# Encyclopædia Britannica, Ninth Edition/United States/Physical Geography and Statistics

PART II.—PHYSICAL GEOGRAPHY AND STATISTICS.

Copyright, 1888, by Josiah D. Whitney.

Physical Geography and Geology.

Plate VIII. Boundaries and area.The area of the United States extends, throughout nearly its whole breadth, from the Atlantic to the Pacific. The northern edge of the vast indentation made in the continent by the Gulf of Mexico marks the southern boundary between the meridians of 83° and 97°. From the edge of this gulf the boundary between the United States and Mexico is partly artificial. Its most essential feature is the Rio Grande, which separates the two countries from 31° 47′ N. lat. to its mouth. The boundary between the United States and Canada is to a large extent natural; though artificial in the north-east (see map), it follows the St Lawrence and the Great Lakes from the 45th parallel to the point where the Rainy Lake river enters Lake Superior, and thence passes up that river to a point on the west side of the Lake of the Woods, whence it goes along the 49th parallel to the Georgia Gulf. The country, as thus limited, excluding Alaska, comprises an area, according to the most recent determinations of the Census Bureau, of 3,025,600 square miles. This total includes 55,600 square miles of water surface (coast waters, bays, gulfs, sounds, &c., 17,200; rivers and smaller streams, 14,500; lakes[1] and ponds, 23,900). The area of Alaska is given in the Census Report of 1880 as 531,409 square miles,—a rough approximation, differing greatly from that given in the 1886 Report of the commissioner of the General Land Office. The total area possessed by the United States is therefore, approximately, 3,557,000 square miles.

Geographical position.The longitude of the most easterly point is about 67° W.; that of the most westerly nearly 125°. The parallels of 29° and 49° N. roughly designate the position of the country in reference to latitude. The extreme southern end of Florida, however, extends as far south as 25° N., while the northern extremity of Maine only reaches to a little beyond 47° N. , the large triangular area between the St Lawrence and the Great Lakes, belonging to Canada, being all south of the 49th parallel.

Coast.As compared with the western coast of Europe, neither side of America possesses a deeply indented coast-line. On the Pacific side there is only one important bay (that of San Francisco) between San Diego and Puget Sound. At San Francisco and San Diego there are commodious harbours, but within the limits of the United States there are no others on the Pacific coast, unless we except that furnished on the extreme north by Puget Sound, and that offered by the mouth of the Columbia river, the bar of which is somewhat formidable except for steamers under the management of skilful pilots.

The eastern coast north of the 35th parallel is a considerably broken one. That of Maine may almost be described as a “fiord coast,” so numerous are the indentations, which are, however, of moderate depths, but which are large enough to afford excellent and commodious harbours, of which that of Portland may be taken as the type. There is an indentation of considerable size formed by the arm of Cape Cod projecting almost at right angles, and enclosing Massachusetts Bay, at the bottom of which lies the commodious, but not specially accessible, harbour of Boston. The situation of the city of New York makes it by far the most important centre of foreign and domestic commerce in the United States. This superiority is due in part to the excellence of its harbour, and in part to its being the terminus of the great natural line of communication between the East and the West, a position which would seem to belong by right to some point on or near the mouth of the St Lawrence, which is the outlet of the Great Lakes, but from which this river is shut out by its north-easterly trend, which carries it into a region beyond that of successful cultivation and populous settlements, and where navigation is suspended during a considerable portion of the year by the freezing of the river.

Long Island, about 120 miles in length, and extending along the southern coast of Connecticut, is the only island of any considerable size on the whole Atlantic coast. Smaller ones—Martha's Vineyard, Nantucket, Block Island, and others—lie adjacent to Long Island on the east, and form, as it were, its prolongation in that direction, while there are indentations of considerable depth on the coast of the mainland opposite these islands.

South of New York there are two indentations of importance—Delaware Bay and Chesapeake Bay,—the former being, as it were, the expanded mouth of the river of the same name, the other being in somewhat similar relations to two large rivers, the Susquehanna and the Potomac. Philadelphia and Baltimore are the cities that command the advantages for commerce which these two bays offer.

The Gulf of Mexico is of large dimensions, and of great importance as affecting the accessibility of the interior, and still more as influencing the climate of the country. The peninsula of Florida, extending 5 degrees south of the mainland, and forming the eastern boundary of the Gulf, is in a considerable part of its surface so low and swampy as to be practically uninhabitable.

Drainage areas. The drainage areas of the country, as given by the United States census of 1880 are—Atlantic and Gulf, 2,178,210 sq. m.; Great Basin, 228,150; Pacific slope, 619,240. The drainage into the Atlantic and Gulf is distributed as follows:—New England coast, 61,830 sq. m.; Middle Atlantic coast, 83,020; South Atlantic coast, 132,040; Great Lakes, 175,340; Gulf of Mexico, 1,725,980 (the Mississippi-Missouri basin being estimated at 1,240,039).

The explanation of this overwhelming preponderance of drainage into the Gulf of Mexico at once becomes evident when we notice the general relief of the country and the positions of the various watersheds. If the level of the ocean were raised 1000 feet, a broad water-way across the American continent would be opened. There would be a great mass of land on the western side, which would comprehend nearly the whole of Mexico, and which, within the limits of the United States, would have a breadth from east to west of from 1500 to 2000 miles. North of the United States boundary line the breadth of this mass of land would diminish rapidly in width as higher latitudes were reached, but its dimensions would still be on a grand scale, although deeply intersected with inlets occupying the positions of the lower portions of the present streams of that region. A similar rise of 500 feet in the ocean would not divide the continent into two decidedly distinct and widely separated parts, but would isolate New England from the land adjacent on the north and west, by opening a channel through the Hudson River and Lake Champlain depressions, would carry the Atlantic coast-line more than 100 miles inland of its present position, and would open a deep bay in what is now the Mississippi valley, the ramifying arms of which would extend north nearly to Chicago, to Cincinnati on the Ohio, to Burlington, Iowa, on the Mississippi, and nearly to Jefferson City, on the Missouri. A rise of sea-level of 2000 feet would not materially change, as to position and size, the great land-mass on the western side of the continent already spoken of. That area would be slightly narrowed on the east, and would have its western edge more deeply indented, with the addition of groups of islands, so that its character would, south of the Columbia, be something like what it is at the present time north of that river. On the eastern side of the continent, however, the most striking changes would be effected. All the present coherent landmass east of the 97th meridian would have disappeared, and in place of it we should have various groups of islands, one of the most important of which would extend from the north line of Georgia north-east into Pennsylvania and New Jersey, where it would terminate in finger-like projections, forming north-easterly and south-westerly trending archipelagos, with various outliers in north-eastern New York, Vermont, and New Hampshire, the highest points of which would rise from 3000 to 4000 feet above the surrounding waters. The distance of these islands from the western mainland would be from 1000 to 1500 miles. Another group of islands north of the United States boundary would extend in a curving line parallel to and north of the St Lawrence and the Great Lakes.

General topography. The all-important fact in the topography of the North American continent in general, and of the United States in particular, is the existence of a central comparatively low and level region, declining gently from a watershed in close proximity to the Great Lakes on the north towards the Gulf of Mexico on the south.

One may ascend the Mississippi to its junction with the Ohio, at Cairo—a distance of 1100 miles from the Gulf of Mexico,—and the elevation attained will be only about 300 feet, an average ascent of about 4 inches to the mile. A journey of almost 1000 miles farther, to Pittsburgh, at the junction of the Allegheny and Monongahela rivers, will only give a total rise of 700 feet above the sea-level. The head of the Mississippi is in a region entirely destitute of mountains, comprising an almost level area, covered in large part by lakes and swamps, and only about 1500 feet in elevation. In ascending the Mississippi to St Louis, a distance of 1250 miles, we have reached an elevation of about 400 feet, and at St Paul, 658 miles above the mouth of the Missouri, one of a little less than 700 feet. If we follow up the Missouri to the western line of the State of the same name, where the river flows from the north, we have the choice, if we wish to keep on directly west, of following either of its great branches from that direction—the Platte and the Kansas. Up either of these we may travel for fully 500 miles, rising so gradually that the difference of elevation from day to day is hardly perceptible, the country preserving all the characteristics of a plain, although declining gently to the east.

We have, therefore, in the central region of the United States a nearly level area, which, roughly speaking, may be taken at 1250 miles square, which has a very gentle downward slope from east and west to the centre, and from very nearly its northern extremity to the Gulf of Mexico, its southern boundary, while its northern edge, with a part of the adjacent region, is occupied by five great bodies of fresh water, communicating by short narrow river-like contractions of the water areas, and all together occupying an area of about 92,000 square miles. This central valley is not entirely devoid of mountains; but these higher areas are so small in extent, as compared with the entire area of the great valley, as to be of extremely subordinate importance. Meanwhile we turn to a consideration of the higher regions east and west of the great valley.

Neither on the eastern nor on the western side of the continent are these of the nature of simple mountain chains. On the contrary, both eastern and western highlands are complicated in their orographic structure, broken into portions, each having a character of its own, and generally—but not always—preserving a certain parallelism with each other, broken longitudinally or in the direction of the prevailing trend of the elevated mass, and separated by valleys and tablelands of very varying dimensions. The western mass, which is much the larger, is also very much the more complicated of the two. It is not only made up of a great number of parts more or less distinctly separated from each other, but it even encloses a large area of plains, valleys, and mountain ranges which have no drainage to the sea. This greatly predominating complexity of the western elevated region is the result of complicated geological conditions, to which are added climatic peculiarities of a striking character. The eastern elevated side of the continent is now generally called the Appalachian region, the western the Cordilleran.

The Appalachian System.

General character.The Appalachian ranges all belong to an ancient system of uplift or disturbance, and have not been invaded by volcanic materials of recent date, presenting in both these respects a most marked contrast to the Cordilleran system. There are volcanic rocks along portions of the eastern slope of the Appalachians; but these eruptive dikes and overflows are not of Tertiary and post-Tertiary but of Mesozoic age.

The name Appalachian is used as designating the entire complex of ranges, valleys, and tablelands in and upon which is situated the watershed from which on the north and east the streams descend to the Atlantic, either directly or through the Great Lakes, and on the west and south-west to the Gulf of Mexico, either as tributaries to the Mississippi or flowing directly to the Gulf. As thus defined, the Appalachian region, system, or complex of ranges extends from the promontory of Gaspé, in a mean direction of north-east and south-west, to Alabama, a distance of about 1300 miles, where it disappears under the much more recent geological formations which form a broad belt along the Gulf of Mexico and extend far up the Mississippi valley.[2]

While the Appalachian system, as a whole, is so nearly continuous, as a feature of the topography of the country, that it seems necessary to include it all under one general designation, yet different portions are extremely unlike each other in some of those features which are generally looked upon as essential to the unity of a mountain system.

Break of the Hudson and Mohawk.One very marked and important break at least in the system is that occupied by the Hudson river, and extending up the Mohawk valley to the west, to a connexion with the Great Lakes, and up the Champlain valley to the north, to a connexion with the St Lawrence river direct. A rise of 152 feet in the sea-level would isolate from the rest of the continent all of New England and that part of Canada lying to the south-east of the St Lawrence, as far as the extremity of Gaspé. A further rise of 278 feet would open a waterway from the Atlantic to the Great Lakes, and leave the mass of the Adirondacks as an island lying adjacent to New England on the east and the Appalachian land-mass on the south. We seek in vain for any other break in the Appalachian system as complete as this, and, when we compare the portions of the system lying on each side of this line of division, we recognize the fact that there are essential points in which they differ from each other, and that these differences are greater than any presented by the various portions of the system in its extension to the south-west of the Hudson river break. This north-eastern division of the Appalachians may therefore properly first have its more important topographical and geological features indicated.

Beginning with the division west of Lake Champlain and north Adirondacks. of the Mohawk valley—the Adirondack region or mountains,—we find, on examination, that this region is not only to a certain extent isolated topographically from the rest of the Appalachian ranges, but that it belongs to a geologically older system.[3] The rocks are all eruptive in the central portion of the mass, and chiefly gabbro and granitic or gneissoid in character. On the exterior, especially on the eastern edge, are deposits of limestone, which are generally believed by geologists to be sedimentary beds, highly metamorphosed, but which, in the opinion of the present writer, are more probably of the nature of chemical precipitates.

The drainage of this region is radial from a central point, but the slope on the east is shorter than on any of the other sides. From Tahawas or Mount Marcy (5344 feet) the drainage is to the north-east by the Au Sable to Lake Champlain, by the Raquette to the St Lawrence, and by the branches forming the head waters of the Hudson to the south. The dominating line of elevations runs nearly east and west, with high spurs and narrow ridges on the north, of which Whiteface Mountain (4871 feet) is one of the most conspicuous. The region is one of numerous lakes and lake-like expansions of the rivers, so that, with short portages, a large part of it can be visited by boat or canoe. The lake region proper of the Adirondacks lies at an elevation of from 1500 to 2000 feet:—Lake Silver, 1983 feet; Placid, 1950; Saranac (Upper), 1606, (Lower), 1557; Smith's, 1738; Tupper's, 1504; Long, 1584; Raquette, 1765; Sandford, 1685; Cranberry, 1570. There are indications of the Appalachian trend in the north-east south-west direction of several of the larger lakes; but many others trend nearly north and south. The number of the lakes is the result of the impermeability of the rock, the general uniformity in height of the region, and the broken character of the surface, which is very irregularly covered by large masses of rolled detritus. The work of water is everywhere distinctly visible, and that of ice hardly perceptible.

New England topography.The mountainous region east of the Hudson in New England is the portion of the system most irregular in topographical features. Two groups of elevations, however, are quite well marked, and have distinctive names—the Green, and the White Mountains. These are separated by the Connecticut river, which has a nearly north and south course parallel to the range of the Green Mountains. Nowhere in this latter range is there a continuous uplift forming a long ridge or crest, but there is a gentle swell of the surface on which here and there rise elongated groups of considerably higher summits. In the extreme south western corner of Massachusetts rises Bald Peak (2624 feet), and in the north-western corner Graylock or Saddle Mountain (3505 feet). Still farther north, in Vermont, are the culminating points of the Green Mountain range: Equinox, 3872 feet; Pico, 3935; Camel's Hump, 4077; Killington, 4221; and Mansfield, 4389 feet. Ascutney, a quite isolated point, near the Connecticut river, has an altitude of 3163 feet.

The Connecticut river makes a very complete separation, in all but the extreme northern portion of its course, between the Green Mountain system and the highlands to the east, which have no collective name, but are in a measure the continuation of the White Mountain range. In Massachusetts this swell of land—for more it can hardly be called—has an elevation of about 1000 feet, the valleys being rarely sunk more than 200 or 300 feet below the general level of the gently undulating higher lands. Occasionally there is a higher point, like Wachusett (2018 feet) or Monadnock (3169 feet). From Monadnock the region east of the Connecticut broadens very much, the coast-line trending rapidly eastward. The country becomes more and more mountainous, but still without continuous ranges. The mountains are grouped around various central points, of which the most important are Moosilauke or Moosehillock (4790 feet), Lafayette (5290), and Washington (6290 feet). The last-named is the highest point in the Appalachian system north of North Carolina, and rises nearly 500 feet above all the adjacent summits.

Farther east, in Maine, and in the neighbouring portions of Canada, the topography has been little worked out in detail. So far as known, in Maine the irregularity of the range is still greater than it is farther to the south-west. There is in this part of New England no coast-region, but a gradual rise from the seashore towards the interior for about 140 miles, to the divide between the waters running into the Atlantic directly and those tributary to the St Lawrence or forming the head of the St John. This divide, which has a general direction of pretty nearly east and west, is at an altitude of about 1800 feet at the western edge of Maine, and declines to about 600 feet on its eastern boundary. The southern slope is a very gradual one to the sea, and though broken and rocky is not diversified by any marked ridges or long elevations. The high points rise sometimes nearly isolated, and sometimes in clusters, having little of the ridge character. Ktaadn (Katahdin) is the dominating peak (5215 feet), and it rises in such isolation as to look in the distance like a volcanic cone. That part of Maine lying south of the watershed is drained by streams running nearly southward. Like the Adirondack Wilderness, it is a district of numerous lakes, as might be expected, since it has no rapid descent in any direction, is underlain by impermeable rocks, and has a considerable rainfall.

The general uniformity of character in the New England portion of the Appalachian system will be evident from what has been here stated. This region is marked by comparatively low swells of ground, on which rise groups of higher points, rather irregularly distributed, nowhere reaching the limit of perpetual snow, and nowhere presenting great obstacles to internal communication.

Geology of New England.The geological structure of this north-eastern prolongation of the Appalachian system has been as yet only imperfectly made out, a circumstance due in part to the extreme scarcity of fossil remains, and in part also to the fact that the various sedimentary beds have been so metamorphosed as to be distinguishable only with great difficulty from the associated eruptive formations, while even the latter have frequently been themselves so much changed by chemical action, since their appearance at the surface, that it is only by the microscope that their real nature can be made out.

With the exception of a narrow belt of Mesozoic rocks in the Connecticut valley and a small basin of similar age in Woodbury and Southbury, Connecticut, and also with the exception of a very limited set of deposits of late Tertiary age on the eastern boundary of Lake Champlain and on the Atlantic coast, there are, so far as known, no rocks in New England more recent than the Palæozoic. Tertiary and Cretaceous formations are, however, found covering small areas on some of the islands adjacent to the coast. Along the western sides of Vermont and Massachusetts the rocks are clearly proved by their fossils to be of Lower Silurian age, and their structure has been made out in part. There are faults and synclinals,—limestones and chloritic and talcose slates being the predominating rocks. The dips are chiefly to the eastward, and the rocks are of more recent age as we go east from Lake Champlain, on the east side of which is a large development of the Potsdam sandstone. In and near the Connecticut valley, to the east of the region just noticed, fossils have been found (at Bernardston, Mass.) of late Upper Silurian age (Helderberg of the New York Survey). The stratigraphical relations of these rocks, however, remain obscure. The same may be said of nearly or quite all of New Hampshire, of eastern Massachusetts, and of a large part of Maine. At one point in New Hampshire (Littleton) fossils have been obtained of the same age as those discovered at Bernardston. In northern Maine, traversing the State in a wide belt running north east and south-west, rocks occur of Upper Silurian and Lower Devonian age, well characterized by fossils, a part of this belt being clearly identical in age with the Oriskany sandstone of the New York Survey. The stratigraphical relations of these rocks are still obscure; and south of this fossiliferous belt is a wide area in which no fossils have yet been found. At one point in Massachusetts (Braintree, near Boston) more than fifty years ago fossils of the lowest Silurian age (Primordial) were found, and there are other indications that the rocks near the coast of New England, from Cape Cod north, are very low down in the fossiliferous series; but in Massachusetts, as well as in New Hampshire and Maine, there is a wide area between the Devonian and Upper Silurian rocks of the Connecticut valley and northern central Maine, of the geological age of which nothing is definitely known, and of which the stratigraphical relations are still very obscure.

The Appalachians south-west of the Hudson.To the west and south-west of the Hudson river, in New York, the intricacy and obscurity of the orographic structure and geological age of the Appalachian system begin to be cleared up; but it is not until we reach Pennsylvania that the characteristic features of this range are fully developed. (1) The first of these is, along its south-eastern edge, as we cross the system from south-east to north-west, at right angles to its trend, a series of elevations, at first comparatively unimportant and more or less detached from each other, but gradually, in going in a south-westerly direction, becoming more and more prominent and continuous, until, towards the extreme southerly end of the system, it forms the most imposing connected mass of high plateaus and still higher ridges anywhere exhibited in the Appalachian region. This south-easterly division of the system has no single distinguishing appellation. It is called in Pennsylvania the South Mountains, in Virginia the Blue Ridge; and in North Carolina and Tennessee it has various names applied to its different portions. (2) The second important feature is the Appalachian region proper of the First Pennsylvania Survey, a region of wave-like folds or corrugations, not so metamorphosed but that the geological sequence can be distinctly made out, in which orographic disturbances and a peculiar erosion have developed an interesting and intricate topography. (3) The third region, in crossing the chain from south-east to north-west, is that of plateaus, bounded on the south-east by an escarpment to which the name of Alleghany has been most frequently given, and by which that portion of the system which lies in Pennsylvania is still almost universally known. These are the three most prominent divisions of the system south-west of the Hudson, and between the first and second there is a pretty well marked depression which is a very conspicuous feature of the topography in Virginia and Tennessee (in Pennsylvania the Kittatinny, in Virginia the Great Valley, and farther south the valley of East Tennessee).

New Jersey and southern New York.We next come to consider the region occupied by south-eastern New York and northern New Jersey. The Green Mountains are generally considered as being prolonged south-westwardly in the Hudson river highlands and the highland range of New Jersey. This latter range in New Jersey occupies a belt of country somewhat over 20 miles in width on the New York line, but narrows down to less than half that on the Delaware. It includes no long unbroken ridges, the one which comes nearest to having this character being the Green Pond range, about 12 miles long; nor are the subordinate ridges of which it is composed in a line with each other or with their axes parallel to the direction of the main range. The highest point is Rutherford's Hill (1488 feet). The Kittatinny valley mentioned above is distinctly marked in New Jersey, where it also bears the same name. It is bounded on the north-west by a range called the Shawangunk in New York, the Blue Mountains in New Jersey, and the Kittatinny in Pennsylvania. It lies on the extreme north-western edge of New Jersey, and forms an almost unbroken straight line for about 40 miles within this State, from the Delaware Water Gap to the New York line. The straightness of this ridge and its almost level crest, from 1200 to 1800 feet in height, mark it as belonging to the peculiar topographical belt of the Appalachian system which occupies central Pennsylvania. This belt is distinctly recognizable still farther east, in the vicinity of Catskill and east of the mountains of that name, where there is a miniature group of hills, called the Little Mountains, only a mile or two in width, and but a few hundred feet in height, which are made up of rocks of the same geological age as those in the central division of the Pennsylvania Appalachians, and with the same characteristic structure.

New York plateau.The plateau region occupies a large portion of the State of New York. Its northern edge extends along the south side of the Mohawk river, forming a distinctly marked escarpment (the Helderberg Mountains). Farther west, in central New York, only a few miles from the central depression through which pass the Erie Canal and the great lines of railroad connecting Albany with the west, are various heights rising 1000 feet above the surrounding plateau, and from 1600 to 2000 feet above the sea (Fenner Hill, in Madison county, 1862 feet; Ripley Hill, in Onondaga, 1968; Niles, in Cayuga, 1623; Milo Hill, in Yates, 1343; and East Hill, in Oswego, 2300). This plateau, with its extension to the south-west into Pennsylvania, forms the highlands in which rise the various branches of the Susquehanna, which traverses the entire Appalachian system, in a general direction from north-west to south-east.

The Catskills.In south-eastern New York there is a remarkable group of mountains, very conspicuous from the Hudson, and apparently quite isolated as seen from the eastern side, but which, as approached from the west, are recognized as being in intimate connexion with the plateau region of central New York. This group is known as the Catskills, of which there are two divisions,—the northern, or Catskills proper, and the southern. The Catskills proper are a massive plateau, enclosed between the Esopus and Catskill creeks—affluents of the Hudson running in a south easterly direction, nearly parallel with each other, and at a distance of about 25 miles apart. The mountain mass thus enclosed consists essentially of two border chains running parallel with the two streams, and from 10 to 15 miles apart. The highest point of the north-eastern border is the Black Dome, 4002 feet; that of the south-eastern range, Hunter Mountain, 4038 feet. The southern group, lying south of the Esopus, the drainage of which on the west and south-west is into the Delaware river, includes Slide Mountain, the highest point of the entire Catskill group (4205 feet).

