generally
by the roots of trees (stigmariae) remaining in their natural position in
the clays which underlie almost every layer of coal.

As some nearly continuous beds of such coal have of late years been
traced in North America, over areas 100 or 200 miles and upwards in
diameter, it may be asked whether the large tracts of ancient land
implied by this fact are not inconsistent with the hypothesis of the
general prevalence of islands at the period under consideration? In
reply, I may observe that the coal-fields must originally have been low
alluvial grounds, resembling in situation the cypress-swamps of the
Mississippi, or the sunderbunds of the Ganges, being liable like them to
be inundated at certain periods by a river or by the sea, if the land
should be depressed a few feet. All the phenomena, organic and
inorganic, imply conditions nowhere to be met with except in the deltas
of large rivers. We have to account for an abundant supply of fluviatile
sediment, carried for ages towards one and the same region, and capable
of forming strata of mud and sand thousands of feet, or even fathoms, in
thickness, many of them consisting of laminated shale, inclosing the
leaves of ferns and other terrestrial plants. We have also to explain
the frequent intercalations of root-beds, and the interposition here and
there of brackish and marine deposits, demonstrating the occasional
presence of the neighboring sea. But these forest-covered deltas could
only have been formed at the termination of large hydrographical basins,
each drained by a great river and its tributaries; and the accumulation
of sediment bears testimony to contemporaneous denudation on a large
scale, and, therefore, to a wide area of land, probably containing
within it one or more mountain chains.

In the case of the great Ohio or Appalachian coal-field, the largest in
the world, it seems clear that the uplands drained by one or more great
rivers were chiefly to the eastward, or they occupied a space now filled
by part of the Atlantic Ocean, for the mechanical deposits of mud and
sand increase greatly in thickness and coarseness of material as we
approach the eastern borders of the coal-field, or the southeast flanks
of the Alleghany mountains, near Philadelphia. In that region numerous
beds of pebbles, often of the size of a hen's egg, are seen to alternate
with beds of pure coal.

But the American coal-fields are all comprised within the 30th and 50th
degree