The investigations made by this division of the Geological Survey during the last two years have been pursued with the object of increasing our knowledge of the physical and historical geology of the West and have had little relation to economic interests. The field of labor is one of the most impressive and instructive in the world—impressive by reason of the magnificent scale on which certain processes of nature have operated, and instructive because the causes, methods, and results of those processes are revealed with a distinctness which is unparalleled. This field comprises the Grand and Marble Cañons of the Colorado and the regions which drain into them. To the entire tract, comprising an area of more than 13,000 square miles, I have given the name of the Grand Cañon District.

The lessons which the geologist finds in this district are many, but the most conspicuous one embraces those subjects which are included under the nearly synonymous names “ Land Sculpture, “ “ Denudation, “ “ Erosion. “ These processes operate upon the land unceasingly, carving out mountains and valleys and giving shape and character to the earth's surface. They represent the work done upon the land by the winds and rains, by flowing water, by the chemical reactions of the atmosphere and of organic life. These processes are operative almost everywhere, and their results in the lapse of immense periods of time attain magnitudes, the statement of which may astonish the ordinary reader and perhaps excite his incredulity, but which at length appear veritable when tested by geological research and deduction. In no other portion of the world are the natural laws governing the processes of land sculpture exemplified so grandly; nowhere else are their results set forth so clearly. The interest excited by the grandeur of the subjects is intensified, and the value of the lessons enhanced, by the exceptionally intelligible manner in which their materials are presented for study.

For convenience of geological discussion Professor Powell has divided that belt of country which lies between the meridian of Denver, Colo., and the Pacific and between the 34th and 43d parallels into provinces, each of which possesses topographical features which distinguish it from the others. The easternmost he has named the Park Province. It is situated in the central and western parts of Colorado and extends north of that state into Wyoming and south of it into New Mexico. It is preeminently a mountain region, having several long ranges of the second order of magnitude. The structure and forms of these mountains are not exactly similar to those of any other region now well known, but possess some resemblance to the Alps, though not a very close one.

As we pass westward of this ranges in Colorado we enter, near the western boundary of that state, a region having a very different topography. The mountains disappear almost wholly, and in their stead we find platforms and terraces nearly or quite horizontal on their summits or floors and abruptly terminated by long lines of cliffs. They lie at greatly varying altitudes, some as high as 11,000 feet above the sea, others no higher than 5,000, and with still others occupying intermediate levels. Seldom does the surface of the land rise into conical peaks or into long narrow crested ridges; but the profiles are long, horizontal lines suddenly dropping down many hundreds or even two thousand feet upon another flat plain below. This region has been very appropriately named, by Powell, the Plateau Province. It occupies a narrow strip in the extreme western part of Colorado, a similar strip of western New Mexico, a large part of southern Wyoming, and rather more than half of Utah and Arizona.

West of the Plateau Province is the Great Basin, so named by Fremont because it has no drainage to the ocean. Its topography is wholly peculiar and bears no resemblance to either of the two just alluded to. I contains a large number of ranges, all of which are very narrow and short, and separated from each other by wide intervals of smooth, barren plains. The mountains are of a low order of magnitude for the most part, though some of the ranges and peaks attain considerable dimensions. Their appearance is strikingly different from the noble and picturesque outlines displayed in Colorado. They are jagged, wild, and ungraceful in their aspect, and, whether viewed from far or near, repel rather than invite the imagination.

The Wasatch, however, is an exception. This noble range is properly a part of the Basin Province, and is one of the finest and most picturesque of the West, but so completely does it contrast with the other Basin ranges that it may be regarded as an anomaly among them. The topographical features of this region are also found outside of the limits which Frémont assigned to the Great Basin, and reached southward into Arizona and northward into Idaho and Oregon. The Basin proper covers the western part of Utah, nearly the whole of Nevada, and a small portion of southern Oregon and Idaho. Its western boundary is the base of the Sierra Nevada.

No attempt will be made here to characterize the Sierra Nevada, partly because it is not thoroughly understood, but especially because it is remote from the region, here to be discussed, and presents few considerations essential to that discussion. The Grand Cañon District is a part of the Plateau Province, and to this province as a whole we may now devote our attention.

As already indicated, it lies between the Park and Basin Provinces, and its topography differs in the extreme from those found on either side of it. It is the land of tables and terraces, of buttes and mesas, of cliffs and cañons. Standing upon any elevated spot where the radius of vision reaches out fifty or a hundred miles, the observer beholds a strange spectacle. The most conspicuous objects are the lofty and brilliantly colored [Fig. 2.—Butte of the Cross. Trias.] cliffs. They stretch their tortuous courses across the land in all directions, yet not without system; here throwing out a great promontory, there receding in a deep bay, and continuing on and on until they sink below the horizon or swing behind some loftier mass or fade out in the distant haze. Each cliff marks the boundary of a geographical terrace and marks also the termination of some geological series of strata, the edges of which are exposed like courses of masonry in the scarp-walls of the palisades. In the distance may be seen the spectacle of cliff rising above and beyond cliff, like a colossal stairway leading from the torrid plains below to the domain of the clouds above. Very wonderful at times is the sculpture of these majestic walls. There is an architectural style about it which must be seen to be appreciated. The resemblances to architecture are not fanciful or metaphorical, but are real and vivid; so much so that the unaccustomed tourist often feels a vague skepticism whether these are truly the works of the blind forces of nature or of some intelligence akin to the human, but far mightier; and even [Fig. 3.—A lateral cañon. Escalante.] the experienced explorer is sometimes brought to a sudden halt and filled with amazement by the apparition of forms as definite and eloquent as those of art. Each geological formation exhibits in its cliffs a distinct style of architecture which is not reproduce among the cliffs of other formations, and these several styles differ as much as those which are cultivated by different races of men.

