THE Story of the Prairies OR THE LANDSCAPE GEOLOGY OF NORTH DAKOTA BY DANIEL E. WILLARD, A. M., Professor of Geology, State Agricultural College, Fargo, North Dakota. FIFTH EDITION. PRINTED FOR THE AUTHOR BY RAND, McNALLY & COMPANY, , CHICAGO. NEW YORK. LONDON. Copyright, 1907, BY DANIEL E. WILLARD, PREFACE. A book justifies its existence if it supplies a need or assists in any way in solving the problem of life. There is a noticeable lack of books suited to the general reader in the branch of science which deals with the earth upon which we live. Splendid contributions to knowledge have been made in this line in recent years, but many of the best things that have been written are practically inaccessible to the average reader both by reason of the technical character of the language used and by the fact that the material is often contained in large volumes unhandy for general use. That these contributions are of great value to the people is indicated by the large amounts which are annually expended by the National 'Government and by the State surveys for their compilation and publication. To present in untechnical language a scientific statement of a subject is not an easy task. Whether the present book accomplishes this or not an intelligent public will soon discover. The author has had in mind, as a class to whom he would make every page readable, those who have reached the degree of maturity represented by the sixth and seventh grades in the public schools. If the book is intelligible to pupils rep- resented by these grades it should be understood by the average citizen who is interested in knowing about his own State. It has seemed impos- sible to avoid the somewhat technical character of certain portions, owing to the intricate and difficult nature of the subject. It may be asked if these passages might not have been omitted. To do this would have marred the book as a whole, and it seemed best to carry out the original plan, leaving to the discretion of teachers what part should be omitted in class work. Such subjects as the causes of the changes of level of Lake Agassiz, the distribution of the lakes of the State, and the chapter on " The Beginnings of North Dakota," may be omitted where these topics are beyond the mental grasp of the pupils. But to have omitted them from the book would have left unanswered questions which the more advanced pupils in the high schools, and many general readers, will be certain to ask. It is the author's opinion that not enough attention is given in our schools to- instruction relative to the character and resources of our own State. Not enough attention to our own State is given by the teach- 808772 iv PREFACE. ers in their private studies, and not enough careful reading is done along this line by the average citizen. In geography instruction in our schools why do we need to go to South America and Asia and the uttermost parts of the earth for illustrations of land forms? Why do we need to study about the river systems, hills, plateaus, lakes, soils, and resources of states which are hundreds of miles away in preference to those of our own State ? When the child has gained a general idea of the earth as a whole, and of North America more partic- ularly, why should he be required to go to states and countries which are far away for concrete examples ? Have we not rivers, lakes, and marshes, hills, valleys, plateaus, and plains in our own State, which are more access- ible and just as real as those of other states ? Have we not types of land- scapes, developing river systems, desiccating lakes, mineral and forest problems ? Indeed, there are no better examples in the world. And the writers of text-books for the use of schools in the Eastern states are now coming to the far West, to the Red River Valley and elsewhere", for examples to illustrate the great principles of geographic science. The author has sought to reach three classes of readers in this book. The primary purpose has been to adapt the language and treat- ment to pupils in the higher grades of public schools. This purpose has been constantly kept in mind, for a large number of boys and girls to whom a knowledge of the resources of the State ought to be of the great- est value will never enter -the high school. While some portions are rather difficult for pupils of the grammar grades it is thought that the book as a whole will make a profitable half year's work in the high school, satisfying the requirement of the State Course of Study in Geology or Physiography. And it is hoped to reach a large class of readers who have entered the practical school of life, but who would be benefited by a fuller and more accurate knowledge of the character and resources of the State in which they live. It has been impracticable in a book of this character to give specific reference in the body of the text to the authors consulted, but the author wishes to give fullest credit for the use of this information. More than all others the author is indebted to the classic work of Prof. Warren Upham, "The Glacial Lake Agassiz," a monograph of the United States Geological Survey. This work leaves little to be added regarding the landscape geology of the Red River Valley and the adjacent portions of the State. He would be a bold student who would attempt to cover the field better than has been done in this comprehensive quarto of more than six hundred pages, but its very elaborateness renders it incon- venient for those to whom its contents should be of the greatest value. The author of this volume has drawn freely from Professor Upham's treatise, hoping to bring its vast fund of useful information within reach of the bus)'- citizen who would not be likely to read a larger work. PREFACE. V Through the very kind permission of Professor Upham a number of the illustrations in his work have been either redrawn and adapted to the purpose of the present work, or copied by the Bureau of Engraving at Washington. The author is also indebted to the valuable bulletins by Prof. J. E. Todd, of the University of South Dakota, for much that is here given regarding the Altamont and Gary Moraines and the landscape features connected with these in Logan and Mclntosh and adjoining counties, and also for facts regarding Lake Dakota in Dickey County, and for several illustrations. The valuable " Report of the State Geological Survey of North Dakota," by Prof. E. J. Babcock of the University of North Dakota, Grand Forks, has been drawn upon in the treatment of the coal deposits of the State, and the author wishes to express his appreciation for the kind permission to use several plates from this Report in the present work. The author acknowledges his personal indebtedness to Prof. War- ren Upham, Secretary of the State Historical Society of Minnesota ; to Prof. Charles M. Hall, of the Agricultural College of North Dakota ; Prof. E. J. Babcock, of the University of North Dakota ; and to Miss Lillian V. Lambert, Instructor in English in the East Side High School, Des Moines, Iowa, who read this book in manuscript, and who by their scholarly and valuable criticisms greatly increased its value. Acknowledgment is made to all those who have so kindly assisted in the preparation of the drawings which illustrate the book. The writer's thanks are particularly due to Miss M. Emma Davis and to Prof. Thomas H. Grosvenor, members of the faculty of the Mayville State Normal School. Many illustrations which needed to be drawn under the author's direction were made possible by their assistance. Pres. Joseph Carhart, under whose supervision the author has for several years had the pleasure of teaching, has given practical suggestions which have been of great value in the preparation of this volume. If this book serves the purpose of making the people of North Dakota better acquainted with their State, and thereby enlarges their appreciation of the opportunities which belong to them as citizens of this growing commonwealth, the author will feel that he is amply repaid for the labor which has been expended upon it. D. E. W. Mayville, North Dakota, State Normal School. May /, igos. PREFACE TO THE FIFTH EDITION, It is a source of gratification to the author that the demand for this work has made the fifth edition possible. The sales of the book prove that the public will read the literature of science provided it is expressed in simple terms and in readable form. The continued demand for the book for class use in schools, by teachers, and by the public generally, makes a new edition necessary, and affords an opportunity for the addition of new matter which the progress of geological investigations in the State, during the five years since the book first appeared, has. made possible. Subjects relating to some portions of the State could not be as fully treated in the earlier editions as it was then felt their importance required, owing to lack of sufficient knowledge. An effort has been made to supply this lack in the present edition. Particularly does this apply to the western half of the State. The number of illustrations has been greatly increased, since it has been the experience of the author that pictures are often the most valuable reading. The author wishes to express his appreciation of the kindly reception the book has received. The cordial adoption of the book by school boards, superintendents, and teachers, and the liberal sales to professional men, real estate men, farmers, and citizens generally, have fully justified the author's notion that there should be more study of our own State by pupils, teachers, and the people generally. North Dakota is preeminently an agricultural state, and the soil is her greatest resource. The soil is the surface geologic formation, and therefore ought to form a part of the course of study of every pupil in the schools, and should be embraced in the private readings of every intelligent citizen. The book is now in use as a class room text in many of the best high schools of the State, and in a great number of the graded and ungraded public schools, as also in several of the State educational institutions. The author wishes to express his obligation to Professor H. V. Hibbard, of Chicago, for assistance rendered by him in the preparation of the chap- ters relating to the plateau region of the western portion of the State, and also for the use which has been made of the work done by him for the U. S. Geological Survey under the author's direction on the Sheyenne and Maple Valleys. D. E. W. State Agricultural College, December, THE TABLE OF CONTENTS. PAGE CHAPTER THE FIRST The Landscape, - - - 11 CHAPTER THE SECOND Excursions Afield, - 20 CHAPTER THE THIRD The Work of Ice, - 30 CHAPTER THE FOURTH An Excursion to Some Glaciers, - - 39 CHAPTER THE FIFTH The Great Ice-Sheet in North Dakota, - -47 CHAPTER THE SIXTH More Excursions, - 61 CHAPTER THE SEVENTH North Dakota, The Old and the New, - 71 CHAPTER THE EIGHTH Glacial Lake Agassiz, 79 CHAPTER THE NINTH The Deltas and Beaches of Lake Agassiz, 92 CHAPTER THE TENTH Other Extinct Glacial Lakes, - 112 CHAPTER THE ELEVENTH The History of Devils Lake, - - 123 CHAPTER THE TWELFTH The Sheyenne River, - 131 CHAPTER THE THIRTEENTH The History of Maple River, _ . _ 143 CHAPTER THE FOURTEENTH The Lakes of North Dakota, - 151 CHAPTER THE FIFTEENTH Salt and Alkaline Waters in Lakes, - 154 CHAPTER THE SIXTEENTH Map Studies: Distribution of the Lakes upon the Landscape, 157 CHAPTER THE SEVENTEENTH Lakes as a Landscape Feature. - 166 CHAPTER THE EIGHTEENTH The Bad Lands, - 173 CHAPTER THE NINETEENTH The Coal Beds of North Dakota, .... 190 CHAPTER THE TWENTIETH The Beginnings of North Dakota, - - 202 CHAPTER THE TWENTY-FIRST The Coteaus of the Missouri, - - 213 CHAPTER THE TWENTY-SECOND The Plateau Region of North Dakota, - - 229 CHAPTER THE TWENTY-THIRD Agriculture West of the Missouri River, - 245 CHAPTER THE TWENTY-FOURTH The Water Supply, 258 CHAPTER THE TWENTY-FIFTH A Study of the Soils, 269 CHAPTER THE TWENTY-SIXTH Minerals in North Dakota, - 291 CHAPTER THE TWENTY-SEVENTH The Future of North Dakota, - - 303 CHAPTER THE TWENTY-EIGHTH Geology from a Car Window: The Great Northern Lines, - 313 CHAPTER THE TWENTY-NINTH Geology from a Car Window: The Northern Pacific Lines, - - 335 CHAPTER THE THIRTIETH Geology from a Car Window: The Soo Line, - 350 APPENDIX Rainfall in North Dakota. 362 A LIST OF THE ILLUSTRATIONS. FIG. i A Geological Map of North Dakota, - - - - _ Frontispiece FIG. 2 Showing Erosion of Young Valleys on Hilly Landscape, - 14 FIG. 3 Diagram of a Young Valley, - - - . - _ !6 FIG. 4 Cross Section of a Young Valley, 17 FIG. 5 A Cutting Coulee, or Young Valley, - - - - - - 18 FIG. 6 In the East They Work the Land on Both Sides, - 20 FIG. 7 In North Dakota Enough Can Be Raised on One Side, - - - - 21 FIG. 8 Three Types of Landscape, --______ 2 2 FIG. 9 Map Showing Position of Extinct Glacial Lakes and Direction of Ice Movement, ---- ____^j FIG. 10 Movement of Pitch Illustrated, - - - - - - - -34 FIG. ii View Along the Top of a Terminal Moraine, - - 36 FIG. 12 The Snow-field on a Mountain Top, - . 39 FIG. 13 A Glacier and Terminal Moraine, - - 40 FIG. 14 An Ice Cave, - 4I FIG. 15 An Ice Cascade, -_-__ --42 FIG. 1 6 Terminal Moraine and Front of Glacier, - - 43 FIG. 1 7 Terminal Moraine and Ice Front Crowding Upon It, 45 FIG. 1 8 Terminal Moraine Wasted by Glacial Stream, - - 45 FIG. 19 An Old Moraine, -----______ 45 FIG. i9a Map Showing Great Ice Sheet of North America, Opposite 47 FIG. 20 Dakota and Minnesota Glaciers, - - - 48 FIG. 21 Cross Section of Valley of Glacial Stream, - 49 FIG. 22 Beaver Lake and Glacial Channels, 51 FIG. 23 The Ice Sheet at Time of Formation of Outer Moraine, 52 FIG. 24 Small Hill Being Planed Down by Ice, - 53 FIG. 25 Formation of Moraine and Stratification of Ice, 54 FIG. 26 Showing Moraine, Being Crowded Upon by Moving Ice, 54 FIG. 27 A Striated and Polished Boulder, 55 FIG. 28 Granite Pebble, Showing Ice Planing and Striae, - 55 FIG. 29 Striae on Quartzite, South Dakota, - 56 FIG. 30 Hills Worn Down by Action of Ice, - 57 FIG. 31 Ideal Sections of the Turtle Mountain Plateau, - 57 FIG. 32 A Veneered Hill, - 58 FIG. 33 In the Hills Southwest of Minot, 60 FIG. 34 A Huge "Foreigner," 62 FIG. 35 Section of a Gravel Pit, - 64 FIG, 36 A Joint Moraine, - - - 68 FIG. 37 A Glacier and Its Moraine, - 69 FIG. 38 Map of North Dakota, Showing Highlands, ------ 73 FIG. 39 The Tributaries of the Red River of the North, 75 FIG. 40 Contour Mayville and Westward, - 87 FIG. 41 Section Across Beach Ridge, - 88 FIG. 42 Profile Across Beaches at Wheatland, - - -> - - - -89 FIG. 43 Section Across Red River Valley at Wahpeton, 89 FIG. 44 Section Across Red River Valley at Fargo, ------ 90 FIG. 45 Section Across Red River Valley at Grand Forks, - - - - 90 FIG. 46 Section Across Red River Valley near International Boundary, - - 90 FIG. 47 Stratified Clay, Bottom of Lake Agassiz, - - - - - - 91 FIG. 48 Profile of Elk Valley Delta, - - - 93 FIG. 49 Section Across Sheyenne Delta, - - 94 FIG. 50 Delta on Campus, University of Chicago, - - 95 A LIST OF THE ILLUSTRATIONS. 9 PAGE FIG. 51 Section, Elk Valley Delta Showing Stratification, - 97 FIG. 52 Angular Outlines, Not Passed Over by the Ice-Sheet, 99 FIG. 53 Smooth Outlines, Showing Effects of Moving Ice, - 99 FIG. 54 Profile of "The Ridge" and Beaches at Inkster, - - 100 FIG. 55 Profile Park River and Westward, - - - - - -101 FIG. 56 Relationship Between Higher and Lower Beaches of Lake Agassiz, - 105 FIG. 57 Multiple Character of Beaches, - - - 106 FIG. 58 Progressive Elevation of Beaches Northward, - - 107 FIG. 59 A Map of Lake Souris, 113 FIG. 60 Sand Dunes Burying Forest, - - - 118 FIG. 6 1 Section of Devils and^Stump Lakes, - - 126 FIG. 62 Map of Devils and Stump Lakes, - - 127 FIG. 63 The Tower Quadrangle, - 129 FIG. 64 Sheyenne Valley, -- 130 FIG. 65 Section, Sheyenne Valley, Valley City, 132 FIG. 66 Section, Sheyenne Valley, Valley City, - - 132 FIG. 67 Terraces, Sheyenne Valley, Valley City, - 133 FIG. 68 Sheyenne Valley, 3^ Miles South of Valley City, - 134 FIG. 69 Sheyenne Valley. 7 Miles South of Valley City, - 134 FIG. 70 Sheyenne Valley, 10 Miles South of Valley City, - 135 FIG. 71 Sheyenne Valley at Daily, ---- -136 FIG. 72 Sheyenne Valley at Standing Rock, - - 136 FIG. 73 Sheyenne Valley, Under-Cut Bank, - 137 FIG. 74 Sheyenne Valley at "The Jaws," - - 137 FIG. 75 Sheyenne Valley, Lisbon Cut-Off, - - 138 FIG. 76 Railroad. Cut, Kathryn, - - 138 FIG. 77 An Outlier of Shale, - - 139 FIG. 78 Sheyenne Valley, Fort Ransom, - 141 FIG. 79 Banks of Sheyenne, Fargo, - 141 FIG. 80 Glacial Channels, Maple River,' - - Opposite 142 FIG. 81 Section, Glacial Channel, - - 145 F.IG. 82 Section, Maple Valley, - 145 FIG. 83 Glacial Channel, Oriska, 146 FIG. 84 Glacial Channel, Tower Tp., - - 146 FIG. 85 Glacial Channel, Terraces, - 147 FIG. 86 Glacial Maple, Clinton Tp., - 147 FIG. 87 Glacial Maple, at Enderlin, 148 FIG. 88 Glacial Maple, Moore Tp., - 148 FIG. 89 Photograph, Maple River, - 149 FIG. 90 Glacial Maple near Enderlin, - - 150 FIG. 91 A Kame, - - - 150 FIG. 92 Lakes Top of Turtle Mountain, - 162 .FiG. 93 Map of Rush Lake, - - 169 FIG. 94 A Butte, - 172 FIG. 95 Bad Lands, Williston, 175 FIG. 96 Clay Butte, - - - --_ --i?5 FIG. 97 Pyramid Butte, - - - - - 176 FIG. 98 "Capped Butte" - 177 FIG. 99 Halting at the Schack, - - 179 FIG. 100 Structure of the Buttes, - - 180 FIG. 101 Masses of Scoria, _____ ____ '182 FIG. 102 Custer Trail Ranche, - - 183 FIG. 103 Bad Lands, Little Missouri, - - - 185 FIG. 104 "The Palisades," - - 185 FIG. 105 Coal in North Dakota, - - - 191 FIG. 1 06 Old Sim's Mine, ------ -192 FIG. 107 Out Cropping Coal, - - 196 FIG. 108 Mouse River Lignite Coal Mine, - - 198 FIG. 109 Section, Lignite Coal Mine, - - 199 FIG. no Section, Lehigh Mines, - - - - 200 FIG. in Section of North Dakota, - - - 202 FIG. T.T.2 Cretaceous Era in North America, - 205 10 A LIST OF THE ILLUSTRATIONS. PAGE FIG. 113 Section, Artesian Well, Grafton, ' - .-.; - - - - 208 FIG. 114 Section, Missouri Plateau, - - 212 FIG. 115 Extent of Glaciation in United States, - - - , - - 215 FIG. 116 Morainic Lake, - - - .-...- - 219 FIG. 117 A Stony Ridge, - - - - - - - .*: v-"- ' - -. 220 FIG. 118 In the Coteaus, - - - - - - - - - - - 221 FIG. 119 A Land Mark, - - 222 FIG. 120 Douglas Valley, - - 223 FIG. i2i Glacial Drainage, --_ 226 FIG. 122 Map of Missouri Plateau, -- -- 228 FIG. 123 West Rainy Butte, - - 232 FIG. 124 East Rainy Butte, - 233 FIG. 125 Alkali Lake, - - 234 FIG. 126 Where the Bad Lands Begin, - 237 FIG. 127 Border of the Bad Lands, 233 FIG. 128 Crown Butte, - 239 FIG. 129 Buttes near Williston, - 240 FIG. 130 Sentinel Butte, - - 241 FIG. 131 Cross-bedded Structure, - - 242 FIG. 132 The Great Stone Face, - - 243 FIG. 133 A Ranch Home, 246 FIG. 134 Cuskelly Ranch, - 248 FIG. 135 Jack Williams' Ranch, - - -251 FIG. 136 Cherry Creek, -254 FIG. 137 Section, Artesian Wells - 259 FIG. 138 Artesian Section, Devils Lake Southward, - - 261 FIG. 139 Section, Fresh Artesian Wells, - - . 263 FIG. 140 Flowing Well, Red River Valley, - - 264 FIG. 141 Flowing Well, Mooreton, - - 264 FIG. 142 Flowing Well, Chaffee Farm, - 265 FIG. 143 Flowing Well, Woods, Cass County, - 265 FIG. 144 Sources of Artesian Water at Grandin, - - - - - -266 FIG. 145 Just Struck Water, - - 268 FIG. 146 Machinery Buried by Eruptive Well, - FIG. 147 Sandstone-capped Butte, - 272 FIG. 148 Clay Butte, ----- - 273 FIG. 149 Muskrat House, 274 FIG. 150 Camp of Soil Party, FIG. 151 Sun Cracks, Missouri River, - - - - - - - -279 FIG. 152 Wind-Blown Sand, FIG. 153 Four Sisters, Holding Claims, - 281 FIG. 154 The Four Claims, - - 282 FIG. 155 First Occupant of the Soil, - 287 FIG. 156 Sandstone Concretions, - - - - 295 FIG. 157 School House of Petrified Wood, ------- 299 FIG. i57a An Agatized Stump, - - 3 01 FIG. 158 Eastern Farm in Small Fields, 34 FIG. 159 "Pictured Rock," Fort Ransom, - 35 FIG. 1 60 The Last of Fort Abercrombie, ------- 305 FIG. j6i The Pioneers. (Sod House), - 3 6 FIG. 162 Stock Farm, Red River Valley, 36 FIG. 163 North Dakota Farm in One Big Field, - - - - - - -3 7 FIG. 164 Ten Feet of Coal in Banks of Missouri River, - 39 FIG. 165 Freeman's Ranch, - 3 11 FIG. 1 66 Russian Home, - - - - - - - - - - -3 11 FIG. 167 Soil Map of North Dakota, - In Pocket THE STORY OF THE PRAIRIES. CHAPTER THE FIRST. THE LANDSCAPE. /' V INTRODUCTION. How many of the readers of this book understand what is meant by the words Landscape Geology? Every one has seen a landscape, but we often hear people speak about Geology as though that meant rocks and stones and minerals and was therefore hard and dry. It is true that Geology deals with rocks and stones and minerals, among other things, and sometimes it is hard and dry. But Arithmetic and Gram- mar are sometimes "hard and dry" also. It may not always be the fault of the subject that it is uninteresting. The trouble may be in the way it is studied. When the author was a boy and sat upon a hard, old-fashioned wooden bench in a little country schoolhouse between the hills, in the state of New York, he used to think the reading lessons were pretty "hard and dry." Since he has become older, however, he has come to think that the fault was not in the subject, for he now finds these same speeches of Webster and Clay and Washington, and selections from Irving and Lowell and Emerson, very interesting. The trouble seems to have been in the way he studied them. He did not see the beauty in them. He saw big words hard to pronounce and harder to spell, and punctuation marks, at which he must stop, put in between the words! When he read it was to pronounce the words and mind the pauses ! The trouble was not with the lessons, for they were beautiful and grand. The trouble was not entirely with the boy, for he tried to do what he was told to do. Perhaps the fault was not altogether that of the teacher, for she did not know any better ! If after the reader has studied this book he finds Geology "hard and dry" the trouble will certainly not be with the subject, and probably not 12 THE STORY OF THE PRAIRIES. with the reader. If the author has not made the landscape, the fields, the roadside, the school grounds, the river and the lake, more interesting because we have come to know more about them and to see something more than mere rocks and stones, sand and water, then it is his fault and not that of the subject. In geography we sometimes think of the things we are studying about as far away, in some other state, or in some other country, the features of some landscape somewhere, but we, may be, do not realize that it means par State, our neighborhood, our school grounds, our door- yard. V ; :' ; *..f t . t , ,cQne 'who, kftpws Botany sees a good deal more in the fields than : \: rgSriss^ and 'grain, 1 weeds, trees and bushes. The psychologist says he "apperceives" more. We do not wish to talk about apperception now, at least we do not wish to call it by that name, but we w r ish to talk about some things which, may be, we have not seen in the fields round about us, in our own State, our own neighborhood, and so perhaps come to see the great beautiful world in a larger and fuller sense, and may be to a larger realization of what is ours to enjoy. Just as the botanist sees more than grass and weeds and trees in the fields, so may we all see more than soil in the ploughed fields, more than a hindrance to farming in the stones in the fields, more than poor land in the hilly farm, more than a misfortune in the rugged coulee which cuts into the level prairie wheatfield, more than a hay-meadow in the level marsh, more than wheat in the waving billowy sea of grain, more than a useless waste in the boggy slough, more than a worthless waste in the sandy tract of dunes upon which barely a vestige of any- thing green exists. There 'is a grand and beautiful meaning in all the varied landscape of our State if we can but read Nature's story book. In these pages the author has tried to make readable a few of the paragraphs of this great book, paragraphs which are not too commonly read and not too fully enjoyed by the average of human kind. Have you ever wondered why the prairies are prairie and the hills are hilly? Or have you, may be, thought about it and said to yourself that you supposed God made the prairies and hills because He saw fit to do so, and made some parts of the world hilly and some parts level because in His great wisdom it pleased Him to so arrange things? This may enable you to satisfy your wondering curiosity, but a little thinking will enable you to see that, while God in very truth made the hills and the prairies and made some parts of the world different from other parts, THE LANDSCAPE. 13 nevertheless this does not answer the question why things are as they are ; for this great universe which an All-wise Creator made and which He rules is governed by laws in accordance with which the prairies and the hills have been formed, the water and the dry lands have assumed their places, the rivers and the lakes have established themselves, and the face of all the landscape has been fashioned. Hills and Valleys. Every one who reads these pages has seen a val- ley, and also what might be called hills. Maybe the valley was only a ditch or small coulee on the prairie and the hills only little banks one or two feet high. But the importance of things is not always measured by their size. Maybe you have been in those parts of our State, or some other state, where there are great rugged hills and broad, deep valleys. Whoever has seen hills has also seen valleys. Have you ever thought that there might be a necessary relation between the hills and the val- leys? Perhaps you have been accustomed to thinking of the earth as "made" in the beginning with oceans and continents and mountains, with plains and rivers of water flowing through them, and have never questioned but that these have always bee/i so. But a little observation and reflection at once teaches that this is not so, for you have not failed to see that the river is constantly changing the land, a little soil is being washed into the valley from the banks along its sides with every rain and this is carried down the stream. All streams transport mate- rials by carrying them or shoving and rolling them along their bottoms. Perhaps you have watched the sand and pebbles creeping down stream on a gravelly bottom, and wondered how long this process has been going on, and when it was that soil and sand began to be carried down stream. And then perhaps you wondered if the stream would ever stop carrying away the soil and sand toward the ocean. By and by you began to think that this carrying away process must have begun as soon as there was any land on which rain fell ; and so also you concluded that this constant wearing away of the land, called erosion, will keep on as long as there is any land left above the level of the sea. It occurs to you that likely this has been going on ever since the beginning of things and you perhaps begin to wonder if the land will not all be car- ried away in time and you wonder if there has not been more land here sometime which has been carried away. When you think -that "the beginning" was a good while ago you are forced to> conclude that a good deal of land has been carried away. And when you think that the land which is nearest the rivers is the first to be carried away, and 14 THE STORY OF THE PRAIRIES. How the Farm is Lost. How the Farm is Regained. How the Farm is Retained. FIG. 2. Showing the Erosion of Young Valleys on a Hilly Landscape. Photographed from a C/iart, by Prof. E. S. Keene. THE LANDSCAPE. 15 that the hills and higher lands are but the parts which are farther away and have not yet been carried away, you see that the river or running stream is the agent which is doing the work of carving and fashioning the landscape. The river is water seeking its level. The rains loosen the soil on the banks of streams so< that it, too, seeks a lower level, or falls. The energy of the sun causes water to evaporate and rise as vapor. This forms the clouds, and the clouds are blown by the winds and carried over the land. Then they fall as rain and again form rivers. Then the rivers, as we have seen, flow off the land and carry with them the soil or fine parts of the earth, the materials of which the hills'are made. So long as the sun furnishes heat the waters will be evaporated, and clouds will be formed, and rains will fall upon the earth, and rivers will flow into the seas. And so the endless cycle goes on, has been going on through the long aeons of the past, and will continue to go on through the lapse of ages to come. And so the continents are being gradually worn down and carried into the seas. The "everlasting hills" are not everlasting. They tarry but a day when time is meas- ured in geologic cycles. In truth, "one day is with the Creator as a thousand years, and a thousand years as one day." The little rivulet which runs by the school-house playground or along the roadside is doing the same kind of work in carrying away the land to the ocean as the river, only on a smaller scale. But it is only a question of time till the level prairies will give way to the hilly landscape, and finally the hills will yield to the constant wearing of the streams. When the landscape has been thus worn away so that the land is but little higher than the ocean-level then it is said to have reached its base- .level of erosion. Beginnings of a Landscape. If a new continent were imagined to arise out of the ocean, upon which were no rivers, no valleys or hills, its surface sloping uniformly to the sea, how would rivers get started? It must be that they would form in some way, for there are rivers or streams on all continents where rain falls. Children have been taught sometimes (let us hope not in the schools of our own State) that rivers were established- in their courses by a gathering of waters in the in- terior of the continent and that this water .flowed across the land wher- ever it could go most easily, and in so doing cut a channel and became established in a definite course. Now, all the water there is on the land in lakes or streams or in the soil comes from the rain which falls 16 THE STORY OF THE PRAIRIES. upon the land. A large part of the rain-water percolates into the soil and rocks of the earth. Some of it collects in low places and forms lakes, pools, and marshes. From these a good deal evaporates and goes into the air to form clouds again. Now, where will a river have its beginning? Where will a definite stream channel first appear? Will it start from the interior and flow toward the sea? What will start it? Does any more water fall on the land in the interior than nearer the sea? Since the land is higher than the sea, the land waters will tend to move toward the sea. Where are the waters which will reach the sea first? It is plain, the waters nearest the sea. * And since moving water always cuts a channel, or erodes the land over which it flows, the first soil to be carried to the ocean and deposited on its bottom as sediment will be the soil which was at the margin, or edge of the land, and the beginning of a channel or valley will be at the edge of the land. The next water to get to the sea will be that which fell on the land near to the edge but a little farther inland. Then that from a little farther inland still, and so on, till finally the water from the interior will get down to the shore. But where now has the valley been cut most? Where is the largest part of the river? Where did the river begin? If we indicate a series of small areas extending from the sea-shore toward the inland by the letters a, b, c, d, e, f, g, h, the waters which fall upon a will be the. first to reach the sea ; those which fall upon b will FIG. 3. Diagram showing how a Valley begins at its own Mouth. be next, taking advantage in their course of the channel made by the waters of a; those falling upon c will be the next, and these will go down by the channel made by a and b; and d will in turn reach the sea coursing down the channel made by a, b and c, making the channel deeper and wider by erosion; and at length e, f, g and h will reach the sea. Let us now compare one part of the valley with another from a to THE LANDSCAPE. 17 //.. How do the amounts of water which have gone over each area compare? Suppose we say the water which falls upon one area is one volume. Then if the whole length of the valley is the distance from a to h, and if we suppose all the water which falls on each area to go down the valley, the water which passes over a will be seven times as much as passes over g, that which passes over b will be six times as much as passes over g, five times as much over c as over g, and so on, while from h will pass only the water which falls upon that area. Where there is the most water, other things being equal, there is the greatest erosion. Where then has the greatest channel been formed? And where is the river largest? And finally, where does the valley of a river begin, in the interior of the continent or at its own mouth ? Let us now think of the series of areas, a, b, c, d, etc., as a thousand, and the extent of each area to be large. From the farthest and highest part of the continent the waters may be thought of as a long time in reaching the sea. There will be then a broad and deep valley nearer the sea, and it will be smaller and smaller as we go inland, and on the thousandth area, or the summit of the continent, it will be only a place where rain falls, with hardly a beginning of a coulee. Let us now go out upon the level prairies of North Dakota and look at the coulees and see what we can observe of the workings of a river system. Let us see if we can find any examples of what we have just been studying. If we select a day when it has been raining for some time so that the land is well covered with water, we shall be able to see in reality what we have been seeing in imagination. Here on the prairie, cutting through level wheat fields, is a coulee, a little valley FIG. 4. Cross Section of a Young Valley. 18 THE STORY OF THE PRAIRIES. having steep sides, growing wider down stream and narrower up stream, its sides becoming less steep towards the mouth and more steep towards its head. In the bottom of this trough or notch in the prairie trickles a tiny stream. Can it be that this stream has carried away the earth which once occupied the space where is now the trough or coulee? Strewn along the bottom are boulders, sand and gravel, the heavier masses which could not so easily be carried away by the waters and which were in the soil or earth which has been carried away. If we go out upon the land some distance from the coulee and look across it we shall see that the whole trough of the young valley is below the FIG. 5. A Cutting Coulee, or Young Valley. Photograph by Prvf. Chas, M. Hall, level of the surrounding country. On the level prairies of the Red River Valley you could imagine a great board or plank to extend across from the prairie on one side to the prairie on the other. The Grand Canyon of the Colorado River is but a great coulee cut down by the river deep into the plain. The materials of which the great Colo- rado plateau is made are of such kind that the moving waters cut it away rapidly, and the walls on either side are steep and high. Canyon is another name for a young valley. Let us now go along the bank of the coulee and see if we can dis- cover how the valley got started. All about upon the level prairie we see water standing in sheets from recent heavy rains. If we ask ourselves if the prairie will by and by be dry again we shall certainly THE LANDSCAPE. 19 answer that it will, for it has often been very wet before and has become dry again. Where did the water go? It soaked into the ground, or a part of it did, and some of it evaporated, and went to help make clouds. But how 7 about the water which was near by the edge of the coulee? Some of it fell down the side into the trough carrying with it always some soil. If it chanced that there was a depression or lower place in the prairie, and there ahvays are such places, this hollow was filled with water, and if the low place is so near the coulee that its waters break over the edge and fall down the side, or if a little rivulet on the bank of the coulee should cut back into the edge of the little "lake" and tap it, then its water would be drained. But in falling down the side of the coulee the water cuts a little channel, and when it rains again the water which falls in this hollow, or lake, will run into the valley through the little channel formed before, cutting this deeper. If this depression were a large one the little channel would become a feeder to the larger stream which made the valley, and it would then be called a tributary to the valley. If we go down the course of the coulee to see where it ends we shall see that it discharges into a larger stream, or maybe runs into a lake. If it joins a larger stream then it is itself a tributary to the larger stream. How then did the coulee or young valley get started? In just the same way as the branch or tributary, for the coulee is only a branch of a larger stream. How does a coulee or valley increase its length? If you watch a little rivulet by the roadside when it is raining hard you will see that the head of the little stream pushes back toward the land as the water from the land falls over into the little valley. In fact it grows longer in just the same way as it got started in the first place, by water falling from a higher to a lower level and carrying the soil along with it. CHAPTER THE SECOND. EXCURSIONS AFIELD A Few Comparisons. North Dakota is one of the "prairie states." Yet those who have seen the various parts of the State often speak of the "hills" in any place as though North Dakota could be said to have real hills! Compared with Pennsylvania or New York or Ver- mont the "hills" of North Dakota are hardly more than knolls. When eastern people think of a North Dakota landscape they often think of broacl-reaching prairies limited to the view only by the distance the eye can reach. North Dakotans will make no serious objection to such opinions being held, especially when the rugged hilly character of many eastern landscapes is considered. And even if it be contended that in the east they can almost "work the land on both sides" because the surface appears to be turned up on edge, yet we are satisfied to answer FIG. 6. In the East they work the land on both sides'. Photograph by McCormick Harvesting Machine Co. 20 EXCURSIONS AFIELD. 21 FIG. 7. In North Dakota enough can be raised on one side! Photograph by McCormick Harvesting Machine Co. that we can raise more on the one side of our prairies than can be raised in the states named on two sides. But those who know the geography of North Dakota know that the whole story has not been told when it is said that ours is a prairie state.. There are prairies and prairies! Level prairies and rolling prairies. And sometimes the "rolling" is so marked that we may ven- ture to speak of it as "hilly." Compare the floor-like level about Fargo or Grand Forks, Cassel- ton or Grafton, or any part of the Red River Valley; the rolling-prairie country about Langdon or Devils Lake, Oakes or Ellendale; the rugged and unploughed hills between Hope and Valley City, or the picturesque "curves" of the landscape along the Sheyenne River in Foster county; the billowy ups and downs on the Coteaus of the Mis- souri west of Minot; the steep and bouldery landscape south of Dog Den Butte in McLean County ; the broken-prairie country about Dick- inson known as the "breaks;" the ragged and rock-ribbed hills, known as "buttes," in the valley of the Little Missouri. We shall see that while North Dakota is a prairie state yet she has much diversity of surface. Again, in some places the fields are very stony, in others hardly a stone can be found over great areas. And not only this but the stones are mostly rounded and smooth, while in some places they are nearly all angular and rough. Some of the lands are called "light," having a dry sandy soil with no stones larger than sand grains, and some are "heavy," with a clayey soil, often with large stones imbedded in the clay or on the surface. And still again, some of the fields are black, 22 THE STORY OF THE PRAIRIES. with a deep loamy soil. And these differences often occur within short distances. One who has ridden over the Great Northern Rail- way westward from Larimore may have observed that there is an abrupt change from the level prairie east of Larimore to the "hilly" prairie to the west. The same kind of a change, though not nearly as great, occurs four miles west of Wheatland on the Northern Pacific BROKEN. FIG. 8. Three Types of Landscape. line, wh'ere that road rises off from the level Red River Valley onto the highland to the west. East of the city of Devils Lake the prairie swells and rolls in grace- ful undulations. Go across the lake and the landscape becomes very hilly, and often the hillsides are strewn with large boulders. From Towner to Minot the country is gently uneven prairie. West of Minot there is a sudden change to a high plateau with an uneven and hilly surface. Along the Goose River east of Mayville the fields are almost as level as the floor of the school-house, and the soil is black EXCURSIONS AFIELD. 23 and when wet exceedingly sticky. Travel west toward Sherbrooke and it will be observed that the soil becomes sandy, and well defined sand-ridges run north and south. About Sherbrooke the hills are sharply rolling and the soil is less black. In many parts of the State fields free from stones and those which are very stony are intermingled. And it is noticeable that the stones are nearly all rounded and smoothed. Cross the Missouri River how- ever in the western part of the State and the hills are seen to be different in shape. Here they are flat on top with trough-like valleys between them very different from the rounded hollows among the hills on the rolling prairies, and the "cobble-stones" or boulders, which are so common over much of the State, soon disappear entirely west of the Missouri River. North Dakota has level and rolling prairies, hills and hollows, lakes and marshes, fields very stony and those free from stones, fertile farm- ing lands the best and richest in the world, other lands more valuable for grazing than for farming, and the most wonderful "Bad Lands," all resulting from geologic agencies. They are not so by accident or chance. They are geologic facts. Their explanation belongs to the science of Landscape Geology. An Excursion Among the Boulders. Everyone has noticed boul- ders scattered here and there over the prairies, big boulders some- times weighing several tons and smaller ones of all sizes down to "cobbles" weighing a few ounces, and pebbles of the size of marbles, and finally gravel and fine sand. A little study of the soil will show that it also is made up largely of tiny particles or grains of sand which are boulders reduced to small size. And the familiar clay which is so common a feature of the soil a little below the surface is but the still finer particles of broken rocks so finely ground or pulverized as to make the separate particles not able to be seen without the aid of a microscope. Boulders are seen scattered sometimes in groups or patches, sometimes a single one with no others near, and big and little are mingled in great confusion. Sometimes a sand pit is seen in which the sand is arranged nicely in layers; and occasionally a stray boulder is found in the sand, sometimes many of them. It has also been no- ticed that the boulders are very unlike in kind. Some of them when broken look very much like broken glass, often having a milky gray appearance. These are called quartz, or quartzite boulders. They are among the hardest of all the rocks commonly found in the fields or 24 THE STORY OF THE PRAIRIES. in quarries. It is the same kind of rock as that from which window glass is made. It is so hard that a freshly broken piece of it will readily cut or mark window glass. A steel knife blade will leave a black mark like a pencil mark on it. By remembering these things you can easily tell which are the quartz boulders in the field. Another kind which is likely to be found in any group of boulders is one which when broken will show a rough surface with little blocks having a somewhat cubical shape, and colored pinkish or reddish, though sometimes white, and often flesh-colored. The surfaces of these little cubes are smooth and shiny, and reflect the sunlight so that they look very bright. These little blocks or crystals, for they are really crystals, are a mineral called feldspar. They may be so small as not to be easily distinguished, and sometimes the little shiny faces are one or two inches across. Mixed with these feldspar crystals may be seen little black specks or plates. These also vary much in size. When they are large enough they may be easily split with the point of a knife into thin scales. This mineral is soft and can be cut or scratched with a knife point. These are crystals of mica, and when they occur in large plates are cut up and split apart into thin pieces and used in coal stoves. The micas used in coal stoves are simply pieces cut out of very large crystals. The mica crystals seen in boulders are some- times black, sometimes clear, sometimes brown, and sometimes green- ish. But they are always soft and can always be split into thin scales. A third mineral which is always present in the kind of boulder we are now describing is quartz, the same quartz as has been before spoken of as making up some whole boulders. It has somewhat the appearance of broken pieces of glass, scattered through the rock among the feld- spar and mica crystals. These particles of quartz are sometimes hard to distinguish from feldspar, but the faces of the little blocks are never shiny like those of feldspar, and it is never in little square blocks like feldspar. Then it may be remembered that quartz is very hard. Feld- spar is hard, but not as hard as quartz. These three minerals, feldspar, mica and quartz, make up the rock called granite, and these boulders are granite boulders, the same kind of granite as is used for making tombstones and for building purposes. It is a very hard rock and is not easily broken. The action of frost and sun has little effect upon it, and it also takes a fine polish. These things make it very valuable for monuments and building purposes. A fourth mineral called hornblende is often found in connection with the EXCURSIONS AFIELD. 25 three named, and this is somewhat like mica in appearance. It is, however, harder than mica and does not split into thin scales so easily as mica and it is generally in thicker masses, and is usually green or greenish-black in color. These two kinds of boulders, quartzites and granites, are among the most common. These are the more familiar "hard-heads" which everyone has observed. Besides these, however, there are others which when broken do not present the glassy, milky or grayish appear- ance of the quartzites nor the flesh-colored, red, brown or specked appearance of the granites. Limestone boulders are common in North Dakota, and in most of the northwestern states. These can be known, however, by their softer character, and usually by being more affected by the action of sun and frost. They dissolve and crumble much more readily than the others. A good deal of the soil of North Dakota is made of ground-up limestone, and as we shall see by and by this ma- terial has helped to make our rich wheat-fields and also to make our wells furnish hard water. Still other boulders there are which have long hard names which we do not need to describe here in particular, but only to say that there are a good many others and nearly all of them are made of hard materials so that they do not easily crumble or break. This fact of their being hard is important, for we shall see later that this helps to explain why they are here. They have not been broken up or dis- solved, because they were so hard. But a fact that we should notice here is that these different kinds are found scattered almost all over our State and over other northern states as well; limestone, granites, quartzites, hornblendes, augites, cherts and many others, large, small, and all sizes, mixed, and scattered singly and in patches, sometimes almost covering the ground and sometimes few and far apart, on the surface and deep in the soil below the surface. This great variety in kinds, in sizes and in the way they are scat- tered leads us to inquire how this has all come about, where have the stones come from and why are they so different in kind and size, and so curiously scattered? Why are huge boulders sometimes found on .the tops of the hills as well as in the valleys? And again sometimes not even a good-sized pebble can be found for miles. Then again it is all sand for miles, suddenly changing to black sticky prairie. It has not required any great skill in guessing to surmise that these rocks, these huge boulders and the great quantities of sand, were not 26 THE STORY OF THE PRAIRIES. "made" in North Dakota, that is, that they did not in the first place belong here, but have been brought here by some means from some- where else. These rocks are not like any of the rocks in the quarries of the State, and then too these boulders, pebbles and gravel, and even the sand grains are all rounded more or less, while the rocks from our quarries or from ledges along the streams where the bed-rock comes to the surface, are all rough and angular. To explain how these things have come about a geological story will have to< be told, a little frag- ment of the earth's history, of the manner in which a great change took place over a large part of North America, and which includes most of the State of North Dakota, all of that part in fact which lies east of the Missouri River. A part of this story will be told in the next few chapters. An Excursion to Some Quarries. Just as it is necessary for us to see, feel, smell, taste and hear in order to think about an object, so it is necessary for us to see, handle, break, dig and walk over the fields, rocks, soils, hills and valleys in order to understand the geography of our own neighborhood or State. But all parts of our State are like all other parts in many respects, and what is true of North Dakota is in a large measure true of other states, and other countries. Since we can- not all visit all parts of our own State, and still fewer can visit all the states or all the countries, let us first study our own neighborhood, and then from this we may be able to understand the parts we cannot visit from what those say who have seen parts we have not seen. He is a good scientist who understands thoroughly his own neighborhood. Let us then go out and pick up a basket full of stones from the fields and roadsides. Let them be collected from all parts of the neighbor- hood, and let big and little and all kinds be gathered. If there is a patch of boulders in the neighborhood which are too large to be moved look carefully at them where they are. In the collection which we have made we have perhaps one hundred, maybe two or three hundred, "specimens," yes specimens, for each one of these humble stones .has its own story to tell, and strange as it may seem scarcely any two of them will tell the same story. Can you find two which are exactly alike in shape or size? Or, what is more wonderful, can you find two in the whole collection which seem to be, when broken, exactly the same kind of stone? If we have two or three hundred specimens gath- ered from about the neighborhood, very likely if you try to sort them, placing them in piles so as to have each kind by itself, meaning by EXCURSIONS AFIELD. 27 kind those which are exactly alike, we shall have a hundred or more piles! Now if you have ever been in a stone quarry you have probably noticed that the stones which were being taken out by the workmen were all very much alike. If the ledge in which the quarry is located is deep, if the wall of rocks is high and you see many layers in order you may have noticed that they are not all alike, but if you look at different parts of the same layer, following it from one part of the quarry to another, you notice it is the same all along. The different layers may also be very much alike. You see no such differences in these layers, or strata as they are called, as you saw in the collection you made from the fields. If you have been in a quarry in Minnesota or Wisconsin or Iowa it may have been a limestone quarry you saw. Among the specimens you collected there are probably several lime- stone boulders. These you will observe are different in shape from the quarry blocks. The boulders are all rounded and smooth, while those freshly broken from the ledges are sharply angular. If you have been in eastern South Dakota may be you have seen the hard reddish building stone which is taken from the extensive quar- ries along the Big Sioux River This rock is of quartzite, the same min- eral as has been spoken of as making some of the "hard-head" boulders. This particular region of South Dakota has no other rocks in the quar- ries. It is known as Sioux quartzite and is famous as a building stone. The city of Sioux Falls gets its name from its location near where the Big Sioux River crosses an outcropping of this rock. Stone quarries are very scarce in North Dakota, for reasons which we shall see a little later. Let us look again to our sister state of Min- nesota. At Kasota, near Mankato, are large quarries where the splen- did reddish-brown sandstone is obtained which is used for trimming the best brick buildings in many towns and cities. The bed-rock at Kasota is of this one kind of sandstone. But around about on the surface, in the fields and by the roadsides, are boulders such as these we have gathered from the fields and roadsides of our own State. So also about Sioux Falls, are boulders in the fields and along the road- sides, but in the quarries there is only the one kind of rock, quartzite. Now if we could dig down deep enough in our own State we should by and by come to bed-rock. In some parts of the State we should find this to be limestone, in other parts sandstone, and in others shale. The sandstone would be different from that at Kasota, however. If 28 THE STORY OF THE PRAIRIES. we should go north into Canada, away to Hudson's Bay, for instance, or about Lake Superior, we should find the bed-rock to be like some of the boulders we have in our collection. In some places we should find granite, in other places quartzite, and hornblendes, and augites. So similar are the bed-rocks in those localities to the pieces or boul- ders which we have collected here, and so much do the scattered boul- ders look as if they had come from some other place, that we almost begin to wonder if in some way our boulders did not come from about the Hudson Bay or Lake Superior country. In a later chapter we shall see that there is reason for thinking that many of our boulders and a large amount of finer materials have really been brought from these far-off regions. All the boulders, pebbles and sand-grains of our prai- ries and fields have come from other places where the bed-rock is the same kind of rock as these boulders. In other words these boulders are pieces broken off from the layers or strata of the bed-rock where these come to, or near to, the surface. They are fragments which have been broken from many different quarries in many different places, and carried sometimes hundreds of miles to where we find them in the fields. Some of the pieces were very large and heavy when first broken. In the process of moving they have become a good deal broken, big blocks being broken up into small pieces, the corners worn off, and the whole surface made smooth. When a large rock is broken into smaller blocks there are always some small fragments formed, and when a corner gets knocked off from a rock by striking against another rock more small fragments are broken off. The only difference between boulders and sand is in the size of the fragments. A boulder may be broken into several smaller boulders, and these may be again broken into pebbles, and these in turn are only larger grains of sand. They all get smoothed and rounded by being jostled and rubbed against each other and against other hard things which are in their w r ay, or which are moved against them. Indeed soil and the clays of the fields and hills are mostly ground-up rock. The softer boulders are more easily worn to powder and broken. The boulders, the larger ones, those which are well rounded and smoothed, and which have been quite correctly called "hard-heads," are the harder masses which have been broken loose from the bed-rock somewhere and by reason of their being so hard have not been worn out and made into soil. If you examine the grains of a handful of sand from a sand-pit you will find it to be made EXCURSIONS AFIELD. 29 up of hard particles of stone. The grains will be largely quartz grains, and bits of feldspar and other hard minerals. You will generally find but few grains of mica or limestone because these are softer and more easily ground to powder. These have been ground into earth and clay. Nearly all the sand patches or sandpits, like the sands of the sea-shore, are whitish, and this is because it is largely grains of hard whitish quartz. Because the boulders, sand and clay of our fields have come from somewhere else, have drifted here from other regions, this material is called "Drift," and the boulders are often spoken of as "foreign" boul- ders or drift rocks to distinguish them from the rocks which have come from our own quarries or from the bed-rock near where the pieces are found. All of North Dakota except that part of the State which lies west of the Missouri River is covered with a great sheet or mantle of "drift." In some parts of the State this covering of drift is very deep, being more than 300 feet in some places in the eastern part of the State. It becomes thinner toward the west till along the Missouri River it is only a few feet thick and further west disappears entirely. The black soil of our fields does not extend down very far, as you have likely noticed. But if you have watched the digging of a deep well or a place where any deep excavation was being made, you have seen that clay and boulders occur down to a much greater depth, and probably no shelf or layer of rock was struck such as you saw in the quarries. All these materials, these many millions of tons of clay, boulders, sand, and gravel and most of the soil also, which cover nearly the whole State, are drift, and the time during which this vast amount of work was being done is known as the "Drift Period," or Glacial Period. It was the last great geologic period before that in which man lives, the period of written history. \Ye shall, in the next few pages, try to see how the boulders, peb- bles, sand, and clay were carried and how they come to be left as they are. CHAPTER THE THIRD. THE WORK OF ICE. The Great Ice Sheet. All of North Dakota east of the Missouri River is embraced in that part of North America which was covered by the ice during the Glacial Period. We have wondered how the boulders and rounded pebbles came to be here, scattered all about as they are, when they are so different from the bed-rocks and also so different from each other. Geologists agree that ice was the agent which transported these rocks here; that it was by the action of the ice that the rock fragments were first broken from their parent ledges and carried, smoothed, broken, and ground to powder; that the way the boulders, gravel and sand are distributed is due to the ice melting and leaving the rocks which it carried; and the peculiar hills and rolling prairies which mark the landscape have been formed by the dumping of these transported materials from the great ice-plow. All the northern portion of North America was covered by this great flood of ice. In all the northern states from North Dakota to Maine and the Atlantic Coast about New York City occur boulders, sand and clay, and peculiar rounded hills such as- are seen between Larimore and Devils Lake, along the line of the Great Northern Rail- way, about Cooperstown in Griggs County, west of Hope in Steele County, at intervals along the line of the Northern Pacific Railway from east of Valley City to Bismarck, east of Lisbon in Ransom County, about Oakes in Dickey County, and, in fact, here and there through- out the whole State west of the eastern tier of counties and east of the Missouri River, occur irregular generally rounded hills, and valleys without outlets. These are hills which mark positions where the edge of the great ice-sheet stood for a time, and, melting, left the mate- rials of which these hills are composed. Wherever such hills are seen the country has been "glaciated." The ice-sheet was a good deal deeper or thicker in some places than in others. \Ve shall get the right idea if we think of the great flood of ice slowly flowing or shoving its way across the country, covering the THE WORK OF ICE. 31 32 THE STORY OF THE PRAIRIES. hills and filling the valleys, planing off the hill-tops and filling the val- leys with the materials of the hills. It may seem a little strange to think of ice flowing over the land, but there are a great many strange things in the world and we should not refuse to study them because they are strange. In another chapter we shall try to see some of the reasons which have led geologists to think that it was a great ice-flood, a vast sheet of snow-ice slowly creeping or flowing from the northeast toward the southwest which has caused all these strange things. We must try to be fair and honest in a study of this kind and not refuse to think about things because we cannot at first understand them, or can- not see how such things can be. No one claims that we know these to be the facts absolutely. No man was on the earth at this time to write a history of what occurred; or if there were any men then at least they did not write any history which we know about. All that we can tell about what occurred is by studying the records left in the rocks and clays and gravels, and the peculiar hills and valleys. The collection of boulders, pebbles and sand, the clay dug up from below the fertile soil, the hills and hollows them- selves which we walked over and through, and the rocks we studied in the field, all enter into the great subject of the history of this period of the earth's changes. Without trying at this time to explain the causes of the extreme cold which made such a gathering of snow and ice possible, let us see what the physicist says, the man who has studied the action of ice and snow and water, and other substances under various conditions, about the behavior of ice in very large masses. Then we may afterwards seek what reasons or evidences there are for thinking that ice was the agent which did all this work; or that what has been called an "ice-invasion" has really at some time occurred. Behavior of Ice Under Pressure. >We are accustomed to think of ice as a brittle substance; and we know that when struck a sharp blow with a hard instrument it will break into pieces. But it can be shown in a laboratory where all things needed are at hand, or in great glaciers where the mass of ice is very great, that when ice is placed under great pressure and acted upon slowly and steadily for a long time it not only does not break into pieces as a brittle solid but actually flows very much as a mass of resin or cold, thick pitch will flow if it is given time, bulging out on all sides from the pressure of its own weight. To get some idea of the way the ice will act let us use some figures. THE WORK OF ICE. 33 A cubic foot of ice weighs about 62.25 pounds. If we imagine two blocks of this size placed one upon the other, the bottom one will hold up a weight of 62.25 pounds. If ten blocks are piled up on top of the first one then the bottom one will be holding up 622.5 pounds. If we imagine the blocks to be piled up as high as the highest grain elevator, say 100 feet high, then the pressure upon the bottom due to the weight of the ice blocks would be 6,225 pounds, or more than three tons. Imagine the whole weight of a load of a hundred bushels of wheat to rest upon one such block of ice. It would be crushed, would it not? Now, suppose that the whole landscape round about were covered with ice to a thickness of 100 feet. Each square foot of land might be thought of as having a pile of one hundred foot-cubes of ice resting upon it. Each bottom cube would be prevented from crushing the way the load of wheat was imagined to crush a single block because there would be more blocks all around it and each one trying just as hard to crush. The lower layer of ice would therefore be under a great stress. Now, in parts of the country where there are high mountains, as the White Mountains in the state of New Hampshire, drift boulders and pebbles on the tops of these mountains show that the ice covered their tops, or, in other words, the thickness of the ice was so great that the high mountains were buried. Some of those mountains which were so covered are more than a mile high, that is, their summits are more than a mile vertically above their bases, and drift boulders and gravel are found upon their sides and up to their very summits. The ice must therefore have been more than a mile deep in those regions. Many careful observations have led to the conclusion that the ice was in some places two or more miles deep. What, then, must have been the pressure upon the bottom layers due to the weight of the ice? One mile is 5,280 feet. The pressure upon the bottom of each square foot, therefore, must be 5,280 times 62.25 pounds, or 328,680 pounds, or nearly 165 tons. Since the ice cannot crush, being hemmed in on all sides by more ice under just the same pressure, the stress upon the bottom layers will be very great. Under these conditions of great pressure ice behaves like a thick, viscous substance, such as pitch or thick tar. An Illustration. Let us imagine a large cask or barrel filled with hard pitch. It appears solid, and if a piece of it is struck a sharp blow it will break much like a brittle rock or a piece of ice. Suppose we 34 THE STORY OF THE PRAIRIES. should knock, the barrel -to pieces and leave the pitch standing in a great block. It will have the form of the inside of the barrel. But let it stand for some time, say a week or a month, and it will be seen to have bulged out at the sides near the bottom. Leave it longer. The mass no longer has the form of the inside of the barrel. It is flattening down and broadening out at the base. Leave it for a still longer time, for a year maybe, or even two years, and it will have flattened out so that no one would ever think that it had once had the form of the inside of a cask or barrel. Now, suppose such a block of pitch is left to stand on a level floor. It would flatten out and flow over the floor from the pressure due to its own weight. If there were some marbles or small stones lying upon the floor scattered about or in little heaps, the pitch would flow over FIG. 10. these and shove them along with itself. If the block of pitch were on the cellar bottom where there were small hollows it would fill these and push on over them. If there were small gravel stones in these hollows some of these would be shoved along up out of the hollows and pushed over the uneven surface. If we now can imagine the pitch to disappear by some means with- out disturbing the pebbles it has moved over the cellar bottom, we should find these pebbles to have been shoved into a somewhat irregu- lar row near where the edge of the spreading pitch had been. In much the same way the ice flowed across the continent, filling the valleys and crossing the hills as the pitch flowed over the cellar bottom and filled and crossed the hollows and hummocks. The great pressure from the accumulation of snow in the interior of the continent caused the outward flow. In the interior of the continent the ice melted on the land when it had flowed southward into the warmer THE WORK OF ICE. 35 climate of lower latitudes. Off the coast of New England the edge of the great ice-sheet pushed off into the sea. In the latter case the rock- fragments carried by the ice were thrust off into the sea. But in the former case, where the ice melted on the land, the broken rock, some of which had been ground to fine powder forming clay, and the small fragments in the form of gravel and sand, together with the large boul- ders, were left where the melting ice dropped them. Alpine Glaciers. Ice can be seen flowing down mountain sides at the present time in many countries, in Switzerland, Norway, Green- land, Alaska and the Rocky Mountains in our own country. Ice- streams flowing down the slopes of mountains are called Alpine Gla- ciers, from the Alps Mountains in Switzerland, where there are splendid examples of glaciers in action, and because it was there that the flow of ice in glaciers was first studied. If you have been on the top of Pike's Peak, or through the Yellow- stone National Park, in the hottest months of summer you have seen great patches of snow here and there among the crags and pinnacles, above what is known as the "snow-line." Where there are high moun- tains with their crests reaching far above the snow-line the summers are not warm enough to cause all the snow to melt, and so it continues to gather in the hollows high among the clouds and craggy peaks. When,o D ^V^-= o o FlG. 21. Cross Section of the Valley of a Glacial Stream. The great Sheyenne Valley is one of the most notable examples of this kind, and one of the grandest in all the Northern States. It re- quires no great effort of the imagination to see that the present small and slow-flowing Sheyenne River did not make the great valley in the bottom of which the river now flows. Many broad, fertile prairies, a little lower than the surrounding roll- ing prairie, and having hills alongside and not very far distant, may be the place where has been a sheet of ice-water from the melting glacier 50 THE STORY OF THE PRAIRIES. a temporary glacial lake. The richness of the soil on such prairies is often due to the fact that waters flowing into the lake carried fine silt or rock-flour and deposited it over the bottom of the lake. This tem- porary lake disappeared after its supply of water from the melting ice ceased. Sometimes, how r ever, a pond or marsh remains as a vestige of the larger lake. The Dakota Glacier flowed south and a little east from the direction of Lake Manitoba and the region west of Lake Winnipeg, and at the time of its greatest extent reached across North and South Dakota. The Dakota Lobe and a part of the Minnesota Lobe at a later stage, when the ice had melted back a long way, is shown in Figure 20. When it stood at the position shown in Figure 22 the outermost moraine, called the First or Altamont Moraine, was formed, along the edge of the ice. On the west side of the lobe was formed the irregular system of hills shown in Figure i crossing the State through Mclntosh, Logan, Emmons, Kidder, Burleigh, McLean, Ward and Williams Counties. Across the State from Ashley in Mclntosh County (see Map, Figure i), northeast to Park River in Walsh County, a line would cross the ten great Terminal Moraines formed by the Dakota Lobe or Glacier of the Great Ice-Sheet in North Dakota. These moraines have been named in their order from the one first formed at the outer edge of the glacier to the one far to the north in Canada. They are numbered as well as given geographic names. The outer or First is the Altamont Moraine, the name meaning high hills; the Second or Gary Moraine, the Third or Antelope, the Fourth or Kiester, the Fifth or Elysian, the Sixth or Waconia, the Seventh or Dovre, the Eighth or Fergus Falls, the Ninth or Leaf Hills and the Tenth or Itasca. The Itasca Moraine was formed after the ice had retreated to the next stage after that represented in Figure 20. The reader need not try to remember these names. They are given here for reference, for convenience later. The names are geographic names from places where the moraines are well developed, as for example the Fergus Falls Moraine is named from the fact that the city of Fergus Falls, Minnesota, stands upon this moraine, where the hills are very conspicuous. The names have no more meaning than the names of persons. Lakes and streams of ice-water skirt the edge of the glacier. Great streams also poured off from the surface of the ice and spread out upon the ground adjoining. Much gravel, sand and finer rock-powder were THE GREAT ICE-SHEET IN NORTH DAKOTA. 