The Lower Carboniferous was brought to a close by a very widespread upheaval, which removed nearly the whole Interior Sea and resulted in a very general unconformity between the Mississippian and the overlying Pennsylvanian. In only a few areas does there appear to be a continuity of sedimentation between them and in some of these there is reason to think that the conformity is apparent, not real. The Kaskaskia faunas are entirely wanting in the western portion of the Mississippian sea, but this should probably be interpreted to mean that the uppermost beds of the Lower Carboniferous were stripped away by denudation during the interval between the two formations, when . so much of the continent was land. If this was not the case, the upheaval must have affected the western portion of the Interior Sea considerably before it drained the central and eastern portions. Some folding accompanied the upheaval in certain areas, as in Iowa and northeastern Pennsylvania.

After a time of quite prolonged erosion over a great part of the continent, a new series of events was inaugurated. In Pennsylvania an orogenic movement took place, raising the low-lying mud flats of the Mauch Chunk, but down-warping their eastern border into a long and narrow trough which extended to Alabama, and in this trough a rapid sedimentation occurred, forming great bodies of gravel and sand, the Pottsville stage. In southwestern Virginia there is, apparently at least, continuity of deposition from the Mauch Chunk into the Pottsville. During the latter epoch the trough continued to subside under its accumulating load of sediment and, from time to time, to transgress westward, in which direction the higher members of the series extend. The subsidence of the trough was intermittent, and fresh-water peatbogs were established upon its surface, resulting in the formation of coal-beds, especially in the southern Appalachians, where the Pottsville is 6000 feet thick. The water which filled the trough varied in character; in the middle of the epoch marine faunas extend as far north as central West Virginia, but the northern portion appears to have been an estuary.

Over the Mississippi valley the Millstone Grit and various sandstones with local names represent the Pottsville, but the evidence of the fossil plants shows that the Millstone Grit is not a single bed, but that in different places it corresponds to different levels of the Pottsville. At the end of the age the sea covered much of the Mississippi valley, perhaps connecting with the Michigan basin, though, this is doubtful, but it spread over western Pennsylvania and a great part of Ohio. "It is highly probable that by this time the Pottsville sea swept across the Cincinnati arch in southern Kentucky and Tennessee so as to connect with the interior region." (D. White).

In Arkansas the Pottsville, as shown by the fossils, is represented by a series of limestones, shales, and sandstones, which have till very lately been placed in the Lower Carboniferous, and so many localities of Pottsville marine animals have been found over the West as to make it probable that the sea extended to Nevada (Girty), while in that state the same stage appears to be represented by beds which contain a fauna transitional between the Lower and Upper Carboniferous.

Fig. 274. - Map of North America in the Upper Carboniferous. Black areas = known exposures; white areas = land, or unknown; horizontal lines = sea; dotted area = Permian of Kansas and Texas.

Though coal accumulation had begun in the Pottsville and even in theMississippian, the time of its formation on the greatest scale was in the second half of the Upper Carboniferous. Vast areas of low-lying swamps bordered the Interior Sea, and in these vegetation flourished most luxuriantly. A very slow subsidence, often intermittent, allowed great thicknesses of vegetable material to accumulate, but frequently a more rapid sinking brought in the sea, or bodies of fresh water over the bogs, killing the trees which grew there. We cannot yet determine how far the different coal regions represent separate basins, and how far their separation is due to the subsequent removal of connecting strata, but even in connected areas we find great differences in the nature and thickness of the beds. This indicates that oscillations of level of different amounts took place in particular parts of the same basin. Thus, in one portion may occur a coal seam of great thickness, divided into two or more layers by exceedingly thin "partings" of shale.

As we trace the coal seam in the proper direction, the partings gradually grow thicker, until, perhaps, they become strata, that intervene between very distinct and quite widely separated coal seams, each of which is continuous with the corresponding portion of the thick seam. The meaning of such a structure is, that while one part of the bog subsided very slowly, permitting the almost uninterrupted accumulation of vegetable matter, other portions sank more rapidly and were inundated with water, which deposited mechanical sediments on the surface of the submerged bog.

Hardly more than 2% of the thickness of the coal measures consists of workable coal. The strata are mostly sandstones, shales, clays, and in some regions limestones, interstratified with numerous seams of coal of very varied thicknesses. This alternation of coal with mechanical deposits does not necessarily, or even probably, imply oft-repeated oscillations of level, but may be explained by assuming a general, slow, but intermittent subsidence. After each submergence, we may suppose, the movement was nearly or quite arrested, and the shallow water was filled up with sediment, until a bog could again be formed. Doubtless, movements of elevation also occurred at times, but the general movement was downward. In the Nova Scotia field are 76 distinct coal seams, each of which implies the formation of a separate bog. Beneath most coal seams occurs what miners call the "seat-stone" or "underclay," which is ordinarily a fire-clay, or it may be siliceous, but is always evidently an ancient soil. The underclay is filled with fossil roots, from which often rise the stumps of trees that penetrate the coal seam, or may even extend many feet above it.

The rock which lies on a coal seam is usually a shale, stained black by organic matter, but may be a sandstone or even a limestone, according to the depth of water over the submerged bog.

That coal is of vegetable origin is no longer questioned. Such a mode of origin is directly proved by microscopical examination, which shows that even the hardest anthracite is a mass of carbonized but determinable vegetable fibres. On the other hand, there has been much difference of opinion concerning the way in which such immense masses of vegetable matter were brought together. Much the most probable view is, that the coal was formed in position in great peat-bogs, added to, no doubt, by more or less drifted material. The evidence for this view is to be found: (1) In the great extent and uniform thickness and purity of many coal seams, which we cannot account for in any other way. Had the vegetable matter been largely drifted together, it must have been contaminated with sediment and could not have been spread out so evenly over great areas. This objection to the "driftwood theory" becomes all the stronger when it is remembered that the process of converting vegetable matter into coal greatly reduces its bulk, a given thickness of coal representing only about 7% of the original thickness of vegetable substance.

Thus a 20-foot seam of coal implies the accumulation of nearly 300 feet of plants, and it is highly improbable that such a mass could have been evenly spread as drift over hundreds (or even thousands) of square miles, without a large admixture of mud or sand. (2) The very general presence of the underclay beneath coal seams points to the same conclusion. An underclay, as we have seen, is an ancient soil, and is of just the same character as that which we find under such modern peat-bogs as the Great Dismal Swamp.

The subsidence of the bogs and the deposition of sediments upon them gradually built up the great series of strata which are called the coal measures. The peat was thus subjected to the steadily increasing pressure of the overlying masses, which greatly aided in the transformation of the vegetable accumulations into coal. Where the coal measures have been folded, the still greater pressure, aided by heat, and perhaps by steam, has resulted in the formation of anthracite. The greater number of the Carboniferous bogs appear to have been covered by fresh water, though some were coast swamps, extending out into brackish or even salt water.

On the other hand, it must be admitted that there are not a few cases to which the peat-bog theory does not apply, for example, to the small coal basins in the central plateau of France. The famous basin of Commentry is, according to Fayol, explained without difficulty by regarding it as a delta formation in a large lake. The coarse gravel was deposited in inclined beds immediately at the mouth of a swift stream, the finer sediment was carried farther out and the floating masses of vegetation still farther. The vegetable matter became water-logged and sank to the lake-bottom, where it was free from sediment. Such a case, however, has little bearing upon the great coal-fields.