The Catskill range is, like the plateau region generally, of the Devonian and Lower Carboniferous formations. The upper 500 feet of the Slide Mountain is occupied by a cap of the Carboniferous conglomerate, the equivalent of the millstone grit, and the nearest approach to the Coal-measures in any part of New York.

Appalachians in Pennsylvania and southward.
Atlantic slope.

The second or upper belt of the Atlantic slope is in large part made up of rocks which are destitute of fossils, and in regard to which it has not yet been clearly made out whether they are really stratified beds older than the lowest Silurian (Azoic or Archæan of Dana), or whether they are highly altered rocks of Palæozoic age. Over most of its area it has very distinctly the character of a plain. In Pennsylvania it is highly cultivated and densely peopled, a “country of rolling hills and gently-sloping vales, with occasional rocky dells of no great depths, and nowhere more than 600 or 700 feet above the sea-level.” It is bordered on the north-west, for a portion of its extent, by a low range of elevations, known as the South Mountains, and generally considered to be the northern prolongation of the Blue Ridge of Virginia and the States farther south. The South Mountains enter Pennsylvania at the Delaware, forming a region of irregularly grouped ridges, which occupy a breadth of somewhat less than 10 miles, and which do not rise to an elevation of more than 400 to 500 feet above the valleys. These hills are made up of massive varieties of gneiss, sandstones,—recognized by the Pennsylvania Survey as being of Potsdam age,—and Lower Silurian limestones. The valleys resting on this latter rock are covered with a highly fertile soil. In Virginia the upper belt of the Atlantic slope broadens out and becomes more and more complicated in its topography. In an official report[4] this region is divided into three portions, called the Middle, Piedmont, and Blue Ridge divisions of the State. The Middle division is said to extend westward from the sea to the foot of the low broken ranges which, under the names of Kitoctin, or Kittoctan, Bull Run, Yew, Clark's, South-West, Carter's, Green, Findlay's, Buffalo, Chandler's, Smith's, &c., Mountains and Hills, extend across the State south-west from Fairfax county on the Potomac to Pittsylvania county on the North Carolina line. These broken ranges, which preserve a general parallelism with the Appalachian ranges proper, and form, as it were, the outliers of this system, are designated by Major Hotchkiss the Atlantic Coast range. This middle country is described as a moderately undulating plain, from 25 to 100 miles wide, and rising from the south-eastern border, where it is from 150 to 200 feet above the sea, to an altitude of from 300 to 500 feet along its north-western edge. It is a succession of low north-east and south-west trending ridges, the valleys between them being sometimes narrow and deep, but the ridges themselves not very prominent. The rocks are metamorphic slates and gneiss, with numerous eruptive masses in the form of dikes, and with many quartz veins, some of which contain considerable gold, although mining has not, on the whole, been successful. The Piedmont division forms a belt of from 20 to 30 miles in width, and may properly be considered as being the foothill border of the Blue Ridge itself. There is a marked tendency to the formation of a continuous valley between the broken ridges already noticed as forming the so-called Coast range and the Blue Ridge proper. In this valley lie Culpeper (400-500 feet), Fairfax (382), Charlottesville (450), Lynchburg (650), and other towns. This foothill region is described as exceedingly intricate. The coast range is succeeded, in the west, by numberless valleys of all imaginable forms, which extend across to, and far into, the Blue Ridge, to which, in point of fact, they topographically belong. Portions of the Piedmont country, however, form quite extensive plains.

The Blue Ridge.The Blue Ridge with its belt of foot-hills—the Piedmont region—forms a conspicuous feature of the topography of Virginia and the States farther south. The Potomac breaks through it at Harper's Ferry, at an elevation of 242 feet, the mountains adjacent rising about 1200 feet higher. The passes, locally known as “gaps,” are numerous, and several of them are traversed by railroads. The James river intersects the Blue Ridge 706 feet above the sea. The elevation of this range is considerable, even in its northern portion,—Mount Marshall, near Front Royal and Manassas Gap, being about 3370 feet. The height as well as the breadth of the range increases rapidly as the southern line of Virginia is approached, the Peaks of Otter, in Bedford county, near Buford's Gap, rising to 4000 feet, and Balsam Mountain, just at the North Carolina line, to 5700 feet. Here the Blue Ridge has already begun to expand into that wide and high plateau, occupying the western portion of North Carolina, in which are found the highest points of the entire Appalachian system. This elevated region is formed by a broadening out and bifurcation of the Blue Ridge, which begins near Christiansburg (2012 feet), opposite the point where the New river changes its course from a direction parallel with that of the Appalachian ranges to one at right angles to this, and breaks through that part of the system which lies north-west of the Great Valley, flowing in that direction to the Ohio. This plateau rises in North Carolina to an average height of 2500 feet, while portions of it are over 3500. It is about 150 miles in length, with a width varying from 15 to 50 miles and averaging about 30, and reaches its highest altitude (3500-4000 feet) at its narrowest part. The plateau is bordered by broken ranges; that on the south-east still continues to be called Blue Ridge; the more or less continuous line of elevations in the south-west has various names—Unaka, Smoky, Bald, and Iron being among the number. Between these exterior ridges run various spurs, with many points over 6000 feet, the culminating one being the Black Dome (6707 feet). The northern portion of this high region is drained by the head waters of the New river, but the principal drainage of the most elevated part is to the north-west, from the plateau, through gaps in the western ridges, to the Tennessee.

The geology of the Blue Ridge division of the Appalachian system is obscure and difficult. Most of the rocks are highly crystalline, but whether of Palæozoic or Azoic age is as yet undecided. These crystalline rocks are more or less intersected by ancient eruptive masses, the range and extent of which are uncertain. Flanking the Blue Ridge on the west side, and involved in the disturbances of the strata by which it has been built up, are sandstones and limestones which in Pennsylvania have been recognized as being of Lower Silurian age. These limestones seem to become more arenaceous farther south, and also to become unfossiliferous. It may be assumed, however, that the range in general is made up of rocks not newer than Lower Silurian.

The Great Valley.To the west and north-west of the Blue Ridge division of the Appalachian system lies the Great Valley. Its north-western limit in Pennsylvania is the Kittatinny Mountain, which separates it from the mountain district lying adjacent to it on that side by a very regular natural wall. The entire length of the valley in that State is about 165 miles, and its width between 10 and 11 miles. Throughout its whole extent it presents a gently undulating surface, approximating to a level plain, with here and there a belt of low hills. In Virginia the Great Valley is a very important feature, having a length of a little over 300 miles, and a quite uniform width of about 20. The mountains on its north-west side (the Kittatinny) are known by a variety of names, their northern portion being designated on some maps as the Great North Mountains. A small portion of this valley is also included within the limits of West Virginia. The drainage of the Great Valley is complicated. The north-eastern portion is the beautiful and fertile valley (about 140 miles in length) of the Shenandoah, drained by two parallel branches of that river,—a well-marked double range, 50 miles in length, dividing the Great Valley longitudinally. Through this range the northern branch of the Shenandoah breaks, turning at right angles to its former course and uniting with the southern branch at Manassas Gap, below which, towards the Potomac, the valley is but slightly broken by hills. South of that portion of the Great Valley drained by the Shenandoah is a region about 50 miles in length, in which are the head waters of the James. The various branches of this river, coming from north-west and south, unite at the western base of the Blue Ridge, and there break through it. The remaining portion of the valley belongs partly to the Roanoke and partly to the New or Kanawha and to the Holston. The rise of the Great Valley to the south-west is very marked, the point where the Shenandoah enters the Potomac being only 240 feet above the sea, while the head of the New river is in a region ranging from 2500 to 3000 feet. According to the official report, “the aspect of this region is singularly pleasant. The great width of the valley, the singular colouring and wavy but bold outline of the Blue Ridge, the long uniform lines of the Kittatinny Mountains and the high knobs that rise up behind them in the distance, the detached ranges that often extend for many miles in this valley, like huge lines of fortifications all these for the outline, filled up with park-like forests, well-cultivated farms, well-built towns, and threaded by bright and abounding rivers, make this a charming and inviting region.”

Middle belt of the Appalachians.The next division of the Appalachians, which may be called the middle belt, is perhaps the most interesting and peculiar although not the most persistent portion of the system. It is not until Pennsylvania is reached that it becomes important. Its character of this division is thus indicated by H. D. Rogers in the report of the First Geological Survey of that State:—“It is a complex chain of long, narrow, very level mountain ridges, separated by long, narrow, parallel valleys. These ridges sometimes end abruptly in swelling knobs and sometimes taper off in long slender points. Their slopes are singularly uniform, being in many cases unvaried by ravine or gully for many miles; in other instances they are trenched at equal intervals with great regularity. Their crests are for the most part sharp, and they preserve an extraordinarily equable elevation, being only here and there interrupted by notches or gaps, which sometimes descend to the water-level, so as to give passage to the rivers. The whole range is the combined result of an elevation of the strata in long, slender, parallel ridges, wave-like in form, and of excessive erosion of them by water; and the present configuration of the surface is one which demonstrates that a remarkable, and as yet little understood, series of geological events has been concerned in their formation. The ridges, which are but remnants of the eroded strata, are variously arranged in groups with long narrow crests, some of which preserve a remarkable straightness for great distances, while others bend with a prolonged and regular sweep. In many instances two narrow contiguous parallel mountain crests unite at their extremities, and enclose a narrow oval valley, which with its sharp mountain sides bears not unfrequently a marked resemblance to a long slender sharp-pointed canoe.” The system of ranges thus described crosses north-western Maryland, and is largely developed in Virginia, with characters in many respects resembling those which it displays farther north. It is a region of long, narrow, parallel valleys, separated by narrow, straight, and quite elevated ridges, of which no one rises greatly above the others. The number of these diminishes as we proceed in a south-westerly direction, and the belt narrows in proportion. The drainage on the extreme eastern border is into the Great Valley. The interior valleys in the northern half of the belt are traversed by streams flowing into the Potomac; the southern half is mostly drained by branches of New river. In some cases the waters flow towards a central depression in the valley, and then break through the enclosing range. The rocks are Silurian, Devonian, and Lower Carboniferous. In its north-eastern extension this division includes the anthracite region of Pennsylvania. The ridges by which the Appalachian belt proper is traversed are the result of a series of flexures of the strata, along numerous axes of elevation and depression, not complicated in Pennsylvania by extensive dislocations of the crust, or faults, either longitudinal or transverse, but gradually becoming so as we proceed towards the south-west. It is a curious fact that, in following in a south-westerly direction the Appalachian belt proper, we find that it disappears or merges in the Great Valley, which in Tennessee occupies the entire space between the Blue Ridge and the Alleghany plateau, there called the Cumberland tableland, and is fully 40 miles in width. The rocks which underlie the Great Valley are of the same geological age as those of which the Appalachian belt is composed farther north, but more closely compressed together, and complicated by great longitudinal faults, the whole area being a comparatively depressed one, and not broken by marked ridges. The true Appalachian belt is therefore limited to Pennsylvania, Virginia, and a very narrow space in Maryland, and here the system has its greatest width and most intricate and interesting topographical features, but not its highest elevations, which occur north and south of this portion, in regions of greater and more irregular disturbance, complicated by metamorphic changes and irregular intrusions of eruptive material on a grand scale, making the task of unravelling the geological structure extremely difficult.

The Appalachian plateau.The most western member of the Appalachian system—the plateau—is the one of which the geology is most easily made out, and, while its eastern border is an important topographical feature, it merges so gradually in the great central or Mississippi valley, that any definition of its limits in that direction is quite impossible. The position and elevation of this plateau region in New York have seen already indicated. Farther south this tableland occupies the western portion of Pennsylvania, nearly all West Virginia, a part of Kentucky and also of Tennessee, in which latter State it is called the Cumberland tableland, or the Cumberland Mountains, since it here presents itself exceptionally with abrupt edges on the west, as well as on the east, but it is narrowed down to a width of not more than 30 or 40 miles in the northern portion of the State and of much less in the southern. The bold escarpment with which the plateau faces the east in Pennsylvania is known as Alleghany Mountains. the Alleghany Mountains. It is continued in Virginia, but with much less distinctness. It is from this tableland that the waters of the Susquehanna descend to the Atlantic, crossing the entire Appalachian system in its course, while with the New river the condition of things is reversed, since this stream heads on the eastern edge of the range and flows across it in the opposite direction from that of the Susquehanna.

Since the dip of the strata in southern New York and western Pennsylvania is generally to the southward, newer rocks occupy the surface as we proceed in that direction. The Coal-measures appear soon after the line of division of these two States is passed, and it is largely with rocks of this age that the tableland is covered through the whole of its southern extension. In fact the Appalachian coalfield, as it is called, presents an almost continuous mass of coal-bearing strata, extending from northern Pennsylvania to Alabama. A part of this field, however, notably in Ohio, reaches far beyond the topographical limits of the Appalachian region.

The Appalachian tableland is, even in Pennsylvania, not entirely destitute of marked topographical features. The axes which characterize the system farther east are not wanting, but the strata are raised and depressed by gentle undulations, and not broken by precipitous ridges. Much intricacy is given to the topography, in a small way, by the streams cutting down into the soft rocks.

Mesozoic belt of the Appalachians.Before leaving the Appalachian region it will be desirable to add a few words on the belt of Mesozoic rocks occurring on the Atlantic slope, which, although not forming a prominent topographical feature, are of much geological and palæontological interest. This belt consists chiefly of sandstones of reddish-brown colour, with which are associated shales, and occasionally, especially in the lower portion, coarser materials—conglomerates,—which are sometimes well rounded by water, but in places almost breccia-like in character. These rocks are first seen, on the north-east, in New Brunswick, in Nova Scotia, and on Prince Edward Island. In New England they are limited to the valley of the Connecticut river, with a small parallel area a little to the west of this in the towns of Southbury and Woodbury, Connecticut. The Connecticut valley Mesozoic area is about 150 miles in length, with a maximum breadth of about 14. The largest belt, however, is that extending from the west side of the Hudson river, along the south-eastern side of the South Mountains and Blue Ridge, through New Jersey, Pennsylvania, and Maryland, to about the centre of Virginia, having a maximum width of about 30 miles and a length of some what over 300. There are other smaller areas of the same rock in Virginia and in North Carolina,—that of the last-named State extending a short distance into South Carolina. Associated with this sandstone is a considerable amount of igneous rock, which occurs in the form of dikes and overflows, which, as the sandstone has been worn away by erosive agencies, occasionally stand out quite conspicuously, although nowhere reaching an elevation of more than a few hundred feet. Some well-known and much visited localities, such as Mount Tom and Mount Holyoke in Massachusetts, the Hanging Hills and East and West Rocks in Connecticut, and the Palisades in New York, are of this character. The fossils which the sandstones contain are not numerous, but are of much interest, and the geological age assigned to this formation by most palæontologists is the Triassic. In several localities, however, great numbers of footprints of animals occur, which were long considered to be those of birds, but are now known to belong—in considerable part, at least—to the Reptilia, some of which had certain features allying them to birds. The paucity of fossil remains, other than footprints, found in these rocks has rendered the working out of their true relations a matter of considerable difficulty. The latest investigations of Prof. Fontaine show that the Mesozoic areas of Virginia are separable into two quite distinct groups, an older and a newer, the floras of the two being quite different. It is in the older Mesozoic of Virginia, and in the most easterly area, near Richmond, that the coal occurs which was the first worked in the United States. The stratigraphical relations of the Mesozoic sandstones are difficult of comprehension, and have been the occasion of much discussion.

The Cordilleran System.

For convenient description, this system may be divided into the following six regions:—I. the Rocky Mountains; II. the Great Basin and the Basin ranges; III. the Northern or Columbian plateau; IV. the Southern or Colorado plateau; V. the Sierra Nevada and Cascade ranges; VI. the Pacific Coast ranges.

The Rocky Mts.I. The Rocky Mountains form the eastern border of the Cordilleran region, a border made up of many subordinate ranges. They may be divided into two parts—the north and south trending portion, and the north-west and south-east trending portion. Between these two subdivisions there is a marked orographic break, in the form of a high plateau region, over which the Union Pacific Railroad passes at an elevation of about 8000 feet. On the north of this plateau are the Sweetwater and the much higher Wind River Mountains, which latter form the culminating region of the continent, since in them head the three great river-systems of the country,—the Missouri, the Columbia, and the Colorado.

Southern division.(A) The southern or north and south trending division is about 600 miles in length from north to south, and about 300 in breadth. Its eastern edge is extremely well marked, the ranges rising abruptly from a very gently sloping plateau. Looking at this division in the most general way, we find on its eastern edge a double range of mountains, quite distinctly marked in Colorado, or between the parallels of 36° and 41°, where they enclose a system of high plateau-like valleys, known as the North, Middle, South, and San Luis Parks, which have an elevation of from 6000 to 10,000 feet, the enclosing ranges rising 3000 to 4000 feet higher. These so-called parks are drained by the head waters of the Platte, Colorado, and Arkansas, with the exception of the San Luis Park or Valley, in which lies the upper course of the Rio Grande, whence it finds its way southward, through New Mexico, having a pretty well marked and lofty range on its eastern side, and more broken ones on the west, the two representing the Front and Park ranges of Colorado.

Colorado range.The Front or Colorado range proper, beginning as a junction at the south of the Medicine Bow range and the Laramie Hills, which are low inconspicuous ranges closing in the plateau above noticed is separating the two divisions of the Rocky Mountain system, is a broad, lofty mass, continuous from about 41° N. lat. as far as Pike's Peak, about lat. 38° 45′, when it runs out into the plain. The best known points in the Rocky Mountains—Long's Peak (14,271 feet) and Pike's Peak (14,147 feet)—are in this range; both are visible from the plains, and are conspicuous landmarks. Gray's Peak (14,341 feet) is the highest point in this range; but, although on the continental divide, it is too far west to be visible from the plains. This divide, which separates the Atlantic waters from those of the Pacific, follows the Front range as far as Gray's Peak, where it is deflected westward for 20 miles to the Sawatch range, which it follows for about 75 miles. In this deflexion the divide passes between Middle and South Parks, the lowest pass in this part being that called the Tennessee (10,418 feet), which leads from the head of the Arkansas to the Grand river branch of the Colorado.

Sawatch range.The Sawatch range is one of the highest and best-marked chains in the Rocky Mountains. It lies west of the head of the Arkansas; and its dominating peaks, along the whole range, exceed 14,000 feet. The most northerly of these, the mountain of the Holy Cross (14,176 feet), was so named on account of the existence on its eastern flank of a large snow-field lying in two ravines which intersect each other at right angles, in the form of a cross, and which in summer is conspicuously visible from a great distance. The highest point is Mount Harvard (14,375 feet), and the passes range from 12,000 to 13,000 feet. The continental divide follows the Sawatch range to its southern end, in lat. 38° 20′, and then runs in a south-westerly direction for about 75 miles, over a high region without any distinctly marked range. Here it turns, and, San Juan range. running south-easterly, follows the crest of the San Juan range, which at many points rises above 13,000 feet. This range forms the western border of the San Luis Park, and, from its north western end, in going either north, north-west, or west the explorer finds a very elevated and exceedingly broken country, which finally merges in the plateau or “mesa” region of western Colorado. Uncompahgre Peak (14,235 feet), a magnificent isolated summit of volcanic materials, is the culminating point.

Elk Mts.West of the Sawatch range is that of the Elk Mountains, a volcanic mass of sharp pinnacles, the culminating point of which—Castle Peak—is over 14,000 feet. Between the Elk Mountains and the Uncompahgre rise the various branches which unite to form the Gunnison river.

Plateau region.The entire western portion of Colorado is a high plateau region of sedimentary rocks, of late geological age, cut deeply into by numerous streams, giving rise to canons or ravines alternating with mesas or plateaus, the whole forming a labyrinthine succession of depressions and elevations. The mesas are sometimes so nearly eroded away that the remaining portion of the flat surface is hardly wide enough for a bridle path, while at other times there is a broad area of nearly level surface, bordered on each side by tremendous precipices. The small valleys on the streams, where there is an area of level land large enough to be a feature in the topography, are called “parks” or “holes.” The area occupied by these, in Colorado, west of the continental divide, is extremely small as compared with the whole area of the region.

The “Parks.”When the “Parks” are spoken of in describing the Rocky Mountains, it is usually the more conspicuous ones—the North, Middle, South, and San Luis Parks—that are intended. The North Park is a tolerably level area, about 40 miles by 20, and quite walled in by high ranges rising at points above 12,000 feet. This park is a favourite resort for hunters. In it rises the North Fork of the Platte. Middle Park, which is drained by the branches of Grand river (not to be confounded with the Rio Grande), is much more broken by elevations than North Park. The continental divide surrounds it on every side excepting the west. There are but few points in this park under 7000 feet. South Park is about 40 miles in length and 15 to 20 in breadth; it is more nearly level than the North or the Middle Park, and has a higher elevation than either (about 10,000 feet at its northern end, and declining gradually southward to about 8000). The San Luis Park, or Valley, is much larger than the more northern parks, and is closed in by high mountain ranges. On its north-eastern side it Sangre de Cristo range. has as a boundary the Sangre de Cristo range, of which the culminating point is Blanca Peak, the highest point in the Rocky Mountains (14,463 feet). The Sangre de Cristo range is almost a continuation of the Sawatch, having the same trend and similar geological characters, the two being separated by the broad depression known as Poncho Pass (about 9000 feet). The north-western portion of the San Luis Valley is closed in by the Garita Hills, which are a part of the great irregular volcanic mass continued north-westerly by the Uncompahgre and south-westerly by the San Juan range. The San Luis Valley is traversed by the Rio Grande, which enters it from the west, rising in the above-mentioned volcanic region. In the northern and wider portion of the valley the mountain streams sink or are lost by evaporation before reaching the Rio Grande, and most of this portion of the valley is sandy, and unfit for cultivation except where it can be artificially irrigated.

Proceeding westward from the San Juan range, in lat. 38°, we soon enter the plateau region which extends to the Colorado. This plateau region is bounded on the north-east by the volcanic masses of the San Juan, Uncompahgre, and Elk Mountains, at the western base of which spread the great tables or mesas of Cretaceous and Tertiary rocks, often capped by volcanic materials, and which thus already begin to exhibit the characteristic features of the plateau region. Here the principal branches of Grand river and of the Gunnison have their sources. The Grand itself rises in the Front range on the western slope of Long's Peak, while the Gunnison heads near Mount Harvard, the two uniting near the western boundary of Colorado, near 39° N. lat.