The character which appeals most strongly to the eye is the coloring. The gentle tints of an eastern landscape, the pale blue of distant mountains, the green of vernal or summer vegetation, the subdued colors of hillside and meadow, are wholly wanting here, and in their place we behold belts of brilliant red, yellow, and white, which are intensified rather than alleviated by alternating belts of gray. Like the architecture,



the colors are characteristic of the geological formations, each series having its own group and range of colors. They culminate in intensity in the Permian and Lower Trias, where dark, brownish reds alternate with bands of chocolate, purple, and lavender, so deep, rich, and resplendent that a painter would need to be a bold man to venture to portray them as they are.

The Plateau country is also the land of cañon, in the strictest meaning of that term. Gorges, ravines, cañadas are found and are more or less impressive in every high region; and in the vernacular of the West all such features are termed cañons, indiscriminately. But those long, narrow, profound trenches in the rocks, with inaccessible walls, to which the early Spaniards gave the name of cajon or cañon, are seldom found outside the plateaus. There they are innumerable and the almost universal form of drainage channels. Large areas of the Plateau country are so minutely dissected by them that they are almost inaccessible, and some limited though considerable tracts seem wholly so. Almost everywhere the drainage channels are cut from 500 to 3,000 feet below the general platform of the immediate country. They are abundantly ramified and every branch is a cañon. The explorer upon the mesas above must take heed to his course in such a place, for once caught in the labyrinth of interlacing side gorges, he must possess rare craft and self-control to extricate himself. All these drainage channels lead down to one great trunk channel cleft through the heart of the Plateau Province for eight hundred miles— the chasm of the Colorado, and the cañons of its principal fork, the Green River. By far the greater part of these tributaries are dry during most of the year, and carry water only at the melting of the snow and during the brief periods of autumnal and vernal rains. A very few hold small, perennial streams, coming from the highlands around the borders of the province, and swelling to mad torrents in times of spasmodic floods.

The region is for the most part a desert of the barrenest kind. At levels below 7,000 the heat is intense and the air is dry in the extreme. The vegetation is very scanty, and even the ubiquitous sage ( Artemisia tridentata ) is sparse and stunted. Here and there the cedar ( Juniperus occidentalis )[1]* is seen, the hardiest of arborescent plants, but it is dwarfed and sickly and seeks the shadiest nooks. At higher levels the vegetation becomes more abundant and varied. Above 8,000 feet the plateaus are forest-clad and the ground is carpeted with rank grass and an exuberant growth of beautiful summer flowers. The summers there are cool and moist’ the winters severe and attended with heavy snow-fall.

[Fig. 4.—The Water Pocket Canon.]

The Plateau Province is naturally divided into two portions, a northern and a southern. The dividing barrier is the Uinta range. This fine mountain platform is, in a one respect, as anomaly among the western ranges. It is the only important one which trends east and west. Starting from the eastern flank of the Wasatch, the Uintas project eastward more than 150 miles, and nearly join perpendicularly the Park ranges of Colorado. Of the two portions into which the Plateau Province is thus divided, the southern is much the larger. Both have in common the plateau features; their topographies, climates, and physical features in general, are of similar types, and their geological features and history appear to be closely related. But each has also its peculiarities. The northern portion is an interesting and already celebrated field for the study of the Cretaceous strata and the Tertiary lacustrine beds. The subjects which it presents to the geologist are most notably those which are embraced under the department of stratigraphy—the study of the succession of strata and co-related succession of organic life. Otherwise the region is tame, monotonous, and unattractive. The southern portion, while presenting an abundance of material for stratigraphical study, and in this respect fully rivaling, and perhaps surpassing, the northern portion, also abounds in the grandest and most fascinating themes for the student of physical geology. In respect to scenery, the northern portion is almost trivial, while the southern is the sublimest on the continent. With the former we shall have little to do; it is the latter which claims here our exclusive attention.

The southern part of the Plateau Province may be regarded as a vast basin everywhere bounded by highlands, except at the southwest, where it opens wide and passes suddenly into a region having all the characteristics of the Great Basin of Nevada. The northern half of its eastern rim consists of the Park ranges of Colorado. Its northern rim lies upon the slopes of the Uintas. At the point where the Uintas join the Wasatch, the boundary turns sharply to the south, and for 200 miles the High Plateaus of Utah constitute the elevated western margin of the Province.

