The passage from Cambrian to Ordovician was gradual, without any marked physical break. Only where the Upper Cambrian is sandy, as in New York, is there a decided change in the character of sedimentation. In the latter part of the Cambrian a great inland sea had been established over what is now the Mississippi valley and, with frequent fluctuations in depth and modifications in form, it was to persist for long periods as one of the salient features of Palaeozoic geography. This sea was separated from the Atlantic by the land mass called Appalachia, and on the western side it was demarcated from the Pacific by islands of undetermined size. A generalized representation of the arrangement of land and water in Ordovician North America is presented in the map, Fig. 260. Such a map, which can only be a rude approximation to the truth, is constructed by marking as water all those areas where Ordovician rocks are known, or confidently inferred to be present, even though concealed by overlying, newer strata, and as land those areas where the Ordovician is wanting. Frequently, however, it is im-• possible to determine whether the absence of the strata is due to their never having been present at the given point, or to their removal by denudation.

On the other hand, the deep-lying and buried extensions of the strata may be subject to many interruptions, of which there is no surface indication, for ancient islands and peninsulas may be covered over and concealed by newer strata. Hence, the shore-lines appear unduly simple, for the details have been destroyed by erosion, or hidden by deposition. Most attempts to reconstruct on a map the long-vanished geography of some ancient period probably err in the direction of not making sufficient allowance for the removal of the strata by denudation. An example will make this clear. At Elmhurst, fifteen miles west of Chicago, from which the nearest known exposure of Devonian rocks is distant eighty miles, and where the surface is made by a Silurian limestone (Niagara), was found a fissure containing Devonian fossils, brachiopods, and fish teeth. This proves that the Devonian once covered that whole region, but has been entirely swept away, leaving hardly a trace behind. "The presence of this Upper Devonian fauna at Elmhurst, buried as it is deep down in the Niagara limestone, indicates with certainty that during the greater part of Devonian time, the region now known as northern Illinois was above sea-level. ... At a later period, near the close of the Devonian, the sea again occupied the region, sand was sifted down into these open joints, and with it the teeth of fishes." (Weller).

Fig. 260. - Generalized map of North America in the Ordovician period. Known exposures of Ordovician rocks in black. Horizontally lined area = sea; white = land. The condition of Mexico and Central America is not known.

The Cambrian subsidence continued into the earliest Ordo-vician, "when more of the continent was under water and the sea probably deeper than at any subsequent period." (Ulrich and Schuchert.) At the end of the Beekmantown stage, extensive geographical changes occurred; two parallel folds, extending from Alabama to Newfoundland and following the trend of the future Appalachian Mountains and the Cambrian trough, were raised as low barriers, which played a very important part in separating the Atlantic from the Interior Sea. Inclosed between these parallel folds was a narrow basin, which was frequently invaded by the sea, though rarely completely submerged from end to end. At the same time there was a widespread elevation, which greatly reduced the area and depth of the Interior Sea, but comparatively soon there followed a less extensive submergence (Chazy) both at the south and in the north. In the latter region a long, narrow gulf extended up the valley of the St. Lawrence, dividing at the Adirondack Mountains and sending one arm where now is the valley of the Ottawa, and the other over the present site of Lake Champlain. Another narrow body of water called the Levis channel, and believed to be separate from the Chazy bay, though very near to it, extended from Newfoundland into the northeastern corner of New York. From the south, the invasion extended into Kentucky and Tennessee and persisted for a longer time.

In the upper Mississippi valley, the lower Ordovician limestones are followed by a very extensively developed sandstone (the St-Peter), which occurs, at the surface or underground, over nearly the whole of Illinois, Iowa, and Missouri, much of Wisconsin, Minnesota, and Michigan and Indiana, and smaller parts of other states from Kansas to Oklahoma.

The Middle Ordovician (Mohawkian series) was a time of limestone-making on an extraordinarily wide scale, which implies that the Interior Sea received less terrigenous sediment than before, and this, in turn, probably indicates that the surrounding lands were low and flat, and that the sluggish streams carried but small loads of fine silt. During the Chazy age the Interior Sea had been reduced in size, and nearly all of New York was above water, but a renewed submergence again extended the sea over most of New York and reopened the northeastern connection with the Atlantic. The Mohawkian limestones, especially the Trenton, occur in New Brunswick, New York, Canada, over the upper Mississippi valley, in the Black Hills, Bighorn, Rocky, Wasatch, and Uinta Mountains, and the Great Basin. In Kentucky and Tennessee unconformities point to oscillations of level, emergence and submergence alternating.

The Upper Ordovician consists largely of a thick mass of shales (Utica and Lorraine) formed from the terrigenous silts spread widely over the sea-bottom, due perhaps to an elevation of the land, which, rejuvenating the streams, increased the loads carried to the sea, and perhaps also to a concomitant shoaling of the sea. These shales and slates are thickest toward the east, and extend along the Appalachian line from the St. Lawrence to Tennessee and westward into Indiana. The northeastern passage to the Atlantic was more freely opened, and a fauna with strong European affinities invaded the Interior Sea, though not for long. The immensely thick mass of shales and slates, which occurs in the valley of the Hudson near Albany, and follows the trend of the Appalachian Mountains to Tennessee, was once regarded as a distinct series {Hudson River) and placed at the top of the Ordo-vician. It is now known, however, to be a separate facies, representing at least the Trenton, Utica, and Lorraine stages, and possibly the whole of the Ordovician.

The uppermost of the Ordovician stages {Richmond) is most typically developed in Ohio and Indiana, with a littoral, and perhaps partly continental, facies in central New York, and along the Appalachians, the Oneida conglomerate and Medina sandstone, which, until very lately, have been regarded as the base of the Silurian. The "eastern Oneida," or Shawangunk grit, has recently been proved to belong to the upper part of the Silurian, and is therefore of much later date than the Oneida-Medina of central New York. "Ulrich has reexamined the Medina deposits of the Appalachian region, more especially in Pennsylvania, Virginia, and Tennessee, and has concluded that they are the eastern shore deposits equivalent to the Richmond series of the Ohio and Mississippi valleys. This result therefore forces stratigraphers to place the line separating the Siluric from the Ordovicic, not at the base of the Medina formation of the New York standard section, but at its uppermost limit and beneath the Clinton." (Schuchert).

In the western portion of the continent the Ordovician is not so well known as in the eastern, because it is so generally buried under newer strata, and, over great areas where it probably once existed, it has been removed by denudation. In the Lower Ordovician great areas of the northwest were land, but these were very extensively submerged in the Middle Ordovician, when the sea probably covered most of the Great Plains and much of the plateau region. The Upper Ordovician is much more restricted, and in many places lies unconformably upon the Middle, pointing to emergences and prolonged denudation. The Great Basin, or that portion of it known as the Nevada trough, seems to have been submerged throughout the Ordovician, as it had been in the Cambrian, and, indeed, during nearly the whole Palaeozoic era. Ordovician rocks fringe the western side of the great northern pre-Cambrian area and occur in the islands of the Arctic Sea.

Aside from the slow and gentle oscillations of level above mentioned, the Ordovician was a period of tranquillity, generally speaking, without violent diastrophic movements, nor have any signs of volcanoes of that date been discovered in North America. Igneous intrusions are rare, though they have been found in New York and in the Wichita Mountains of Oklahoma, where the deposition of Trenton beds was followed by the upheaval of the mountains and the intrusion of a great mass of granite, which has metamorphosed the overlying sedimentaries.