Uintah range.Directly west of North Park, and separated from it by the mass of the Park range, and also by a broad belt of high mesa country, is the Uintah range, remarkable as having an east and west trend, and thus forming a sort of connecting link between the eastern edge of the Rocky Mountain system and its western border, of which the Wahsatch range is the most strongly marked division. Starting from the eastern side of the Wahsatch, where it inosculates with it, it runs east for a distance of 150 miles, when it sinks and becomes lost in the Tertiary and Cretaceous mesas lying west of the Park range. It has the Bridger basin on the north—a continuation of the Laramie Plains,—the two together forming an important member of the orographic break between the north and south divisions of the Rocky Mountain system already indicated. The Bridger basin is underlain by rocks of Eocene age, and has an elevation of 6000 to 7000 feet, Fort Bridger itself being 6,753 feet above the sea. South of the Uintah range is the Uintah Valley, also underlain by Tertiary rocks, and forming the north-westernmost division of the Colorado plateau region. The Uintah range itself is of very simple geological structure, being a low flattened anticlinal, complicated by one or more faults on the northern edge. The Tertiary and Cretaceous strata have been so much eroded away on the higher portion of the flattened arch that the chief rock exposed in the body of the range is the underlying Carboniferous. The highest points, as given by the Fortieth Parallel Survey, are Gilbert's Peak (13,687), Tokewanna (13,458), and Wilson's Peak (13,235 feet). The southern slope of this range is drained by the affluents of Green river, which unites with the Grand, about 175 miles farther south, to form the Colorado.

Wahsatch range.The Wahsatch range is one of the most conspicuous of the Rocky Mountain system, and, as it borders the Great Basin on its eastern side, it may properly be considered as forming the western limit of the Rocky Mountain southern division of the Cordilleran system. This range has a nearly north and south trend, rising with a bold escarpment to a height of nearly 12,000 feet in the portion of the range just east of Salt Lake City, but falling off gradually towards the north and not being recognizable as a distinct range beyond Bear river. This stream rises on the northern slope of the Uintah range, in nearly the same latitude as that of Salt Lake City, then flows northward for more than 100 miles, to lat. 42° 40′, when it turns and follows an almost exactly opposite course, finding its way round the north end of the Wahsatch range and emptying into Great Salt Lake. In the loop thus made by Bear river is the Bear River range, in which North Logan Peak rises to 10,004 feet, and others are of nearly equal altitude. The whole of the Wahsatch region—the range of that name as well as the parallel ranges and spurs on the east—is one of difficult and complicated topography. It forms the connexion between the north and south divisions of the Rocky Mountains, and connects by spurs and irregular lines of elevation with the Wind River range, the range of the Tetons, and the Snake River Mountains.

Although Bear river runs for more than a hundred miles in a northerly direction before crossing the Wahsatch range, the Weber river, which rises in the Uintah range, within three or four miles of the head of Bear river, runs with a pretty direct north-westerly course across that range, breaking through it where it is from 8000 to 9000 feet in elevation, and affording an easy route for the railroad from the Bridger and Green river basin to Salt Lake.

Northern division of Rocky Mts.(B) The northern division of the Rocky Mountains has been much less fully explored than the southern. It also is made up of a large number of ranges, having a general though by no means uniform north-west south-east trend. As a whole, this division is lower and less impressive from the grandeur of the masses than the southern; and as we advance north-westerly we find more monotony in the scenery, more uniformity in the height of the ranges, and an almost entire absence of dominating peaks. Striking exceptions to this condition are offered by the Wind River range and the Yellow stone geyser region.

Northern Pacific Railroad.The Northern Pacific Railroad, by which access is had from the east to this portion of the Cordilleras, strikes across the plains from the western end of Lake Superior, directly west to the Missouri, which it crosses in 101° W. long., at Bismarck, near the centre of Dakota. From this crossing it runs in almost a straight line to the Yellowstone river, which it follows for a distance of 340 miles. It then crosses to the Missouri at Gallatin, and follows that stream to near Helena, a distance of about 100 miles. From here the ascent of the main divide of the Rocky Mountains is made by way of Mullan's Pass, the summit being crossed by a tunnel 3850 feet long, at an elevation of 5548 feet. Thence the line follows Hell-Gate river, the Missoula, and Clarke's Fork, to Lake Pend d'Oreilles (2059 feet), which it curves round on the north, and then strikes directly south-west to the junction of Snake river with the main Columbia.

The Great Basin.II. The Great Basin is the name now given to a region embracing an area of about 225,000 square miles, and having no drainage to the sea. Its shape is roughly triangular, the apex of the triangle being near the mouth of the Colorado river, and its base extending in an irregular line, approximately east and west in direction, from near the north-eastern corner of California to a point on the northern slope of the Uintah range, where, as already mentioned, Bear river has its source. The length of the east side of the triangle thus designated is approximately 600 miles. From the northern side or base the drainage is into Snake river; on the south-eastern side rise various branches of the Colorado; and the south-western is very distinctly marked, for the greater part of its length, by the crest of the Sierra Nevada.

Geology of the Great Basin.The Great Basin is an interesting field for the geologist. The most important feature is the entire absence of the marine Cretaceous and Tertiary formations, which play such an important part in the Rocky Mountain division. With the exception of the late freshwater Tertiary of the Humboldt river and some of the areas farther west, and of the post-Pliocene detrital accumulations of the valleys, there is nothing more recent than Jurassic, and very little of this, the most recent really important fossiliferous formation being the Alpine Trias. The stratigraphical relations of the formations, especially with reference to the building-up of these ranges, are mostly simple, as in the Appalachian and Jura ranges, or even simpler still. Some ranges are simple monoclinals, others anticlinals, and others again synclinals, or a combination of two or more of these forms of structure. They are rarely or never closely compressed and only moderately faulted. The striking peculiarities of Appalachian erosion, due in large part to the repetitions of hard and soft strata, are not to be found as important elements in the Great Basin topography. The Basin ranges differ, however, in a marked degree from those of the Appalachian and Jura in the almost constant presence, and sometimes overwhelming importance, of the volcanic masses throughout the whole region. In some instances these formations make up the whole range; or, at least, the whole interior skeleton of older rocks, if such exist, is concealed by them; in other cases the eruptive materials have been poured forth along the base of the uplift, and there form great plateau-like masses; or they have issued from the summit of the range and spread themselves there in sheets, or flowed down the flanks of the central mass. In this respect the Basin ranges maintain a unity with the other portions of the Cordilleran system, throughout which the exhibition of the results of volcanic energy during the later geological periods is everywhere manifested on a scale perhaps unequalled elsewhere.

Northern plateau.III. The Northern or Columbian plateau embraces the region enclosed between the northern extension of the Rocky Mountains on the east and the Cascade range on the west. It is the basin of the Columbia river, which drains it by means of two principal branches, one of which retains the name Columbia to its source beyond the boundary of the United States, while the other, originally named the Lewis, is now almost universally known as the Snake river. The Columbia itself forks near the boundary line, the main river coming down from the north, and being joined by Clarke's Fork from the south-east. The Columbia and the Snake, after uniting, flow westward for about 100 miles, before breaking through the Cascade range. This area is the portion of the United States of which we have the least topographical knowledge, and therefore only its more striking features can be indicated. The north-westerly trend of the northern division of the Rocky Mountains reduces the width of the Cordilleran system as we go north, since the Cascade range remains unchanged in its direction from the southern line of Oregon to the northern boundary of the country. The area between the two systems is more or less completely filled with mountains of which little is definitely known. Of these there are two principal groups, the Blue and Salmon River ranges, of which the former lies in the angle made by the Snake in its northerly course before reaching the Columbia, while the latter forms an intricate mass, extending from the westernmost ridge of the Rocky Mountain system westward towards the Snake.

Columbia river.The Columbia river rises only 100 miles north of the boundary line, but runs nearly 200 miles farther in a north-westerly direction before turning to go south again in a course nearly parallel to that it had before. The Okanagan joins it about 70 miles south of the boundary line and from here the course of the Columbia is southerly, parallel with the Cascade range for about 160 miles to the Great Bend, when the river takes a nearly westerly direction, which it keeps until, after having passed through the range, it reaches the Pacific. All the region lying north and west of the river and between that and the Cascade range is mountainous. The topography is veiy irregular, but there is a general tendency to a north and south trend, which is still more marked in the region between the Columbia and Clarke's Fork. Here the ranges on each side of the north-flowing Colville rise to from 5000 to 7000 feet in height.

Colorado plateau region.IV. Enclosed between the ranges of the Rocky Mountains on the east and the Sierra Nevada on the west there are—as has been seen—numerous high plateau-like districts, the beds of old freshwater lakes, some of which were of large dimensions. The strata deposited at the bottom of these lakes have been cut into by erosive agencies in numerous places, so that the geological structure stands fully revealed, while the wealth of organic remains which they contain has made these old lake-beds wonderfully attractive to palæontologists. The drainage, desiccation, and subsequent erosion of these areas have given rise to a remarkable type of scenery, to which the name of Mauvaises Terres was applied by the fur-hunters. These Bad Lands, which lie south and south-west of the Missouri and along its tributaries coming in from that direction, may be considered as the precursors and representatives of lands which, from the agricultural and business point of view, are bad enough, but which to the geologist and lover of the picturesque are in the highest degree good, and which are perhaps, on the whole, more striking than anything which the continent elsewhere exhibits. The essential features of their unique and striking type of scenery (of which the Grand Cañon of the Colorado is the grandest and most complete example) are these:—a heavy mass of stratified materials, several thousand feet in thickness, and covering many thousand square miles, has been cut into and eroded away, so as to give rise to a labyrinthine series of gorges, or “cañons,” having a depth of from 1000 to 5000 feet, the walls of which are almost always extremely precipitous and in places perpendicular, and are by no means flat surfaces, but are worn and sculptured into forms almost always peculiar and striking, and often fantastic in the highest degree. And to a variety and complexity of form which seem to find a parallel nowhere on the earth is added the attraction of colour, the various groups of strata forming the cañon walls presenting a gay adornment of tints of red, yellow, purple, brown, and grey, the depth and brilliancy of which surpass belief.

The region in which these wonderfully picturesque forms of landscape occur lies to the south and east of the Great Basin, between Great Salt Lake and the Colorado, to the west of the Green river branch of that river, extending west, with a gradual disappearance of its characteristic features, to near the border line of California. The Uintah Mountains may with convenience, although somewhat arbitrarily, be taken as the northern limit. But, in point of fact, the characteristic type of scenery begins to be developed in Colorado, where the Book or Roan plateau, which rises to a height of 9000 feet, is deeply cut into by the White river in the north and the Grand river in the south.

The peculiar interest of the topography of this region is due in part to the manner in which great blocks of strata have been raised or depressed between long faults, which have given rise to differences of level amounting to thousands of feet, and in part to the extraordinary amount of erosion which the region has undergone, first over its whole surface where not protected by overlying masses of harder volcanic material, and, later, in the channels of the streams, which channels have been gradually growing narrower with the lapse of time, the streams diminishing in volume, until during the present epoch they have either shrunk to nothing or have become absolutely insignificant in comparison with what they were in later Tertiary times. In fact, we have in this region the best possible illustration of the progress and effect of that stupendous desiccation of the climate which has long been manifesting itself all over the world, and of which the results may easily be traced far back in geological history. The contrast between the plateau region south and south-east of the Great Basin and that lying to the north—between the region of the Colorado and that of the Columbia—is a most striking one. In the north the volcanic outflows have filled the depressions in the corrugated and folded strata, covering over the whole of the lower portions of the region, from which the older mountain ranges project like islands from the great congealed sea of lava. The rivers could not subsequently cut very deep into these overflows, because the material is so hard and the general level of the region so low. In the Colorado region, on the other hand, the strata have not been crumpled, folded, and metamorphosed, but raised en masse to a high elevation, and not hardened so as effectually to resist erosion; and, possessing just enough variety of lithological character to prevent uniformity of wearing away and give complexity to the resulting forms, they have, under the simple influence of eroding agencies, assumed the wonderful condition in which we now behold them. Here, too, volcanic agencies have been active; but the molten material has been poured out from orifices at a great elevation, and has built up cones, some of which are of nearly as grand dimensions as the mightiest of the Sierra Nevada and Cascade range; but the valleys and lower regions have not been filled up by them, nor have there been in the southern plateau region any such enormous overflows as those which characterize the northern volcanic district.

Geology of the Mississippi Valley.

Mississippi valley.The area enclosed between the Appalachians and the Cordilleras, extending to upwards of 1,500,000 square miles, the drainage basin of the Great Lakes and St Lawrence on the north and of the grand Missouri-Mississippi river-system on the south, cannot here be discussed in detail from the topographical point of view. The general features do not present the diversity seen in the regions already considered. All that can be done here is to indicate the salient points of the geology.

Political and Natural Subdivisions.

As politically organized at present the area included within the limits of the United States is divided into forty-nine subdivisions, including Alaska. There are thirty-eight States, eight Territories, and three subdivisions, neither States nor Territories, each of which stands in a peculiar relation to the general Government namely, the District of Columbia, the Indian Territory, and Alaska.

In the following table of the States and Territories the names are followed by their customary abbreviations. The dates are those of admission into the Union as States; in the case of the thirteen original States (printed in small capitals), they are the dates when those States ratified the constitution. The names of the Territories are printed in italics.

Table I.—Area and Population of the States and Territories.[5]

 States and Territories. Date of Admission. Area in  Square Miles. Population,  June 1, 1870. Population,  June 1, 1880. Increase  per Cent. Population  per Square  Mile (1880). Alabama (Ala.) 1819 52,250 996,992 1,262,505 26.6 24.5 Arizona (Ari.) 113,020 9,658 40,440 318.7 0.4 Arkansas (Ark.) 1836 53,850 484,471 802,525 65.6 15.1 California (Cal.) 1850 158,360 560,247 864,694 54.3 5.5 Colorado (Col.) 1875 103,925 39,864 194,327 387.4 1.9 .mw-parser-output .nowrap,.mw-parser-output .nowrap a:before,.mw-parser-output .nowrap .selflink:before{white-space:nowrap} Connecticut (Conn. or Ct.) 1788 4,990 537,454 622,700 15.8 128.5 Dakota (Dak.) 149,100 14,181 135,177 853.2 0.9 Delaware (Del.) 1787 2,050 125,015 146,608 17.2 74.8 District of Columbia (D. C.) 70 131,700 177,624 34.8 2960.4 Florida (Fla.) 1845 58,680 187,748 269,493 43.5 5.0 Georgia (Ga.) 1788 59,475 1,184,109 1,542,180 30.2 26.1 Idaho (Id.) 84,800 14,999 32,610 117.4 0.4 Illinois (Ill.) 1818 56,650 2,539,891 3,077,871 21.1 55.0 Indiana (Ind.) 1816 36,350 1,680,637 1,978,301 17.7 55.1 Indian Territory 64,690 ... ... ... ... Iowa (Ia.) 1846 56,025 1,194,020 1,624,615 36.0 29.3 Kansas (Kan.) 1861 82,080 364,399 996,096 173.3 12.2 Kentucky (Ky.) 1791 40,400 1,321,011 1,648,690 24.8 41.2 Louisiana (La.) 1812 48,720 726,915 939,946 29.3 20.7 Maine (Me.) 1820 33,040 626,9l5 648,936 3.5 21.7 Maryland (Md.) 1788 12,210 789,894 934,943 19.7 94.8 Massachusetts (Mass.) 1788 8,315 1,457,351 1,783,085 22.3 221.8 Michigan (Mich.) 1837 58,915 1,184,059 1,636,937 38.2 28.5 Minnesota (Minn.) 1858 83,365 439,706 780,773 77.5 9.8 Mississippi (Miss.) 1817 46,810 827,922 1,131,597 36.6 24.4 Missouri (Mo.) 1821 69,415 1,721,295 2,168,380 25.9 31.5 Montana (Mon.) 146,080 20,595 39,159 90.1 0.3 Nebraska (Neb.) 1867 76,855 122,993 452,402 267.8 5.9 Nevada (Nev.) 1864 110,700 42,491 62,266 46.5 0.6 New Hampshire (N. H.) 1788 9,305 318,300 346,991 9.0 38.5 New Jersey (N. J.) 1787 7,815 906,096 1,131,116 24.8 151.7 New Mexico (N. M.) 122,580 91,874 119,565 30.1 1.0 New York (N. Y.) 1788 49,170 4,382,759 5,082,871 15.9 106.7 North Carolina (N. C.) 1789 52,250 1,071,361 1,399,750 30.6 28.8 Ohio (O.) 1802 41,060 2,665,260 3,198,062 19.9 78.5 Oregon (Or.) 1859 96,030 90,923 174,768 92.2 1.8 Pennsylvania (Pa.) 1787 45,215 3,521,951 4,282,891 21.6 95.2 Rhode Island (R. I.) 1790 1,250 217,353 276,531 27.2 254.9 South Carolina (S. C.) 1788 30,570 705,606 995,577 41.0 33.0 Tennessee (Tenn.) 1796 42,050 1,258,520 1,542,359 22.5 36.9 Texas (Tex.) 1845 265,780 818,579 1,591,749 94.4 6.1 Utah (Ut.) 84,970 86,786 143,963 65.8 1.7 Vermont (Vt.) 1791 9,565 330,551 332,286 0.5 36.4 Virginia (Va.) 1788 42,450 1,225,163 1,512,565 23.4 37.7 Washington (Wash.) 69,180 23,955 75,116 213.5 1.1 West Virginia (W. Va.) 1862 24,780 442,014 618,457 39.9 25.1 Wisconsin (Wis.) 1847 56,040 1,054,670 1,315,497 24.7 24.2 Wyoming (Wy.) 97,890 9,118 20,789 127.9 0.2 ⁠United States 3,019,140 38,558,371 50,155,783 30.08 17.29 Alaska 531,409 33,426

As long as the population of the country was limited to the Atlantic and Gulf coasts there was no difficulty in classifying the divisions geographically; the Northern, Middle, Southern Atlantic, and Gulf States constituted a natural grouping. The almost unknown and at that time not easily accessible region beyond them to the west was known as “the West,” and by this term until more than a quarter of the present century had elapsed the valley of the Mississippi and its tributaries on the east was designated. It was not until about the middle of the century that a still farther “West” began to be taken into consideration.

Early in the history of the country the six north-eastern States (Maine, New Hampshire, Vermont, Massachusetts, Rhode Island, and Connecticut) received the still current designation of New England. And till after the Civil War there was the division of the States into Southern or Northern, according as slavery was or was not permitted.

The suggestion was made by Mr Gannett, geographer of the census of 1880, “to divide the country into three great divisions,—the Atlantic region, the region of the Great Valley, and the Western or Cordilleran region.” The region of the Great Valley he calls the Central region, and this is again subdivided into two parts—the Northern Central and the Southern Central—by the Ohio river and the southern boundary of Missouri and Kansas. The Atlantic division is also subdivided, by a line following the south boundary of Pennsylvania and New Jersey, into the North Atlantic and South Atlantic divisions. The Western or Cordilleran division is limited on the east by the eastern boundaries of Montana, Wyoming, Colorado, and New Mexico. A farther subdivision will be found convenient at times, the Northern Central region being divided into two parts (the North-Eastern and the North-Western) by the Mississippi, and the Southern Central also into two parts (the South-Eastern and South-Western) by the same river. The Western or Cordilleran division may be naturally divided into the Rocky Mountain, the Plateau, and the Pacific Coast regions. Adopting the scheme thus suggested, we have the following grouping of all the States and Territories of the United States (Table II.), the only differences between this scheme and that of Mr Gannett, besides those already indicated, being that the Atlantic States are divided into three subdivisions the Northern, Middle, and Southern, and that West Virginia is placed with the Central States because its drainage is to the Ohio and in its physical characters it is allied to the North-Eastern Central group:[6]

Division. Subdivision. States and Territories. Area. Population.

Gross
Area.
Square
Miles.
Per
cent.
of
Total.
Land
Surface.
Square
Miles.
Number. Per
cent.
of
Total.
No. per
Sq. m.
of Land
Surface.

 Atlantic ${\displaystyle \scriptstyle {\left\{{\begin{matrix}\ \\\\\ \ \end{matrix}}\right.}}$
Northern Atlantic  New England States, New York, New Jersey, Pennsylvania 168,765   5.6 162,065  14,507,407  28.9 89.5
Middle Atlantic  Delaware, Maryland, Virginia, District of Columbia 57,400   1.9 52,005  2,771,740   5.5 53.3
Southern Atlantic  North Carolina, South Carolina, Georgia, Florida 200,975   6.6 191,970  4,207,000   8.4 21.9

427,140  14.1 406,040   21,486,147 42.8 53.8

 Central ${\displaystyle \scriptstyle {\left\{{\begin{matrix}\ \\\\\ \\\ \ \end{matrix}}\right.}}$
North-Eastern Central  West Virginia, Ohio, Indiana, Illinois, Michigan, Wisconsin 273,795   9   269,195  11,825,125  23.5 43.9
North-Western Central   Minnesota, Dakota, Iowa, Nebraska, Kansas, Missouri 516,840  17.1 509,000  6,157,443  12.3 12.1
South-Eastern Central  Kentucky, Tennessee, Alabama, Mississippi 181,510   6   179,630  5,585,151  11.1 31.1
South-Western Central   Arkansas, Louisiana, Texas, Indian Territory (including “unorganized territory”)  438,780  14.5 430,585  3,412,362   6.8  7.9

1,410,925  46.6  1,388,410   26,980,081  53.7 19.3

 Cordilleran ${\displaystyle \scriptstyle {\left\{{\begin{matrix}\ \\\\\ \ \end{matrix}}\right.}}$
Rocky Mountain  Montana, Idaho, Wyoming, Colorado, New Mexico 555,275  18.4 553,280  406,450    .8  0.7
Plateau  Utah, Nevada, Arizona 308,690  10.2 304,850  246,669    .5  0.8
Pacific Coast  Washington Territory, Oregon, California 323,570  10.7 317,420  1,114,578   2.2  3.5

1,187,535 39.3  1,175,550 1,767,697   3.5  1.5

Climate.

Plate IX. Temperature.From the Atlantic seaboard west to near the base of the Rocky Mountains the lines of equal mean temperature have a considerable degree of regularity, running approximately east and west. When, on the other hand, we reach the borders of the Cordilleran region we find the isothermal lines suddenly deflected from their normal course, and in passing across the mountain and plateau belt we find them irregular, often concentric over large areas and through great ranges of temperature, according as the altitude, width, and general trend of each separate range or system of ranges make their influence felt. Hence there are three distinct climatic divisions of the United States:—(1) the eastern region, from the Atlantic to the foot of the high plateaus at the base of the Rocky Mountains; (2) the plateau and mountain region of the Cordilleras; (3) a narrow strip on the Pacific coast, lying west of the Sierra Nevada and the Cascade range. These three divisions are of very unequal size and importance. The first embraces about three-fifths of the entire country, and contains fully nineteen-twentieths of its population; the second is also much larger than the third, containing not much less than a million square miles, but is very sparsely peopled. The third is more densely peopled than the second, but small in area, although its limits are not capable of being accurately defined. These three divisions will here be designated the Eastern, Cordilleran, and Pacific.

Isothermals of the Eastern division;In the Eastern division the passage from one type of climate to another is gradual and uniform, though rapid. The difference in climate between the eastern and western coast of the Atlantic was long ago noticed and commented on. It was George Forster who first controverted the prevailing idea that the New World in general was colder than the Old, and recognized the analogy between the climates of the eastern coasts of the Atlantic and Pacific. Humboldt afterwards investigated the facts and published a tabular statement, which, as enlarged by Hann, is here presented (Table III.):—

Place.  Latitude.   Mean Temperature of the   Difference.
Year.

Year.  Coldest
Month.
Warmest
Month.

