248 
EOCENE PERIOD. 
[Ch. XVIII. 
always produce very complex results ; but in proportion as it is 
more difficult in these cases to discover any fixed order of 
superposition in the associated mineral masses, so also is it more 
easy to explain the manner of their origin and to reconcile their 
relations to the agency of known causes. Instead of the suc- 
cessive irruptions and retreats of the sea, and changes in the 
chemical nature of the fluid and other speculations of the 
earlier geologists, we are now simply called upon to imagine a 
gulf, into one extremity of which the sea entered, and at the 
other a large river, while other streams may have flowed in at 
different points, whereby an indefinite number of alternations 
of marine and fresh- water beds were occasioned. 
Second or Ujiper marine group. — The next group, called 
the second or Upper marine formation (No. 4 of the tables), 
consists in its lower division of green marls which alternate 
with the fresh-water beds of gypsum and marl last described. 
Above this division the products of the sea exclusively pre- 
dominate, the beds being chiefly formed of micaceous sand, 
80 feet or more in thickness, surmounted by beds of sandstone 
with scarcely any limestone. The summits of a great many 
platforms and hills in the Paris basin consist of this upper 
marine series, but the group is much more limited in extent 
than the calcaire grossier. Although we fully agree with 
M. C. Prevost that the alternation of the various marine and 
fresh-water formations before described admit of a satisfactory 
explanation without supposing different retreats and subsequent 
returns of the sea, yet we think a subsidence of the soil may 
best account for the position of the upper marine sands. 
Oscillations of level may have occurred whereby for a time the 
sea and a river prevailed each in their turn, until at length a 
more considerable sinking down of part of the basin took place, 
whereby a tract previously occupied by fresh water was con- 
verted into a sea of moderate depth. 
In one part of the Paris basin there are decisive proofs that 
during the Eocene period, and before the upper marine sand 
was formed, parts of the calcaire grossier were exposed to the 