It is from the summits of the High Plateaus that we gain our first comprehensive view of those grand facts which are the principal subjects of this discourse. But let me first ask the reader to endeavor to frame some conception, however crude, of three lines, each 200 miles long, placed in the positions of three sides of a square; the fourth side being for the moment neglected. Upon the eastern side conceive the Park ranges of Colorado; upon the northern, the Uintas; and upon the western side the southern portion of the Wasatch and the High Plateaus of Utah; and all these highlands having altitudes ranging from 9,000 to 12,000 feet above the sea, while the included area varies from 5,000 to 7,000 feet high. The space thus partially bounded may represent the northern part of the southern Plateau Province. Along the line required for the fourth and south side of the complete square there is no boundary. The topography continues on beyond it to the southward, and also widens out both west and east and overspreads an additional area more than twice as great as that already defined. From the eastern crests of the High Plateaus we may obtain an instructive overlook of the northern portion of the southern Plateau country.

The easiest line of approach is from Salt Lake City. Proceeding south from that town along the western base of the Wasatch, we reach the southern end of that fine range about 90 miles from Salt Lake. The last mountain pile is Mount Nebo, and skirting around its southern flank we soon perceive to the southeastward a long and very lofty ridge 20 to 30 miles distant. This is the Wasatch Plateau, the northernmost member of the group of High Plateaus. It has nothing in common with the Wasatch Mountain range, being wholly disconnected from it and standing with a wide interval en échelon to the southeastward of it. The Watsatch Plateau presents a long, straight, horizontal summit projected against the sky without peaks or domes, resembling somewhat the ridges of Pennsylavania and Virginia, but on a grander scale. We perceive along its entire western front a rapid slope, descending to the bottom of the San Pete Valley at its foot. It is not deeply incised with ravines and amphitheaters, nor notched with profound transverse gorges, as are ordinary mountain ranges, but shows a slightly diversified slope in every part. As we draw nearer we begin to see the attitudes of the strata composing its mass, or, as the geologists say, its “structure.” The strata are inclined at the same angle as the slope of its flanks. In the valley below, the beds are horizontal; as they approach the base of the plateau they flex upwards and ascend the slope; a they reach the summit they flex back to horizontality. If we ascend the plateau and ride eastward a very few miles, there suddenly breaks upon the view a vast and impressive panorama. From an altitude of more than 11,000 feet the eye can sweep a semicircle with a radius of more than 70 miles, and reach far out into the heart of the Plateau country. We stand upon strata of Lower Tertiary age, and beneath our feet is a precipice leaping down across the leveledges of the beds upon a terrace 1,200 feet below. The cliff on which we stand stretches far northward into the hazy distance, gradually swinging eastward and then southward through a course of more than a hundred miles, and vanishing below the horizon. It describes, as we well know, a rude semicircle, around a center a bout 40 miles east of our standpoint. At the foot of this cliff is a terrace of greatly varying width, rarely less than 5 miles, consisting of Upper Cretaceous beds nearly but not quite horizontal. They inclined upwards towards the east at angles rarely so great as 3°, and are soon cut off by a second cliff plunging down 1,800 feet upon Middle Cretaceous beds. This second cliff describes a semicircle like the first, but smaller and concentric with it. from its foot the strata still rise gently towards the east, through a distance of about 10 miles, and are cut off as before by a third series of cliffs concentric with the first and second. For the fourth and fifth time this process is repeated. In the center of these girdling walls is an elliptical area about 40 miles long and 12 to 20 miles broad, completely surrounded by mural escarpments more than a thousand feet high. This central spot is called the San Rafael Swell , and it is full of interest and suggestion to the geologist. From its central point the strata dip away in all directions, the inclinations, however, being always very small.[2]* This configuration of the strata (dipping away from a central point in all directions) is technically termed “quaquaversal.”

The accompanying diagram (Plate XI) shows the relative masses and positions of the strata as they would appear in vertical sections cutting [U. S. GEOLOGICAL SURVEY. Sections from San Pete and Sevier Valleys across the Wasatch Monoclinal to the San Rafael Swell. ANNUAL REPORT 1881. PL. XI.] east and west through the Wasatch Plateau to the San Rafael Swell. It will be observed that the lower Tertiary is found only on the summit and western flank of the plateau. The Cretaceous extends further out, but is at last cut off in turn; and as lower and lower beds are exposed to daylight, they too are similarly cut off until the summit of the Carboniferous is nearly or quite exposed within the swell itself. The approximate length of the section here given is about 55 miles, and the thickness of the strata from the summit of the Carboniferous to the top of the Lower Tertiary is nearly 11,000 feet.

The geologist who becomes aware through observation of the general facts thus set forth quickly reaches the following conclusion: The beds which are successively terminated in the terrace cliffs once reached further eastward, and in all probability every one of them extended in full volume and without a break entirely over the locus of the swell to regions far beyond it. Upon the eastern side of swell, and at varying distances from it, the missing strata reappear in inverse order, with terminal cliffs facing the westward. From the intervening space they have been swept away by erosion.