° ′ ° ° °  °
Nain, Labrador 57 12  25.16   3.82 51.08
 ${\displaystyle \scriptstyle {\left.{\begin{matrix}\ \\\ \end{matrix}}\right\}\,}}$ 21.6
Aberdeen, Scotland 57 12 46.76 37.22 57.74
St John's, Newfoundland  47 36 40.10 22.46 59.54
 ${\displaystyle \scriptstyle {\left.{\begin{matrix}\ \\\ \end{matrix}}\right\}\,}}$ 13.5
Brest, France 48 24 53.60 42.44 64.76
Halifax, Nova Scotia 44 42 43.34 22.64 64.40
 ${\displaystyle \scriptstyle {\left.{\begin{matrix}\ \\\ \end{matrix}}\right\}\,}}$ 11.7
Bordeaux 44 48 55.04 42.44 69.08
New York 40 42 51.08 28.94 75.56
 ${\displaystyle \scriptstyle {\left.{\begin{matrix}\ \\\ \end{matrix}}\right\}\,}}$ 10.6
Naples 40 48 61.70 48.20 77.18
Norfolk, Virginia 36 50 59.18 40.28 78.62
 ${\displaystyle \scriptstyle {\left.{\begin{matrix}\ \\\ \end{matrix}}\right\}\,}}$ 4.3
San Fernando, Spain 36 30 63.50 52.70 76.10

From the above table it will be seen that the difference between the mean annual temperature of places in high latitudes on the opposite sides of the Atlantic is very large, and that it diminishes as we go south. About lat. 30° the two sides of the Atlantic have nearly the same mean temperature, the difference in climate being very great, but chiefly dependent on differences in the amount of precipitation. Nearly the whole area of the United States is included between the annual isothermals of 44° and 68°—a difference of 24°, the corresponding difference of latitude being about 15°. The average change of temperature is, therefore, 1°.6 for each degree of latitude,—the most rapid change of temperature with the latitude known in any region of anything like equal extent. The causes of the rapid increase of temperature in going south along the Atlantic seaboard are the position of the Gulf of Mexico, the high temperature of its waters, and the increasing predominance of south-westerly winds. From these circumstances the southern portion of the Atlantic coast of the United States is decidedly warmer than the regions corresponding to it in latitude on the west side of the Pacific, while farther north places in the same latitude on the west sides of the two oceans have approximately the same temperature. This similarity of temperature on the corresponding sides of the Atlantic and Pacific is the result of causes now easily understood, the chief being the position of the mass of the land with reference to the direction of the prevailing winds. From the Atlantic coast to the eastern base of the Cordilleras the isothermal curves for the year are nearly parallel, and have a general east and west course, being only interrupted in this regularity and deflected to a certain moderate extent in passing across the Appalachian chain, which nowhere rises high enough to give a chance for permanent accumulation of snow. These curves, of course, are roughly parallel to the coast-line of the Gulf of Mexico, which over a breadth of fourteen degrees of longitude does not vary much from an east and west direction. The region over which a higher mean annual temperature than 68° F. prevails includes nearly the whole of Florida and a narrow strip along the Gulf, which widens rapidly in Texas, where the trend of the coast-line suddenly becomes nearly north and south. The extreme south end of Florida, which just touches 35°, has a mean temperature of over 72°, the isothermal of 76° being nearly on that parallel. The isothermal of 64°, which meets the Atlantic coast near the borders of North and South Carolina, keeps nearly on the parallel of 34° as far west as about 100° W. long., where it is rapidly deflected southward, in conformity with the direction of the other isothermals, by the gradually increasing elevation of the country when the plateau region is encountered. The isothermal of 60° is nearly parallel to that of 64°, except that it manifests the influence of the high southern extremity of the Appalachians, and is in consequence considerably deflected to the south between the meridians of 83° and 87°. The isothermal of 52° is, to the west of the Appalachians, nearly coincident in position with the Ohio river as far as Cincinnati, and thence follows an undulating course, with a nearly westerly general direction, through Indiana, Illinois, northern Missouri, and along the northern boundary of Kansas to the border of Colorado, where it is suddenly deflected and runs with a nearly southerly course for a distance of fully 500 miles along the eastern base of the Rocky Mountains. Those portions of the country which lie between the isothermals of 44° and 52° are New England, with the exception of Maine and the northern part of New Hampshire and Vermont; New York, excluding the extreme north-eastern corner (the Adirondack region); the Appalachian plateau region on the borders of New York and Pennsylvania; nearly all Ohio; two-thirds of Indiana and Illinois; nearly all Michigan and Iowa; southern Wisconsin; south-eastern Minnesota; nearly all Nebraska; and the southern half of Dakota. The isothermal of 40° passes through the centre of Maine, cuts off the extreme northern end of New Hampshire and Vermont, then passes out of the United States, re-entering at the west end of Lake Superior, passing through the centre of Minnesota, making a large loop to the south in eastern Dakota and then trending north-westwardly until it passes beyond the boundary line of the United States in 107° W. long,

of the Cordilleran division;Within the Cordilleran region, or west of the 105th meridian, the position of the isothermals is largely dependent on that of the mountain ranges, which rise high enough profoundly to influence the climate, though it is only at a few points, especially round the summits of the lofty volcanic cones near the Pacific coast, that they reach the region of perpetual snow. This deficiency of lasting accumulations of snow, however, is in very considerable part due to the smallness of the precipitation. Observations of temperature on the higher ranges are extremely deficient. On Mr Schott's temperature chart (Plate IX.) the isothermal of 44°, which, as already mentioned, east of the Cordilleran region nearly coincides with the northern boundary of the country, encloses within a great southerly-reaching loop the whole of the higher portion of the Rocky Mountains, extending as far south as the 34th parallel, or to about the position in latitude of the isothermal of 60° in the eastern division of the country. The crest of the Sierra Nevada, Cascade, and Blue Mountain ranges is also within the curve of 44°. The highest portion of the Rocky Mountains, as far south as 39° N. lat., is laid down as having a mean temperature lower than 36° F. The whole of the Great Basin and the Columbian plateau is indicated as having a considerably higher temperature than the dominating system of ranges which enclose it on the east and west. Considerable bodies of snow remain on the summits of the ranges during a large part of the year, at least as far south as 39° N. lat. In the plateau region of Arizona, Utah, and Nevada the decline of the ranges, the generally lessening elevation of the region, and the facility of access which the topographical conditions allow to the heated air from the south give a high temperature, and the isothermals form irregularly concentric loops extending from the head of the Gulf of California northwards. The isothermal of 52° reaches as far north as Virginia City, in lat. 39°, and that of 72° extends to Fort Mohave, in lat. 35°.

of the Pacific division.In strong contrast with the Eastern division, we find in the region bordering on the Pacific a very marked tendency to a parallelism of the isothermals with the trend of the coast; consequently, a very moderate change in the mean annual temperature may be met with over a large range of latitude. The character of the isothermals here is greatly modified by the position of the two parallel ranges, the Coast Mountains and the Sierra Nevada, which enclose valleys of great extent but of low altitude. In general the temperature of the Pacific coast-belt is much more uniform and higher than that of the Atlantic side of the United States. The isothermal of 60° runs nearly parallel with the coast, and not far distant from it, from the southern line of California north through nearly three degrees of latitude. The isothermal of 52° approaches San Francisco in lat. 37° 48′, and keeps near the coast to as far north as lat. 47°. A higher mean temperature than 48° prevails over the region adjacent to Puget Sound, at the northern boundary of the country, in lat. 49°, while the mean temperature of the northern part of Maine, between the parallels of 45° and 47°, is below 40°. Thus it may be said with truth that near the Pacific coast we have a difference of only 12° in mean temperature in a range of over sixteen degrees of latitude. And if we pass from the immediate vicinity of the coast in lat. 35° into the San Joaquin and Sacramento valleys, we may range over five degrees of latitude and keep in a region of which the mean temperature is not below 60° and nowhere much higher. The causes of this are the proximity of the great area of water from which the prevailing winds blow, the modification which the temperature of this ocean undergoes near the American coast by the Asiatic coast current and the northern or Arctic coast current, and the position of the mountain ranges near the coast. Uniformity of climate along the edges of the land is still further aided by the peculiar nature of the currents along this coast. The influence of the warm Asiatic current—the Kuro-Siwo—is distinctly felt in raising the temperature as far south as the northern border of California, while farther south the cold Arctic current, which apparently emerges from under the warm current, makes its cooling presence felt along the coast nearly or quite as far as the southern boundary of the country.

 VOL. XXIII. UNITED STATES PLATE IX.

 Page 804 W. & A. K. Johnston. ENCYCLOPÆDIA BRITANNICA, NINTH EDITION

Summer isothermals.The isothermals for the summer months (June, July, and August) are much more irregular than those of the year, especially in the Eastern division. The powerful heating influence of the Gulf of Mexico, swept over in summer by southerly winds, makes itself extremely apparent in the summer isothermals, which bend to the north-west in a most remarkable manner, that of 72° reaching as far as the centre of Dakota, or beyond lat. 45°. A mean summer temperature of 80° and upwards prevails over Florida, a considerable portion of the Gulf States, and nearly all Texas. The belt adjacent to the Ohio, extending north as far as the Great Lakes, south along the Appalachian tableland into Tennessee and the north-western corner of Georgia, and west through Iowa, Nebraska, and northern Kansas, lies between the summer isothermals of 68° and 76°. The summer isothermals along the Pacific coast are much less considerably changed in position and character from their mean annual character than they are on the Atlantic side, for reasons which have been already given, while the irregularity and complexity of the summer curves in the Cordilleran region generally would be very distinctly noticed if the data were at hand and could be exhibited with some detail. An extraordinarily high temperature prevails in summer in the southern portion of the Great Basin and in the Arizona plateau region, the isothermal of 88° surrounding with its northerly-reaching loop a large area in the lower valley of the Colorado river and extending north as far Winter isothermals. as lat. 35°. The winter (December, January, February) isothermals in the Eastern division have more of the regularity of the annual curves than have those of the summer. The winter isothermal of 52° coincides very nearly with the mean annual curve of 68°, keeping near and closely parallel to the Gulf of Mexico. The winter isothermal of 32° runs from Cape Cod across Long Island to New York city, and across New Jersey, thence making a large loop to the south so as to surround the Appalachians, and, after ascending northerly again on the west side of that range to near the Ohio, passing through Indiana, Illinois, Missouri, and Kansas, thence descending in a south-westerly direction and sweeping around the Rocky Mountains, and through the centre of the Great Basin in a very irregular course. On the Pacific coast the form of the winter curves closely resembles that of the yearly isothermals. The winter curve of 52° very closely coincides with that of 60° for the year, and the winter curve of 40° runs from near San Francisco, closely parallel to the coast and at a little distance from it, as far as Cape Flattery, or through a distance of over ten degrees of latitude.

Irregular fluctuations of temperature.The irregular, non-periodic fluctuations of the temperature are of great interest, and without knowing what these are one would form a very false idea of the real character of the climate. It does not appear that these fluctuations greatly affect the general salubrity of the country, but they have a marked effect on the character of the vegetation, as well as on the methods of cultivation. The occasional occurrence of very low temperatures in low southerly latitudes where the mean winter temperature is quite high is one of the most striking phenomena in the climate. Savannah, as Hann remarks, has a mean winter temperature the same as that of London and Cadiz, although this latter city lies 4½° farther north. But the vegetation of the two regions is essentially different, because frosts do not occur in that part of the Spanish peninsula. Orange trees are liable to become entirely frozen everywhere in the United States except in southern Florida; this is not the case in Spain. The cotton plant is a perennial in the south of Spain, while, on the other hand, the stem and branches are killed every year by frost in the United States, so that the fields have to be annually replanted.

The following table (IV.), from data arranged by Hann, gives an idea of the range of temperature in various parts of the country:—

 Latitude. Mean Monthly Range. Mean YearlyExtremes. Winter. Summer. ° ′ ° ° ° ° Fort Sully 44 39 103.5 90.1 −25.2 108.7 Fort Snelling 44 53 93.2 70.3 −25.2 93.0 St Louis 38 37 90.3 73.2 − 5.3 99.7 New York 40 42 79.0 71.2 0.7 93.0 Macon 33 46 80.2 59.7 18.3 97.2 Charleston 32 45 73.4 52.3 24.1 92.7 New Orleans 30  0 78.6 55.6 23.2 96.2

The region of lowest winter temperature is that along the eastern border of the Rocky Mountains in the northern portion of the country, where the temperature not unfrequently sinks so low as to freeze mercury. The lowest temperatures observed in this region, as given by Schott, are—at Fort Sanders, in Wyoming, −50°; Fort Ellis, Montana, −53°. A temperature low enough to freeze mercury is occasionally observed in Wisconsin and Michigan, and on the borders of Canada and New York. The hottest region is that along the lower portion of the Colorado and Gila rivers in Arizona and the adjacent part of California.

Cold waves.An excellent illustrative example of the suddenness and severity of the “cold waves” which occasionally pass over the country is afforded by the facts gathered by the Signal Service in regard to an occurrence of this kind in January 1886.

The barometer was high from the Rocky Mountain region to the Pacific coast on the 2d, and from that date to the 5th a slow north-easterly movement of this high area was observed; after the 5th there was an apparent increase of this high area from the region of the Saskatchewan valley and Manitoba. On the afternoon of the 6th the observers in Wyoming, Colorado, Nebraska, Kansas, and Missouri were warned of the approach of a “cold wave,” accompanied by a “norther,” and of a probable fall of temperature of 20° to 25° in the next twenty-four hours. The centre of greatest barometric pressure remained north of Dakota from the 6th to the 12th, but the cold wave had reached the Gulf coast and Florida before that date, causing in many places a lower temperature than has been observed in many years, and in some a lower one than had ever before been known. In Kansas many persons were frozen to death, and the loss of stock was very great; at Dodge City the wind blew with a velocity of 40 miles an hour, the thermometer averaging during the day 10° below zero. In Mahaska county, Iowa, from the 7th to the 11th twenty persons perished with the cold, and much stock was lost. Similar reports came from other parts of Iowa. In Memphis, Tennessee, the thermometer fell to 8° below zero. In Nashville, from the 9th to the 10th, the cold was the severest on record. In New Orleans the cold wave struck the city at 3 A.M. on the 8th, and the thermometer stood at 15°.2 on the morning of the 9th. At Indianola, Texas, the coldest weather experienced for several years occurred from the 8th to the 13th; on the 12th snow fell to the depth of 3 inches. At Galveston the cold was the greatest ever known, the mercury falling to 11°, being a fall of 54° in less than eighteen hours. A heavy snowstorm set in on the morning of the 12th, covering the ground to the depth of 6 inches, and causing much loss and suffering. At Mobile, Alabama, the minimum on the morning of the 9th was 11°, and at Montgomery, 5°.4. In Florida the cold was very severe; ponds were frozen over, and much fruit frozen on the trees. At Atlanta, Georgia, the mercury fell to 2°.4 below zero. At Savannah it stood at 12°, the lowest ever recorded at that place. At Charleston, S.C., it stood at 10°.5; ice 3 inches thick formed on the ponds. On the morning of the 11th, the curve uniting points of which the temperature was zero ran from Dakota south nearly to Arkansas, thence across to the Atlantic, passing south of Knoxville, and up the coast to Nova Scotia. On the St Lawrence and beyond it to the north-west, the mercury stood at from 10° to 30° below zero. This cold wave was remarkable, not only for its severity, but because it extended so far to the south and caused so much damage. The whole country east of the Rocky Mountains was brought under its influence. Of the rapidity of its progress an idea can be formed from the statement that the first warning was issued from the Signal Office at 12h. 2m., January 7th, for the extreme north-west, and that for New England just two days later. This area of high barometer moved east ward, after the 12th, to the Atlantic coast, following the coast-line, passing over Nova Scotia, and disappearing to the eastward on the 16th.

It appears from Prof. Loomis's working over of the records of the Signal Service that throughout the greater part of the United States there is occasionally observed a difference of as much as 40° between the maximum and minimum of the same day, and that there are a few places where such changes are remarkably frequent. These places seem to be all west of the 95th meridian, and at or near the base of the Rocky Mountains. Thus, in 1874 there were thirty-eight stations at which a difference of 40° on the same day between the maximum and minimum temperature was observed. At Colorado Springs (5935 feet) this happened fifty-six times, at Denver (5135 feet) forty-five times, and at Cheyenne thirty-three times; at seventeen stations it happened only once. At Denver, 15th January 1875, the thermometer fell 48° in one hour; and an observer “who is pronounced perfectly reliable” reported a fall in temperature at that place of 36° in five minutes. These changes of temperature felt at Denver were the concomitants of a considerable storm, which came from the north-west, and whose centre passed about 250 miles east of that place.

Hot waves.The occasional occurrence of “hot waves” which sweep over large areas of country, raising the temperature much above its normal height, is one of the most striking and most disagreeable features of the climate of the country, and especially of its northern and north-eastern portions. There is rarely a year in which one or more of these abnormal occurrences are not observed; and, although they do not usually last more than two or three days, they are sometimes prolonged for a month or more, in a succession of heated periods with little or no interval between them. Thus, for example, in July 1885 the thermometer at West Las Animas, Colorado, rose on the 15th to 105°.2; at Albany, N.Y., on the 17th, to 96°.6; at New London, Conn., on the 18th, to 92°.4; in New York city, on the 21st, to 95°.9; in Baltimore, Md., on the 20th and 21st, to 98°.3 and 98°.7; at Dubuque, Iowa, on the 20th, outdoor work was suspended on account of the intense heat. Again, a little later, in Dayton, Washington Territory, on the 28th of the same month, the temperature rose to 102°.6; at Milwaukee, Wis., on the 28th, to 92°.8; at Fort Sully, Dakota, on the 29th, to 104°.5; at Yankton, Dakota, on the 30th, to 100°.7; at Dubuque, Iowa, on the 30th, to 97°.1; at Des Moines, Iowa, on the 30th, to 100°.1. All through the country many cases of sunstroke occurred, eighteen fatal cases having been recorded in Baltimore during the week ending with the 25th.

Winds.The prevailing winds, as in other regions lying in the latitude of the return trades, are westerly. The extreme southern part of the country is just on the border line where the influence of the causes by which the trade-winds are originated cease to be felt. In the autumn, however, in the southern Atlantic States there is some approach to the conditions of the trade-wind region. At that season the winds in Florida and along the northern edge of the Gulf are decidedly north-easterly as far as 33° N. lat. Farther south the Florida Keys and the northern Bahamas belong, to a certain extent, to the trade-wind region.

Along the whole extent of the Atlantic coast region westerly winds predominate during the entire year, but they are chiefly south-westerly in summer and north-westerly in winter. In the following table (V.) the direction of the summer and winter winds is given in percentages of the total amount, for the districts named:—

 Summer. Winter. N. N.E. E. S.E. S. S.W. W. N.W. N. N.E. E. S.E. S. S.W. W. N.W. New England 5 10 8 10 12 24 14 16 9 11 4 7 7 14 15 33 Middle Atlantic States—New York to North-east Virginia 8 10 6 11 14 19 16 15 9 12 5 6 7 14 19 28 South Atlantic States—South-east Virginia to Georgia 7 12 8 12 17 26 11 8 13 13 7 6 11 18 14 17

In the region between the Mississippi and the Appalachians, southward as far as the Cumberland range and north to Lakes Michigan and Huron, south-westerly and westerly winds prevail during both summer and winter. There is an extensive region in the south-west of the United States, embracing an area equal to about one-third of the whole country, in which the winds of summer are chiefly southerly, varying between south-east and south-west, while in the winter they are mostly north and north-west. This region extends from the extreme south-east of California, through Arizona, New Mexico, southern Utah, Texas, Arkansas, eastern Colorado, eastern Wyoming, Kansas, and Nebraska, to Missouri. Farther north, in Wisconsin, Minnesota, and northern Michigan, south winds prevail in the summer, but in winter there is no such marked predominance of northerly and north-westerly winds as in the region to the south-west. The influence of Lake Superior is clearly indicated in northern Wisconsin, where the prevailing winds in summer are from the lake and in winter from the land. On the Pacific coast the winds have a decidedly westerly character; but in the winter this preponderance is much less marked than in summer. On the coast of Washington Territory south-east is the prevailing direction, these winds being probably the south-west winds of the Pacific coast deflected by the mountains which lie close upon the ocean. In the interior of Washington Territory south-west is the prevailing direction in both summer and winter. On the California coast the winds are very strong and steady from the north-west in the summer, but more to the south-west in winter. In summer the intensely heated plateau to the east draws the air from the Pacific, which blows with violence through every depression in the coast ranges towards the heated land-mass. There is no “wind-gap” in the Coast ranges from the Columbia river to Santa Barbara so deeply and widely cut as that of the Golden Gate at San Francisco. At this point the cool winds from the sea find entrance to the Great Valley of the Sacramento and San Joaquin, and the mass of air thus set in motion spreads itself out fan-like after passing through the Gate, so that the prevailing winds in those valleys are in summer always from the Bay of San Francisco towards the mountains. The hotter the weather in the interior the more violent is the wind at San Francisco. But this condition is limited to the daytime. At night the rapid cooling of the higher plateau checks or stops altogether the indraught of air, and an almost entire calm prevails at San Francisco, while the cool air flows in a gentle breeze down the slopes of the mountains, in a reverse direction from that which it had during the daytime. In the winter the westerly direction of the winds in this region is still greatly predominant, but the prevailing westerly current of air is not intensified in its motion as it is during the summer. Over the plateau and mountain region included between the Sierra Nevada and the Rocky Mountains the surface winds are irregular, being governed by the topography of the country; but the upper currents are, in general, from the west. In the southern part of this region, in the valley of the Gila and the lower Colorado, there is a large area which is intensely heated in summer, and towards which the winds blow from the lower region to the south, and especially from the Gulf of California. Here the predominance of southerly winds in summer is very great; but the mountain ranges to the west have so declined in height in this southern region that westerly winds are nearly or quite as common as northerly ones. Farther east and north-east, as has been seen, the preponderance of northerly winds in winter is very great.

Rainfall.In reference to precipitation the territory of the United States may be divided into two nearly equal portions by the meridian of 100°, the region to the east of that meridian being one of sufficient and pretty regularly distributed rainfall, while that to the west is irregularly and insufficiently supplied, with the exception of a narrow belt on the Pacific coast, over a part of which the precipitation is irregular, but fairly sufficient, while another portion is very abundantly supplied with moisture.