51 SMMKse W - Sip !:-'$, 1^4^*1 ^ *'' I >^ -fei fift - m * 52 THE STORY OF THE PRAIRIES. washed by such streams from the ice-front and spread as "over-wash plains" upon the land. The streams cut wide and deep channels, for the waters were kept at flood by the continued melting of the ice. When the ice had melted of course the streams ceased to be, but their channels were left and they mark the landscape to-day in many parts of the State. South of Devils Lake are some of the largest hills in the State. There was probably a large range of hills there before the ice-sheet ICE .SHEET AT THE TIME Of THE FORMATION OF THE OUTER MORAINE IN SOUTH DAITOTA PIG. 23. After Todd. covered the country. When the Dakota Glacier extended across North and South Dakota these hills were buried in the ice. At the time of the formation of the Leaf Hills Moraine, or when the Dakota Glacier reached as far south as is shown in Figure 20, the ice edge stood upon the hills south of Devils Lake, not being deep enough to flow over them. East of these hills it pushed farther south. The ice of the Dakota Glacier moved from the north in the direction of Lake Manitoba toward the south and a little east, and that of the THE GREAT ICE-SHEET IN NORTH DAKOTA. 53 Minnesota Glacier from the region beyond Lake Superior and south of Hudson's Bay toward the south and west. The two lobes of the Great Ice-Sheet thus met along the Pembina Mountain highland. It was in the hollow or ice valley between the lobes that the Glacial Elk River flowed, at first probably on the top of the ice, and later formed what is now known as the Elk Valley. It was this great glacial river which carried down the sand and finer rock-flour which made the Elk Valley Delta, from about McCanna and Larimore south to Portland. The Work Done by Moving Ice. Let us now inquire as to the effect of a great moving mass of ice upon the land-surface it passes over. If there are rough places on the rock surface these will be ground off and smoothed, and the fragments which are torn away will be shoved or carried along with the moving mass. The rubbing of the moving ice FIG. 24. A Small Hill Being Planed Down by the Ice. serves to give a peculiar polish to the stones carried in it. Such smoothed and polished rocks are very common among glacial gravels and boulders. In fact, nearly all the boulders in the fields are smooth, at least the sharp, angular corners have been rounded, and many of them are distinctly polished. It is common also to find boulders and pebbles not only smoothed but having straight lines cut in their sur- faces. These lines have been caused by the stone being shoved against another hard rock. Boulders or pebbles having marks made in this manner are said to be "striated," and the fine lines or furrows are called "striae." Boulders or fragments which are carried or shoved along the bot- tom of the ice upon a hard rock floor will indeed receive severe treat- ment. Not only will their rough corners be ground off, but any except those which are very hard will be likely to be ground to powder. Much 54 THE STORY OF THE PRAIRIES. FIG. 25. Showing Formation of Moraine, and Stratification of the Ice Photograph by Prof. T. C. Chamber tin. FIG. 26. Showing Moraine, which is being Crowded upon by the Moving Ice. Photograph by Prof. T. C. Chamber -tin. THE GREAT ICE-SHEET IN NORTH DAKOTA. FIG. 27. A Striated and Polished Boulder. Photograph by M. B. Erickson. of the clay of our fields is rock-flour thus ground by the great glacier- mill. While these rock fragments which are carried along by the moving ice are being thus ground to powder, what is the effect upon the under- lying bed-rock? It must be getting a pretty hard scouring! Figure 29 is a photograph of striae on a surface of hard quartzite rock in South Dakota. If a hummock or little hill lies in the path of the glacier, and if its width and height are SO' great that it cannot be broken off, then the ice will surround it and flow over it. The hummock will be combed and rasped by the ice and by the pieces of rock which are being car- ried in it. If the hummock should withstand the harsh treatment, when the glacier disappears by melting and leaves the once ice-covered land- scape, the little hill or hummock may look something like A Figure 30 or like B Figure 30 the ice having moved in the direction of the arrows. The Turtle Mountains furnish a good example in our own State of a large and broad "hill" which was covered by the ice and "veneered." FIG. 28. Granite Pebble, Showing Ice Planing and Striae. Drawn by Miss Jessie Dawson. 56 THE GREAT ICE-SHEET IN NORTH DAKOTA. 57 FlG. 30. Hills Worn Down by Action of the Ice. The "Mountains," so-called, really are not mountains at all, but a plateau. Before the ice-invasion this plateau looked something like A Figure 31 standing upon the prairie like a great, broad biscuit on a table or floor. After the ice had passed over it, it looked more like B Figure 31 which is about as it appears to-day. The steep side at the left is near Bottineau and the section extends northeast across the In- ternational Boundary. Devils Heart Hill and Sully's Hill south of Devils Lake are "veneered" hills. East of the Missouri River many long hills with FlG. 31. Ideal Sections of the Turtle Mountain Plateau, A before, and B after, being Cro&be d by the Ice-sheet. smooth outlines have a core of stratified rock, but have been combed across by the ice and strewn with boulders and finer glacial gravel and sand, and so are "veneered" with drift. The great fertility of the Red River Valley and the eastern part of our State comes not alone from the fact that the land in the Valley 58 THE STORY OF THE PRAIRIES. was once covered by a lake, but we have inherited a large amount of limestone, in the form of soil, from the limestone beds in western Mani- toba. This limestone has been ground to powder by the ice as it shoved it along. We have noticed that the Dakota Glacier moved south and a little east from the region about Lake Manitoba west of Winnipeg. This gave the Red River Valley and the eastern portion of the State a valuable "shipment" of the best wheat-producing limestone soil from our neighboring Province to the north. This pulverized and ground limestone is the best and most fertile known for wheat grow- ing- Advance and Retreat of the Ice Front. In the chapter on the Glaciers of Norway something was said about the advancing or pushing ahead of the front of a glacier due to. the movement of the ice being greater than the melting, and again the ice melting away more rapidly than it flowed down, causing a retreating or moving backward of the edge of 3 FIG. 32. A Veneered Hill, Ideal Section of Mauvais or Big Butte. the glacier. We may now apply what we saw then to the great Con- tinental Glacier. If melting were for a season more rapid than the onward movement of the ice, then the edge of the ice would slowly retire backward and leave its supply of earth, sand, gravel, boulders and clay to show where it had been. If the edge had stood for some time at one place there would be a long heap or ridge of materials forming what has been called a moraine. If the ice melted back somewhat rapidly there would be scattered boulders, gravel, sand and clay over the area between the moraine and the ice front. If the forward movement of the ice and the melting should now balance for a time so that the ice front became stationary again, here would be formed another morainic ridge. If this should occur again, this formation of a moraine would be THE GREAT ICE-SHEET IN NORTH DAKOTA. 59 repeated, and so there might come to be a series of morainic ridges more or less nearly parallel to each other. If, however, the ice should move ahead more rapidly than it melted away at the front, the ice would O'verride these ridges, leveling them down and pushing their materials .along. This melting back and pushing ahead have occurred a great many times, as a study of the terminal moraines of our State and of other Northern States show. Such advance and retreat of the ice front would tend to cause the terminal moraine to become not a simple line or long heap of earth and stones, but a belt of such materials. And as not all the earth and rock of the hills and ridges would be shoved along in front of the advancing ice but would be run over by the ice, the depth of the material in a moraine-belt becomes often very great, and so much material piled up in front of the ice would act as a dam to the on-flowing ice and hinder its advance. The terminal moraines which mark the places where the edge of the Great Ice-Sheet stood are not merely ridges of earth and rocks, but are belts of ridges and hills. The hills may be of all sizes and all heights up to 150 feet or even 200 feet. Between theni are little hollows and large hollows, "kettles," they have been called, sometimes containing water, sometimes dry, and sometimes what have been lakes have given place to marshes or "hay-meadows" by the blowing in of dust and the continued growth of rushes and water plants till the lake has been filled. Sometimes the hills are long, graceful swells, and sometimes their sides are very steep. So also the hollows may be round or they may be elongated and irregu- lar, and they may be deep or shallow. Sometimes the hillsides are strewn thickly with boulders, and sometimes no pebble larger than a toy marble can be found. Figure 33 is a photograph taken in "The Hills" southwest of Minot in Ward County. Boulders are seen scattered in abundance over the hills. In the foreground is a patch of boulders which have been brought together from far away. Limestones, granites and quartzites are here side by side. The limestones came from over in Manitoba, perhaps a hundred miles away. The granites and hard quartzites may have come from much farther away, possibly 200 miles or more. Such boulders have sometimes been traced back to their parent ledges over a distance of more than 300 miles. The hill at the right where the carriage stands is one of the highest, if not the highest, in this section of the State. It can be seen from 60 THE STORY OF THE PRAIRIES. more than twenty miles distant on the prairie. Many lakes of small size and hay-sloughs can be seen from its crest. The smoke rising from the chimneys of the shops at Minot can be seen also twenty miles away. To the left of the centre of the picture is a small circular lake now nearly filled so that it is a marsh. Two others can be seen, one at the right and one at the left on the margin of the picture. These are the "meadows" from which the ranchmen get their supplies of hay during summers when there is not too much rain. They are lakes dur- ing wet seasons. If we imagine that the ice pushed ahead, leaving its burden of earth and stones and then in turn melted more rapidly so that the edge of the FIG. In the Hills Southwest of Minot. ice was farther back; and if we imagine that it, so to speak, stood still here for some time so as to leave another mass of earth and stones; and if again the ice should advance and plough through and over the nearer masses of morainic material, and this process should be repeated again and again, when the ice should have finally all disappeared and left the landscape to become covered with plants and trees, we should expect that a very rough and hilly landscape would be the result. And if, as has been suggested, the materials piled up at the ice edge stood in the way of the forward movement of the ice, the tongues of ice would push out where there was less material in the way, and this would help to form the irregularities such as we now see in terminal moraines. CHAPTER THE SIXTH. MORE EXCURSIONS. Shore Boulder Chains. We may now understand better perhaps why the soil changes in character so much in going short distances. A farm may be located in a rnorainic region and its soil be stony, gravelly or sandy, or all of these, and it may be very hilly and rough, with small lakes or sloughs and marshes. Another farm only half a mile away, or even only a few rods distant, may be nearly level, of fine black loamy soil, and almost entirely free from stones. Still another may have a gentle slope or undulating surface, with almost no stones, but the soil may be very sandy, so that when the wind blows it may drift into dunes or heaps of sand. The first farm may be on what was the land barrier or moraine which hemmed in a temporary lake on one side; the second, where was once the deep water of the lake and hence received the fine sediments ; and the third may be on what was a delta in the lake. Sometimes again a chain of boulders may lie in great collections along some parts of a farm or section of land, and other parts near by be entirely free from such boulders. Such chains of rocks are often seen along the shores of lakes, especially of lakes whose waters are shallow. During cold "winters such lakes freeze to their bottoms. And lakes which are deeper in some parts and so do not freeze to their bottoms will freeze to their bottoms in, the more shallow parts nearer shore. If rocks and boulders are lying on the bottom, these become frozen into the ice. The sheet of ice cracks and breaks during the winter and the cracks become filled with water and this freezes and in freezing ex- pands, and so the ice sheet covering the whole lake becomes larger and it therefore shoves outward upon the shore. In so doing the bl-jcks of stone and boulders which were frozen into the bottom of the ice are shoved toward shore with the ice. This not only moves them a little way shoreward, but it serves also to loosen them from the bot- tom. When the ice "breaks up," in the spring, these rocks will be carried with the floating ice cakes until by melting of the ice they are again dropped. Whichever way the prevailing winds bk>w the ice 62 THE STORY OF THE PRAIRIES. cakes will tend to be moved and the rocks with them. The result is that the boulders are moved toward the shore in the direction of the prevailing wind. Winter after winter they are caught by the ice and shoved and carried a little way toward shore. Finally they are stranded along the bottom near the shore. Then they are frozen in and shoved up on the shore by the expansion process spoken of until finally there is a great chain of rocks and boulders piled along the shore, shoved up above the water's edge and left there by the melting of the ice. Hence if often happens that there is a great shore chain of rocks piled along the windward shore of a lake, as though they had been hauled there and dumped by some titanic force. Such chains of boulders were sometimes piled along the shores of glacial lakes, and when these lakes disappeared and the lake bottom became a dry field here were left the boulders to mark where once had been the lake. Boulder-Strewn Prairies. There are many places where boulders of all sizes are scattered over the land in great numbers. Great blocks weighing many tons often lie upon the prairie as though they had been dropped there by some gigantic force. Sometimes the land is strewn with boulders so that one can walk for a considerable distance with- out stepping upon soil at all. Large and small sizes, and different kinds, granites, quartzites, limestones and others, appear as though they had been carried there and thrown down. Just how these boulders, these huge masses and the smaller blocks, FIG. 34. A Huge " Foreigner." Photograph by Prof. Chas. M. Hall. MORE EXCURSIONS. 63 came to be distributed just as we now find them we need not now trouble ourselves about, only to observe that they are all "drift" boul- ders or "foreigners," and that they have been transported from some other place by the great ice-sheet, and when the ice melted they were left just where the ice happened to drop them. Their corners are nearly always rounded and their surfaces smoothed by the rubbing and grinding of the great ice-mill in which they were carried. Buffalo Boulders, It quite frequently happens that a large boulder lies in the center of a small basin or hollow, as though the basin had been dug around the rock. Such hollows are usually not large, ex- tending only a few feet each way from the stone. This has suggested the idea that buffalo, wandering in herds over the once unbroken prairie, rubbed their bodies against the sides of the rock, and in tread- ing about it ploughed up the soil. Loose soil is easily carried by the wind, and so the hollow might easily have been formed by the joint action of the hoofs of the buffalo and the wind. The rocks are sometimes polished on their sides with a sort of greasy polish, but no such thing is seen on the tops of the boulders beyond the reach of the animals' heads. Sometimes when these hollows become quite deep, or are on low ground, water collects in them during wet seasons and they become "buffalo wallows." When this is the case the soil would be carried away on the bodies of the animals. Stratified Gravel and Sand in Sand-Pits. Probably all have seen a gravel- or sand-pit. Here the little fragments of stone we call gravel or sand are arranged in beautiful layers, one above another like the boards in a lumber pile. Some of the layers are very thin, perhaps only a small fraction of an inch in thickness, and again they are several inches thick, or even several feet. Occasionally, also, a boulder is found imbedded in the layers. The size of the particles in any particular layer or stratum it is noticed is about the same, though the next layer above or below may be much finer or coarser. If we follow the line oi'9~*--:'-i'.-?-~:.'~'-^Z 500 FIG. 43. Section Across the Red River Valley on the Latitude of Breckenridge and Wahpeton. Horizontal scale, 25 miles to an inch. U. S. Geological Survey. 90 THE STORY OF THE PRAIRIES. * - * * ? * * ' ** " * " * L " v " V v . * ' * J.^ FIG. 44. Section Across the Red River Valley at Fargo. After Upham. FIG. 45. Section Across the Red River Valley at Grand Forks.- After Upham. FIG. 46. Section Across the Red River Valley near International Boundary. After Upham. the highest Herman Beach is shown in Figure 42. A section across a beach ridge is shown in Figure 41. The ridge is made up of sand and gravel arranged in layers. Underneath the ridge is the boulder-clay called "till," the unstratified drift which underlies the materials nearer the surface which were arranged in layers by the waters of the lake. Sections across the Red River Valley at Wahpeton, Fargo, Grand Forks, and along the International Boundary, are shown in Figures 43, 44, 45, 46. These sections show the till or boulder-clay under- lying the wave-washed materials, and underneath the till the layers of the stratified rocks, the top of which was the land surface, the pre- glacial landscape, before the great ice-sheet spread over the land. The upper portion of the clay which makes up the deeper sub-soil of the Red River Valley is arranged in layers, as is shown in Figure 47. This is due to the fact that the upper part of the drift clay of the Red GLACIAL LAKfc AGASSIZ. 91 River Valley was deposited in the water and spread over the bottom of Lake Agassiz. This material was dropped from the melting ice while the ice-sheet was receding and the lake was increasing in size, and was washed by the waves and deposited in layers upon the bottom. FIG. 47. Stratified Clay, Sediments of Bottom of Lake Agassiz. Excavation in City of May vilie. Photograph by the Author, CHAPTER THE NINTH. THE DELTAS AND BEACHES OF LAKE AGASSIZ. Three deltas were formed on the western side of Lake Agassiz on that part of the bottom now embraced in North Dakota. These are known as the Sheyenne, the Elk Valley, and the Pembina Deltas. They were formed by the Sheyenne River, the Glacial Elk River, and the Pembina River. These rivers were flooded by the waters from the melt- ing ice-sheet, and when their swift currents entered the still waters of Lake Agassiz their speed was checked and they threw down the burden of materials they were carrying, the coarse gravel and sand first, and later the fine sand and silt. The finer sand was carried for many miles into the lake and spread out as a great fan, and the finest silt was spread over all the bottom of the lake, being distributed by the waves and currents. Not all the streams which flowed into Lake Agassiz formed deltas. It is interesting to inquire, therefore, why the Sheyenne and Pembina Rivers, and also the glacial Elk River, which ceased to be a river at all after the ice-sheet had melted away, should have formed deltas^ while other streams flowing into Lake Agassiz formed no deltas. We have seen that much earth material was carried by the ice, and that much water flowed away from the edge of the ice-sheet from the melting. If a river had its head near the edge of the ice-sheet, or flowed along its edge so as to receive these waters, then whatever gravel, sand, and earth the ice contained might be in considerable part carried to the river. Some parts of the ice-sheet probably carried more gravel and sand than other parts, depending upon the kind of land surface it had passed over. Then, too, the edge of the ice-sheet was very irregular and indented by jagged places made by the melting, and so there would be many small hollows and lakes in which the earth materials from the ice would be deposited, so that not all the streams which flowed at flood height from along the ice-sheet's edge received such great burdens of gravel and sand. When, therefore, a river had its head near a portion of the edge of the ice-sheet where a good deal 9-2 THE DELTAS AND BEACHES OF LAKE AGASSIZ. 93 of sand was left so that it was washed into the river's channel this stream, having a swift current because its channel was kept flooded, would carry much sand and gravel down its course. FIG. 48. Profile of Elk Valley Delta and Beaches at Larimore and Arvilla. After Upham. Again the Elk River was a stream which at first probably flowed on the surface of the ice-sheet in the hollow between the Dakota and Minnesota Glaciers (see Figure 20), though it later formed a channel in the drift between these Lobes, and this river formed a delta thirty to thirty-five miles long and from five to twelve miles wide. In Minnesota, of the rivers entering Lake Agassiz, only the Buffalo and Sand Hill Rivers formed deltas, although the Red River, the Wild Rice River, and the Red Lake River, on that side of the lake were as large or even larger than these. The Sheyenne Delta. When the ice-sheet had receded so that its edge rested upon the high hills south of Devils Lake, the Sheyenne River received a great influx of water from the melting ice, and with it the finer materials which were in the ice. These were carried by the stream down its course. The water was muddy, something as the waters of our streams now are muddy after a hard rain, or when they are swollen from melting snows. Not only this, but sand and gravel which were too coarse to be carried any distance by the current would be rolled along the bottom, or taken up and carried for a short distance and thrown down again, perhaps forming a sand-bar, to be in turn taken up and carried on again by the varying current. So at the mouth of the stream where the current met the still waters of the lake these were thrown down, first the coarser gravel and then the finer sand. These became the delta. Little by little the river kept adding more materials to the delta, the coarser being dropped nearer the shore or head of the delta, the finer being carried farther out, and the finest, which would remain in sus- pension in the water for a long time, being carried far out and dis- tributed over the lake bottom as the so-called lacustrine silt. 94 THE STORY OF THE PRAIRIES. The delta is made up mostly of sand and gravel arranged in layers. Whoever has traveled along the lower Sheyenne River south and east of the Big Bend has noticed how sandy is the soil, also the hills along the river and over great areas farther from the river. The sand has been blown and piled into heaps by the wind, forming the "dunes" which are a conspicuous landscape feature. The Sheyenne Delta covers an area of about 800 square miles, be- ing mostly in Ransom and Richland Counties, but extending also into Cass and Sargent Counties. From the Big Bend eight or nine miles below Lisbon the Sheyenne River flows north for ten miles along the western edge of the delta, then flows east and north across its surface, leaving the delta front about three miles south of Kindred. The town of Sheldon is located on the western edge of the delta plain, and its western edge extends from here north about three miles into Cass County. Thence the northern edge extends eastward a little to the north of Leonard, and eastward and southward near Walcott, Colfax, and Barrett, on the 'Great Northern Railway; thence south to Moore- ton, on the Milnor branch of the Northern Pacific Railway, and a little east of Hankinson to the Lightning's Nest, a very large wind-blown sand-hill or dune. The southern and western edge of the delta extends from the Lightning's Nest west and north by Taylor, Willard, and Swan Lakes, to a point about four miles northeast of Ransom in Sar- gent County, and from here northwest to Milnor and the bend of the Sheyenne River. Much of the surface of the delta is now marked by wind-blown sand piled into dunes of all sizes from little choppy knolls two to four feet high to large hills fifty to one hundred feet high. FIG. 49. Section Across the Sheyenne Delta After Upham. Along the northeastern front the waves of Lake Agassiz cut a cliff or bank, so that in approaching the delta from the northeast the land- scape rises suddenly in passing from the adjoining prairie onto the delta plain in some places as much as seventy-five feet. Figure 49 shows a cross section of the delta in which the valley of the Sheyenne River is shown at the left, and near this the Herman Beach, which THE DELTAS AND BEACHES OF LAKE AGASSIZ. 95 marks the highest level of Lake Agassiz. A tract of dunes more than ten miles across is near the center of the cut, and the steep delta front si,xty to seventy feet high is shown at the right. FIG. 50. Delta on University Campus, Chicago. Photograph, 1894, by the Author. Figure 50 is a photograph of a small delta which was formed during a single night. The current of water from under the sidewalk was slackened as it poured out upon the low fiat area in the foreground, and the materials carried by the little stream were thrown down layer upon layer in the same manner as the sand and gravel of which the great Sheyenne Delta is composed. The Pembina Delta. The Pembina Delta was formed by the Pem- bina River after the ice-sheet had melted back so as to leave the Pem- bina Mountain uncovered, the delta lying along the foot of that Moun- tain. The delta plain rises quite abruptly from the level prairie of the valley bottom to the east, and is locally known as "First Pembina Mountain." It covers an area of about eighty square miles, or only about one-tenth of that of the Sheyenne Delta. Its average depth is estimated to be about 150 feet. The average depth of the Sheyenne Delta is estimated to be about forty feet, so that the volume of the Pem- bina Delta is more than one-third that of the great Sheyenne Delta. The materials of which the Pembina Delta is composed are not only sand and gravel brought from the melting ice-sheet, but shale from 96 THE STORY OF THE PRAIRIES. the underlying rock-formations of Pembina Mountain, into which the river has cut a very deep valley, and also pebbles of granite and other hard rocks up to six inches in diameter. Large boulders of granite lie upon its surface, dropped perhaps from blocks of floating ice from the lake. The delta extends from the foot of Pembina Mountain about four miles south of the International Boundary east and a little south to near Walhalla, thence curving south and east to its widest point, and south and west to the foot of Pembina Mountain again a mile south of Tongue River, being about eight miles wide at its widest part. Its western boundary thus lies along the foot of Pembina Mountain. Its highest point is about six miles southwest of Walhalla and a little more than a mile south of the Pembina River. It is here 1,270 feet above sea-level. The highest, or Herman, shore-line of Lake Agassiz is about two miles east of this point, and about fifty feet lower. This shows that the river piled its burden of sand, gravel, shale, and pebbles up to a height at the head of the delta greater than that of the level of the lake. The surface of the delta slopes gradually to the north, east, and south from this highest point or head. Along the foot of the delta front run the Norcross, Tintah, Camp- bell, and McCauleyville Beaches, marking the height of the waters of Lake Agassiz during those stages of the lowering of the lake imme- diately following that during which the delta was formed, the highest or Herman Stage. The waves of the lake washed against the front or edge of the delta plateau and eroded the loose materials, forming a steep bank or wave cliff which on the northeast side of the delta is more than 150 feet high. In crossing the delta from the level prairie east of Walhalla to Olga, about twelve miles southwest, after crossing well marked McCauleyville Beaches, the road rises suddenly up the steep face of the wave-washed and tree covered cliff 150 feet, from the top of which the surface of the delta plain spreads out as a great undulat- ing plain with scattered clumps of trees here and there. From Beau- lieu on the delta plain the road leads up the steep face of Pembina Mountain (called Second Mountain, to distinguish it from the delta plateau which is called First Mountain) a height of about 300 feet. The outcropping Cretaceous shales are exposed by the roadsides and in the coulees, and drift boulders of granite are scattered upon its sides. Where the Pembina River cuts across the crest of Pembina Moun- tain the valley has been cut 350 to 450 feet into the soft shales and THE DELTAS AND BEACHES OF LAKE AGASSIZ. 97 clays which underlie the drift, and tributary streams which have also eroded deep valleys, give to the landscape, which is covered with trees, a wild and picturesque appearance. The delta plain or plateau is also much cut up by streams. The Pembina and Little Pembina Rivers have cut deep gorges in the delta, even cutting down into the till which underlies the delta and on which it was built upon the lake bottom, so that a section through the 150 feet of delta sand and gravel is shown. The Cretaceous shales and clays (these belong to the Fort Pierre group of the Cretaceous series) are well exposed in the sides of the valley of the Pembina River where it cuts through the (Second) Mountain, and FIG. 51. Section Showing Stratified Sand of the Elk Valley Delta. Erosion by Tributary of the Goose River. Photograph, iqoo, by M. B. Erickson. 98 THE STORY OF THE PRAIRIES. farther down its course the layers of the delta sands and gravels are similarly exposed. It is worthy of notice here that what seems to be a small "butte" stands about a mile north of the northern end of the delta and three miles south of the International Boundary, a half mile east of the face of Pembina Mountain. It has much the appearance of the small rounded buttes of the Bad Lands. It looks from a distance much like a large haystack, being thinly covered with grass. Badger holes near its top and on its sides showed clean shale such as that of the Mountain, If this is its true character it is an outlying fragment of Pembina. Moun- tain, and so was a tiny island in Lake Agassiz when its waters washed the eastern face of the Mountain. It is interesting also as being the most eastern "butte" in the State, and perhaps in the United States. The Elk Valley Delta. The Elk Valley Delta covers an area of about 300 square miles, extending from McCanna east of Larimore and south to Mayville and Portland, and covering the area west to the shore of Lake Agassiz. No river which could have formed this delta now exists. The stream which formed it, the glacial Elk River, is no more. The reasons for thinking that such a stream did once exist are found in the structure of the delta itself, the materials of which it is composed, and the form- of the landscape near to the delta, The delta is higher north of Larimore and its surface slopes gently toward the south, west, and east, as though the "head" or place where the materials of which it is composed were poured into the lake was at this point. The mate- rials making up this delta are of, a finer character than those of the Sheyenne and the Pembina Deltas, being mostly fine sand and silt brought from the ice of the great ice-sheet, and are not mixed with shale graVels from the Cretaceous rocks underneath the drift, as are those of the other deltas. And then, extending north from Larimore and McCanna, just where a river ought to have been to have formed this delta as it is, the broad flat bottomed valley extending for more than forty miles to Edinburg and Gardar is what is known in its southern portion as Elk Valley and farther north as Golden Valley. It is, however, all one valley, varying in width from about four miles along the greater part of what is called Elk Valley to two miles at Ram- sey's Grove, where begins the part, called Golden Valley, and this por- tion varies in width from one to two miles. Figure 20 shows the positions of the Dakota and Minnesota Gla- ciers, or lobes of the Great Ice-Sheet, at the time of the formation of THE DELTAS AND BEACHES OF LAKE AGASSIZ. 99 FIG. 52. Angular Outlines, not Passed over by the Ice-sheet. . Photograph by Prof. T. C. Chamber lin. FIG. 53. Smooth Outlines, Showing Effects of Moving Ice. Photograph by Prof, T C. Chamber lin.' 100 THE STORY OF THE PRAIRIES. the Elk Valley Delta. At this time the Leaf Hills Moraine was formed at the edge of the ice. Remember what has been said about the thick- ness of the ice-sheet, and that the surfaces of these lobes were higher along their axes or centers and the ice thinner near the edges. The arrows indicate how the ice spread out or flowed toward the south, east, and west, near the southern ends of the lobes. There was melting of the ice on the surface of the ice-sheet as well as at the edge. Water would therefore collect in the hollow along the line where the two lobes met. When the ice-sheet reached farther south, as at the time of the formation of the Fergus Falls Moraine, the ice extended across from one lobe to the other, in the region shown in Figure 20. As the ice melted and the edge came to be farther back a hollow came to be upon the surface of the ice-sheet. At the time the Leaf Hills Moraine was being formed a large stream flowed in the hol- low where the two glaciers met, having its bottom and sides of ice. This was the glacial Elk River. Soon the ice valley became deeper from the melting due to the stream and from the melting at the edges of the lobes. The sand and silt which were elsewhere left at the edge of the ice as moraines were washed away by the swiftly flowing river. This was added to the ma- terial of the delta. In time, however, this glacial river came to flow upon the ground between the two glaciers, being kept at high flood by the waters from the melting ice, which poured in from both sides. As the ice of the two glacier edges on each side of the hollow kept moving toward each other, and each delivered its burden of sand and silt, a large amount of earth material was left along the course of the stream only to be quickly carried away by the rapid current of the stream which was constantly renewed by the inpouring of waters from the ice. So it would seem that the conditions must have been such as to form a large river bur- dened with a great load of earth, and when the still waters of the lake were entered a delta must result. FIG. 54. Profile of " the Ridge " and Beaches at Inkster. After Upham. THE DELTAS AND BEACHES OF LAKE AGASSIZ. 1Q.1. There was, however, more drift piled into this valley from the melt- ing ice than could be carried away by the river. The wdst side of the valley is the highland of Cretaceous rocks which for,nred >; tne we-stenv boundary of Lake Agassiz. Along the top of this hi-ghla/id was' left the. Leaf Hills Moraine of the Dakota Glacier. On the east side of the valley the Minnesota Glacier piled its moraine, a chain of hills which is now locally known in its southern portion as "The Ridge" and the northern part as "The Mountains." This chain of hills extends from McCanna north to Edinburg, a distance of about thirty-five miles. "The Ridge" is a series of three morainic hills from one to three miles in length and from a half mile to three-fourths of a mile in width. "The Mountains" are two long, large hills, one about six miles long, lying west of Conway, the other about fourteen miles long and two to three miles wide, lying west of Park River and extending north to Edin- burg. After the Leaf Hills stage of Lake Agassiz, when the ice-sheet had receded to the position of the Itasca Moraine, this chain of hills formed islands in Lake Agassiz, and the valley of the Elk River was a great sound or strait between these islands and the western shore of the lake. This is shown by the beach ridges which mark the height of the water on the sides of the islands and the west shore of Elk Valley. Figures 54 and 55 show profiles across the Elk and Golden Valleys, the Ridge and the Mountains, and the upper beaches of the Lake. Tlie Mount a\ns" Island. I Mo-1-airnC.Ti U /^arx^S^H.&Hcvr FlG. 55. Profile across Beaches at Park River and Westward. Horizontal scale, 3 miles to an inch. After Upham. The northern mountain is crossed by the south branch of Park River, west of the city of Park River, in a well-marked valley. Farther south to the west of Conway and Inkster is a gap two or three miles wide between the southern mountain and the northern hill of the ridge. The three branches of Forest River send their waters through this gap after they have united into one stream, cutting across the beach ridges 102 THE STORY OF THE PRAIRIES. which extend along the east side of the chain of hills. West of Orr the ridge is broken in two, but Lost Creek, which is formed by several small tributaries from the higher land west of the shore of Lake Agas- si?:, fails' to flow across but becomes "lost" on the flat, marshy prairie trie old sand and silt bottom of the Elk Valley. Between these islands were straits or necks of water connecting the main lake with the large sound west of the islands. The bottom of Elk and Golden Valleys is a level tract forty miles in length and from one to four miles in width with no stream on its bottom representing the great Elk River which once surged down its course and built the broad delta at its mouth. It has so little slope that no stream flows upon the level bottom for more than a few miles. In fact, Lost Creek, after it enters the flat bottom of this valley, struggles toward the north instead of south in the direction of the Elk River, and after two or three miles gives up and becomes a marsh. West of the northern mountain several small streams flow into the valley from the highland to the west and become "lost," spreading out into a marsh. The Pembina Delta was formed after the ice of the Great Ice-Sheet had melted back so that Lake Agassiz extended north beyond the International Boundary to the city of Winnipeg, but the lake remained at about the same level, for the same beaches which run across the eastern side and along the front of the Elk Valley Delta also cross the eastern side and run along the steep front of the Pembina Delta. And the Herman Beach, which marks the highest level of the lake, runs along the western or shore side of both deltas. And similarly the Nor- cross, Tintah and Campbell Beaches run across the eastern side of the Sheyenne Delta, and the McCauleyville Beach along its front, while the Herman Beach runs near its western or shore side. The highest or Herman stage of Lake Agassiz therefore continued during the sev- eral stages of "retreat" or melting of the ice-sheet, which are marked by the Dovre, Fergus Falls, Leaf Hills and Itasca Moraines. The stages of Lake Agassiz should, therefore, not be confused with the stages of retreat or melting of the ice-sheet. Stages and Beaches. It has been previously explained how Lake Agassiz came into existence by the hemming in of the waters of the melting ice-sheet by the higher lands which formed the sides of a great pre-glacial valley. These formed the shore boundaries of the lake on the east, west and south, while the great wall of ice formed its northern shore. Since the lowest place in the rim of the surrounding highlands THE DELTAS AND BEACHES OF LAKE AGASSIZ. 103 was at the south here was established the first outlet. And the waters must needs find escape to the sea to the south because the great ice- sheet prevented any drainage toward the north. The first great stage of the lake was begun when the ice had melted back to the position of the Fergus Falls Moraine. During this time the highest beach or shore line, known as the Herman Beach, began to be formed. As has been before explained the Sheyenne Delta began, to be built up as soon as the lake began, and its level had not changed much when the Elk Valley and Pembina Deltas were formed. The outlet of the lake was across the soft drift materials of the Dovre Moraine. Lake Traverse now lies in the north end of the old outlet channel, near the southeast corner of North Dakota and on the boundary between the states of South Dakota and Minnesota. The lake grew larger by the melting of the ice-sheet, or the "retreating" of the ice-wall which formed the northern shore. The water remained at the same height during all the time the lake was increasing in size, the outlet channel being cut down during the time five or ten feet. The beach which marks the next lower stage or level of the lake is the Norcross. At the time this beach was formed the level of the lake was about twenty feet lower than during the time of the formation of the Herman Beach, the outlet having been cut down this amount. The lake stood at this level for quite a long time, as is shown by the well- defined shore lines or beaches. Then the outlet was cut clown again about fifteen feet, causing a lowering of the lake this much below the Norcross stage. At this level the higher of two Tintah Beaches was formed, followed by another lowering of the water-level of about fifteen feet and the forming of the lower Tintah Beach. Again the level of the water was lowered about fifteen or twenty feet and the Campbell Beach was formed. And finally about the same amount of cutting down of the outlet brought the level to the lowest stage while yet the waters escaped to the south, the McCauleyville Beach being formed at this lowest level. Thus a beach was formed at each stage of the lake. The names of these beaches are a little awkward, and have no mean- ing except that they are names. They were applied to the beaches from towns which are built upon the beaches or which are near to them. The five names applied to the higher beaches of the lake are the names of towns in Minnesota. Other and lower beaches were named from towns in North Dakota and Manitoba, as the Blanchard, the Hills- 104 THE STORY OF THE PRAIRIES. boro, Emerado, etc., in North Dakota, and Gladstone, Burnside, etc., in Manitoba. The next lower stage than the McCauleyville was about twenty feet below the bottom of the southern outlet channel, and the melting of the ice at the north had allowed the waters to find escape by another outlet. At this time were formed the Blanchard Beaches, and it is known as the Blanchard stage of the lake. The outlet was probably to the northeast, the waters escaping into Lake Superior, thence to Lake Ontario, and by way of the Mohawk Valley and the Hudson River to the Atlantic Ocean. The ice had not yet melted off from the Valley of the St. Lawrence and hence escape of the waters by that course was impossible. It was noted above that during the time of the forming of the Her- man Beach the outlet channel was cut down only five or ten feet, al- though the water stood for a considerable time at this level. Then while the outlet was being cut down fifteen or twenty feet no shore line what- ever was formed. While the water stood at this second level, the Nor- cross stage, another beach was formed. Again the outlet cut down rapidly, leaving no beach ridges on the shores because the water did not stand at any one level long enough for the waves to pile up a shore ridge. This is the upper beach of the Tintah stage. Again the outlet deepens suddenly while no shore lines are formed, and then the water stands at the second level of the Tintah stage while the lower Tintah beach is forming. Then, still again is the outlet cut down rapidly to the Campbell stage, and the Campbell Beach. And finally another lowering of the outlet to the McCauleyville stage, when the last beach was formed while the waters discharged by the southern outlet. But the next level of the lake is below the bottom of the outlet. It was not, then, the cutting down of the outlet channel which caused these changes of level of the lake, for this outlet could not drain the lake below its own bottom. It is evident, therefore, that some other outlet had been found for the waters at a lower point in the rim of the lake. This occurred when the ice melted back at the north so as to un- cover a lower place in the surrounding highlands which kept the waters hemmed in. This, however, does not explain why the lake stood at certain levels long enough for the waves to build up distinct beach ridges while the outlet was cut down but little, and then the outlet cut down so rapidly that the waves left no shore marks at all. The outlet was changed and the old River Warren became an aban- THE DELTAS AND BEACHES OF LAKE AGASSIZ. 105 doned channel. This is shown by the fact that those beaches which were formed after the McCauleyville stage, the lowest stage while the waters were drained to the south by the River Warren, run across the axis or central part of the old lake bottom (where is now the Red River of the North) instead of running down along either side of the old channel, as do the McCauleyville and the higher beaches. O 4? FIG. 56. During the Higher Stages the Lake Outflowed Southward. The Lower Beaches Cross the Red River of the North. Figure 56 shows the relation of the higher beaches formed while the lake discharged toward the south and the first two (Blanchard) beaches formed after the lake had ceased to overflow southward and had formed a lower outlet into Lake Superior. The explanation of these rather remarkable things is somewhat dif- ficult, and those who' do not care to attempt to follow it may omit the next few pages. Causes of These Changes. The cause of these changes of level of the lake is a, somewhat difficult one to understand. It is no less a matter than changes in the form of the earth's crust, changes in the altitude or level of the surface of the earth itself. It has been observed that in fol- 106 THE STORY OF THE PRAIRIES. lowing the beach lines from south to north that they are not simple or single ridges at the north as they are in their southern parts, but they become double and multiple as they are followed northward. The Her- man Beach, for instance, which is a single ridge in its southern portion, becomes five distinct beaches near Maple Lake in Minnesota, and still farther north in Manitoba becomes seven distinct beaches. And simi- lar facts are observed on the west side of the lake. The five beaches near Maple Lake are separated from each other by vertical distances FIG. 57. Map of Portion of the Herman and Norcross Beaches, near Maple Lake, Minn., showing the Multiple Character Northward. The five Herman Beaches become one Beach, and the four Norcross Beaches one. of eight, fifteen, thirty and forty-five feet; that is, the highest Herman Beach is there eight feet higher than the next lower, that is, fifteen feet higher than the next lower than this, making the highest twenty-three feet above the third one, and this third one in turn is thirty feet higher than the fourth, making fifty-three feet from the highest to the fourth lower, and the fourth is forty-five feet higher than the fifth, so that the first or highest is ninety-eight feet higher than the fifth or lowest. And all these merge into the one single Herman Beach in the southern portion of the lake. Similarly the Norcross Beach, which is a single THE DELTAS AND BEACHES OF LAKE AGASSIZ. 107 FIG. 58. Diagram Showing the Progressive Elevation of Beaches Northward in Vicinity of Maple Lake, Minn. Continue the lines to the right and the upper five meet in one, and the lower four in one. beach ridge in the southern portion, becomes double at the north, as does also the Tintah, while the Campbell and McCanleyville Beaches each become separated into three distinct ridges at the north. The five stages of the lake, while it discharged its waters by the southern outlet, are represented in the southern portion by the five beaches named, the Herman, Norcross, Tintah, Campbell and McCau- leyville. These five beaches in the south are represented by seventeen beaches in the north. The highest, or Herman Beach, near the old out- let at Lake Traverse, is about 90 feet higher than the lowest or Mc- Cauleyville Beach, while the vertical distance between the highest of the Herman beaches, 300 miles to the north, and the lowest McCauley- ville Beach is nearly 300 feet. In traversing these beaches from south toward the north it is observed that they rise gradually northward. They were formed at the waters edge and were therefore in the first place level.* The ascent or rise is more gradual toward the south and more rapid toward the north. The uplift of the crust of the earth was, therefore, going on at the time Lake Agassiz was here and forming the beaches, and it, the uplift, was greater toward the north. The movement of elevation of the country at Lake Traverse during the time of formation of the five beaches while Lake Agassiz outflowed to the south was about ninety feet. On the International Boundary at Pembina Mountain it was 265 feet. At Gladstone, in Manitoba, about 350 feet, and 200 miles north of the International Boundary on the east side of Duck Mountain, nearly 500 feet. * The surface of the lake was not perfectly level, for the waters were drawn by the attraction of the great mass of ice toward the north, making the water "pile up" toward the north, and hence the shore lines would ris.e a little in going north, but for our study they may be considered as horizontal. 108 THE STORY OF THE PRAIRIES. To explain these remarkable changes of level it is necessary to con- sider a somewhat difficult geological problem, that of the changes of level of the earth's crust before referred to. This is the rising in one place and sinking in another, over large areas, or regional elevation and subsidence, called "epeirogenic movements," of the crust of the earth. That the form of the earth's outer layers or "crust" is not fixed or "solid" is a well established fact. The sea creeps upon the land, or withdraws from the shore as the land rises or sinks, very slowly, to be sure, but none the less truly. The movement is more easily recognized at the seashore because the sea-level forms a convenient base-line for mak- ing comparisons. It is thought that the great basin in which Hudson's Bay lies is being uplifted at the present time, probably a continuation of the same great movement by which the beaches of Lake Agassiz were lifted out of their level positions. This uplift of the basin of Hud- son's Bay has been estimated to be from five to ten feet in a century.* If the great weight of the vast ice-sheet caused the crust of the earth to bend down or sink, then the melting of the ice and the flowing away of the water would relieve the pressure and so allow it to rise again. The ice was deeper at the north and the rise of the land, as we have seen, was much greater at the north. The Herman stage of Lake Agassiz represents that period of the lake during which all the beaches at the north which unite into the one Herman Beach near the outlet at Lake Traverse were formed. But during all this time the water was pouring out at the Lake Traverse outlet without cutting the channel down very much, which means that the current was not very swift at the outlet. The elevation at the north may be likened to the slow tipping of a broad pan or dish filled with water so as to just keep the water steadily flowing out at the side. But then there followed a more sudden and widespread elevation which affected the whole area of the lake. The whole basin was lifted up, which had the effect to increase the rate of flow of water at the outlet, and so the channel \vas cut down rapidly to the level of the next stage of the lake, the Norcross stage. Here the same process was repeated, the outlet staying just about the same during the time that the several Norcross Beaches were being formed at the north. These beaches, like those of the Herman stage, unite into one in the southern portion of the lake, showing that the * Dr. Robert Bell. THE DELTAS AND BEACHES OF LAKE AGASSIZ. 109 uplift during this stage did not extend to the southern end of the lake. The close of the Norcross stage is marked by another comparatively sudden uplift of the whole lake bottom, followed again by the rapid cuttino- down of the outlet channel. o This series of changes, viz., the uplifting of the northern portion of (the lake area during the time of each stage while the outlet remained at 'just about the same depth, followed by a somewhat sudden uplifting of the whole region of the lake so that the water passing through the outlet channel increased in speed so as to cut down its depth a consider- able amount, to the level marking the next lower stage, continued dur- ing the five great stages while the outlet remained at the south. The two Tintah Beaches at the southern outlet mark substages, there being a lowering of the outlet between the two periods of the Tintah stage when the two* beaches were formed. Finally, at the close of the McCauleyville or lowest stage of the lake while the outlet remained at the south the uplifting of the bottom coin- cided with the uncovering of a place in the rim of the lake lower than the bottom of the Lake Traverse outlet, and so the outlet was changed to the northeast. The several beaches at the north which belong to one stage and which unite to form one at the south, mark intervals of quiet or pauses in the uplifting which affected the more northern region only and not the whole area of the lake. This means that the uplifting was progres- sively greater toward the north. The succeeding beaches, which mark the stages of the lake after the water had ceased to be discharged by the southern outlet, are three Blanchard Beaches, representing three stages of the lake, each being lower than the preceding, the first being fifteen feet lower than the McCauleyville Beach, the second twenty feet lower than the first, the third fifteen feet lower than the second, the Hillsboro twelve or fifteen feet lower still, the Emerado thirty feet, the Ojata twenty-five feet, the Gladstone twenty feet, the Burnside twenty feet, the Ossawa fifteen feet, the Stonewall twenty feet, the Niverville forty-five feet/ and from the Niverville Beach still another fall of forty-five feet reaches the earliest level of Lake Winnipeg, and the cutting down of the Nelson River outlet has lowered Lake Winnipeg still further twenty feet. Let us now briefly review the history of Lake Agassiz. The lake first began as a body of water from one to three miles wide and about thirty miles long, and was little more than a broadening of the Shey- 110 THE STORY OF THE PRAIRIES. enne River. The melting back of the ice-sheet to the position of the Fergus Falls Moraine increased the size of the lake and the first and highest Herman stage of the lake was ushered in. When the ice melted back to the position of the Leaf Hills Moraine it became still larger; and again the rapid recession of the ice to the Itasca Moraine increased its area still further. And when the Mesabi Moraine was formed the lake extended to the southern ends of Lakes Winnipeg and Manitoba, and still later embraced all the vast territory adjacent to these lakes. Most of the melting away of the ice occurred during the time of the formation of the Herman and Norcross Beaches, as these beaches have been traced from Maple Lake, Minnesota, south to Lake Traverse, and north through North Dakota to Duck Mountain in Manitoba, a distance of more than 700 miles. The deltas which have been described, the Sheyenne, Elk Valley and Pembina, and also the Buffalo and Sand Hill Deltas in Minnesota, and the great Assiniboine Delta in Manitoba, were formed mostly dur- ing this earlier time of the lake, as they are crossed by the Herman and Norcross Beaches, whereas the others which mark lower levels of the lake mostly pass around them, leaving them to the landward. The changes in level of the lake were caused by changes in the form of the earth's crust, an uplifting of the floor of the lake causing more rapid cutting down of the outlet and draining away of the water, the successive stages or levels of the lake being marked by shore lines or beach ridges. The northern portion was uplifted more than the south- ern portion, as is shown by the beaches which become double and mul- tiple at the north. Finally the floor of the lake was uplifted so that escape of the waters by the southern outlet was cut off and the waters overflowed to the northeast, the ice melting at the north so as to allow the waters to escape by a new outlet at the same time the outlet to the south was elevated. Successive stages in the level of the lake are marked by beaches. At the time of formation of the Gladstone Beach the southern point of the lake was about as far south as Buxton, the Red River of the North flowing into the lake there. The western shore of the lake in North Dakota is marked by the Gladstone Beach west of Grafton and Minto. At the time of the formation of the Niverville Beach the lake did not extend south of the International Boundary, and the Red River of the North flowed into the lake near Morris, Manitoba, twenty-five miles north of Neche and. Pembina. The entire area covered by Lake THE DELTAS AND BEACHES OF LAKE AGASSIZ. Ill Agassiz was about 110,000 square miles, or an area equal to more than one and a half times the whole State of North Dakota, and the greater part of this vast expanse was covered during the highest or Herman stage of the lake. The depth of the waters of Lake Agassiz above the present surface of the south end of Lake Winnipeg during its higher Herman stages was about 600 feet. At the time the waters ceased to discharge by the southern outlet and began to overflow toward the northeast the depth at this point was about 300 feet. At the time of the Niverville stage, the last before the waters fell to the highest level of Lake Winnipeg, the depth was about sixty-five feet. Finally the ice disappeared, uncovering the present Nelson River outlet and the waters lowered to the highest level of Lake Winnipeg, and then by the cutting down of the Nelson River channel the waters were lowered to the present level of Lakes Winnipeg, Manitoba and Winnipegosis, which remain as a last vestige of the once great Lake Agassiz. CHAPTER THE TENTH. OTHER EXTINCT GLACIAL LAKES. Glacial Lake Souris. Glacial Lake Souris occupied the Valley of the Souris, or Mouse River, from which river it gets its name. It was formed by the waters from the melting ice-sheet, as was Lake Agassiz, and, like that lake, had the wall of ice for its northern shore, the ice acting as a dam preventing the escape of the waters northward. After the ice had melted .back from the position of the First or Al- tamont Moraine, the waters began to fill the basin between the higher land along the eastern front of the great Missouri Plateau, the Coteau du Missouri, and the edge of the ice. The First or Altamont Moraine lies on the top of the eastern portion of the great plateau, extending in a northwest and southeast direction across Ward and McHenry Counties, being crossed by the Great Northern Railway between Tagus (Wallace) and Palermo, the distance between these stations represent- ing the width of the Moraine. West of Balfonr and Anamoo'se the high hills of the Second or Gary Moraine appear, marking the second halting place of the edge of the Dakota Glacier or lobe of the ice-sheet as it slowly melted off from the landscape. It was probably in the inter- val between the times of formation of these two moraines that Lake Souris began, being at first a long, narrow lake fed by the waters flow- ing directly from the melting ice-sheet and the then great glacial river, the Des Lacs. The upper course of the Souris or Mouse River was probably at this time covered by the ice. The earliest outlet of Lake Souris was to the south by the broad valley which extends from near Velva at the southern point of the Ox- Bow or Big Bend of the Mouse River south and west of Balfour and Anamoose, and then conducted the waters across to the Missouri River probably by Pony Gulch, or the channel to the west of Dog Den Butte, in which lie Strawberry, Long and Crooked Lakes, form- ing a channel across the great Altamont Moraine. This valley south from Velva is a broad, level tract of prairie, low in many places and covered with lakes and hay-sloughs. It varies in width from a quarter to a half mile or more. That this was the outlet for a considerable time 112 113 114 THE STORY OF THE PRAIRIES. is shown by the fact that the old channel is well marked, having clearly defined banks and a broad, flat bottom. After the ice had melted back farther, probably to the position of the Fourth or Kiester Moraine, and at the time when the edge of the great ice-sheet rested upon the high hills south of Devils Lake, the outlet was changed so that the waters escaped by the channel of the present Big Coulee and Girard and Buffalo Lakes to the upper course of the James River. This old outlet channel is about 125 feet deep and a third of a mile wide. The Big Coulee, which now occupies this valley, is one of the head streams of the Sheyenne River. A well marked channel a half mile in width leads across from the valley of the Sheyenne to the valley of the James in northern Wells County, by which the waters of Lake Souris were carried to the James Valley from the upper course of the Sheyenne, the lower valley of the latter being at this time still buried beneath the ice. This old channel connecting the Sheyenne and James Valleys is now a "dry" waterway. In time of high water there is a stream on its bottom flowing toward the Sheyenne, the valley of the Sheyenne being now a deeper and larger valley than that of the James. The Big Coulee now extends as a well marked valley to> within about twelve miles of the Mouse River. Here, then, is an old waterway, from the basin of Lake Souris to the Valley of the James, fifty miles in extent, in which now lie the Big Coulee, and Girard and Buffalo Lakes, the upper valley of the Sheyenne and the abandoned channel which connects the valleys of the Sheyenne and James Rivers. Afterward, when the ice-sheet had melted back so that its edge extended from west of the Turtle Mountains to the high hills south of Devils Lake, and south and east through Nelson, Steele and Barnes Counties, the Sheyenne River, now receiving flood waters from the melting ice, was cutting its broad and deep channel and building its delta in Lake Agassiz. With the deepening of its channel the waters from Lake Souris were diverted from their course to the James, and Lake Souris became connected with Lake Agassiz by the Sheyenne River. The old channel which formerly carried the waters of Lake Souris to the James now became a reversed waterway. The Valley of the James is lower than the former mouth of this old channel, so that the headwaters of the James are not cut off and drawn away by the Sheyenne, although in time of high water a sluggish current moves in this channel toward the Sheyenne. OTHER EXTINCT GLACIAL LAKES. 115 When at a later time the ice had melted back so that the Turtle Mountain plateau was uncovered and the lower land north of these mountains was freed from its burden of ice and was covered by the waters of the now larger Lake Souris for the lake grew larger as the ice melted back still another outlet lower than that by the Big Coulee was formed north and east of the Turtle Mountains, about twenty miles north of the International Boundary, and the waters of Lake Souris flowed south by the course of Badger Creek in Manitoba, through Lac des Roches in Towner County, and thence south by the Mauvaise Coulee into Devils Lake. At this time Devils Lake drained into Stump Lake, and Stump Lake drained into the Sheyenne River. So Lake Souris still furnished water to keep the Sheyenne at flood while it was cutting its deep channel and building up the great Shey- enne Delta in Lake Agassiz. On the northwest side of the lake the Assiniboine River was pouring in its waters and building a delta upon the bottom of Lake Souris, and its waters also were added to the volume of the Sheyenne. During the time that Lake Souris was discharging its waters by the Big Coulee outlet into the James River, and. later into the Shey- enne, and probably also at the time of the later outlet north of the Turtle Mountains to Devils Lake, another large glacial lake far north in Canada, Lake Saskatchewan, was sending its waters into Lake Souris also, so that there was a vast waterway from 200 miles north of the International Boundary in Canada by the way of Lake Souris and the Sheyenne to the southern part of Lake Agassiz, and from Lake Agassiz south by the River Warren and the present course of the Minnesota River into the Mississippi and so to the Gulf of Mexico. At a still later stage in the melting of the ice Lake Souris was drained by the Pembina River into Lake Agassiz, and its waters helped to build up the Delta of the Pembina in Lake Agassiz, and deepened the channel of the Pembina River where it crosses the crest of Pem- bina Mountain to a depth of 350 to 450 feet. The Dakota Glacier or ice-lobe still lay upon the northeast corner of North Dakota at the time Lake Souris was drained by Lac des Roches and the Mauvaise Coulee into Devils Lake, the southern end of the ice-lobe forming a point which rested on the highland west of Park River, Conway, and Inkster, and formed the moraine which ex- tends from west of Inkster northwest to the northeast corner of Tow- ner County. But the ice had entirely melted off from North Dakota 116 THE STORY OF THE PRAIRIES. at the time of the Pembina River outlet. The water of Lake Souris had by this time lowered so that the southern end of the lake did not reach south of the International Boundary, and the lake was finally en- tirely drained while the waters outflowed by this outlet. Thus, while Lake Souris began a long time before Lake Agassiz, Lake Agassiz was still at nearly its highest stage when Lake Souris was entirely drained. Lake Souris began after the ice had melted back from the position of the First Moraine so that there began to be a basin between the highland of the Missouri Plateau and the edge of the ice-sheet. The region of the Red River Valley was buried beneath the ice of the Minnesota Lobe till after the time of forming of the Seventh or Dovre Moraine. About the time when the Sheyenne River began to broaden out to form the first long, narrow lake, which was the beginning of Lake Agassiz (page 80), Lake Souris was changing to its third outlet and the waters of the Lake covered the whole region of the Ox-Bow of the Mouse River. During all the time that Lake Souris was being drained into the Missouri River and then by the Big Coulee into the James River, Lake Agassiz had not yet begun to exist. But the long time that Lake Agassiz continued while all the beaches below the Herman and Norcross stages were being formed makes it much older in length of time of existence. By reference to the Map (Figure i), it will be seen that the shore- line of Lake Souris crosses the International Boundary near the north- west corner of Bottineau County, extending south to the city of Minot. thence follows nearly parallel with the Ox-Bow of the Mouse River to Rugby, thence north to the Turtle Mountains, and skirts the base of these mountains around their south, west and north sides, and then extends north for forty miles, making the area in Manitoba about the same as that in North Dakota, and this is just about the same as the area covered by Lake Agassiz in North Dakota, It will thus be seen that all the great expanse of prairie from Rugby, Willow City and Bottineau west and south to Towner, Velva and Minot, and north to the International Boundary, is lake bottom. This vast region embraces a natural basin of nearly 4,000,000 acres within the State of North Dakota and an area of about equal extent in Canada. When the Great Ice-Sheet covered the continent this great basin and the surrounding highlands were filled and covered by the ice. It was the melting of this enormous mass of ice which furnished the water which, hemmed in by the higher lands on three sides and by the ice on OTHER EXTINCT GLACIAL LAKES. 117 the fourth or north side, caused the lake. When, therefore, the ice of the great glacier had melted back farther north than where Velva now stands a lake began to be formed, growing larger as the ice continued to melt away toward the north. There was at this time, of course, no Mouse River, because the land which it now drains was buried under ice probably half a mile deep. Just how long it took for the ice to melt away, and therefore how long the lake lasted we do not know, but it was a long time measured in years, perhaps several centuries, for the outlet channels were cut down a good way into the land surface, and the shore marks made by the washing of the waves indicate that the water stood here for a long- time. It will be noticed upon the Map (Figure i) that the moraines stop at the edge of the lake. The ice left these earth materials upon the lake bottom as well as upon the land outside the lake. But the waves and currents of the lake leveled down the hills to a. large extent and spread the materials upon the bottom. The lands are not perfectly level on the old lake bottom just west of these hills, buf are quite roll- ing, or undulating. They have been leveled a good deal so that the rough and high places have been softened and toned down, giving a gracefully curved contour to the surface. One can trace the line of hills of a moraine across the lake bottom in many places by the rolling and undulating character of the land surface. The lake bottom hills have low-rounded, smooth surfaces quite different from the rugged and irregular heaps and ridges of clay, sand and boulders which make up the moraines east of the lake shore. Since the waters of the lake have gone the old bottom has become a : grassy prairie. Boulders are scattered upon its surface as they were left by the melting ice of the glacier, or were dropped from floating cakes of ice formed on the surface of the lake during winters and so scattered over the prairie. The hills on the lake bottom are often quite sandy, and dune tracts are common. The sand comes from the Turtle Mountains and the region south, for the ice-sheet in crossing the Turtle Mountain plateau combed off great quantities of the soft sandstone rock which makes up a large part of the rock layers of these mountains. The underlying rock south of the mountains is sandstone also, so that as the ice ploughed over the landscape, sandstone rock was broken loose and ground up, and so when the ice melted it left the sand in morainic 118 THE STORY OF THE PRAIRIES. heaps. The waves and currents of the lake washed away much of whatever clay and earth was carried by the ice. The clean, almost white, sancl was therefore left. The "heavier" soils became covered with grass and other vegetation, and the sod so formed prevents the wind from carrying the particles. The sand does not readily become sodded over and so it is taken up by the wind and blown and piled into dunes. Some large dune tracts are crossed by the Great Northern Railway FIG. 60. Sand Dunes, North of Towner, McHenr> County. The Sand is Carried by the Wind over the Crest of the Hills, and is Burying the Forest. Photograph by F. N. Molyneitx. where it passes over the old lake bottom from Rugby to Minot. A scant growth of scrubby timber holds a footing on many of the dunes. Some of these hills are made up of almost perfectly clean whitish sand and they are moved across the country in drifts in the same manner as drifting snow travels with the winds. During "the time that the Antelope Hills were being formed at the edge of the ice Lake Souris still discharged by the Spring Creek outlet to the south, and to the Missouri River. If the lines of the moraines shown on the Map are extended across the lake bottom to show where the edge of the ice was across the lake it will be seen that the lake was OTHER EXTINCT GLACIAL LAKES. 