In restricted localities of a few square miles, in a river valley, in the open glades of a hill-country, the most unscientific observer maybe easily convinced that the waste of thousands of years has broken the continuity of the strata and quarried away large masses of rock. But in the wide expanse before us even the mind of the geologist may falter before accepting a conclusion so portentous. The magnitude of the work is oppressive, and cautious philosophers are reluctant to take up and carry the burden of unusually large figures. They prefer to cast about in order to see whether some easier conclusion may not discovered. The one already stated is to the effect that a body of strata more than 10,000 feet thick and more than 500 square miles in area have been swept off from the surface of the swell; that nearly 9,000 feet have been removed from a much larger annular space around it; 7,000 feet from a still larger and remoter space; and so on with expanding annuli, from which successively decreasing amounts have been denuded. It is needless to define just here the limits of the denuded region, even if it were possible. It is sufficient to say that its extent is much more than 10,000 square miles, and that the thickness of the strata removed varies from a few hundreds to more than 10,000 feet. Nor does the conclusion stop here. The San Rafael region is only one of the considerable number of the subdivisions of the Plateau Province where the same enormous extent of erosion has taken place. It is not the largest of those subdivisions, nor is the thickness of the removed strata the greatest there. It is merely an example, and whatsoever it reveals in regard to erosion is but a group of events common to the entire southern province with its vast area of nearly 100,000 square miles. I have selected it for discussion because its array of facts in evidence is more easily handled and can be more lucidly presented than those of the other subdivisions. Let us, then, examine in detail the arguments upon which this deduction of a great denudation is based.

If we stand before one of the great marginal cliffs which bound the several terraces, we shall speedily detect abundant evidence that time and the elements are slowly robbing its face of the materials which compose its mass. Fragments have spawled off and fallen, and they now lie at its base in great quantities, forming a talus. Cliff and talus alike are seamed and scored with rain-gullies, and if we are fortunate enough to observe the effects of a shower we shall see the waters trickling, spouting, or rushing down through every seam and gully, carrying sand, mud, and small fragments with which they are charged to their utmost capacity. The meaning of this is that the cliff is wasting away, and its locus, through the ages is farther and farther back. This backward movement of the line of frontage by slow waste is very happily named by Powell the Recession of Cliffs.

It may seem at first as if the rate of recession must be so exceedingly slow that when we are asked to consider the possibility of a recession of thirty or forty miles the argument would break down under the weight of its time-factor. But it will be shown hereafter that in the total process of denudation the rate of recession is rapid enough to satisfy the temper of such geologists as may be parsimonious or even very stingy in their allowances of time. It is sufficient here to advert to the very obvious fact that the cliffs are receding, and that at some former geological period they once stood nearer the center of the denuded district. Now, it is sufficiently obvious that if we allow the imagination to range back indefinitely into the past and reverse the process of recession, restoring the material which has been denuded, the continuity of argument will at length bring us to an epoch in which the cliffs which now face the center of the San Rafael Swell came together, and the strata which those cliffs now terminate stretched unbroken from west to east across the whole width of the Plateau Province. It is only a question of time and continuity of the process. The geologist may, however, raise a very pertinent inquiry. Admitting that the cliff-bound strata once reached out in advance of their present limits, may they not have grown thinner as they approached the center; may they not have attenuated rapidly so that their former thickness over the swell was but a small fraction of the aggregate thickness disclosed in the present escarpments? May not the higher beds have thinned out and disappeared entirely a few miles from their present boundaries? In all other well-studied regions it is a general and almost universal rule that the strata vary greatly in thickness when traced from place to place, and attenuate as they extend away from their shore-lines. May they not have done so here?

Answering the questions directly, it may be said that the Permian, Trias, and Cretaceous certainly did not grow perceptibly thinner as they approached the center of denudation. The Jurassic did thin out quite notably from west to east, and it is possible that the Tertiary may have thinned a little; but this loss of thickness in the Jurassic and Tertiary has been abundantly discounted in the estimate given of the mass of strata denuded. As regards the general rule that strata vary greatly in thickness, it may be stated that the Plateau country is a remarkable exception to it. One of the most striking features in its stratigraphy is the wonderful persistency with which its formation maintain their volumes and lithological features over great areas. In this respect the province has no parallel, not even in the calm and undisturbed terrains of the Mississippi Valley.

Further support of this conclusion may be found by reverting to the section (Plate XI). On the eastern side of the swell the section shows a great monoclinal flexure where the strata extending eastward rapidly bend downward and subsequently flex back to horizontality. Before this flexure began to form, the Cretaceous strata had already been deposited. Possibly, also, the Tertiary had been laid down, but of this we are not as yet certain. But we know that it was formed after the Cretaceous age, for the strata abundantly betray it. If we could bend back the strata now inclined upward in that flexure, we should have a wall about 8,000 feet of more, looking down from the east upon the central amphitheater, and in that wall would appear the broken edges of the Permian and Mesozoic beds, though the upper part of the Cretaceous and Tertiary would be wanting at the summit. Thus nearly four-fifths of the denuded strata appear upon the eastern side of the swell, in very close proximity to it, and the remainder make their appearance at varying distance beyond. There is no appreciable loss of volume in the exposed beds of the monocline as compared with the corresponding beds to the westward.