Regions of less than 20 inches of precipitation must be essentially pastoral, or, where the amount falls considerably lower, uninhabitable or even deserts. For regions where the precipitation is between 20 and 25 inches cultivation of the soil may be on the whole possible, but will be liable to serious drawbacks, since the smaller the rainfall the greater the liability to a series of years when it will fall below the mean, with partial or total failure of the crops and consequent suffering. Of course, in regions favourably situated for artificial irrigation much may be accomplished in the way of making up for deficient precipitation. If in the light of these preliminary remarks we consult Mr Schott's rainfall charts of the United States we find that the whole of the Eastern division of the country is well supplied with moisture. The isohyetal of 26 inches, which may be taken as approximately the dividing line between a sufficiently and an insufficiently watered area, crosses the northern boundary to the north-west of Lake Superior, runs south-westerly to the 97th meridian, which it strikes in about the latitude of St Paul (45°), and runs thence very nearly south, with a slight westerly inclination, so that when it reaches the northern border of Texas it has advanced westward as far as the 99th meridian, near which it remains through four degrees of latitude, to the parallel of 31°, when it again advances about four degrees to the westward, and then runs south-easterly to the Gulf of Mexico, near the mouth of the Rio Grande. As thus indicated, the isohyetal line of 26 inches leaves to the east, or in the moister region, a large part of Minnesota, the eastern edge of Nebraska, rather less than half of Kansas, most of the Indian Territory, and about half of Texas. The line of 20 inches crosses the northern boundary of the country at about the 97th meridian, and runs south with moderate undulations, gaining a little in westing, so that in the centre of Texas, on the 31st parallel, it is in about longitude 102°. Thence its course is south-easterly to the Gulf, in a course nearly parallel to the isohyetal of 26 inches, and at a very short distance from it. The isohyetal curve of 32 inches, or that marking the western limit of abundant precipitation, is in general pretty nearly parallel to that of 26 inches, and not far distant from it, so that in general it may be said that we pass from a region where precipitation is abundant to one where it is decidedly insufficient in traversing a belt of country having an average width in longitude of about three degrees. The only important exception is that towards the north the distance between the isohyetal lines widens rapidly, that of 32 inches having an almost easterly course along the southern shore of Lake Superior and the northern of Huron. Moreover, there is in the lines of 26 and 32 inches a marked loop running to the south-east, so that almost the whole of Minnesota is brought within the area over which the precipitation ranges between 20 and 32 inches, considerably the larger portion having over 26 inches. The position of the curve of 32 inches is such that a small part of eastern Wisconsin, a portion of eastern Michigan, and a small irregularly shaped belt in New York south of Lake Ontario lie in a region of less than that amount of rainfall.

The regions of largest precipitation are those bordering on the Gulf of Mexico and the Atlantic. Along the Gulf the rainfall between the meridians of 85° and 92° exceeds 56 inches in amount, and the curve of 56 inches extends northward so as to embrace a portion of Arkansas, Tennessee, Georgia, and South Carolina. There is no part of the Atlantic coast, except the extreme end of Florida, where the precipitation is as large as 56 inches. At various points the average is above 50, as in eastern North Carolina, the line of 44 inches running nearly parallel to the coast, and not far from it, as far south as lat. 37°, when it bends westwardly. The greater part of the Eastern division of the United States thus enjoys a sufficient but not over-abundant amount of precipitation, namely, that coming within the limits of 32 and 44 inches. Small areas in several of the States, however, have somewhat over 44 inches of rainfall. In the region of sufficient and in places abundant rainfall thus designated there is, on the whole, no such thing as a clearly-defined rainy season. Along the Atlantic sea-coast from Portland to Washington, through the Hudson river valley, Vermont, northern and western New York, in the Ohio valley from western Pennsylvania to Missouri, south to Arkansas and down the Mississippi to its mouth, the rainfall is pretty uniformly distributed throughout the year. There are, however, local peculiarities in the distribution. Thus, in the Atlantic sea-coast region, as far south as Washington, there are three nearly equal maxima, about the middle of May, August, and December. In the region adjacent to the Hudson river valley through to western New York two maxima are indicated, one early in July and one about the middle of October, while there is one principal minimum, early in February. In the Ohio river valley, west to Missouri, there is one principal maximum and one principal minimum, the former early in June, the latter early in February. In the lower Mississippi valley and in that of the Red River there is one principal maximum and one principal minimum, the former early in December, the latter about the middle of October; there is also a secondary maximum in July, and a secondary minimum in June. In the Mississippi delta and along the Gulf coast eastward in Alabama and Mississippi there are two maxima, the principal one about the end of July, the secondary one early in December, while there are a principal minimum early in October and a secondary one towards the end of April. Along the upper Mississippi, in central Minnesota and part of Wisconsin, there is a decided tendency to a condition of summer precipitation and winter drought; there are two maxima, a principal one about the end of June and a secondary one about the middle of September, and a principal minimum about the beginning of February. This is a similar condition of precipitation to that prevailing in the Hudson river valley and westward, except that in the upper Mississippi region the range is much larger. Again, on the Atlantic coast from Virginia south to Florida there is also a strongly-marked prevalence of summer rains, there being one maximum of very large range late in July or early in August, with two small adjacent minima about the middle of April and late in October. There are also subordinate maxima in March and December.

On the Pacific coast the increase in the amount of precipitation as we go northward is a very marked feature of the climate. Thus at San Diego the mean of the series from 1850 to 1874 is given at 9.31 inches; that of San Francisco, for nearly the same years, at 21.49; that of Astoria at 77.61. Along the coast of California, as well as in the interior of that State in the valley and on the western slope of the Sierra Nevada, there is an almost entire absence of rain during the summer months, and a strongly marked maximum in December. Farther north, with the great increase in the total annual amount of precipitation already noted, there is also an increase in the rainfall of the summer, which amounts in the extreme north-western corner of Washington Territory to 10 or 12 inches during the three summer months. A large portion of the precipitation in the higher region of the Sierra Nevada is in the form of snow, of which the amount in different years appears to be very variable. Indeed the same thing may be said of precipitation in general on the coast of California. The largest amount of rainfall at San Francisco during the years 1851 to 1874 is given by Mr Schott as 36.02 inches, the smallest 11.73. All through the Cordilleras, from the summit of the Sierra Nevada east to the Rocky Mountains, the statistics of the precipitation are meagre, and have been very irregularly taken. The amount in general is quite small. No doubt the precipitation on the higher portions of the Cordilleran mountain ranges is considerably higher than it is in the valleys, as is indicated by the records kept by the Signal Service at the station on the summit of Pike's Peak (14,134 feet), the average for 1874-80 being 31.65 inches. In the Cordilleran region generally the fact that the precipitation is larger on the mountain ranges than it is in the valleys, and that it is chiefly in the form of snow, is a matter of great importance. When the ranges are lofty and wide enough to collect and store away a large supply of snow, this by ice melting furnishes water enough to irrigate the slopes and valleys, so that they can be cultivated; when, on the other hand, the ridges are low, they, as well as the valleys at their bases, are absolutely sterile.

Storms.Those abnormal disturbances of the atmosphere which are accompanied by rain and wind may be classed under two heads,—ordinary storms, and those of destructive violence, or tornadoes. The former extend over wide areas, and are ordinarily attended by no evil results; the latter are limited to comparatively narrow belts, and are often very destructive. The ordinary storms of the United States begin with the formation of areas of low barometer, which are first heard of in the far west or south-west, and move towards the east or north-east with a velocity averaging for the entire year, as shown by Loomis's investigation of the Signal Service Records for the years 1872-84, 28.4 miles per hour, the velocity being greatest in February and least in August, the former velocity 50 per cent. greater than the latter, and the velocity varying also very greatly for the same month in different years, the average velocity for the entire year being about two-thirds greater than it is in Europe. The direction in which these storm centres advance in the remote western stations—as, for instance, Bismarck, long. 100° 38′; Fort Sully, long. 100° 36′; Breckenridge, long. 96° 17′—is towards a point considerably south of east, but at the more eastern stations it is a little north of east. In general, probably about half the storms of the country advance from the extreme north-west in great curved lines beginning with a south-easterly direction, and passing out of the country in a direction a little north of east, or, in general, following a track nearly parallel in position to the Great Lakes and the St Lawrence. The remainder of the storms of the Atlantic coast region begin in the south-west and travel north-east, or begin in the south and follow the coast-line pretty closely. In general the area of rainfall attendant on the advance of the centre of low barometer is in advance of the progress of that centre nearly in the direction of its average progress. The diameter of the rain area is variable, often much over 1000 miles. In the case of the great rain storms happening between the years 1873 and 1877, as investigated by Loomis, there were found to be, in many cases, quite a large number of independent rain centres prevailing simultaneously within the general rain area. In one case there were as many as eight of these, and there were only nine cases in which there was not more than one area in which the rainfall exceeded half an inch. The average distance of the principal rain centres from the centre of low pressure was about 400 miles.

Tornadoes.The occurrence of tornadoes in the United States is a matter of importance on account of their frequency and their destructiveness, and much has been published in regard to them. A large amount of information will be found in a publication of the Signal Service, prepared by Mr J. P. Finley, and issued in 1882. These storms are not limited to any one month or season; but they are most frequent in summer, especially in the months of June, April, July, and May, and least so in the months of December and January. Of 600 tabulated by Mr Finley, occurring from 1794 to 1881, 112 were in June, 97 in April, 90 in July, 81 in May, and only 9 in December and 7 in January. They are most frequent in the afternoon, between noon and six o'clock; the hour in which the greatest number occurred was that from 5 to 6 P.M. The course of more than half of the 600 (310) was from south-west to north-east, and only 38 moved in the opposite direction. Only 46 had a course directed from the eastern side of the meridian towards the western. The width of the path of destruction varied from 40 to 10,000 feet, the average being 1085 feet. The velocity of progression of the storm-cloud, in 130 cases in which this item is given, varied from 12 to 60 miles per hour, the average being 30 miles. The time consumed by the tornado in passing any given point varied from 10 seconds to 30 minutes, the average of 50 occurrences being 6.52 minutes. The velocity of the wind within the cloud vortex was variously estimated at from 70 to 800 miles an hour. The whirling motion of the cloud was invariably from right to left. Of 600 tornadoes investigated, 134 were reported as being “unusually destructive.” Of these 64 occurred within the States of Kansas, Illinois, Iowa, and Missouri, and this region, lying adjacent to the Mississippi river, seems to be that in which the conditions are most favourable to the development of these phenomena. There are also two areas one in Georgia and one in New York where tornadoes are more frequent than they are elsewhere in the eastern States. Of the destructiveness of these tornadoes some idea may be formed from the statement that in many of them buildings and everything else projecting from the surface are levelled to the ground, fragments of the materials thus uptorn being carried often to great distances. In the tornado of April 18, 1880, the effects of which were felt along a path more than a hundred miles in length through Illinois and Missouri, in one town over which it passed, 65 persons were killed, over 200 wounded, and more than 200 buildings were demolished. The loss of property in two counties of Missouri was over a million dollars.

The series of destructive storms which took place on the 19th of February 1884 is probably the most remarkable occurrence of this kind which has taken place in the United States since the country was settled by the whites. The loss of property was not less than $3,000,000 to$4,000,000, while 800 persons lost their lives, and about 2500 were wounded. From 10,000 to 15,000 were rendered homeless, as many as 10,000 buildings having been destroyed. Great quantities of live stock also perished. A central area of barometric depression moved between 7 A.M. of the 18th and 7 A.M. of the 19th from Fort Keogh to the vicinity of Chicago; at the same hour on the 20th it was about 150 miles north-west of Montreal. On the 19th, at 7 A.M., another extremely elongated area of barometric depression had been formed, extending almost north and south across the whole United States, and having its centre near Davenport, Iowa. Towards this centre the winds blew from north and south, the isotherms indicating very great contrasts of temperature between the areas of northerly and southerly winds, this condition of things being an invariable precursor of tornado development. The two centres of barometric disturbance were, as is commonly the case in occurrences of this kind, widely separated. At 3 P.M. of the 19th the centre of the north and south trending barometric depression was near Indianapolis, the contrasts of temperature remaining extreme, and violent winds developing themselves at various points south of Indianapolis, especially along the Ohio river from Cairo to Louisville, in the vicinity of Nashville, and in northern Alabama. At 11 P.M. of the same day the barometric trough had diminished somewhat in intensity, and the entire area of disturbance was passing rapidly off to the north-eastward. Between 3 P.M. and sundown the area devastated was chiefly in eastern Alabama and northern Georgia. Before 11 P.M. the destructive storms in North and South Carolina had reached their maximum violence; those in southern Virginia were most destructive about midnight. The Signal Service charts for the day indicate about thirty distinct areas of violent tornadoes, most of them between the eastern border of Alabama and the southern boundary of Virginia.

Vegetation.

No portion of the United States attains so high a latitude that the forest growth should be necessarily dwarfed by the cold, or disappear altogether. The northern boundary is, however, practically nearly the limit beyond which valuable timber cannot be expected. The portions of the United States where altitude is fatal to the growth of forest vegetation are insignificant as compared with the area of the whole country. The Appalachian ranges—which originally were densely forested from extreme north-east to extreme south-west, and which still continue to be so over a considerable portion of their extent—only rise at a very few points high enough to cause the forests to disappear. This is the case particularly with Mount Washington and the higher adjacent peaks, and with the summits of the most elevated part of the system in North Carolina. The Adirondacks are densely wooded, even almost to the highest summits. In the most elevated mountain-chains making up the Cordilleran system, want of moisture appears to co-operate with elevation in thinning out the forests on their flanks and causing them to disappear entirely on the highest ranges. The timber line on the most elevated peaks of Colorado reaches from 11,000 to about 11,500 feet,—the summits themselves rising from 2000 to 3000 feet higher. The Sierra Nevada is bare of forests in its highest portions. The high region about Mount Whitney is, where not snow-covered, nothing but an entirely bare mass of granite domes and needles. In the central part of the Sierra, in the vicinity of the Yosemite valley, forest vegetation is extremely scanty above 9000 feet, and the upper 3000 feet of the highest peaks is entirely bare of trees. If large areas of the United States are destitute of trees, and other regions but very poorly supplied, the chief cause of this is want of sufficient moisture.

Appalachian forests.In briefly indicating the nature and distribution of the forests of the United States, we may begin with the Appalachian region, which here must be taken as embracing also the country to the west and south-west, including the valleys of the Mississippi and Missouri, as far west as the western boundary of the State of Missouri, or about the 95th meridian, to the east of which lies, coincident with the region of generally abundant and every where sufficient rainfall, that portion of the United States which is almost everywhere densely forested, and the only portion which is so, with the exception of a comparatively narrow strip on the Pacific coast. Included within this densely-forested region of the Appalachian system and Mississippi valley there is quite a large area destitute of continuous forests,—the so-called “prairie region” (see below). The portion of the United States first settled by Europeans was, almost without exception, a densely-forested region, over which the aboriginal inhabitants roamed, without having interfered to any perceptible extent with the natural forest growth of the country. Their numbers were small, and their habitations were, almost without exception, either on or near the shores of the ocean and its bays and indentations, or along the river bottoms, in such places as were naturally grassed and not forested. This densely-forested region extends throughout the whole length of the Atlantic coast from Maine to Florida, west through the region of the Great Lakes to beyond Lake Superior, and to the south-west through Louisiana and for some distance into Texas. It differs from the densely-forested region of the Pacific in that it is essentially a region of deciduous or hardwood forests, while the latter is essentially one of coniferous trees; it differs from the forested region of the Rocky Mountains in that the latter is not only essentially a region of coniferous trees, but one where the forests do not by any means occupy all the area, neither do they approach in density or economic importance those of the eastern division of the country. Again, the forests of the east embrace a great variety of species, which, as a rule, are very much intermingled, and do not, unless quite exceptionally, occupy areas chiefly devoted to one species; while, on the other hand, the forests of the west—including both Rocky Mountain and Pacific coast divisions—exhibit a small number of species, considering the vast area embraced in the region; and these species are, in quite a number of instances, extraordinarily limited in their range, although there are cases in which one or two species have almost exclusive possession of very extensive regions. The eastern forested region, while continuous from north-east to south, south-west, and west, is of course marked by changes in the species corresponding with the changes in temperature between the extreme north-east and the extreme south. These changes, however, are almost without exception gradually made, and we pass almost imperceptibly from a northern to a southern forest. This condition is, in a measure, the consequence of the breadth and high elevation of the Appalachian system in its southern extension, along which elevated belt the northern aspect of the arboreal vegetation is prolonged into a region almost semi-tropical.

The following hardwood trees may be mentioned as being the most prominent and important in the forests of the eastern division of the country. The sugar-maple (Acer saccharinum), called also the hard and rock maple, ranges as far south as northern Alabama, but is of the most economical importance in New England and the region of the Great Lakes. On the southern shore of Lake Superior, in the higher portions of the country, on and near the divide between the waters flowing into the lake and those which descend to the Mississippi, the forest, over large areas, is almost exclusively made up of this species, the “bird's-eye” variety—formerly much prized for cabinet work—being there abundant. The other species of maple of less importance are the soft maple (A. dasycarpum), having a wide range, and attaining its greatest development in the valley of the lower Ohio, and the red maple (A. rubrum), also ranging from New Brunswick westward to the Lake of the Woods and south to Texas, and being largest and most abundant in the central portion of the Mississippi valley. The oaks range over the entire eastern forested region from Maine to Florida, and west nearly as far as arboreal vegetation extends. The number of species is large. The white oak (Q. alba) ranges over nearly the whole forest region of the east, reaching its greatest development along the western portion of the Appalachian belt, and in the valley of the Ohio and its tributaries. The burr oak (Q. macrocarpa) has almost as wide a range as the white oak, extending farther west and north-west than any oak of the Atlantic forests; it forms, with the scarlet oak (Q. coccinea), the principal growth of the “oak-openings” in the prairie region. The red oak (Q. rubra) has also a wide range; it extends farther to the north than any other species. The jack oak or black jack (Q. nigra) is a small tree of little value except for fuel, but widely disseminated in the west and south-west of the eastern forest region, and forming with the post oak (Q. obtusiloba) the growth of the so-called “cross timbers” of Texas. The live oak (Q. virens) is an evergreen tree of considerable value, chiefly developed along the Gulf coast and through western Texas into the mountains of northern Mexico. The chestnut oak (Q. Prinus) ranges through the Appalachian region, from Lake Champlain to northern Alabama, and west to central Kentucky and Tennessee. Its bark is used in preference to that of the other North American oaks in tanning. The ash is represented by several species. The white ash (Fraxinus americana) is of special value, and its range is very extensive, namely, east and west from Nova Scotia to Minnesota, and south-west to the extreme border of Texas. This species has its greatest development in the bottom lands of the lower Ohio valley. Towards the west and south-west it diminishes in size and importance, and is replaced to a considerable extent by the green ash (F. viridis). The range of the red ash (F. pubescens) is nearly as large as that of the white ash, except that it does not extend quite so far to the south-west. Its wood is less valuable. The chestnut (Castanea vesca, var. americana) is an important tree, with a wide range. The American chestnut is smaller and sweeter than the European. The species ranges from southern Maine west to Indiana, and south along the Appalachians to northern Alabama, attaining its greatest development along the flanks of the mountains in North Carolina. The birch is represented in the eastern forest region by several species. The white, canoe, or paper birch (Betula papyracea) reaches a higher latitude than any other tree of the American deciduous forest. It ranges south to the mountainous region of northern Pennsylvania, and west to British Columbia. The yellow or grey birch (B. lutea) is one of the largest and most valuable trees of the New England forest, ranging south along the higher portion of the Appalachians to North Carolina, and west to southern Minnesota. There are in the region several species belonging to the two genera of the Juglandaceæ, Juglans and Carya, which have a wide range, and are of importance both for their wood and for their fruit, and which also are among the most attractive ornaments of the forest. Prominent among these are the hickory (Carya alba), the butternut (Juglans cinerea), the black walnut (J. nigra), and the pecan (G. olivæformis). The pecan does not occur to the north-west of Indiana, has its greatest development in the rich bottom lands of Arkansas, and is the largest and most important tree of western Texas. The butternut occurs in New England, but is by no means an abundant tree in that region; farther west, especially in the valley of the Ohio, it attains its maximum development. The black walnut is hardly known in New England, unless on its extreme western border; but south-westward along the Appalachians and west to the Mississippi it is a tree of great value and importance. It attains its maximum development on the western slope of the southern portion of the Appalachian range and thence to Arkansas. Hardly any other wood is ever used for gunstocks. The American elm (U. americana) has a wide range, extending from southern Newfoundland to Texas and west to central Nebraska. This species is especially the tree of the river bottoms, and specimens occurring isolated in natural meadows often attain great size. The rock or white elm (U. racemosa) is a tree hardly occurring in New England, but largely developed in the region of the Great Lakes, west to north-eastern Iowa, and south to central Kentucky. Its wood is considerably denser than that of U. americana. The beech (Fagus ferruginea) occurs through nearly the whole of the eastern forest region, ranging from Nova Scotia south and south-west to Florida and Texas, and west to Missouri. The linden, lime, bass-wood, or white-wood (Tilia americana) is a tree of wide range, occurring more and more abundantly as we go west from New England through the region south of the Great Lakes into the Ohio valley, and found south along the Appalachians to Georgia. It has its maximum development towards the west and south-west in the rich bottom-lands. The tulip tree (Liriodendron tulipifera), called also yellow poplar and white-wood, is one of the largest and most beautiful trees of the eastern forest region. It is rare in New England, but has its maximum development from New Jersey south along the slopes of the Appalachians to Tennessee and North Carolina, and west in the Ohio valley. The genus Magnolia is represented by several species, two of which are of importance, especially for the great beauty of the tree and its flowers. These—M. glauca and M. grandiflora—like the other species of the magnolia, are pretty closely limited to the Atlantic coast and Gulf region, and the lower portion of the Mississippi valley. M. glauca, which has a variety of names, among which those of sweet bay and white laurel are most common, is found over a small area on Cape Ann in Massachusetts, and in no other place in New England,—its range being from New Jersey southward, chiefly along the coast to Florida, and west to Arkansas and Texas. M. grandiflora, called the big laurel or the bull bay, an evergreen, and one of the finest trees of the region, is pretty closely limited to the southern and south-western coast, ranging from North Carolina south to Tampa Bay, westward to south-western Arkansas, and along the Texas coast to the valley of the Brazos. There are two trees known familiarly as the locust which are of considerable importance. One is the Robinia Pseudacacia, commonly called either simply the locust or the yellow locust; the other is Gleditschia triacanthos, to which the popular names honey locust, acacia, sweet locust, and black locust are given. The former occurs naturally in the Appalachians from Pennsylvania to Georgia, reaching its maximum development on the western slopes of the mountains of West Virginia, but has been introduced and cultivated over the whole region east of the Rocky Mountains, wherever trees can be made to grow. This tree, however, over an extensive portion of the region where it was formerly cultivated has been exterminated by the attacks of the “locust borer” (Cyllene picta). The other locust, the three-thorned acacia, ranges from Pennsylvania, along the western flanks of the Appalachians, south as far as Florida, south-west through northern Alabama and Mississippi to Texas, and west from Pennsylvania through southern Michigan to eastern Kansas. It is the characteristic tree of the “barrens” of middle Kentucky and Tennessee, and attains its maximum development in the lower Ohio bottom-lands. It is widely cultivated throughout the region east of the Appalachians for shade and ornament, and for hedges. There are certain trees and shrubs in the eastern forest region of little or no economical importance, but which, especially when in flower, are highly ornamental. Of these only a few can be mentioned: the mountain ash (Pyrus americana), ranging over nearly the whole region, and much cultivated as an ornamental tree on account of the beauty of its fruit, of dark reddish or scarlet colour, and remaining long upon the branches; the sumach (Rhus glabra), a handsome shrub, from 4 to 10 feet in height and very striking both for foliage and fruit, and a very characteristic feature of the New England landscape, as seen along the borders of the forests and by the sides of country roads; the mountain laurel (Kalmia latifolia), covering extensive areas of half-cleared forests in the hilly regions and very conspicuous at the flowering season, June and July, one of the most beautiful of all the characteristic native American shrubs; the dog-wood or cornel (Cornus alternifolia), a beautiful shrub, rising occasionally to sufficient height to be called a tree, ranging from the St Lawrence to Alabama, and in certain regions, especially in parts of New Jersey and Pennsylvania, very conspicuous at the time of its flowering, the landscape from a distance looking as if it had been snowed upon. The red-bud (Cercis canadensis), a small tree, is a conspicuous feature of the forest in the extreme south-west, especially in southern Arkansas, the Indian Territory, and eastern Texas.