119 as yet only a small sheet, for these lines mark the position of the ice- shore on the north side of the lake. It was, however, supplied with water by the Des Lacs and Mouse River drainage from along the edge of the ice-sheet far to the northwest. The outlet and shores were high at this time, and it seems likely that the well marked ridge which ex- tends about fourteen miles from south of Balfour north and west to Pendroy at the Mouse River is a beach ridge formed on the east side of a bay which formed the southern end of the lake, and which ex- tended south until the waters broke over the summit near Balfour. This ridge, the famous "Balfour Ridge," is as smooth and well-defined as a railroad grading, becoming higher and broader toward the north. It rises six to eight feet above the prairie at the south end about Bal- four and rises gradually and evenly till at the north end, where it is abruptly cut off by the Mouse River, it is thirty feet high. Such a beach would be built higher where the lake was wider and the waves rolled higher, and this accords with the form of this southern bay, which had its narrow point ten to fifteen miles south of Velva, near Balfour. That this shore, and the southern Balfour or Spring Creek outlet, were higher than the Big Coulee outlet, which was opened after the Antelope Hills had been formed, or which likely began to be cut while the last ridges of these hills were being formed, is shown by the fact that Wintering Creek flows from along the east side of the Balfour Ridge toward the Big Coulee outlet, several small coulees which enter it flowing in deep cuts across the ridge. Glacial Lake Dakota. North Dakota has a "majority" in the number of old lake bottoms within the limits of the State. Besides Lake Agas- siz and Lake Souris. a third lake, which lay mostly in South Dakota, extended over a small area in Dickey County in North Dakota. This old lake has been called Lake Dakota. Like the other two large glacial lakes which have been described, it \vas caused by the flood waters from the melting ice-sheet, but not in just the same way as were these. Lake Dakota was formed when the edge of the ice on the western side of the Dakota Glacier stood at the position of the Third or Ante- lope Moraine, when the Valley of the James River had but just been uncovered from the ice. The lake lay in the Valley of the James River. It \vas an enlargement on a very large scale of the James River. The waters could not escape at the south fast enough, being dammed by a 120 THE STORY OF THE PRAIRIES. ridge of hard rock, the Sioux Quartzite, the rock which is spoken of in Chapter Two as being at the surface in central eastern South Dakota. This rock is very hard, and the ice-sheet in passing over it did not plane it off as it did the softer rocks to the north. The result was that a ridge or low hill of this rock lay across the course of the James River and acted as a dam, causing the waters to accumulate and spread out north of it, thus forming the lake. The edge of the ice-sheet lay along the east side of the James Valley and so there was much water flowing down its course from the melting of the ice. The ridge of hard quartzite was at Alexandria, South Dakota, and the lake extended north from here along the present Valley of the James River to Oakes in North Dakota. Its length was about 175 miles, and it varied in width from eight or ten to thirty miles, and its depth in the deepest part was probably 175 feet. Only the northern end of this lake extended into North Dakota. Where the southern boundary line of the State crosses the old lake bottom it is about eight miles in width. It extends a few miles north of Oakes in North Dakota, and covers a territory in this State of a little more than 100 square miles. Glacial "Lake Sargent." In the interval after the draining away of Lake Dakota and before the beginning of Lake Agassiz, a glacial lake covered the greater part of Sargent County, a small part of Ransom County, and extended about ten miles into Marshall County, South Dakota. No name having been given to this extinct lake, it is here called "Lake Sargent." The broad morainic belt on the western line of Sargent and Ran- som Counties served as the western shore of this lake, this moraine and the Coteau des Prairies the southern " shore, and the wall of the melting ice-sheet the northern and eastern shore. As the ice melted on the eastern side of this moraine adding its waters to those of the lake the area of the lake extended eastward following the melting ice, till the Dovre Moraine was formed. This moraine became the eastern shore and was washed on its western side by the waves of the lake. The shore-line thus extended from Nicholson and Straubville south across the State Line to Burch, South Dakota, then north and east around the head of the Coteau des Prairies to Lake Tewaukon or Skunk Lake, and north by Cayuga and Ransom, covering the Stormy Lakes, and extending north into Ransom County, its area covering probably between 600 and 700 square miles. OTHER EXTINCT GLACIAL LAKES. 121 Lake Dakota had been drained away before the beginning of Lake Sargent, and the James River was flowing across its old bed. Lake Sargent at first discharged to the southwest across the now dry bottom of Lake Dakota into the James River. Later when Lake Agassiz had begun to be formed a lower channel of discharge probably was found to the east from Lake Tewaukon close north of the Coteau des Prairies highland and south of the high range of hills (Dovre Moraine) which extends south of Lidgerwood, passing through a low place in the mo- raine, and entering Lake Agassiz about four miles south of Hankin- son, and twenty miles east of Lake Tewaukon. The depth of the lake at the time of its highest stage was probably about 50 feet at Forman, 100 feet at Perry and 150 feet along the northeast side in the vicinity of the Stormy Lakes, though the eastern outlet may have lowered the water before it became as deep as these figures indicate. The eastern two-thirds of Sargent County is now drained into the Red River of the North by the Wild Rice, which enters the area of ^Lake Agassiz near Wyndmere. A cut of twenty-five feet in the mo- raine east of the James River twelve miles south of the State Line at Amherst, South Dakota, would permit the waters of the James River to be carried by the course of the Wild Rice to the Red River of the North. The elevation at Amherst is 1,312 feet above sea-level. Wild Rice station, near the mouth of the Wild Rice River, where it enters the Red is 911 feet above sea-level, so that there would be a fall of about 400 feet from the James River to the Red River of the North in a distance of about 100 miles, a fall about four and one-half times as great as that of the Red River from Lake Traverse to Lake Winnipeg. Had it not been that the James River cut a channel deep enough to prevent it breaking over to the east while the ice-sheet still covered the land to the east and was forming the large moraine which lies east of the Valley of the James, that river might have taken an easterly course to the Red River of the North instead of its present southerly course. This is an interesting example of the way the ice-sheet changed and directed the course of rivers. At Nicholson a broad channel widens out onto this old lake bottom, a channel by which a large glacial river entered this old lake. This old channel was occupied by the Sheyenne River before the ice-sheet had melted back far enough to allow this river to cut its present chan- nel south of Valley City to Lisbon. It is about two and a half miles 122 THE STORY OF THE PRAIRIES. wide where it is crossed by the Fargo Southwestern Branch of the Northern Pacific Railway at Englevale. It extends north into the Fort Ransom Military Reservation, and south at Nicholson broadens out onto the bottom of Lake Sargent. Ridges of drift formed islands in the broad river, and deep channels cut in its flat bottom perhaps by cur- rents of the old river during winters when the melting of the ice-sheet was less rapid, are now filled with water and give to the old valley the name of Big Slough. The Sheyenne River received water from all along the edge of the ice-sheet north to Devils Lake, and probably during this time received the waters from Lake Souris by the way of the Big Coulee outlet. It did not, however, cut a channel so deep but that, when the ice had melted back farther than to the position of the Dovre Moraine and the course of its present valley was uncovered and Lake Agassiz began to be formed, it cut its deep channel south and east to Lisbon and began to build up its delta at Milnor. When, therefore, Lake Sargent had been lowered by the opening of its eastern outlet from Lake Tewaukon to the east into Lake Agassiz and the Sheyenne River had ceased to pour its waters into Lake Sar- gent this lake rapidly ceased to be, and later still the drainage from the Lake Sargent area was established by the course of the present Wild Rice River. So Lake Sargent came into existence after the formation of the large moraine which consists of the combined Fourth, Fifth and Sixth Moraines along the western boundaries of Ransom and Sargent Counties, and continued to grow larger during the time that the ice was melting back to the position of the Seventh or Dovre Mo- raine, after which it quickly disappeared by the drawing off of its waters to the east and by the changing of the course of the Sheyenne River so that its waters did not enter this lake. Lake Dakota began at or before the time of depositing of the Fourth Moraine and had disap- peared before the beginning of Lake Sargent at the time of the Sixth Moraine. Lake Agassiz began with the same events which caused the closing of the existence of Lake Sargent, that is, the withdrawing by melting of the edge of the ice-sheet farther east than the Dovre Moraine so that the basin of the Red River of the North began to be uncovered, and the withdrawing of the waters of the Sheyenne River from Lake Sargent to that basin and at the same time causing the drawing aw 7 ay of the waters of Lake Sargent. CHAPTER THE ELEVENTH. THE HISTORY OF DEVILS LAKE. The history of Devils Lake is interesting not only because it is "The Great Salt Lake" of North Dakota and the largest lake in the State, but its history forms an interesting chapter in the geology of the State. The Cause of the Lake. A line connecting Stump Lake and Devils Lake and extending northwest through Ibsen, Hurricane, Grass, Island and Long Lakes probably marks the place of an old river valley which once extended from near the Turtle Mountains to the Red River. This old valley was filled or nearly filled with drift. The Blue Hills south- west of Stump and southeast of Devils Lake, the high and massive hills south of Devils Lake, Mauvais Butte, or Big Butte, south of Lake Ibsen, the eastern end of which is about eight miles west of the western end of Devils Lake, a hill about ten miles long, and the high land northwest from Mauvais Butte which forms the watershed or divide between the Mouse Valley and Mauvaise Coulee, which is the highest point crossed by the Great Northern Railway between Grand Forks and the Missouri Plateau west of Minot, form a series of highlands which were probably the southern and western side of this valley. This it will be understood was a valley upon the landscape of "Old North Dakota," or the pre-glacial landscape. When the ice-sheet melted off from the land the valley was nearly filled with drift. It was not entirely filled for its course is still able to be traced for 100 miles from the east end of Stump Lake to* Long Lake south of the Turtle Mountains. All these lakes lie lengthwise of this valley as we should expect them to do if this were the partially filled valley of an old river. The Blue Hills are veneered hills, that is, hills which were there before the ice-sheet came, and which have been covered with a mantle of drift. So also the high hills south of Devils Lake were hills before the ice melted in trying to cross over them and dumped the drift hills on top of them. Mauvais or Big Butte is also covered with a mantle or coating of drift, but is not itself made up of drift. 123 124 THE STORY OF THE PRAIRIES. The Blue Hills, which rise from 100 to 200 feet above Stump Lake, the massive hills south of Devils Lake, the highest of which, Devils Heart and Sully's Hill, rise 275 to 290 feet above the water of Devils Lake, and Mauvais Butte, which rises at its higher western end nearly 300 feet above the prairie at its base, are all elevated masses of Cretace- ous (Fort Pierre) shale, their rough surfaces having been combed off and smoothed by the ice-sheet passing over them, and leaving a cover- ing of drift as it melted. Devils Lake and Stump Lake occupy deep hollows in this old val- ley where it was less filled with drift. Stump Lake is said to be nearly 100 feet deep in its deepest place, and Devils Lake in the centre of the widest portion of the eastern end is 75 to 80 feet deep. The drift on the surrounding prairies is from 10 to 50 feet deep on the general land- scape and as much as 100 feet deep in the morainic hills. These lakes thus lie in a trough in the rocks which underlie the drift materials. They are, in fact, lakes formed by the damming of a river valley. At least it seems probable that they lie in such a valley. Some of the arms or bays of Devils Lake are very likely the partially filled valleys which were 'tributary to the main valley. Some of the bays are caused by moraines which were dumped into the valley and now form the bluffs on the north side of the lake, but it does not seem that these gave the. general form to the outline of the lake. The sides or shores on the south side seem to be the underlying rock only thinly covered with drift. Devils and Stump Lakes were much larger bodies of water during the time when the ice-sheet was melting north of these lakes and a flood of ice-waters was being poured into them. There are marks made by the waves twenty-one to twenty-five feet higher than the present sur- face of low-water in these lakes. If the water in Devils Lake should rise sixteen feet above low-water a connection would be made across from its eastern end near Jerusalem to Stump Lake, and if it should rise five. to eight feet higher still, Stump Lake would also drain into the Sheyenne River. An old channel connects Stump Lake with the valley of the Sheyenne, as also a lower channel which connects Devils and Stump Lakes, showing that there has in time past been an outflow to the Sheyenne as well as connection between the two lakes. If we follow the Chain-of-Lakes and the Mauvaise Coulee to Lac des Roches near the international boundary, and thence by Badger Creek to Pelican Lake in Manitoba, and to the Souris River, a natural THE HISTORY OF DEVILS LAKE. 125 waterway is seen to almost connect the Souris with Devils Lake. This was the course of the outlet of Lake Souris to Devils Lake and the Sheyenne River before the ice-sheet had melted off from the northeast corner of North Dakota. At the time the waters of Lake Souris flowed by this course, Sweet- water, Dry, and De Groat Lakes, which are fenced in by the terminal moraine which lies south of them, were higher than now because of the flood of ice-water they received from the melting ice-sheet, the edge of which was but a little north of them at this time, and flowed across to> the south into Devils Lake. Fluctuations of Level. The waters of all lakes vary in the height of their water-level during periods of years. Devils Lake has been much higher than it is now, as is shown by beach-lines marked by the waves at levels considerably higher than the present high-water level. It has also been much lower than the present low-water, as is shown by forests which are now submerged along the shores below low-water. Such trees now stand with their roots imbedded in the mud in Stump Lake, and also in the Washington Lakes a few miles south of Devils Lake in Eddy county. It is said that the name Stump Lake came from this fact. The year 1889 marks a low stage in the waters of Devils and Stump Lakes, while about sixty years before, in 1830, the waters of these lakes were sixteen feet higher than in 1889. This was about the time of the highest known flood of the Red River of the North, when its wa- ters rose so high that they covered the land on which the City of Win- nipeg stands to a depth of- five feet. The waters of Devils Lake rise and fall through a height of four feet in a dozen years. Since Fort Totten was built, about thirty-five years ago, the lake has fallen ten feet. At the time of the high water in 1830 the height of the water in Devils Lake was limited by an overflow into Stump Lake, a channel about sixteen feet above low-water, as has been stated, connecting the two lakes. It is likely it has risen high enough to discharge into Stump Lake many times in the period since the Ice Age. At the time the melting ice-sheet was pouring its waters into these lakes their level was twenty-one to twenty-five feef higher than the low stage in 1889, and Stump Lake then discharged into the Sheyenne River. The channel from Stump Lake to the Sheyenne has a nearly flat bottom 150 feet wide, and hills rise on either side fifty to seventy- five feet high. The bottom of this old channel is higher than the ;L ?tfM IM fl i| $ s 1 Wil'!' 1 "!,,! fe^lS i i ! S I'lilllr ^! PT ,.,.,. v.^ .^i^i^ ;'i-'r.iii;ii ,ii'i|V.^ & fiSs PI. I'M 126 128 THE STORY OF THE PRAIRIES. beaches which were formed by the waves during the probably quite long time when the waters were at the high stage of 1830, which are sixteen feet higher than the low stage of 1889, but these beaches are higher than the bottom of the channel connecting the east end of Devils Lake at Jerusalem with the west end of Stump Lake, and these beaches are marked on the sides of this channel showing that the two lakes were then united or joined by a strait. The heavy and older forests which border these lakes extend across the highest shore-line which marks the height of the waters at the time of the melting away of the ice-sheet, and down to the beach which marks the high stage of 1830. Below this shore-line are only smaller and scattering trees, one of the largest of which is reported to have been cut by Captain Heerman and to .-have had fifty-seven annual rings of growth. During the igth century, therefore, these lakes probably have not stood above the high shore-line of 1830. The old submerged forests may date back 200 years earlier to the time of the great period of drought when Pyramid and Winnemucca Lakes in the Great Basin of Nevada were dried up. NORTH DAKOTA Tower Sheet T.140V. T. 139N. T. 138H. T.137N. T. 138N. 51 = T 136N, FIG. 63. The Tower Quadrangle. 129 CHAPTER THE TWELFTH. THE SHEYENNE RIVER. The Sheyenne River is one of the most interesting of rivers. Some- thing of its history has been given in the earlier pages of this book. A brief statement of a few important facts, as illustrating the effect of the great ice sheet in determining the courses of modern streams, is given here, though at the risk of repetition. Only the portion of the River between Valley City and Lisbon is considered. An Example of a Glacial Valley. The casual observer who has but little knowledge of geology cannot but be impressed with the very great size of the Sheyenne Valley when viewed in comparison with the very small modern stream that occupies it. The valley that has been excavated by the waters that have passed down this course is as much as 5 miles in width at Valley City, and the depth of the valley below its highest floodplain (now known as a terrace) is as much as 200 feet. The great valley is marked by terraces, often broad, which are remnants of the floodplains of the river when it flowed at much higher levels than any reached by the modern stream. These floodplains were left high above the river as it eroded its channel deeper. The Lanona Plain. The highest floodplain of the river is known as the Lanona plain. This plain is nearly 200 feet higher than the bot- tom lands of the present site of Valley City. When the river flowed at this stage, and before the present deep valley had been eroded, it was a stream 5 miles in width in the vicinity of Valley City. The waters that kept this tremendous stream at flood tide came from the melting of the great ice sheet. At this time Devils Lake and Stump Lake were much larger bodies of water than now, and these were joined by a con- necting channel so that the waters of Devils Lake passed to Stump Lake, and the waters in turn passed from Stump Lake to the Sheyenne. Devils Lake was at this time a body of fresh water and not a salt lake as now. The waters of this great stream were not at this time discharged in- to the Red River .of the North, for there was as yet no Red River of the 131 II THE SHEYENNE RIVER. 133 North, neither was there yet any Glacial Lake Agassiz occupying the Red River Valley. On the other hand the Red River Valley was filled and covered with the ice of the great ice sheet, and the Sheyenne at this time discharged to the south instead of turning east, as at present, at Fort Ransom, its waters being mingled with those of the James and thence passing to the Missouri. At this time. the edge of the great ice sheet probably lay upon the Alta Ridge, and southward to Standing Rock in Preston township, the ice covering the whole course of the present river below Fort Ransom. Bears Den Hillock, immediately west of Fort Ransom, and the range of morainic hills that runs southward from this large morainic hill FIG. 67. Section of Sheyenne Valley at Valley City. Generalized section from the butte east of city northwest about z l / z miles, through sections 22, 21, 16, 17, Valley township. through the townships of range 58 to and beyond the southern boundary of the state of North Dakota, represent the deposits made at the edge of the ice at this time. It will be easily seen therefore that the Sheyenne River could not have followed its present course east of Fort Ransom, as this whole region was buried underneath the ice. But it has been stated that the river at this time was probably 5 miles in width in the neighborhood of Valley City. Where, then, was the river 20 miles south? The Sand Prairie Spillway. The nearly level Lanona plain ex- tends south of Valley City about 8 miles, bordering the moraine on Al- ta Ridge, the Fergus Falls moraine. Southward from here the moraine lies close upon the east bank of Sheyenne Valley. Standing Rock and Bears Den Hillock are regarded as belonging to this moraine, and it has been stated that the moraine extends many miles to the south. Evidently then the earliest Sheyenne River was compelled to discharge 134 THE STORY OF THE PRAIRIES. to the south instead of turning to the east at Fort Ransom, as now. At this time the moraine extended across where is now the deep valley between Standing Rock and Bears Den Hillock. The edge of the ~i t e _ -. - ' _zi FIG. 68. Section of Sheyenne Valley, 3^4 Miles South of Valley City. great ice sheet and the moraine which had been deposited together formed a great dam against which the waters could not prevail. The waters therefore gathered along the edge of the ice till they covered a broad tract of land, and finally when compelled to go somewhere they flowed south on the west side of*this moraine, and entered Lake Dako- ta and the James River. This broad tract where the waters ponded is now known as Sand Prairie. It was a sort of spillway or lake caused by this piling in of the waters of the great glacial river. FlG. 69. Section of Sheyenne Valley, j Miles Below Valley City. On the map. Figure 64, a broad expanse of water is represented west of the present site of Fort Ransom. The land here is now an almost level sandy plain, locally known as Sand Prairie. For several miles along the west bank of the Sheyenne Valley, in Bear Creek and Oak- ville townships, this level plain comes up to the edge of the valley, stop- ping abruptly as though it had been cut off. And this is really what has happened, for this level plain was once the bottom of the Sheyenne THE SHEYENNE RIVER. 135 River, and the north end of this flat plain was eroded away as the river cut its valley deeper. The River Ransom The river thus far described represents the earliest stage of the Sheyenne. After a time, we do not know how long, the ice melted so that the water did not all escape by the Sand Prairie spillway and southward by the course west of this moraine, but formed a channel on the east side of the moraine. This means that a passage had to be cut through the moraine that had been formed during the time the river was discharging south by the Sand Prairie spillway and Lake Dakota. This gap or passage through the moraine is now shown by the steep high morainic hills such as Standing Rock and Bears Den Hillock, which stand close upon the banks of the valley, and by the rolling hills in southern Bear Creek township which were partially worn away and leveled by the action of the river. The narrow canyon-like char- acter of the valley from Standing Rock southward to Fort Ransom also shows that the river had a hard time cutting a channel through this re- gion. The waters could not escape to the eastward, however, even after they had found a way across the moraine, because the ice had not melted to the eastward of Fort Ransom. There was formed at this time a new channel to the south from Fort Ransom, the bottom of this channel being now about 80 feet lower than the earlier channel rep-resented by Sand Prairie. The stream that eroded this ancient channel has been called the River Ran- som. It extended south from Fort Ransom on the east side o-f the mo- raine that has been referred to as extending to the state line in range 58, and between this moraine and the ice front. This channel is a large broad ancient water course that can be easily traced southward to Englevale and beyond, till it finally broadens out into the plain of the ancient bottom o>f Glacial Lake Sargent a little south of the northern EflST WS/E3T FIG. 70. Section of Sheyenne Valley, 10 Miles Below Valley City, 136 THE STORY OF THE PRAIRIES li THE SHEYENNE RIVER. 137 boundary line of Sargent County. The River Ransom was half a mile to 2 or 3 miles in width. The town of Englevale stands in the midst of this old river, or in other words, the gravelly and sandy plain upon which Englevale is built is the bottom O'f this old river. This ancient bottom is called the Big Slough by the people who live in this vicinity. In sections 12, 13, and 24, Bear Creek township, an old channel lies about thirty feet below the level of Sand Prairie, showing how much the Sheyenne River had cut down its channel at this time. This old valley is as perfect a ditch as could be made with shovels and grading crew. It is now a dry channel, in part a hay meadow and in part ploughed fields. Another channel crosses this same terrace plain about a mile to the east, and is about 20 feet lower than the channel last described. EflST WEST FIG. 73. Section of Sheyenne Valley, 13 Miles Below Valley City. FIG. 74. Section of Sheyenne Valley at "The Jaws." Section 36, Bear Creek Township, 25 Miles Below Valley" City. 138 THE STORY OF THE PRAIRIES. The broad terrace into which these channels were eroded lies in a turn or bend of the valley, this terrace representing" the floodplain of the river before the deeper valley had been excavated around the bend. In sections 35, Bear Creek, and 2 and u, Fort Ransom townships, terraces of an old river bed can be seen high up on the steep side of the 1200 FlG 75. Section of Sheyenne Valley Above Lisbon. The cut-off Passes West of Lisbon about z miles. present valley 70 to 80 feet below the level of Sand Prairie, and nearly 200' feet above the present bottom of the Sheyenne Valley. This old channel represents a later stage in the development of the valley than the channels just referred to, being considerably lower, but the waters were discharged by the channel of the River Ransom during all this time. \t Fort Ransom this ancient channel hangs on the steep side of FIG. 76. Railroad Cut East of Kathryn. Shale shown at right, and the Terrace Floodplain continuous with this at left. THE SHEYENNE RIVER. 139 the valley 186 feet above the water of the river below. This seems such a big story that it was determined by leveling, so that no one could say it was merely guess-work! The bottom of this channel is now a flat shelf or terrace, its east side having been eroded away, its west side being the great wall o>f the valley side. From this shelf the leveling instrument was pointed southward dow 7 n the big channel of the River Ransom, and this broad flat bottom was found to be at the same level as this shelf. FIG. 77 . An Outlier of Shale, West Bank of Sheyenne Valley. The flat top represents a portion of an Old Floodplain (terrace) ; the channel of the river has been eroded into the Lower Floodplain (the lowest terrace.) The Sheyenne Valley East of Fort Ransom In course of time the great ice sheet grew less, and its margin retreated by melting farther toward the east and north. Finally the waters could escape eastward near the present site of the city of Lisbon, and the old channel of the River Ransom became an abandoned channel. It was about this time that the Maple system of glacial channels, de- scribed in the next chapter, began to be formed. When however the first Maple River discharged its waters southward by the channel of the South Branch the Sheyenne did not follow its present course by the city of Lisbon. A large "cut-off" channel crosses the bend about 2 miles west of Lisbon, which is 60 to 80 feet higher than the bottom of the present valley at Lisbon, the valley not having been yet eroded to its present depth. The Hanging Valley of the Maple. The channel of the South Branch, which joins the Sheyenne Valley about 4 miles above Lisbon, conveyed the waters of the Maple to the Sheyenne while the Sheyenne 140 THE STORY OF THE PRAIRIES. was yet flowing about 75 feet above the present valley bottom. This is the reason that the channel of the South Branch is said to enter the valley of the Sheyenne by a "hanging" valley." The Sheyenne Valley has been eroded 75 feet deeper since the waters of the Maple ceased to be discharged into this valley, and so the ancient mouth of of the South Branch Channel is left hung up 75 feet above the present Sheyenne Valley bottom. This old mouth has since been dissected by a recent coulee, but still the bottom of the ancient channel can be easily seen high above the Sheyenne bottom. Terraces of the Sheyenne Valley The history of the Sheyenne River is revealed by the form of its present valley. The high terraces that mark its sides indicate the former levels at which the stream flowed. A terrace shelf is a part of what at one time was the floodplain of the river. True it is that no man was there to write a record of the size of the river. Nevertheless the width of the river is stated quite con- fidently to have been as much as 5 miles at Valley City, because the floodplain of the ancient stream is seen to extend to this distance. The terrace called the Lanona plain is clearly the one-time bottom of the ancient river. (Figures 64, 68, 69.) Lower terraces show where the bottom of the valley was at later stages. The higher the terrace above the bottom of the present valley the older it is, that is, the farther back it dates in the history 'of the river. Old channels that are now far above the bottom of the valley were at one time the lowest parts of the valley. A valley has been defined as the excavation made by a stream and its antecedents. The present Sheyenne River would never erode such a valley as that in which the river now flows. The river is now a small and sluggish stream. But its antecedents, the ancient channels which were kept at flood by the glacial w r aters, did a tremendous work of erosion. When the river was at the Lanona stage the great valley had not yet been cut. Where are now the towns of Valley City, Fort Ransom, and Lisbon were then the firm solid earth. (Figures 66, 71, 78.) This remained to be excavated during the long time that the river has been working at the task of eroding* its valley. The terraces at Valley City, shown in Figures 65, 66, 67, represent stages in the down-cutting of the stream at this point. The top of the butte in Figure 66 was a part of the river's bed at one time. The erosion by which it has been left as an isolated hill with a fragment of terrace THE SHEYENNE RIVER. 141 FIG. 78. The Sheyenne Valley at Fort Ransom. Pliotograph by Rex Willard. FIG. 79. Banks of Sheyenne River West of Fargo. 142 THE STORY OF THE PRAIRIES. for its top has been accomplished during the time since the river had excavated its channel to lower levels. The terraces on the opposite side of the valley represent floodplain levels at later stages of the river. The hill represented in Figure 77 shows in its flat top the oldest floodplain and the earliest stage of the river. 'This fragment of the old floodplain was cut around so as to form an outlying hill by erosion of main stream and tributaries during later stages. The small hill that stands above the terrace and which the terrace surrounds, is a small fragment that was not carried away by the earliest stream. At one time the Sheyenne River flowed east of the water tank at Valley City (Figure 65). This of course was long before the great valley forming the bend around to the westward had been excavated. Similarly, a cut-off channel shows where the river once flowed many feet above the town and to the west of the city of Lisbon. Many such abandoned channels representing the ancient water courses occur along the Sheyenne Valley. In a general way the higher these old channels above the bottom of the present valley the more ancient they are. Many ox-bow cut-offs occur on the alluvial flood- plain of the present valley. If the stream should erode its valley deeper and some of these abandoned ox-bows should be left higher up on the sides of the valley they would be the same kind of relics showing the former course of the river as those referred to and shown in the cuts. That the Sheyenne Valley has been excavated by a larger stream than the modern Sheyenne River is shown by the broad terraces, and that it has been eroded deeply into the shale underlying the drift is shown by the shale exposed in the sides of the valley at many points. (Figures 65, 76, 77.) The sides of the valley are steep and the shale rock has not been greatly changed by the action of the weathering agencies. These show that the valley has been excavated in recent geologic time. That it was eroded by glacial waters forming a large stream is shown by the long course of the stream, the few branches, and the deep and large character of the valley. R 5 (, W R 55 W 10 FIG 80. Map Showing the Glacial Channels of the Maple River. CHAPTER THE THIRTEENTH. THE HISTORY OF MAPLE RIVER. The history of the Maple River is closely associated with the progress of recession of the great ice sheet. The region traversed by this stream and its branches so well illustrates the close relationship between the present drainage and that established by the streams of ice water from the great ice sheet that a study may well be. made of this as a typical region. The modern Maple is a small stream, but like the Sheyenne, it occupies a very large valley in comparison. The main stream and its branches comprise a small system, but for this very reason it can be more readily studied. The same principles of ice movement and melting that determined the positions of these small streams have also deter- mined many large streams. The drainage of the greater part of North Dakota east of the Mis- souri River has been determined in much the same manner as has that of the Maple basin. This means that the streams have* been located upon the land in a very different manner from that of drainage streams gen- erally. In regions untouched by the ice sheet the river valleys have been formed by the run-off from rainfall and melting snows. In the case of this region the channels were formed by water delivered in relatively large quantities at the edge of the melting glacier. As a result the rivers are long, their channels are large, and there are very few tributaries. This is another way of saying that the land at the present time is not drained at all. The land is crossed by stream chan- nels that have no relation to modern drainage. This is why it is possi- ble to have large river valleys and yet have almost no drainage. It is as though systems of valleys had been superimposed or let down upon the landscape fully developed. And this is really about the fact. The vast floods of water that came from the melting of the great continental ice sheet had to get away somewhere, and they stood not on the order of their going. They just went wherever they could go most easily. On the comparatively level plain in front of the glacier the accumulating waters found no ready means of escape. There were no channels 143 144 THE STORY OF THE PRAIRIES. already established, for such as had been excavated had all been filled up during the passage of the great ice sheet over the land. Now as this tremendous mass of ice melted the waters had to escape as best they could, and therefore many large channels were eroded by the ice waters as they passed. Once these channels were established they had to stay. Now when the rains fall upon the land behold here are these old channels ready made. There is not half enough rain to make streams for all of them. The result is that here are the large channels or valleys with little water, and during much of the year often with none at all. The Beginnings of the Maple River. At the time that the edge of the great continental glacier lay upon the Alta Ridge and the Fergus Falls moraine was being formed, there was as yet no Maple River. The region now occupied by this system of channels was buried underneath the vast mass of the great ice sheet. The Sheyenne was then being" flooded by the waters that came from the melting of the ice along its edge and upon its top. The channels of the Maple system began to be eroded after the ice had melted back toward the east so that its edge was somewhere along an irregular line extending from the vicinity of Buttzville north and west to Fingal and Oriska. It is not thought, how- ever, that the edge of the ice was anything like as nearly uniform as would be suggested by even a crooked line from Buttzville to Fingal and northward. It seems more likely ^that the ice border was very irregular, and not only this but it probably melted back and then ad- vanced again locally. When the edge of the ice was at Buttzville it was probably as far west as Fingal, the morainic hills in Raritan Township, and those north of Fingal in Binghampton and Springvale Townships, and also those north and west of Oriska, being thought to have been formed at this time. The accompanying map, Figure 80, shows some- thing of the supposed relation of the development of the channels to the retreating ice front. This sketch map is not claimed to be strictly correct, but a careful study of the relations of the channels to the morainic hills and ridges in the field lends assurance that the map sug- gests an approximation at least to the true history of the events. A broad channel having a gravelly bottom, with gravel exposed in its banks, takes its origin north and west of Fingal and extends to the south and east along what was probably the edge of the ice to the south- west corner of Pontiac Township. Here it entered the channel of what THE HISTORY OF MAPLE RIVER. 145 is now the modern South Branch of the Maple. In Section 26, Raritan, a channel from the west joins the Fingal Branch. This channel takes its origin five or six miles west in Thordenskjold Township. Before the ice had melted so as to uncover the region later occupied by the Fingal channel this stream flowed in a southeasterly direction from Section 27, Raritan, entering the channel of the South Branch in Sec- tion 14, Moore Township. This channel now contains no water except in standing pools during time of heavy rains and melting snows. It FIG. 81. Section of Glacial Channel (Branch of Maple). Section 32, Binghamton Township. Bottom of Valley is a hay meadow; no modern stream channel; sand and gravel exposed in sides; gravel probably under the soil of the bottom. was the earliest channel of the Maple system to be formed. It was the water escaping by this course that first begun the task of eroding the large channel of the South Branch. The earliest Maple River therefore was a glacial stream emerging from the ice border in eastern Thordenskjold Township, and having a southeasterly course to Section 14, Moore Township, and thence south southwest to the Sheyenne River. (See i-i, Figure 80.) This was soon reinforced, however, by the Fingal branch, after another recession of the ice front, the two uniting in Section 26, Raritan, and flowing near to the ice border to Section 31, Pontiac Township, about four miles *- Drijf FIG. 82. Section of Glacial Channel (Branch of Maple). Section 34, Weimer Township. Gravelly banks; no modern stream channel; a little water in pools. (See pp. 37, 51-54). north and west of Enderlin, where it was forced by the wall of ice to turn southward. It pushed its way to the eastward by an irregular channel for a distance of about two miles, thence turning west and 146 THE STORY OF THE PRAIRIES. south till this stream entered the channel already eroded south and west to the Sheyenne from Section 14, Moore. An Embayment in the Ice Margin When the ice had melted back still farther a channel was formed extending a mile east, but again turned south into the South Branch. The channel from Springvale and eastern Binghampton Townships, which enters the main valley in Sec- tion 1 6, Pontiac, was probably opened while that part of the main val- PIG. 83. Section of Glacial Channel (Branch of Maple). Section 36, Oriska Township. Channel eroded in drift; no modern stream; a little water in pools. ley where Enderlin now stands was yet buried beneath the great mass of ice. When the ice had melted back sufficiently so that the Thor- denskjold and Fingal branches occupied the more eastern of the two channels in southwestern Pontiac, the ice margin is thought to have been far enough east so that the branch from Springvale and Binghamp- ton was developed, the present main channel being thus opened for a distance of about two miles in Pbntiac Township, but the waters of this stream were deflected to the west sharply by the barrier of ice in Sec- tion 28, Pontiac. The waters of this branch probably joined with those SOrcls,. FIG. 84. Section of Glacial Channel (Maple River Valley). Section n, Tower Township. Very small modern channel with a little water standing in pools; bottom of valley a hay meadow. of Thordenskjold-Fingal branch and passed to the South Branch through Sections 32, Pontiac, and 6, Liberty. At a later time the waters of these combined streams seem to have passed still farther to the east by the large channel in Sections 32, 33 and 34, Pontiac, and thence passed south by the present site of the city of Enderlin, only to again turn sharply to the westward to join the channel already developed to the Sheyenne. THE HISTORY OF MAPLE RIVER. 147 That the ice still lay to the south of this sharp bend in the ancient stream is suggested not only by the character of the channel, but also by the broad belt of morainic hills in western Liberty and Casey and eastern Moore and Fuller townships, and it was clearly impossible for the waters to escape eastward by the present Maple Valley, or by any eastward course, since the great ice sheet lay upon all the land to the east. That the main channel of the Maple was not yet open north of Enderlin, and that the waters made the longer journey around by the course indicated seems to be shown by the deep 'and broad character of these channels and the smaller and narrower channel of the main valley north of Enderlin. Before the deep embayment into the ice front at Enderlin was formed the edge of the ice probably lay over and west of the city of Enderlin, FlG. 85. Section of Glacial Channel (Maple River Valley). Section 10, Clinton Township. Modern stream channel 10 feet deep, dividing the old floodplain into terraces; old floodplain 20 feet below general prairie level. a lobe or tongue of ice lingering north of the city where the main chan- nel was later opened, and the morainic knobs and rounded hills in western Clinton were probably being formed. It seems likely also that the knobs west of Tower City were also formed in the marginal portion of the ice sheet at this time. The Main Channel Opened The next important recession of the ice sheet uncovered the region of the upper course of the main channel FIG. 86. Section of Glacial Channel (Maple River Valley.) Section 33, Clinton Township. Hay meadow on bottom underlain by 18 feet of gravel and sand; several old channels on the flood plain; rolling topography of prairie along the sides of the valley shows that there is no relation between the prairie and the valley. 148 THE STORY OF THE PRAIRIES. of the Maple as far south as the present city of Enderlin. The halting place of the ice edge is probably represented by the hills in eastern Casey and Liberty Townships, the rolling morainic hills in the region of the Alice Chain-of-Lakes in eastern Pontiac and Clinton Townships, and the round symmetrical knobs in western Tower and Cornell Townships. The main channel of the Maple from the vicinity of Tower City southward to Enderlin, together with the western branches (it will be observed that all the branches of the Maple join the main stream from ----_=_- -^T--^f ^'^^^.4 : - t 'r.* V;;.y ( . ; ;'.V> :' v. ^'-^/^C-^Vx^v^V, <>.'' -" FIG. 87. Section of Glacial Channel (Maple River Valley). City of Enderlin, section 4, Liberty Township. Showing terraces, and gravel underlying valley. the west) was now open. The ice sheet still formed a dam just east of the present site of the city of Enderlin, and the waters were compelled to turn westward away from the ice wall, and were conveyed by the present course of the South Branch to the Sheyenne. Lake Agassiz Opened It was not until another recession of the ice border had occurred and the valley which subsequently came to be occupied by the waters of Glacial Lake Agassiz had begun to be relieved of its burden of ice that the Maple finally became free to discharge its waters eastward by the present course into Lake Agassiz in Section 32, Highland Township. At the time of the opening of the main channel east of Enderlin the Sords. *$$$ ;0.f-<\"i-.->B- FIG. 88. Section of Glacial Channel (South Branch of Maple). Section 27, Moore Township. Valley bottom 90 rods in width; hay meadow on bottom; sides 30 to 40 feet high, with very small modern stream channel; no water except in pools. THE HISTORY OF MAPLE RIVER 149 great channel now occupied by the South Branch, which had been car- rying the waters brought away from the ice border during a long period of recession of the ice, and had been a part of the trunk or main line of the Maple during this time, now ceased to be a part of the course of the main stream and became what it now is, a reversed water-way occu- pied by a small intermittent modern stream. The South Branch now enters the main valley at Enderlin. It is an interesting example of a great river from which its glory has departed. It used to carry vast floods of water from Enderlin to the Sheyenne. Now it brings a little water (very little during most of the year) from within a mile or two of the valley of the Sheyenne in the opposite direction to the Maple. The channel is a large, well defined valley, having steep sides thirty to forty feet high in northeastern Moore township, and from fifteen to thirty feet high in southern Moore and Fuller Townships. (See Figures 88 and 89.) Very little water passes through the chan- FIG. 89. Valley of South Branch of Maple River. A Glacial Channel. Photograph by Rex IVillard. nel now except during the spring season. In fact, throughout much of its course the channel bottom is a fine level hay meadow. Gravel shoul- ders occur along its course, showing the action of a large stream at flood tide. This old channel is a marked landscape feature, and can be easily followed from Enderlin to Section 20, Fuller Township, where it joins the Sheyenne Valley by a hanging valley about seventy-five feet above the present Sheyenne River. It is called a hanging valley because it is literally "hung up" above the present river, the present valley having been eroded since this channel was formed. 150 THE STORY OF THE PRAIRIES. It was not, therefore, till the ice had melted off from the entire region occupied by the Maple system, with the possible exception of the extreme northern headwaters, that the full Maple system became estab- lished in its present course. By this time Glacial Lake Agassiz had begun to be formed by the enlargement of the Sheyenne River in the vicinity of the present town of Milnor, the waters of the Sheyenne being discharged between the wall of the retreating ice sheet on the east and the higher land on the west, ten miles south of the southeast corner of the area shown on the map (Figure So). FIG. 90. Section of Glacial Channel (Maple River Valley). Sections 2 and 3, Liberty Township. One mile east (below) City of Enderlin. The rolling morainic topography in the region of the more eastern headwaters of the Maple system has a north-northwest trend in the alignment of the hills and ridges, which is considered evidence that the region of the eastern branch of the head streams of the Maple sys- tem had not yet been uncovered, this region being the latest to be freed from the burden of ice. FIG. 91. A Kame. One mile west of Sheldon. Photograph by C. M. Hall. CHAPTER THE FOURTEENTH. THE LAKES OF NORTH DAKOTA. The Kinds of Lakes. If we glance at a map of North Dakota it will be seen that all that portion of the State west of the Red River Valley and east of the Missouri River, except the Mouse River Valley, is dotted with lakes, and there are hundreds, yes, thousands, of small lakes not shown on even the largest maps. These are "glacial lakes" that is, lakes which occupy basins or hollows amongst drift hills. They are more com- mon among the hills of terminal moraines, and hence are often called "morainic lakes." Lake Agassiz, which covered the Red River Valley, Lake Souris, which covered the lower Mouse River Valley, and Lake Dakota, which occupied the Valley of the Lower James River, a small part of which lake extended into North Dakota, and Lake Sargent, covering most of Sargent County, were glacial lakes; but these owed their existence to the presence of the melting ice-sheet, and they lasted only so long as the ice-sheet remained to fill their basins with water, and at the same time to dam the northern drainage courses, except in the case of Lake Dakota, which, as we have seen, was dammed at its southern end by a ridge of hard rock. These lakes disappeared with the final melting of the ice-sheet; they are therefore called extinct lakes. The Cause of Existing Lakes. All existing lakes in North Dakota owe their being to the fact that the rainfall is greater than the evapora- tion, and the hemming in of their waters by morainic hills or other land barriers which form the sides of their basins. They are "glacial lakes," therefore, not because their waters come from the melting of the ice of a glacier, but because the glacier which was once here caused their basins to be formed among the heaps and ridges of earth left where it melted. A good deal of the drift is clay, and this holds water about as well as a porcelain dish. Wherever there is a hollow in which more water 151 152 THE STORY OF THE PRAIRIES. falls or collects than disappears by evaporation or soaking into the ground there will be a lake, and it is called a "glacial lake" if its basin was formed by the action of the ice of the great ice-sheet. All the lakes in North Dakota are glacial lakes. It is not necessary that the land forming the basin of a glacial lake should be entirely in drift deposited from the melting ice in order for it to be a glacial lake. The materials from the glacier may cause a lake to be formed without the entire rim of the lake being of drift. A river valley may be partly filled with drift so as to dam the stream and thus cause a lake above the dam. Such a lake would owe its existence to ice action and hence would be a glacial lake. It is likely that Devils and Stump Lakes were formed in this manner. Jim Lake and Arrow Wood Lake in Stutsman County were formed by the partial filling of the channel of the James River by the drift so that the river is com- pelled to spread out above the obstructions till the water rises high enough to flow over. The lakes of North Dakota vary in size from tiny ponds only a few rods across to those several miles in diameter. Devils Lake, the Lake Superior of North Dakota, is forty miles in length, measured in a direct line, and it is more than three hundred miles around its shore. Des Lacs Lake in Ward County is nearly thirty miles long, while only from a quarter to a half mile wide. Sometimes the depths of glacial lakes are very great in proportion to their sizes and sometimes they are large and shallow, broad, flat clay- pans filled with water. Sometimes the bottom drops suddenly to a great depth, and sometimes there is a gradual slope of the bottom from the shore toward the centre. In a similar manner, on "glaciated" land surfaces hollows are some- times deep with their sides abrupt and steep, and sometimes a broad "flat" merges gradually into surrounding hills. The deeper and steeper sided hollows in glaciated regions have been called "pots and kettles." The broad and more shallow ones might as properly be called "pans." "Pots and kettles" are very common in terminal mo- raines, and "pans" are common on rolling prairies between moraines. Exactly the counterpart or opposite of the "pots and kettles" are the steep, rounded knobs or knobby hills of terminal moraines. Pots and kettles and knobby hills wherever seen are a pretty certain indica- tion of a terminal moraine. A gently undulating prairie with shallow depressions generally indicates a ground-moraine. THE LAKES OF NORTH DAKOTA. 153 The great irregularity of the shores of many lakes in North Dakota is due to the fact that they are hemmed in by knobby hills, and if the lake is large there may be several ''pots" covered by the water of one lake, the water being very deep where are the pots and quite shallow between them, or knobs may rise up, forming islands. Lakes may diminish in amount of water they contain during dry, hot seasons, or they may dry up entirely during the driest part of the summer. Such are often called "dry" lakes. Lakes may also be "dry" for a period of years when the summers are seasons of unusual drought, and become lakes again during a series of rainy seasons. If a hollow is not deep enough to hold sufficient water to form a lake but rushes and marsh grasses grow upon its bottom it will be a slough or bog. There are thousands of such sloughs in North Dakota, and they afford some of the most valuable "hay-meadows" in the State. Sometimes a stream flows from higher land onto a tract of land so nearly level that the water is unable to cross it and so spreads out and forms a marsh or sw r amp. Such marshes, also often making valuable hay-meadows, occur upon the bottoms of old glacial stream channels. Good ex^ amples of this kind are the flat bottoms of the old outlets of Lake Souris west of Balfour, and the Big Coulee, and very many over the great Missouri Plateau where glacial channels were cut by the waters from the melting Glacier flowing across to the Missouri River. Since the walls which hem in the waters of glacial lakes are the materials dumped from the melting ice, and since these materials are often left in very irregular piles and ridges, the outlines or shores of glacial lakes are often very irregular, the shore-line of the lake winding around all the irregularities of the hills which hem in the waters of the lake. Sweetwater Lake, in Ramsey County, is a good example of such a lake having very irregular shore, though there are many hundreds of smaller lakes in the State which are equally good examples. In the case of a lake formed by the damming of a river valley by the drift the shore-line will follow not only the windings of the stream course and the curves around the hills dumped into the valley, but will reach out into the tributary valleys forming bays. The very irregu- lar shore-line of Devils Lake is probably due to all three of these causes. CHAPTER THE FIFTEENTH. SALT AND ALKALINE WATERS IN LAKES. The Salts in Lake Waters. The waters of many lakes are not only "salt," but they are often bitter. This is because there are bitter "salts" in the water. Our common table salt is what the chemist calls Sodium Chloride. This gives the "salt" taste to the w^ater. There is also Sodium Sulphate and Magnesium Sulphate in the water of many lakes, and this is bitter to the taste and affects the digestive organs of animals that drink it. There are also other salts such as the Sulphates of Potas- sium and Calcium (lime), and the Carbonates of Magnesium, Potas- sium and Calcium. If common salt or Sodium Chloride is present in the water in larger quantity than any of the others the water is called "salt" water. If it contains a larger quantity of some salt which is bitter to the taste it is apt to be spoken of as "bitter" or "alkali" water. Waters which are "hard" contain some kind of salt, usually Calcium Carbonate or Calcium Sulphate (gypsum). Rain water is "soft" be- cause when water is evaporated the mineral salt is left behind, and when the vapor condenses into clouds and falls as rain it is free from any salt. Not all waters which contain salts are "hard," nor are all "soft" waters free from salts. The waters from the artesian wells at Jamestown and Devils Lake are "soft," but they contain a large amount of salts. These waters are not hard because the salts in them are not such as to give the water the character of "hardness." Hard water is not good for washing because the salt in it forms a chemical combination with the soap and a new "soap" is formed which will not dissolve in water. The soap thus formed floats on the surface of the water, forming a greasy "scum." Hard waters are agreeable to the taste and are generally good for drinking if not too hard. Water which is hard from the presence in it of Calcium Carbonate can be "softened" or "purified" by boiling, which causes the limestone to fall to the bottom as a fine, white powder, or to 154 SALT AND ALKALINE WATERS IN LAKES. 155 collect in scales on the sides of the vessel in which it is boiled. This is called "temporary" hardness. Water which contains Calcium Sulphate or gypsum is "permanently" hard for it is not affected by boiling. The Sources of the Salts and Alkalies. The explanation of the origin of the salts in "alkali" waters lies in the fact that these minerals are in the rocks of the earth. The Cretaceous shales contain them, for they were present in the sea-waters at the time these rocks were deposited on the bottom of the ocean. We shall see in a later chapter that a great arm of the ocean once covered North Dakota and the rocks which underlie the drift were deposited as sediments on its bottom. The ice of the Great Ice-Sheet ploughed up these rocks and ground them into the fine soil, sand and clay which now covers the old land surface. What has been called in a former chapter "New North Da- kota" has been made from the broken and pulverized top of "Old North Dakota." The till or drift earth which was thus ploughed up from the Cretaceous shales has given to the soil its alkaline character. The salts, Sodium Sulphate and Magnesium Sulphate, are among the minerals in the soil, but other salts which dissolve in water, such as Potassium Sulphate and Sodium Carbonate, also occur, and altogether make up the "alkali" which distinguishes the soils and the waters of this region from those of the northern states farther east. The minerals or salts which make the water "hard" are Calcium Carbonate (limestone) and Calcium Sulphate (gypsum). These have been derived also from the Cretaceous shales. Pure limestone is the mineral Calcium Carbonate, and the drift which has come from a lime- stone region contains this rock pulverized in the soil, and so this be- comes a cause of hardness of the waters. In the Red River Valley and also farther West the drift contains a large amount of this rock which has been ground to powder, and this adds greatly to the productive- ness of the soil. These salts are therefore seen to be in the soil and when the rain falls upon the ground it dissolves them and becomes "hard" or "salt" again, and as the waters flow down the coulees or streams into the lakes and there again are evaporated the lakes become "salt" or "al- kali." If the lakes have outlets then the salt is carried on in the water which flows out of the lakes and away to the ocean, and as the ocean cannot have an "outlet" the waters of the seas become salt. Salt Beds on Dry Lake Bottoms. Sometimes a large inland lake be- comes so salt from the long continued evaporation of the waters, a little 156 THE STORY OF THE PRAIRIES. salt being generally present in the waters of the earth's surface, that the lake becomes a great tank of brine, and after a while becomes so "strong" that it cannot hold any more salt in solution, and finally salt begins to fall to the bottom. Or if the lake is small so that it frequently becomes dry the salt left by evaporation upon the bottom may not all be re-dissolved when the waters again fill the basin. If but little mud or fine earth is carried into the lake by streams and the "salt" in the water is mostly "common salt," beds of salt will accumulate on the bottom of the lake. These may become of considerable thickness and may be almost pure salt. Now, if for any reason a lake where this process has been going on for a long time should permanently dry up here might be salt beds of great value. Such salt beds occur in some of the Western States, where the dry salt can be shoveled from the ground in great quantities. It is said that salt has been shoveled up and hauled away in wagons for stock purposes from such salt lakes. CHAPTER THE SIXTEENTH. MAP STUDIES: DISTRIBUTION OF THE LAKES UPON THE LANDSCAPE. Map Studies; The Lakes of North Dakota. A map ought to mean more than dots and lines and shaded areas. We ought to be able to see in a map of the State a picture of the landscape. The "map stud- ies" in our geographies do not sometimes mean as much as they ought to. Let us notice the distribution of the lakes of our State and see if we can make these have a meaning as landscape features. In the light of the studies we have made in the preceding pages it will not be difficult to see that all the lakes in the State, while they are all "glacial" lakes and hence all belong in one great class, yet they fall into about a dozen groups, in each of which groups there is a meaning as a landscape feature. The McLean County Group. Look first at the group of lakes in McLean county. Does their position strike you as having any sug- gestion in it? Look at the Map. Figure i, and you see that the great Altamont Moraine, the one called the First, or the outer one formed at the edge of the great Dakota Glacier of the ice-sheet, makes a turn or loop toward the big elbow where the Missouri River turns south- ward. Some of the highest and most rugged and stony drift hills in the State are here. You notice that these lakes are in chains or sort of crooked rows. This is more than accident. When the ice of the great ice-sheet had its edge here great glacial streams poured from it into the Missouri River, and cut large valleys in the drift which had been left from the melting ice. Some of them also \vere probably val- leys before the ice came and were not entirely filled by the drift. These streams did not last long because the ice melted back so that the water ceased to flow through them. A short time though as used in geol- ogy is usually a good many years. Their bottoms were not in many cases made smooth by the streams, and when the ice had melted and the water was no longer compelled to flow through these channels the ii 157 158 THE STORY OF THE PRAIRIES. hollows remained and became filled with water and formed lakes. When the water is high in the spring it often overflows from one to another and may even pass to the Missouri River in some of these old channels. It may escape from Strawberry Lake near Dog Den Butte across by a long series of lakes and sloughs to the river, and in a similar manner from Brush and Pelican Lakes to the Missouri River. About forty miles west of Fessenden is Pony Gulch, a broad val- ley extending for many miles across the great Missouri Plateau, the Coteau du Missouri. This is a valley in which probably a stream flowed eastward before the drainage systems were changed by the ice filling them, but when the ice-sheet lay over all the eastern part of the State, filling all the river valleys, a glacial river probably flowed west- ward into the Missouri River, which you will remember was not cov- ered by the ice. The waters from Lake Souris were very likely car- ried across by this channel to the Missouri for a time, as we saw in another chapter. The hollow places along the bottom of this old chan - nel are now beset with lakes. The Kidder and Logan County Group. In northern Burleigh, Kidder, western Stutsman, Logan, and northern Mclntosh Counties is another group of lakes some of which also mark old glacial channels where the ice-waters surged over into the Missouri River. These lie in hollows among the hills of the First and Second or Altamont and Gary Mo- raines. These are all upon the top of the great Coteau, or Plateau of the Missouri, and hence are on the "Missouri Slope." It is probable that the James River flowed across by the Hawk's Nest in southeastern Wells County by this group of lakes in Kidder County to the Missouri River at the time of the formation of the Third or Antelope Moraine, for at this time the ice-sheet covered the land as far west as Carrington, and its edge lay upon the plateau to the south, so that the river could not follow its present course southward. The Chains of Lakes Another group consists of the lakes in Foster, eastern Stutsman, and western Barnes Counties. The James River flows for nearly thirty miles along the course of the Fourth or Kiester Moraine. The river probably begun to cut its channel here when the moraine was being deposited from the melting ice and the river flowed along the edge of the ice. Sometimes the materials from the moraine were dumped into the channel of the river so< that its waters were dammed up and lakes were formed. Such lakes are the Jim and the Arrow Wood, in northern Stutsman. MAP STUDIES. 159 The Spiritwood Chain of Lakes and four other chains of lakes which cross or lie near to the Northern Pacific Railway between Valley City and Spiritwood station, lie in deep channels which were the places of large glacial streams during the time the ice-sheet was melting back from the position of the Fourth or Kiester to the Seventh or Dovre Moraine. Lake Eckelson lies in one of these old channels which is five miles long extending south to Walker Lake. Another lies about two miles west, and the old channel is six miles long. Another also about six miles in length is just east of Sanborn, and there is still another extending south from Hobart. These lakes are along the bottoms of channels forty feet below the general land surface. These channels may mark the places of old val- leys on the pre-glaeial landscape which were not filled by the drift so but that the flood waters from the melting ice flowed in their courses and cut these channels in the soft drift which partly filled them. A Picturesque Group in Griggs County. One of the prettiest groups of lakes in the State and surrounded by the most picturesque morainic hills is that in Griggs County, and extending also north into Eddy County. The group consists of Lakes Jessie, Addie, Sibley, Clear, and Red Willow, besides many small ones, and also the North and South Washington Lakes in Eiddy County, and Free People's Lake, on the Indian Reservation north of the Sheyenne River. From Devils Heart Hill across the Sheyenne at the Morris ford to McHenry and Coopers- town is a continuous series of lakes and hills. West of Cooperstown are the high, steep, rounded knobs of the Dovre Moraine, rising sev- enty-five to one hundred and fifty feet above the surrounding prairie, covered often thickly with large granite and limestone boulders, and among these hills are the silvery sheets of water of the lakes named. The Washington Lakes in Eddy County are walled in between the hills of the Elysian and the Waconia (Fifth and Sixth) Moraines. These lakes are interesting as having old forests with their stumps still stand- ing below the water along shore, showing that the water has been much lower in them at some time. The cut banks or cliffs on the sides of these and others of the group show that the water has also been considerably higher than it is now. The Devils Lake Group. The long series of lakes extending from the small sheets east of Stump Lake for more than 100 miles northwest nearly to the Turtle Mountains, including from the east the two small lakes east of Stump Lake, Stump, Devils, Ibsen, Hurricane, Grass, 160 THE STORY OF THE PRAIRIES. Island, and Long Lakes, as has been explained before, probably were all formed in the valley of an old or pre-glacial river by the partial filling of this valley with drift. These lakes, therefore, have a quite different meaning as landscape features from those in the Griggs County group just described, which are "morainic lakes" pure and simple. The Group North of Devils Lake North of Devils Lake is a group of lakes which are cut off from draining into Devils Lake by the range of morainic hills which lies between it and them. These lie in broad flat hollows or "pans." This range of hills, which belongs to the Itasca (Tenth) Moraine, lies close along their southern shores and holds their waters from escaping into- Devils Lake, their waters pushing up into the hollows between the hills forming many small bays. Quite a large area to the north is drained into these lakes, and in times of high water or during periods of years when the amount of rainfall is greater, these lakes increase in size, rising and spreading out in area, and become connected by sluggish streams. They may thus at times become connected with the Mauvaise Coulee and so drain for a time into Devils Lake. Sweetwater Lake has sometimes risen high enough so that its waters overflowed the rim of its basin and dis- charged directly across to the south into Devils Lake. If the position of these lakes in relation to Devils Lake is noticed it will be seen that the three larger, Sweetwater, Dry, and the Twin Lakes (Lake Irwin and Lac aux Morts or Lake of Death), lie directly north of the three large bays or arms of Devils Lake. It has been suggested already that Devils Lake probably lies in the hollow of an old partially filled valley and that its larger bays or arms may be due to tributary valleys entering the old main valley. The position of these lakes, with the moraine forming a barrier to prevent their draining into Devils Lake, suggests that they may lie in the same tributary valleys in which the three large arms of Devils Lake lie, and that the moraine which crosses these tributary valleys in an east and' west direction dammed their courses and so caused the lakes to gather above where the valleys are filled. This suggestion of valleys in the old or pre-glacial landscape of this region is further strengthened by the fact that wells which are dug or drilled about the City of Devils Lake and in the surrounding country vary very much in depth within short distances, but all penetrating clown below the drift to the old land surface, the Cretaceous shale. MAP STUDIES. 