But we may with advantages pursue the tasks of restoring the beds to the position they held during the period of their deposition by straightening out or bending back the strata in those parts where they have been tilted and flexed since their accumulation. This is readily done here. They were deposited originally in layers which were quite horizontal. We know this by reasoning upon the following facts. From the summit of the Permian, and I think we may say quite confidently from the summit of the Carboniferous upwards, the whole series was deposited in very shallow waters. The evidence of this is overwhelming. We find proof that the surfaces of deposition throughout Mesosoic time oscillated repeatedly a little below and a little above seal-level. The cross-bedded sandstones of the Trias and Jura, the sandy shales wonderfully ripple-marked, the occurrence of bands containing the silicified remains of forest trees, the occasional recurrence of contacts showing “unconformity by erosion’ without any unconformity of dip,[3]* the occurrence of brackish-water types of mollusca in the Jurassic, the lignites, fossil leaves, and carbonaceous shales of the whole Cretaceous system, the brackish-water fossils of the lowest Tertiary, leave no doubt as to the verity of the foregoing inference. The final restoration, then of the strata to their original positions leaves them horizontal.[4]*

If we draw a section of the strata restored to horizontality, we shall find that the strata now remaining require, in order to perfect their continuity, the restitution of large masses fully equal to those which we have inferred to have been swept away by erosion. Any hesitation to do this would leave us without resource. Any other hypothesis, so far as I can conceive, would be not only without support in the facts presented, but in opposition to their entire tenor and purport.

The geologist who is familiar through long field-study with the physical problems presented in the West would not need further argument to become satisfied of the reality of the great erosion here inferred. Perhaps he would consider that too much has been said in support of it already; especially since the subject of this paper is not the San Rafael but the Grand Canñon district. But I have devoted so much discussion to the San Rafael district because it is a type of a congeries of districts which make up the Plateau Province, and because it exemplifies in the most intelligible, compact, and complete manner the broad facts and laws which are to engage our attention hereafter. These facts and laws apply to the Grand Cañon district; but to take the facts there presented and arrange them in a clear view before the mind of one who has never visited that region, and make them definite and convincing, would be extremely difficult without preparatory exercises on problems similar in kind but simpler in form. For this reason I proposed, before leaving the San Rafael district, to bring out another category of facts which it exemplifies. They involve a generalization very interesting in itself, and of the greatest utility in solving many problems presented in all parts of the Plateau country. This generalization—or law in the sense of an observed order of facts—may be called the Persistence of Rivers.

The rivers of the Atlantic States, from the Hudson southward, cut through the Appalachian ridges by narrow gorges, or gaps, which seem to have been quarried out for the purpoes. Geology, however, does not take account of “purposes” or “design,” but seeks its explanation in “natural” causes alone. It asks by what natural processes were those gorges made?

The answer it finds is, that the rivers themselves scoured them out, and that secular decay has widened them somewhat. A reader not versed in geology might be led to ask a further question. How can a river attack a mountain wall, or even a gentle declivity, and quarry though it a pathway giving a continuous descent for the flow of its waters? The reply is that no river ever does that. To understand how it all came about we must go back to the beginning. The rivers were born with the country itself. The land emerged from the sea; and when it emerged the rains or melting snow sought whatever channels were determined by the slight inequalities of the newly-risen surface and flowed seawards. These lofty ridges, gashed with noble ravines, had then no existence. The rivers are older than the mountains. As time ran on the mountains grew upward, athwart the courses of the streams. But a flowing river has a power to fight for and maintain its right of way, which becomes apparent only when we have carefully studied and analyzed it. This power is inherent in the descent of its waters—is literal water-power. The weapons or tools are the sand, gravel, and silt which the waters carry, and which act after the manner of a sand-blast, except that in the sand-blast the grit is impelled by air or steam, while in the river it is impelled by water. This power, inherent in the fall, increases rapidly as the fall increases. When the declivity is feeble the power to grind down the channel—to “corrade,” as Powell terms it—is correspondingly feeble, or even annihilated. When a barrier like a ridge rises across the track of a stream the declivity is increased at that point. Increased velocity and corrasive power is at once developed in the stream, and it cuts down the barrier. Perhaps a lake may be formed above the barrier, but its outlet will be cut down and the lake drained.

In a low country the slopes are, with rare exceptions, feeble, and this corrasive power by which the stream maintains its locus is in such countries correspondingly feeble. Here we may expect to find many cases where streams have been deflected largely from their courses; but in a high country the reverse is the case. In a region newly risen from the waters the positions of the streams may be very inconstant; but as the elevation increases they gradually fasten their grip upon the land and hold it.

It would be difficult to point out an instance where a great river has ever existed under conditions more favorable to stability of position than those of the Colorado and its tributaries. Since the epoch when it began to flow it has been situated in a rising area. Its springs and rills have been among high mountains, and its slope since the earliest period of its history has always been great. The relations of its larger tributaries have, in these respects, been the same; and indeed the river and its tributaries have been a system, and not a mere aggregate; for the latter are dependent upon and responsive to the physical conditions of the former. And now we come to the point. The Colorado and its tributaries run to-day just where they ran after the region emerged from the waters. Since that time mountains and plateaus have risen across their tracks, whose present summits mark less than half their total amounts of uplift. The rivers have cleft them to their foundations.