The deciduous trees of the Sierra Nevada and the Cascade range, as well as of the Pacific Coast ranges from California north to the boundary line, are of comparatively little importance. There are several species of oak, but of little value, among them being the Coast live oak (Q. agrifolia), the largest and most generally distributed oak in the south-western part of California; the black oak (Q. Kelloggii), ranging along the coast mountains of Oregon, and the most characteristic hardwood tree of the western slope of the Sierra; and the chestnut oak (Q. densiflora), occurring in the Coast ranges from Oregon to central California. An evergreen tree, very characteristic of the coast ranges of Oregon and California, and very ornamental, is the California laurel (Umbellularia californica), of which the wood is hard and strong, and of a very pleasing light-brown mottled colour. The tree called the madrona or madroño (Arbutus Menziesii) occurs from British Columbia south through the Coast ranges to the Santa Lucia Mountains, and is a very characteristic tree of the region, with its red bark and beautiful glossy foliage. The wood is used in the manufacture of gunpowder, and the bark to some extent in tanning.

In the area included between the two heavily-timbered regions described above, or between the summit of the Sierra Nevada and Cascade range and the western border of the great eastern forest region, the paucity of rainfall corresponds to paucity or almost entire absence of forests over much the larger portion. We may here distinguish, first, the Rocky Mountain region; then the Great Basin and the plateaus north and south of it; then the “Plains,” or the nearly level country lying east of the base of the Rocky Mountains; and, finally, the “Prairie” region, or that portion of the scantily timbered area which for the most part lies enclosed within the eastern forested belt, and where other causes than the absence of moisture have operated to bring about the growth of a peculiar vegetation.

Forests of the Rocky Mountains.The Rocky Mountain belt is not destitute of forests, but these are very irregularly scattered over the surface, and the species are few in number, chiefly belonging to coniferous genera, and of little economic value. The species of conifers have been already mentioned, and their range indicated in speaking of the forests of the Sierra Nevada and Cascade range. Few forests in the Rocky Mountains at all compare in density or in the size of the individual trees with those of the Sierra and the Cascade range. What trees there are usually grow most densely in the moist places at the foot of the ranges, where the streams debouch from them, in the ravines and gorges, and on the lower slopes. By far the most common deciduous tree throughout this region is the aspen (Populus tremuloides), often called cottonwood and sometimes poplar, which most commonly springs up, forming dense thickets, throughout the Rocky Mountains, wherever the coniferous forest has been burned off. It ranges from Newfoundland to Arizona, and is the most widely distributed of North American trees, and highly characteristic of northern and elevated regions. In various portions of the Rocky Mountains there are scattered oaks. The scrub oak (Q. undulata, var. Gambelii) occurs in some quantity on the mountains of southern New Mexico and Arizona, and is also found in Colorado and along the Wahsatch range. The black oak (Q. Emoryi), the white oak (Q. grisea), and a few other species are found here and there in the southern part of the Rocky Mountains, as also in Arizona, and ranging south into Mexico. The most densely forested portions of the Rocky Mountains are the extreme northern in north-western Montana, the north-west corner of Wyoming, the higher part of Colorado, the eastern slope of the range in New Mexico, and the higher portions of Arizona.

Enclosed between the densely-forested Pacific belt and the Forests of the Great Basin. poorly-timbered region of the Rocky Mountains is an extensive area, including the northern and southern plateaus and the Great Basin, which practically is nearly destitute of trees. The coniferous species occurring in the Rocky Mountains are found here and there along the moister tracts of the higher ranges in the Great Basin, especially in its eastern and higher portion, but by far the larger part of the slopes and nearly all the valleys are treeless, being chiefly occupied by the well-known “sage-brush” (Artemisia tridentata), which covers many thousands of square miles, especially in Nevada and Utah. Two trees are, however, very characteristic of the Great Basin, especially of its western portion—the juniper (Juniperus occidentalis) and the piñon or nut pine (Pinus monophylla). These two species, usually much dwarfed in size and scrubby in appearance, are almost the only trees of western and central Nevada, where they occur hidden away in the cañons. Everywhere in a wide sweep adjacent to the mining districts all this vegetation has been completely cleared away.

Vegetation of the Southern Plateau.An order of plants peculiarly American, and characterizing in a most marked manner the hot dry region adjacent to the lower Colorado, is that of the Cactaceæ. The cactus ranges from the extreme north of the plateau region to the extreme south, but its most abundant and striking development takes place in southern Nevada, southern California, and in general the region adjacent to the Mexican boundary line. The so-called prickly pear (Opuntia) is the cactus family which has the widest range, being found from the upper Missouri through the Great Basin to Arizona. It has many species, that which ranges farthest north being O. missouriensis. The genera Mamillaria, Echinocactus, and Cereus are found in various localities in the Great Basin, as well as in southern California, in portions of which region, as well as in lower California and Arizona, there are large areas where various kinds of cactus form almost the exclusive vegetation, often rising to such a height as to be properly called trees; the loftiest of all is the singularly striking Cereus gigantcus. Mingled with them are yuccas (called the “Spanish bayonet”), mezquites (Algarobia glandulosa), and the creosote bush (Larrea mexicana), which are among the most abundant and characteristic plants of this region.

“The Plains.”The vast area extending east from the base of the Rocky Mountains to near the 95th meridian is the district universally known as “the Plains,” and one not at all to be confounded with the “Prairies,” which are almost entirely included within a region of dense forests, and over which the partial absence of trees is due to a cause entirely different from that which has made the plains the home of the grasses and not of trees. The transition from the forested region of the east to the region of the plains is, almost without exception, coincident with the diminution in the precipitation, which as we proceed westward goes on rapidly, and, on the whole, pretty regularly. Thus Dakota, between 97° and 104° W. long., is practically destitute of timber, except in its river bottoms and the small territory between the north and south forks of the Cheyenne, the region of the Black Hills. In Minnesota, which lies east of 97°, only the north-eastern portion, especially that adjacent to Lake Superior, is heavily timbered. The south-western corner of the State, embracing about one-third of its area, and the area west of the 96th meridian, are classed in the census report as having less than two cords of wood to the acre. Nebraska and Kansas, still farther south, are almost destitute of forests. In Nebraska only a narrow strip along the Missouri, near the meridian of 96°, is given as having from one to two cords of wood per acre. The heavy forest growth of the Mississippi basin just reaches the extreme south-eastern corner of Kansas. North of this, and along the eastern border of the State, there is a belt of from thirty to a hundred miles in width in which there is valuable timber on the borders of the streams. West of 97° W. long, the trees are confined to the immediate banks of the large streams, and are small and of little value. West of 99° we find the typical vegetation of the plains, with only a few small stunted willows and cottonwoods scattered at wide intervals along the streams. The yearly isohyetal of 26 inches forms a limit beyond which arboreal vegetation is almost entirely absent, while in going east there is little of value until we reach the belt in which the rainfall is over 32 inches. The same may be said of the Indian Territory and Texas, the bending of the isohyetal curves to the west as we approach the Gulf of Mexico being, however, as might be expected, accompanied by a corresponding extension of the forest belt in that direction. Thus, in Texas, the limit of what may be designated as the well-timbered region lies between 96° and 97°, while the line marking the entire disappearance of the forests may be placed somewhere between 99° and 100°, and pretty closely adjacent to the isohyetal of 26 inches.

Prairie region.The French word “prairie,” a meadow or grassy plain, was employed by Hennepin about 1680, in his excellent description of the prairies of Illinois. The word has become current in the Mississippi valley, and still farther west. In the northern portion of the Rocky Mountains the small grassy areas adjacent to the streams and surrounded by mountains are called “prairies,” while farther south they are known as “parks,” the still smaller areas being frequently denominated “holes.” In general, however, the term prairie is used to designate tracts of land nearly or quite destitute of forests, or over which the trees are, as a general rule, limited to the “bluffs” the more or less precipitous slopes which separate the upland, or prairie proper, from the river bottom—which treeless areas occur in the midst of a well-forested country. Illinois is par excellence the Prairie State, and may be considered the centre of the prairie region, the adjacent States, on all sides, having more or less prairie and also areas of dense forest.

All through the prairie region the precipitation is abundant and pretty equally distributed through the year. The vicinity of Chicago, a typical prairie region, comes within the isohyetal of 44 inches and upwards, and the same is true of a part of the treeless region of Iowa. Large areas in the more southern States—Arkansas, Alabama, Mississippi, and Louisiana—are also prairies, portions of which are entirely destitute of forests, while others have small “clumps” of trees sparsely scattered over their surface. This is in a region of the largest rainfall, that of 56 inches and upwards. The cause of the absence of trees on the prairies is the physical character of the soil, and especially its exceeding fineness, which is prejudicial to the growth of anything but a superficial vegetation, the smallness of the particles of soil being an insuperable barrier to the necessary access of air to the roots of a deeply-rooted vegetation. Wherever in the midst of the extraordinarily fine soil of the prairies coarse or gravelly patches exist, there dense forests occur. The theory that fineness of soil is fatal to tree growth finds its most remarkable support in the fact that in south eastern Russia the limits of the “black earth” and the treeless region are almost exactly identical, and not at all in harmony with the position of the isohyetal lines. The black soil of Russia is an earth of exceeding fineness—so fine indeed that it is with the greatest difficulty that the air can penetrate it so as to oxidize the organic matter which it contains,—the essential cause (in the opinion of the present writer) of its dark colour. The peculiar mode of decay of the organic matter in such fine soils seems analogous to that by which vegetation has been turned into coal or lignite when so buried under detrital material as to greatly impede the access of the air. It is now easy to see why plains are likelier than mountain slopes to be treeless, it being towards the plains that the finer particles of the material which is abraded from the higher regions are being constantly carried. The more distant the region from the mountains and the broader its area the more likely it is that a considerable portion of it will be covered with fine detritus, whether this be of subaerial origin or deposited at the bottom of the sea. The exceedingly fine soil of the typical prairie region consists, in large part, of the residual material left after the removal, by percolation of the rain and other atmospheric agencies, of the calcareous portion of the undisturbed stratified deposits, chiefly of Palæozoic age, which underlie so large a portion of the Mississippi valley. The finer portions of the formations of more recent age in the Gulf States have also, over considerable areas, remained treeless. There are in various parts of the country beds of lakes which have disappeared in consequence of their very slow filling up with fine sediment, and these are not occupied by the forest, although surrounded by the densest arboreal growth.

Economic importance of the forests.The principal use of the forests of the United States is for fuel; and there is no part of the settled portion of the country where the consumption for this purpose is not of some importance. In the middle and northern Atlantic States coal is the chief fuel in the cities and large towns, and the almost exclusive fuel of the larger cities on important lines of communication, but supplemented, to a greater or less extent, by wood. But in the country, on the farms, and in the small towns wood is almost exclusively used, except on the coal-fields or in their immediate vicinity. In the coal-producing States of Illinois and Iowa, where forests are limited to certain areas, and in Nebraska and Kansas, which have coal of inferior quality and where forests are still scarcer than they are in Illinois and Iowa, coal is the dominating fuel. The same is the case in certain parts of Missouri. In all the States south of Virginia and Kentucky wood is almost the exclusive fuel, except in a few of the very largest towns. Wood is also almost exclusively the material of which houses and barns are built over the whole United States, the exceptions being the larger cities (chiefly of brick, with some stone), the business portions of the towns of second rank, and occasional important buildings in towns of the third rank. The other uses of wood are obvious. The following figures given by the census of 1880 relate to the manufacture of sawn lumber:—establishments, 25,708; capital, $181,186,122; average hands employed, 147,956; feet of lumber produced, 18,091,356,000; laths, 1,761,788,000; shingles, 5,555,046,000; staves, 1,248,226,000; sets headings, 146,523,000; feet of bobbin and spool stock, 34,076,000; total value of specified products,$230,685,061; value of other products, $2,682,668; total value,$233,367,729. The consumption of wood for domestic fuel is given in 1880 as amounting to 140,537,439 cords, with an estimated value of $306,950,040, and the total consumption of wood as fuel as 145,778,137 cords, valued at$321,962,373.

Mineral Resources.

Historical sketch of early mining.In 1619 the erection of “works” for smelting the ores of iron was begun at Falling Creek, near Jamestown, Va., and iron appears to have been made in 1620; but the enterprise was stopped by a general massacre of the settlers in that region. In 1643 the business of smelting and manufacturing iron was again begun at Lynn, Mass., where it was successfully carried on, at least up to 1671, furnishing most of the iron used in the colony. From the middle of the 17th century the smelting of this metal began to be of importance in the vicinity of Massachusetts Bay, and by the close of the century there had been a large number of iron-works established in that colony, which, for a century after its settlement, was the chief seat of the iron manufacture in America, the bog ores, taken from the bottoms of the ponds, being chiefly employed. Early in the 18th century the industry began to extend itself over New England, and into New York and New Jersey, the German bloomery or forge being employed for reducing the ore directly to bar iron, and by the middle of that century it had taken a pretty firm hold in the Atlantic States. About 1789 there were fourteen furnaces and thirty-four forges in operation in Pennsylvania. Before the separation of the colonies from the mother country took place the manufacture of iron had been extended through all of them, with the possible exception of Georgia. As early as 1718 iron (both pig and bar) began to be sent to Great Britain, the only country to which the export was permitted, the annual amount between 1730 and 1775 varying ordinarily between 2000 and 3000 tons, but in one year (1771) rising to between 7000 and 8000 tons.[7]

So far as known, the first metal, other than iron, mined by the whites within the territory of the United States was copper.[8] The first company began work about 1709, at Simsbury, Conn. The ore obtained there and in New Jersey seems to have been mostly shipped to England. A few years later attempts were made to work mines of lead and cobalt in Connecticut and Massachusetts, but none of these enterprises seem to have been conducted with much vigour or to have met with any success. The first metal, other than iron, mined and smelted on any scale of importance was lead. The ore of this metal—galena—occurring in considerable quantity, and in many localities, on or near the Mississippi, and being easily smelted by the roughest possible methods, was made use of at an early date. While this region—then called Louisiana—was in possession of the Spanish, some mines were opened and worked, although in a very rude manner, the ore being taken out from mere pits and smelted on log-heaps. In 1774 Julien Dubuque began operations in the region of the upper Mississippi, at the place where is now the flourishing city which bears his name; but no real development of the mining interest took place in that region until half a century later.

The first mining excitement of the United States dates back to the discovery of gold by the whites in the southern States, along the eastern border of the Appalachian range, in Virginia, and in North and South Carolina. The existence of gold in that region had been long known to the aboriginal inhabitants, but no attention was paid to this by the whites, until about the beginning of the present century, when nuggets were found, one of which weighed 28 ℔. From 1824 the search for gold continued, and by 1829 the business had become important, and was attended with no little excitement. In 1833 and 1834 the amount annually obtained had risen to fully a million of dollars. A rapid development of the lead mines of the West, both in Missouri and on the upper Mississippi in the region where Iowa, Wisconsin, and Illinois adjoin one another, took place during the first quarter of the present century and as early as 1826 or 1827 the amount of this metal obtained had risen to nearly 10,000 tons a year. By this time the make of iron had also become important, the production for 1828 being estimated at 130,000 tons.

Mining industries about 1850.The following details show the development of the mineral resources of the country at the middle of the present century. In 1850 the shipments of anthracite amounted to nearly 3,500,000 tons; those of Cumberland or semi-bituminous coal were about 200,000 tons. The yearly production of pig iron had risen to between 500,000 and 600,000 tons. The annual yield of gold in the Appalachian belt had fallen off to about $500,000 in value, that of California had risen to$36,000,000, and was rapidly approaching the epoch of its culmination (1851-1853). No silver was obtained in the country, except what was separated from the native gold, that mined in California containing usually from 8 to 10 per cent. of the less valuable metal. The ore of mercury had been discovered in California before the epoch of the gold excitement, and at New Almaden, about 100 miles south of San Francisco, was being extensively and successfully worked, the yield of this metal in the year 1850-51 being nearly 2,000,000 ℔. At this time the copper mines of Lake Superior were being successfully developed, and nearly 600 tons of metallic copper were produced in 1850. At many points in the Appalachian belt attempts had been made to work mines of copper and lead, but with no considerable success. About the middle of the century extensive works were erected at Newark for the manufacture of the oxide of zinc for paint; about 1100 tons were produced in 1852. The extent and value of the deposits of zinc ore in the Saucon valley, Pennsylvania, had also just become known in 1850. The lead production of the Missouri mines had for some years been nearly stationary, or had declined slightly from its former importance; while that of the upper Mississippi region had, in the years just previous to 1850, risen to from 20,000 to 25,000 tons a year, but was gradually declining, having in 1850 sunk to a little less than 18,000 tons.

Coal.Coal.—Coal exists in the United States in large quantity in each of its important varieties, hard coal, or anthracite; soft or bituminous coal; and lignite, or brown coal. Semi-bituminous coal, which stands midway between hard and soft coal, is also an article of importance, being especially well adapted for blacksmiths use, and also for ocean-going steamers. Geologically, the anthracite, semi-bituminous, and bituminous coals nearly all belong to the same formation, the Carboniferous par excellence. All the coal of the Appalachian region and Central Valley is of this geological age, excepting the small field near Richmond, Va., and two in North Carolina, the Deep River and Dan River fields, which are of Mesozoic age. That of Richmond was the first coal-field worked in the United States; but it is no longer of any importance. The North Carolina Mesozoic areas have never been developed to any extent. All the Cordilleran coal and that at the eastern base of the Rocky Mountains is either Tertiary or belongs to the very uppermost portion of the Cretaceous. Some of it is decidedly lignite, and is so called by the people who use it; but most of it, although of so recent geological age, is called coal, and, in point of fact, does not differ essentially from Palæozoic coal in external appearance. The area underlain by the Coal-measures in the United States is very large.

The areas of the various coal-fields are, in round numbers, as follows:—Rhode Island, 500 square miles; Appalachian, 59,000; Central (Illinois, Indiana, Kentucky), 47,000; Western (Missouri, Iowa, Kansas, Arkansas, Texas), 78,000 ; Michigan, 6700; total, 191,200. Of these fields the Appalachian is, and is likely long to remain, by far the most valuable. Those of Rhode Island and Michigan are practically of very little importance. Different portions of the fields are of very different value, as respects quality and quantity of coal, and portions of them do not contain coal-beds of sufficient thickness or of good enough quality to be worked with profit.

The following table (VI.) shows the amount of coal produced in the several States and Territories (not including the local and colliery consumption), and the value at the mines in 1885:—