161 Along the International Boundary. A number of lakes lying near the International Boundary and east of the Turtle Mountains are good examples of a class of lakes which owe their existence to the action of the ice-sheet, and hence are ''glacial" lakes, but which are not mo- rainic. They lie in shallow 7 pan-like depressions in the region between moraines. Their basins often have their bottoms and sides in glacial clay of till, but they may also lie in hollows which were scooped out in the Cretaceous shales by the moving ice. There are many such lakes in the State. Those here described lie between the ranges or belts of hills of the Itasca (Tenth) Moraine. It has been before explained that the Itasca Moraine extends across the northeast corner of the State in a northwest and southeast direc- tion. It is a compound moraine, being made up of several ranges or belts lying between Devils Lake and Pembina Mountain. Lying between the belt or wide range of hills which extends from the shore of Lake Agassiz west of Inkster northwest to the International Boun- dary in the northeast corner of Towner county, and the moraine lying upon the top of Pembina Mountain which was formed on the east side of the Dakota Lobe or Glacier, are Rose Lake, six miles east of Lang- don, Rush Lake, near Hannah, a small lake near Mt. Carmel, east of Hannah, and a fourth about five miles north of Osnabrock. Between the range which lies west of these lakes and the next large range still farther west, which extends from north of Lakota to the east of Cando and to the Turtle Mountains, lie Lac des Roches, near the Interna- tional Boundary in Towner county, which has been before spoken of as lying in the line of outlet of old Lake Souris, and Rock Lake, about four miles west of Lac des Roches. All these lakes have inflowing streams or small coulees feeding them. The four first named are drained by outlet streams, Rose Lake being drained by the Tongue River, Rush by the Pembina River, Mt. Carmel by the Little Pembina River, and Osnabrock by Park River. All these streams have cut deep channels into the Pembina Mountain highland, for in flowing down its steep front their currents become swift. They have worked back and "tapped" these lakes since the ice-sheet melted, that is, since the close of the Glacial Period. In a short time short as time is measured in geology these lakes will have been drained and become meadows, for their outlets will be cut down and their waters will be drawn off. But with Lac des Roches and Rock Lake the case is different. No 162 THE STORY OF THE PRAIRIES, stream has worked back so as to tap them. Streams have worked back from their basins into the higher land which surrounds them and now bring water to them but this only makes them spread out the more. The old channel along the course of Mauvaise Coulee is almost up grade till after it crosses the belt of morainic hills to the south. When the flood waters of Lake Souris came this way it forced a pas- sage over the moraine after Lac des Roches had spread out so that it and Rock Lake were probably united into one large lake. The Turtle Mountain Group. When it is remembered that this plateau rises about 600 feet above the surrounding prairie it will seem FIG. 92. Township 163, Range 74. Top of Turtle Mountain, showing the great number of small Morainic Lakes. the more surprising that there are so many lakes on its top, for there are probably not less than 200 large enough to be shown on a map. Fish Lake, twelve miles north of Bbttineau, its north end extending a little across the International Boundary, is one of the prettiest sheets of water in the State. Other lakes worthy of note are Kippax, Constance, Butte, Magog, Waukastian, Nemo and many others. It will be seen from Figure i that a broad belt of the Itasca Mo- raine crosses the Turtle Mountains, and many of the lakes are "mo- MAP STUDIES. 163 rainic" lakes scattered amongst the rounded hills in little round and irregular hollows. The Turtle Mountain Plateau is about forty miles long and twenty- five miles wide in its widest part, lying mostly in North Dakota. Its top is forest covered, and very much broken and rolling due to the drift hills of the moraine just spoken of, and also to the fact that the plateau was cut up by creeks and coulees before the ice-sheet pushed across it, and many of these hills were too large to be entirely leveled down. In the hollows among the old, that is, the pre-glacial, hills, which are often only partially rilled with drift, occur many of the lakes. The Big Coulee Group. Buffalo and Girard Lakes and a series of small lakes in Pierce county lie in the valley which was the outlet of Lake Souris before the time of the outlet by Lac des Roches. These lakes differ in character from the great number of small lakes which are scattered among the hills in that neighborhood which are morainic lakes. These which lie along the bottom of the old channel are formed by the water which collects there from rains, and which cannot escape because there is not enough water to flood the old channel so as to cause a current. The water, therefore, stands upon the bottom in the low places, forming lakes and sloughs. "Big Coulee" Creek is a small stream lying in the valley, one of the head streams of the Sheyenne River, but it is a mere little pool which soaks along the bottom of the great wide valley. As the Sheyenne cuts its channel deeper and low- ers the mouth of the Big Coulee Creek this creek will become more swift at its mouth and drain the lakes which lie upon the bottom of the old channel. A Group of Typical Morainic Lakes. The great number of lakes in western Benson, Pierce, and eastern McHenry Counties are morainic lakes, small and larger basins, or "pots and kettles," of water hemmed in by the steep, irregular, and knobby rounded morainic hills. Many of these hills are sandy, from the ground-up sand-rock of the Fox Hills Sandstone (one of the Cretaceous formations) which is the underlying rock south of the Turtle Mountains. Some of the lakes in this group lie among the sand blown hills (dunes) of the Lake Souris bottom. These hills, which were dumped into the lake as moraines, were not entirely leveled by the waves of the lake, but the sand of the hills on its southeastern shore was washed and assorted by the waves, and this is now blown by the winds into dunes, sometimes filling the 164 THE STORY OF THE PRAIRIES. lakes which lie in the hollows. The lakes are sometimes entirely filled, just as trees are buried by the drifting sand. The "Alkali Flats." South and west of Balfour, Anamoose, and Harvey, lying along the foot of the high front of the great Missouri Plateau, are what are known as the "Alkali Flats." Many broad flat bottoms are occupied by shallow lakes or sloughs. The water is strong of alkali, as are all the lakes farther west which have no outlets. The headwaters of the Sheyenne River have a sluggish beginning in this region, but there is almost no fall toward the Sheyenne in the flat surface from Dog Den Butte to western Wells County. A coulee from Pony Gulch and others from off the high front of the plateau flow out upon the "flats" and spread out as lakes. The shores and dry bottoms of the lakes are white from the "alkali" salt left by evapo- ration during the dry season of summer. These salts dissolve again in the water when the wet season returns. When either a natural or an artificial system of drainage shall carry away the surface waters from these regions "the flats" will become valuable lands. They are rendered nearly valueless now by the ac- cumulation of alkali by evaporation of the waters. The "Alkali" Lakes in the Far West. The lakes in western Ward and Williams Counties which lie upon the top of the high northern portion of the great Missouri Plateau, are among the most strongly alkaline, if not the most so, of any in the State. This is because the rock strata of which the plateau is composed, known as the Laramie formation (Cretaceous), are even more alkaline than those rocks which underlie the drift in the central and eastern portions of the State. Many of these alkali lakes lie in hollows in the underlying rock, which is covered by only a thin mantle of drift. The water of some of these lakes is a bitter brine. The River of Lakes. Another group of lakes in Ward County is of more than usual interest. This is a group nearly forty miles in length which lies in the old glacial river valley which once brought the waters of Lake Saskatchewan from far north in Canada, and also the waters from the edge of the melting ice all along its course, to Lake Souris. There are three lakes in this series, the one farthest east and south being a small, pretty sheet of water one or two miles long. This is followed by a marsh and low meadow which separate this from the next lake, which is about three miles long. Then a marsh and a meadow again follow for a mile or two, and then a continuous and beautiful MAP STUDIES. 165 sheet, or silvery ribbon, of water extends to the northwest for thirty miles, having a width of about half a mile, its northern end extend- ing about two miles into Canada. Des Lacs River, which drains (?) these lakes is a small, narrow ditch winding back and forth across the flat bottom of the broad and deep valley, and enters the Mouse River at Burlington, about five miles west of Minot. Salt Lakes From Artesian Springs. A remarkable group of lakes lies upon the level prairies in Grand Forks and Walsh Counties, between the city of Grand Forks and Grafton and north of Graf- ton. These are salt lakes which owe their origin and the saltness of their waters to the same causes as those which produce artesian wells in the Red River Valley. Springs which furnish salt water burst out upon the level prairie, the water having the same source far west of the Red River Valley as the water which is obtained by drilling artesian wells on the Red River bottom lands. In fact, these springs are natural artesian wells, the water being forced up through gravelly veins in the drift or till which fills the valley, and having its "head" or source in the high lands which flank the Rocky Mountains. These springs make the streams which start upon the highlands which formed the western shore of Lake Agassiz streams of salt water. There being not enough fall to the almost level prairie to cause drainage into the Red River their waters spread out into marshes and lakes, and the water which comes to the surface in the region of the lakes in springs adds to their volume, and hence the salt marshes. CHAPTER THE SEVENTEENTH. LAKES AS A LANDSCAPE FEATURE. The Meaning of Lakes on a Landscape. Lakes as a landscape feature mean "youth," that is, the landscape is young in the sense that there has not been time for river systems such as were described in the first chapter to be developed. The landscape is as yet largely undrained by streams. A comparison of that portion of the State lying west of the Missouri River with the great portion east of it will show the difference between an "older" and a "younger" landscape. We have noticed already the many lakes scattered over that part of the State which is east of the Missouri River. West of the river we see none marked on the map, for there are none. If there were once lakes there they have been drained. All of the hollows have outlets, and are valleys. East of the river most of the hollows do not have outlets, and are basins. The landscape west is therefore "older;" that east is "younger." West of the river drainage systems have become estab- lished, and streams have cut the landscape into hills, and these hills are being worn down and carried to the sea. East of the river few streams mark the landscape, and the cutting of the prairies into hills has just begun. West of the river the hills have been carved upon the face of the landscape. East of the river the hills are mostly "dumped" hills, or heaps and irregular ridges piled upon the landscape. What has been the cause of these marked differences we have al- ready seen. It was the great ice-plow which leveled down the hills and filled the valleys of the original landscape and piled these hills on the surface as it melted away. As this great ice-sheet reached only to the Missouri River the region west of this river has not been ploughed down and leveled and covered with dumped hills. There the landscape is "older" because the processes which carve and fashion all landscapes have been going on longer than east of the river, where they had to begin all over again after the Glacial Period. LAKES AS A LANDSCAPE FEATURE. 167 The rocks are not any older in years west of the river than they are east of it; in fact, the oldest rocks in the State, as to the time they have been in existence as rocks, are in the eastern part of the State, as we shall see in a later chapter. It is the form of the landscape which is older. When the hills west of the Missouri River have all been washed away, or nearly so>, so that there are no high, steep, flat-topped hills, and the whole region is worn down to base-level, then the landscape will have reached its old age. In all the State east of the Missouri River drainage systems are just getting started. These are the "coulees" which, starting from the river valleys, old channels and lakes, have pushed back upon the landscape. Wherever there is a low place water collects from the falling rain and lit- tle streams begin to work back into the surrounding land. In time larger streams will become established and their heads will work back into the surrounding land and tap the lakes. The lakes will be drained by the cutting down of their outlets, and so in time there will cease to be any lakes, and the prairies will have been cut up into hills. The rapidity with which river systems get started in any particular region depends upon the mouths of the streams. If the streams pour their waters into a deep basin, or if they fall suddenly down from a highland or plateau upon a considerably lower plain, they will cut their channels down and push their heads back rapidly, and the highland will become soon dissected into hills. The landscape may be said to "grow old" rapidly. But if the whole region is low, that is, if there is no place which is quite a good deal lower into which the waters can discharge, then streams will push back upon the landscape and cut their channels very slowly, and the rain which falls upon the land will lie upon its surface and in the soil till evaporation removes it. Nearly perfect examples of landscapes which are "growing old" rapidly are the plateau top of Pembina Mountain, and the top of the Turtle Mountain Plateau. Of those which are lingering long in the youthful stage are the almost perfectly flat plain of the Red River Val- ley, and the region of the group of lakes north of Devils Lake. To the latter class, however, belongs most of the State east of the Missouri River. All these regions began their "infancy" nearly at the same time, which was after the close of the Ice Age, or the Glacial Period. But the region of eastern Cavalier and western Walsh Counties, and the top of the Turtle Mountains, will be cut up into hilly landscapes and be 168 THE STORY OF THE PRAIRIES. reaching "middle age" while yet the plain of the Red River Valley and the region north of Devils Lake, as also much of the State else- where, will still be in the age of youth. This is because the Reel River has so little fall that it cannot deepen its channel, and so the coulees upon the prairies cannot lower their mouths, and the water which falls upon the broad level prairies stands in sheets until removed by evaporation. In the Devils Lake region the fall in any direction is so slight that only the faintest beginnings of drainage have been developed. Mauvaise Coulee enters Devils Lake from the north, but it cannot be said to drain the lakes with which it is connected. It is itself a long-drawn-out slough or pool which is broader at those places where it spreads out into lakes. In the case of the Pembina Mountain top all the streams which fall down its steep front have cut deep channels. The Pembina River, the Little Pembina, the Tongue, the head streams or coulees of the Park, the Forest and the Turtle, have all cut deep channels down through the drift into the underlying shales. This is because of the fall from the top of the high plateau down to the low prairie. These same streams all become sluggish pools after they get upon the valley plain and their channels become long, puddling ponds. The high prairie upon the plateau top of Pembina Mountain will become cut up into hills while the Red River Valley still remains almost undrained. All the larger lakes upon this plateau have already been tapped by the head coulees of the streams named. The Turtle Mountain Plateau is being cut into by the coulees which push their heads back from the prairie up the steep slope of the high front. All around the mountain on any good map streams are shown which are pushing their heads back onto the higher land. The old valleys which were partially filled with drift, many of which were dammed, forming the small lakes, will be cut out anew, and the lakes which are scattered among the hills in the hollows will be drained. Fish Lake and the series of lakes lying near it are in an old valley which was partly filled by drift. Oak Creek has cut a deep coulee into the side of the mountain and already draws away water from several small lakes in the series. When these have been drained by the deep- ening of the channel it will later draw off the water of Fish Lake, and finally the whole valley will be re-opened something as it was before the great ice-plow moved across the mountain's top. A good illustration of the tapping of a lake by the cutting down of LAKES AS A LANDSCAPE FEATURE. 169 a coulee channel and the pushing back of its head is furnished by Rush Lake, on the Pembina Mountain highland near Hannah. It will be seen in Figure 93 that it has two outlets. The lake is a shallow clay-pan of water only a few feet deep. The north outlet is the old outlet, one which was established when the water from the melting ice-sheet made the lake larger than it is now. Pembina River, FIG. 93. Map of Rush Lake, Cavalier County, showing two outlets. From a Drawing by W. A. Hillier. which is only a few miles away, flows in its old, or pre-glacial, channel. This was partially filled with drift, which has been mostly carried away by the river, and the river has cut its valley still deeper. Snowflake Creek has cut back from the Pembina Valley as a tributary, and it, too, has cut a deep gorge or channel, because its mouth is made low by the deep gorge of the Pembina into which it empties. Now, it chanced that a low place in the land surface caused a small tributary, /, to cut back from Snowflake Creek at the fork where the two outlets now meet. Snowflake Creek at first had its source in Rush Lake through the north outlet. But this little tributary has cut down more rapidly than the north outlet owing to the fact that more water falls over its sides, it being in a slight depression, and so it has pushed its head rapidly back. 170 THE STORY OF THE PRAIRIES. It happened that a coulee, c, leading into the lake at o marked a little valley. This had its head about where the bend in the south outlet is now. At length the little tributary coulee from Snowflake Creek pushed back and began to draw the water of the little coulee the other way. So the little coulee which at first flowed into the lake was reversed and its channel became a part of the new south outlet. This is just about the stage in which the two outlets are now. The north outlet is still the main outlet of the lake, that is, it carries a little more water from Rush Lake to Snowflake Creek than does the newer south outlet. But the south outlet at t has a deep gorge and it is rapidly cutting this gorge back so that it will soon lower the channel of c, and this will then become the principal outlet, and soon the north outlet will cease to carry away water from the lake entirely. Because the channel of Pembina River is deep Snowflake Creek is able to cut deeply its channel, and soon the rim of Rush Lake at o will be cut down and the waters of the lake will be drawn away, and its bottom will become a meadow. 172 CHAPTER THE EIGHTEENTH. THE BAD LANDS. Bad Lands to Travel Through. No part of North Dakota is perhaps more widely known or less understood than that part of the State styled the "Bad Lands." Probably about no part of the State are there more mistaken notions than about this region. In the first place the lands are not "bad" for the purposes for which nature has fitted them, viz., for stock-raising. This is claimed by those who ought to know to be one of the best parts of all the State for profit- able cattle- and horse-ranching. The region was called by the early French travelers who crossed the country in wagons "Bad Lands to Travel Through," which is a very fitting and appropriate title. The lands are not, in fact, so "bad," but they are bad to travel through. The whole region is so much cut up by deep valleys with steep sides that it is almost impossible to travel there with wagons. And the tourist who attempts to travel on horseback without a guide is very likely to "get lost." Mistaken Notions About the Cause of the Bad Lands. Many strange stories have been invented by travelers to explain how the lands of this region came to be so very rough. They are often described as having been made rough and jagged by great volcanic up- heavals or earthquake shocks ! There is scoria in the hills or "buttes," and this has given color to the notion that great volcanic fires have raged here, and the high crags and rugged hillsides with deep, narrow valleys appear to those who have keen imaginations like great rents or fissures in the rocks caused by earthquakes. Then there are many veins or beds of coal in the region and some of these are burning, and this has given rise to the idea that great fires have burned out the chasms, or that the coal has burned out underneath and the rocks over- lying have then fallen in, causing the steep-sided, ragged gullies. But careful observation and study will show that none of these causes is the true one. The earth has not been formed in the way we now see it by sudden changes. Great upheavals of the earth's crust 173 174 THE STORY OF THE PRAIRIES. forming mountain ranges and volcanic outbursts causing floods of lava to pour out upon the surface from the depths of the earth, have occurred in many parts of the earth, and the form and appearance of the earth have been greatly changed by such processes. But these are not the causes which have made the landscape features of the Bad Lands, nor of any part of North Dakota. The Real Cause of the "Bad Lands." The agent which has fashioned the landscape in the "Bad Lands" is the same as that which has been working ever since the solid crust of the earth first appeared above the seas, and dry land began to receive rainfall and to be worn away by it. The "Bad Lands" have been cut up into "Bad Lands to Travel Through" by the action of running water, just as the plateau top of Pembina Mountain is being cut up into hills by the action of streams. The buttes or flat-topped hills of this remarkable region, the deep valleys or gorges which surround the buttes, often so steep that neither man, horse nor wild beast can cross them, have been made by the eroding action of running water. The same processes of valley cutting which were studied in Chapter One are the explanation of the "Bad Lands." The "Bad Land" Region. A belt of country from ten to twenty miles wide in North Dakota along the course of the Little Missouri River is deeply intersected or cut into by this river and its tributary streams. The channel of the Little Missouri has been cut by the river deeply into the landscape so that the streams which flow into this river have steep bottoms, that is, they descend rapidly, and this causes them to erode or cut down their beds rapidly. Their sides, therefore, be- come steep and rugged. These tributary streams push back their heads into the land, as has been explained before, and often their heads work back into the plain so that they meet, and so a portion of the prairie becomes cut around by the streams forming a table-land. If this bit of land thus surrounded by deep valleys is large it is called a plateau or a "mesa." If it is a small area so that it is simply a flat- topped hill it is called a "butte." Buttes and mesas are flat on top because the original plain or prairie was flat. The sides of the valleys or coulees are steep and rugged because the streams which form them cut down rapidly, and we have seen before that swiftly flowing streams erode their bot- toms much more rapidly than do streams having slow currents. In the spring when the snows are melting, and during seasons of heavy rainfall these streams are swollen and flow very swiftly. The THE BAD LANDS. 175 FIG. 95. View South of Missouri River near Williston. Photograph by Rex Willard. FIG. 96. Naked Clay Butte with Fluted Side. McKenzie County. Photograph by Rex Willard. 176 THE STORY OF THE PRAIRIES. rocks which make up the landscape are clay and sandstone and shale, and such rocks erode very easily under the cutting action of swift cur- rents of running water. During the hot months of summer the coulees become mostly dry, the clays become "baked" and cracked by the sun's heat, and the sandstones and shales become crumbled. Then when the rains come and the snows melt the rocks are easily broken and car- ried away by the waters. Grass and other vegetation do not have time to get much foothold on the steep sides of the buttes because the earth wears away too rapidly, and so the sides of the buttes are gen- erally naked of vegetation, except in crevices where the washing is less. The layers of clay, sandstone, shale, and. often of lignite coal, are seen in parallel series one above the other just as they were laid down on the bottom of the ocean. Different Forms of Buttes. The sides of the buttes are worn away year by year as the rains continue to wash their sides, and the sun shin- FIG. 97. Pyramid Butte. Photograph by Prof. W. E.Johnson. ing upon their unprotected sides, and the frosts of winter, crack the clay and crumble the sandstone and shales, and the areas of the flat tops become smaller and smaller. By and by the flat top is entirely worn away and the butte "comes to a peak," and then the peak becomes lower and lower as the wearing process goes on. Thus there may be large mesas and small mesas, the small mesas grading in size THE BAD LANDS. 177 into large buttes, and large buttes differ from small buttes only in the lesser areas of their tops. A large mesa may be cut up into smaller mesas or large buttes, and larger buttes may be cut into smaller buttes, by coulees pushing back and cutting up their tops. Finally the flat tops become rounded tops, and then the buttes begin to get lower, so that there are higher and lower buttes, and as the low, rounded buttes become still lower and smaller they in time wear away and become mere little naked, rounded hillocks or "bee-hives." If there should be a harder layer of sandstone running through the butte the edges of this harder layer will not be worn away as fast as the rest of the softer materials, and so this layer will come to project out of the sides of the butte as a shelf. If the shelf is at or near the top of the butte then the butte will become a "table rock" or "capped butte." Sometimes a harder part of a sandstone layer or of lava form- ing a crag or jutting mass, stands out on the side or at the top of a butte, and so a "pinnacled" butte is formed. FIG. The Butte becomes a "Table Rock " or "Capped Butte." Pltotograph by Miss Nellie T. Cruden. Outside the Bad Lands. The landscape about Dickinson, forty miles east of the Little Missouri River, is that of a broken prairie. The val- leys are not so deep because the headwaters of the Heart River have to go a long way to the Big Missouri at Mandan, and this means that the "fall" is not so rapid so that the streams cannot cut their channels down so rapidly. If the headwaters of the Heart River did not have 178 THE STORY OF THE PRAIRIES, to travel the long journey of about 100 miles to the Missouri River, that is, if there were a place as low as the Missouri River at Mandan, only say ten miles east of Dickinson, the country about Dickinson would rapidly become "bad lands," for the streams would quickly cut down their channel bottoms and the prairie surface would soon become the tops of buttes. This is what has happened thirty miles west of Dickinson at Fry- burg, where the streams flow west into the Little Missouri, descending 530 feet in a distance of about ten miles. About twenty miles north of Dickinson are what are called "the breaks." Here the gently rolling and grass-covered prairie changes suddenly to a much broken and rugged landscape with narrow valleys and buttes with naked sides. Here a stream with a deep valley and many tributary coulees, the Knife River, has pushed back into the landscape from the Missouri River at Stanton in Mercer County, and the development of "bad lands" has well begun. Farther north, after crossing the region drained by the Knife, there is rolling prairie again. At a distance of about sixty miles north of Dickinson the prairie suddenly drops off, as abruptly as off the end of a bridge, into the val- ley of the lower Little Missouri at the bend where it turns east to enter the Big Missouri. Here the Little Missouri has cut its valley down like a great trough 400 to 500 feet into the prairie, and tlie side streams have cut the landscape on each side of the river into the most striking and majestic buttes anywhere to be seen in the North Dakota Bad Lands. The change from the grass-covered prairie to the steep and naked jagged buttes of the Bad Lands of the side of the valley-trough is as marked as stepping off from the edge of a plank platform. The traveler often has to go along the edge of the prairie for many miles before finding a coulee he can descend to the river, although the dis- tance to the river in a direct line is less than four miles. There are only one or two places in a distance of thirty miles where it is practi- cable to get down to the river, ford the stream with its treacherous quick-sands, and get up again upon the prairie on the other side of the valley. Yet it is possible on a clear day to see across from the prairie on one side to the grass-covered prairie on the other, the dis- tance across, which in this region represents the whole width of the "Bad Lands," being from seven to ten miles. But the journey down from the prairie to the river is a most diffi- cult one. Jagged, rough and steep, down into holes cut out by tor- THE BAD LANDS. 179 rents of water, around slippery clay buttes, down deep and steep gorges, over hard crags of sandstone which have resisted the wearing action of sun, frost and y/ater, passing sometimes a butte in the sides of which glisten countless crystals, passing with caution over a ledge under which burns a vast natural furnace of coal, till at length the bot- tom of the valley is reached, where roll the waters of the Little Mis- souri, yellow with their burden of sand and clay. Halting at the hospitable door of a ranchman's log "schack," glad to rest and hear again the sound of a human voice, one may well gaze FIG. 99. " Halting at the Hospitable Door of a Ranchman's Log 'Schack.' " back in awe and wonder at the lofty gray precipices which have been passed. "Bad Lands to Travel Through" indeed! But there is never a lack of a cordial welcome at the humble, thatched cottage of one of these ranchmen "cattle kings." True, there is nothing to drink but the warm water of the river, and this is so muddy from its sediment- laden current even in mid-summer that it is impossible to see the bot- tom of a spoon which is filled with it. But anything is good enough, and the best the ranchman has he deems none too good for the welcome traveler. Here rolls the swift-flowing and sediment-laden Little Missouri, at once the cause and the explanation of the "Bad Lands." Its bed de- scends rapidly so that its waters flow swiftly, carrying a great burden of sand and clay, in some places little else than a great moving stream of quicksand creeping down the valley. Rolling on and on, bearing its mighty burden of sand and clay down its steep course to the Big Mis- souri, it adds to the muddiness of that great dirty river this pudding of 180 THE STORY OF THE PRAIRIES. rock, its waters stirred to a soup with clay, a burden which it has brought from all the coulees which girdle the buttes of all the "Bad Lands" from its long course in South and North Dakota. The Structure of the Buttes. One of the most striking things which the traveler observes in the Bad Lands is the arrangement of the rocks in the naked buttes in horizontal layers. So far from the region being one which has been rent and broken and upheaved by volcanic or earth- quake action, so far from the rugged form of the hills being due to heat from eruptions of the earth, as has sometimes been said in descriptions of this region, the rocks are all horizontal in position and FIG. ioo. "One Layer above another like Boards in a Lumber Pile." Pyramid Park. Photograph by Prof. W. E.Jshnson. one layer or stratum above another in as systematic order as boards in a lumber pile. Rocks which have been upheaved and crumpled and melted in the processes of mountain making are upturned, broken, and bent, and the character of the rocks themselves changed, so that what had been soft clay or shale has been changed into slate, and sandstone into quartzite in which the grains of sand cannot now be distinguished. But no such- changes have occurred in the Bad Lands. The rock-layers are in horizontal position just as they were laid down as sediments on the THE BAD LANDS. 181 bottom of the sea long ages ago. The layers of clay are still clay, and the sandstone strata are sandstone now, made up of the same grains of sand as they were when the waves of the sea washed them. All these layers of rock belong in what is called the Laramie forma- tion, the highest in the series, or latest formed, of the Cretaceous rocks. That these layers of rock, these sandstones, shales, and clays, were formed on the bottom of a great body of water there can hardly be doubt, for nothing but water can form clay or fine sand into such layers. Follow along any particular layer in the side of a butte and then- look across to the other side of the valley and the same layer occurs there at the same height. Follow It on and it has the same position in all the buttes. There are the same layers above it in all its course, and those layers which are below it can be seen below it in all the buttes. The edge of a particular layer as it is seen in the side of the hill is like a great ribbon stretched along the side of the valley. It keeps just the same thickness from one butte side to another as far as it is fol- lowed, till it finally plunges into the ground below the level of the stream bottom, if it is followed up-stream, or rises, a little toward the surface or top of the side of the valley if it is followed down-stream. This is what would be expected if the layer itself is horizontal, for the stream bed at the bottom is not horizontal but rises up-stream, and so the layer seems to come down to meet the bottom of the valley, though really the stream bottom rises to meet the layer. The Rat land at the top of the buttes has a slope down-stream or else the stream would not have been started, and so the layer tends to rise more and more toward the top when followed down-stream. The only explanation of layers of rock so extensively horizontal is that they have been deposited in water upon the bottom of an ocean. Veins, or beds, of lignite coal occur along with the layers of clay, sandstone and shale. They are of various thicknesses from less than an inch to eight feet or more, and these can be followed along the naked sides of the hills or buttes like the other rock layers. Now if lignite coal, while under the great pressure of the weight of the rocks above it were to be greatly heated, as it would be if volcanic action or earthquakes had caused great upheavals and rents in the earth, it would be changed to anthracite coal and cease to be lignite. The fact that there are beds of lignite coal all through the Bad Lands, therefore, is a 182 THE STORY OF THE PRAIRIES. proof that heat from earth eruptions was not the cause of the Bad Lands. In many places in this region clay has been heated by the burning coal mines so that it has been baked into brick, and sometimes also it has been melted so that it looks much like lava. Where the sides of the buttes have crags of this melted rock projecting in great masses the region has sometimes the appearance of having been rent by volcanic eruptions. Natural brick, which has been baked by the heat of burning coal mines that have smouldered in the bosom of the hills during centuries, and FIG. 101. "Masses of Scoria lie upon the Surface, forming Crags and Pinnacles." Photograph by Miss Nellie T. Cruden. scoria, which is melted clay, are extensively used on the Northern Pacific Railway as ballast for the road-bed, and also at many of the ranches for making walks and the floors of stables. Six miles east of Medora at Scoria the buttes look as though they had been deluged with blood, and immense masses of the hard scoria lie upon the surface crowning the buttes, and forming huge ragged crags and pinnacles. Many outcroppings of scoria and burned-clay brick occur also south of Medora in the buttes along Custer Creek and other small streams entering the Little Missouri River, shown by the red THE BAD LANDS. 183 color which extends down over the lower rock layers, having washed from the red layers above. Some Places of Interest. The Northern Pacific Railway crosses the Little Missouri River at Medora. This is often spoken of as "the heart of the Bad Lands," though it is not in fact so "bad" here as at the point described farther down the river, for, as the valley is deeper farther toward its mouth, the buttes are higher and the chasms deeper. Me- dora is an interesting spot, and the traveler who wishes to see and study the Bad Lands will find no more favorable place so easily accessible. It was at Medora that the French nobleman, the Marquis de Mores established his once famous stockyards and slaughtering houses, intending to make this a shipping point for dressed beef from this great cattle-raising district. The name Medora was given to the town in honor of his wife, who was an American lady. The baronial residence, the Mores Castle, still stands on a beautiful bluff overlooking the river and the town. The buildings which were intended to be used for the slaughtering and packing industry are still standing. About two miles south of Medora is the old trail by which the ill- fated General Custer led his army across the Bad Lands in the famous FIG. 102. " Custer Trail Ranche " is a good place from which to see the " Bad Lands." Photograph byj. J. Freeman. 184 THE STORY OF THE PRAIRIES. campaign against Sitting Bull in 1876. The buttes on which the picket guards were stationed while the army was encamped here are pointed out, and the marks of the trail made by the wagon wheels, and also the marks of the tent-pins at the camping place, are still visible. Custer Trail Ranche, two miles south of Medora, named from its location on the line of the old Custer trail, is an unique place, and worth the tourists' time to visit, both because it is a typically ideal ranche, and because it is a good place from which to see the "Bad Lands." The ranche buildings, mostly made of logs, constitute a picturesque villa standing upon the plain where Custer Creek enters the Little Mis- souri, and surrounded by an amphitheater of buttes. The proprietors, the Eaton Brothers, are three gentlemen, and tourists will find here everything needed for their convenience for study or recreation. A carload of saddles and riding equipage, and all the things which go to made up the accessories of an ideal ranche headquarters, comfortable quarters, good food, congenial company, in fact, everything except the unpurchasable ability to ride a "broncho," are at the service of guests. A herd of riding horses of all degrees of docility, from the wild and unbridled broncho to the placid "old stager," which is suited to the novice, who may wish to see the Bad Lands, are "rounded up" early each morning from a pasture which is enclosed by twenty-eight miles of ware fence, into a corral built of logs so high and strong that the wildest deer, buffalo, or untamed broncho could neither scale it nor break through it. A few miles farther up the river stands the log "schack" which was once the headquarters and home of President Roosevelt. It has been stated before that the Northern Pacific Railway crosses the Little Missouri River at Medora. The railroad descends from the divide, or watershed, between the Heart and the Little Missouri Rivers at Fryburg, creeping down the steep bottom and between the jagged sides of Sully's Creek. The brakes upon the wheels of the great roll- ing city of parlor cars creak and grind as the train follows the curves of the track down the steep grade of more than fifty feet to the mile. At .length the porter calls out the poetic name of Medora. Step- ping upon the platform of the little station the great nearly perpendicu- lar wall of a large butte meets the gaze, its ribbon-marked side standing like a great curtain 300 feet high behind the town. The top of the butte appears to be perfectly flat, as though the upper part of a great mountain had been sawed off and taken away and this great massive .. . . * /.^^^i&:. ..-.- .* - * . FIG. 103. The Bad Lands, Little Missouri Valley. FIG. 104. "The Palisades," Medora, Billings County. 185 186 THE STORY OF THE PRAIRIES. base left. The ribbon-like marks across the steep side are the hori- zontal layers of the outcropping clay, sandstone, shales, and coal, of which the butte is composed. After crossing the river the railroad suddenly bends northward or down the river, hugging closely against the bank at the foot of the overhanging buttes till the mouth of a coulee is reached, when it winds its laborious way up the steep path of the coulee to the prairie beyond the Bad Lands, that is, to the level of the tops of the buttes at Meclora. A View from the Top If the tourist secures a saddle-horse, and there is nothing else, for there is little use in this country for wheeled vehicles, he may go by a winding course around to the top of a butte. There he finds a level grass-covered prairie, as fine a field for a base- ball ground as college student could desire. Away down there in the valley not away off there but away down there is the muddy Little Missouri, houses, the railroad, the bridge! He is now on the top of the same butte which was first seen from the depot platform. Now he is looking clown the face of the same wall from the upper edge, and not looking against it from near its base. Look off toward the horizon and the scene is that of a great prairie cut up by little grooves or scratches, for. the eye cannot see down into the valleys, and only the edges of the flat tops of the buttes tell where the valleys are. The other buttes are like this one, they are all little segments or blocks of prairie separated by deep and jagged valleys. Looking away across the distant landscape there spreads out over the tops of the buttes the vast prairie, the great Plateau which embraces all of western North Dakota and extends west to the Rocky Moun- tains. But look at the nearer landscape and it is deeply cut up by valleys. Go down into* a valley and the traveler is lost to the world. He sees only the edges of the little prairies which are on the tops of the buttes. It is not the butte tops which are high, it is the valley bot- toms which are low. They have been sunken down into the earth. They are furrows or troughs cut deeply into the bosom of the prairie. Let us look off once more to the far distant horizon. Away to the south rises a huge dark mass higher than the general level. To the southwest is another dark mass, and against the western sky two other great blocks can be seen above the general horizon. These are higher buttes, buttes standing on the shoulders of buttes, as it were. They are so far away that they do not appear so very high upon the horizon, but when we approach nearer to them they are seen to stand 400 to 600 THE BAD LANDS. 187 feet above the surrounding landscape, that is, higher than the tops of the buttes on the shoulders of which they stand. On the Map of the State (Figure i) it will be noticed that there are buttes or hills scattered over the southwestern portion of the State. These are higher buttes standing considerably above all the surround- ing landscape. The Killdeer Mountains, forty miles north of Dickin- son, are high buttes of this class. They are more than 700 feet above the surrounding prairie, their sides steep and ruffled with crags, their top a broad level meadow. These higher buttes, the Killdeer Mountains, Camel's Hump, Sen- tinel Butte, Square Butte T Round Butte, and the many in the south- western corner of the State which are higher than the surrounding landscape, tell an important story of the history of this region. West- ward in Montana are many such high buttes. The story in brief is that the whole vast region extending west to the Rocky Mountains was once lower than it is now, that is, its elevation above sea-level was not as great. The land had been worn away by erosion till there were left only scattered patches of upland. The region had all been reduced to base-level except these few remaining parts. "Base-level" means that the general level of the landscape has been lowered by the streams till it is so little above sea-level that erosion has practically ceased. The high hills, these highest buttes, are, there- fore, vestiges of a former landscape, higher places which were not worn away, just as the Turtle Mountain Plateau was left during the long ages preceding the Glacial Period as a fragment of an older landscape which was nearly all carried away. The general level of the tops of the buttes in the Bad Lands is the old base-leveled plain, such a plain as was the great region of the Mouse Valley, and the central part of the State which is now crossed by the James and Sheyenne Rivers,- before the ice- sheet swept over the landscape. Now, this whole region was uplifted. This is what is called an epeirogenic movement of the crust of the earth. When the uplifting of this region occurred erosion began actively again, and then began to be formed the coulees by which the Bad Lands are dissected into buttes. All the "Bad Lands" along the Little Missouri River, therefore are the result of erosion since the region was uplifted. It is this uplift which gives the steep gradient to the Little Missouri and so makes the deep cutting of its channel and those of its tributaries possible. Thus the Bad Lands are a new feature. They represent the "sec- 188 THE STORY OF THE PRAIRIES. ond childhood," or a beginning of the development of a new land- scape from one which had become old. They have not always been "Bad Lands to Travel Through!" The region was once a great broad prairie lowland. The level meadows which are now left in patches on the tops -of the buttes are fragments of the old base-leveled plain of a former time. It should be said that the "Bad Lands" are not really so bad after all. They are, indeed, "bad to travel through," but it would be diffi- cult to convince the ranchmen who have become wealthy grazing herds of cattle and horses here that they are "b^d." They claim that these lands are better for grazing, area for area, than the smooth and un- broken prairie. The coulee bottoms yield excellent pasturage, for here the grass grows abundantly, and the deep valleys furnish protec- tion for the animals in winter, and the snows which gather in the win- ter protect the grass so that more grows than there are cattle enough to eat. The burning coal mines also have their advantage, for these act as great furnaces warming the air near by them, and the cattle congregate about them in the cold weather to enjoy the warmth. The Petrified Forests Another chapter in the history of the past is revealed in the "Petrified Forests," the remains of which are scattered over the landscape, or still stand as stumps in the places where they grew. Huge logs, looking so much like natural wood as to be easily mistaken for it, occur in great numbers. Many stumps still stand with their "roots" buried in the earth just where they grew. We have seen before that beds of lignite coal occur in the rocks of the region. These were formed from the forests which grew during the times when these rocks were being formed. The "petrified for- ests" are trees which grew upon the landscape but which were not buried under such conditions as to form coal. They have become "petrified," or "stone trees." When a tree dies in the forest, but remains standing, it does not become dry or "seasoned," but takes up water from the ground and becomes "sap-soaked." Such trees dry out by the action of the sun and wind, as do all trees, but they continually take up more water from the earth and so do not become dry or seasoned. The water which was taken up by the trees which become "petrified trees" contained mineral matter in solution. This mineral matter cannot evaporate from the tree with the water and so it is left behind in the pores or cavities of the wood. As this process went on for a long time the tree THE BAD LANDS. 189 trunk, and sometimes the larger limbs also, became slowly filled with the mineral matter. Logs which lay upon the ground or became bu- ried in the soil absorbed water which contained mineral matter, and these became "petrified" also. The wood did not change into stone; this is not what is meant when it is said that the trees became "stone trees" or the log became a "petrified log." The wood decayed and particles of mineral matter were left in place of the wood, and so the tree trunk or log came to be replaced by stone having exactly the form and the structure of the original tree trunk or log. It thus happens that a log or piece of "pet- rified wood" can sometimes hardly be told from actual wood till it is examined closely. It is no joke that travelers on the western plains where no trees now grow, have been deceived by the petrified logs into thinking that they had found fuel ; for such logs, falling to pieces under the action of frost and sun, so closely resemble slivers and pieces of a log of wood that only handling shows them to be stone. When a block of petrified wood has been polished, or when a thin slice is examined with a microscope, the grain of the wood or the cell- structure can be seen just as it was in the original tree. In this way it is possible to tell what kinds of trees grew on the landscape long ages ago. And in the same way the kinds of trees which make up the coal in the coal beds can be found out. CHAPTER THE NINETEENTH. THE COAL BEDS OF NORTH DAKOTA. The Early Landscape. One of the great sources of wealth with which North Dakota has been endowed by Nature lies beneath the surface, and so is not exactly a landscape feature, yet it is so directly related to the landscape, and to the resources which belong to the surface, that it can hardly be omitted from a study of the landscape geology of the State. Its bearing upon the development of the wealth of the soil is so direct that it becomes a part of our subject. This is the great wealth of coal which lies buried beneath the surface of the western half of the State. To understand the formation of the great deposits of coal we need to go back to an earlier chapter in the story of the rock formations of our State, to a time long before the present landscape was formed, and before the landscape which has been called "Old North Dakota," or the pre-glacial landscape, was formed, back to a period whose history is only known to us through the rocks which were then deposited. In fact the "date" of the history we now study goes away back to the great Middle Time of the progress of the North American Continent, and of the World, to a time when a great Inland Sea or arm of the ocean covered nearly half of the continent, and the rocks which are now the shales and sandstones underlying the drift formations were being de- posited. This is the period in the earth's history known as the Creta- ceous Era, the closing part of the great Mesozoic, or Middle Life, Period of the earth's history, known also as the Age of Reptiles. The beginning of the landscape of North Dakota, as of all land- scapes, was beneath the sea. The continents were first sea-bottoms and afterward became the dry land. The rock layers which are passed through in drilling an artesian well are the old mud-floors of the ancient oceans, and the different kinds of rock in these layers and the plant and animal remains they contain tell the history of the time in which they were formed. The Map, Figure 112, shows the portion of North 190 THE COAL BEDS OF NORTH DAKOTA. 191 OS 573 " Burlington 1,590 " Kenmare I >799 " Williston 1,859 " Bismarck 1,668 " Wilton 2,158 " Sims T 958 " Lehigh (near Dickinson) 2,342 " The accompanying Map of the State shows the area where coal has been mined, and where there is not much doubt but that it can be found wherever ?. stream cuts deeply into the rock layers, or wherever a shaft may be sunk. In the Bad Lands In the Bad Lands where the streams have cut deeply into the strata, coal beds are frequently seen in the sides of the buttes. They range in thickness from an inch or less to six or eight feet, or even more. It is a common thing for the ranchmen in this part of the State to have coal mines on their own lands or within short distances of their houses, so that they haul their fuel supply directly from the mines, shoveling it at first hand into wagons, just as in the eastern states farmers go to the woodlands on their own farms for loads of wood. Sometimes a coal bed is cut across by a small stream on the bank of which stands the house, so that coal is brought directly from the mine in the coal-pail and put into the stove! The writer has stopped at a ranche for dinner while traveling in this part of the State, and when fuel was wanted for the kitchen stove a small boy was de- spatched to the coal mine in the back yard to get the coal! It is not a joke that in digging a cellar for a house a coal bed may be dug into only a little below the surface, so that in the winter the owner of the house may go to the coal mine after a scuttle of coal without even going out of his own house! A point of advantage the western farmer has over his eastern cousin lies in the fact that in the west the fuel comes from a forest which lived and flourished thousands of years ago, and the land at the surface, over the coal bed, may be cultivated, or used for grazing, while at the same time the coal forest underneath furnishes the supply of fuel. But in the east the woodland occupies a special preserve so that the land can- not be used for farming purposes! CHAPTER THE TWENTIETH. THE BEGINNINGS OF NORTH DAKOTA. The Sea Bottom on Which the Rocks Were Deposited The great In- land Sea in which the rock formations of North Dakota were laid down as sediments extended from eastern Minnesota over North Dakota and Montana to Idaho and Washington, and south to northern Texas. FIG. in Generalized Section across Northern Portion of North Dakota, showing the Formations. From a Crayon Drawing by Miss Bessie M. Willis and the Author. Re-drawn by Prof. Thomas H. Grosvenor. Into this great sea were borne the sediments from the surrounding land areas, and the waves of the great shallow sea beat upon the shores and eroded the rocks into sand and mud and distributed them over its bottom, forming the rocks which now make up the sandstones and shales of the Cretaceous series, or system. North Dakota was then all under water. The Cretaceous system, or series of rocks, is divided into Lower THE BEGINNINGS OF NORTH DAKOTA. 203 and Upper, the basis of this separation being the different conditions of the sea bottom during the earlier and later times of the Cretaceous Era. The Lower Cretaceous rocks do not, so far as we know, occur in North Dakota. The division into Lower and Upper is, therefore, made from the rocks in other states. The Upper Cretaceous, or what will here be called simply the "Cretaceous" series of rocks, is subdivided into several formations, each distinguished by certain characteristics which separate it from the others. The lowest of these formations, the Dakota Sandstone, is at the bottom of the series, so far as we have got down to the "bottom" in North Dakota. The other formations follow in the order in which they were deposited from below up, each formation being described as "shale" or "sandstone," etc., according to the kind of rock most common in that formation. A "shale" forma- tion often contains some sandstone, however, and a "sandstone" forma- tion often has layers of shale. Clays occur also in nearly all the formations. The thickness so far as it is known is given for each formatioi The Geological Formations. Thickness. 6. Laramie Sandstone, Shale, and Clay, with Lignite Coal 1,000 feet. 5. Fox Hills Sandstone 100 " 4. Fort Pierre Shale, with Beds of Clay 600 " 3. Niobrara Shale, Calcareous (Lime) 150-200 " 2. Fort Benton Shale 200 " i. Dakota Sandstone, with Lignite Beds 600 " The Fort Benton and Niobrara formations are together called the Colorado formation, and the Fort Pierre and Fox Hills formations are together called the Montana formation, in the western states, but in North Dakota it seems more convenient to use the names and divisions here given. The total thickness of all the Cretaceous series in North Dakota is thus seen to be nearly 3,000 feet. These formations, however, are thinner to\vard the east. The artesian wells at Devils Lake and James- town passed through about i ,400 feet from the upper layers of the Fort Pierre Shale to the Dakota Sandstone. When, in speaking of the rocks which come to the surface, or out- crop, at any place, any one of these names is given to the rocks, it shows in what part of the Cretaceous series it belongs, and hence whether it is older or more recent than some other of the series. The 204 THE STORY OF THE PRAIRIES. lowest was deposited first, and, therefore, is the oldest, and so on up through the series. The lowest and oldest, the Dakota Sandstone, and the highest and most recent, the Laramie Sandstone, are fresh-water or brackish formations, that is, they were deposited as sediments either in ponds and pools of fresh water, of else upon the bottom of a very shallow sea in which the water was only slightly salt, or brackish. The land now embraced in North Dakota was slowly sinking, and the sea was creeping upon the land when the Dakota Sandstone was being formed. The land was rising, and the sea was drying off from the bottom when the Laramie rocks were deposited and the great forests grew which formed the coal beds. The other formations, those formed after the Dakota Sandstone and before the Laramie, are marine or sea forma- tions deposited when the whole region of North Dakota was a sea bottom. It is to the fact that the rocks of the Fort Benton, Niobrara, Fort Pierre, and Fox Hills formations are salt sea sediments that the \vater of the lakes and streams of a large part of the State contain so much salt and alkali. The salt and alkaline substances were in the sea water, and so, as the sediments were deposited, they were saturated with salt and alkali water, and when the sea dried off from the mud and sand of the bottom, and these became the shales and sandstones of these forma- tions, they contained the salts and alkalies which now dissolve out into the waters of the lakes and streams. The highland which formed the western shore of Lake Agassiz, extending from the Pembina Mountain on the north to the Coteau des Prairies on the south, called the Manitoba Escarpment, is an outcrop- ping of the edges of the horizontal layers, mostly of the Fort Pierre formation. This outcropping was caused by the erosion of the great pre-glacial valley in which now lie the level prairies of the bottom of Lake Agassiz. We have seen how this great valley was filled with the ice of the Great Ice-Sheet, and how as the ice melted this basin came to be filled with \vater because the course of the river to the northward was blocked by the ice, and Lake Agassiz came to occupy the great valley, its western shore being ^the escarpment, or cut off edges, of the Fort Pierre, Niobrara, and Fort Benton formations. We are now study- ing a much earlier period, when the rocks were deposited in which the valley was afterward cut. THE BEGINNINGS OF NORTH DAKOTA. The "Manitoba Escarpment." The great Inland Sea during the Cre- taceous Era spread over all of North Dakota and a large part of Minne- sota, although all of Minnesota, and probably a little of the eastern edge of North Dakota, had before been raised above sea-level so that it had been dry land. But the Dakota Sandstone was deposited over a large part of western Minnesota, showing that the sea not only North.A7n.encd in tKe Cretaceous $rd. Areas Covered by FIG. 112. After Dana. covered North Dakota and the states west to where the Rocky Moun- tains are now, but extended east, covering much of Minnesota. So the sea-bottom formations, the Fort Benton, Niobrara, Fort Pierre, and probably the Fox Hills, were deposited over all of North Dakota and western Minnesota, but during the long period following the Cre- taceous Era, known as the Tertiary Era, and before the time of the 206 THE STORY OF THE PRAIRIES. Glacial Period, the great valley of the Old Red River of the North was eroded, carrying away the sediments which had been deposited over eastern North Dakota and western Minnesota, so that the outcrop- ping edges of these formations now occur along the west side of the Red River Valley. ' The strata, or layers, which are at the top of this highland under- lying the drift in its northern and higher portion, the Pembina Moun- tain, and extending south more than half way across the State, and also the outcropping edges along the northern half of the highland, are Fort Pierre shales. The Niobrara and Fort Benton formations outcrop lower down on the old valley wall, but they are deeply buried by the drift so that we do not readily see them. About ten miles east of Lisbon, be- low the Big Bend, just after the Sheyenne River enters upon the plain of the Lake Agassiz bottom, this river has cut a deep gorge in the Fort Benton shale. This formation is also penetrated in drilling artesian wells in the southeastern part of the State, lying beneath the drift. The Dakota Sandstone forms the floor of the old Valley beneath the great depth of drift in the part of its course lying between Grand Forks and Larimore and southward to Casselton and Fargo, though patches of shale, which are probably Fort Benton, were struck by arte- sian wells at Fargo and Mayville. These probably represent the tops of higher places or low hills on the old (or pre-glacial) valley bottom. Farther south in the higher part of the old valley the floor of the valley is probably the Fort Benton shale. This shale is struck by artesian wells in the vicinity of Wahpeton. In the lower (northern) portion of the valley the floor is older rock than the Dakota Sandstone, the arte- sian well at Grafton passing through the drift into limestone belonging to the Lower Silurian, which is much older than the Cretaceous. The section through the formations of the northern part of the State (Figure 74) will make this more clear. West of the Manitoba Escarpment, in the central portion of the State, the eroded surface of the Fo-rt Pierre and Fox Hills formations underlie the drift. The Sheyenne, James, and Mouse Rivers have cut down their channels in many places so that the strata of these forma- tions have been cut into. The deep valley of the Sheyenne River has cut into the Fort Pierre Shale through much of its course from Devils Lake south to the Big Bend east of Lisbon, and a large amount of shale was added to the Sheyenne Delta, eroded along the course of this val- ley during the time of the glacial flood waters. The Valley of the THE BEGINNINGS OF NORTH DAKOTA. 207 James is not nearly as deep, and is cut through much of its course in North Dakota in the Fox Hills Sandstone. The Fox Hills Sandstone extends east underneath the drift prob- ably nearly to Devils Lake. From the fact of this sandstone being the surface rock from the vicinity of the Turtle Mountains south across the State, comes the sandy character of the drift hills, and the tracts of sand dunes along the eastern side of the old Lake Souris bottom, the soft sand-rock being easily ploughed up by the moving ice-sheet, and dumped in the lake by the melting of the ice. The Missouri Plateau Farther west rises abruptly the great hill- country known as the Plateau du Coteau du Missouri, or the Plateau of the Missouri Hills. This highland is composed of Laramie rock- strata, and the sudden rise from the lower land of the James and Shey- enne Valleys of 300 to 400 feet is due to the erosion of the eastward continuation of these rocks, just as the Fort Pierre, and the formations below it, in the eastern part of the State, were eroded by the pre-glacial Red River of the North, forming the Manitoba Escarpment. The Turtle Mountains, on the International Boundary about mid- way between the Coteau du Missouri and Pembina Mountain, is a plateau of Laramie strata, surrounded on all sides by great wide-spread- ing prairies, the old valley bottoms of the rivers which eroded the inte- rior portion of the State, and carried away the upper part of the Fox Hills and Fort Pierre formations, in the region east of these moun- tains, and the Laramie strata west to the highland of the Coteau du Missouri. Thus the Turtle Mountain Plateau is a fragment of the great Missouri Plateau which was not carried away by the erosion- which lowered the whole country round about it. Dog Den Butte, the Mauvais or Big Butte, south of Church's Ferry and Leeds, and probably Devils Heart and Sully's Hill south of Devils Lake, are fragments of the Laramie strata of the great Missouri Plateau which have not been entirely carried away by erosion. All the great plateau country to the westward is Laramie. The Bad Lands along the Little Missouri River, and the Yellowstone in Montana, are Laramie rocks, made up of sandstones, shales, and clays, with beds of coal and lava. This great upper part of the Cretaceous series or system of rocks extends westward to the Rocky Mountains. It extended once much farther east than now also, and it probably cov- ered all the State. At least it reached farther east than the Turtle Mountains, for the form of this plateau shows that the rock layers once 208 THE STORY OF THE PRAIRIES. & 66V : C2L L , I 150 fetf fir far $tr*1'if't Chy, {fydnyin to 71/1 5 flow. Ml fett DRIFT ftfaf 4 f feet sfafe Mferf Red Jejti(f~ JfCfff LOWER BILUmH UPPER CAMBRIA ARCH EM FlG. 113. Section showing the Rock Formations passed through by the Artesian Well at Grafton. After Upham. THE BEGINNINGS OF NORTH DAKOTA. 209 extended farther east. This broad valley between the Turtle Moun- tains and the Missouri Plateau was eroded during the same time that the Old Red River Valley was being formed farther east. The Older Rocks Underlying the Eastern Portion of the State, Below the Dakota Sandstone in the Red River Valley are still older rock formations. The Jura-Trias, the Carboniferous, the Devonian, the Silurian, the Cambrian, and finally the oldest of all, and the oldest in the world, the Archaean, lie one below another under the rocks of the State, and their thinner eastern edges extend along the eastern portion of the State. These are shown by borings for artesian wells. An arte- sian well at Graf ton, 915 feet deep, after passing through nearly 300 feet of drift penetrates several older formations, into the granite at the bottom, which may indeed be called "the bottom," for it was the first formed and hence the oldest of all the solid rocks of the earth. This oldest Archaean granite comes to the surface in Minnesota about Lake Superior, and northward in Canada. It was the old, first beginning of the Continent, being at first an island raised above the sea. Other formations lie all around it and flank or lap upon its sides. The ice of the Great Ice-Sheet ploughed its way across it and broke off huge masses, which are the "hard-head" boulders now scattered over the prairies. West of Winnipeg in Canada the Silurian, which is a limestone for- mation, is the surface rock, and from it were broken off and carried away limestone fragments by the great ice-plow, and these were ground up to make the fertile wheat lands of our State. Many of them are scattered over the prairies, not having been entirely ground up by the ice-mill. Boulders of the softer and more easily crumbling shales and sandstones were soon broken and pulverized into clay or ground into sand. How Old Is North Dakota ? It is natural to ask how long ago it was that the great Inland Sea covered North Dakota, and how long it has been since the forests grew which have become the coal beds. It is a fair enough question, and one which any thoughtful person is bound to ask, in his mind at least. But it is one which the most learned scien- tist cannot answer with accuracy, as time is measured in years. We do not know how long it is since civilization began upon the earth because we have not a written record from the beginning. We can only infer from the marks left in buildings and implements and other things which show man's handiwork. So we can only infer from the great handi- 210 THE STORY OF THE PRAIRIES. work of Nature how long the time has been that geologic processes have been fashioning the -earth. We do not know how long the time has been since the Glacial Period, or Ice Age, but we know that it is only a little while as compared with the time since the coal beds of North Dakota were formed. Many attempts have been made to get a basis of comparison by which the time since the Ice Age could be measured in years, but no conclusion which can be considered as fact has been reached. Nothing more than estimates can be said to have been made. A method of studying the problem is this: The gorge of the Mississippi River from Fort Snelling to the Falls of St. Anthony has been formed since the closing stages of the Ice Age. This is known because the River was forced out of its old channel by the drift which filled its valley, and when the river re-entered its old channel at Fort Snelling a "falls" was formed. The falls have been "moving back," by cutting the rock ledge over which the water passes, ever since that time. The gorge at Niagara Falls, New York, has been formed in a similar manner, the gorge having been cut back from Lewiston to the present cataract. Now it would seem a simple matter to see how far the falls cut back in one year, and then by measuring the length of the gorge (from Fort Snelling to the falls at Minneapolis, or from Lewiston to the Niagara cataract), divide this distance by the amount of cutting in one year. This would give the time in years since the close of the Ice Age. But the problem is not as simple as it may at first appear. Geologists have reached estimates ranging from 6,000 to 10,000 years (Upham) to more than 30,000 years (Gilbert). So that the result at best is only an esti- mate. But suppose we assume a rather low estimate of 10,000 years for the time since the close of the Ice Age, then how long has it been since the coal beds were formed during the closing stages of the Cretaceous era? How long was North Dakota under the sea after the Dakota Sandstone had been deposited, while the salt and alkaline sediments, which now make up the shales and sandstones of the Fort Benton, Nio- brara, Fort Pierre and Fox Hills formations, were being deposited? Attempts have been made to estimate the length of geologic periods by measurements of the rate of accumulation of sediments on the sea bottom at the present time, but these estimates are quite as variable as those of the time required for the cutting of the gorges referred to. Without considering the methods of computing by which the estimates THE BEGINNINGS OF NORTH DAKOTA. 