The Green River, passing the Pacific Railway, enters the Uinta platform by the Flaming Gorge, and after reaching the heart of the chain turns eastward parallel to its axis for 30 miles, and then southward, cutting its way out by the splendid cañon of Lodore. Then following the base of the range for a few miles a strange caprice seizes it. Not satisfied with the terrible gash it has inflicted upon this noble chain, it darts at it viciously once more, and entering it, cuts a horseshoe cañon in its flank 2,700 feet deep, and emerges near the point of entrance; thenceforward, through a tortuous course of 300 miles, it flows southward through gently inclined terraces, which rise slowly as the river descends. Along this stretch it runs almost constantly against the dip of the beds, cutting through one after another, beginning with the Upper Eocene until its channel is sunk deep in the Carboniferous. Further down, the Kaibab Plateau rose up to contest its passage, and a chasm 5,000 to 6,000 feet deep is the result. It is needless to multiply instances; the whole province is a vast category of instances of river channels cutting through plateau, mesas, and terraces where the strata dip up stream. The courses of the cañons are everywhere laid independently of the topographical inequalities, whether these inequalities be due to the broader features of land sculpture or to displacement and unequal uplifting. On the north and west side of the Colorado the tributaries generally run counter to the structural slopes; on the east and south sides, they ran more nearly with them.

It is clear then that the structural deformations of the surface, the uplifts and downthrows had nothing to do with determining the present distribution of the plateau drainage. The rivers are where they are, in spite of them. As irregularities rose up, the streams turned neither to the right or to the left, but cut their way through in the same old places. The process may be illustrated by a feeble analogy with the saw-mill. The river is the saw and the rising strata are the timber which is fed against it. The saw-log moves while the saw vibrates in its place. The river holds its place as rigidly, and the rising strata are dissevered by its ceaseless wear. What, then, did determine the situations of the present drainage channels? The answer is that they were determined by the configuration of the surface existing at, or very soon after, the epoch of emergence. Then, surely, the water-courses ran in conformity with the surface of the uppermost (Tertiary) stratum. Soon afterwards that surface began to be deformed by unequal displacement, but the rivers had fastened themselves to their places and have ever since refused to be diverted.

[U. S. GEOLOGICAL SURVEY. HORSESHOE CAñON, GREEN RIVER. From Powell's Exploration of the Colorado River. ANNUAL REPORT 1881. PL. XII.]

This theorem is of great utility in the study of the Plateau Province, for it throws light upon many problems which would otherwise be obscure. The course of a river is the index of the slope, and, to a great extent, the configuration of the primitive unmodified surface of a tract. It betrays the amount of tilting or flexing which the strata have undergone, and also conveys information as to the amount of strata which have been denuded. This information, however, is in many cases incomplete, but when placed in relation with other facts it frequently becomes conclusive. The application of the theorem to the San Rafael district is a beautiful instance of its validity.

Across the San Rafael Swell extend two river channels, one crossing it near the northern and the other near the southern end. They head in the High Plateaus, and pass through the successive terraces in deep cañons; then crossing the swell, enter the high cliffs on the eastern side and flow on. The northern stream—the San Rafael River—ultimately joins the Green River; the southern one—Curtis Creek—enters the Frémont, River, a tributary of the Colorado. A glance at the map and an interpretation of the topography as expressed by the contours will quickly show that these streams are quite independent of the existing topography and could not have had their situations determined by it. They must have been laid out upon some ancient surface differing widely from the present. To find that surface is not difficult. It must have had a continuous descent, though doubtless of slight declivity, from the western margin of the province to the line of the Green and Colorado Rivers. We shall obtain precisely that surface configuration be reducing or bending back the flexures, and depressing the tilted strata until the Cretaceous beds are everywhere horizontal, and then filling up the gaps made in the continuity of the strata by erosion. Thus we shall reach, by argument from the persistence of rivers, the same conclusion which we reached by studying the effects of the recession of cliffs, and by the independent study of the displacements.

The example of erosion thus given by the San Rafael Swell illustrates, as a sharply defined type, the denudation of the Plateau Province. The thickness of the strata removed varies greatly in different portions. In the High Plateaus it has amounted to only a few hundred feet. In large areas it amounts to two or three thousand feet and in others of considerable extent it reaches more than 10,000 feet. Preliminary comparisons of known facts derived from nearly the entire extent of the southern province lead to the conclusion that on the average 5,500 to 6,000 feet of strata have been removed from its entire expanse. Our knowledge of the geology of some portions of it is at present very imperfect. Still, enough is known to justify us in believing that this summary estimate will not be much affected by future investigation.

We may for special purposes of convenience regard the province as consisting of districts or spots of maximum erosion separated from each other by high mesas or dividing platforms where erosion has been at its minimum. The San Rafael district may be regarded as one of these areas of which the central part is an area of maximum erosion while its peripheral parts are areas of minimum erosion. The Grand Cañon district is another, and there are still others which we need not here specify.

Before concluding the introductory part of this paper it will be desirable to recite briefly the succession of geological events which the study of the region has thus far brought to light, selecting only such as will hereafter be of special utility.

Throughout the great Carboniferous age the entire area of the Plateau Province was submerged beneath the ocean. Deposition of strata went on continuously. The thickness of the strata accumulated in that age appears to have varied greatly, and the deposits were laid down unconformably over the surface of a country which had been ravaged by a great erosion. Such exposures of the Carboniferous as now exist, however, exhibit for the most part a remarkable evenness of stratification. In the interior spaces of the province the beds are either horizontal, or if disturbed, give full evidence that the disturbances took place long after their deposition. The close of this age evidently left a subaqueous surface, which was exceedingly flat, and, except around the borders of the province, quite free, so far as we now know, from any appreciable inequalities.