 Tons. Value. Pennsylvania— $Anthracite 32,265,421 72,274,544 Bituminous 23,214,285 24,700,000 Illinois 8,742,745 11,456,493 Ohio 6,978,732 8,206,988 Maryland 2,865,974 3,209,891 Missouri 2,750,000 3,850,000 West Virginia 3,008,091 3,369,062 Indiana 2,120,535 2,731,250 Iowa 3,583,737 4,819,230 Kentucky 1,700,000 2,094,400 Tennessee 892,857 1,100,000 Virginia 567,000 666,792 Kansas 1,082,230 1,410,438 Michigan 45,178 75,000 Alabama 2,225,000 2,990,000 Georgia 133,929 180,000 Colorado 1,210,769 3,051,590 Wyoming 720,828 2,421,984 New Mexico 271,442 918,606 Utah 190,286 426,000 California 63,942 214,845 Oregon 44,643 125,000 Washington 339,510 950,615 Texas 133,928 300,000 Arkansas 133,928 225,000 Montana 77,179 302,540 Dakota 23,214 91,000 Idaho 893 4,000 Indian Territory 446,429 750,000 Totals 95,832,705 152,915,268 The amount consumed for local and colliery use would increase the total five or six per cent. Including all the coal thus consumed, the figures for the four years ending 1885 stand as follows, the value at the mine being added for each year, except that for 1885 the value of the “commercial coal” only is given:—1882, 92,219,454 tons,$146,632,581; 1883, 102,867,969 tons, $159,494,855; 1884, 106,906,295 tons,$143,768,578; 1885, 99,069,216 tons, $152,915,268. The Appalachian coal-field does not occupy any portion of the State of New York, but extends from near its southern boundary south-west into Georgia and Alabama:—Pennsylvania, 12,302 square miles; Maryland, 550; Ohio, 10,000; West Virginia, 16,000; Virginia, 1000; Kentucky, 8983; Tennessee, 5100; Georgia, 170; Alabama, 5530; total, 59,635 square miles. Pennsylvania, to which State the anthracite of the country is practically limited, produces more than half the coal raised in the United States, and about one-eighth of the total yield of the world, the increase during the past five years having been astonishingly rapid. Ohio stands next in importance among the States so situated as to have a portion of this field within their borders,—its yield, however, being only one-eighth of that of Pennsylvania. Maryland produces about one-third as much as Ohio. The production of the remaining States over which the Appalachian field extends is small as compared with the extent of their areas underlain by coal. In Kentucky and Alabama, however, some progress has been made within the past few years. About three-quarters of the coal mined in the United States is from the Appalachian field. Next in importance to the Appalachian is the Central coal-field, from which Illinois draws its supplies of fuel. The yield of that State is somewhat less than a fifth of that of Pennsylvania, while Indiana produces about one-fifth as much as Illinois. The coal of the Central field is decidedly inferior to that of the Appalachian, since it contains on the average considerably more ash and more water. The quality of the coal of the Western field is somewhat variable, but on the whole inferior to that of the Illinois field. The region over which it is spread is, however, not well supplied with forests; and hence the amount annually raised is considerable (nearly 4,000,000 tons for Iowa and 2,500,000 for Missouri). The southern extension of this field through Arkansas and Texas has, as yet, been very little explored or developed. The coal-producing areas of the Cordilleran region are all of comparatively small size, and no one of them is capable of furnishing a large supply for any considerable number of years; but taken together they are of great value, not only for local consumption, but for supplying the Mexican plateau. The Cordilleran coal (some call it lignite) is all newer than the Carboniferous; and it is not known that there is any coal at all, either in the Rocky Mountains or farther west, in that portion of the geological series which is the palæontological equivalent of the true “Carboniferous.” A large part of the Cordilleran coal has a geological position close upon the line between the Cretaceous and Tertiary. By some palæontologists it is referred to the one, by some to the other. The quality of this newer coal is very variable: in some localities it is quite good, but in general it is decidedly inferior to the average coal of true Carboniferous age; some portions are distinctly lignitic in character. While the amount of coal in the United States is large, it is not by any means so much larger than that of England as it has been always inferred to be from the simple consideration of the comparative dimensions of the areas over which coal is known to exist in the two countries. In the Central and Western fields the number of beds is small, and they are never of great thickness; nor is it known how far the total area embraced within the limits usually assigned as that of these fields was really originally underlain continuously by coal-seams, or how much of these seams has been removed by erosion. It is only in regard to the anthracite fields that even an approximate statement of the total remaining available quantity of coal can be made. Mr P. W. Sheafer, a mining engineer of long experience in that region, has made a statement that the anthracite fields originally contained about 25,000,000,000 tons of coal. Mr Ashburner, the assistant in charge of the State survey of the anthracite coal district, has stated that, up to 1st January 1883, the total production of that district had amounted to 509,333,695 tons. He also estimates that two-thirds of the coal actually attacked had been lost in the mining. Making allowance for increased consumption and other considerations, it would appear that 200 years must be taken as the maximum time during which the anthracite fields will hold out, while it is probable that they will be practically exhausted considerably earlier. As regards the quantity of available coal in any portion of the bituminous coal-fields of the country, the only estimate we have is that of Prof. Lesley, who (February 1886) estimates the amount of available coal in the Pittsburgh seam at 5,000,000,000 tons, but confesses that any exact calculation is impossible. If the consumption remains at the figure at which it stood in 1884 (11,000,000 tons), this quantity will last about 450 years. Petroleum.Petroleum.[9]—The oil-producing districts of present or past importance are nearly all in Pennsylvania; but there are small productive areas in the adjacent portions of New York. (1) The Allegany district, including the Richburg and several small outlying pools in Allegany county, New York, has a productive area of 31 square miles, and up to January 1885 produced 15,000,000 barrels. (2) The Bradford district, embracing the oil-pools in central and northern M‘Kean county, Pa., and southern Cattaraugus county, N.Y., has a productive area of 133 square miles, 121 of which are included in the Bradford field proper. The geological position of the oil-bearing strata is indicated by the fact that the uppermost oil-sand is, at Bradford, 1775 feet below the bottom of the lower member of the millstone grit, or Pottsville conglomerate, which in western Pennsylvania is one of the most persistent and best-recognized geological horizons, and is there known as the Olean conglomerate. This district had up to January 1885 produced 109,000,000 barrels of oil. (3) The Warren district lies in eastern Warren and north-eastern Forest counties, Pa.; it has an area of 35 square miles, and up to January 1885 had produced 12,000,000 barrels. The oils from different subdivisions of this district vary considerably in colour and gravity, although generally spoken of as “amber oils.” They come from “sands” (sand rocks) of varying geological horizons, from 1100 to 1850 feet below the Olean conglomerate. (4) The Venango district—the scene of all the earlier oil developments—has an area of 65 square miles, and includes forty distinct and well-recognized oil-pools, the largest of which lies between Oil City on the south and Pleasantville on the north, and covers an area of 28 square miles. The production of this district up to January 1885 had been about 55,000,000 barrels. The oil of the Venango district comes from three principal sand beds, of which the uppermost one lies about 450 feet below the base of the Olean conglomerate. They are all contained within an interval of 350 feet. The oils are generally green, but frequently black, and sometimes amber. The pebbles of the sand rocks are water-worn, sometimes as large as hazel-nuts, loosely cemented together and bedded in fine sand; but the sands are not so regular or homogeneous as in the Bradford and Allegany fields; consequently, the risk of obtaining unproductive holes and variable wells has always been greater in the Venango than in the Bradford and Allegany districts. (5) The Butler district includes the oil-pools in Butler and Clarion counties and in south-eastern Venango county. The area is 84 square miles, and up to January 1885 the production had been about 69,000,000 barrels. The oil here comes from the same group of sand rocks as in the Venango district. (6) The Beaver district includes the two principal oil-pools known as Slippery Rock and Smith's Ferry, having an area of about 16 square miles, with a production of 1,000,000 barrels up to January 1885. In both the oil-pools of the Beaver district heavy oil is obtained from the representative of the Pottsville conglomerate, and amber oil from the Berea grit, a member of the Sub-Carboniferous series. The geological position of all the other oil-fields is considered by the geologists of the Pennsylvania Survey as Devonian. The total area of the productive oil areas is given by Messrs Carll and Ashburner at 369 square miles, and the general boundaries of the oil-regions of Pennsylvania are now regarded as established. That production is declining is shown by the following figures. In July 1883 the number of producing wells was 17,100, and the average daily product per well was 3.8 barrels; in 1884 the corresponding figures were 21,844 and 3.0, and in 1885 they were 22,524 and 2.5. Since July 1882, when the maximum average daily production for any one month was realized (105,102 barrels), there has been an irregular but steady decline. In 1884 the shipments of petroleum were more than 1,000,000 barrels in excess of the production. At the end of August 1884 the stock of oil on hand had reached its maximum, 39,084,561 barrels; in September 1885 it had declined to 35,343,771 barrels. The price of petroleum had not, up to July 1885, been influenced by this condition of things, crude oil being at that time worth 92½ cents a barrel, 13¼ cents less than the average price for 1884. The extraordinary fluctuations in the price of petroleum during 1880 to 1886 are shown by the following figures:—in 1880 it ranged from 124⅜ to 70⅝ cents per barrel for crude oil; in 1881 from 100½ to 72; in 1882 from 135 to 49¼; in 1883 from 125 to 84¾; in 1884 from 115¾ to 51; in 1885 from 111⅝ to 68; in 1886 from 92¼ to 59¾. These figures, however, have but little reference to changed conditions of production. The fluctuations are simply the result of a colossal system of gambling, the magnitude of which may be inferred from the statement that the “clearances” on the “consolidated stock and petroleum exchange” for 1886 amounted to about 2,275,000,000 barrels. The daily average exports of petroleum for that year are given at about 44,300 barrels. The tendency to lower prices in 1886 was due in part to the remarkable yield of the oil-wells at Baku, on the Caspian Sea, and in part to discoveries, supposed to be of importance, in Ohio. Natural gas.Natural Gas.—The use of natural gas for illumination, and even for metallurgical purposes, has lately become a matter of importance. The existence of outflows or springs of gas in the region west of the Alleghany range has been long known, and the gas obtained from wells or bore-holes was used for illumination in Fredonia, N.Y., as early as 1821. One well after another was bored for this purpose at that place, until, in 1880, the supply had reached the amount of 110,000 cubic feet per month. The following figures, reported by T. P. Roberts, show the estimated value of the coal displaced by natural gas in the region where this source of heat and light was in use for the years mentioned:  1882 Pittsburgh region,$75,000; elsewhere, $140,000; total,$215,000 1883 Pittsburgh region, 200,000; elsewhere, 275,000; total, 475,000 1884 Pittsburgh region, 1,100,000; elsewhere, 360,000; total, 1,460,000

Gas seems to be a general concomitant of the oil all through the petroleum region, but for a long time the outflow of gas from the oil-wells was looked upon as a nuisance. According to Mr Ashburner,[10] the amount of gas at present flowing from the explored sands of Pennsylvania is probably two or three times greater than is required to meet all present demands. The same authority gives an account of the development of the natural gas resources of Ohio. Their amount is not yet ascertained with any degree of certainty, but it seems likely to be large. All the gas comes from the Palæozoic strata, from the Upper Coal-measures down to the Trenton limestone, the most prolific gas-bearing rocks being the Berea grit in the Sub-Carboniferous and the Trenton limestone of Lower Silurian age. Natural gas has been obtained in numerous localities in New York, but nowhere in considerable quantity except in the vicinity of the Allegany oil-district, in the county of that name. Portions of West Virginia, especially the Kanawha valley, give promise of being regions of large production.

Iron.Iron and Steel.—The following table (VII.) will convey an idea of the condition of the iron industry at the date of the last census, and of the progress made during the ten preceding years:—

 No. of Establishments. Amount ofCapitalinvested. Value ofProducts. Weight of Productsin Tons. 1870 808 $121,772,074$207,208,696 3,245,720 1880 1005 230,971,884 296,557,685 6,486,730 Increase per cent 24.38 89.68 43.12 98.76

The “weight of products,” as given above, includes the products of all the various processes or operations; hence there is evidently a certain amount of duplication (rolled iron, for instance, being mainly produced from pig). The following table (VIII.) gives the production in each branch of the steel and iron industries:—

 1870. 1880. Pig iron and castings from furnace 1,832,876 tons. 3,375,901 tons. All products of iron rolling mills 1,287,347 tons. 2,101,114 tons. Bessemer steel finished products 17,324 tons. 794,550 tons. Open-hearth steel finished products 0 tons. 83,163 tons. Crucible steel finished products 25,061 tons. 62,784 tons. Blister and other steel 2,040 tons. 4,424 tons. Products of forges and bloomeries 98,935 tons. 64,784 tons.

The distribution of the iron industry is extremely irregular. West of the Mississippi, with the exception of the angle between that river and the Missouri adjacent to St Louis, the amount of iron made is very small. The percentage of total production in 1880 was distributed as follows: Pennsylvania, 50; Ohio, 13; New York, 8; Illinois, 6; New Jersey, 3; Wisconsin, W. Virginia, Michigan, and Massachusetts, each nearly 2; Missouri, Kentucky, and Maryland, between 1.5 and 2; Indiana and Tennessee, about 1; all the other States and Territories, an aggregate of about 4 per cent. New England now makes but little pig iron, and the South scarcely any rolled iron; the West has largely embarked in the manufacture of steel by the Bessemer process, while New York has not a single Bessemer establishment; New York manufactures chiefly from ore, and Pennsylvania from pig and scrap iron; Michigan is the leading producer of charcoal pig iron, and now makes no other kind; only five States make Bessemer steel; and two States, Pennsylvania and New Jersey, produce nearly all the crucible steel.

The census year 1880 was one of exceptional prosperity for the iron and steel industries of the country. The production of pig iron and of Bessemer steel ingots and rails since 1880 is shown in the following table (IX.), from the statistics collected by the American Iron and Steel Association:—

 1881. 1882. 1883. 1884. 1885. 1886. Pig iron 4,144,253 4,623,323 4,595,510 4,097,868 4,044,526 5,683,329 Bessemer steel ingots 1,374,247 1,514,687 1,477,345 1,375,317 1,519,426 2,269,190 Bessemer steel rails 1,187,769 1,284,066 1,148,709 996,465 959,470 1,562,409

The total number of completed Bessemer steel works at the close of 1886 was 33, with 69 converters. Pennsylvania in that year made 59 per cent. of the ingots produced, Illinois 21, and other States 20. In the following table (X.) the amount of steel of all kinds produced is given (in tons) for the years stated:—

 Bessemer Steel Ingots. Open-Hearth  Steel Ingots. Crucible Steel Ingots. All other Steel. Total. 1870 37,500 1,339 31,250 70,089 1875 335,283 8,080 35,179 11,256 389,799 1880 1,074,261 100,850 64,664 7,558 1,247,334 1885 1,519,430 133,375 57,599 1,514 1,711,919 1886 2,269,190 218,973 71,972 2,366 2,562,502

By far the largest production of iron ore in the United States is from the rocks which lie below the Lower Silurian—the Azoic series of Foster and Whitney and the Archæan of Dana. In this formation the ore occurs in immense quantity—in what may without exaggeration be called mountain masses, which in many cases exhibit all the evidences of an eruptive origin, as in the case of the Iron Mountain of Missouri, or in some of the localities in the Marquette and Menominee regions of Lake Superior. At the first-mentioned locality the ore is intimately associated with an undoubted eruptive rock, with which it is intermingled in such a manner as to show that the two—ore and rock—must have had one and the same origin. A similar condition of things is revealed on Lake Superior, where the ore occurs in repeated interlaminations between sheets of unquestionably eruptive material. The ores thus occurring are hæmatite or specular ore and magnetite, with some hydrated oxide, or limonite, the result of the action of water on the previously formed hæmatite. They are in general extraordinarily free from deleterious ingredients, especially phosphorus and sulphur, although usually containing a small amount of silica. The purity of these ores is a strong indication of their origin differing from that of ordinary sedimentary deposits. Many of the analyses of Lake Superior ore show the presence of only a few hundredths of one per cent. of phosphorus. The most important district in which these ores occur is the south shore of Lake Superior, and the most important port of shipment is Marquette, a considerable amount being also shipped from Escanaba on Lake Michigan. From 1856 to 1886 the total shipments of iron ore from the region amounted to 31,030,160 tons. A district known as the Vermilion Lake iron district, in which the ore has similar characters, and where the quantity is believed to be very large, has been recently opened in Minnesota, on the north shore of Lake Superior. The Iron Mountain region, a little less than a hundred miles south of Saint Louis, although small, is of considerable importance. A very large portion of the Lake Superior ores goes to the Appalachian coal-field to be smelted; and this was formerly the case with the Iron Mountain ore, but the latter is now used in nearer localities. There is a very important and apparently inexhaustible deposit in Lebanon county, Pennsylvania; the ore is chiefly magnetite, and its mode of occurrence in close connexion with an eruptive rock is to the present writer strong, if not absolutely convincing, evidence of its igneous origin. There are important deposits of iron ore, on the eastern border of New York, especially in the Adirondacks, and along the Hudson River. The geological position in which a portion of these ores occurs is certainly the same as that of the ores of Lake Superior—namely, the Azoic. That is the character of the Adirondack ores, which have long been and still are extensively worked. The localities are chiefly in Clinton, Essex, and Franklin counties. In the last-named county are the very extensive Chateaugay mines. The ores of this region are chiefly magnetite, but they often contain too much phosphorus to be used in the manufacture of steel. There are important occurrences of magnetic ore near New York city and also near the Hudson river, in Orange, Rockland, Putnam, and Columbia counties. Some of these ores are adapted to the manufacture of Bessemer steel. The mode of occurrence of the ores in southern New York and northern New Jersey is peculiar; they are not so distinctly eruptive as are the ores of Missouri and Lake Superior. One authority regards the New Jersey ores as unquestionably of sedimentary origin. This New Jersey district is not in a flourishing condition at present, since the ores, as a rule, are not adapted to the manufacture of Bessemer steel. Similar ores occur at many points, and often in large quantity, in the Azoic or crystalline belt of the Appalachian system, in the States lying farther south than New Jersey. In Mitchell county, North Carolina, is a deposit known as the Cranberry Bank, of which the quality is excellent, and the quantity is believed to be very great. Up to the present time, however, but little of this ore has been shipped. Next to the Azoic ores in importance, but decidedly inferior in quality, arc those of the Clinton group, a member of the Upper Silurian series. The ore occurring in this geological position is known by various names, the most common ones being fossil or dyestone ore. It is a red hæmatite, not crystalline like the specular variety of the peroxide, but usually in a more or less granular or concretionary form, that of flattened grains resembling flax seed being a common mode of occurrence, whence the name “flax-seed ore.” This deposit occurs at many points along the outcrop of the Clinton group all the way around from Wisconsin, through Canada and New York, into Pennsylvania, and down the eastern slope of the Appalachian range to Georgia, and is said to be the most extensive deposit of iron ore in the world yet discovered. The fossil ore, though large in quantity, contains too much phosphorus to be used for making steel in the ordinary method; but, being in places favourably situated with regard to fuel, and yielding as it does a satisfactory quality of cast iron, it is quite extensively mined at various points. Next to the Clinton ore in importance comes the brown hæmatite ore (limonite), which occurs in numerous localities, but of which the most extensive deposits are those occurring in the Lower Silurian limestones of the Appalachian system, and especially along the line of the Great Valley. Some of the ores are of excellent quality, notably those of Litchfield county, Connecticut, in the so-called Salisbury district. The iron made in charcoal furnaces in this region is considered as being of the highest value for articles in which strength and toughness are essential. Carbonate of iron, in the form of spathic iron, is of rare occurrence. The argillaceous carbonates (clay ironstone) are also of comparatively little importance, although used to some extent, especially as mixed with other ores, in western Pennsylvania and Ohio. The coal-fields west of the Appalachians are very poorly supplied with iron ores. Blackband ore is also of somewhat rare and limited occurrence in the United States.

In spite of the abundance of iron ore in the United States and the existence of a heavy protective duty (75 cents per ton), a large quantity of ore is imported from abroad—chiefly from Spain, Elba, Algiers, and Cuba (1,039,433 tons valued at $1,912,437 in 1886). Almost all this was used in Pennsylvania, and chiefly in the manufacture of Bessemer pig iron and spiegeleisen. Up to 1884, at least, no pig iron suitable for the manufacture of Bessemer steel by any process now in use in the United States had ever been made south of the Potomac or south of Wheeling. Neither has iron suitable for crucible steel been made in the United States which can compete with that of Sweden. The annexed table (XI.), published by Mr Swank, shows the production of iron ore in tons in the leading ore-producing districts for the years 1884, 1885, and 1886:— 1884. 1885. 1886. Lake Superior (Michigan and Wisconsin) 2,455,924 2,231,064 3,258,174 Vermilion Lake (Minnesota) 62,124 225,484 304,396 Missouri 233,225 169,162 379,776 Cornwall (Pennsylvania) 412,320 508,864 688,054 New Jersey 393,710 330,000 500,501 Chateaugay (Lake Champlain, New York) 214,394 143,278 214,800  Crown Point (New York) ${\displaystyle \scriptstyle {\left.{\begin{matrix}\ \\\\\ \ \end{matrix}}\right\}\,}}$ Port Henry (New York) Other Lake Champlain mines (New York) 290,500 235,799  ${\displaystyle \scriptstyle {\left\{{\begin{matrix}\ \\\\\ \ \end{matrix}}\right.}}$ 60,084 298,868 15,000 Hudson River Ore and Iron Company (New York) 90,000 55,000 75,000 Tilly Foster (New York) 35,964 18,910 17,728 Forest of Dean (New York) 20,370 18,274 18,000 Salisbury region (Connecticut) 25,000 32,000 36,000 Cranberry (North Carolina) 3,998 17,839 24,106 Tennessee Coal, Iron, and Railroad Company's Inman mines (Tennessee) 70,757 94,319 81,650 Total 4,308,286 4,079,993 5,972,137 Gold and silver.Gold and Silver.—The washing of the high or Tertiary gravels by the hydraulic process and the working of mines in the solid rock did not, on the whole, compensate for the diminished yield of the ordinary placer and river diggings, so that the produce of gold in California continued to fall off, and by 1860 had decreased to about half what it had been ten years before. Discoveries in other Cordilleran Territories, notably in Montana and Idaho, made up, however, in part for the deficiency of California, so that in 1860 the total amount of gold produced in the United States was estimated at not less than$45,000,000. In the latter part of the decade 1850-59 the Territories adjacent to California on the east, north, and south were overrun by thousands of miners from the Sierra Nevada gold-fields, and within a few years an extraordinary number of discoveries were made, some of which proved to be of great importance. The most powerful impulse to mining operations, and the immediate cause of a somewhat lengthy period of wild excitement and speculation was the discovery and successful opening of the Comstock lode in 1859, in the western part of what is now the State of Nevada, but was then part of the Territory of Utah, and known as the “Washoe Country” (from a small tribe of Indians of that name). The locality of the lode, where there soon grew up a large town called Virginia City, is about 20 miles east of the boundary of California, and nearly due east of the northern end of Lake Tahoe. As early as August 1860 two mills were at work stamping and amalgamating the ore from the lode; mining had begun on a large scale; and many ingenious metallurgists were endeavouring to ascertain experimentally how the somewhat complex metalliferous combinations there occurring could be best and most economically treated. So far as quantity of bullion produced is concerned, these operations were eminently successful. The various estimates of the total yield of the lode from the time of its discovery to 30th June 1880, collected and published by Mr Eliot Lord, run from $304,752,171.54 to$306,181,251.05. The total production to the end of 1886 may be estimated as having been not far from $320,000,000. The bullion obtained from the milling of the ores contains about an equal amount in value of silver and gold. The mines on this lode thus added to the bullion stock of the world$100,000,000 more in eighteen years than the whole Freiberg district in Saxony had furnished in a little over 700 years (1168-1875). The lode is an ore channel of great dimensions, included within volcanic rocks of Tertiary age, which themselves have broken through pre-existing strata of Triassic age. It exhibits some of the features of a fissure vein, combined with those of a contact deposit in part and of a segregated vein in part. The gangue is quartz, very irregularly distributed in bodies often of great size, and for the most part nearly or quite barren of ore. The metalliferous portions of the lode, “bonanzas,” as they are here generally called, are usually of great size, but extremely irregular in their position. Their number has been about twenty, most of them lying near the surface; but the last important one discovered was struck at a little over 1000 feet below the surface. Its dimensions were—according to Mr Church—about 700 feet in length, by 500 deep, and 90 wide; the average yield $93.55 per ton; and the total value of the bullion obtained from it$104,007,653. The mines on this lode have been worked to a greater depth than any mines in the world extending over an equal amount of ground. Up to October 1886 work was still being carried on at several points below the depth of 3000 feet. The lower levels, below that of the Sutro tunnel, which intersects the lode at a point about 2000 feet below its outcrop, are now abandoned. Workings are still going on, however, above the great adit level, where there yet remain considerable bodies of low grade ore, which can perhaps be extracted with moderate profit, since more economical methods have been introduced both in mining and milling. The yield of the Comstock lode at present, although small as compared with that of its prosperous years, is still much larger than that of all the mines in the Freiberg district of Saxony.

The success of the Comstock lode workings led to the active exploration of the whole adjacent region, with the result that a great number of localities were discovered where auriferous and argentiferous ores occur, and some of these have been extensively wrought and have produced largely. In very few, however, do the ore deposits bear the distinguishing characters of true or fissure veins. Special mention may be made of the Meadow Valley and Raymond and Ely mines in Lincoln county. The first-named mine is in Pioche, 273 miles south of Palisade station on the Central Pacific Railroad; the enclosing rock is quartzite of Lower Silurian age, the veins varying in width, but averaging from 2 to 2½ feet, the ore being carbonate of lead near the surface with chloride of silver, passing into sulphurets, as usual, in going down. At the end of 1873 this mine was 1100 feet deep, and a large body of ore had been struck yielding $300 in value of silver per ton. The culminating years of the prosperity of these mines were 1872 and 1873, and the yield of the Ely district for those years respectively was$5,321,007 and $3,735,596. The Eureka district in the central part of Nevada is of importance on account of the magnitude of its yield, and because mining and smelting operations have been carried on uninterruptedly since 1869, thus affording great facilities for scientific investigations. Its ores are chiefly galena, accompanied by the various oxidized combinations resulting from its decomposition. Similar ores of zinc are also present, but in much smaller quantity. These ores are rich in gold and silver (average—15 per cent. lead, 0.079 silver, and 0.00248 gold). The rocks in which they occur are chiefly limestones, and of Lower Silurian age; the deposits are very irregular in form. In many of their features they closely resemble the so-called pipe-veins of the North of England lead mines. Much of the ore has become decomposed, and has been subjected, since decomposition, to a rearrangement by water, analogous to stratification. The total yield of the Eureka district from 1869 to 1883 is stated at about$60,000,000 in value of gold and silver, and about 225,000 tons of lead. According to present indications, this district is approaching exhaustion.