211 have been made we may think of the time of the Ice Age, that is, the length of time that the cold of the Glacial Period continued, as five to ten times as long as the time since the ice finally melted, or the time dur- ing which the gorge of the Mississippi River below Minneapolis, and the Niagara gorge from the Falls to Lewiston, were being cut, or 50,000 to 100,000 years (Upham, Prestwich). The time since the formation of the coal beds in North Dakota would be from fifteen to twenty-five or thirty times as long as that which has passed since the beginning of the Glacial Period, or nearly 300 times as long as the time since the ice finally melted away and Lake Agassiz began to drain toward the north and the present Red River Valley began to appear as dry land.* This would make the age of the shales and sandstones and coal beds of North Dakota nearly 3,000,000 years, and the time during which the salt-sea sediments which occur between the Dakota Sandstone and the Coal Measures, the Fort Ben- ton, Niobrara, Fort Pierre and Fox Hills formations were being formed, may have been 1,000,000 years.f Of course, no one knows how long it has been. These figures are only estimates, but they will at least serve as a suggestion that time is long. They should not be taken by the reader as settled facts, for they are not. But that geologic time is im- mensely long as compared with human standards of years we may safely admit. Perhaps a better idea of the great length of geologic time may be gained from this, that the greater part of the Rocky Mountain region was under the sea during the Cretaceous era and perhaps till after the depositing of the rock strata of the Laramie formation with its coal beds in North Dakota. In fact, it is likely that the great uplift by which the Laramie rocks in North Dakota were raised so that the region became dry land was a part of the beginning of the great move- ment by which the Rocky Mountains were heaved up. And since the Laramie strata were deposited and the coal beds were buried the region of the Colorado Canons has been elevated from 10,000 to 11,000 feet, and erosion has cut down 10,000 feet (Dutton). And in British Col- umbia it is estimated that an elevation of 32,000 to 35,000 feet has taken place since Cretaceous time, and canons 5,000 to 6,000 feet deep have been eroded (G. M. Dawson). * Based on Walcott's estimate of the length of Caenozoic Time, and Upham's esti- mate of the length of Glacial and Post-Glacial Time. f Based on Walcott's estimate of 27,240,000 years for the whole of Mesozoic Time. 212 CHAPTER THE TWENTY-FIRST. THE COTEAUS OF THE MISSOURI. What the Coteaus Are A great region lying east of the Missouri River was called by the early French explorers "Les Coteaux du Plateau du Missouri," or The Hills of the Missouri Plateau. In. the popular mind the hills or "coteaus" and the plateau are often confused, the term "coteaus" being applied to the great hilly upland, which is really the plateau, while the coteaus, or hills, are a surface feature of the plateau. A glance at the accompanying diagram (Figure 114) will assist in making this relation clear. It will be seen that the plateau is of immense size as compared with the largest hills of the "coteaus." The plateau is made up largely of layers or strata of sandstone or shale rock, these layers or strata being in nearly horizontal position, as they were laid down upon the bottom of the ancient sea. The sloping front of the plateau is simply the place where these rock layers come to an end. These strata of rock once extended farther east, we do not know how far, but probably a good many miles. The great sea bottom upon which they were deposited probably extended over nearly or quite all of North Dakota. The great basin known as the Mouse River Valley and also the basin of James River, were formed by erosion, or the wearing away of the rocks by the action of streams and the weathering processes, long after the sea had disappeared and the sands and muds of the ancient bottom had become dry land. A great uplift of the crust of the earth such as that which made the sea bottom dry land, raised the land high enough so that it is a plateau, an elevated plain instead of simply a plain. When the front of the plateau is spoken of, by' this is meant the cut- off edges of the strata of the eastern part of the plateau. Crossing the State of North Dakota from the northwest corner south and east to about the middle point of the southern boundary is the edge of the plateau. The rock layers of which the plateau is composed might pro- ject out, or come to the surface in an outcropping of rock, only that a great mantle of drift covers the whole region. The coteaus are the hills of a terminal moraine, and these are on the plateau, but they are 213 214 THE STORY OF THE PRAIRIES. no part of the plateau proper. Look now at the diagram (Figure 114) and recall what a terminal moraine is, and if need be re-read Chapter Five, and it will be clear what is meant by "the coteaus" as distinguished from the Missouri Plateau. "Les Coteaux du Plateau du Missouri" may be briefly and for con- venience styled the coteaus of the Missouri, or the Missouri coteaus: Les Coteaux des Prairies, commonly called the Coteau des Prairies, should be carefully distinguished from Les Coteaux du Missouri, or the Missouri Coteaus. The former is another and quite different feature of the landscape of the States of North and South Dakota. The Coteau des Prairies is an immense hill many times larger than any one of the Missouri coteaus, though not as large as the Missouri Plateau. It lies mostly in northeastern South Dakota, but extends across the boundary into North Dakota in Sargent County. The Coteau des Prairies is a large preglacial hill, having its surface covered with a mantle of drift. Some of the drift is in the form of morainic hills much like the coteaus of the Missouri, only not generally as large. The Coteaus of the Mis- souri are morainic hills. There are also morainic hills on the Coteau des Prairies. Le Plateau du Missouri, or the Missouri Plateau, as has been stated, is a vast upland of preglacial origin. The eastern edge of this upland extends across North Dakota in a generally northwest and southeast direction. The front rises quite abruptly from the plain to the east from 300 to 400 feet. This steep slope or front appears west of Ellendale, Edgeley, Jamestown, Carrington, Fessenden, Minot and P'ortal, distant from these places twenty to thirty miles. Lying on the top of the plateau is a great belt or tract of hills, drift hills, formed by the action of the great ice sheet, which together make up what is known as a moraine. This moraine in North Dakota is a portion of the great continental moraine which was formed during what is known as the Wisconsin stage of the great ice age. This moraine extends across the continent from the Canadian northwest ter- ritories to the Atlantic ocean. The moraine in North Dakota is no more a part of the great Missouri plateau than is this same moraine in Penn- sylvania a part of the Allegheny Plateau. The moraine is a deposit of earth materials stones, clay, sand and soil ploughed up and trans- ported by the great moving ice sheet, and left in heaps and piles or spread out as rolling prairie, to a depth of thirty to 100 or even 150 feet. THE COTEAUS OF THE MISSOURI 215 216 THE STORY OF THE PRAIRIES. The continental moraine in North Dakota lies in such relation to the plateau that it suggests that something more than accident caused the moraine to lie just upon the edge of the plateau through a distance of 300 miles in North Dakota. The front or edge of the plateau, it has been stated before, extends across the state in a northwest-southeast direction. The moraine also extends across the state, lying almost par- allel to the edge of the plateau, and nowhere more than a few miles back from the slope which marks the edge of the plateau. Often in fact the coteaus or hills of the moraine are encountered immediately upon entering upon the higher lands of the plateau top The direction of movement of the ice in this part of North America was probably nearly at right angles to the front of the great plateau so that when the great ice sheet, in its onward course toward the south and west, flowed against the edge of the plateau, which, as has been stated, was 300 to 400 feet high, this, acting as a great wall or barrier, served as a dam to hold back the ice. Thus it came about that the moraine occurs along the edge of the plateau, because the ice could not advance beyond this position. It will be remembered that a terminal moraine is formed when the edge of the ice is stationary, that is, when the conditions are such that the ice melts at the margin as fast as the general mass moves onward. Because of the fact that the great plateau existed in the western half of North Dakota, the southwestern portion of the state was not passed over by the great ice sheet. It is because of the occurrence of the moraine upon the edge of the plateau that so much confusion has arisen regarding the true nature of the coteaus. The term "coteaus" has been applied to the hills in this region. The altitude of the plateau above the prairie to the eastward has easily made this seem a part of "the hills," whereas the coteaus are hills on the top of the plateau and entirely different in their origin. The extent of territory embraced by this great moraine within the vState of North Dakota is probably approximately 7,000 square miles. This region was for many years mostly a grazing range, native grasses adapted during the ages to the soil conditions of such a landscape grow- ing in abundance, and eaten by the herds of cattle and horses which during the last thirty to forty years have succeeded the herds of buf- falo and antelope that formerly roamed and grazed on these lands. The agricultural value of the lands was an unknown factor. The ranchman THE COTEAUS OF THE MISSOURI. 217 was the only settler. Little was known of the character of the lands, and little question was asked by homeseekers about these lands because there were other lands open to homestead entry that were thought to be more desirable. Within the last few years the desire for free homestead lands has led to the settlement of these lands by farmers. The region was long regarded as adapted only to grazing, and no attempt at general farm- ing was made until quite recent years. The settlement of even a small portion of the land by farmers overthrows the large ranching enter- prises where cattle, horses and sheep in great herds wandered at will over a range unbounded by fences. An Unique Part of North Dakota. The region known as "the coteaus" is unlike any other part of North Dakota. There are other morainic lands in the State, but none that can vie with this region in rugged character, in abundance of the number of sloughs and lakes, and in the "everlasting monotony." It is not like the rugged hill-land west of the Missouri River, for there the hills tend to have flat tops, and the land is nearly all drained by streams and is often dissected by deep coulees. The hills west of the Missouri River are for the most part hills of erosion, and crags of sandstone and hard layers of rock project in shelves from the tops and sides of the butte-like hills. Among the coteaus streams are unknown. The hills are all rounded in form, and never flat on their tops. Rock ledges or shelf rock never project from the sides or tops of these hills, and their tops and sides are often strewn with boulders of granite and other rocks unlike any that are native to this region. This region is one that marks the halting-place of the great con- tinental ice sheet in its passage across the northern portion of North America, and here was deposited the great mass of morainic material rocks, sand, gravel, clay and soil ; huge boulders so hard that they would phase the hardest stone-cutter's chisel and weighing many tons side by side with small rounded pebbles and sand grains; masses of clay and soil piled in heaps; hollows filled with water or grown up with reeds and .rushes. It is truly the region of hills coteaus. They are the coteaus of the Missouri Plateau because they are on the plateau. They are the high- est and most rugged of morainic hills in the state because they were 218 THE STORY OF THE PRAIRIES. formed at the edge of the great ice sheet at a stage when the edge remained stationary for a longer time than at any other stage. The Region Described The region to which this chapter refers embraces a somewhat indefinitely limited tract which crosses the State from the northwest corner east and south to the State boundary in Emmons and Mclntosh Counties. The tract extends eastward from the Missouri River, and varies from forty to sixty miles in width, and erribraces the greater part of Williams County, about one-half of Ward, nearly all of McLean, a portion of Wells, most of Burleigh, all of Kid- der, the western half of Stutsman, Emmons, Logan and Mclntosh, and the western portions of LaMoure and Dickey Counties. In this region are embraced three types of landscape: (a) The eastern slope of the edge of the Missouri Plateau, a region which was passed over by the ice, but which is not marked by moraines; (b) a region of rugged morainic hills closely set with lakes and sloughs, the coteaus; and (c) a region marked by broad and deep channels with little water, known as the "Missouri Slope," which lies outside, or west of the great moraine or coteaus, a region over which the watery from the melting ice of the great glacier passed on their way to the Missouri River. The first of these regions, the plateau front the region that would show outcropping ledges of rock if it were not covered with the drift of ground moraine is dissected by coulees. It has no lakes or sloughs, or exceedingly few. The coulees are deep, due to the fall from the high plateau region toward the Mouse and Des Lacs Valleys. This region is comparatively well drained. The highland of the Missouri Plateau rises distinctly upon the hori- zon in the west, and confronts the eye boldly from the vicinity of Ellen- dale, Edgeley, the prairie outside the valley at Jamestown, from nearly any point on the Jamestown Northern Branch of the Northern Pacific from a few miles north of Jamestown to Carrington, and from the car window nearly all the way along the Soo Line from Carrington to Por- tal, except in the vicinity of Minot, where the deep valleys of the Mouse and Des Lacs Rivers cut off the view. The second, or morainic, type of landscape, the coteaus, embraces an irregular belt generally from ten to twenty or thirty miles in width, lying west of the plateau front. Here will be observed an utter absence of streams or coulees, but many lakes and sloughs. This broad but irregu- THE COTEAUS OF THE MISSOURI. 219 lar tract of hilly land divides the western portion of North Dakota from the eastern by a natural division. This hilly region forms the divide or watershed that parts the drainage of the continent between Hudson's Bay and the' Gulf of Mexico in the northern part of the State, and that between the Missouri and the James Rivers in the southern part. In this belt, however, there is no drainage whatever. The hills and hollows are scattered in confusion. Many of the hollows contain water, and are therefore lakes. Others are hay-sloughs only. The lakes often occur at different levels within short distances, yet with no drainage from one to the other. There are no streams among the coteaus. East of this FIG. 116. A Small Lake in Morainic District. hilly region the land is drained, the waters ultimately becoming a part of the Hudson Bay drainage. The streams that flow westward from the coteaus are tributaries of the Missouri River, and so> finally discharge into* the Gulf of Mexico. The third type is that of the "Slope" region. This is the region that lies between the coteaus, or the great moraine, and the Missouri River, and was, therefore, crossed by the waters that came from the melting of the great ice sheet during the time that the coteaus were being formed. It is the eastern drainage area of the Missouri River now as it was in the time w r hen the ice waters were seeking escape to the river. 220 THE STORY OF THE PRAIRIES. The Eastern Slope It will be observed that there are but few streams having their heads in the coteaus and flowing eastward. What- ever stream valleys may have once been there have been filled with drift so that they no longer appear. East of the plateau front are the large valleys of the Des Lacs, Mouse, Sheyenne, James and Pipestem Rivers. These show the tremendous work of erosion that was accomplished by the waters from the melting ice, for their valleys are broad and deep beyond all comparison with the rivers now occupying them. They are glacial channels cut by the flood waters from the melting ice. They have been eroded 100 to 200 feet into the plain that lies east of the great Missouri Plateau. Their bottoms are often below the lowest portions of the mantle of drift, as is shown by the occurrence of shale beds, sand- stone ledges, and lignite coal seams in their banks. The plain which is crossed by these channels is the "rolling prairie" for which the state is- noted, and which makes North Dakota fittingly called a "prairie State." To the west of the heads of these streams is the sloping front or edge of the great Missouri Plateau. While the prairie adjacent to the deep valleys of these streams is from 100 to 200 feet or more above the bottom lands along the immediate stream beds, still the plateau top 30 miles west is 300 to 400 feet higher than the plain of the prairie. FIG. 117. Looking Along a Stony Ridge. In the Coteaus. THE COTEAUS OF THE MISSOURI. 221 It is this slope from the edge of the plateau toward the deep valleys of the Mouse and Des Lacs that gives the dissected character to the plain bordering the plateau in Ward County. A rise of nearly 700 feet in a distance of forty miles in the railroad grade of the Great Northern Railway from Minot westward has made possible the erosion of the deep and narrow V-shaped coulees which characterize the region. Description of the Moraine The extremely hilly tract of this region, the coteaus, varies from 10 to 30 or 40 miles in width, including sometimes mter-morainic tracts that are comparatively level. The hilly iregion is exceedingly irregular in outline. The term "coteau" means a hill, and the hills are the most conspicuous thing about this region. But FIG. 118. A Stony Ridge. In the Coteaus. hills are not the only feature about a terminal moraine by which it may be distinguished. This morainic tract is made up of hills, ridges, roll- ing or even gently undulating- prairie, lakes, sloughs, and hay meadows. The edge of the moraine is also very irregular, having long- projecting lobes and being 1 indented by deep sinuses of comparatively level land. A moraine is thus seen to be a quite complex thing. The general aspect of the landscape after entering! the morainic re- gion is distinctly hilly. Many of the hills are high and their sides very steep and rugged. Often they are so closely set together that there is 15 222 THE STORY OF THE PRAIRIES. no space between the bases of the hills, but the bottom of one merges into its nearest neighbors. Occasionally a hill is so decked with stones large and small that the face of the hill appears like a vast stone heap. A few years ago, before settlers had occupied the land, travel through the hills to one unaccustomed to the region was almost impossible, not only because of the roughness of the landscape and the frequent large FIG. 119. A Landmark. Such cairns are common on the tops of the highest hills, placed there for the guidance of travelers through the coteaus. rocks, but the hills all have such a resemblance that the inexperienced traveler easily mistakes the hill he thinks he is traveling toward, another insidiously substituting itself for the one he started to reach, while the traveler unconsciously changes his course to go around a hill' or avoid a slough. Where definite trails do not serve as a guide to the traveler he is almost helpless, and a journey is well nigh impossible. THE C.OTEAUS OF THE MISSOURI. 223 The hills are sometimes so stony and steep that progress even on horseback by one who is not accustomed to "the range" is almost im- possible. Following a well worn trail one can often see his way but a few rods ahead, so crooked is the way among the hills. Leave the beaten path but a few steps and the unaccustomed traveler is as one adrift on the rolling sea. One may be lost and pass within a few rods of a ranchman's shack and not see it, for the shack may be and often is located in a hollow between the hills so that it cannot often be seen, even from a short distance. The writer speaks from experience in seeking to find a ranch house while traveling a stranger and alone in this solitary region. A miss of a single dim fork in the trail caused him to pass by the last FIG. 120. Douglas Valley, South of Douglas Postofflce, McLean County. house in many miles, and as a result he lay down fatigued to the point of exhaustion upon the hard bosom of Mother Earth, and slept with the picket rope by which his saddle pony was held tied around his body till the cold of the small hours of the morning compelled him to travel on eagerly looking for the dawn which should enable him to find food and what was more intensely needed, water. A common custom is to place upon the highest points of the highest hills piles of rocks with often a pole supported in the midst of the rocks as a guide to the traveler. Such a landmark is always known to the 224 THE STORY OF THE PRAIRIES. inhabitants of the country, and if lost in fog or storm- and one of these marks is seen it will indicate quite as accurately as section corners in the agricultural portions of the state where a ranch house is located. The Missouri Slope The portion of region known as the "Mis- souri Slope" included in this chapter embraces southern Williams, south- western Ward, western McLean, Burleigh, and Emmons Counties. This region is drained into the Missouri from the east. The region is marked by deep and broad valleys having extensive gravelly flood- plains upon their bottoms. It is a region over which the ice did not pass during the Wisconsin stage of glaciation, the stage dur- ing which the coteaus were formed, but over which vast floods of water passed, outwash from the melting of the great ice sheet. It was by these floods of ice water that the large channels were formed and extensive deposits of gravel were made. The valleys of Little Muddy Creek in Williams County, White Earth, Little Knife, and Shell Creeks in Ward County, Douglas, Snake, and Painted Woods Creeks in Mc- Lean County, Apple Creek in Burleigh, and Beaver Creek in Emmons County, are examples of broad and deep valleys, generally with large flood-plains of sand and gravel, that were formed by the waters from the melting ice during, or immediately following, the Wisconsin stage of glaciation. The Older and the Newer Drift. The great moraine which is known as the coteaus, and which, as has been stated, was formed at the edge of the great ice sheet when it had advanced so far south and west as to lie upon the eastern portion of the great Missouri Plateau, repre- sents the limits reached by the great flood of ice during this stage of the Glacial Period. But there is drift farther west than the coteaus. The drift of the overwash region west of the coteaus does not represent the most western deposits of drift materials. There are deposits of drift soil, sand, gravel and large boulders far beyond the Missouri River. Soil that is thought to be of glacial origin, pebbles that show by their form and composition that they are drift pebbles, and boulders of gran- ite which are entirely unlike any rocks of which the hills are composed, occur over large areas in Morton, Oliver, Mercer, Dunn, and McKenzie Counties, How then is this to be explained? Does the great range of hills that has been described really represent the halting-place of the edge of the ice of the great continental ice sheet ? Yes, the coteaus repre- sent the great terminal moraine formed at the edge of the ice at the THE COTEAUS OF THE MISSOURI. 225 time of the greatest advance of the ice sheet at this stage of the Glacial Period. But the Glacial Period was probably very long, and it was made up of several stages, each stage representing a long time. It is as though a great battle between heat and cold had been going on through long ages. When the cold gained the mastery and the snow and ice did not melt as fast as they accumulated then the ice sheet grew larger and deeper, and spread out over more of the land. Then in turn a warmer condition of climate might have occurred causing the snow and ice to melt more rapidly than they gathered, and thus the amount of ice composing the great ice sheet would grow less. Far from the edges of the great sheet the ice might become less deep from melting, and great streams of water probably ran off from the ice or down through cracks or crevasses to the ground, and probably may have formed streams under the ice. At the edges of the great sheet the melting ice would cause streams of ice water to form and flow away. If there continued to be more melting than there was snow-fall then the mass of the ice would tend to grow smaller, and the edge of the ice would retreat back toward the center of accumulation or point from which the ice came. Now, if these conditions of advance and retreat continued, each for a long time, these would constitute two stages of the great Glacial Period. The time when the coteaus were formed at the edge of the great ice sheet, a time that represents a stage of advance, is known as the Wisconsin stage of the Glacial Period. The time following this stage when melting was more rapid than the onward movement, so that the edge "retreated" and the mass of the ice of the great ice sheet grew less, is known as an interval or stage of deglaciation or melting. Prob- ably the climate was warmer during this stage, and the time was probably very long, as we measure time in centuries. How long it was, or how warm the climate, or what caused the changes, we do not know. It is the facts that concern us now rather than the causes, and we may leave the question of causes till another time. We may not live to learn the nature of the causes, though these may be ascertained sometime. There still remains unexplained the drift west of the Missouri River, and all that drift west of the coteaus which was not washed down (Jirect- iy from the edge of the ice or from the coteaus themselves after their deposition. When was the drift deposited here? We say deposited, for it is evident from the character of the material the soil, sand, 226 THE STORY OF THE PRAIRIES. gravel, and boulders, that it was not formed here, but has been trans- ported here and deposited. Was there an earlier stage of glaciation than the Wisconsin? Did the ice of an earlier stage push farther west, even beyond where the great Missouri River now is? Yes, this is what is supposed to have occurred. Let us be reminded again that we are not asking about causes now, but about facts. Did the ice once extend fifty or a hundred miles west of the great moraine which has been described? No man was there to write a history, or if man did exist on the earth at that time (and there are geologists who think that he did) he did not leave any record of his observations that have since been discovered. We can therefore only know of the ice having been farther west than the Missouri River by the character of the things we can find in the field. This leads often to long and painstaking investigations to determine. But we may take the results of many studies made at different times by different -ob- servers, going into the field ourselves to verify our conclusions. The peculiar soil, the pebbles, and gravel, the granite boulders, the rocks with marks on them such as have been recognized in all "glaciated" regions all tell of something having transported a large amount of earth materials from some other region and deposited them here, and nearly all geologists now agree that the agent that did the work was ice. FIG. 121. Channel of Glacial Drainage. THE COTEAUS OF THE MISSOURI. 227 We conclude then that there has been in this region an earlier ice invasion. The stage of the Glacial Period when the continent from Hudson Bay to this region west of the present Missouri River was all covered with a vast sheet of ice is known as the Kansan stage of the Glacial Period. The drift that lies west of the Missouri River, and much of that lying between the coteaus and the Missouri River, is Kansan drift. Just how it differs from the drift of the Wisconsin stage, and just how geologists distinguish between the two deposits we shall not attempt to discuss at this time. It may be stated merely that the Kansan drift is thought to be much older than that of the Wisconsin stage, and differs from it in important respects. This drift has been in some places 'nearly or quite all washed away so that but little or none at all remains covering the original land sur- face. It has been deeply eroded in many places by the streams of ice water which flowed away from the great glacier during and following the Wisconsin stage. Little Muddy, White Earth, Little Knife, Shell, Douglas, Snake, Painted Woods, Apple, and Beaver Creeks are exam- ples of such channels that were eroded into the older drift in their upper courses, and further toward the Missouri have cut entirely through the mantle of older drift into the underlying rock, as is shown by the out- cropping ledges of sandstone, shale, and lignite coal along the banks of these streams. 228 CHAPTER THE TWENTY-SECOND. THE PLATEAU REGION OF NORTH DAKOTA. State boundaries are arbitrary lines agreed upon by men. Geographic districts are regions that have some natural reasons for being separated from other regions. The state of North Dakota includes within the arbitrary boundary of its rectangular outline parts of two distinct geographic districts. The line separating these districts extends in a northwest-southeast direction in a general way parallel with and about 60 miles east of the course of the Missouri River. More accurately described it crosses the state from the extreme northwest corner, south- eastward through western Ward, eastern McLean and central Stuts- man Counties, to the southern boundary in the western part of Dickey county. This line marks very nearly the eastern-facing slope or escarp- ment of a great westward rising bench of upland, the Missouri Plateau. The escarpment rises abruptly to an altitude of 300 to 400 feet above the generally level plain on the east. The geographic significance of this natural boundary lies in the fact that here the prairie plains end and the great plateau begins. Ex- tending north far into British America and south almost to the Gulf of Mexico, though less conspicuous as a landscape feature farther south, this great earth bench or escarpment is the continental threshold from the low prairies of the central west to the Great Plains which rise thence westward to the foothills of the Rocky Mountains. Boundary. The southwestern part of North Dakota is thus in- cluded within the geographic district of the Missouri Plateaus. A brief review of the principal features of this relatively large physiographic district is here given in order to set forth more fully the geographic relations of that part within the state of North Dakota. This plateau- province extends a short distance across the International Boundary line as far as the divide between the head waters of the Saskatchewan river and the southward-flowing tributaries of the upper Missouri River. Its southern boundary extends through central Wyoming as the divide 229 230 THE STORY OF THE PRAIRIES. between the Platte and Yellowstone rivers, and thence eastward, south- ward of the Black Hills, to the Missouri Escarpment in central South Dakota. Great Plains The region known as the Great Plains extends from some distance beyond the Rio Grande River in Mexico through the United States and far in British America. This constitutes what is called a geographic province, and is one of the natural divisions of the North American continent. The northern portion of this Great Plains belt is what is known as the Missouri Plateau district. This is a great natural district because it is separated from the regions that surround it by some sort of natural boundaries. The surface features of this region are simple, but developed on a grand scale. That is, there is nothing particularly strange or difficult to understand about the structure of the landscape, only that everything is big and laid out on a large plan. General Topography and Structure. A notion of the general struct- ure and topography of the district may be formed by considering the plateau a vast earth block two miles in thickness and 500 miles square, built up of level-bedded rock layers. This seems like a pretty large thing to talk about in common words, or rather it seems a rather large thing to grasp in the imagination. But its size is the most awk- ward thing about it. It is really a simple problem a very large earth block. The western border of the plateau, where it merges into the eastern flanks of the Stony Mountains, has an altitude ranging from 4,500 feet to 6,000 feet above sea level. The western somewhat ragged edge is warped up against the mountain slope and beveled off by slow decay of the rocks and the erosion of mountain streams. The warping of the layers of rocks on to the mountain slopes is due to the bending of the strata when the mountains were uplifted. The strata thus lap up on to the mountain sides. This is because the mountains were upheaved after these rocks were deposited, and so they were bent up in the upheaval that produced the mountains. The rocks that were thus bent, and that now flank the mountains on the east have been beveled off by erosion and weathering. The upper layers in this western portion of the great block have been worn away so that the thickness of the rock layers there from top to bottom of the block is not now as great as it once was. THE PLATEAU REGION OF NORTH DAKOTA. 231 The eastern margin of the great block has been worn down by ero- sion during! long ages to a steep slope facing! the prairies on the east, the slope that rises west of Ellendale, Jamestown, Carrington, Fessen- den, Balfour, Kenmare, and Portal. The surface of this huge block has been trenched or cut into by winding streams, and sculptured by weath- ering. From beneath, that is, in the lower depths of the rocks of which this great block is composed, the layers in the western half of the block have been riven and thrust into by plugs and dikes of molten rock, and in many places the layers above have been pierced entirely through by the forces from below so that the lavas flowed out upon the surface. So great was the force from below that the strata were lifted bodily, the different layers wedged and spread apart, and the upper layers up- turned. This means that the rocks have been much broken and pushed out of their original positions, openings like volcanic necks or throats have been made through the overlying: strata, and molten rock or lava forced into these tubes or openings. When the region again became cool these openings were left "plugged" with lava. Such lava is often hard, and when the softer rocks surrounding are later worn away these hard masses are left sticking out as "volcanic plugs." When great cracks or fractures were made in the rocks by the tremendous forces that built the mountains these spaces were also filled with moulten rock forced up from below. These masses of molten lava when cooled are also fre- quently very hard and resistant to the weathering agencies, and so they often appear as great solid stone walls projecting out of the earth, and are known as dikes. When, however, the molten rock or lava was forced up through these tubes or cracks clear to the very surface and was poured out in such quantities that it flooded the whole region round about them this became what is known as a lava-flow, and the result now is a sheet of lava covering the earth. Laccolite Mountains Some times the force from below was so very great that immense masses of lava were forced in between the lay- ers of rock and pushed them apart like a huge wedge. This resulted in causing an uplift of the rock layers above, and sometimes the uplift was so great that the upper layers were turned up into a vertical position, and even sometimes completely overturned. These upwellings of molten rock in many cases assumed the form of vast reservoirs and lakes of lava ; such fluid masses, since hardened into 232 THE STORY OF THE PRAIRIES. solid rock, were so enormous in dimensions that out of them the agen- cies of rain and rivers have carved whole mountain ranges. The mass of solidified material out of which the sculpturing agencies have since formed a mountain is called a laccolite, or laccolith, a term which means a lake of lava. An example of an upthrusted laccolite mountain is the Madison Range northwest of Yellowstone Park. a The peaks of this range now stand at an altitude of a mile above the general level of the plateau or surface of the earth block. Lesser Plateaus. The larger streams and their tributaries have cut deeply into the surface forming wide valleys with steep bluffs bor- FiG. 123. West Rainy Butte, Billings County. Photograph by A. L. Fellows. dering their sides. Irregular areas of the great plateau have been thus cut around by the streams, and when the top layers of rock are hard and resistant to the action of the weathering agencies of frost, wind, and rain flat topped table lands or small plateaus are formed. In the western portion of this district many such blocks of rock that have been cut around by the streams are large enough to be called moun- tains. In the eastern part of the region, and along the lower courses of the rivers erosion has been more effective in broadening the slopes a Three Forks Folio, Montana, U. S. Geol. Survey. THE PLATEAU REGION OF NORTH DAKOTA. 233 and removing the original plateau surface. Here therefore such plateaus are less common, and when they do occur they are generally smaller in area and less high. Erosion has more completely removed them. Monadnocks Where the old landscape surface has been nearly worn away and removed by erosion, but occasional small plateaus or large buttes are left as hills, these are called monadnocks, a name derived from Mount Monadnock, a hill in New Hampshire which is of this FlG. 124. Brow of East Rainy Butte, Billings County. Photograph by A. L. Fellows. type. Monadnocks are a characteristic feature of the landscape in west- ern North Dakota. They will be referred to again later in this chapter. They are referred to in Chapter Eighteen as older buttes standing on the shoulders of the younger buttes. They sustain the same relation to the old landscape surface of the plateau before it was eroded that the Turtle Mountain Plateau does to the present surface of the Missouri Plateau. Work of Rivers In the eastern portion of the plateau more es- pecially than elsewhere the Missouri River and its tributaries have been the controlling agencies in making the landscape what it is today. The hills, ridges, valleys, and plains make up what is called the topography, or the form of the landscape. The work of excavation accomplished by 234 THE STORY OF THE PRAIRIES. the streams is represented by v the amount of earth necessary to fill their valleys up to a level with the highest hill-tops. The hills down to the level of the lowest streams of the region represent the unfinished work of the rivers. This not only means to fill up the stream valleys to the level of the highest land along the bluffs, but to fill in the whole region till it is all brought to the level of the highest hilltops of the region. Look up at the top of the highest hill you know of and imagine a line extending horizontally from this highest point, and that all the region is filled in with earth up to this level. On the other hand the work that remains for the rivers to do yet is represented by the carrying 1 away of all the earth of which the hills and all the land surrounding them are composed down to the level of the lowest stream bottoms of the region. Divisions The Missouri River flowing diagonally across the state divides the North Dakota plateau region into two parts, viz., a wide plain sloping eastward to the river and embracing all of the southwest- ern portion of the state lying between the western boundary line of the state and the river, and a narrow belt bordering the river on the east and extending to the edge of the plateau. These two divisions of the FIG. 125. Alkali Lake, Northern Williams County. The lake bottom is entirely dry, the surface being covered with alkaline salts. Photograph by Rex Willara. THE PLATEAU REGION OF NORTH DAKOTA. 235 plateau comprise what is popularly known as the "Missouri Slope" in North Dakota, and is frequently referred to as "the Slope." Slopes In the western portion of North Dakota the surface of the plateau is characterized by slopes, wide and gentle for the most part, but steep and rugged in the vicinity of streams. They appear at first to be arranged without order, yet every slope, gentle or steep, leads down- ward ultimately to the channel of a stream. These slopes were all formed by running water, except in cases where landslides or other dis- turbing factors have entered in. Every hillside is a part of trm side of a valley. The surface of the country is therefore completely drained. Lakes and marshes do not exist. If they existed once, as they very likely did, they have all -been drained by the streams. Streams have formed and are still controlling the features of the landscape plain. Comparison with Ice Plains. To fully appreciate this type of land- scape it should be compared with that of the prairies, the ice-made plains, such as those that extend eastward from the escarpment of the Missouri Plateau to the valley of the Red River of the North. Wide re- gions in this part of the State are wholly undrained. Since the melting of the great ice sheet, by which all former drainage lines were effaced, the modern streams have not had sufficient time to extend their valleys and drain the depressions. Here hills exist without valleys. Slopes lead to hollows, but not to valleys made by streams. Hollows occur without outlets. Lakes, sloughs and marshes are a common feature of the landscape. The Rivers of Western North Dakota The Cannon Ball, the Heart, the Knife, the Little Missouri and the "Big" Missouri are the prin- cipal rivers of the plateau district in North Dakota. The Cannon Ball, Heart and Knife Rivers rise in a narrow divide which runs north and south parallel with the Little Missouri River. These streams flow thence eastward in very tortuous channels to the Missouri River. Wide valleys bordered by steep bluffs have been eroded in the plateau sur- face. Extensive flood-plains and terraces have been formed in the val- leys. A remarkable feature about each of these three rivers is its rela- tively long and narrow drainage basin. The tributaries are short ; they join the main valleys at nearly right angles, and in nearly every instance are intermittent streams, i. e., water flows in them only during the sea- sons of heavy rains or melting snows. The streams flowing in the main channels during the summer season are exceedingly small when 236 THE STORY OF THE PRAIRIES. compared with the valleys which they occupy. These marked charac- teristics of the streams, viz., their parallel courses side by side, the long narrow region which each drains, their winding courses, wide valleys, extensive flats and terraces, find explanation in three important facts, viz., the climate of the region, the structure of the rocks, and the slope of the plateau surface. The drainage basins of these rivers are narrow and parallel and approximately equal in extent because the long uniform slope and struc- ture of the plateau gave to one stream no advantage over the others dur- ing the process of development, but distributed the run-off equally with- out showing favor to one valley more than to another. The small size of the streams during most of the year in comparison with the large valleys is due to the great variation in the volume of water carried at different seasons. It is a general truth relating to streams that by far the greater part of the work done in the erosion of valleys is done during time of floods. During a single day more work may be done in eroding the bottom and banks than in all the rest of the year. During times of high water the current is more swift, and swift currents erode with an increased power much beyond the simple ratio of the increase in the velocity of the current. Thus the stream at flood carries stones much larger than could be moved at ordinary times, and can carry many small stones rapidly. These latter become implements for cutting and breaking the banks and bottom. Thus the swift stream comes to carry an immense burden of earth materials, and when the current slackens the materials are thrown down, forming flood-plains, which later often become terraces. The small and gentle stream at low water does little except to modify the results of its more vigorous work- ing days. This general law of streams is more especially applicable in the semi-arid plateau country where the precipitation in violent rain- falls and rapidly melting snows is carried by the rivers only during limited seasons of the year. Another important factor in producing the relatively wide valleys with their attendant deposits is the long courses of the streams and the slight fall, together with the soft and easily erodable character of the rocks in this region. The rivers, laden with earth materials from the soft rocks of the region, do little downward cutting because of the slow movement of the current at ordinary times. Any cause that slackens the speed of the current will result in sediment which was being THE PLATEAU REGION OF NORTH DAKOTA. 237 carried by the stream being dropped, only to be in turn taken up again and carried farther whenever the current becomes swift again. Under such circumstances the streams will swing from one side of their val- leys to the other, undercutting the banks and widening the valleys. The low angle at which the tributaries enter the main valleys is also due to the slight decline of the slope of the main valleys. If the slope of the plains were more steep the tributaries would not enter the main streams so nearly at right angles. FIG. 126. Where the Bad Lands Begin. View looking east. Photograph by A, L. Fellows. Rivers are said to pass through three stages in their growth and development. These three stages are known as youth, maturity and old age. So also the topographic features of the landscape pass through three corresponding stages in the development of hills and slopes, viz., young, mature and old. Young rivers have narrow V-shaped valleys with flat-topped hills between their branches. They are vigorous streams, cutting downward for the most part, but with large un drained territory surrounding. Much of their work remains to be done. This is the stage of the streams in the western part of the Missouri Plateau, in western Montana and Wyoming. A mature river has a wide valley; the vigor of down-cutting is diminished; all the region intervening is carved into valley slopes; none 16 238 THE STORY OF THE PRAIRIES. of the upland surfaces are flat; the river swings from one side to the other of its valley, thereby broadening rather than deepening it. After this stage the river declines in activity until old age, when it erodes its valley and transports earth no more; the steepness of the grade of its channel bottom, which gave to it vigor in youth and matur- ity, is gone; hence it flows slowly; the intervening hills are cut away mostly to expressionless slopes on a featureless plain. The river's work is done. The Cannon Ball, the Heart, and the Knife Rivers have passed the stage of maturity and are approaching old age. Only in isolated places does any of the old plateau surface remain. The remnants are the hills with flat tops, the "old buttes" of the region, the buttes that "stand upon the shoulders of the younger buttes." They represent the fragments of the unfinished task of maturity. For the most part the plateau surface has been cut away and the hills lowered, or worn away altogether. FIG. 127. Where the Bad Lands Begin. View looking west. Southwestern Stark County. Figs. 126 and 127 were taken from the same point, the camera being merely turned on the tripod. Photograph by A. L. Fellows. The Little Missouri Eiver The Little Missouri River rises near the junction of the Missouri Plateau with the mountains in northeast- ern Wyoming. It flows thence north by east to central McKenzie County, North Dakota, where it swings broadly to the east and enters the Missouri River. The basin of the Little Missouri, like that of the other rivers of this region which have been described, is long and nar- row. It is about 320 miles in length, and in width varies from a maxi- mum of fifty miles to a minimum of about twenty-five miles. Through- THE PLATEAU REGION OF NORTH DAKOTA. 239 out much of its course the river drains an area much nearer the latter figure in width than the former. Eelation to Other Streams By reference to the map (Figure 122) it will be seen that the Little Missouri stands in a peculiar relation to its neighboring streams. It flows toward the north obedient to the long northward slope of the plateau, but dangerously near the hip, or edge, where the plateau breaks down to the east-facing slope. It thus crosses close against the head-waters of the Moreau, Grand, Cannon Ball, Heart and Knife Rivers, separated from their head valleys only by a very low and narrow divide. The rivers above mentioned have pushed their heading valleys westward up the slope, encroaching upon the territory of the Little Missouri until at several points the ridge parting their waters is less than six miles from the banks of the latter stream. FIG. 128. Crown Butte. A remnant of the old landscape which has been protected by the hard rock forming the crown of the butte. Photograph by A. L. Fellows. Piracy The Little Missouri is in imminent danger of having its basin invaded and upper waters "pirated" away by any one of its unfriendly neighbors on the east. This process of "beheading" a river, or of diverting the waters of one stream into the valley of another, is called "piracy." The invading river is, of course, the "pirate." The Lit- tle Missouri is most unfortunately located in this respect, having such a threatening number of would-be pirates on its eastern flank. This pro- cess of robbing the Little Missouri of its waters has already been initi- ated by the Belle Fourche River, a tributary of the Cheyenne.* The upper 150' miles of the Belle Fourche, that portion extending from the sharp bend northwest of the Black Hills, originally belonged to * Aladdin Folio, U. S. Geol. Atlas. 240 THE STORY OF THE PRAIRIES. the Little Missouri. This cutting up of the Little Missouri's valley piecemeal and diverting the sections, once begun at its head-waters, is more likely to be continued by the streams successively heading against its course lower down; each removal weakens the power to intrench its channel deep enough to be beyond the reach of its foe. Bad Lands. The basin drained by the Little Missouri is, through- out its larger part, the region of the Bad Lands. The river flows through the center of the Bad Land belt. The tributaries descend to the deeply intrenched main stream by steep gradients; these short branches fed by storm waters have cut deep into the soft clay beds. The FIG. 129. Buttes South of Williston. Photograph by Rex Wtilard. heading coulees extend out most intricately in all directions. There is little weathering, no soil except in the valley bottoms, and erosion in its most vigorous phase is the controlling factor. The result is a jum- ble of topograghic forms that beggar description. There is no beauty here. Steep hills with ugly bulging flanks stand foot to foot, corru- gated up and down their naked sides with rain gutters. Sharp-crested ridges wind in and out forming cirques and amphitheatres at the heads of streamless valleys below. The divides between tributary valleys are often sharp-crested ridges not inappropriately called knife edges. Ver- tical pillars and walls of clay, veritable mud fences, stand along the sides of deeply worn channels. From within the valleys no extended view can be obtained; the observer is surrounded by vertical or steep- rising slopes on all sides. From the top of a lofty butte the landscape THE PLATEAU REGION OF NORTH DAKOTA. 241 appears a myriad of hilltops closely set together and massed back of each other until they blend far away in a level sky line. Structure In this region where the streams have deeply dissected the plateau the structure and composition of the rocks are readily seen. The formations are built up of horizontal strata, "one layer above another like boards in a lumber pile." The materials of which they are composed is fine sediment, beds of clay and sand for the greater part alternating with each other. Name The Bad Lands were so named by the early French explor- ers because they found them lands difficult to travel through. The country seems to have sustained that reputation ever since. No one who has ever traveled through this region will question the propriety of FIG. 130. Sentinel Butte. Photograph by H. V. Hibbard. the name. There is practically no direction of traversing the region except by way of the waters. Roads, when roads there are, follow the courses of winding streams. A Burning Mine. In a few places in the Bad Lands country the outcropping beds of coal have become ignited and are now burning back under the hills. As the fire advances beneath the surface the baked earth above opens in great crevices, admitting a down draft of air and main- taining further combustion. Such a "burning mine" is at present located close to the tracks of the Northern Pacific Railway near Sully Springs. Here a crater-like depression, about 500 feet in diameter, has been formed above the burning coal. The sides of wide fissures opening 242 THE STORY OF THE PRAIRIES. deep into the earth glow white hot and red from the fires of the sub- terranean furnace. Great volumes of gases^ with stifling and sulphur- ous odors, arise from the crater, but no smoke or flame; the combustion is complete. The fires advance slowly but with great persistence against the bed of coal. The vein at Sully Springs is known to have been burn- ing during the past twenty-five years. In that time the crater has moved perhaps 100 feet northward. The advancing fire leaves behind it a trail of red-baked clay and earth fused into scoriaceous masses. This process if extensive enough in the past ages would account for the red strata so widely shown outcropping on the hillsides of the Bad Lands. FIG. 131. Crest of Hill in Ragged Buttes, McKenzie County. Showing cross- bedded structure. Photograph by Rex Willard. Sentinel Butte -Throughout the southern part of the plateau, standing back from the streams, usually on or near their divides, are sharp hills, serrated ridges, and flat-topped buttes. Their summits, capped with resistant sandstone, mark the level of the old plateau sur- face upon which the streams began their work of valley development. They are remnants of the unfinished work of the streams at the stage of maturity, and are termed by geographers monadnocks. Such a monad- nock is Sentinel Butte, located three miles south of a village of the THE PLATEAU REGION OF NORTH DAKOTA. 243 same name on the Northern Pacific Railway. From its flat top, 600 feet above the surrounding undulating and rolling-hilly plain, the land- scape beneath appears as a great map in midsummer, hills showing the gray of range lands, and valleys the green and yellow of wheat and forage crops. The hill was used in the early Indian fighting days of the northwest by the United States troops as a vantage point of observation ; hence the name Sentinel Butte. Two of Custer's scouts, killed by the Indians, are buried near the top of the hill. Their grave is marked by a little cairn and a rough slab of sandstone. A vein of lignite coal ten feet in thickness underlies apparently the entire base of the butte. From an outcrop of the vein on the north side of the hill a coal mine has been developed and furnishes a good grade of fuel at the cost of mining only. Other monadnocks, or remnants of the old plateau, whose tops rep- resent the original surface of the plateau, are : Camel's Hump, White Butte, Red Butte, East and West Rainy Buttes, Square Butte, the H. T. Butte, Buillon Butte (pronounced Bool-yong), in Billings County; Kill FIG. 132. The Great Stone Face, McKenzie County. Photograph by Rex Willard. 244 THE STORY OF THE PRAIRIES. Deer Mountains, in Dunn County; Brenchaud's Butte and Ragged Buttes in McKenzie County; Short Medicine Pole Hills, Pommes Blanches Hills, in Bowman County ; Black Butte, Tepee Buttes, Whet- Stone Buttes, Wolf Butte, in Hettinger County ; Hailstone Hill, Heart Butte, Dogsteeth Buttes, in Morton County. These buttes do not all have the same elevation above the adjacent plain by which they are surrounded. In some cases the tops have been lowered by erosion and weathering, but generally their tops may be said to represent fragments of the original plain of the plateau surface. The Plateau East of the Missouri Eiver That part of the Missouri Plateau extending east from the Missouri River to the escarpment joining the prairies, differs in its topographic features and drainage from the section to the west of the river. During that time in the earth's history known as the glacial period the front of a great continen- tal glacier lay along the crest of the Missouri escarpment. The south- westward advance of the ice over the prairie region from the enormous snow fields of Labrador was doubtless stopped by the rising margin of the plateau. The debris carried by the glacier, as fragments of rock, clay and boulders, was lodged in the form of a broad hilly belt at the edge of the melting ice. This belt of hilly topography, with its accom- panying marshes, lakes and undrained depressions, constitutes the first and outermost moraine of the continental glacier. The region is known as "The Hill Country of the Missouri Plateau." From the melting ice front there flowed to the Missouri river glacial streams carrying great quantities of sand and gravel. The valleys of these short rivers, built up with extensive terraces, are prominent features of the topography between the hill country and the Missouri River. Both the hill country and the valley region are excellent grazing lands, and large tracts are being rapidly brought under cultivation. Wells from twenty to seventy feet in depth yield abundant water. Industries All the plateau country of North Dakota is compara- tively new, but settlers are moving in and land values are rapidly ris- ing. Stock raising is at present and will probably continue to be the leading industry. Fine herds of cattle, horses and sheep feed on the nutritious grasses of the plains. Ranchmen have introduced the better breeds of stock and realize large profits on this branch of industry. When crops adapted to the soil and climate are introduced fanning will become a larger element of industry than at present. CHAPTER THE TWENTY-THIRD. AGRICULTURE WEST OF THE MISSOURI RIVER. West of the Missouri River in North Dakota is a vast region about which many erroneous opinions have been held by people living in other States and in the eastern portion of our own State. North Dakota is so large in extent that it is not surprising that there should be many people who have lived two decades in the State and yet have no definite knowledge of the character of many parts of the State. Many persons who have lived and prospered during a goodly period of their lives in the Red River Valley, or other eastern portions of the State, have never been west of the Missouri River. It is not strange, therefore, that the idea has gained wide acceptance that the country west of the Missouri River is "Bad Lands." This vast domain has but recently been recognized as an agricultural region. So much success has been attained in the grazing of horses, cattle, and sheep, and so little attention has been given to general agri- culture that until recently the whole country west of the Missouri River has been generally regarded as a vast grazing domain. Within the past few years a great influx of settlers whose purpose has been general farming rather than exclusive grazing, has largely changed the sentiment regarding this country, as it has also changed the character of the pursuits of the residents in this part of the State. Ranching, by which is meant that phase of agriculture in which the grazing of large herds of horses, cattle, and sheep is the principal indus- try, has largely given place to the more intensified methods of diversified farming except in those localities where, clue to the natural roughness of the land and its consequent unadaptability to> diversified farming, stock-raising is still the dominant industry. The quest of homeseekers, both from the older and more thickly settled States and from foreign lands, for free government lands has made the region a mecca for immi- gration. Farmers now live where but one, two or three years ago was the open range and the unbroken sod. 245 246 THE STORY OF THE PRAIRIES. Questions about the climate, rainfall, soils, water supply, available fuel, and means of transportation are frequently asked by the prospective homeseeker or the newly settled homemaker, and an effort will be made in this chapter to give a brief and simple statement regarding the questions indicated. Climate The climate of the western part of North Dakota is of the healthful and invigorating kind that characterizes the whole State, though there may be said to be a little odds in favor of the western portion as compared with the eastern. Warmer currents of air from the Pacific Ocean flow southeastward after crossing the great mountain FIG. 133. Family and Ranch Home of E. Paulson, Knife River. Photograph by A. L. Fellows. axis in the Canadian northwest, and these warmer winds influence the climate. On the whole the climate is wholesome and satisfactory, and few States afford more healthful and invigorating conditions. The winters are cold, and there are times of extreme severity, as indeed there are in any of the northern States. The weather is not more trying, how- ever, than that of northern Illinois, Iowa, southern Minnesota, or west- ern New York. Extravagant reports of temperatures are given wide circulation and are often accepted as true without verification. AGRICULTURE WEST OF THE MISSOURI RIVER. 247 The error, it may be explained, grows out of the common use of cheap and unreliable thermometers. These thermometers have not been tested and vary from one to two degrees to as much as ten or twelve degrees. Thus a thermometer that hangs at the "corner store" may register minus 46, 48, or minus 50 degrees or even more, while a reg- istered thermometer at the same moment and in the same locality reg- isters minus 34 or 36, or rarely minus 39 degrees. Temperatures lower than minus 39 degrees F. are very seldom experienced in North Dakota. An agreeable feature of the climate in this region is its uniformity. The greater evenness and continuity of the conditions in North Dakota than in many states farther east saves much of the trying character of winter. Rainfall The question of the annual rainfall of this region is one about which there is no little misconception. It has long been cur- rently accepted that the rainfall west of the Missouri River is not suf- ficient to make general farming safe and successful, and therefore profit- able. This generally accepted opinion, though requiring evidence to remove, is not, in the opinion of the write*, a true verdict. The fact of the amount of the annual rainfall is one that cannot be readily, deter- mined by the average person from general observation, nor yet by the closer observation of the frequency and severity of storms. Even if one were to keep a written record of the days when rain fell, a thing that is very rarely done, still he would not have a reliable record of the rainfall. It is only by the daily reading of a scientifically constructed gauge that a correct record can be obtained. Within recent years a considerable number of gauge stations have been established, and these give a reliable basis for estimating' the rainfall for any particular local- ity. Probably one reason for the unfavorable impression regarding the sufficiency of the rainfall in this region has been that during the time since this part of the state has been occupied by white settlers almost no attempt at cultivation of the soil has been made, and when such attempts were made no particular attention was paid to the matter of cultivation or the adaptation of seed to the conditions of this region. Stock-rais- ing was the principal industry, and general agriculture was not con- sidered at all. When, therefore, an occasional small area was ploughed, seeded, and neglected, and no satisfactory crop harvested it was as- sumed that the trouble was due to lack of moisture. It might be said 248 THE STORY OF THE PRAIRIES. in this connection that if similar methods of farming were used in Illinois or Iowa loss in quality and quantity of crops and financial fail- ure would most certainly follow. Systematic records of rainfall covering any considerable number of years have been kept at only a few stations in this newer part of North Dakota. Comparison of the average rainfall, as recorded at these stations, with the records of stations in other parts of the state where no question has ever been raised as to the sufficiency of the rainfall to produce profitable crops, furnishes a basis for an opinion which at least would seem to contain the elements of fairness. FIG. 134- Cuskelly Ranch and Killdeer Mountains, Dunn County. Photograph by A. L. Fellows. The annual precipitation* for four stations located at points either on the Missouri River or west of it are given below, these records being for the period of fourteen years, from 1892 to 1905 inclusive. These are the only stations in this portion of the state for which complete rec- ords are available for this period.** 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1892 1903 1904 1905 Bismarck 18.17 13.74 14.32 16.92 16.63 14.33 13.67 15.47 17.88 15.59 15.95 17.96 14.17 17.19 Dickinson Fort Yates .... Williston *** 21.23 14.26 11.64 16.30 15.45 *** 12.20 17.76 *** 12.96 17.07 18.48 20.08 22.04 *** 18.32 12.19 11.92 19.55 14.44 17.27 17.71 12.61 11.78 16.80 15.81 12 92 13.42 18.36 16.49 16.85 15.00 17.69 15.19 17.30 9.44 16.55 *** 10.66 * By precipitation is meant rain, snow, hail, sleet, or any form of moisture that is "precipitated" or falls from the clouds. All frozen forms of water are reduced to equivalent amounts in the liquid form. ** Compiled from the records of the United States Weather Bureau, Bismarck, N. D. *** Records wanting for certain months, so that the correct amount for the year could not be given. AGRICULTURE WEST OF THE MISSOURI RIVER. 249 Soils. The soils of a great part of the region lying west of the Missouri river are different in origin from the soils of most of the state, and differ very considerably in character from those of other parts of the State. The soils of any region are determined by the geological conditions that have prevailed in the region. The soils of this region therefore differ from those of other portions of the State because of a different set of geological conditions through which the region has passed. Most of the State of North Dakota falls within that great portion of North America which was covered by the ice of the great ice sheet dur- ing what is known as the Glacial Period. The surface formation in the region over which this great sheet of ice passed is often spoken of as "drift." This drift formation is made up of the broken and pulverized fragments of the rocks of the land surfaces over which the ice passed together with the original soil that covered the face of the landscape before the invasion of the ice. In those regions, on the other hand, where the ice sheet did not extend the soils have not been modified by this agency, and consist of the residual material arising from the dis- integrating or weathering of the rocks which form the foundation of the landscape. The soils in such a region are spoken of as residual soils, as distinguished from the drift soils just referred to. The soils of any particular locality in a region where the great ice sheet has never been will therefore be made up of the same materials that composed the rocks in that locality, minus whatever has been re- moved by the process of erosion. The rocks of the great Missouri plateau, of which the region west of the Missouri river is a part, are mostly shales, sandstones, and clays. The soils are residual formations derived directly from the disintegra- tion of these rocks, and consist generally of an admixture of sand and clay, with organic matter added, principally from the decomposition of vegetable matter such as has grown upon the landscape during long ages. Among the most widely distributed types of soil in the northern states are those belonging to the classes of loams and clays. There are sandy, stony, gravelly, silt, clay, etc., loams, and stony, gravelly, sandy etc., clays. These terms have to do with the character of the soil as to texture, structure, and quality. The character of soils has generally 250 THE STORY OF THE PRAIRIES. speaking been determined by the processes through which the rocks of the region have passed. Sandy soils are derived, directly or indirectly, from sandstone rocks or from rocks which when broken up by weathering yield sand fragments. Clay soils are derived from rocks that contain argillaceous or clayey materials, such as shales, slates, and clay-rock. Loams are soils that contain some clay and some sand, and these are further described as sandy loams or clay loams according to the relative amounts of clay and sand. Sandy soils are often spoken of as "light," and clay soils as "heavy." Not that one is lighter or heavier so far as actual weight is concerned, or lighter or darker as to color, but looser or more compact in texture. Light soils may be dark in color, and heavy soils may be light in color, but light soils are generally sandy and porous, and heavy soils are clayey and compact in texture. Subsoil is that portion of the surface formation which lies below the superficial few inches, and differs from the soil proper in color, texture, and amount of organic matter contained. In the classification of soils the subsoil as well as the surface por- tion is considered, inasmuch as the character of the soil, so far as agricultural conditions are concerned, is determined by the subsoil as well as by the character of the soil proper. The soils of Oliver, Mercer, McKenzie, northern Dunn, and east- ern Morton Counties are in part glacial soils, that is, the soils are not entirely residual (derived from the rocks in the localities where they occur), but have been in part transported from some distance by the great moving ice sheet. A belt having an indefinite edge to the west- ward lies along the west side of the Missouri river, which belt repre- sents the western limits of the glaciated area of North Dakota, and of the continent of North America. This "belt" of land along the west side of the river shows by the character of the soils and the rocks that lie upon or near the surface that the great continental glacier was once here. Toward the west the belt fades out and becomes indistinguish- able from the land farther west over which the ice did not pass, but the eastern part of the belt is sufficiently modified as to the soils and the landscape features to be readily recognized. For a distance of 15 to 30 miles west of the river the soils are con- siderably modified by the presence of drift. Farther from the river the presence of drift, and therefore the presence of the one-time ice sheet, AGRICULTURE WEST OF THE MISSOURI RIVER. 