The thickness of the Carboniferous system is from 4,500 to 5,000 feet in the interior of the province, but around its borders, and in the Uinta Mountains, it is sometimes found in far greater volume. Its strata consist of impure limestones, occasionally of enormous thickness in the individual beds, and alternating with fine-grained homogeneous sandstones. Extensive beds of gypsum also occur.

After the Carboniferous came the Permian age, in which were laid down from 800 to 1,500 feet of sandy shales. The stratification was wonderfully even and everywhere horizontal. The Permian beds are often ripple-marked and betray many evidences that they accumulated in shallow waters. Among these evidences are the appearances at several horizons of indications that for a time the sea-bottom was laid bare by the recession of the waters, or by the elevation of the platform itself; for we may discern evidences of slight erosion at the contacts of the beds. But the horizontally of the beds appears never to have been notably disturbed.

The same state of affairs continued through the Trias. There, too, we find evidence of alternations of emergence and submergence in the shape of slight unconformities by erosion, and in the occurrence of extensive remains of silicified forests. The Triassic series is composed almost wholly of sandstones, the only calcareous matter being thin seams of gypsum. The sandstone beds are very numerous and often shaly. They are usually of no great thickness individually, but there is one very notable member of which we shall see more when we come to view the Vermilion Cliffs.

Directly upon the Trias rests the Jurassic. A wonderful bed of sandstone 800 to 1,200 feet thick and very white and sugary in color represents the principal part of this series. It is a very notable formation because of its remarkable homogeneity, the persistent way in which it preserves its lithological characters through great distances, and the absence of divisional planes of stratification—the mass being solid from top to bottom. But most striking of all its wonderful cross-bedding, far surpassing in beauty, extent, and systematic character, any similar phenomenon elsewhere, with which I am acquainted. The summit of the Jurassic suddenly changes to calcareous and sandy shales, abounding in fossils. This series, as well as the Trias, appears to have been laid down horizontally in shallow waters.

Next comes the Cretaceous system—a mass of yellow sandstones with clayey and marly shales, aggregating from 4,000 to 5,000 feet thick. In this series we find an abundance of plant remains, many beds of good coal, and much carbonaceous shale. The conditions during the Cretaceous appear to have been quite similar to those which prevailed in the Appalachian region during the Carboniferous. Perhaps the conditions which attended and rendered possible the accumulation of coal are not sufficiently well understood to enable us to say confidently just what they were, but there seems to be general agreement that they involved a flat, low, moist country lying almost exactly at mean sea-level, and subject to alternate emergence and submergence. No other supposition seems to meet the requirements of the case, or to be capable of explaining how a mass of strata could be so accumulated, consisting of alternations of thin seams of coal and carbonaceous shale with layers of sand stone containing marine fossils.

We have now the following remarkable state of affairs. From the close of the Carboniferous to the close of the Cretaceous there is strong evidence that the surface of deposition was always very near to sealevel, sometimes a few feet above it, but for the most part a little below it. And yet in the interval about 9,000 feet of strata accumulated with remarkable uniformity over the entire province, and always in a horizontal position. From this it necessarily follows that the mass of material thus deposited sank or “subsided” at a rate which, in the long run, was exactly or sensibly equal to the rate of deposition.

At the close of the Cretaceous we find evidence that the long calm which had characterized the action of the physical processes was invaded. Some extensive disturbances took place, resulting at some places in the dislocation and flexing of the strata, and the elevation of some portions of the region to considerable altitudes. Erosion at once attacked the uplifted portions, and around the borders of the province we find numerous localities, usually not very extensive, which were greatly devastated. At some of these places the entire local Cretaceous series was denuded, and even a portion of the Jurassic; and the Tertiary is seen lying upon the Jurassic across the beveled edges of the flexed Cretaceous strata. But even these localities were again submerged, as the presence of the Tertiary fully attests. The disturbances were not general—did not extend to the entire province, but appear to have occurred around, or a little within, its marginal portions.

The last period of deposition was marked by the accumulation of the Eocene beds, which form such a striking feature in the stratigraphy of the peripheral parts of the Plateau country. Around the southern flanks of the Uintas their aggregate thickness exceeds 5,000 feet, but southward the upper members disappear, and 80 miles north of the Grand Cañon only about 1,000 to 1,200 feet, representing the lowest portion of the series, make their appearance. It is highly probable that the middle and upper portions of the Eocene were never deposited there. But the lowest beds, most probably, once covered the entire province, while the middle and late Eocene were confined to its more northerly portions. The lowest members were deposited in brackish water, as their fossils amply attest; but in the succeeding beds the fossil forms are entirely those which live in fresh water. From the epoch to the present time there has been no recurrence of marine conditions.

We now reach a turning point in the history of this region. That long continuance of marine conditions lasting from the beginning of Carboniferous time to the close of the Cretaceous came gradually to an end. The waters became brackish and then fresh. During the prevalence of the marine condition it seems to be a necessary conclusion that the waters which covered it had abundant access to the ocean. Whether its waters were wide open to the ocean, like the Gulf of Mexico or Hudson's Bay, or whether they formed a broad expause, with a comparatively narrow outlet, like the Mediterranean, we do not know, and it would be useless to conjecture at present. At all events the communication was sufficiently free to maintain a degree of saltness suitable to the existence of molluscan forms of the ordinary marine types. When the water became brackish, we infer that the straits became greatly narrowed; when they became quite fresh, we infer that the access of the ocean to the area was wholly shut off, and that the water brought by the rivers and rains outflowed, and the region became an inland lake of vast proportions. For the deposition still went on. Through Eocene time from 1,000 to 5,000 feet of lacustrine beds, containing an abundance of fresh-water fossils, were deposited. Among them are also found layers of coal and carbonaceous shales, and sandstones thickly imprinted with the traces of arboreal vegetation.