Utah has important mines resembling to a considerable extent those at Eureka. The Hornsilver mine at Frisco, in southern Utah, is a large contact deposit, 30 to 50 feet wide, between dolomitic limestone and an eruptive rock called rhyolite or trachyte. The ores are sulphate of lead with some carbonate, associated with heavy spar, which occurs chiefly near the wall of eruptive rock. This mine has paid $4,000,000 in dividends, but none since 1884. The Little Cottonwood district, at a height of over 10,000 feet in the Wahsatch range, shows a record of about 3500 “locations” made within an area 2½ miles square. It includes the famous Emma mine, where a large body of ore occurred occupying an egg-shaped cavity in the Carboniferous limestone, which yielded largely (over$2,500,000); but it was soon worked out. The Flagstaff mine in the same cañon, a “pipe-vein” or deposit occupying several irregular cave-like openings in the limestone, seems also to have been nearly or quite worked out. One of the largest and most productive mines in the country, and the most important one in Utah, is the Ontario, in the Uintah district, Summit county. The vein seems to be, in the lower levels at least, a contact mass between walls of quartzite and highly decomposed eruptive rock (called porphyry), and it varies in width from a few inches to 15 feet. It was discovered in 1872, has been worked by a company with $15,000,000 capital since 1877, and up to February 1887 had paid$8,075,000 in dividends. The Silver Reef district in Washington county is a region of remarkable interest, where sandstones of Triassic age have been broken through and invaded by eruptive masses (andesites and trachytes), and where the ore, according to Prof. Reyer, occurs in flat masses and impregnations between the strata, adjacent to the eruptive rock, and is especially largely developed in contact with the remains of plants with which the rock is filled. The ores are carbonate of copper with chloride of silver, passing into the sulphuret at depths. According to the same authority, these mines produced a few years ago over a million of dollars a year; but they have much fallen off of late. The census report of 1880 gives the total product of the district up to 1st June 1880 at a little over $3,250,000. The metalliferous deposits of Colorado are important from their magnitude and variety.[11] Of the actual precious metal production, by far the largest portion is derived from pyrites and galena and their decomposition products. The telluride ores of Boulder county and the auriferous pyrites of Gilpin county, with a few individual deposits in the southern portion of the State, constitute the source from which it is derived. With these exceptions its mineral deposits may be considered as essentially silver-bearing. The principal source of silver is argentiferous galena and its decomposition products, while argentiferous grey copper or freibergite is, next to this, the most important silver-bearing mineral. The sulphides of silver also occur, and in some cases bismuth is found in sufficient quantity to constitute an ore. As yet, so far as known, no copper is extracted except as an adjunct in the reduction of silver-bearing copper ores. Placer deposits are generally confined to the valley bottoms among high mountain ridges; their present yield is relatively inconsiderable. Before the silver ores of Leadville were discovered, mining in Colorado was principally confined to approximately vertical veins in the Archæan rocks of the Front range or in the eruptive rocks of the San Juan region; but since the limestone deposits of the Mosquito range have proved so exceptionally rich attention has been more and more turned to the ores in sedimentary rocks, and many new districts have been discovered, but none to rival that of Leadville. The Leadville ores are flat sheets, of the kind often designated in the Cordilleran mining regions as “blanket deposits,” and appear to be contact deposits between the Carboniferous limestone and the overlying felsite, with additional or incidental ore accumulations in the limestone in irregular cavities. According to reports locally published the product of the Leadville “smelters” or smelting works in 1886 was—lead, 51,925,546 ℔; silver, 4,569,013 oz.; gold, 22,504 oz.; total value,$7,515,148. From the mode of occurrence of the Leadville deposits it seems probable that they will within a few years become practically exhausted. The area over which the blue limestone of Emmons (dolomite of Rolker) has been found in places productive is, however, very large, having been estimated at 225,000,000 square feet. The Montana deposits, which up to 1880 are reported to have yielded fully $50,000,000, seem now to have been pretty much worked out. The workings from which the ores are now chiefly obtained appear to be of the class of segregated veins. The ores consist largely of auriferous pyrites in a gangue of quartz, oxidized in their upper portions and there easily manipulated, but in depth passing into the more refractory sulphurets. Unlike the ores described as occurring in Colorado and Utah, they are accompanied by copper rather than by lead, and they are also rather manganiferous than ferriferous. Much yet remains to be done before their nature and value can be fully understood; and the same may be said of the adjacent Territory of Idaho, the auriferous gravels of which resemble those of Montana, and have been next to those of California and Montana in importance (total yield to date estimated at$30,000,000). The deep gravels of Boise basin seem to be exceptional, and to resemble the deep or high gravels of California. A very large portion of the mines (other than placer) of Idaho appear to be of the fissure class, and to be enclosed in a country rock of granite, resembling in many respects the veins of the vicinity of Austin in Nevada. In Arizona, which ranks along with Idaho in the production of the precious metals, and is next to Michigan and Montana as a producer of copper, the mode of occurrence of the metalliferous deposits is complicated and varied, and is still very imperfectly known. They appear to be largely of the nature of contact deposits, dependent on the presence of some ancient or modern eruptive mass. The famous Tombstone district, in Pima county, has been a productive one, but seems at present to be declining.

Table XII.—Gold and Silver Production of the Different States and Territories for 1885 (in Thousands of Dollars).

 Gold. Silver. Total. Alaska 300 2 302 Arizona 880 3,800 4,680 California 12,700 2,500 15,200 Colorado 4,200 15,800 20,000 Dakota 3,200 100 3,300 Georgia 136 ... 136 Idaho 1,800 3,500 5,300 Montana 3,300 10,060 13,360 Nevada 3,100 6,000 9,100 New Mexico 800 3,000 3,800 North Carolina 152 3 155 Oregon 800 10 810 South Carolina 43 ... 43 Utah 180 6,750 6,930 Washington 120 70 190 Other States and Territories 90 5 95 ⁠Total 31,801 51,600 83,401

The production of gold in the United States for each of the seven years 1880-86 inclusive is estimated (in millions of dollars) as—36, 34.7, 32.5, 30, 30.8, 31.8, 35. Similarly the coining value of the silver produced during the same years is returned as—39.2, 43, 46.8, 46.2, 48.8, 51.6, 51. The commercial value of silver has, however, of late been considerably less than the coining value—42, 42.5, and 39.4 in 1884-85-86.

Mercury.All the quicksilver produced in the United States comes from California, although small quantities of the ores of this metal have been obtained at various points in Colorado, and also in New Mexico. A little mercury has also been produced in Oregon. The Californian mines are all in the Coast ranges, in rocks of Cretaceous age. Small veins of quartz containing a little cinnabar have been found in the Sierra Nevada; but this ore is not known to exist anywhere in that range in workable quantity. The mercurial ores of the Coast ranges occur in very irregular deposits, in the form of strings and bunches, disseminated through a highly metamorphosed silicious rock. The first locality where this metal was successfully mined was New Almaden, about 100 miles south of San Francisco. Another locality—New Idria—about 100 miles still farther south, has also been productive, but in a less degree. Mercury ores have also been mined at several points north of San Francisco, in the neighbourhood of Clear Lake, where they occur, not only in metamorphic Cretaceous strata resembling those of New Almaden, but also in recent volcanic rocks, where gold has sometimes been found in intimate association with the cinnabar. The New Almaden mine has been productive since 1850, but the yield has varied greatly from year to year, partly on account of the irregularity of the mode of occurrence, and partly on account of interference on the part of the United States based on a question of title to the property. The most productive year was 1876, when the number of flasks (of 76½ ℔) obtained was 47,194. The number produced in 1880 was 23,465, and in 1886 18,000. The total produce of the Californian mines was 59,926 flasks in 1880, 60,851 in 1881, 52,732 in 1882, 46,725 in 1883, 31,913 in 1884, 32,073 in 1885, and 29,981 in 1886. No new discoveries of localities of importance have been made during the past few years, and the mines now worked in California are all in a depressed condition.

Tin.Tin has been discovered in numerous localities, and various attempts have been made to open mines—in Maine, New Hampshire, Virginia, Alabama—but hitherto the amount of the metal produced has been quite insignificant. The region from which most has been expected is the Black Hills of Dakota, about 20 miles south-west of Rapid City. The occurrence of the tin ore and the associated minerals at the mine to which the name of Etta has been given is very similar to that of the ores in the Erzgebirge. The cassiterite is disseminated through a granitic or granitoid rock in irregular bunches, strings, and even masses, associated with the usual minerals. There has not hitherto been any production of commercial importance from this source.

Copper.The present sources of copper are almost exclusively the Lake Superior region, and the Territories of Montana and Arizona. The mines of Lake Superior are the only important mines in the world in which the metal is exclusively obtained in the native state. The mode of occurrence of the copper varies, however, considerably in different portions of the mining district, which extends from Point Keweenaw along the southern shore of the lake to a little beyond the Ontonagon river. The most productive mines at present are those in the vicinity of Portage Lake, about half-way between the eastern and western extremities of the cupriferous range. The rock in which the metal occurs is an old basalt, much metamorphosed from its original condition, and in the form now generally called melaphyre. This belt, commonly known in the region as the “trap range,” is a volcanic material, interbedded in numerous alternating layers with sandstone and conglomerate, equivalent geologically to the Potsdam sandstone of the New York Survey or the Primordial of Barrande. The metal occurs along nearly the whole extent of Keweenaw Point in veins crossing the formation, and having all the characters of true fissure veins, the gangue being a mixture of quartz and calcite with various zeolitic minerals. Copper in large masses has been found in various mines on Lake Superior, but in none of such great dimensions as at the Minnesota mine, near the Ontonagon river. The largest mass discovered here weighed about 500 tons. Its length was 46 feet, its breadth 18½, and its maximum thickness 8½. While a considerable amount of copper has been obtained on Lake Superior in large masses, and lumps too small to be shipped separately, and therefore put in barrels and called “barrel-work,” much the larger portion occurs in small grains, scales, and strings, disseminated through the rock. For crushing rock of this character a new form of stamp known as the Ball stamp was invented, one head of which is capable of crushing from 220 to 250 tons of rock in twenty-four hours. The most interesting mine on Lake Superior, and indeed the largest and most important—not being an open-work—in the world, is that of the Calumet and Hecla Company. The mode of occurrence of the copper, which is all in the metallic form, and like that of the other Lake Superior mines almost chemically pure, is peculiar. The cupriferous mass is a bed of eruptive material, interstratified with other masses of similar origin, but itself a conglomerate made up chiefly of more or less rounded pebbles of eruptive rock—rhyolite, trachyte, and basalt cemented together by native copper. The Calumet and Hecla mines, which form one connected work, have been opened over a length of about 1¾ miles, and to a depth, on the inclination of the metalliferous bed, which is about 39°, of about 3300 feet. The number of men employed is about 2800, and the production for 1886 was 22,552 tons (the total amount since 1866 being 201,529 tons).

The copper mines of Montana are chiefly in the vicinity of Butte City. The really important operations seem to be pretty closely limited to an area only 2½ miles long by 1 mile wide, within which are three important silver mines, as well as the copper mines which make the district so famous. There are two great classes of mineral occurrences, cupriferous veins, carrying more or less silver, and silver veins, with a manganese gangue, which carry little or no copper. These two groups have certain features in common. They all occur in granite, and are all accompanied by zones of decomposed country rock, which run parallel to and usually form the walls of all the large copper veins that have been opened to any great depths, and which are called porphyry dikes by the miners, but are really granite altered by the chemical changes which have accompanied the formation of the lode. They all pitch vertically or nearly so, and lack entirely the well-defined walls and the selvages which are characteristic of fissure veins. The cupriferous veins appear on the surface as wide bands of quartzose rock, much decomposed and stained with gossan. The surface ore always carries silver, is almost entirely in a free-milling condition, and generally in paying quantity. Nearly all the lodes at present worked for copper were at first worked for silver, and this condition continued until the water-line was reached, when the base ores—mainly the ordinary sulphuretted combinations of copper and of copper and iron, especially copper glance or vitreous copper and erubescite or variegated, peacock, and horseflesh ore—set in. The veins are of great size, being often thirty feet wide for several hundred feet in length. The average width of pay ore in the copper veins is stated to be not less than 7 feet; the Anaconda—the widest of any yet opened—averages over 12 feet of profitable ore, and in many places widens to 30 or 40 feet for a great distance, showing no diminution in richness at the depth of 800 feet. In a part of these veins—as, for instance, the Anaconda and Liquidation—the ore is copper glance in a gangue of quartz and decomposed feldspathic rock; while another type of veins, represented most perfectly by the Parrot vein, has nearly its entire metallic contents in the form of erubescite. This last-mentioned vein is also of great importance for its silver. The veins of the manganese-silver group lie all within a small area, but are of much interest and value. The gangue is quartz, heavily charged with its various oxidized combinations of manganese, all more or less argentiferous, the amount of silver ranging from three or four to several hundred ounces of that metal per ton. In these manganese veins the transition from decomposed oxidized combinations to the hard silicate and carbonate of that metal at the water-line is as sudden and striking as that from oxidized to sulphuretted ores in the previously-mentioned class of veins. An exceptional occurrence in this district is that of the Gagnon mine, of which the gangue is chiefly quartz, and the ore argentiferous zinc blende.

There has of late years been a falling off in the production of copper in Arizona, but this appears to be due to unfavourable situation with reference to a market rather than to exhaustion. The Santa Rita mines, in New Mexico, near the Arizona line, were the first worked in that region, but are at present idle. The original workings were for metallic copper, occurring near the contact of a bed of limestone with an eruptive rock resembling felsite, the whole deposit being one of irregular character. Near the junction of the felsite and the limestone there is a series of parallel veins, in which the copper occurs in the form of carbonates and oxides. The Clifton district has been the scene of continuous mining operations since 1872, and is at present the largest producer of copper in the south-west. Its cupriferous deposits have been divided into three classes,—those occurring in limestone, those associated with porphyry, and those in the granite. The ores of the first class are the red oxide in a gangue of compact hæmatite and the carbonates in a gangue of brown manganese ore. The ores of the second and third class are oxides and oxysulphides, changing into copper glance at a trifling depth, and into yellow sulphurets in the deepest workings. The deposits are irregular, in some respects resembling contact deposits, and in others the metalliferous occurrences in the North of England lead mines. They are called by many mining geologists “pockets”; but other authorities consider them true veins. The most productive mine of the district—the Longfellow—is described as being an almost vertical fissure in stratified limestone, at or near its junction with a dike of felsite.

Table XIII.—Copper Production of United States (in Tons) from 1882 to 1886.

 1882. 1883. 1884. 1885. 1886. Lake Superior 25,438 26,653 30,961 32,209 35,593 Montana 4,043 10,658 19,238 30,267 25,719 Arizona 8,029 11,010 11,935 10,137 6,989 Other localities 2,957 3,253 2,574 1,438 1,340 Total domestic copper 40,467 51,574 64,708 74,051 69,641 From imported ores 446 726 1,276 2,271 2,000 ⁠Total 40,913 52,300 65,984 76,322 71,641

Zinc.Zinc has become within the past few years an important article of production. The ores are found in very numerous localities, usually in connexion with those of lead, both in the Appalachian range and throughout the comparatively undisturbed Palæozoic regions of the Mississippi valley. The Illinois zinc furnaces are at Peru, La Salle, and Collinsville; those of Missouri mostly at Carondelet near St Louis, but there is one at Rich Hill and one at Joplin. Those of Kansas are at Pittsburg, with the exception of one at Weir City. There is also a small establishment at White River in Arkansas. There are also zinc works at Bergen Point, N.J., and at Bethlehem, Pa., and one small establishment in Tennessee.

Table XIV.—Total Zinc Production (in Tons) from 1881 to 1885.

 1881. 1882. 1883. 1884. 1885. Illinois 14,509 16,250 15,037 15,708 17,345 Kansas 4,464 6,576 8,500 7,017 7,591 Missouri 2,455 2,232 5,116 4,670 4,175 Other States ... 5,087 4,770 7,018 7,216 ⁠Total ... 30,145 33,423 34,413 36,327

Lead.The lead production was for many years, as already mentioned, limited to two districts near the Mississippi,—one, the so-called “Upper Mines,” covering an area of 3000 to 4000 square miles included within the States of Wisconsin, Iowa, and Illinois; the other, the “Lower Mines,” in south-eastern Missouri. The yield of the Upper Mines reached its culminating point about 1845; and in 1852 it had fallen off to about 15,000 tons. That of the Missouri mines also fell off, so far as the south-eastern district was concerned, but the loss was more than compensated by discoveries of ore in south-western Missouri, and later in the adjacent State of Kansas. The production of all these districts (the lead of which contains but a trace of silver) for the years 1880-1884 is given by Mr E. A. Caswell as follows:—24,700 tons in 1880, 27,470 in 1881, 25,900 in 1882, 19,300 in 1883, and 17,567 in 1884. A considerable portion of the recently renewed activity in the Missouri mines is due to the increased utilization of the zinc ores associated with the galena. The numerous lead mines opened and worked in various States in the Appalachian region, from Maine to North Carolina, have nearly all proved unsuccessful. Yet, on the whole, the country has largely increased its product—a result due, chiefly, to the discovery and successful working of various lead ores containing silver in sufficient quantity to pay for separation, in several of the Cordilleran States and Territories. The total yield was 87,340 tons in 1880, 104,540 in 1881, 118,650 in 1882, 128,515 in 1883, 124,908 in 1884, 115,546 in 1885, distributed in 1883-84 as follows (Table XV.):—

 1883. 1884. Utah 25,900 25,000 Nevada 5,350 3,570 Colorado 63,000 56,400 Montana 4,500 6,250 Idaho 5,350 6,700 New Mexico 2,100 5,350 Arizona 1,350 2,400 California 1,500 1,420 Mississippi Valley 19,300 17,567 Virginia 190 220

To the non-metallic mineral substances mined or quarried in the United States, apart from coal and petroleum, which have been already considered, only very brief reference can here be made. The value of the lime and building-stone used in the country in 1885 was, for each of these articles, at least as great as that of the petroleum. Coal alone constitutes nearly seven-tenths of the value of the non-metallic minerals mined, and coal, petroleum, building-stone, and lime together make up almost nineteen-twentieths of the same total. Other important articles are salt (value $4,825,345 in 1885), cement ($3,492,500), phosphate-rock ($2,846,064), and limestone for flux in iron manufacture ($1,694,656).

Salt.The utilization of the brine springs of New York and Virginia was begun towards the end of the last century, and has become extensively developed. To this development has been added that of similar saline resources in Ohio and Michigan. Previous to this, however, some salt had been made by the evaporation of sea-water at points along the coast, and especially in the neighbourhood of Massachusetts Bay; and the census of 1880 showed still six establishments of this kind in existence, producing nearly 10,000 bushels of salt per annum. In California the evaporation of sea-water has attained some importance (884,443 bushels in 1880), the climatic conditions being much more favourable than along the northern Atlantic. A very small amount of salt was also made on the Florida coast. There are various lakes and partially or wholly dried-up beds of former lakes in the Cordilleran region, which are capable of furnishing a large quantity of salt, and some of these have been worked to a limited extent for use in metallurgy. By far the larger portion of the salt manufactured in the United States comes from the evaporation of brine, obtained by boring. The produce of the four brine-producing States (in bushels of 56 ℔) is given as follows in the census report for 1880:—Michigan, 12,425,885; New York, 8,748,203; West Virginia, 3,105,333; Ohio, 2,650,301; total, 26,929,722; total production of all the States, 29,800,298. The New York salt region is in the centre of the State, near Syracuse. The brine is obtained in a detrital deposit, varying in size from the coarsest gravel to the finest sand, which fills depressions in the Onondaga or Salina shales, to the depth in places of 300 or 400 feet. These shales are of Upper Silurian age. No rock-salt has been struck by boring; but farther west, at various points in New York, Canada, and Michigan, the presence of large bodies of salt has been proved by the aid of the diamond drill. As yet, however, this source of supply has only been utilized to a limited extent. The salt of Michigan comes from several distinct geological horizons. The uppermost one is in the Coal-measures, the next lower in the Lower Carboniferous, and the lowest in the Onondaga or Salina salt group, the salt-bearing formation of New York. It is only in the last-named formation that rock-salt has been found in Michigan, namely, at Bay City, on Saginaw Bay, where, at a depth of 2085 feet, a bed of salt 115 feet in thickness was bored through. In the Saginaw valley salt can be produced more cheaply than anywhere else in the country; the business of salt-making is associated with that of sawing lumber, and the refuse of the saw-mills feeds the fires under the salt-pans.

The following table (XVI.) shows the value of the metallic products of the United States for the years 1882, 1883, 1884, and 1885, as reported by the chief of the Division of Mining Statistics of the United States Geological Survey:—

 1882. 1883. 1884. 1885. Pig-iron $106,336,429$91,910,200 $73,761,624$64,712,400 Silver 46,800,000 46,200,000 48,800,000 51,600,000 Gold 32,500,000 30,000,000 30,800,000 31,301,000 Copper 16,038,091 18,064,807 18,106,162 18,292,999 Lead 12,624,550 12,322,719 10,537,042 10,469,431 Zinc 3,646,620 3,311,106 3,422,707 3,539,856 Mercury 1,487,042 1,253,632 936,327 979,189 Nickel 309,777 52,920 48,412 191,753 Antimony 12,000 .. .. .. Platinum 600 600 450 187 Aluminium .. 875 1,350 2,550 $219,755,109$203,116,859 $186,414,074$181,589,365

The value of the iron is the spot value; that of the gold and silver the coining value; that of the copper, lead, and zinc the value at New York; and that of the mercury the value at San Francisco.

The total values of the mineral products of the United States for the same years are returned on the same authority as amounting respectively to $235,461,580,$249,049,889, $226,800,674, and$246,931,991. (J. D. W.)

1. Under this head no portion of the Great Lakes is included.
2. The first person to describe topographically the Appalachian system as a whole was Guyot, who, in 1861, published the results of several years' work among these mountains, giving the heights of numerous points obtained by the aid of the barometer, accompanied by generalizations in regard to the orography of the region, and a map (1 : 6,000,000), which was first published in Petermann's Mittheilungen (1860). The various geological surveys carried on by the States included in the Appalachian region have furnished a large mass of material both geographical and geological in regard to portions of the range; but, since this work has not, as a rule, been based on accurate topographical maps, most of it is rather of the nature of a reconnaissance than of a finished survey. The State of New Jersey is, however, publishing a map based on an accurate triangulation, but on a small scale (1 : 63,360). The first and second geological surveys of Pennsylvania have also done much to extend our knowledge of the topography of the most complicated and interesting portion of the whole Appalachian region, but as we go either south or north from this central region we find the data less and less complete, and in regard to the extreme south-western portion of the Appalachian range we are little if at all in advance of Guyot. What is most lacking at present is an accurate map of that portion of the system which lies to the south of the region embraced within the State of Pennsylvania.
3. The Adirondack region—or Adirondack Wilderness, as it is often called—has been more or less completely covered by a topographical survey made by authority of the State of New York; it was begun in 1872, but the final results, in the form of a map, have not yet been published.
4. Hotchkiss, Virginia, a Geographical and Political Summary, based on information obtained during the Geological Survey by Prof. W. B. Rogers in 1835-41.
5. The total area (excluding Alaska) is 3,025,600 square miles,—made up by the addition of unorganized territory (5740), Delaware Bay (620), and Raritan Bay, &c. (100). The last column shows the population per square mile of the land surface, which amounts in all to 2,970,000 square miles.
6. The name Cordilleran is preferred for the Western division, because thereby any confusion is avoided which might arise from the fact that the people of the eastern States are still more or less inclined to call any portion of the region lying to their west by that name. No grouping in which all the States and Territories are included can be entirely satisfactory; but in that here suggested they are, while geographically connected, in most respects pretty closely allied to each other by their physical, climatic, and agricultural characters.
7. Throughout this article, by “ton” is understood the ton of 2240 ℔, unless the contrary is expressly stated.
8. This metal had also been extensively mined in the Lake Superior region long before the first visit of the English.
9. The substance of the following paragraph is mainly derived from a paper read by Mr C. A. Ashbarner, of the Pennsylvania Geological Survey, at a meeting of the American Institute of Mining Engineers in September 1885.
10. Paper read before the American Institute of Mining Engineers, Oct. 1886.
11. See Census Report for 1880 (1885).