251 is shown by the occurrence of scattering boulders of granite and other hard rocks. Where only a few boulders occur, and these scattered widely so that the traveler sees only an occasional specimen, and these only at intervals of many rods or even miles as the western limit of the drift is approached, the soil does not appear to have been greatly modi- fied by the presence of the drift, but is largely residual in its origin. In the region nearer the river the occurrence of boulders is common, and the soils show evidence of the influence of drift sands and gravels, silts and clays. FlG. 135. Jack Williams' Ranch, near Little Missouri River, McKenzie County. Photograph by Rex Willard, The soils nearer the river resemble the soils on the east side of the river and the interior portion of the state. The soils in the region far- ther west than any boulders of granite occur constitute a different series of soils. These last, as indicated before, are residual soils formed from the rocks of the region. Along the western portion of the indefinite belt of drift west of the river the soils are influenced less and less by the drift toward the west and more and more toward the east. The soils, therefore, in the belt bordering the Missouri River on the west constitute a transition type from the glacial soils of the eastern 252 THE STORY OF THE PRAIRIES. portion of the State to the non-glaciated or residual soils of the south- western portion of the State. The soils in any region are an expression of the geological processes that have occurred there. The soils in the regions beyond the limits of the great continental glacier have had a different geological history from the soils where the action of the great ice sheet affected the whole land surface. In western Morton, southern Dunn, Hettinger, Stark, Billings and Bowman Counties the soils are residual in character, that is, they have been derived from the layers of rock which underlie the landscape. These rocks were originally deposited on the bottom of a shallow sea, and when first thrown down as sediments were in the form of muds and sands. During the long lapse of time they became solidified into more or less compacted rock. The layers or shelves of rock that jut from the sides of buttes, and the crags and pinnacles that project from the crests, are eroded and broken edges of the rock layers that were once the muddy or sandy bottoms of the sea. All the rock that used to be between the hills, that is, the material that once filled the present valleys, has been removed by erosion and carried away by the land waters that have flowed off from the land. The result of the erosion processes upon the rocks, weathering, transportation, and deposition, is the soils as they exist today. In places where there was sandstone rock the soils are generally sandy, because the sandstone is merely sea sand compacted into stone. When the stone breaks up by the action of the weathering agencies of heat, frost, air, and water the rock becomes sand again. If a good deal of water flows over any region the soluble and finer clayey parts are carried away by the water, and the heavier particles of sand are left. It is this process of disintegrating and carrying away the finer parts of the soil that makes the flood waters of the streams roilly or muddy. The soils in the region west of the drift covered belt, have been determined in character by the kind of rock in the particular region and the character of the destructive work of erosion that has taken place. The soils west of the Missouri River may for the purposes of this study be divided into six groups, though this classification will be quite gen- eral in its application. These are : ( i ) the soils that have been de- posited by the great ice sheet, or which have been largely modified by it. These may be called for convenience glacial soils; (2) the heavy lands AGRICULTURE WEST OF THE MISSOURI RIVER. 253 occupying the lower places, often called "gumbo;" (3) the sandy loam of the rolling lands; (4) the sandy or stony areas of the higher ridges and hills; (5) the eroded and broken regions, called ''bad lands;" and, (6) the bench and bottom lands along the stream courses. The glacial or drift soils are described in other parts of this book and need not be again described here. Their occurrence in low broad hills, often small, and distributed in the irregular fashion of drift hills in the eastern portion of the State may be recognized by any one who has studied drift deposits anywhere. They have been so far modified in this region by the erosion of the streams that their true character may not always be readily recognized. The presence of boulders of granite is, however, always an indication of drift, and gravelly deposits composed of pebbles of granite and other rocks not native to this region are pretty sure indicators of drift, and of the glacial origin of the soil. In this connection it may be observed that in Bowman, Billings, and western Hettinger Counties pebbles occur in great abundance which at first appear much like glacial or drift pebbles, both in their petrologic or rock character and in their rounded and often smoothed form. These rocks, however, occur most abundantly in the extreme southwestern part of the state, and only scatteringly toward the Missouri River. They will not be mistaken for drift pebbles after a little experience. The "gumbo" soils in the valleys are often in nearly level tracts. In fact it is not infrequently the case that the land is so nearly level that drainage is difficult. The texture and -character of the soil in these places is that of a fine grained and compact clay, and is not infrequently somewhat alkaline. The clay appears to have been derived from a shale- clay formation which is one of the rock formations of this region. In this shale-clay are some alkaline mineral substances which were depos- ited with the muds upon the ancient sea bottom. These mineral sub- stances accumulate in the low places where there is not drainage suffi- cient to carry them away. The sandy-loam soils, when underlaid with a clay subsoil, are the best soils of this region for general agricultural purposes. And this type of soil embraces by far the greater part of the agricultural lands* throughout this region. The soil is described as a sandy loam. This means that in texture the soil is more light and porous than the heavier clay soils of the low * By agricultural lands is meant all those lands that are in any reasonable sense fit for general farm- ing, and the term includes all the land that might reasonably be ploughed, but does not include rocky hill crests or "bad lands." 254 THE STORY OF THE PRAIRIES places described above, and yet contain enough clay to give body and firmness to the soil. It is usually dark in color due to the presence of organic matter. When underlaid with a clay subsoil this gives support to the soil by holding the soil moisture from draining away tco rapidly and furnishes a supply of moisture from below to the soil above dur- ing the warm summer months when crops are growing. In the third type of soil referred to, if indeed it may be called a soil type, the surface formation of the hills and ridges, is included the high or stony parts of the hills or buttes, and the higher ridges which are capped with sand derived from a sandstone rock layer j but FIG. 136. Valley of Cherry Creek, McKenzie County. Photograph by Rex Willard. still covered with grass or other vegetation. This type is such as will furnish pasturage, but will not generally be of much value for other fanning purposes. The fourth or "bad lands" type is that characteristic kind of land- scape which is widely known as "bad lands," but which in reality should be called, as first named by the early French explorers "bad lands to travel through." These lands are not adapted to general farming, and are not included in the general classification of agricultural lands. They are lands well adapted to grazing, but are inaccessible to general farm- AGRICULTURE WEST OF THE MISSOURI RIVER. 255 ing. The hillsides are often naked of any vegetation, the layers of shale, shale-clay, and sandstone extending to the surface without cover- ing of any trace of soil. Soils derived directly from the erosion or wash- ing and disintegration of these naked rocks accumulate in the valley bottoms, and these soils support nutritious and valuable varieties of grasses, and produce large crops of alfalfa, timothy, oats, barley, and garden crops when cultivated. The areas that are available for cultiva- tion are, however, usually limited, so that the general fact remains as stated, that these lands are best adapted to grazing and not to general farming. The soils of the stream bottoms and benches comprise a group of soil types falling in a class by themselves, and differing from those above described. The flats and benches along the stream courses rep- resent deposits made during the flood stages of these streams. By flood stages is here meant those times in the past when vastly larger streams flowed down these water courses than any that ever flow in them now even during the highest freshets. It is not necessary to give here a geological history of these streams, but merely to refer to the facts that have a bearing on the present char- acter of the soils. The soils of the broad level benches that border most of the larger streams frequently have a gravelly subsoil. This renders the problem of their successful use for farming lands a somewhat difficult one, since the gravelly subsoil permits of ready under-drainage, and does not sufficiently conserve the moisture of the soil for the grain crops during the drier seasons. No general rule can be laid down however for all these lands. The soils and subsoils represent floodplain deposits of streams heavily laden with sediment. Such streams deposit their burden of earth materials whenever the current is slackened. The soils will differ therefore in different places, and often within short distances, according to the con- ditions which affected the rate or flow of the waters of the stream. The particular consideration of these stream deposited soils must await the more detailed investigation of a systematic soil survey before they can be correctly mapped and the types defined. In many instances the flats or bottom lands constituting the lowest extensive floodplain are sufficiently heavy to make valuable farming lands, and fine hay-meadows may be developed on these lands. On the 256 THE STORY OF THE PRAIRIES. other hand, some of the bench lands, while level and beautiful to look upon, are too sandy in texture and the subsoil too loose and porous to make farming by the ordinary methods profitable. Water Supply The supply of water from streams and springs in the region of North Dakota west of the Missouri River is much greater than that in many parts of the State east of the river. None of the streams west of the Missouri are large except during the rainy seasons and times of melting snows. However, there is a constant supply of water for stock during the entire year in the larger streams. Streams and springs have furnished the water supply for the stock of the ranges during the past, but with the settlement of the country by farmers this supply will be insufficient, and owing 1 to the distances it would be im- practicable for the settlers generally to depend upon the water of the streams. Some fine springs furnish excellent water and an abundant supply, but this supply can be only made available to the few who are so fortunately situated as to have access to a spring. Obviously, therefore, the question of a water supply from wells is soon bound to become a practical question. So far as this problem has been solved by the practical test of experience the results seem very favorable and satisfactory. Few tests for artesian water have been made, and the question of artesian or flowing wells is therefore largely an open one, though there seems to be ground to doubt if artesian water will prove to be available in this region generally. However, the sandstone layers underlying the country should, and probably do, contain abundant supplies of water, so far as geological deduction gives a clue. Experience in digging wells wherever observa- tions have been made seem to substantiate this view. Good supplies of water are obtained in wells from 25 to 70 feet in depth, so far as records are at hand to show. Nothing like a complete summary of the records of the known diggings has been made as yet, and the knowledge at hand is limited to the occasional observations made in various parts of the region. These observations lead to the tentative conclusion that there is probably an abundance of water of good quality to be had at depths, as before indicated, of 25 to 70 feet, though it may be found necessary to go to greater depths in some cases. The water should be good, based on geological evidence, for it is almost always the case that water that is derived from sand or gravel beds is of good quality, often AGRICULTURE WEST OF THE MISSOURI RIVER. 257 soft, and generally free from alkaline or other undesirable mineral sub- stances. Fuel Western North Dakota is abundantly blessed in the matter of natural fuel supply. It is probably not an overestimate to say that there is reasonable assurance, based upon geological data, that coal seams underlie practically all of the region west of the Missouri river and, not only this, but it is so situated that mines can be operated almost anywhere. There are scores and hundreds of outcroppings of lignite coal of good quality that have never been touched by way of develop- ment, simply because there are so many accessible openings, so many workable mines distributed wherever there is a demand for coal. Many farmers own their own coal mines, just as in the eastern states farmers own a "wood lot." Coal can be had in hundreds of places for the mere labor of mining. In other cases farmers drive to the mine, have their wagons filled, and then haul home their supply, paying perhaps $1.00 per ton for the coal loaded at the mine. Or it can be purchased delivered for a price that merely pays for the mining and hauling. Lignite coal burns without the smoke, soot, and disagreeable black- ening that accompanies the use of the bituminous coals of the Mississippi valley. A supply of fuel for a North Dakota winter costs not to exceed one-fifth of what it costs for southern Minnesota, Iowa or Wisconsin. Transportation One of the most serious hindrances to the agri- cultural development of the great domain west of the Missouri River in the past has been the problem of transportation of farm products and the inaccessibility of local markets. At the present time the counties of Hettinger, Oliver, Mercer, Dunn, and McKenzie are without rail- road facilities, and the result is a long and tedious journey to and from local markets. There is reason to believe that in the near future better railroad facilities will be afforded. As soon as there are products to be trans- ported and supplies to be brought in, as demanded by an increased popu- lation, a natural result would be that railroads would seek these avenues of business. Already several surveys have been made in various parts of the vast region west of the Missouri River, and it may confidently be predicted that soon there will be lines of railroad traversing* these broad and fertile lands. CHAPTER THE TWENTY-FOURTH. THE WATER SUPPLY. Conditions Necessary for Artesian Wells. In a prairie country more than in a broken or hilly country the water supply for men and animals comes from wells. In a prairie country there are generally few streams and these are apt to be small and often sluggish so that their waters are not good for drinking, and there are not usually many springs. In North Dakota a large part of the water supply for towns and cities as well as farms comes largely from wells. This condition makes the pos- sibility of obtaining artesian wells over a large part of the State a very fortunate thing. An immense saving to the people of North Dakota results each year from the fact that the water flows from the depths of. the earth without being pumped. Artesian wells have been in use^for hundreds of years, but the fact that they have been long known does not make it possible to obtain them in every place. It is only where the structure of the earth deep below the surface is such as to cause an upward pressure of the water that an artesian well can be obtained. The word "artesian" is borrowed from France from the province of Artois, because such wells were first known there. When flowing wells began to be found in other parts of the world they were called Artois wells or Artois-ian wells, and by usage the word has become "artesian." Artesian wells differ irom common wells in that the water flows from them naturally, that is, without being pumped. They are often deep, but there are many wells not artesian which are much deeper than some artesian wells. Sometimes artesian wells are a mile or more in depth, and there are many in North Dakota which are less than fifty feet in depth. There are even natural artesian wells, in which the water rises to the surface as springs, but yet the flowing of the water is due to the same causes as those which make the flow from a boring. The reader will be able to understand the conditions which are necessary for an artesian well from Figure 137. The section shows the THE WATER SUPPLY. 259 relation of the underlying rocks from the Red River Valley westward to the Rocky Mountains. The source of the water supply is the region along the base of the Rocky Mountains where the rain which falls upon the ground soaks into the soil and travels underground along the porous gravel and sand of the Dakota Sandstone. The water follows this layer at first for the same reason that it flows down hill on the sur- face. It fills all the little cavities or spaces in the loose rock because of the pressure due to the weight of the water. Now, if a boring is made from the surface down through the over- lying rock layers the water will rise in this opening and there will be FIG. 137. Section showing Water Supply of Deep Artesian Wells. After Upham. a flowing well. How rapidly the water will rise above the ground and flow out at the surface or how high it will rise depends upon the pressure or "head" which the water has, for the same reason that the height of the tower on which the water tank stands in a city determines how rapid a stream of water can be poured from a hose in time of fire, or how high a stream can be thrown, or how high it can be made to run in pipes in the houses. Not taking into account the friction of the water in its passage in the rocks it will rise as high as the source or collecting ground along the foot of the mountains from whence it comes. But it will not ac- tually rise nearly as high because of the friction, but we may think of the flow from an artesian well being determined by the "head" or height of the land where the rainwater soaks into the ground, and if this is a good deal higher than the surface where the well is the water will flow out with considerable force, but if it is not much higher and is a long distance away then the water may rise only part way in the bor- ing and not flow out at all. There must be a layer of clay or shale or some rock through which water does not pass readily both above and below the gravel and sand layer or else the water would soak away into the other layers of rock 260 THE STORY OF THE PRAIRIES. and so would not rise in the boring and flow out at the surface. There are, therefore, certain conditions necessary for an artesian well; there must be, firstly, a collecting ground higher than the surface where the boring is made; secondly, there must be a layer of rock both above and below the layer from which the water flows through which water can- not readily pass; and, thirdly, there must be enough difference between the height of the collecting ground and the place where the boring is made to overcome the friction or resistance to the passage of the water. Deep Artesian Wells West of the Red River Valley. The artesian wells at Devils Lake, Jamestown, Ellendale and Oakes, and a large number in South Dakota, obtain their water supply from the Dakota Sandstone by deep borings through the overlying formations. The borings vary in depth from less than one-fifth of a mile at Oakes to more than one- third of a mile at Devils Lake and Jamestown. These pierce through the Fort Pierre, Niobrara and Fort Benton formations, which are mostly shale and through which water does not readily pass. From the section showing the formations of the State (Figure in) it will be seen that the top of the Dakota Sandstone in the northern portion of the State is nearly at sea-level. The depot at Devils Lake is 1,468 feet above the level of the sea, and the surface about the well is six or eight feet higher. The depth of the well is 1,511 feet. The boring penetrates eighty feet into fine white sand. The top of the Dakota Sandstone is, therefore, at this point about forty-five feet above sea-level. The depot at Jamestown is 1,395 feet above sea-level, and the well reaches a depth of 1,476 feet, penetrating into the top of the sand- stone. The surface about the well is about eight feet below the depot. The upper part of the Dakota Sandstone beneath Jamestown is there- fore about eighty-nine feet below sea-level. Ellendale is 1,449 ^ eet above the sea and the well penetrates the Dakota Sandstone at a depth of 1,087 f eet > so tnat the upper portion of the Dakota Sandstone at this place is 362 feet above sea-level. At Oakes the elevation is 1,322 feet and the Dakota Sandstone is reached at a depth of 944 feet, so that the Dakota Sandstone beneath the sur- face at Oakes is 378 feet above the sea. Farther south in South Da- kota the sandstone is reached at still less depths, showing that its upper portion is nearer the surface southward. At Vermilion in the south- east corner of South Dakota the sandstone is reached at 323 feet below the surface, or 818 feet above sea-level. THE WATER SUPPLY. 261 It would, therefore, seem that artesian wells may be expected to be obtained anywhere over the central and eastern portion of the State, and much farther west, by penetrating to the depth necessary to reach the Dakota Sandstone. It is not always possible, however, to get a flow of water even when the general conditions are such as to warrant, the expectation. Sometimes the sandstone is pierced in a place where from some local cause, such as an unusually hard place in the sandstone rock, the water is not able to pass readily through the rock and so cannot rise in the boring with force enough to cause a flow. The artesian well at Grafton penetrates through the drift to a depth of 298 feet, but instead of entering the Dakota Sandstone passes next through 137 feet of limestone belonging to the Lower Silurian forma- TI^T^^ 204 FIG. 138. Section showing the Series of Artesian Wells from Devils Lake and Jamestown southward to Yankton and Vermilion. Horizontal scale, 90 miles to an inch. U. S. Geological Survey. tion, and obtains its flow of water from a sandstone layer still lower in the Lower Silurian series. The well had a depth of 915 feet when first drilled, a small flow of very salt water being obtained at a depth of 898 feet, from a sandstone layer next to the Archaean Granite. The boring was filled, however, below the sandstone layer which yields the very large flow of brackish water. A section of the rocks passed through by the boring is shown in Figure 76. It is interesting to note that the Dakota Sandstone was not struck at all, showing that this sandstone does not, in this part of the Red River Valley, extend as far east as this. Artesian Wells in the Red River Valley. In the Red River Valley there are many artesian wells which range in depth from 250 feet to 400 feet, and which, 'like the deep wells at Devils Lake, Jamestown, Tower City, Oakes and Ellendale, derive their water supply from the far-distant foothills of the Rocky Mountains. From Blanchard north to southern Manitoba most of the artesian wells are of this class. The source of the water is the same as that of 262 THE STORY OF THE PRAIRIES. the deep wells farther west, that is, the Dakota Sandstone. To obtain an artesian well, therefore, it is needful to penetrate through the mass of drift. The greater part of the material of the deeper portion of the floor of Lake Agassiz consists of boulder-clay or till. Water cannot pass through clay much more readily than through a porcelain dish. The glacial clay has, therefore, to be drilled through, and when the sandstone is reached a flow of water usually results. This is not always the case, for sometimes the sandstone is hard and compact, so that the water is not able to soak through it readily, and so it is not always possible to obtain an artesian well. The drill penetrates through a few feet of soil and fine silt and soon enters blue clay. Occasionally there are layers of sand and gravel, and large boulders are sometimes struck. But always a harder layer of clay known to the driller as "hard-pan" is found at the bottom of the clay, and then beyond this is the water-bearing sandstone. This bot- tom "hard-pan" is the part of the drift which is next to the underlying rock, and is always passed through in drilling or digging wells either in the Red River Valley or west of it wherever the drift lies upon the surface. Several wells in the vicinity of Mayville and Blanchard, and northward, range in depth from 300 to 400 feet, water being obtained from white sandstone just below the hard-pan. Mineral Substances in the Water of Artesian Wells. The water from artesian wells in North Dakota generally contains some mineral matter. That from some wells is very salt, that is, it contains the kind of salt we use in our food, or "common salt." . Other wells contain a greater amount of salt, but do not taste "salty" or like brine; they contain other kinds of salt. Some of these give a bitter taste to the water, and some- times the water is a bitter brine, and still other salts give a sparkling and pleasant taste such as those of "hard" waters in limestone regions. Hard limestone waters are sometimes called "pure" because of their clear and sparkling character. Such waters are far from pure though they may be good for drinking and general uses. Some wells furnish water that is soft so that it is good for washing. It is "soft" because it does not contain those salts which make it "hard," though it may contain more salts of other kinds. The water from the deep well at Devils Lake contains seven times as much common salt (Sodium Chlo- ride) and three and a half times as much Glauber's salt (Sodium Sul- phate) as the water from the Jamestown well, and yet the water from the Devils Lake well is called "soft." THE WATER SUPPLY. 263 The salts which are in the lake water of Devils Lake are much the same as those which are in the water of the artesian well at Devils Lake city. The waters, therefore, which soak into the ground and become the source of the artesian well 600 or 700 miles away dissolve the salts from the rocks through which they pass, in a similar manner as the rains falling upon the ground dissolve the salts from the soil and carry them into the lake. It should be remembered that the rocks in North Dakota, the Cre- taceous formations, were deposited in a great inland sea or ocean, and ocean waters are always salt. Our artesian waters are, therefore, much like the sea water of the ancient oceans. The water from all the wells is not the same because it does not all pass through the same kind of rock. Wells, therefore, in different localities furnish water differing in quality. Different rock layers con- tain some more and some less of a certain kind of salt. The rain water soaking into the ground and passing slowly through it dissolves out different kinds of salt. Another Class of Artesian Wells. There is another class of artesian wells in the Red River Valley in which the source of the water supply is probably not the same as that of the deep wells west of the valley and the wells which yield salt water in the valley. They are obtained at depths of even less than forty feet and from this up to 250 feet. Some- times within a distance of only a few rods flowing wells are obtained at depths varying greatly, and the water in the wells of this class is gen- Lt-oel o$La.li -V-..".'i'. : 'r.'>;: : : ; ''ic < -:t S^S /?-:% . 'SAND, - PIG. 144. Diagram indicating the probable Relationship of Sources of Artesian Water at Grandin. U. S. Geological Survey. and sand to lower levels in the Red River Valley. Beds of gravel and sand serve as underground water courses much like stream beds on the surface. In Figure 80 four artesian wells located at Grandin, Cass County, are represented which have depths of 105 feet, 158 feet, 187 feet and 248 feet. These wells are only a few rods apart, and the water from them is fresh, containing but little of any kind of salt. It is good for drinking or for any purpose. The water of these wells, as of many other shallow artesian wells in the Reel River Valley, has probably come from the higher land west of the valley not so very far away where the soil is sandy or gravelly, and it has followed the loose layers of sand and gravel which extend between beds of clay, down to the THE WATER SUPPLY. 267 lower valley, and thus a head is given which causes the flow when these "veins" are tapped by the drill. These veins are probably long, nar- row beds of gravel or sand, but in some parts of the valley wells are obtained at about the same depths, showing that the beds are in some places not long, narrow strips of gravel or sand, but wide sheets. Such shallow artesian wells yielding plenty of good water are found over a large part of the southern and eastern part of the Red River Valley south of Blanchard, and north to Crookston in Minnesota. The waters are fresh because they have not passed through salty rock layers for any great distance. Waters which flow underground for long distances and through different kinds of rocks which contain salts are salty or alkaline because there are many salts and alkaline sub- stances in the Cretaceous formations, and the waters in passing through these slowly dissolve the mineral substances from them. Common Wells. On the whole, North Dakota has an abundant sup- ply of good, wholesome water. Almost anywhere west of the Red River Valley, which means nearly the whole State west of the eastern tier of counties, the supply of water from surface wells is abundant and of good quality. The mantle of drift over the underlying shales is not so deep but that it is an easy matter to dig or drill through it, and plenty of water is generally obtained as soon as the overlying drift is passed through at depths of fifteen to seventy-five feet. The writer thought on a hot afternoon that he had never tasted better water than the clear, sparkling liquid which he drew from an old-fashioned chain pump at a rancher's cottage in the hill country south of Dog Den Butte in McLean County, and many such wells dot the prairies in the great interior portion of the State. 268 THE STORY OF THE PRAIRIES. FIG. 145. Flowing Well, Just Struck, Trott Farm, Casselton Quadrangle. (Depth 418 feet. Flow at first, 4,000 barrels per day. Section 10, Township 140, Range 53.) FIG. 146. Old Farm Machinery Buried by Sand Thrown Out of Budke Artesian Well, South of Wheatland. CHAPTER THE TWENTY-FIFTH. A STUDY OF THE SOILS. What is soil? How have our soils been formed? What is the dif- ference between good soil and poor soil? What is it that makes one farm worth more than, another only a short distance away? Why are some soils naturally rich and productive while others are in their natural condition poor and can be made to produce crops only by great care in cultivation and often by considerable expense in fertilizing? These are questions of vital importance to the practical farmer. They are all questions that he can answer, and upon his answer to them may depend much of his success in farming. A farmer does not have to be a labeled geologist in order to under- stand something of the geology of soils. Indeed he may think that he does not understand geology at all, but still think that he knows some- thing about soils, their structure, texture, and qualities, good or bad. It may very likely be the case that he does know a good deal about soils, and in this knowledge he probably possesses more skill in geologic observation than he has supposed, for the knowledge of soils that has most to do with their productiveness, and hence with their value, is, after all, principally geologic knowledge. We may, therefore, address ourselves to the study of the geology of soils with the understanding that this study will deal with those facts and principles that have to do with the character of soils in their rela- tion to the growth of plants. The structure, texture, and composition of soils are three important considerations. Upon these depend the fer- tility and productiveness of the land. No one of these considerations can be unfavorable and still have the best soil. It will be seen that these are all geological considerations, as they are all the result of geologic processes. The best method of treatment of any soil is, there- fore, often revealed through a study of the geological factors concerned in the production of the soil. Definition What then is soil? Geologically it is rock. Prac- tically it is material in which plants will grow. Let us look at a hand- is 269 270 THE STORY OF THE PRAIRIES. ful of soil from a field where crops have been grown for many years. It is sometimes spoken of as dirt, rich dirt, poor dirt, black dirt, yellow dirt, brown dirt, etc. Then it may be heavy or light, hard or mellow, cold or quick. These are all terms that are in common use in describing the qualities of soils. A little study of a sample of soil from the field will show that these qualities are all determined by the geologic con- ditions under which the soil was formed. Before inquiring as to the causes by which the soil has come to be what it is let us note carefully what this sample contains, or in other words, let us see of what it is composed. As soon as it is examined closely it is seen that a large part of it consists of small particles of stone, tiny rock fragments, that are just like large boulders that we have seen except in the matter of size. Here are many bits that look like particles of broken glass. Then there are other particles that are black or dark, some rounded in form, others flat like thin scales. If the par- ticles are examined with a microscope or hand magnifying glass they will appear much like the boulders that lie in the fields, only small. If a hundred small bits are picked out and compared with the same num- ber of pebbles from a gravel or sand-pit they will look much like these except that the soil grains are smaller. Put a handful of soil in a tumbler of water and stir it up. What happens? In a few i minutes most of the soil has settled to the bottom. The water, however, remains roilly for some time. Let it stand 24 hours and then carefully pour off this water into another dish. Put in more clean water and stir again. Allow to settle and pour off as before. The same result follows. Repeat the process several times. After the soil has been thus thoroughly rinsed what remains ? Just such particles as those that were picked out at first. This simple experiment shows that the greater part of the soil is made up of small bits of rock. Now take all the "dirty" water that was saved from the washing and let it stand several hours in a glass tumbler or bottle. What finally gathers on the bottom? Fine mud. Now put a little of this mud on a glass and look at it with a good microscope. What is the mud made of? If this experiment has been conducted carefully and all the materials that settled to the bottom are weighed, it will be found that the amount of material that has been carried off in the water is indeed very small. A STUDY OF THE SOILS. 271 What then do you conclude that the soil, as you see it in the field, is principally made of? From this simple study of a handful of soil it appears that > the greater part of the soil is small particles of rock. And we may now ask again what is the difference in soils that makes some worth so much more than others, if after all the soil is largely made up of small bits of rock? A little further study of soils in the field will help to make this matter clear. Every one has seen gravelly soil, and sandy soil, and soil that is neither gravelly nor sandy but just good every day soil. Such a sample was that we studied when the experiment was made. Then there are also soils that are heavy and compact, called clayey soils. Of these classes of soils those that are gravelly or sandy are not considered the best for farming pursuits. They are too much affected by drought, among other things. The clayey soil is too compact, becoming hard in the sun, and unsuited to the growth of plants. In one case the rock particles of which the soil is composed are too coarse, and in the other they are too fine. The one lets the water soak in and disappear too rapidly; the other holds the water too long. The latter class of soils is too heavy; the former class is too light. Just the right mixture of the clayey and sandy materials would make the best soil, and this in fact is the most valuable class of soils in the world, the mixture of clay and sand, called loam. Origin of Soils Since the small particles of rock of which the soils are so largely composed are seen to be just like the larger stones in the field except that they differ in size, it may very naturally be asked where these tiny bits of rock have come from that make up the soils of the field. This is an important question, and in order to understand it, it will be necessary to know something of the processes by which soils have been formed. This means that it will be needful to inquire how the rocks become broken up into small fragments. We may consider now two ways by which the rocks are broken up. One is by the actiori of frost, wind, heat, and rain. This is called the process of weathering. The other is that of mechanical breaking, in which the rocks are broken by some mechanical force that strikes one rock against another, or some substance strikes the rocks so as to break them. Both these processes have been at work in forming the soils of our prairies and fields. The weathering of rocks is going on all the 272 THE STORY OF THE PRAIRIES. time all around us, slowly it may be, but it is going on. New soil is being formed all the time from the rocks. All the streams are working away carrying fine particles of soil down stream from the lands toward the sea. The finer parts of the soil are continually being carried away, and if it were not that more soil is being formed to take the place of that which is carried away the fields would by and by become very thinly covered with soil, and in time there would be only bare rock left. The mechanical processes, by which rocks are broken, crushed, and FIG. 147. Clay Butte, Capped with Sandstone. Yields soil containing stones. ^Township 149, Range 96.) Photograph by Rex Willard. ground to fine powder, are closely related to rock weathering, and are among the most important soil forming processes. One of the most stupendous things that has ever occurred to chan'ge the face of the landscape is that known as glaciation, or the movement of a great ice sheet over the land, by which rocks were broken, crushed, and ground to fine powder. At the same time they were often carried long dis- tances and so were mixed and stirred, so that when the ice finally melted the material was left in piles large or small, and scattered over large areas. By this process a great variety of soils was deposited upon the A STUDY OF THE SOILS. 273 landscape, and this transported material, broken, crushed, and ground up to all degrees of fineness frorii large boulders to fine sand and clay, now makes up the great body of soil of many of our northern states. This work done by ice in fashioning and changing the landscape and modifying its soils belongs to that part of the earth's history which is known as the Glacial Period. The work that was done by the ice during this period, and the forms in which the landscape was left after the ice had melted, has been described in the earlier chapters of this book. Re-read chapters three and five, having in mind the effect of the FIG. 148. Clay Butte. Yields soil free from stones. (Township 149, Range 96.) Photograph by Rex Willard. things described upon the soil, and you will probably be convinced that this has been one of the greatest factors in making North Dakota soils what they are. Kinds of Rock from which Soils are Derived From what has been said it will be apparent that the particular kind of soil is determ- ined in considerable measure by the kind of rock that was weathered or mechanically broken up to form the soil. The chemist finds that soils differ very much in composition, or the kind of substances of which they are composed. For example, limestone rock when weathered or 274 ,THE STORY OF THE PRAIRIES. broken and ground up will form a soil which will be composed of very different substances from soil formed by weathering or grinding up of shale, or of quartz rock. And since these kinds of rock differ very greatly in hardness the texture of the soil, by which is meant the coarseness or fineness of the grain, will differ very much. Soil formed from limestone will be very different from soil formed from shale or quartz rock not only in the sizes of the little grains or particles of rock of which it is made up but it will differ also in chemical composition. So also soil formed from shale will differ from that formed from lime- stone or quartz, and soil formed from quartz will differ from the others FIG. 149. Morainic Lake Filling with Vegetable Matter. (Muskrat house in center.) both in texture and in chemical composition. And since both the chemi- cal composition and the texture of the soil makes a difference in the fertility, or in the value of the soil for plant growth, the amount of each kind of rock that entered into its formation becomes a very important matter. Thus we shall find ourselves driven back to the geology of the orig- inal rocks from which our soils have been formed in order to find out their nature. In the Seventh and Twentieth chapters something is said about the ages of the rocks in North Dakota, and the conditions under which these rocks were formed. Not all limestones, nor all shales, nor A STUDY OF THE SOILS. 275 all quartzites, are alike in their manner of formation, and all are not made of the same substances. The soils that are formed from the weathering or the mechanical breaking and grinding of these rocks therefore are not the same. Whether a particular farm is good for the raising of wheat, or whether or not the soil contains alkali, may depend upon the conditions that existed in the ancient seas upon the bottom of which these rocks were originally laid down as sediments. The materials that lie at or near the surface in North Dakota east of the Missouri River are often spoken of as drift. The shale or clay or sandstone that lies beneath the drift is called the underlying rock. The story of the drift has been told in earlier pages of this book. The history of the underlying rock also has been given in Chapters Seven and Twenty. The drift has been explained to have been derived largely from the underlying rock in the vicinity where the drift occurs. This means that the drift is rock that has been mechanically broken, pulverized, and mixed by the action of the great ice sheet. Soluble Salts or Alkali Let us first consider the materials of the soil which may be dissolved in water. These are often spoken of as the soluble salts of the soil. They are also frequently called "alkali." The subject of alkaline waters in lakes, and the origin of the alkali in the rocks of the ancient sea bottoms, is explained elsewhere in this book. Probably all the soils in North Dakota contain alkali. In fact, while many farmers are afraid of 'alkali' land as not being productive, still these mineral salts are necessary for the food of plants, and it is only when they are present in too great quantities that they become injurious to growing crops. The best methods of cultivation for alkali lands, and the cause of there being too much alkali in the soil, will be considered later when the different types of soil are considered. It is important now to understand what is the source of the alkali, and the natural relation of these mineral salts to the soil. Since the alkalies have been derived from the waters of the ancient seas upon the bottom of which the rocks, from which the soils have been formed, were deposited as muds or sediments, and since the character of the soil, as sandy, clayey, gravelly, etc., has been largely determined by the action of the great ice sheet in forming the drift over a great part of the land, it fol- lows that the study of the alkalies in the soils takes us into a study of the origin of the soils and the processes by which the particular soil type has arisen. The cure for a disease is often most certainly found out by 276 THE STORY OF THE PRAIRIES. learning the cause of the disease. When we know the causes that have led to alkali being present in the land we are on the right road to find a way to overcome the trouble, if indeed there is any way to overcome it. Organic Matter in Soil The organic matter in soils is one of the very important considerations relating to soil fertility. Organic matter in the soil may be described as partially decayed remains of plant and animal matter. If a boulder of granite be broken up and ground to fine powder and wet with rain water it will be found that plants will FIG. 150. Soil Survey Party in Camp. Photograph by The Author. grow in this soil. This would be soil formed by mechanical breaking and grinding in very fact. But if some kind of decayed matter, such as old leaves or rotted green plants, are mixed in the soil this adds greatly to the fertility of the soil. It should be borne in mind that substances in the earth that dis- solve in water will rise from considerable depths due to the process of capillarity. In other words as the rain waters soak into the earth the salts become dissolved in the water, and as the surface of the ground becomes dry the water rises from below, bringing the salts with it. A STUDY OF THE SOILS 277 Then the water dries off from the surface by evaporation and the salts are left. This rising of the salts from the subsoils goes on in all our fields. It may not do any harm, because it may be washed away from the surface so that too much does not gather. On the other hand there may be too much washing of the surface, and not only the salts that rise from the deeper subsoils may be washed away, but the organic matter that comes from the decay of roots and leaves and stems, which add fertility to the soil, may also be washed away. The salts that come to the surface through the rising of the ground water and the or- ganic substances that gather from the growth and decay of living things are valuable ingredients of the soil. How a Soil Survey is Made It may be of interest to the reader to know how a soil map is made. The work of soil surveying, like any other investigation of a scientific nature, requires patience, perseverance, and skill. Soil surveying depends upon the judgment of the one doing the work rather than upon the accurate use of ipechanical instru- ments, as in land surveying. The instruments that the soil surveyor finds the greatest use for are his eyes and his fingers, in other words the skillful use of his senses of sight and feeling. To be sure he uses instruments for certain parts of the work, but what he seeks to do is to use his senses in such manner that he can form a correct judgment of the character of the soil as to its structure, texture, and quality. Making a Soil Map The soil map is made by representing the different soil types, in colors, using one color for each type of soil. A map drawn on a large scale is used, and the colors for each type are ap- plied in the field section by section as the work progresses. . Samples of the soils and subsoils from different parts of the area are sent to Washington to be analyzed in the laboratories of the Bureau of Soils. Several samples are taken of each soil type, and these are analyzed with much care by very delicate methods. The analyses show accurately just the amount of clay, silt, sand of different grades of fineness, the amount of organic matter, and the chemical constituents. These things can be determined in the field by an experienced man with a fair degree of accuracy, but the analysis in the laboratory shows exactly the amount of each kind of material in the sample. The soil' map when completed shows every section of land on the area, and the kind of soil and subsoil on every part of each section 278 THE STORY OF THE PRAIRIES. down to areas as small as 10 acres. This map is sent to Washington along with the written report of the surveyor who had charge of the work. The report gives a full and detailed description, of each soil type and the agricultural conditions. The written report and the col- ored map and the samples are all carefully gone over in Washington by members of the staff of soil experts, and then the final report is ready for publication. This is the method that has been used by the men who have made the several soil surveys in North Dakota. Reports containing the col- ored maps, the results of the analyses of the various soil samples, and the descriptions of the soils, and of the agricultural conditions, have been published and distributed to the public. These surveys made in different parts of the state have made it possible to classify the soils of the state, and have formed the basis for the soil map which accompa- nies this volume. The Soil Regions The soils of North Dakota may be classified in a general way by the character of the regions where they occur. There are three great soil regions in the state, and each differs in import- ant particulars from the others because of different conditions that have prevailed in each region. These three regions are the level lake bottom lands, of which a good example is the Red River Valley ; the rolling prairie region which has been transformed and fashioned to a large extent by the action, of the great ice sheet, in which region is included the large basins of the Sheyenne and James rivers; and the regions where the land has not been affected by glaciation, west of the Missouri River, and where the high plain has long been subject to the agencies of weathering and erosion, and the land dissected by streams. If these three regions be compared in a little further detail it will be observed that the geological conditions which have prevailed in each region have been very different, and the present soil conditions in each of the three regions are very different as a result. The territory lying west of the Missouri River gives some suggestion of what the character of the landscape would have been over all the state if the invasion of ice of the Glacial Period had not occurred. The conditions today in the Red River Valley and the Mouse River Valley are what they are because when the ice of the great continental ice sheet began to melt away great floods of water accumulated in the lower places forming large lakes, and the sediment that was borne into these bodies of water by streams A STUDY OF THE SOILS. 279 from the great melting mass was assorted and arranged upon their bot- toms, and the present soils of the level prairies where were once the still deep waters are the result. Now, each of these three great regions includes smaller regions that have soils peculiar to the special conditions that exist in these particular places. It will be seen presently when the different classes and types of soil are considered that certain types occur wherever a given set of con- ditions existed. These conditions have been studied elsewhere in this FIG. 151. Sun Cracks in Muddy Bottom of Missouri River, near Williston. Photograph by Rex Wtilard. book in connection with the geologic history of the different regions. Now it will be the soils as they exist today that will claim our attention. Soils of the Glacial Lakes The soils formed from sediments de- posited on the bottoms of ancient lakes, such as Glacial Lakes Agassiz, Souris, Sargent, and Dakota, constitute what is here called the lacustrine or Lake Agassiz series. These soils fall into two general groups, viz., those that are finer in texture, which were deposited in the central deeper still waters of the ancient lakes, and those which are coarser in texture that were deposited nearer shore and were washed and assorted by the 280 THE STORY OF THE PRAIRIES. waves and off-shore currents. These groups of soils are represented on the accompanying soil map, and embrace several clay loam types and several sandy or gravelly loam types. Just what is meant by the terms loam, clay loam, sand, sandy loam, etc., will be explained later when the various soil types represented on the map are considered. It will be observed on the soil map that the soil types in these regions lie in tracts that are longer north and south and comparatively narrow east and west. The regions traversed by the Red River of the North and the Mouse River are the axial portions of the ancient lake bottoms FIG. 152. Wind-Blown Sand, Bottoms Missouri River, near Williston. Photograph by Rex Wtllard. across which these streams now sluggishly meander. The parts of the great glacial lakes Agassiz and Souris that lay in the United States had their longer axes in a north-south direction. The deepest portions of the lakes were in these axial regions. The shores were in a general way parallel with these 1 deeper axial lines where now extend the sluggish Red and Mouse Rivers. In these deepest central parts of those great lakes the finest sediments were deposited, and in these regions as a result now occur the heavy clay and clay loam soils. Farther toward the shores the sediments deposited were slightly coarser, but they were A STUDY OF THE SOILS. 281 quite uniformly fine-grained sediments consisting mostly of such fine rock powder as silt or clay. At the margins of these great bodies of water, however, materials much coarser in character were deposited, and the change from the one to the other is often, though not always, quite abrupt. Here were deposited the coarser sands and gravels brought down by the streams or thrown into the lakes directly from the melting glacier after these materials had been washed and assorted by the waves and currents. It thus comes about that the several soil types are distributed in belts running north and south. FIG. 153. Four Sisters, Each Holding a "Claim," Williams County. Photograph by H. V. Hibburd. The division df all the soil types of the lacustrine or Lake Agassiz series into two groups, those formed from the deep water deposits, called the clay loam group, and those formed from the shore deposits, called the sandy loam group, is therefore a natural one. The division is, however, somewhat arbitrary, since it is often impossible to tell just where the one begins and the other ends. In the former group are included three principal soil types, as deter- mined by the United States Bureau of Soils, two of which are classi- fied as clays and one as clay loam. The latter group includes seven 282 THE STORY OF THE PRAIRIES. different soil types, as they have been mapped in critical soil surveys, but these are all included in the grades between loam and sanely or gravelly loam, and the shore soils may therefore for geneial purposes be represented by two forms, viz., loam and sandy loam. It will thus be seen that the soil types in the lacustrine or Lake Agassiz series embrace soils ranging in texture from clay, in the axial or central deep water portion of the lake bottoms through clay loam, loam, sandy loam, to gravelly loam, and that these represent no less than ten different soil types when critically analyzed, but that these may FIG. 154. The Four "Claims" of the Four Sisters, Williams County. Photograph by H. Y. Hibbard. be divided into two general groups, the clay loam group on the one hand and the sandy loam group on the other. Soils of the Rolling Prairies In the great rolling prairie district several types of soil occur, but they fall mostly into two series. Three or four phases of the landscape include all the principal soil types. ( i ) The more roughly rolling hills or moraines, (2) the broadly rolling prairie with gentle swells and extensive nearly level tracts, and (3) the places where running waters (mostly from the melting ice sheet) have modified the soils by washing and re-depositing them, represent the A STUDY OF THE SOILS. 283 most important phases of the landscape that find expression in the different soil types. Lowland tracts occur also among the hills of the rolling prairie, and these often constitute distinct soil types. One of the principal series of soils in the rolling prairie portion of North Dakota is known as the Marshall series. The Marshall series includes a number of different types, in fact there is almost a complete series from the clay types to coarse sand and gravel represented in the rolling prairies of North Dakota. In the morainic hilly regions there is commonly either Marshall stony loam or Marshall sandy or gravelly loam. One of the most widely distributed and most valuable soil types in the state is that known as Marshall loam. Marshall silt loam is one that occurs also over quite large areas, and this is also a most excellent soil. Some of the best land for cereal grains, such as wheat, oats, bar- ley, and the like, is Marshall silt loam. Marshall loam is an excellent soil for general farming purposes. It differs from Marshall silt loam chiefly in having a little larger percentage of sand and a smaller per- centage of silt. Marshall clay loam and Marshall clay are two types that occur, but not over as wide areas as the loam and silt loam just men- tioned. Clay loam and clay are heavier soils, and generally require arti- ficial drainage to make the lands most fully productive. All the types of the Marshall series are soils that have been formed directly from material that has been transported or ploughed up and ground to powder by the great ice sheet. The soils of the terminal moraines and of the ground moraines therefore generally belong to the Marshall series. The soils that have been deposited from running water along the stream courses are not classed in the Marshall series. The conditions under which they have been formed are quite different from those of the Marshall series, and so the resulting structure of the soil is* quite different. The Marshall series consists of soils that have been formed by the action of the sun, rain, wind, frost, and other natural agencies, from the materials that were left when the great ice sheet melted. The Sioux series includes those soils that have been formed from materials of the drift that have been worked over by running water from the melting ice. The soils of the Sioux series are generally dark colored, with a sub- soil of gravel or sand. These soils occur in the bottoms of valleys like the Sheyenne, the James, the Pipestem, the Des Lacs, the upper Mouse, 284 THE STORY OF THE PRAIRIES. and along many ancient channels that are tributary to the Missouri River from the east. The Clyde series is another important series of prairie soils. These consist of soils that are of glacial origin, but they differ from those of the Marshall series in that, while they have been formed from materials left by the melting of the great ice sheet, they have been modified by the action of the ice waters. They are distinguished from the Sioux series also in that, while they have been modified by the action of the glacial waters, they have not been deposited along stream channels but have been re-worked by glacial flood waters and deposited in ponds or lakes of still waters. The soils of this series are often dark in color from the large amount of organic matter that they contain. The Sioux soils are generally well drained naturally by reason of the fact that they are underlain by coarse gravel or sand as a subsoil, while the subsoil of the Clyde series is generally a quite compact clay, gray, yellow, or mottled in color. The Clyde soils are therefore often heavy in texture, and re- quire artificial drainage to make them most fully productive. These soils occur in the low areas among the rolling hills and broad swells of the ground moraine. Soils West of the Missouri River. West of the Missouri River a different set of soil conditions exists from those of any other part of the state. Here the great ice sheet did not go. No such great change has been wrought upon the landscape as that which has affected all the land east of the river. Upon any good map of the state it will be observed that there are more streams than in any other region of the state, and no lakes. East of the river are comparatively few streams and very many lakes. Any one who has traveled in that part of the state has observed that the appearance of the whole landscape is differ- ent. The hills are different in form and different in structure, that is, they are different in the way they are made and in the materials they are made of. Here the hills are flat topped, or may be craggy and rough from projecting rough rocks, while east of the river they are rounded in form, and often strewn with boulders of many different kinds of rock. West of the river there are few glacial boulders. ( a ) When the hills are strewn with rock or capped with crags the rock is always of the same kind as that of which the hills are composed. The boulders in the (a) This statement is a general one. See "The Older and Newer Drift," Chapter Twenty-one. A STUDY OF THE SOILS. 285 rolling country east of the river are of different kinds, and none of them are like the rock of which the hills are composed. The boulders are of glacial origin, or drift boulders. There is no drift after passing 40 to 50 miles west of the Missouri River, and the rocks that lie upon the surface are fragments that have been broken from the underlying rock which was originally formed there, and is the rock that lies underneath the soil of the whole region. The soils in these western counties are known as residual soils as distinguished from the glacial or drift soils of the rolling prairie regions. The soils west of the Missouri River are called residual soils be- cause they are the result of the action of the sun, frost, wind, and rain upon the rocks of which the landscape was originally made. This is the meaning of residual as compared with drift, the drift soils being those that have "drifted" from some other place, or have been trans- ported by the great ice plough. The rock layers of which the landscape west of the Missouri River is made are mostly sandstones, shales and clays. Under the action of the weathering processes these break up into sand and clay, some harder portions remaining as "rocks" or "stones" lying upon the surface or in the soil. If the sandstone rock in any region is very hard there are apt to be many fragments or hard pieces lying upon the surface, and this is called stony land. In such a region of hard rock there are apt to be also hills capped with hard rough crags of sandstone. Many times also there will be flat-topped hills or buttes, a hard layer of rock serving as a cap or protecting cover. If, however, the sandstone in a .region is easily broken up, becoming sand under the action of the weathering influences, then the hills are generally less high and rugged and the sides of the hills less steep. It was stated before that the rock layers making up the hills are mostly sandstone, shale, and clay, and that shale breaks up into clay in the process of forming soil. A mixture of sand and clay has been called loam. This forms the basis for the best soils. The quality of the soil therefore depends in large measure upon the kinds and amounts of rock that have entered into the formation of the soil. There are places where the proportion of clay is too lax'ge and that of sand is too little, with the result that the soil is a heavy clay soil. This may be a clay loam or a clay. Many so-called "gumbo" spots are 19 286 THE STORY OF THE PRAIRIES. places where there is not enough sand to "lighten" the soil. This is not the only cause of the "gumbo" spots, but it is one of them. Shales generally contain soluble mineral substances known as salts. These are popularly known as "alkali." Some of these salts are really alkaline in their nature, but not all of them. They are all of value, however, in the soil if they are not present in too large quantities. It has been stated before that alkali is a valuable ingredient of soil. "Gumbo" lands are generally lands that are not naturally well drained, so that the alkalies and other salts accumulate owing to the fact that too little water runs away from the land to carry away the salts. Clayey soils that contain considerable alkali are generally slippery and sticky when wet, and the land bakes into hard blocks when the sun shines upon it in summer after it has been much wet earlier in the spring. Often "gumbo" lands can be recognized on slopes by the fre- quent occurrence of peculiar little spots that look as though the soil had been dug up or pawed by animals. These spots are sometimes thought to have been formed by the pawing of buffaloes in the days when these animals roamed in herds over the prairies. These spots are really little landslides where the soil has slipped when wet because of the slippery greasy-like character of the soils when saturated with water. "Gumbo" lands generally in this region owe their peculiar quality to the presence of alkali in the soil. The most widely distributed soil on the rolling lands west of the Missouri River is loam. In many cases where the soils have been studied clay subsoil has been found. This affords the best condition for general farming. A loam soil with gravelly or sandy subsoil is likely to suffer from drought where the same soil underlain with a clay subsoil with the same rainfall would be profitable farming land. There are four phases of landscape west of the Missouri River with corresponding soil features. These four kinds of landscape are : first, the rolling lands, which are naturally good grass lands, the soil of which is loam; second, the low lands in the hollows among the rolling hills, usually in their native condition good grass lands, the soil being often more clayey; third, the stone lands of the high hills and ridges; and fourth, the lands in the larger stream valleys, in which the soils range from clayey to fine sandy, coarse sandy, and gravelly, the soils representing flood deposits of streams. A STUDY OF THE SOILS. 287 Classification of Soils. The classification of soils is ? based pri- marily upon the physical properties, but all factors that influence the relation of soils to crops are taken into consideration, so far as they can be determined. Three factors that are of great importance in soil classification are texture, the structure, and the organic matter con- tained in the soil. By texture is meant the coarseness or fineness of the particles of rock making up the soil. By structure is meant the arrange- ment of the materials. The organic matter is that which comes from decaying vegetable or animal matter. Then the factors of origin, depth, FIG. 155. One of the First Occupants of the Soil. color, natural fertility, and the topography and natural drainage of the land, all enter into the determination of the character of the soil. The Soil Classes. Soils are divided into classes on the basis of their texture, that is of the relative amounts of finer and coarser parti- cles of rock of which they are composed. Thus there may be clay, clay loam, loam, sandy loam, sand, gravelly loam, and gravel, according to the amounts of materials of different grades of fineness. The difference between a clay soil and a sandy soil is one of the size of the particles only, and the specific determination of the type will depend upon the relative amounts of the materials of different degrees of fineness. Wheat- 288 THE STORY OF THE PRAIRIES. land sand, for instance, contains a large percentage of small bits of rock of the size of sand grains. Marshall clay contains a large percentage of clay with very little of sand or gravel. In addition to the fine particles of earth of which soils are com- posed there may be also large particles, such as smaller or larger peb- bles and stone fragments. If these fragments are small in size yet larger than sand grains they are called "gravel;" if they are too larjge to be called gravel then they are known as "stones." Thus there may be stony loam, gravelly loam, sandy loam, stony clay, gravelly clay, sandy clay, etc. A further study of the different classes of soils will be made later in connection with the consideration of the various kinds of soil in North Dakota. The Soil Series Soils are arranged in series according to the manner in which they have been formed, and by conditions of topo- graphy, drainage, etc. The soils in any region may include all the classes, but may belong to one series, because all were formed by the same general process. For example, all the soils may be of glacial origin, or all might have been formed from sediments deposited in a large body of water. In another locality there may be the same classes but belonging to different series, because they have been formed by dif- ferent processes. For example, in the western part of North Dakota there may be seen the residual soil, or those formed from the rocks of the region where the soils occur ; there are glacial soils, or those that have been ground and transported by the great ice sheet; and there may also be the soils that result from the floods of the Missouri River, or stream deposited soils; and each of these series may be represented by any or all the different classes of soils that have been referred to. Upon the rolling prairies of North Dakota there occur Marshall stony loam, Marshall gravelly loam, Marshall fine sandy loam, Marshall loam, Marshall silt loam, Marshall clay loam, and Marshall clay. These different soil types have all had the same general history as to the pro- cesses of their formation, and they are therefore all classified in one series. Another series known as the Clyde series differs from the Marshall series in that, while the soils are of glacial or drift origin, they have had a different history, or have been formed by different processes from the Marshall series, and this makes often a very marked difference in the quality of the soil. There may be the same classes of A STUDY OF THE SOILS. 