 vegetation quickly became as luxuriant as at many places between Mont- 

 lucon and Yiviez, within the schist area ; that even the walls of the early 

 valleys were covered by a dense growth, such as is seen in close gorges 

 within the same area. One may well imagine this condition, since the 

 Coal Measures flora is taken to be proof of a warm and rather humid 

 climate. One may picture to himself the upland region as covered in 

 great part by a forest, its surface imperfectly drained and its borderf^ 

 indented by short, abrupt streams. 