But at length the deposition of lacustrine strata ceased; not, however, at one and the same time in all parts of the province. The evidence indicates that in the southern and southwestern portions it stopped after about one-fourth or one-third of the Eocene horizons had been laid down. In the central portions it appears to have ceased after about one-half to two-thirds of those horizons had been deposited. In the northern portions, in the vicinity of the Uintas, the entire system of Eocene strata is found in immense volume. These facts lead us to infer that the great Eocene lake, soon after its water became quite fresh, began to shrink its area, and that its bottom became through a slow progression dry land. The southern and southwestern portions were the first to emerge; then the middle portions; the lake gradually retracting its boundary to the northward, until in the latter part of the Eocene it occupied a greatly diminished area in the vicinity of the San Rafael country and the southern base of the Uinta Mountains. At the close of the Eocene this remnant of the lake also disappeared.

We now reach another turning point in the history of the region. Hitherto and for an immense stretch of geological time it had been an area of deposition and of subsidence. It now became an area of elevation and denudation, and these processes have been in operation ever since. In the periods of deposition and subsidence, from the Carboniferous to the Eocene, both inclusive, the thickness of the strata accumulated varied from 14,000 to 20,000 feet, and the subsidence of the base of the Carboniferous was of nearly equal extent. In the periods of elevation and denudation these vast masses of strata rose bodily up again; the amount of elevation varying according to locality from 6,000 to 18,000 feet. The havoc wrought by erosion has been as already shown, stupendous; the thickness of strata removed exceeding 10,000 feet in some considerable areas, and averaging probably 5,500 to 6,000 feet over the entire province.

The points which it is desirable to notice in this chapter concerning the progress of the Tertiary and Quaternary erosion of the province, are few and of the broadest nature. In truth it is necessary to speak very guardedly. For while the most general features of the work have left well-marked traces which can be interpreted, yet when we come to details the vast erosion has swept away so much of its mass that a large portion of the evidence of the details has vanished with the rocks. There is reason to believe that the greater part of the denudation was accomplished in Miocene time. This was a period of slow but continuous uplifting, reaching a great amount in the aggregate, and it was most probably also a period of rapid erosion. The uplifting, however, was unequal in the different parts of the province. The comparatively even floor of the old lake was deformed by broad swells and plateaus rising above the surrounding country. As we shall see hereafter, the action of the denuding agents is much more vigorous and efficient upon the higher than upon the lower parts of a region; and consequently these up-swellings at once became the objects of special attention from the destroying forces and were wasted more rapidly than the lower regions around them. Here were formed centers or short limited axes, from which erosion proceeded radially outwards, and the strata rising gently towards them from all directions were beveled off. Thus were formed those areas of maximum erosion, already spoken of, and of which the San Rafael Swell is the most perfect and simplest type.

We have also reason to believe that the climate of the Miocene was moist and subtropical, conditions favorable under the circumstances to a rapid rate of erosion. We do not indeed find the proof of this in the province itself, for it contains no Miocene strata of fossils; but in surrounding regions the strata and fossils of that age are found in abundance, and they clearly indicate that the climate had that character; and it would be quite untenable to suppose that so limited a tract as the Plateau Province was an anomaly in respect to the climate of the broader regions of which it is a part, unless special reasons for it could be adduced. I know of no such special reasons. But near the close of the Miocene, or not long thereafter, the climate of almost the entire West underwent a change, becoming arid, as it is at present. In this change the Plateau country no doubt shared. The more important results of the Pliocene and Quaternary erosion, however, will be among the principal themes of the following chapters.

1* Botanists inform me that the predominant upland juniper of the Plateau Province, as the species are now distributed according to Dr. Engelmann's revision published in 1877, would be Juniperus California, var. Utahensis, rather than J. occidentalis, some of the varieties of which may, however, occur there. Until that revision was made the western junipers were little known, and several distinct species were indiscriminately classed as J. occidentalis.
2* Upon the eastern margin of the swell is one of those great “monoclinal flexures” with a high inclination so characteristic of the Plateau Country. These will be adverted to hereafter.
4* It would be very instructive, if space permitted, to elaborate this discussion of the original horizontally, and I am tempted to point out in the hastiest manner some obvious consequences of the deduction. It appears that if this deduction be true the deposits must have settled or subsided as rapidly (in the long run) as they were accumulated. The surface of deposition appears never to have varied much from sea-level. But the total accumulation of Permian, Mesozoic, and Tertiary beds was nearly 11,000 feet, and when the deposition ended (supposing that it ended in the Middle Eocene, though I think it more probably continued here until the close of the Eocene) the Permian must have sunken more than two miles below sea-level. The gradual subsidence of large bodies of sediment as they accumulate in strata is a fact now generally recognized, and is of universal application. That it is caused by the gross weight of the enormous masses of deposited material sinking into the yielding earth seems a most natural explanation.