289 the Clyde series, as Clyde stony loam, Clyde gravelly loam, etc. So also the Sioux series is another series that occurs in North Dakota, and this too has been through a different set of conditions from any of the others. The gravelly and sandy loams and clays of the Sioux series have been formed by the action of glacial waters in streams. Thus while there may be a sandy loam in any one of several different series, and the texture of the soil might be about the same in all, yet the structure and the resulting quality of the soil will be different. The Soil Map of the State Several soil surveys have been made in the state, and these have been upon quite widely separated areas so as to make as representative a study of the soils of the state as possible in the shortest time. By taking the soil maps of these areas it is possi- ble to construct a map of the whole state with a fair degree of accuracy. Two areas have been surveyed in the Red River Valley, one embrac- ing a portion of Grand Forks County, and the other a portion of Cass County. These two surveys give a fairly representative study of the soils of the Red River Valley. The Ransom County area, consisting of twenty-four townships, embraces a portion of the Sheyenne delta and the lower Sheyenne Valley, and the rolling prairie adjoining the Red River Valley on the west. The Jamestown area extends from the Sheyenne Valley at Valley City westward ten miles beyond James- town, and includes the eastern edge of the Missouri Plateau. The Carrington-Cooperstown area includes two tiers of townships extending across Griggs and Foster Counties. This area extends from the Shey- enne Valley east of Cooperstown westward across the fine farming lands of these counties, crossing the James Valley and extending to the edge of the Missouri Plateau. The Cando area includes eight town- ships in southern Towner County. The Ransom County area, the Jamestown area, the Carrington-Cooperstown area, and the Cando' area thus represent the soil types of the great east central portion of the state. The Minot area is so located as to include the soils of the famous Mouse River Valley and the lands to the westward which were once the shore of Glacial Lake Souris. A preliminary survey of seventeen townships in Ward and McLean Counties, embracing a portion of the hill country south and west of Minot commonly known as the coteaus, or the Coteau du Missouri, and including also a portion of the Missouri 290 THE STORY OF THE PRAIRIES. slope to the westward, represents a large region in the west central portion of the state. The Williston area, including seventeen townships in southern Will- iams County, represents the upland glacial soils of the older or Kansan drift, and also includes most of the irrigable land in the Williston irri- gation project. A survey in the Cannon Ball district in Morton, Het- tinger and Adams Counties furnishes data on the soils of the great empire west of the Missouri River and south of the main line of the Northern Pacific railway. This area is so located that the soil types include all those likely to occur between the Missouri River on the east and the Little Missouri on the west. It will be recalled that the older or Kansan drift extends for some distance west of the Missouri River, the drift deposits being represented by scattering granite boulders and thin deposits of sand and gravel. The boulders become less numerous and the sand and gravel thinner westward till they disappear entirely in western Morton County. The soils are thus largely residual, having been derived from the weather- ing and erosion of the rocks of that region. The McKenzie County area is situated west of the Little Missouri River and about midway between the main lines of the Great Northern and Northern Pacific railways, and abuts the western boundary of the state. This area lies beyond the limits of the territory supposed to have been crossed by the Great Ice Sheet. The soils are therefore the residual soils formed from the weathering and erosion of the native rocks. Thus an aggregate area embracing about one hundred fifty town- ships has been examined in detail, and accurately mapped. This is 5,400 square miles, or approximately three and one-half million acres. From the data thus furnished it is possible to compile a soil map repre- senting the whole state. Such a map is designed to accompany this chapter. It is in preparation and will soon be published. Those who are interested in any particular section of the state will be able to make a very satisfactory study of the soils of any region by use of this map and the more detailed maps which represent the particular area described. Copies of these detailed soil maps and the accompanying reports describing the soils can be obtained, unless out of print, by applying to the director of the Agricultural College Survey, Fargo, N. D. CHAPTER THE TWENTY-SIXTH. MINERALS IN NORTH DAKOTA. The search for valuable minerals in the earth is as old as civiliza- tion itself, and is as young as the progressive, alert, and active age in which we live. Men have always searched for valuable or curious minerals in the earth and they probably always will. There is probably no land or people that has been free from the search and the searchers after valuable or curious minerals. There is no state in the union in which valuable, rare, or curious minerals have not been looked for, and in a very large number of states they have been found. In fact there are few if any states in which there is not some mineral of some kind found that is of value. As long as human nature is the same as it ever has been the search for hidden wealth will probably go on. North Dakota in common with other states and countries will have people who will search (in vain or otherwise) for hidden treasure. And the writer will not attempt to say that these searchers for hidden wealth are doomed to certain disap- pointment. He does not care to make any such statement as that there never can and never will be any valuable metal found deep in the earth in North Dakota, or that some valuable mineral that as yet has ' never been thought of may be buried beneath North Dakota soil. Things just as strange have happened. Geologists like other people make mis- takes. There is no evidence known at the present time that points to the probability that any great discovery of precious metals, rare gems, or valuable minerals other than those that are now known to occur (coal, underground water, clays, and building stone) will ever be found. But too little is known about this great earth of ours to make sweeping statements safe about what never can and never will be found. What rich deposits of gold, silver, copper, or iron; or valuable beds of rock salt, gypsum, or asphalt; or caverns of petroleum, or natural gas, may lie hidden from view beneath thousands of feet of rock it may be safely said we do not know. We may as well say frankly that we do not know, for we do not. It may be said that the odds are tremendously against 291 292 THE STORY OF THE PRAIRIES. the probability of any of these substances occurring in large amounts, but as a matter of known fact we know the rocks that lie beneath the surface in North Dakota only very superficially, and our inferred knowl- edge of them, gathered from regions surrounding, while this seems to give us good assurance for believing that there are no valuable min- erals hidden deep below our lands, yet this does not enable us to say we know. North Dakota is pre-eminently an agricultural state, that is, her chief sources of wealth lie in agricultural lines rather than in mineral resources. Nevertheless North Dakota is not without valuable re- sources of a mineral character. And in a state where there seems to be little to interest the mining prospector, there may yet be discoveries made that will prove to be of great importance and value. No one knew a few years ago what valuable clays lay hidden below the surface. Even yet we do not know what resources there are in our clays, because they have not been fully tested. So lignite coal is a mineral the full value of which has not been fully determined. Underground water and building stone are mineral substances of great importance, and belong in the catalogue of North Dakota's mineral resources. While we are seeking to develop the resources of the clays, coal, building stone, and underground water we may come upon something of great value, the presence of which has never before been suspected. The purpose of this chapter, however, is not to dwell upon the possi- bilities of future development of mineral wealth in North Dakota, nor to explain how the mineral resources that we know we possess may be best utilized. The object is rather to explain the nature and describe the appearance of some of the common minerals that are often mis- taken for something valuable, some of which would have commercial value if they occurred in large quantities, and upon which many times considerable money and time are spent in the mistaken notion that the mineral contains some precious metal or valuable gem. There are many mineral substances found in North Dakota which in appearance seem to resemble gold or other valuable metal, or which seem to resemble gems. Many specimens of such minerals are sent to the writer every year with the inquiry whether the sample contains anything valuable, or has value as a gem. A fe\v of these minerals will be described and something of their mode of occurrence in the earth given. A little knowledge of the geology of the mineral, in MINERALS IN NORTH DAKOTA. 293 connection with a description of its appearance, will often enable one to determine both the name of the mineral and whether it has any value. Among the minerals that occur in North Dakota that are most com- monly mistaken for the ores of the precious metals are pyrite (iron pyrites), gypsum, mica and other minerals occurring in granite rock, and lime crystals. These minerals occur in such varieties of form, and often have such bright, metallic, or shiny appearance that they are often mistaken for precious metals or gems by those who are not informed as to the forms in which very common minerals occur. Iron Minerals. Let us note some of the common minerals of iron. It should be understood that this is not an attempt to describe all the ores of iron, nor yet to describe all the forms in which iron minerals or ores occur in North Dakota, but rather to describe those that are most commonly found, and which are most likely to be mistaken for something else, or mistakenly supposed to be some form of precious metal. Pyrite, or Iron Pyrites Pyrite or iron pyrites, known to the chemist as sulphide of iron, or more strictly, disulphide of iron, is now and ever has been one of the greatest deceivers of the searchers after precious metals that the world has ever known. It has come to be known as "fool's gold" from the fact that it has been so many times mistaken for gold. Every year many specimens of this mineral are taken to chemist or assayer or geologist in the eager expectation that he will pronounce it the king of metals. Instead of this, however, it is one of the most worthless of substances in the manner in which it often occurs, and more than this it .is frequently a great nuisance inter- fering with the utilization of some other material. It occurs often in beautiful crystals that have a yellow color, and yellow suggests gold. It occurs also in massive form, and has many times a beautiful yellow color. It may be found in hard quartz rock; and gold occurs in hard quartz rock. Or it may be found in beautiful shiny yellow bits in sand or gravel beds. Placer gold occurs in such beds of sand and gravel. The conclusion is thus readily reached that this is probably gold. Pyrite occurs also too commonly and abundantly in connection with lignite coal. This is the substance that gives the sickening sulphurous odor to burning lignite. Many times it occurs in cracks or crevasses 294 THE STORY OF THE PRAIRIES. in coal seams in beautiful yellow crystals. It is a nuisance worse than worthless when it occurs in this manner. This same mineral is encountered in drilling artesian wells, and is often referred to by the driller as "iron ore." In this case it is not generally crystalline, and is very hard. Some times well or spring waters have a. distinctly sulphurous smell and taste derived from the water passing through rock in which pyrite is present. This mineral is also found in the form of nodules, the mineral hav- ing collected around some particle of shell, stone, or other substance which acted as a nucleus. When these nodules are freshly broken the mineral appears a bright and shiny yellow, and readily excites the imagination of the searcher for rare gems or precious metals who may not have learned the lesson of "fool's gold." Of this mineral it may with much truth be said "all that glitters is not gold." There is, however, a redeeming feature about pyrite, or fool's gold. It is commercially valuable as a source of certain chemical substances. Sulphur may be obtained from it by roasting the . mineral in a retort, the sulphur easily vaporizing and passing off into a collecting receptacle. The black slag that remains is an oxide of iron. It is not a common source for obtaining metallic iron owing to the difficulty of separating all the sulphur. It is, however, one of the sources from which the world's supply of sulphur is obtained. Pyrite or iron sulphide also has high importance in the arts as a source from which green vitriol, or copperas, iron sulphate, is obtained, and also in the manufacture of oil of vitriol, or sulphuric acid. These substances are of considerable com- mercial importance owing to their wide use in the arts, and fool's gold therefore has its value. Iron sulphate, or green vitriol, differs from iron sulphide, or pyrite, in that the latter has been oxidized to produce the former. Oxidizing is the chemist's term for burning. Sometimes this oxidized form, the sulphate, is seen in crevasses in outcropping seams of lignite coal where the pyrite has been exposed to the air and moisture. Under these condi- tions the sulphide oxidizes or burns to the sulphate. This may often be recognized on the walls of old or damp mines, and frequently in bins of lignite where it is stored in damp places. The sulphate appears as a nearly white powder, but if dissolved in water it is green. If the min- eral is allowed to crystallize out of the water the crystals will be green. This is the reason it is called green vitriol it looks like green glass. MINERALS IN NORTH DAKOTA. 295 The question is often asked by those who send specimens to be analyzed whether this material, pyrite, has any value. It may be answered that it does, if it occurs in sufficient quantity and where it can be handled. The process of manufacture is simple. The sulphur can be mostly driven off by very moderate heat, and saved without much trouble. The black residue has simply to be thrown into a broad flat heap and allowed to lie exposed to the weather. The valuable iron sul- phate which will slowly form will now leach out from the heap and may be obtained by collecting as it escapes. FIG. 156. Sandstone Concretions, near Schafer, McKenzie County. Photograph by A. L. Felloivs. It should be stated that true iron ore occurs in the western part of North Dakota in seams or beds in the sandstone. Sometimes it is in very hard layers which project out of the sides of the buttes. This ore should not be confused with pyrite. It might be commercially val- uable as a source of metallic iron, hence regarded as a true ore, if it were more accessible. It does not appear as though it would ever be of much 296 THE STORY OF THE PRAIRIES. commercial importance, as it occurs, so far as yet observed, in thin seams, and in regions quite inaccessible to railroads or means of man- ufacture. Samples of iron-bearing sand have also been taken in different places in North Dakota, and excitement has sometimes been aroused that these sands might represent some valuable deposit. But in all cases observed these samples have been from drift sands, and therefore are to be ex- plained as glacial in their mode of occurrence, and hence local and limited in amount. True magnetic ore has been observed in such sands, indicating that fragments of magnetite or other hard ore has been broken up by the ice and transported from some iron-bearing region to the north. Gold and Things Not Gold In "prospecting" for gold there are two considerations of importance, viz., the occurrence of gold (as dis- tinguished from minerals that may be mistaken for gold), and its occurrence in paying quantities. These are matters to be reckoned with anywhere, and they therefore apply in North Dakota. As has already been observed pyrite is sometimes mistaken for gold, and there are other minerals also that not infrequently deceive the inexperienced prospector. But gold has been found in North Dakota, and it is reasonably safe to say that more will probably be found. This is not prophesying that any profitable mines will ever be developed, 'however. Gold Gold occurs in some sands in North Dakota, as has been shown by scientific assays. The writer does not know of any instance where anything like a paying quantity was found, but still enough has been found to say "gold." The sands referred to were drift sands, and the gold was "free" gold, that is, it was in the form of small bits of yellow metal. Its occurrence in drift sands or gravels means, as in the case of the magnetic iron sands referred to, that the gold has been transported by the great ice sheet, and deposited with the other earth materials borne by the ice. Gold is heavy, and so when borne by the ice waters that escaped from the melting glacier it would naturally sink sooner than the lighter earth and stones. Whatever gold there may be therefore is not evenly distributed through all the gravels and sands of a stream bed where it may be found to occur. It would be collected somewhat into pockets and basins because of its weight, and if one of MINERALS IN NORTH DAKOTA. 297 these pockets happens to be discovered there may be, and sometimes is, enough gold found to create an excitement. Gold in Hard Boulders Gold is sometimes observed in a hard boulder rock on the prairie. If there is some gold there must be more, some one may reason, and so begin to dig and search for more. The most elementary knowledge of glacial geology should enable one to distinguish a "hardhead" or drift boulder, and once this is recognized it will be seen that the extent of this particular mine will be determined by the size of the boulder. This rock, like the gold-bearing sand, has been transported by the ice from some rock ledge far to the north in which gold occurs. There might not be another like this within many miles. However, the discovery of a small bit of gold in a boulder has more than once been the cause of a "gold excitement," and has been the innocent occasion of many claims being "staked out" all about the place where the great continental glacier permitted the boulder to come to rest. Gold in Shale Several "gold scares" have occurred in North Da- , kota, and no doubt in other states, by discoveries of gold in pockets or ' cavities in shale rock. The explanation of the occurrence of gold, if gold there be, must be looked for in the surroundings where the "find" is made, and from a study made upon the ground. However some sug- gestions as to the manner in which gold might occur under such condi- tions may be made. There is a very small trace of gold distributed very widely through nature. It is said by good authorities that all sea waters contain a trace of gold. If this has been the case through past geologic ages then there might have been a slight trace of gold in the sediments deposited upon the bottoms of the sea. The shales and sandstones of North Dakota are sea sediments. If there was therefore ever so slight an amount of gold deposited with the muds and sands on the bottoms of these seas then there must be a little gold in the rocks of North Dakota which lie underneath the drift. Wherever favorable pockets for the concentration of the heavy gold might happen to be there might be enough gold so that it could be recognized. This again might be enough to cause a "discovery" to be made. Several such "finds" have been made in North Dakota, and in some cases considerable time and labor have been spent in attempting to develop mining properties. 298 THE STORY OF THE PRAIRIES. Mica Mistaken for Gold. One of the most common mistakes made by those who see gold in every thing that is yellow is that of assuming small bits of bright yellow material to be gold just because it is yellow, and, as they think, looks like it. Again, the least knowledge of field geology or practical prospecting would show that bits of yellow mica have none of the properties that distinguish gold. In the first place gold is heavy, and the thin scales of mica are light, so much so that so simple a test as that of blowing with the breath would show them to be too light for gold. A further test by washing or "panning" would add proof to probability that the material is not gold. Gold is the last thing to be washed out of the pan in the panning process sim- ply because it is heavy. By washing any suspected sand in a common basin or dish it will be shown pretty conclusively whether there are any heavy grains of yellow 'metal. Then if there is still any doubt or ques- tion it may be further tested with a few drops of mercury. If the sus- pected yellow particles are really gold they will quickly disappear when touched by the mercury. Gold forms an amalgam with mercury, which is about the same thing as saying that gold dissolves in mercury. If the yellow bits are not affected by the mercury then they are probably either mica, or fool's gold or pyrite. Copper and Lead There are two metals besides iron and gold that have been found in North Dakota, and therefore that may be found again. If they are ever discovered in paying quantities it will be neces- sary to revise our supposed knowledge of the geology of the state .some- what, as this would be in the nature of an addition to present knowledge of the geology of the state. These metals are copper and lead, the former has been found in the form of small nuggets of the native metal in drift deposits. It is not known to occur in the stratified rocks of the state. Its explanation is therefore to be found in the same conditions that make it possible for iron or gold to be found in drift deposits. Lead has been found in the rocks of North Dakota in the region west of the Missouri River where the ice of the great glacier did not extend. Its occurrence is therefore a suggestion that more might be found. As has been stated before, it is not wise or safe to make sweeping statements about what never can and never will be found. Lead may prove to have been in the ancient sea waters, and so may occur in appreciable quan- tities in the rocks of North Dakota. A few years ago the presence of lead in the limestone rocks of Wisconsin was unsuspected. In fact it MINERALS IN NORTH DAKOTA. 299 had been supposed by those who were thought to know that no mineral of value would ever be discovered in the rocks of these prairies, since no volcanic outburst or mountain upheaval had ever occurred in the region, and it was thought that there was no possibility of any valuable ore being there. Nevertheless very valuable lead and zinc mines have been developed in the rocks underlying those prairies. Non-Metallic Minerals It seems fairly certain from the geologic structure of North Dakota that no metallic minerals of commercial im- portance lie hidden in the rocks. At any rate none have been discovered. FIG. 157. Stone School-House, Made of (Petrified) Wood, New England, Hettinger County. Photograph by A. L. Fellows. There are several minerals of non-metallic character, however, the more important of which have already been mentioned. There are a number of minerals that are quite widely distributed which would have con- siderable commercial importance if they occurred in sufficient quantity to pay to mine them. Some of these occur in the form of crystals and attract attention by reason of their appearance. Some of the common ones will be described. Lime Minerals. Among the common minerals that occur in crys- talline form are various kinds of lime or calcium minerals. These min- 300 THE STORY OF THE PRAIRIES. erals may occur also in the massive or uncrystalline form. They occur in the drift deposits and also in the stratified rocks that underlie the drift, and in the rocks west of the Missouri River where no drift occurs. They are not of much importance owing to the fact that they do not comprise large enough deposits to make them available for any practi- cal purpose. They are described here because they are so common and are so often sent to our laboratory for analysis. They attract attention* by their bright crystal faces and transparent character, and they sug- gest the quality of gems to those who are not familiar with the nature of such minerals. Gypsum Gypsum, or calcium sulphate, is perhaps one of the most common crystalline minerals in our native rocks, and it is one which attracts attention because of its transparent crystalline character and bright shiny surfaces. It occurs in such abundance in the rocks west of the Missouri River that the sides of naked buttes frequently present the appearance of gem-studded hills from the reflection of the sunlight from the glistening surfaces of these crystals. The shales that are exposed in the P'embina Mountain escarpment abound in these crystals. They are likely to be found wherever the shale rocks are exposed at the surface. So also they are frequently found in cavities in the clay of the drift where the clay has been formed from the grinding up of the shale by the passage of the great ice sheet over the landscape. This explains why they are frequently found in digging wells. Sometimes in such cases they are found in roughly roundish aggregations of crys- tals. This peculiar form has caused them to be supposed to grow in the earth after the manner of potatoes, the projecting prong-like crystals being thought to represent roots. In answering the question whether this mineral has any value it may only be said that it would have if enough could be obtained at one place to make it worth while to handle it. The mineral occurs very widely distributed in the rocks, but the crystals, which represent nearly pure mineral, are generally thinly distributed. They form in crevasses and cavities in the rocks from ground waters that percolate through the earth, the waters first dissolving the mineral and later, when condi- tions change, depositing it as crystal. The crystals when heated give off water and break down into a white powder which is known as Plaster of P'aris, which substance is much used in the arts. Gypsum is also MINERALS IN NORTH DAKOTA. 301 used as a fertilizer, being first ground to a fine powder, and is sold under the name of land plaster. Calcium Carbonate Calcium carbonate, a common form of which is limestone, is a mineral which is widely distributed in the drift of North Dakota, and while it is not one of the native rocks, in the strict sense, yet it has been transported within our borders in such quantity as to become an important element in our soils and underground waters. Some late observations indicate that there are limestone strata in some of the higher buttes or monadnocks west of the Missouri River, but they do not occur over a large area. Limestone boulders are so common in some sections of the state that they have been collected and burned in kilns for the lime they con- tain. The Lime Kiln buttes, southeast of Dickinson, are said to have gained their name from the burning of limestone rock from strata in those buttes. It is, however, as rock either for burning for lime or for building stone that limestone has played the most important role in North Dakota. It is widely distributed through our soils, and so> great is the amount that has been mixed with the drift that our well waters and the waters of most of our lakes and streams are hard from the presence of this mineral. Its wide distribution in the soil adds a valu- able ingredient to the soil also, as this mineral is important in the 20 FIG. 1573. An Agatized Stump, Morton County. Photograph by Rex Willard. 302 THE STORY OF THE PRAIRIES. growth of many plants. It is probably due in part to the limestone that has been brought into our state from Canada and ground to powder and mixed with the soil that North Dakota is noted as a wheat pro- ducing region. Calcium carbonate in the crystalline form known as Calcite is quite commonly found, and these crystals are not infrequently supposed to be gems of some kind by the finders, Iceland Spar and Dog'-tooth Spar are forms of the crystal of common occurrence. Calcium carbonate may be observed in almost any gravel pit, where it occurs as a white coating on boulders and pebbles, and often masses of gravel and sand are cemented into huge blocks of conglomerate by this mineral. Small pebbles are frequently found clinging tightly to larger boulders, cemented in like manner. It will be borne in mind that this is the mineral that forms the basis of the mortar used in the construction of walls of brick and stone. Quick lime is calcium carbonate after the carbon and oxygen have been driven off by heat, as in the process of burning in kilns. Mortar is formed by adding water, when the lime slakes and sets or hardens. As the mortar stands in the wall it slowly changes back to its original character as calcium carbonate. The cement that binds together masses of gravel and sand in gravel pits is natural mortar. CHAPTER THE TWENTY-SEVENTH. THE FUTURE OF NORTH DAKOTA. Geology a Practical Science. The geological features of any region of country have much to do with the industries of that region, and determine in large measure the value of the lands. A farmer may or may not understand the geology of his farm, but if he is a successful farmer he is controlled very largely in his methods of farming by those agricultural principles which have their explanation in geology. A farmer knows in a very practical way that certain crops do well upon certain soils, that others cannot be raised to advantage upon those soils. Oftentimes this knowledge is painfully practical, because he has learned it at the expense of much toil and labor. The best knowledge in the world is often gained by experience, that is, by experiment. It is because of this that experimental stations or laboratories have been established for the study of soils, and experiment in the best methods of cultivation of crops, looking toward a better knowledge of the natu- ral resources of the land. The laboratory of the landscape geologist is the field. % The farmer has to do with soils, rain and sunshine, and hence his laboratory is also the field. He must be a geologist. He may not know it; he may not believe in "science;" he may know nothing of geology as such, but he is a practical geologist, nevertheless. In buying a farm the geology of that farm is of much interest, not for the sake of the geology, but because this determines for all time certain points of value about the farm. A landscape is a more complex thing than is often thought, and more things enter into its character and so into the quality of the fields determining their use and value than is by many supposed. The Character of the Lands, That there is a large variety of types of landscape and a great diversity of soils in North Dakota is apparent from what has been said in the preceding pages. The diversity in kinds of soil as well as in forms of landscape adapts the State to a diversity of farming interests. Few states offer a greater range of opportunities 304 THE STORY OF THE PRAIRIES. for agricultural pursuits than does North Dakota. The richest wheat lands in the world, the most profitable flax and oat fields, ranges for grazing herds of horses, cattle and sheer^, of immense extent, vast areas of meadow, fuel in inexhaustible supply, clays suitable for brick for building purposes unlimited in extent, make any forecast except that of permanent prosperity, under the judicious management of an intelli- gent population, seem absurd. Let us briefly pass in review some of the things worth remember- ing about the landscape of North Dakota. The landscape owes its present form in a large measure to the fact that the Great Ice-Sheet spread over the State and leveled down the FIG. 158. In the East the Farm is cut up into Fields containing Acres! Photograph by McCormick Harvesting Machine Co. hills and filled the valleys, and left the surface in the form of level, undulating, or rolling prairies. The fertility of the soil is due in large part to the grinding and pulverizing of the rocks by the Great Ice- Sheet, forming the finest of rock-flour. No long, high, sweeping hill- sides such as are common in the Eastern States, formed by the wearing of streams during long ages, greet the eye in North Dakota. No long stretches of clayey hard-pan hillsides off from which the fertile soil is annually washed into the rivers, enter into the farm scenes in this Northwest land. The land is mostly free from stones in the fields because the rocks which were ploughed up by the Great Ice-Sheet were mostly soft rocks which were easily pulverized into fine rock- THE FUTURE OF NORTH DAKOTA. 305 FIG, 159. "Pictured Rock," Fort Ransom. After the Rocks Came Man. (The surface of the boulder was polished by glacial action. The marks are Indian hieroglyphics. ) FIG. 160. The Last Standing Vestige of Old Fort Abercrombie. 306 THE STORY OF THE PRAIRIES, FIG. 161. After the Aborigines Came the Pioneers. (Sod House Southeast of Lisbon.) FIG. 162. Flowing Well and Stock Barn, Red River Valley. (This is what man builds on North Dakota soil after the days of pioneering have passed.) THE FUTURE OF NORTH DAKOTA. 307 powder and soil. It is estimated that not more than one-twentieth as many boulders strew the fields in North Dakota as in the Eastern States which were passed over by the Great Ice-Sheet. The soil is good in North Dakota because the old sea-bottom sediments of the Cretaceous formations contained those mineral "salts" which are needed for the growth of wheat, flax and oats, and other cereal grains. The same "alkali" which sometimes renders the water not good for drinking, when present in small quantities distributed through the FIG. 163. In North Dakota it is One Big Field containing Sections ! Photograph by McCormick Harvesting Machine Co. rocks, helps to make the great fertility of the soil. The limestone which is the surface rock in portions of the Canadian Provinces lying to the north, and off from which the great ice-plow broke vast quanti- ties, which was carried over into this State and the sister state of Min- nesota, when ground into rock-powder forms the most fertile wheat soil in the world. From the circumstance of great bodies of water standing upon portions of the State in glacial lakes of large extent, which have now disappeared and their bottoms become dry land, there is spread out the most nearly level and among the most productive large areas of land in the world. These lands are level because the waves and currents of the lake waters distributed the earth materials evenly over their bottoms. They are the richest lands known because the finest of rock-powder was carried by the waves and currents and distributed in layers over their bottoms, and the gathering of vegetable remains upon the bottoms of these lakes added the black matter which gives the final touch of fertility to the soil and makes it the strongest crop-producing soil known. These lake bottom lands are not confined to the eastern portion of the State known as the Red River Valley, but the Mouse River Valley, embracing more than one-third of a million acres of land within the State of North Dakota, and the area covered by Lake Sargent, embracing most of Sargent County and a portion 308 THE STORY OF THE PRAIRIES. of Ransom County, and Lake Dakota, covering a portion of Dickey County, are Jike the Red River Valley in their geological character, viz., old lake bottoms. The even character of the bottoms of these lakes leaves almost the entire area s-uitable for cultivation in crop rais- ing and meadow. Less "waste land" exists in these portions of North Dakota than in almost any equal areas in any of the Northern States, being estimated over large areas not to exceed one-fiftieth of the whole. Mineral Resources. The geologic situation of the State makes pos- sible the obtaining of abundant flows of artesian water over most of the entire eastern half if not, indeed, over the whole State, and in all parts of the State water is obtained in unlimited supply from common wells at moderate depths. An inexhaustible supply of coal underlies the surface of the western half of the State, making abundant and cheap fuel within reach of all citizens of the State. Railroads traverse the State so that a never-ceas- ing fuel supply is available at small cost to the eastern half of the State from the vast coal fields in the western half. Clays suitable for the finest quality of building brick, and for fire- brick and tile purposes, and also for the finer processes of cement and pottery manufacture, lie but a little beneath the surface, and within easy reach of fuel for the manufacturing processes. Natural forests of growing timber abound along nearly all the stream courses, and experience has demonstrated that groves of any extent desired may be grown by suitable cultivation upon prairie lands remote from the larger streams. The Stockman's Paradise. Stock raising finds a paradise in the vast pastures of natural prairie grass, and hay in almost unlimited quantity can be cut, during dry seasons on the bottoms of the sloughs and marsh-lands, and in wet seasons on the higher lands, which yield more grass by far than there is stock of any kind to eat during the grazing season. The coulees or deep valleys in the western portion of the State furnish protection to animals from the storms of winter so that the expense for the construction of stables for horses and cattle, which attends stock farming in the East, is largely saved to the ranchman farmer in North Dakota. That the occurrence of storms and severe weather will bear favor- able comparison with other Northern States farther east is shown by the statistics of the U. S. Weather Bureau extending over the decades since accurate records have been kept in the Northwest. THE FUTURE OF NORTH DAKOTA. 309 310 THE STORY OF THE PRAIRIES. The Question of Rainfall. That the rainfall in North Dakota is suffi- cient in amount and distributed over the growing months of the year so as generally under good cultivation to produce reasonably sure- re- turns in bountiful harvests is shown by the records of annual and monthly rainfall for many years past. The question of rainfall is an important one in any agricultural dis- trict. To all appearances North Dakota has almost unlimited resources in the matter of fertility of soil. Records of the amount of rainfall have been kept for many years. A study of the U. S. Weather Bureau rec- ords will show that failure in farming, if failure there be, is due more largely to careless or unscientific farming than to either quality of soil or amount of rainfall. If any one has labored under the impression that the Northwest is subject to drought and hence is not adapted to profitable farming, a study of the climatic records will tend to dispel that idea. The idea has gained acceptance that these lands have good soil but that the rainfall is insufficient. Statistical figures compiled from official sources tell their own story of the amount of rainfall, and largely outweigh the notion that the rainfall is scant. Thirty years ago a vast region lying west of the Mississippi River embracing what are now portions of Kansas, Nebraska, South and North Dakota and Minnesota and some of the best parts of those now great and wealthy states was considered a great arid waste unfit for cultivation and not capable of supporting an agricultural population. Tn fact, many citizens of those states whose heads have silvered a little as they approach the latter half of life, remember when they studied about the "Great American Desert" in the geography lessons in the little schoolhouse on a clay hillside in the East, and were taught that it was a great arid waste a vast wild and irreclaimable buffalo range, best adapted to gophers, badgers, wolves, wild horses and red men ! It is said that Horace Greeley, while traveling across the barren desert plains of Arizona and New Mexico, said to a friend, with char- acteristic dry humor, that "all that that country needed to make it a desirable place to live was good people and plenty of water." "Yes," remarked his facetious friend, "all that hell needs to make it a desirable place to live in is the same two things!" Western North Dakota has been thought of by those who know the West only through fanciful tales as needing good people and plenty of water. That the State has room for more good people, thrifty, in- dustrious people, and that there is a large amount of land waiting for THE FUTURE OF NORTH DAKOTA. 311 f FIG. 165. Freeman's Ranch, in Little Muddy Valley. FIG. 166. Russian House and Stable. Photograph by Rex Willard. 312 THE STORY OF THE PRAIRIES. settlers, none will deny, but the water it seems to have. The geology of the region shows that the character of the soil is such as to be capable of a high degree of productiveness, and the records bear out the statement that the rainfall is sufficient under proper cultivation and with intelligent attention to scientific farming, with stock raising and crop rotation, to insure large harvests and profitable returns from general farming. The Days of the "Great American Desert" No More So far as the great Northwest is concerned there is no more "Great American Desert." The phrase has given place in modern geography to 'The Bread Bas- ket of the World." It is beyond our purpose to traverse the history of the development of the great Northwest, but the writer wishes herein to record his modest prediction that before another thirty years shall have rolled around the vast domain known as Western North Dakota will be occupied by bona fide settlers, will be owned and occupied by somebody, railways and highways will intersect it, schoolhouses and churches always an accompaniment and mark of American civiliza- tion will stand in different parts of all the counties; where are now small towns will be larger towns; what are now prairie postoffices at the intersections of trails will have grown to be agricultural and shipping centers. When this region becomes settled with an industrious and intelligent people who shall take advantage of the conditions as they are and adapt the mode of cultivation to the character of the lands, this will become a great and prosperous part of a vast commonwealth. Di- versified farming will be made successful. These lands are capable of sustaining a great population. A great population will occupy them. Here w r ill be the homes of successful and well-to-do farmers, as is now the case in the greater portion of those states which were once con- sidered to be irreclaimable desert wastes. CHAPTER THE TWENTY-EIGHTH. GEOLOGY FROM A CAR WINDOW. Prefatory Note. The pleasure and the benefit to be derived from travel are not always measured by the cost of the journey, the distance traveled, or the great cities passed through. A journey of a hundred miles may afford more real pleasure to the lover of nature than a thousand miles of travel to one who simply endures the time till he arrives at the end of his journey. There are people to whom a tour across the continent is a matter to be wished soon over because there is nothing to do but to ride and ride and wait for the miles to roll by. There are others to whom every mile reveals something new, to whom the changing view is a continuous panorama of delight from the begin- ning of the journey to the end of it. One who cannot enjoy pictures and statuary does not care to visit the art gallery. To one who does not enjoy the beautiful in architec- ture and mural decoration a cathedral is a no more fitting place to wor- ship than a hay-barn. To one who does not enjoy the beauty of the landscape, or who does not know the meaning of landscape gardening, who does not recognize nature's handiwork in the fashioning of land- scapes, does not derive the fullest value he pays for when he buys a railroad ticket. Just as a guide-book is an indispensable requisite to the tourist abroad, so it is thought a word about what may be seen by the way, a description made from the car window, will help to make more real what has been known in a general way before. Just as a catalogue of the pictures and statuary in the art gallery is helpful to the visitor, so the traveler on a business trip or the tourist in search of health or pleasure may get more from his journey if he has pointed out to him some of the simpler things of the landscape, so that there is added 313 314 THE STORY OF THE PRAIRIES. meaning in the hills and valleys, the forests and prairies, the sand- plains and lakes, a meaning which includes much more than merely seeing the things. The notes which follow have been gathered from observations ac- tually made from the car window, the rear platform, the cupola of the "caboose," or the top of a box car. This does not mean that the "geol- ogizing" which made the notes possible was all done from the moving train. Many days of field work have been necessary to make it possible to interpret what could be seen from the passing train. But to make the notes valuable for their purpose, and that they should not include what could not be actually seen by the passing traveler, the author has journeyed over all the lines of railroad here described and made the notes from first hand observation. It is hoped that the reader will find these observations helpful as furnishing particular illustrations of what has been said in the text. The notes are intended not only for the tourist and the traveling citizen who may be interested in knowing what he sees at the time when he sees it, but it is hoped that they may serve to give local touch and color to the descriptions in the body of the preceding pages. As teachers and all other persons who may read these pages travel more or less it is hoped that such notes may have a value in pointing out local examples of landscape features and that the teacher will find some assistance in local geography lessons in the facts which are here compiled. The figures following the names of stations indicate the number of miles from the point where the Red River is crossed by the particular line of railroad. The figures in parenthesis indicate the number of miles east from the point where the line leaves the State, unless other- wise specified. The figures representing feet show the altitude of each station above sea-level. All the figures representing distances and alti- tudes are taken from the official surveys of the several railroads. The population of all incorporated cities, towns and villages, census of 1900, is also given. County seats are distinguished by black faced type, and the station which is nearest to the boundary line of any county traversed by the particular road is indicated. THE GREAT NORTHERN LINES. 315 THE GREAT NORTHERN LINES. Grand Forks (Grand Forks County.) Distance from St. Paul (via Crookston), 324.1 miles; distance east of Montana line, 350 miles. Altitude, 835 feet. Population, 7,652. Grand Forks, on the Red River of the North, is surrounded by the level black plain of the axis of the bottom of Lake Agassiz. The prairie from here to the International Boundary, eighty miles north, falls only a little more than seven inches to the mile. At the time of the highest stage of Lake Agassiz, i. e., when the highest Herman Beach was being formed, the water was probably more than 300 feet deep where the city now stands. OJATA. Distance from Grand Forks, n.i miles. (Distance east of the Montana line, 338.9 miles. ) Altitude, 864 feet. ; Level prairie, with alkaline marshes. Two miles west the prairie becomes broken into irregular hummocks six to eight feet high. These mark the place of the lower Ojata Beach. The shore sand has been piled by the wind into small dunes. A little farther west the upper Ojata Beach rises about twenty feet, sloping at first, then rising quite suddenly. EMERADO. 15.6 miles. (334.4 miles.) Altitude, 910 feet. Popu- lation, 236. The Emerado Beach is crossed one-fourth mile east, about ten feet high. Hillsboro Beach, two miles west, just after crossing Hazen Creek. The east side of the beach rises as a beautiful slope, falling again a little west of its crest. ARVILLA. 21.0 miles. (329.0 miles.) Altitude, 1,022 feet. Popu- lation, 199. The front of the Elk Valley Delta rises distinctly to view, from south window of car. Within a mile east of Arvilla two conspicuous beaches are crossed, the McCauleyville and the Campbell, lying near together. The railroad makes a deep cut in crossing these. A gravel pit has been opened north of the track from which beach sand and gravel are taken. .The McCauleyville Beach rises eighteen feet, falling a little west of its crest, then the Campbell Beach rises twenty-five feet, falling about one- third of this west of its crest, to the level prairie. A fine view of the delta front is afforded from the south window, and the deep valley of the Turtle River from the north window. The Tintah Beach is crossed two miles west, where it runs along the delta front. The railroad here ascends upon the delta-plain by a heavy- 316 THE STORY OF THE PRAIRIES. grade. Farther west, long, low hills represent the Norcross Shore-Line, the beach sand having been piled into low dunes by the wind. LARIMORE. 27.7 miles. (322.3 miles.) Altitude, 1,138 feet. Pop- ulation, 1,235. Larimore stands upon the crest of the Elk Valley Delta, the plain of which is here about eight miles wide. The highest Herman Beach is four miles west, skirting the highland. Two lower Herman Beaches pass near the city, skirting its eastern suburbs. North of the city these beaches are crossed by the Turtle River, beyond which they rise as a long, low hill. Traveling westward it will be noticed that a broad, level plain ex- tends to the north. This is the mouth of the Elk Valley, and was the bed of the great glacial Elk River, which formed the delta in Lake Agas- siz. On the western side of this valley is the Herman Beach, marking the highest shore of Lake Agassiz. An extensive gravel pit has been opened in this beach near the railroad. As the ascent is made up the face of the highland to the west a fine view is afforded of the mouth of Elk Valley, "The Ridge," which forms its eastern side, north of Mc- Canna, and the Norcross Shore-Line south of the Ridge. A ride on the rear platform is here worth while, for a few miles. NIAGARA. 41.7 miles. (308.3 miles.) Altitude, 1,444 f eet - In a distance of eight miles an ascent of more than 300 feet is made, to Niagara. This is the Manitoba t Escarpment, the southern continua- tion of the Pembina Mountain Highland. Rounded, irregular hills, very different from anything which has been seen in the Red River Valley, occur on both sides of the track. These are morainic hills, and belong to the Ninth, or Leaf Hills, Moraine. Many granite boulders are no- ticed, and sloughs and ponds such as are common among the hills of a terminal moraine, are frequent. PETERSBURG. (Nelson County), 47.9 miles. (302.1 miles.) Alti- tude, 1,527 feet. Population, 182. MICHIGAN CITY. 53.7 miles. (296.3 miles.) Altitude, 1,523 feet. Population, 309. In the distance^ north the higher knobs of the Itasca (Tenth) Moraine can be seen. Many small lakes and grassy sloughs dot the prairie, glacial "pans" which show the undrained character of the landscape. A belt of high rounded knobs is crossed a little west of Michigan. MAPES. 58.4 miles. (291.6 miles.) Altitude, 1,531 feet. One of the principal ranges of the Itasca Moraine lies north one THE GREAT NORTHERN LINES. 317 mile. Same seen in distance north of Michigan City. Crosses line of railroad near Lakota, and continues to the Odessa Narrows of Devils Lake. Lakota. 64.0 miles. (286.0 miles.) Altitude, 1,520 feet. Popula- tion, 576. The Itasca Moraine is about a mile wide where crossed by the rail- road, just west of Lakota. It is a fine example of a morainic "range." It is followed by undulating prairie again farther west. Before reach- ing Bartlett another range of low hills of this moraine is crossed. BARTLETT. (Ramsey County), 68.0 miles. (282.0 miles.) Alti- tude, 1,536 feet. Many small shallow lakes dot the prairies between the moraines. In summer these are often "dry" lakes, with sheets of white alkaline salts covering their bottoms. The rounded knobs of another large range of the Itasca Moraine can be seen in the distance north. A branch connect- ing the moraine north with the one which was crossed at Lakota, lies in north and south direction across the line of the railroad east of Sidney station. West of Sidney is another similar range. CRARY. 77.9 miles. (272.1 miles.) Altitude, 1,490 feet. Popu- lation, 284. East of Crary a finely developed range, another branch connecting the two principal moraines mentioned. West of Crary a few miles is another range, with fine rounded hills, and little basins filled with water. These hills and hollows continue west of Keith station. The hills and hollows have the appearance of having been formed by the dumping of giant loads of earth from gigantic wheelbarrows ! This is a good exam- ple of a terminal, or "dump," moraine. Devils Lake 88.5 miles. (261.5 miles.) Altitude, 1,468 feet Population, 1,729. Along the line for fifteen miles approaching the city the high hills south of Devils Lake can be seen against the sky. Devils Heart, the highest of all, holds its summit 175 feet above its base, and 290 feet above the level of the water of the lake. Eight miles west, and about nine miles south of the city, is Sully's Hill, rising 275 feet above the ilake. Fine groves of oak, elm, linden, and cottonwood grow on the bluffs about the lake. In one of these groves, on a morainic swell on the north shore of the lake, the North Dakota Chautauqua grounds are located, one of the prettiest places in the State. The City of Devils Lake stands upon the low hills of a moraine, one 318 THE STORY OF THE PRAIRIE^. of the branches of the Itasca Moraine. The hills of one of the principal ranges of this moraine are seen about two miles north of the city. GRAND HARBOR. 95.8 miles. (254.2 miles.) Altitude, 1,460 feet. The large morainic range seen north of Devils Lake is crossed at Grand Harbor, the town being built upon the rounded knobs of this mo- raine. The railroad crosses another branch half way between Devils Lake and Grand Harbor. The bluffs which indent the lake, and which are generally tree-covered, are the ends of these smaller moraines which are too high to be covered by the waters of the lake. PENN. 101.7 miles. (248.3 miles.) Altitude, 1,473 ^ ee ^- Gently undulating prairie, enclosed in a wide loop of the moraine which was crossed at Grand Harbor, the western side of the loop being crossed two miles west of Churches Ferry. CHURCHES FERRY. 107.4 miles. (242.6 miles.) Altitude, 1,464 feet. Population, 264. Splendid level prairie, among the finest in the State and the northwest. The Mauvaise Coulee (meaning ''Bad Valley") is an old glacial drainage way which now connects the lakes to the north with Devils Lake in time of high water. This was the outlet stream of Glacial Lake Souris when that lake drained into Devils Lake and the Sheyenne River. ST. JOHN BRANCH. Cando (Towner County), 15.4 miles north of Churches Ferry. (43.5 miles south of International Boundary.) Altitude, 1,488 feet. Population, 1,061. Cando is located on the broad and fertile prairie which extends from the loop in the range of hills south of Grand Harbor and Churches Ferry north and west to the Turtle Mountains. About eight miles north and east lies one of the ranges of the Itasca Moraine, about five miles in width, which extends northwest across the Turtle Mountains, and south- east nearly to Lakota, thence northeast along the highland west of Con- way and Inkster. Holla. (Rolette County), 47.4 miles. (11.5 miles.) Altitude 1,823 feet. Population, 400. About six miles south of Rolla the railroad crosses the moraine re- ferred, to. About six miles northwest, the plateau of the Turtle Moun- tains rises suddenly against the horizon, having a height here of 300 to 400 feet above the prairie. THE GREAT NORTHERN LINES. , 319 ST. JOHN. 54.8 miles. (4.0 miles.) Altitude, 1,950 feet. Popula- tion, 1 68. City located on eastern slope of the Mountains. Deep coulees inter- sect the sides of the plateau. Morainic hills, of the range crossed ten to twelve miles south, three miles west. Mountain top clad with forests. ##*##***#** LEEDS. (Benson County), 119.0 miles. (231.0 miles.) Altitude, 1,516 feet. Population, 349. Big Butte, or Mauvais Butte, rises with broad sweeping outlines six or seven miles south. This immense hill is about ten miles in length, rising 150 to 200 feet above the surrounding prairie at its eastern end, and 250 to 300 feet in its western portion. This is an old hill, i. e., it was a hill before the Ice Age, as is shown by the fact that morainic ridges cross its surface and many boulders lie scattered upon it. It be- longs in the class of hills such as Devils Heart and Sully's Hill, south of Devils Lake. Occasional morainic hills stand upon the flat landscape between the Big Butte and the railroad. A well marked, broad glacial drainage channel west of Leeds. Only a small stream in it now. YORK. 125.3 miles. (224.7 miles.) Altitude, 1,619 feet. The high western end of Big Butte rises in the southeast. Its sur- face is seen in the distance to be rough, and marked by coulees and mo- rainic hills. The prairie between Leeds and York is undulating, with an occasional round, morainic knob standing like a great mound upon the generally flat landscape. A school house stands on the top of one of these knobs, just south of the railroad. Many fine hay-meadows occur in the shallow depressions. KNOX. 131.1 miles. (218.9 miles.) Altitude, 1,610 feet Fine example of moraine crossed east of Knox. Along the horizon in the west, the high knobs and ridges of the combined Fergus Falls (Eighth) and Leaf Hills (Ninth) Moraines, the town standing upon a beautiful, gently undulating prairie. A fine view of a moraine is afforded from the south window of the car, west of Knox. The higher rounded hills of another moraine seen in the distance north. About two miles west is the highest point of the Great Northern Railway between the Red River of the North and the eastern slope of the Missouri Plateau, 1,660 feet. This is the "divide" between Devils Lake and the Sheyenne on the east, and the Mouse River on the west. It can hardly be called a watershed, for the land in this region is undrained 320 THE STORY OF THE PRAIRIES. by streams. Drainage courses have not had time to. become established since the Glacial Period. PLEASANT LAKE. 136.7 miles. (213.3 miles.) Altitude, 1,608 feet The change from the gently undulating prairie to the extremely irreg- ular topography of the morainic belt which is here entered, is very marked. The hills are steep and rugged, and strewn with boulders. A typical morainic region. This belt of hills, the compound Fergus Falls and Leaf Hills Moraines, crosses the Indian Reservation south of Devils Lake, the two moraines continuing separately from the hilly region south of Devils Lake across Nelson, Steele, and Barnes Counties, and thence, after crossing the Red River Valley, where they are recognized only as slight undulations, they continue across Minnesota, and become a part of the great "Kettle Moraine" in Wisconsin, Illinois, Michigan, and Indi- ana, and extending east to the Atlantic Ocean near New York City. The higher hills, which lie along either side of the railroad, rise from fifty to one hundred feet above the hollows at their bases, and from 100 to 150 feet above the general level of the prairie outside the moraine. Scores and hundreds of lakes of all sizes, from small ponds up to lakes two miles in diameter, occupy the hollows between the hills. Which one of the great number which lie within a few miles is the most "pleasant" would be hard to say. South of Pleasant Lake station is Broken Bone Lake, named from the piles of broken buffalo bones which were left on its shores by the Indians, from the manufacture of "pemmican," in which the marrow of the bones was used. Rugby. (Pierce County), 145.7 miles. (204.3 miles.) Altitude, 1,567 feet. Population, 487. Rugby stands where was the eastern shore of Old Lake Souris. The aspect of the prairie suddenly changes to the west. The rolling and hilly landscape gives way to a broad, gently swelling prairie. The hilly mo- raines stop abruptly at the shore of the old lake. The hills were depos- ited upon the lake bottom just as they were beyond its shores, but they were leveled down by the action of the waves. From Rugby to Minot the railroad crosses the bottom of Lake Souris in the same manner as the western half of the bottom of Lake Agassiz was crossed from Grand Forks to Larimore. High morainic hills are seen both north and south of Rugby. The region about the city was occupied by a bay of the lake. A long slough lying north of the railroad can be traced from the car win- dow most of the way between Pleasant Lake and Rugby. This is the THE GREAT NORTHERN LINES. old channel of a glacial stream which flowed into Lake Souris from the melting ice at the time the moraine about Pleasant Lake was being formed. BOTTINEAU BRANCH. WILLOW CITY. (Bottineau County), 21.2 miles from Rugby. (36.9 miles south of the International Boundary.) Altitude, 1,478 feet/ Population, 476. The railroad runs near the eastern shore, on the Lake Souris bottom. The horizon line to the east is an irregular line, marked by the crests of the hills and ridges, and the deep intervening hollows. To the west the sky meets the earth upon an almost unbroken horizontal line, the nearly level surface of the old lake bottom. O'MEMEE. 30.4 miles. (27.7 miles.) Altitude, 1,545 feet. Popu- lation, 320. East of Willow City and Omemee the undulations of the prairie rise often ten feet above the adjoining hollows. These swells are the mo- rainic hills which were not entirely leveled down by the waves of the lake. The high and rugged hills of the same moraines beyond the lake shore show how much the action of the waves leveled the hills which were deposited in the waters of the lake. Bottineau. 38.1 miles. (20 miles.) Altitude, 1,646 feet. Pop- ulation, 888. Bottineau stands upon the lake bottom, about four miles from where the waters washed the base of the Turtle Mountains. The Turtle Moun- tain Plateau rises 400 to 600 feet above the level of the prairie at Bot- tineau. The highest hills on the top of the plateau, Butte St. Paul and Bear Butte, rise 700 and 600 feet, 1 respectively, above the prairie at the foot of the plateau, or Mountains. 2 SOURIS. 52.1 miles. (6 miles.) Altitude (about), 1,550 feet. West of Bottineau is a group of morainic hills which seemed to be too large to be leveled by the waves of the lake. They probably formed an island, or a group of islands, during the receding stages of the lake. Souris is on the nearly level prairie of the deeper lake bottom. 1 Andreas' Atlas of Dakota. 2 The U. S. Boundary Commission gives the highest point of the Turtle Mountains as 2,534 feet above sea level. 322 THE STORY OF THE PRAIRIES. BERWICK. (McHenry County), 157.4 miles. (192.6 miles.) Alti- tude, 1,487 feet. West from Rugby the railroad grade falls rapidly from the shore of the old lake toward its center. From the north window, or better, from the rear platform, the ends of the high moraines can be distinctly seen along the shore, cut off abruptly by the waves of the lake. North, and south, of the railroad, ranges of low, broad hills, with gracefully curv- ing outlines, are the continuation of the moraines south of Rugby, across the lake bottom. They all have a trend northwest by west, so that they appear to approach the railroad from the south, and diverge to the north. Fine nearly level prairies lie between these long ground-swells. A mo- raine of sandy hills is marked by dunes farther west. Towner 164.7 miles. (185.3 miles.) Altitude, 1,482 feet. Pop- ulation, 331. Towner is on the lowest part of the old lake bottom crossed by the Great Northern Railway. The valley of the river, i. e., the trough cut by the river, is very small here compared with the broad and deep valley west of Minot. Here it is a small valley, cut since the lake disappeared. At Minot and westward the large valley was eroded by the great glacial river which emptied into Lake Souris. DENBIGH. 173.4 miles. (176.6 miles.) Altitude, 1,520 feet. Choppy sand hills, wind-drifted. High, ragged dunes, morainic hills of sand partly leveled by the waves of the lake, and modified by the action of the wind. Higher hills, grass covered, are more clayey, and so not affected by the winds. Medicine Lodge Hill, and Buffalo Lodge, seen from the north window, are morainic hills which were too large to be leveled by the lake. They were at one time islands in Lake Souris. Tracts of dunes, sparsely covered with scrubby timber, and alkaline lakes, lie along the course of this moraine. About a mile north of the railroad, west of Riga station, is a high morainic hill, its crest irregular in form, grass covered, and abruptly cut off at its eastern end. This was an island in the lake, as is shown by the form of the hill, for its crest has not been made smooth by the waves. Its eastern end was washed away by the waves, hence the level plain sur- rounding it. GRANVILLE. 185.1 miles. (164.9 miles.) Altitude, 1,516 feet. Range of morainic hills west, their outlines smoothed and rounded by the waves, but not leveled much. Little bumpy dunes of sand, the sandy character of the hills being due to the ploughing up of the Fox Hills THE GREAT NORTHERN LINES. 323 Sandstone by the ice-sheet, and ground up and deposited as a moraine in the lake. Nearly level prairies lie in the broad tracts between the mo- raines. NORWICH. 192.2 miles. (157.8 miles.) Altitude, 1,531 feet. Town built on hills of moraine. Hills modified in outline by action of lake waters, giving them a beautifully curved contour. Moraine crosses railroad from southeast to northwest. Splendid prairie, nearly level, both east and west of this moraine. SURREY. (Ward County), 198.9 miles. (151.1 miles.) Altitude, 1,635 feet- Steep slope of the western side of the valley of the Mouse River can be seen to the west, from south window. Prairie here intersected by coulees leading into the Mouse. Minot 206.3 miles. (143.7 miles.) Altitude, 1,562 feet. Popu- lation, 1,277. Minot stands at the western edge, or shore, of Lake Sour is, in a ba> formed by the broad mouth of the Mouse River, where that great glacial stream emptied its waters, collected from the melting ice-sheet, into Lake Souris. The valley of the Mouse is here more than a mile wide, and its banks on either side are hills rising 150 feet or more to the level of the adjacent prairie. The valley is here many times larger than the same valley sixty miles farther down its course at Towner. Many coulees, or little tributaries, have cut into the valley sides giving a beautiful combed, or grooved, appearance. Viewed from the bottom of the valley the land- scape on either side- appears hilly. The coulees which mark the banks are little notches cut by the rains which flow into the valley from the prairie. Some of these have worked back into the landscape several miles. About four miles west of Minot is the junction of the Mouse and Des Lacs Rivers. Both these streams are now small,- but their valleys are broad and deep. They are glacial valleys, cut when the waters from the melting ice-sheet kept the streams at high flood. ( See Tenth Chapter. ) DES LACS. ^220.0 miles. (130.0 miles.) Altitude, 1,902 feet. Railroad ascends by heavy grade to the high prairie west of the Mouse Valley, rising 339 feet in less than fourteen miles. A fine view of the broad Valley of the Mouse is afforded from the north window. The view from the rear platform will repay the inconvenience. A deep coulee is spanned by a high iron bridge four miles west, beyond which the grade rises through many cuts to the prairie at Des Lacs. Fine ex- 324 THE STORY OF THE PRAIRIES. amples of coulees, or "young valleys," deep troughs with steep sides, are best seen from south window. The great Missouri Plateau rises against the western horizon fifteen miles distant. BERTHOLD. 229.2 miles. (120.8 miles.) Altitude, 2,087 feet. A vast evenly sloping prairie, rising quite rapidly toward the Plateau front, extends as far'as the eye can reach, to the northwest and southeast. TAGUS (or Wallace). 238.8 miles. (111.2 miles.) Altitude, 2,187 feet. Here is the front of the Missouri Plateau, or Coteau du Missouri. The Outer or Altamont Moraine lies upon the eastern edge of the plateau. Rounded hills and many small lakes mark the rough morainic topogra- phy. A small lake having a distinct wave-worn cliff, and a shore-boul- der chain at its foot, lies near the station. DELTA. 245.4 miles. (104.6 miles.) Altitude, 2,263 ^ eet - Delta is on the top of the plateau, 700 feet higher than Minot, forty miles east. The high, steep, and boulder-strewn hills illustrate finely the landscape of a terminal moraine. Many small lakes and sloughs at dif- ferent levels, without outlets. PALERMO. 252.6 miles. (97.4 miles.) Altitude, 2,200 feet. Station stands at western edge of the Altamont Moraine, which is here about ten miles wide. The landscape is a rolling prairie westward. No more morainic hills will be seen farther west, for here was the limit of the western movement of the great ice-sheet. Boulders and glacial gravels occur as far as the Missouri River, "over-wash" materials from the Outer Moraine, or else deposits from the older ice-sheet, which is often spoken of as the "Old Drift." 1 STANLEY. 260.7 miles. (89.3 miles.) Altitude, 2,253 ^ eet - The vast upland prairie of the Coteau du Missouri. Many alkaline lakes in shallow basins. Fine grazing lands; hay-meadows dotted with stacks. WHITE EARTH. 279.9 miles. (70.1 miles.) Altitude, 2,092 feet. 1 By " Old Drift" is meant the materials left upon the surface by an older ice- sheet, which belonged to an earlier epoch of the Glacial Period. This earlier ice inva- sion extended farther west in North Dakota than did the great Ice- Sheet by which the moraines described in this book were formed. The Altamont Moraine, the outermost of those formed by the later Ice-Sheet, nowhere occurs west of the Missouri River, but the Older Drift extends as a thin mantle to a distance, as usually mapped, of 40 to 60 miles west of the river. The drift may have extended farther west, however, and have been carried away by erosion since the disappearance of the ice. The writer has ob- served what seemed to be drift pebbles 100 miles west of the Missouri River at Dickin- son, and nearly 40 miles still farther west, 85 miles south of the Missouri River at Williston, on the tops of buttes in the valley of the Little Missouri River, 10 miles south of Medora, what appeared to be drift deposits occurred to a depth of 8 to 12 feet, rounded pebbles of quartzite and granite up to 5 or 6 inches in diameter being observed, THE GREAT NORTHERN LINES. 325 The Valley of White Earth Creek has been cut down 150 feet into the plain. Strata of whitish sandstone and clay exposed % in the naked sides of the valley, Hence the name. Springs, which burst out of the hillsides of this and similar deep valleys, are of much value to the ranchmen. Protection is also afforded for stock during the winter season in these deep valleys. The only town between Minot and Williston is this. All the other stations are names upon the map merely. TIOGA. (Williams County), 287.9 miles. (62.1 miles.) Altitude, 2,237 feet. The high prairie again reached, after traversing the course of a side coulee of White Earth Creek. WHEELOCK. 305.4 miles. (44.6 miles.) Altitude, 2,380 feet. Highest point of the Great Northern Railway in the State. The higher points on the prairie rise to an elevation above sea-level of 2,500 to 2,600 feet, which is more than 1,000 feet higher than the plain of the Mouse Valley where it is crossed by the railroad 100 miles east. SPRING BROOK. 316.0 miles. (34.0 miles.) Altitude, 2,066 feet. Railroad descends more than 500 feet in a distance of a little more than tw r enty miles, to the Valley of the Missouri River, crossing and re- crossing the channel of Stony Creek. Williston 327.3 miles. (22.7 miles.) Altitude, 1,859 feet - Pop- ulation, 763. High, steep, naked butte-like hills south of the river. River channel meanders over a broad belt of changing sand-bars. Dense forest of red willows and heavy timber along the river, the favorite haunt of deer and other wild game. The city stands upon a terrace formed by a glacial stream which spreads as a belt of gravel and sand for a distance of more than thirty miles along the course of Little Muddy Creek. The bottom land of the valley is about thirty feet lower than the ' 'bench" or terrace, and the Little Muddy flows in a small channel on the "bottom." So slight is the fall of this flat bottom that an irrigation ditch eleven miles in length is necessary to raise the water out of the small channel to the level meadow bottom-land at the mouth of the valley. BUFORD. 348.0 miles. (2.0 miles.) Altitude, 1,950 feet. Railroad follows the picturesque valley of the Missouri. South and west rise the naked walls of the buttes, cut in the high prairie by the mighty rivers, the Missouri and Yellowstone. The two rivers meet be- yond the plain south of the station. 326 THE STORY OF THE PRAIRIES. FARGO TO GRAND FORKS AND NECHE. Fargo. (Cass County.) Distance from St. Paul, 242.0 miles. (Distance south of International Boundary at Neche, 159.1 miles.) Altitude, 900 feet. Population, 9,589. The Great Northern Railway crosses the Red River of the North at Fargo, thence running down the Red River Valley generally at a distance of about ten miles from the river, to Grand Forks, and the International Boundary. The entire distance is over the almost level plain of the bot- tom of Lake Agassiz, broken only by beach ridges, or shore lines. The soil is the fine silt of the lake bottom, blackened by the accumulation of organic matter. Its fertility is unexcelled by anything in the world. For fifteen miles out from Fargo the prairie is nearly level, and un- drained. The plain is crossed by the Sheyenne River at Harwood, but this river can hardly be said to drain the land. It is the relic of a large glacial stream, but which has now but a very slight current. The sta- tions of Harwood and Argusville are both lower than Fargo. The river would have to flow up hill to discharge its waters directly into the Red. GARDNER. Distance from Fargo, 20.7 miles. (Distance south of the International Boundary, 138.4 miles.) Altitude, 886 feet. Popula- tion, 266. GRANDIN. 27.0 miles. (132.1 miles.) Altitude, 891 feet. Hillsboro Beach, rising ten feet from the prairie, two miles west of railroad. With some difficulty distinguished from the passing train. KELSO. (Traill County), 33.2 miles. (125.9 miles.) Altitude, 897 feet. North branch Elm River. Hillsboro Beach finely developed west, a typical shore-line of gravel and sand. HILLSBORO. 38.9 miles. (120.2 miles.) Altitude, 902 feet. Pop- ulation, 1,172. Goose River, north of the city. The Hillsboro Beach rises fifteen feet from the level prairie, one mile west. CUMMINGS. 46.7 miles. (112.4 miles.) Altitude, 904 feet Two miles south of Cummings the railroad crosses the Hillsboro Shore-Line. This is here not a beach ridge but an eroded cliff" eight to ten feet high. East of Cummings a few rods it is a well defined beach ridge of gravel and sand, rising ten feet on the east side and falling five feet on the back, or west side. THE GREAT NORTHERN LINES. 327 BUXTON. 52.8 miles. (106.3 miles.) Altitude, 931 feet. Popu- lation, 470. North of Cummings the railroad runs upon the crest of the Hillsboro Beach for a short distance, then the beach diverges eastward. West of Buxton one mile is the lower Blanchard Beach. This is the high ridge which is crossed four miles west of Cummings, on the stage route to Mayville. REYNOLDS. (Grand Forks County), 57.6 miles. (101.5 miles.) Altitude, 910 feet. Population, 389. The Hillsboro Beach is crossed again by the railroad one and one-half miles south of Reynolds. The beach is here about thirty rods wide, a large gravel pit having been opened in it. It is about eight feet high on the east side, falling six feet on the west side. The Emerado Beach is crossed one and one-half miles north of Reynolds, and three miles far- ther north the upper Ojata Beach is crossed. One mile north of Thomp- son the lower Ojata Beach is crossed. A mile north of Merrifield is the Gladstone Beach, which marks the shore of the lake when its southern point was where this beach crosses the Red River east of Buxton. GRAND FORKS. 78.2 miles. (See page 198.) SCHURMEIER. 84.3 miles. (74.8 miles.) Altitude, 833 feet. Irregularly ridged surface, showing wave action at time of forma- tion of Burnside Beach. MANVEL. 90.7 miles. (68.4 miles.) Altitude, 827 feet. The irregular surface of the Burnside Beach continues east of the railroad from Schurmeier, and extends through the west side of Manvel. ARDOCH. (Walsh County), 102.4 miles. (56.7 miles.) Altitude, 832 feet. Population, 298. The Burnside Beach, a ridge two to three feet high, and thirty rods wide, crossed by railroad north of station. MINTO. 108.8 miles. (50.3 miles.) Altitude, 828 feet. Popula- tion, 860. Burnside Shore-Line about a mile west, but it is not easily traced on the level prairie. Cross Forest River. Grafton 117.7 miles. (41.4 miles.) Altitude, 834 feet. Popu- lation, 2,378. Burnside Beach passes through east side of city. Gladstone Beach about four miles west, but neither beach is well marked. The region represents the almost perfectly level plain of the lake bottom. The black soil is unexcelled in fertility in the world. 328 THE STORY OF THE PRAIRIES. ST. THOMAS. (Pembina County), 131.8 miles. (27.3 miles.) Al- titude, 847 feet. Population, 66 1. The Pembina Mountain, or northern and higher portion of the Mani- toba Escarpment, rises clearly on the horizon twenty miles distant. HAMILTON. 144.8 miles. (14.3 miles.)' Altitude, 827 feet. Pop- ulation, 224. Burnside Beach is crossed by railroad one mile south. BATHGATE. 149.9 miles. (9.2 miles.) Altitude, 828 feet. Popu- lation, 756. Cross the Tongue River. Burnside Beach about one mile west, not easily recognized. Two or three low ridges two to four feet high, mark- ing the shore-line of Lake Agassiz during the Ossawa Stage, are about two miles east, and appear from passing train as slight undulations in the surface. Pembina Mountain rises high in the west, distant about twen- ty-five miles. NECHE. 157.7 miles. (1.4 miles.) Altitude, 837 feet. Popula- tion, 396. Pembina River north of the town. In a journey of 160 miles from Fargo the traveler has seen the finest agricultural land in the world. His route has been along the axial line of the Red River Valley. The fine sediment of the lake bottom makes the soil of incomparable richness. Nowhere on the Western Hemisphere, if in the world, is there such an extent of so nearly level and so fertile land, with so little waste. WINNIPEG. North of Neche, 74.5 miles. Altitude, 757 feet. WAHPETON TO LABIMORE AND HANNAH. Wahpeton (Richland County.) Distance from St. Paul, 215.4 miles. (Distance south of International Boundary, 228 miles.) Alti- tude, 955 feet. Population, 2,228. Bois des Sioux River crossed just above its junction with the Red to form the Red River of the North. The highest shores of Lake Agassiz are twenty-eight miles west, near Wyndmere, and seventeen miles east, in Minnesota. Lake Traverse is thirty-five miles south, lying in the trough of the southern outlet of Lake Agassiz. The Red (called also Otter Tail) and the Bois des Sioux meet to form the Red River of the North, which flows north along the axis of the bed of Lake Agassiz, 285 miles in a direct line, or about 700 miles following its winding course, to Lake THE GREAT NORTHERN LINES. 329 Winnipeg. At the time of the formation of the highest shore-line of Lake Agassiz the water was about 100 feet deep where the City of Wah- peton now stands. DWIGHT. Distance from Wahpeton, 6.9 miles. (South of Interna- tional Boundary, 221.1 miles.) Altitude, 946 feet. The front or edge of the plateau of the delta formed by the Sheyenne River in Lake Agassiz rises along the horizon eight miles west. COLFAX. 20.1 miles. (206.9 miles.) Altitude, 951 feet. South two or three miles the railroad crosses the lower McCauley- ville Beach, which is broken into dunes of wind-blown sand five to fifteen feet high, the sand having been washed from the edge of the delta, which is about a mile west. WALCOTT. 26.3 miles. (201.7 miles.) Altitude, 948 feet. The delta front rises in the near distance west, so that the horizon is near. The horizon to the east is far out over the broad expanse of the level lake bottom. Low swelling dunes with beautifully curved outlines south. Dunes with ragged crests north. These are the wind-piled sands of the upper McCauleyville Beach. This beach was formed dur- ing the lowest stage of the lake while it yet discharged to the south. The delta was formed during the higher stages of the lake. The Campbell and Tintah Beaches cross the delta beyond view to the west, where their sands have been blown into dunes ten to thirty feet high. KINDRED. (Cass County), 34.2 miles. (193.8 miles.) Altitude, 932 feet. Population, 348. The lower McCauleyville Beach turns westward between Walcott and the Sheyenne River. It is a large ridge broken into hills twenty to thirty feet high. A house stands on the top of one of these hills just west of the railroad. After crossing the river a fine view of the delta plateau is obtained from the level prairie south of Kindred. DAVENPORT. 39.1 miles. (188.9 miles.) Altitude, 910 feet. Pop- ulation, 245. Crossing Fargo and Southwestern Branch N. P. R'y. The front of the delta can be seen from the west window as it extends away to the northwest. DURBIN. 47.3 miles. (180.7 miles.) Altitude, 908 feet. Two miles north of Durbin Maple Ridge rises ten feet above the level prairie, where it is crossed by the railroad. This is a beach ridge which was built out from the shore of Lake Agassiz as a "spit," or bar. It is 330 THE STORY OF THE PRAIRIES. a broad swell about fifty rods across, and follows the course of Maple River for twenty miles. CASSELTON. 54.8 miles. (173.2 miles.) Altitude, 927 feet. Pop- ulation, 1,207. Crossing main line N. P. Ry. Prairie level as the floor of a house. An ideal landscape. ARTHUR. 68.6 miles. (159.4 miles.) Altitude, 979 feet. The Campbell Beach is crossed by the railroad south where it is a ridge rising ten feet on its east side. Just before reaching Arthur the beach west of the track appears to be double, one ridge being higher than the other, like stairs. The lower one is the upper McCauleyville, the higher the Campbell. The crest of the Campbell Beach lies close east of Arthur. This beach is again crossed north of Arthur as it turns west- ward, the railroad then running east of and near to the ridge for about ten miles. Marshy lagoons occur frequently west of the ridge. HUNTER. 74.6 miles. (153.4 miles.) Altitude, 965 feet. Popu- lation, 407. The upper McCauleyville Beach is a low bank, or cliff, east of Hun- ter. It can be traced from the east window for about two miles south, lying half a mile from the railroad. The Campbell Shore-Line is a well marked cliff half a mile west. BLANCHARD. (Traill County), 85.2 miles. (142.8 miles.) Alti- tude, 935 feet. South of Blanchard a fine illustration of the development of a drain- age system on a level plain is afforded. The prairie is intersected by the north fork of the Elm River, and tiny hills, just like the big hills in Penn- sylvania and New York except in size, are separated by little coulees, or young valleys. (See Chapter One.) MAYVILLE. 97 miles. (131 miles.) Altitude, 967 feet. Popula- tion, i, 1 06. South of Mayville the rise toward the western shore of Lake Agassiz is clearly seen. The Campbell, Tintah, Norcross, and lower Herman Shore-Lines rise one above another, the crest of the last named forming the horizon line. The heights of these beaches, as shown in Figure 40, p&ge 85, give an idea of the depth of the water of the lake where the City of Mayville now stands, at the time when these beaches were being formed. The Campbell Beach passes through the City of Portland, two miles west, as a wave-cut cliff. The Bruflat Academy stands upon THE GREAT NORTHERN LINES. 331 its crest. North of Mayville half a mile the railroad crosses the upper McCauleyville Beach, which here has a nearly northeast course. HATTON. 108.6 miles. (119.4 miles.) Altitude, i, 068 feet. Pop- ulation, 430. The elevation of the railroad from Mayville to Hatton, a distance of less than twelve miles, rises more than 100 feet. It may be observed also that the soil becomes more sandy. The sudden rise in the elevation and the more sandy character of the soil are due to the fact that the railroad here passes upon the plain of the Elk Valley Delta. This delta spreads over the lake bottom from the mouth of the Elk Valley north of Larimore south to Portland, its southern and eastern edge thence extending near Mayville north and a little east, crossing the main line of the Great Northern Railway about a mile west of Arvilla. The Norcross Beach crosses this delta plain at Hatton. It can be recognized south of Hatton where it is marked by small dunes of wind-blown sand. NORTH WOOD. ( Grand Forks County), 116.8 miles. (11.2 miles.) Altitude, 1,1 01 feet. Population, 697. From Hatton north is a beautiful prairie^ the top of the delta being here but slightly undulating. The highland of the Manitoba Escarp- ment, the southern extension of the Pembina Mountain, rises clearly in view in the west. The highest, or Herman, shore-line of Lake Agassiz lies at the foot of this long hill. LARIMORE. 129.5 miles. (98.5 miles.) Altitude, 1,138 feet. Population, 1,235. (See page 199-) McCANNA. 137.7 miles. (90.3 miles.) Altitude, 1,145 f eet - The broad level plain is the mouth of the Elk Valley. The highland west is the Manitoba Escarpment. North four miles and half a mile east of the railroad, rises the first of the series of hills locally known as 'The Ridge." These hills were islands in Lake Agassiz. The Ridge formed the eastern side, and the Manitoba Escarpment the western side, of the Elk Valley. It was the Glacial Elk River, which once flowed in this broad valley, which formed the delta south of Larimore. North of this first island the railroad crosses a high beach ridge which extends between the first and second islands. The second island rises as a large hill west of the railroad after crossing the beach ridge. ORR. 143.0 miles. (85.0 miles.) Altitude, 1,103 feet. Fine view of the great plain of the bottom of Lake Agassiz, from east window, south of Orr. A few rods south of the station the railroad ,332 THE STORY OF THE PRAIRIES. crosses the Norcross Beach, which is here a well defined ridge of gravel and sand. INKSTER. 148.5 miles. (79.5 miles.) Altitude, 1,041 feet. Pop- ulation, 376. Two miles north of Orr the railroad crosses the Tintah Beach, which is well marked. West of the railroad it is double, the upper crest being two to five feet higher than the lower. Inkster stands upon the flat top of the Campbell Beach, which is here a wave-cut cliff, or bank, fifteen to twenty-five feet high. Half a mile east, on the level prairie, is the Mc~ Cauleyville Beach, a gravelly and sandy ridge. The railroad crosses the latter north of Inkster near the crossing of Forest River, where it is a conspicuous ridge, its front rising eight feet from the prairie, and falling five feet on the west. CON WAY. (Walsh County), 154.5 miles. (73.5 miles.) Altitude. 993 feet. Population, 216. From the crossing of Forest River north to Conway the McCauley- ville Beach lies a few rods west of the railroad. A half mile west and parallel with it is the Campbell Beach. The first of 'The Mountains"- the name locally applied to the large hills which are the northern continu- ation of "The Ridge" rises high two and one-half miles west of Con- way as a large long hill. PISEK. 160.0 miles. (68.0 miles.) Altitude, 1,006 feet. Popu- lation, 132. The wide southern end of the second "Mountain" rises three miles west. The crest of this "mountain" is 200 to 225 feet higher than the surrounding prairie, and seventy-five to one hundred feet higher than the highest point reached by the waters of Lake Agassiz. PARK RIVER. 166.1 miles. (61.9 miles.) Altitude, 1,003 ^ eet - Population, 1,088. From two miles south of Pisek the railroad runs on the natural grade of the McCauleyville Beach. The Campbell Beach lies half a mile west, and is a massive beach rising twenty to thirty-five feet. The Mountain is hidden from view by the high crest of this beach. After crossing Park River the top of the Mountain is seen above the crest of the beach. About a mile north the railroad crosses the Campbell Beach by a deep cut. The prairie surface beyond is marked by hummocks, the irregular Tintah and Norcross Beaches, broken by the action of the waves beating upon the base of the Mountain, which now rises high to the west. THE GREAT NORTHERN LINES. 333 EDINBURG. 175.5 miles. (5 2 -5 miles.) Altitude, 1,194 feet. Pop- ulation, 286. Edinburg stands at the north end of the Mountain. This "Moun- tain." which is a part of the great ridge which forms the eastern side of Elk Valley (called Golden Valley, or Pleasant Valley, in northern portion) should be clearly distinguished from the Pembina Mountain, which is a part of the Manitoba Escarpment, which formed the western shore of Lake Agassiz. "The Ridge" and "The Mountains" on the east side of Elk and Golden Valleys, are drift hills (moraines). The Mani- toba Escarpment is a hill of the Cretaceous rocks. (See Seventh Chap- ter, page 69. ) The higher part of the Manitoba Escarpment, north from about where the railroad passes upon it, is called Pembina Mountain. MILTON. (Cavalier County), 1 88.1 miles. (39.9 miles.) Altitude, 1,591 feet. Population, 384. In the distance from Edinburg to Milton, a little more than twelve miles, the railroad rises 397 feet, the grade being forty-two feet to the mile for seven or eight miles. Here are many boulders of granite, and gravel and sand, and the irregular ridges and rounded hills which always indicate a terminal moraine. This is one of the ranges of the Itasca, or Tenth, Moraine. Boulders are especially abundant about Union. Deep cuts along the railroad show sections of drift hills, often with stratified sand, some of the excavations reaching into the shales of the bed-rock. Deep, jagged sided coulees, and forests of poplar trees, lend a pic- turesque grandeur to the scene. Magnificent panoramic views of the level plain of the valley bottom of Lake Agassiz are obtained at different points along the line. A fine example of a "young valley" occurs west of Milton, one of the head streams of Park River. Its course can be followed for several miles (better traced when traveling east) from west of Milton, where it has its "head," growing deeper and wider down its course south and east, many small "tributaries" entering it. OSNABROCK. 193.7 m iles. (34.3 miles.) Altitude, 1,625 feet. Population, 228. Fine undulating prairie, marked by occasional low morainic hills. EASBY. 199.2 miles. (28.8 miles.) Altitude, 1,652 feet. A small but noticeable range of morainic hills extends across the line of the railroad in a northwest by north direction. Langdon. 205.4 miles. (22.6 miles.) Altitude, 1,615 feet Pop- ulation, i, 1 1 8. Range of morainic hills east half a mile. Rolling prairie landscape, 334 THE STORY OF THE PRAIRIES. fine farming lands. Several low, gently rising moraines, belonging to the Itasca Moraine, lie in nearly parallel courses a few miles apart. HANNAH. 226.5 miles. (1.5 miles.) Altitude, 1,568 feet. Pop- ulation, 596. Terminus of railroad. The vast plain comprising the top of Pem- bina Mountain is crossed by several small ranges of the Itasca Moraine. These diverge from the line of the railroad going north and approach it going south, from the right hand. CHAPTER THE TWENTY-NINTH. GEOLOGY FROM A CAR WINDOW THE NORTHERN PACIFIC LINES. y ; . Fargo (Cass County.) Distance from St. Paul, 251.5 miles. (Distance from Montana Line, 376.2 miles.) Altitude, 902 feet. Popu- lation, 9,589. Fargo is situated upon the axis of the Red River Valley, surrounded by the almost perfectly level prairie of the bottom of Lake Agassiz, the great wheat belt of the Northwest and the world. The Northern Pacific Railway runs nearly due west across the State. For more than forty miles across the level plain of the lake bottom the track is without a curve, said to be the longest stretch of straight track in the world. MAPLETON. Distance from Fargo, 12.5 miles. (363.7 miles.) Altitude, 905 feet. Population, 322. Prairie rises imperceptibly toward the west. At Greene, two miles west, the railroad crosses "Maple Ridge," which rises ten feet from the prairie to the east. This was an off-shore bar known as a "spit," built at the time of the Blanchard and Hillsboro stages of Lake Agassiz. CASSELTON. 20.0 miles. (356.2 miles.) Altitude, 931 feet. Popu- lation, 1,207. Crossing Breckenridge Division Great Northern Railway. Broad, level, fertile prairie; the finest wheat land in the world. WHEATLAND. 25.6 miles. (350.6 miles.) Altitude, 992 feet. Level prairie continues west of Casselton, till at Wheatland it rises suddenly fifteen feet onto a conspicuous gravelly ridge sixty rods wide, the Campbell Beach. MAGNOLIA. 29.7 miles. (346.5 miles.) Altitude, 1,078 feet. The watertank stands upon the Herman Beach, the highest shore- line of Lake Agassiz, and the western limit of the Red River Valley. The Tintah and Norcross Beaches were crossed between this point and Wheatland. Beach sand and gravel are taken from an extensive pit which has been opened in the Herman Beach. 335 336 THE STORY OF THE PRAIRIES. BUFFALO. 35.2 miles. (341.0 miles.) Altitude, 1,204 feet. Popu- lation, 213. Buffalo is 212 feet higher than Wheatland, less than ten miles east, and 126 feet higher than Magnolia, five and one-half miles east, at the highest shore-line of Lake Agassiz. This is the Manitoba Escarpment, the continuation -of the Pembina Mountain highland, which formed the western side of the pre-glacial Red River Valley. (See Seventh Chap- ter. ) The traveler can easily distinguish the heavy grade as the engine toils westward, or rolls with easy speed toward the east. At Buffalo a distinct range of low hills is crossed, the Fergus Falls, or Eighth, Moraine. The moraine makes a loop south of the railroad, being crossed again fifteen miles west near Alta. TOWER CITY. 41.0 miles. (335-2 miles.) Altitude, 1,172 feet. Population, 468. Scattered morainic "knobs" give a varied aspect to the prairie, out- lying hills from the moraine just crossed. ORISKA! ,( Barnes County), 46.5 miles. (329.7 miles.) Altitude, 1,269 f eet - One of the ranges of the Fergus Falls Moraine is well shown east of Oriska about two miles. North of Oriska are seen broad, low hills differing in appearance from the "morainic" hills. These are "pre- glacial" hills, that is, they were hills before the Ice Invasion, and while they were passed over by the ice, yet were not leveled down entirely. f . They are therefore "veneered hills," being covered with a mantle of drift. ALTA. 51.3 miles. (324.9 miles.) Altitude, 1,430 feet. More hills of the Fergus Falls Moraine between Oriska and Alta. The elevation at Alta is 161 feet higher than at Oriska, five miles east. This rapid rise means that here was a hillside on the old, or pre-glacial, landscape. Broad hills with smooth surfaces north are veneered hills. Pilot Mound, seven or eight miles north of Alta, is such a veneered hill. Valley City 57.0 miles. (319.2 miles.) Altitude, 1,221 feet. Population, 2,446. Crossing main line of Soo Railway. The railroad descends 209 feet from Alta to the bottom of the Sheyenne Valley, rising again 204 feet to Berea, five miles west. At the edge of the valley east of the city a fragment of the prairie has been cut around by coulees so as to form a flat-topped hill, or "butte." This is capped with drift to a depth of ten or twelve feet. The line of separation between the drift and the under- THE NORTHERN PACIFIC LINES. 337 lying Cretaceous shales can be traced by the difference in the vegetation above and below the line. A similar line can be followed along the sides of the coulee down which the railroad descends to the valley bottom from the east, and also along the sides of the valley of the Sheyenne. Large . and small boulders are strewn upon the top and sides of the "butte," and also along the coulee. Outcroppings of the blue shale can be seen along the sides of the coulee, and in places in the steep sides of the Sheyenne Valley. The Sheyenne Valley has a most interesting history. Here once rolled a mighty river, many times larger than the present small stream which occupies the great valley, because kept at flood by the waters from the great melting Ice-Sheet, during the closing stages of the Glacial Period. This glacial river carved its broad channel deeply into the Cretaceous shales which underlie the drift, and bore the materials thus eroded, together with sand and finer rock-powder from the melting ice, into Lake Agassiz, and there built up the great delta which bears its name. (See Ninth Chapter.) SANBORN. 68.0 miles. (308.2 miles.) Altitude, 1,445 feet. Popu- lation, 259. The hills of the Dovre, or Seventh, Moraine lie west of the Sheyenne River, and cap the hills which border the coulee up which the railroad rises to the prairie from Valley City. Once out upon the prairie numerous steep, rounded knobs are noticed. South of the railroad the hills of the Waconia, or Sixth, Moraine rise from fifty to seventy-five feet above the prairie. This moraine continues south of the railroad and nearly parallel with it for several miles. A long lake extends south from Hobart, and several lakes, one of which is crossed by the railroad a mile east of Sanborn, represent ancient watercourses, probably pre-glacial valleys, which were partially filled with drift. ECKELSON. 74.4 miles. (301.8 miles.) Altitude, 1,464 feet. The railroad crosses Lake Eckelson, which also lies in an old water- way. The hills surrounding the lake rise twenty-five to forty feet. Wave- worn beaches border the lake, and boulders are perched upon the shores, shoved up by the action of ice during winters. The Waconia Moraine is crossed by the railroad just west of Eckelson. URBANA. 78.2 miles. (298.0 miles.) Altitude, 1,471 feet. Between Eckelson and Urbana the railroad crosses another ancient waterway a fourth to a half mile in width, with hills rising forty feet on each side. This old valley is occupied by a lake north of the track, and 338 THE STORY OF THE PRAIRIES. by a slough south. More small lakes lie in the valley to the south and west. SPIRITWOOD. (Stutsman County.) 81.0 miles. (295.2 miles.) Altitude, 1,478 feet. West of Spiritwood the railroad crosses a broad and deep valley having steep sides and a level bottom, extensive hay-meadows occupying the flat bottom. This is the valley in which lies the Spiritwood Chain of Lakes. These lakes lie along a course from five or six miles north of the railroad fifteen miles or more to the northwest. This also is an ancient drainage channel. East of Spiritwood station the low rolling hills are those of the Elysian, or Fifth, Moraine. West of the station are similar hills of the Kiester, or Fourth, Moraine. Jamestown 92.0 miles. (284.2 miles.) Altitude, 1,397 ^ eet - PP~ ulation, 2,853. The buildings of the North Dakota Asylum for the Insane stand upon the west bank of the James River, seen from south window. A fine view of the deep James Valley, its sides serrated with coulees, is afforded as the approach is made to the city. The Valley of the James is not as deep as that of the Sheyenne, being from seventy-five to 125 feet deep, cut in the drift through most of its course, but in places having its bed in the Cretaceous rocks which underlie the drift. Like the Sheyenne Valley, it is a large channel eroded by the flood-waters from the melting Ice-Sheet. From Bloom, five miles east of Jamestown, the highest point between the James and Sheyenne Valleys, the railroad descends 101 feet to Jamestown. West the grade rises 132 feet in five miles, to the prairie. The coulee up which the railroad passes to the prairie to the west fur- nishes a fine illustration of the development of a river system. The coulee, itself a' "young" valley, has its sides serrated with many smaller coulees, still "younger" tributaries. The prairie bordering the Valley is thus being cut up into hills. DEVILS LAKE BRANCH. JAMESTOWN NORTHWARD. The railroad follows the valley of the Pipe Stem River, and a coulee which enters this, to the general level of the prairie at Parkhurst, six miles from Jamestown. The great plateau, the Coteau du Missouri, rises in the west six to ten miles distant, the railroad running at about this distance from the steep slope of its front for twenty-five miles to the northern boundary THE NORTHERN PACIFIC LINES. 339 of Stutsman County. At times the horizon line is rendered broken and irregular along the top of the plateau by the morainic knobs and ridges which lie upon it. Deep coukes also intersect the face of the sloping front of the great highland. The prairie is a fine level expanse such as is common in the broad tracts between moraines. This gently undulating- tract continues northward for forty miles, to Carrington. MELVILLE. (Foster County.) 33.4 miles north from Jamestown. Altitude, 1,602 feet. From this point the railroad diverges from its course parallel with the front of the great plateau. Hawk's Nest, a large outlying hill belong- ing to the plateau, lifts its Iplue and hazy head on the horizon twelve miles west. Carrington. 42.4 miles. Altitude, 1,579 f eet - Population, 1,150. (See p. 236.) New Rockford (Eddy County.) 58.5 miles. Altitude, 1,529 feet. Population, 698. Cross the James River, here a small stream with sluggish current. A splendid expanse of gently undulating prairie extends fifteen to twenty miles both east and west. About two miles north the railroad passes upon a tract of morainic hills. Between this point and Minnewa'ukan, twenty-five miles north, the railroad crosses several morainic belts, which represent the Kiester, Elysian, Waconia, Dovre, Fergus Falls, and Leaf Hills Moraines. Between the broad belts of hills are tracts of nearly level prairie varying from two to seven miles in width. SHEYENNE. 69.6 miles. Altitude, 1,470 feet. A broad inter-morainic belt, traversed by the Sheyenne River. A fine illustration of river terraces is observed west of the railroad bridge, there being two distinct terraces, or "benches," one higher than the other. These are best seen from the west window, or from the rear platform. OBERON. (Benson County.) 78.3 miles. Altitude, 1,559 ^ ee ^- Population, 217. Oberon lies at the eastern end of the Antelope Valley, a fertile inter- morainic tract of gently undulating prairie, from five to six miles wide, lying between ranges of hills from one to three miles wide. No stream occupies this valley, nor is it a glacial drainage course. It is a nearly level belt of prairie between moraines, an inter-morainic tract. Two miles north the railroad crosses a morainic range about two miles in width, then passes upon an inter-morainic tract, on which Lallie, or Fort 340 THE STORY OF THE PRAIRIES. Totten Station, stands. This prairie is about eight miles long and two miles wide, and entirely surrounded by ranges of morainic hills. Minnewaukan. 89.1 miles. Altitude, 1,461 feet. Population, 432. One and one-half miles north of Lallie the railroad passes amongst the hills, which continue for six miles, to Minnewaukan, except for two small inter-morainic areas which are nearly level. Minnewaukan stands on the western shore of Devils Lake, at the northern edge of the great morainic region just crossed. A broad expanse of prairie lies north. LEEDS. 107.3 m il es - Altitude, 1,516 feet. Population, 349. The railroad passes near to the eastern end of Big Butte, or Mauvais Butte, and along the west shore of Lake Ibsen t Big Butte is a large hill which belongs in the same series of hills as Devils Heart and Sully's Hill, south of Devils Lake, large pre-glacial hills which are covered with a mantle of drift, but are not themselves "drift hills'." The western and higher end of the Big Butte is crossed by well marked morainic ridges, with many large boulders. *********** ELDRIDGE. 99.0 miles. (227.2 miles.) Altitude, 1,542 feet. After rising from the Valley of the James River, the great western plateau, the Coteau du Missouri, forms the blue and hazy horizon west. The Antelope, or Third, Moraine is crossed just west of E., the moraine lying along the foot of the great plateau. The engine will be noticed to toil heavily now, as the ascent is made onto the plateau. From three miles west of Eldridge the grade rises 277 feet in a distance of six miles to Windsor. WINDSOR. 108.0 miles. (268.2 miles.) Altitude, 1,839 ^ eet - The slope of the plateau front is marked by many coulees, small lakes and marsh hay-meadows being numerous among the hills of the Antelope Moraine. The eastern edge of the plateau marks the "divide" between the James Valley and Missouri "Slope." The landscape is now a high rolling prairie, with hills rising fifteen to forty feet, and sometimes seventy-five to 100 feet. This is a splendid grazing country. Ranche buildings, and fine herds of cattle, horses and sheep may be seen upon the hillsides and prairies. What have been glacial lake bottoms are now the best of hay-meadows. Some of these flat bottoms show well defined wave-worn beaches around their margins, now nicely grassed over. Some of these lakes probably represent the bottoms of glacial drainage channels. THE NORTHERN PACIFIC LINES. 341 MEDINA. 120.5 miles. (255.7 miles.) Altitude, 1,794 feet. Occasional high morainic hills, but the landscape generally is the "swell-and-sag" topography of glacial regions. Large granite boulders are frequent. Fine hay sloughs and alkaline lakes occupy low places. To the west the railroad crosses a channel in which lies a long lake, well grown with rushes, a long morainic hill of white sand lying in this lake. This is probably a drainage channel by which water from the melting ice- sheet escaped to the Missouri River. CRYSTAL SPRINGS. (Kidder County.) 128.8 miles. (248.2 miles.) Altitude, 1,796 feet. The railroad passes amongst the hills of the Gary, or Second, Moraine, to the west. The main range lies south, but high hills also occur north. Hills rise to a height of 125 feet above prairie. The track runs for several miles along the course of an old glacial channel now occupied by a long irregular lake, and low, marshy lands. TAPPEN. 136.7 miles. (239.5 miles.) Altitude, 1,765 feet. Large hills of the Gary Moraine both north and south of railroad. West of Tappen another broad valley is crossed, having sloping sides and an extensive hay-meadow on its bottom, another glacial drainage channel. DAWSON. 142.0 miles. (234.2 miles.) Altitude, 1,746 feet. Fine level prairie about the town. South are the high hills of the Altamont, or First, Moraine. High hills of this moraine are also seen north in distance. West the railroad makes a long cut through a range of hills belonging to this moraine, then suddenly comes into a broad shallow glacial channel, its bottom marked by lakes and hay marshes. This is Long Lake Valley, an old drainage course which extends south- west to Long Lake, and- the Missouri River. Steele 150.0 miles. (226.2 miles.) Altitude, 1,856 feet. Popu- lation, 185. Fine level tract about Steele. Hills of the Altamont Moraine south. A small ridge crossed west. High rugged hills north. West of Steele a few miles a grassy lake bottom, probably a part of the Long Lake glacial drainage system. That this has been the place of a larger sheet of water is shown by the boulders perched high on its shore. DRISCOLL. (Burleigh County.) 161.0 miles. (215.2 miles.) Alti- tude, 1,873 f eet - Hills of Altamont Moraine south of railroad approaching Driscoll. West the train glides swiftly down to a broad, level, marshy meadow. In 342 THE STORY OF THE PRAIRIES. this valley lies a chain of lakes representing a large channel of glacial drainage which opens southwest into the valley of Apple Creek. STERLING. 169.6 miles. (206.6 miles.) Altitude, 1,812 feet. A deep cut is here made through the high crest of a ridge of the Altamont Moraine, the crest being eighty-two feet above the track, and the highest hill crossed by the railroad between Fargo and Bismarck. Many boulders, some of immense size, lie along the surface. Some of these knobs rise 200 feet above the prairie. Butte-like hills capped with the Fox Hills Sandstone can be seen in the distance. These are hills of erosion, and not drift hills. The last ridge of morainic hills, westward, has now been passed. At McKenzie a fine level hay-meadow is crossed, which belongs to the Apple Creek Valley glacial drainage system, lead- ing to the Missouri River. BURLEIGH. 180.0 miles. (196.2 miles.) Altitude, 1,722 feet. Burleigh stands on a terrace plateau, the old flood-plain of the large glacial river in the valley of which now sluggishly meanders the small Apple Creek. About a mile west of Burleigh the "bottom" of the valley is crossed, about seventy-five feet below the old flood-plain on which Burleigh stands. BISMARCK. 192.7 miles. (183.5 miles.) Altitude, 1,670 feet. Population, 3,319. The State Penitentiary stands upon what appears to be a terrace about fifteen feet above the bottom of the creek to the east, and sixty feet lower than the terrace on which Burleigh stands. The depot at Bismarck stands upon another terrace-like plateau ten or twelve feet higher than that at the Penitentiary, and about two miles west the eleva- tion is about the same as that at Burleigh. The Missouri River is a majestic stream. Its broad sandy bottom spreads out in the distance south. Along its banks the Cretaceous rocks which underlie the drift are exposed in many places. A fine view of a section of these rocks is obtained at the east end of the railroad bridge which spans the river. The strata of shale extend up nearly to the top of the high bank, being capped w r ith a thin mantle of drift. This shows that the broad valley east of Bismarck is really a valley cut in the under- lying rocks, the drift merely forming a surface covering. North and east of Bismarck, along the line of the Bismarck, Washburn, and Great Falls Railway, many fine examples of butte-like hills may be seen, their flat tops capped with the Fox Hills Sandstone.* The extensive coal mines *J. E. Todd. THE NORTHERN PACIFIC LINES. 343 at WiltOn, and other points north, are in a higher series of rocks than the Fox Hills formation. The coal formation is the Laramie. This means that the shelves of sandstone rock which cap the hills would be found deep below the coal to the north. (See Eighteenth Chapter.) Mandan (Morton County.) 199.5 miles. (176.7 miles.) Alti- tude, 1,644 f eet - Population, 1,658. Mandan is located on the broad plain at the mouth of the valley of the Heart River, near where that stream enters the Missouri. The city is picturesquely located among the hills, which have been formed by the deep cutting of the Heart River and its tributaries. The hills rise abruptly 300 to 400 feet above the bottom of the valley. Ascending the Heart Valley, shale and sandstone outcroppings occur in the sides of the drift-capped hills. Numerous drift boulders lie along the bottoms and sides of the coulees. SWEET BRIAR. 214.8 miles. (161.4 miles.) Altitude, 1,799 ^ eet - The railroad passes up Sweet Briar Creek from the valley of the Heart River. A shelf of sandstone outcrops along the north side of the valley. Where the side of the valley is grown over with grass a mark on the hillside shows the edge of the sandstone shelf. Going up the valley the sandstone layer approaches the bottom, and other layers of rock are above. The layers do not become lower, but the bottom of the valley rises. Farther west many capped buttes are observed, and the layer of hard sandstone which caps these becomes lower in relation to the surface, and finally disappears beneath the surface. JUDSON. 220.5 miles. (155.7 miles.) Altitude, 1,948 feet. The railroad still follows the branching coulee, not yet having reached the general prairie surface. Many boulders lie upon the surface, and crags of sandstone project from the hills. NEW SALEM. 227.0 miles. (149.2 miles.) Altitude, 2,160 feet. Population, 229. Here the high prairie is reached. The highest peak, which stands far above the surrounding landscape, shows outcropping horizontal layers of sandstone. Some idea of the amount of earth which has been carried away by erosion can be gained from this, for the projecting edges of this high peak are part of the horizontal layers which once extended over the whole landscape, and have been all carried away except this. Some of this rock is now the sand along the valley of the Missouri River, and some has been carried far down toward the Gulf of Mexico, and some may be resting on its bottom. Occasional large boulders lie upon the 344 THE STORY OF THE PRAIRIES. surface, but most of the finer drift materials has been carried away, the boulders remaining because too heavy to be transported. SIMS. 234.2 miles. (142.0 miles.) Altitude, 1,959 ^ eet - West of Sims the railroad turns abruptly south and descends a coulee six- miles, turning again to traverse another valley for twenty miles to the prairie surface west of Glen Ullin. About Sims are some small hills of drift thickly strewn with boulders, the last drift hills and large boulders observed, westward bound.* A projecting chimney from a coal mine shows that it is the Laramie formation that is now being crossed, a formation which lies above the sandstones and shales farther east, and in which are the North Dakota coal beds. ALMONT. 239.1 miles. (136.3 miles.) Altitude, 1,914 feet. North of railroad a fine "park" of sandstone-capped hills. The lower hills, those which have been eroded more, are rounded at their tops. Farther west a pretty row of beehive-shaped buttes, tipped with red from colored sandstone or clay. Two high, pointed buttes, the first genuine "Bad Land" buttes observed. Same range of hills becomes grass- covered farther west, as the valley becomes less deep and less erosion has occurred. GLEN ULLIN. 265.6 miles. (119.6 miles.) Altitude, 2,067 f eet - Population, 272. Low flat buttes south. Some lofty peaks mar-k the higher level of an older landscape which has been mostly carried away. Eagles Nest is one of these, a few miles west. HEBRON. 269.1 miles. (107.1 miles.) Altitude, 2,157 feet. Pop- ulation, 182. East of Hebron is the divide between streams tributary to the Big Muddy and the Heart on the south, and the Knife River on the north. The streams have not yet cut deeply, and the general landscape is an expanse of grassy prairie. ANTELOPE. (Stark County.) 278.5 miles. (97.2 miles.) Altitude, 2,405 feet. The railroad here runs upon the divide between the Heart and Knife Rivers. Tributaries from both these rivers push up upon the prairie. This is therefore what may be called the high prairie. North of the rail- road stands a pointed butte with naked sides, its crest reach-ing far above the general landscape. Another is south of the track, thinly covered with grass. These isolated peaks, standing alone on the landscape, their tops * See footnote, page 324 THE NORTHERN PACIFIC LINES., 345 composed of horizontal layers of sandstone, show that a vast amount of erosion has occurred, for once the layers of sandstone were continuous over the whole region and the general level of the landscape was above where these tops now are. TAYLOR. 289.5 miles. (86.7 miles.) Altitude, 2,484 feet. Taylor marks the highest point of the railroad east of the divide between the Heart and Little Missouri Rivers. GLADSTONE. 297.2 miles. (79.0 miles.) Altitude, 2,345 feet. Well marked terraces are shown along the streams south and west from the railroad. West toward Lehigh two naked buttes stand south of railroad. One is nearly "worn out," being a mere thumb standing upon the prairie. The other is larger, and has hard sandstone shelves projecting from its sides which protect it from so rapidly wearing away. The flat tops and projecting shoulders of many buttes are thus explained. LEHIGH. 304.0 miles. (72.2 miles.) Altitude, 2,343 feet. The extensive Lehigh Coal Mines on south side railroad. Dickinson 308.6 miles. (67.6 miles.) Altitude, 2,401 feet. Pop- ulation, 2,076. Three broad table lands, or mesas, covered with grass, northeast. Northwest of the city the small hills are crested with sandstone crags. High, naked buttes south in distance. FRYBURG. (Billings County.) 334.3 miles. (41.9 miles.) Altitude, 2,761 feet. Steadily ascending all the way from Mandan, the summit, or divide, between the Heart and Little Missouri Rivers has now been reached. Mandan is at the mouth of the Heart River, with an altitude of 1,644 feet; Fryburg is 117 miles (by section lines, 136 miles by rail) west, with an altitude of 2,761 feet, or 1,116 feet higher. Now, in a distance of twelve miles (by section lines, fourteen miles by rail) a fall of 500 feet is made into the valley of the Little Missouri. Then, in a distance of twenty miles the ascent is made to the high prairie again through a vertical rise of 750 feet to an altitude west of Sentinel Butte of 2,801 feet. Within this narrow but deep valley lies the famed "Bad Lands." And this deep valley is at once the cause and the explanation of the Bad Lands. The Little Missouri River has a steep bed, and it therefore cuts down rapidly. This gives to its inflowing tributaries a high gradient, and these in turn cut their channels rapidly. The result is that the landscape along the course of the Little Missouri is deeply intersected by streams. The flat-topped buttes are the hills which have not yet been worn away 346 THE STORY OF THE PRAIRIES. so as to make their tops round. The high table lands are fragments of the old prairie which has been thus cut up by the streams. The hard sandstone layers, such as were observed in the journey west from Man- dan, give to the hills many of their remarkable features, just as they gave the jutting shoulders and projecting crags to many hills along the Heart River. SULLY SPRINGS. 340.0 miles. (36.2 miles.) Altitude, 2,571 feet. Just west of the station the railroad passes through the Petrified Forest. No shade is afforded by this "forest," and the trees have long since ceased to shed their leaves! These ancient monarchs of a ''dead past," these giants whose branches once wafted in the breezes of the Cretaceous Age, have fallen, and their immense trunks now strew the ground. The stumps on which they grew still stand, buried in the rocky soil in which they grew, mutely testifying to a glory long past, of a "forest primeval." Truly it may be said: "This was the forest primeval. The murmuring pines and the hemlocks, Bearded with moss, and in garments green, distinct in the twilight, Stood like Druids of old, with voices sad and prophetic, Stood like harpers hoar, with beards that rest on their bosoms. Loud from its rocky caverns, the deep-voiced neighboring ocean Spoke, and in accents disconsolate answered the wail of the forest ! Filled was the air with a dreamy and magical light; and the landscape Lay as if new-created in all the freshness of childhood." But this was long, long ago. What we see is the tomb in which the "Forest Primeval" was buried, the strata of rock which were deposited over them, and which have in later time been removed by erosion. On both sides of the railroad many stumps stand where they grew. Immense logs four feet in diameter lie near the track. Pyramid Park lies to the left (south), a magnificent view. Pyramid shaped buttes, large and small, white, naked walls of rock, flat topped tables and smaller rounded cones. Descending the valley the buttes become red on their crests, capped with lava. Rough crags of scoriaceous rock project from many sides. At Scoria the buttes appear as though dyed in blood. The small stream, Sully's Creek, has cut down through a lava bed which was spread over the ancient sea-bottom which was here "before ever the hills were formed." THE NORTHERN PACIFIC LINES. 347 Medora. 348.2 miles. (27.5 miles.) Altitude, 2,261 feet. Medora lies in the "heart of the Bad Lands." The nearly perpen- dicular walls of rock rise 400 feet from the Little Missouri River to their tops, which represent the prairie level. The horizontal strata of the Cretaceous rocks are magnificently exposed to view. Sandstone, clay, shale, and lignite coal lie in alternate bands from base to top of the buttes. SENTINEL BUTTE. 365.8 miles. (11.4 miles.) Altitude, 2,703 feet. After crossing the river the railroad follows down its west bank to the mouth of Andrew's Creek, up which it passes to the high prairie. High rugged buttes rise steep on either side the narrow deep valley. A fine view of a group of haystack-shaped buttes is obtained from the north window, several miles west. Sentinel Butte stands high above the sur- rounding prairie six miles south of the station. The top of this butte is said to be the highest point in North Dakota. The butte gets its name from a pathetic incident in the campaign of 'the lamented General Custer against Sitting Bull, in 1876. The "pass" through the top, which may be seen from east of the station, was guarded by two soldiers during a night when it was anticipated an attack might be made by the Indians upon the army encamped upon the plain north of the butte. In the morn- ing when relief was sent to the guards their bodies were found pierced with arrows. Their bodies were buried one on either side of the pass they had guarded, and their graves are marked by two large piles of stones. STATE LINE. 376.2 miles. Altitude, 2,811 feet. The highest point of the Northern Pacific Railway in North Dakota is reached at the instant of crossing the State line into Montana. The western boundary is almost exactly on the watershed, or divide, between the Little Missouri and Beaver Creek, the railroad then descending to the valley of the Yellowstone. *********** FARGO SOUTHWESTERN LINE. Fargo (See p. 218.) The railroad crosses the level plain of the bottom of Lake Agassiz for twenty-five miles. LEONARD. Distance from Fargo, 28.0 miles. Altitude, 1,052 feet. The railroad here passes from the level lake bottom upon the plateau of the Sheyenne Delta, rising more than eighty feet in a distance of two miles, and crossing the McCauleyville, Campbell and Tintah Beaches of 348 THE STORY OF THE PRAIRIES. Lake Agassiz, which extend along the delta front. South the delta sand is in places piled into dunes by the action of the wind. SHELDON. (Ransom County.) 40.5 miles. Altitude, 1,079 ^ eet - Population, 318. Surface of the delta slightly undulating. The sand is piled into dunes, notably south. The western edge of the delta, and of the lake, is at Sheldon. The change in the landscape is at once seen toward the west, where the hills become the familiar drift hills, with occasional high knobs. Lisbon 55.3 miles. Altitude, 1,089 feet Population, 1,046. Deep valley of the Sheyenne River. A well marked broad terrace of the larger glacial river on west side of valley. Suburbs of the city built upon its top. Terrace seen also north of the city on east side of valley, from west. The Dovre Moraine lies along the western bluffs of the river. A cut in the gravelly terrace west of the city shows finely stratified sands. From the prairie west of the valley the broad outlines of White Stone Hill may be seen eight miles- south. This is a pre-glacial hill rising 150 feet above the prairie and veneered with drift. ELLIOTT. 62.8 miles. Altitude, 1,330 feet. West of Elliott is another pre-glacial hill rising about sixty feet. ENGLEVALE. 67.7 miles. Altitude, 1,342 feet. An old channel about three miles in width, having a flat bottom, was occupied by the Sheyenne River when that stream discharged into Lake Sargent. A deeper part of the channel a mile to a mile and a half west of Englevale is known as the Big Slough. When the glacial Sheyenne River flowed here the ice of the great Ice-Sheet had not melted off from the .region about Lisbon, and the Dovre Moraine west of Lisbon was being formed. VERONA. (LaMoure County.) 75.6 miles. Altitude, 1,385 feet. Between the channel of the glacial Sheyenne River and Verona the Waconia, or Sixth, Moraine is crossed. LaMoure 87.0 miles. Altitude, 1,308 feet. Population, 457.- The city lies in the broad valley of the James River. Terraces occur along the west side of the valley. Fine undulating prairie west of the valley. BERLIN. 97.0 miles. Altitude, 1,469 feet. That trees can be successfully grown on these prairies is proven by a fine grove of cultivated trees at Berlin. Good farm buildings bespeak the thrift of the farmers in this section. In the distance along the western horizon rises the blue outline of the great plateau, the Coteau du Missouri. THE NORTHERN PACIFIC LINES. 349 EDGELEY. 108.5 miles. Altitude, 1,567 feet. Population, 306. The Coteau highland rises eight to ten miles west. East of Edgeley the hills of the Antelope, or Third Moraine are crossed. Drift hills also occur along the foot of the plateau west. MONANGO. (Dickey County.) C, M. & St. P. Railway. 122.6 miles. Altitude, 1,501 feet. Low round hills and hollows of the Antelope Moraine crossed near Monango. Crossing Bismarck Branch of Soo Railway. Ellendale 135.0 miles. Altitude, 1,446 feet. Population, 750. Low rolling hills of the Antelope Moraine six miles west. In the distance fifteen to twenty miles west the highland of the Coteau du Missouri rises suddenly to an elevation of more than 2,000 feet. All about Ellendale, and east and south is a fine farming section. CHAPTER THE THIRTIETH. GEOLOGY FROM A CAR WINDOW THE SOO LINE. FAIRMOUNT. (Richland County.) Distance from St. Paul, 201.3 miles. (Distance east of Portal, 359.0 miles.) Altitude, 983 feet. Population, 284. The Soo Line enters North Dakota at Fairmount, in the southeast corner of the State, and passes in a northwesterly direction, crossing the International Boundary at Portal. Fairmount is located a mile and a half west of the Bois des Sioux River, on the level axis-plain of Lake Agassiz. OSWALD. Distance from Fairmount, 5.8 miles. (Distance east of Portal, 353.2 miles.) Altitude, 987 feet. Level prairie from Fairmount westward, but now begins to be slightly undulating, and soon broken into short, choppy sand billows, the wind blown sands of the Sheyenne Delta. HANKINSON. 14.5 miles. (344-5 miles.) Altitude, 1,068 feet. Population, 713. Rising conspicuously from the prairie south of Hankinson stands a large dune 150 feet high, covered scatteringly with trees, known as Lightning's Nest. A tract of dunes from ten to thirty feet high and mostly covered with grass extends to the northwest. Lightning's Nest has been built up from the sands of the Herman Shore-Line, where it crosses the Sheyenne Delta. The Norcross Beach passes close east of Hankinson as a well defined sand ridge. The hills of the Dovre Moraine are a prominent range south and west. MANTADOR. 22.0 miles. (337.0 miles.) Altitude, 1,027 ^ ee ^- The railroad crosses the tract of dunes which extends from Light- ning's Nest. Beyond the dune tract fine level prairie. Wild Rice River is crossed north of the dune tract. WYNDMERE. 31.9 miles. (327.1 miles.) Altitude, 1,060 feet. Fine level prairie about Wyndmere and south. North the surface is 35 THE SOO LINE. 351 broken into swells and hollows of wind-blown sand. Farther north still the knolls become sharper in outline and the road-bed cuts through them, showing the white delta and beach sand of which they are composed. For forty-five miles from -Hankinson to Anselm the Soo Road passes over the delta plain of the glacial Sheyenne River. The sand of this delta is piled by the wind into knolls often fifty feet high, and this gives the landscape its peculiar appearance. Nothing but wind-blown sand could form hills such as these. Where the surface does not become grass- covered, the sand drifts precisely as does dry snow, the hills slowly travel- ing across the plain as the sand is continually carried up over the crests of the hills and falls down the other side. SANDOUN. (Ransom County.) 44.2 miles. (314.8 miles.) Alti- tude, 1,074 feet. Billowy dunes rise ten to fifty feet or more, the hollows between being often filled with water. Hills mostly thinly covered with grass. VENLO. 53.8 miles. (305.2 miles.) Altitude, 1,062 feet. The Sheyenne River, which runs at the west side of the Sheyenne Delta, is crossed between this and the next station. The Herman Beach, the highest shore-line of Lake Agassiz, is a little south of the River, but is so much broken up that it is not easily recognized. The Big Bend of the Sheyenne, ten miles south, marks the place where the great glacial river discharged into Lake Agassiz at the time the large delta was built. ANSELM. 57.6 miles. (301.4 miles.) Altitude, 1,085 ^ eet - Cuts in the valley side made in grading the railroad show the strati- fied sands and gravel of the delta. High dunes rise in the distance east. Anselm is just off from the delta plain, and the different aspect of the landscape is at once apparent. The prairie becomes gently rolling with occasional high round knobs, morainic hills. There are no hills of this character, no morainic "knobs," on the area covered by Lake Agassiz, which is what is known as the "Red River Valley." ENDERLIN. 66.1 miles. (292.9 miles.) Altitude, 1,082 feet. Pop- ulation, 636. Enderlin is situated in the valley of the Maple River, here a deep glacial valley such as the Sheyenne. Fine prairie north to Lucca and Fingal. FINGAL. (Barnes County.) 79.2 miles. (279.8 miles.) Altitude, 1,277 f eet - Population, 376. The rapid rise in elevation here is due to the railroad passing upon 352 THE STORY OF THE PRAIRIES. the Manitoba Escarpment, the highland which formed the western side of the pre-glacial Red River Valley. CUBA. 84.8 miles. (274.2 miles.) Altitude, 1,352 feet. The hills of the southern loop of the Fergus Falls Moraine appear in the east and north. LANONA. 90.9 miles. (268.1 miles.) Altitude, 1,387 feet. Knobs and irregularly-shaped hills of the Fergus Falls Moraine. Valley City 95.8 miles. (263.2 miles.) Altitude, 1,227 feet - Population, 2,446. Valley City is located in the broad and deep valley of the Sheyenne. The station is two miles north of the city. The hills which form the sides of the valley rise 150 to 200 feet. Much of the material which was eroded by the great glacial stream which cut this large valley makes up the delta plateau which has just been crossed. (See p. 219.) ROGERS. 109.5 miles. (249.5 miles.) Altitude, 1,422 feet. The railroad ascends by a narrow, crooked, deep, and boulder-strewn coulee nine miles to the beautiful level prairie about Rogers. A rise of nearly 200 feet has been made in this distance. Blue shale is exposed in many cuts. LEAL. 115.4 miles. (243.6 miles.) Altitude, 1,465 feet. Morainic hills of the Waconia, or Sixth, Moraine. WIMBLEDON. 123.6 miles. (235.4 miles.) Altitude, 1,468 feet. Population, 226. Fine level tract of prairie lying between the Waconia Moraine, which is seen in the distance to the east, and the Elysian, or Fifth, Moraine, the low swells of which are seen here. Shallow lakes occupy many hollows. COURTNEY. (Stutsman County), 129.8 miles. (229.2 miles.) Altitude, 1,523 feet. Population, 346. Broad tract of prairie again, between the Elysian and Kiester, or Fourth, Moraines. Many boulders strew the prairies, and shallow lakes without outlets show that as yet drainage systems have not become established. KENSAL. 139.4 miles. (219.6 miles.) Altitude, 1,541 feet. Kensal is situated among the hills of the well-marked Elysian Mo- raine. From here the railroad descends to the Valley of the James River. The course of the river for more than twenty miles lies amid the hills of the Kiester Moraine, the hills and the river seeming to be in a struggle for the mastery! South of the railroad bridge the river broadens gut to form Arrowood Lake, the valley being blocked by the THE SOO LINE. 353 drift, which nearly fills it. North of the bridge the river is a broad, sluggish, pooling, lake-like stream. Fine examples of morainic hills are here displayed. Some high knobs of this moraine are nearly 200 feet high. Upon rising from the Valley of the James the great Missouri Plateau, the Coteau du Missouri, appears in the distance west about twenty miles. BORDULAC. (Foster County), 152.4 miles. (206.6 miles.) Alti- tude, 1,530 feet. Low morainic hills of the Kiester Moraine in distance east. Lakes George and Bordulac are broad shallow glacial pans, their waters hemmed in by low morainic hills. Carrington 161.1 miles. (197.9 miles.) Altitude, 1,579 ^ eet - Population, 1,150. The Soo Railway passes through a fine tract of farming country in Foster and Wells Counties, along the upper James and Sheyenne Rivers. The highland of the Missouri Plateau rises thirty miles west. In the distance to the east the high knobs of the Kiester Moraine can be seen, fifteen miles away. Hawk's Nest, an outlying fragment of the great plateau, stands fifteen miles southwest. It is a high drift-covered pinnacle of the old landscape before the Ice Age. LEMERT. 168.8 miles. (190.2 miles.) Altitude, 1,594 feet. Small ranges of low morainic hills cross the broad prairies. Many large granite boulders are strewn along the. track between Lemert and Cathay. It is noticeable that many of these have one side planed off smooth and flat. When examined closely many of these flat surfaces are found to be marked with parallel lines, or striations. This shows that they have been carried long distances, and planed off in the process of being shoved over hard surfaces. CATHAY. (Wells County), 176.2 miles. (182.8 miles.) Altitude, 1,584 feet. Beautifully undulating and rolling crest of the moraine which was noticed west of Carrington now plainly in view west. EMRICK. 181.6 miles. (177.2 miles.) Altitude, 1,597 feet. Morainic ridges west, probably belonging to the Antelope, or Third, Moraine. Fessenden 188.6 miles. (170.4 miles.) Altitude, 1,610 feet. Altitude James River, low water, 1,591 feet. Population, 637. Low morainic ridges south and west, well defined. Occasional hum- mocky knobs, such as are characteristic of terminal moraines. The mo- 354 THE STORY OF THE PRAIRIES. raine which was crossed at the James River, and which filled its valley, is still seen in the east and north frbm about Fessenden. MANFRED. 195.0 miles. (164.0 miles.) Altitude, 1,605 feet. A prominent cluster of hills west, a morainic heap, the crests of the ridges giving the moraine a rugged appearance. A small "glacial lake" hemmed in by the hills lies close north. HARVEY. 205.0 miles. (154.0 miles.) Altitude, 1,596 feet. Shey- enne River, low water, 1,527 feet. Population, 590. Low water of the Sheyenne is sixty-nine feet below the prairie sur- face, while that of the James, less than ten miles south, is only nineteen feet. The Sheyenne is thus shown to have been the great avenue of escape for the waters of the melting ice-sheet, from this portion of the State. Moraine with high rugged hills west. Moraine also east in dis- tance. Fine tract of prairie intervening between. ANAMOOSE. (McHenry County), 221.5 miles. (137.5 miles.) Al- titude, 1,620 feet. Population, 430. Approaching Anamoose high and rugged moraines lie on both sides of the railroad. A lake hemmed in by the hills south. High knobs mark the surrounding hills. West of Anamoose, east of the railroad, is a large valley having distinct terraces on its sides, and a broad, flat bottom, with no stream upon it, and many boulders scattered upon the terraces. Such a deep and well-defined valley, having no stream on its bottom, and having well marked terraces, shows by its form that it is the channel of a glacial stream, a stream which ceased when the waters ' from the melting ice-sheet had disappeared, and the terraces mark the flood-plain of the stream during its earlier stages, before its channel had been cut down to the present bottom. Still further west the railroad runs upon the bottom of a broad level channel, having extensive hay meadows and shallow lakes along its course. A dry lake southwest from Balfour lies in this channel, which is the old southern outlet of Lake Souris, at an earlier stage than that when its waters escaped to the Sheyenne by the Big Coulee outlet. BALFOUR. 236.1 miles. (122.9 miles.) Altitude, 1,613 feet. A few rods west of the station the track crosses the famous "Balfour Ridge," a beach of gravel and sand, which extends from two to three miles south of Balfour, north by northwest for about fifteen miles, to the Mouse River, at Pendroy. Where the railroad crosses it a section showing the sand and gravel in layers is exposed. The ridge rises six to eight feet above the prairie at its southern end, and becomes gradually THE SOO LINE. 355 higher toward the north. It is about thirty feet high at its northern end. The smooth and uniform surface and sloping sides, about equal in height, make it appear much like a railroad grading, and it has been surveyed and set apart for use as a public highway. The level crest of the ridge can be seen from the east window for some distance west of Balfour. Southwest of Balfour about twenty miles rises the blue and hazy head of Dog Den Butte. This is a large outlying hill of the Mis- souri Plateau. Its crest is crossed by several morainic ridges, and small lakes are on its top. The low and level prairie east of Dog Den is known as the "alkali flats." Many shallow alkaline lakes with shores of white sand lie along the tract. VOLTAIRE. 251.3 miles. (107.7 miles.) Altitude, 1,587 feet. Westward from Balfour the old outlet channel of Lake Souris is a mile or more in width, seen west of the railroad. Extensive hay-meadows lie upon its bottom. From Voltaire the railroad descends to the Mouse River "bottoms" by the steep, sharp valley of Spring Creek. This coulee is a notch or channel cut in the bottom of the old outlet channel since the waters of Lake Souris disappeared. VELVA. 256.3 miles. (102.7 miles.) Altitude, 1,525 feet. West of Velva deep coulees with steep sides border the valley of the Mouse. The highland along the valley on the west was the shore of Lake Souris. Minot (Ward County), 277.4 miles. (81.6 miles.) Altitude, 1,557 ^ eet - Population, 1,277. The valley of the Mouse is a great trough eroded by a great glacial river, its bottom being one to two miles across, and the hills forming the sides of the valley rising 1 50 to 200 feet above the flat bottom. The sides of the valley are beautifully serrated by little coulees, as though a giant hand had drawn a coarse comb across the hillsides and made the little furrows, or coulees. (See p. 206.) BURLINGTON. 285.2 miles. (73. 8 miles.) Altitude, 1,590 feet. Here is the point of meeting of the Des Lacs and Mouse Rivers, both of which flow in large glacial valleys. The Soo Road follows the course of the Des Lacs Valley, traversing the bottom of the valley more than forty miles. The valley was eroded by the glacial flood waters deeply into the shales which underlie the drift. Coal mines are opened by tun- neling from the hillsides along the valley. The Burlington mines are extensively operated. (See Fig. 71, p. 167.) FOXHOLM. 295.2 miles. (63.8 miles.) Altitude, 1,657 feet. 356 THE STORY OF THE PRAIRIES. Fine examples of terraces, marking the flood plain of the river at an earlier time before its valley had been eroded to the present bottom. One such terrace plateau lies on the east side of the railroad west of Burling- ton, rising twelve to fifteen feet above the railroad grade. It is almost per- fectly flat on top, and strewn with many boulders. West of Foxholm another fragment of the older flood plain occurs, about ten feet high. Approaching Carpio, the terrace is well developed on west side of valley. CARPIO. 306.6 miles. (55.4 miles.) Altitude, 1,696 feet. Layers of brown sandstone outcrop in the sides of the high banks west. Black layers indicate lignite coal in thin seams. Nearing Donny- brook are seen fine examples of "alluvial fans," soil carried down to the valley bottom by streams flowing in the steep coulees and spreading out upon the flat plain. DONNYBROOK. 3i2.7miles. (46.3 miles.) Altitude, 1,760 feet. The town stands upon one of the alluvial fans just mentioned. On the west side of the valley, where two coulees enter, a butte has been formed, its sides steep, its top flat. Sandstone and clay are exposed in horizontal layers on the south and east sides. On the east side of the valley opposite Donnybrook a line may be traced along the side of the valley about half way to the top of the hillside, by the difference in the vegetation above and below the line. This marks the depth of the drift which overlies the sandstones and shales which are the "bed rock." The Des Lacs River is a very small stream, little more than a meadow ditch which a small boy could jump across. Its course is very crooked, show- ing that it flows very slowly. Such is the modern representative of the great glacial stream which carved the deep broad valley on the flat bottom of which this tiny rivulet now meanders. The bottom of the Des Lacs valley in the upper half of its course is covered by a series of lakes. These are there simply because there is not enough fall to the bottom of the valley to cause the water to run. The first of these lakes is a pretty sheet extending from one side of the valley to the other and about two miles long, lying about eight miles west of Donnybrook. KENMARE. 327.4 miles. (3 1.6 miles.) Altitude, 1,799 feet. Alti- tude, low water, Des Lacs Lake, 1,783 feet. Altitude, top of hill adjoin- ing, 1,950 feet. Population, 300. Kenmare stands upon the hillside overlooking the second lake. The lake covers the entire width of the valley bottom so that the shores come down abruptly to the water's edge. A hay-meadow lies upon the valley bottom between the second and third lakes, the latter being a long ribbon THE SOO LINE. 357 of water half a mile to a mile in width, and extending along the valley bottom thirty miles to the International Boundary. Along the sides of the valley at Kenmare are the machinery and tracks, and the openings to the very extensive coal mines for which Kenmare and the State of North Dakota are noted. Sandstone rock suitable for building purposes is obtained from the hillsides, evidences of w r hich are seen in the substantial buildings constructed from this stone in the city of Kenmare. West of Kenmare the railroad leaves the Des Lacs valley, laboriously climbing up the side of the valley and emerging upon the beautiful level prairie. At once the lake has disappeared from sight, for the eye scans the prairie only to look directly across the valley, which lies entirely below the gen- eral prairie level, a deep flat bottomed trough cut in the great plain. If any further proof were needed that this is a valley of glacial erosion this would serve the purpose, for the country along the stream course would be cut into hills if the valley had been cut by the ordinary development of a drainage system. BOWBELLS. 339.6 miles. (19.4 miles.) Altitude, 1,958 feet. Pop- ulation, 398. A splendid prairie, unbroken by any coulee or mark of drainage. The horizon line in the east is a straight line unbroken by any hill or eleva- tion. In the west, twenty miles a\vay, rises the highland of the great Coteau du Missouri. The horizon line in the west is rendered undulating and wavy by the knobs and ridges of the Altamont, or First, Moraine, which lies along the plateau top. FLAXTON. 349.3 miles. (9.7 miles.) Altitude, 1,956 feet. Between Bowbells and Flaxton the railroad crosses a low belt of morainic hills, which give to the prairie a gently rolling aspect. A long slough t a lake during very wet seasons, a fine hay-meadow usually, has been a lake, the waves of which have beaten upon the shores forming a wave-cut terrace along its border. It is a relic of the flood waters from the melting ice-sheet. PORTAL. 359.0 miles. Altitude, 1,954 feet. Portal, as its name indicates, is the "gate city," standing upon the International Boundary between North Dakota and Assiniboia. A range of sandy morainic hills is crossed by the railroad just as it passes into Canada. Extending far south is a fine tract of prairie. The "Hills," the great Missouri Plateau, rise majestically against the sky twenty miles distant in the west. 358 THE STORY OF THE PRAIRIES. BISMARCK EXTENSION HANKINSON. (Richland County), 14.5 miles from Fairmount. LIDGERWOOD. 26.6 miles. Altitude, 1,090 feet. Population, 585. About four miles west of Hankinson is the Herman Shore-Line of Lake Agassiz. Three miles southwest the hills of the Dovre Moraine are 100 to 150 feet high. This moraine lies south of the railroad about two miles and parallel with it for nearly twenty-five miles. The Herman Beach is about six miles north of Lidgerwood. On the broad flat prairie between the moraine on the south and the Herman Shore north are many broad shallow lakes. RANSOM. (Sargent County), 38.0 miles. Altitude, 1,128 feet. Cross the Wild Rice River. The Dovre Moraine is here crossed by the railroad, a belt of rolling hills about a mile wide. Many granite boulders occur scattered upon the prairie east and west of the moraine. On the western side of the moraine was the eastern shore of Lake Sar- gent. (See p. 117.) Forman 49.7 miles. Altitude, 1,247 f eet - Population, 257. At Forman the depth of the water of Lake Sargent was about fifty feet. At Perry, six miles east, it was about 100 feet, and along the east- ern side of the lake near the Dovre Moraine, about 150 feet. Many boulders are scattered upon these prairies. NICHOLSON. 61.3 miles. Altitude, 1,305 feet. Nicholson is near the western shore of Lake Sargent. To the north is a hill two to three miles long, north and south, covered with a smooth veneering of drift, a pre-glacial hill. Westward from Nicholson the railroad crosses a belt of morainic hills about six miles in width, a com- pound moraine representing the Waconia, Elysian, and Kiester Moraines. OAKES. (Dickey County), 72.6 miles. Altitude, 1,320 feet. Popu- lation, 668. West of the compound moraine mentioned the northern end of the plain of the bottom of Lake Dakota is crossed, here about six miles in width. The James River flows south across this old lake bottom. (P. 116.) MERRICOURT. 106.3 miles. Altitude, 1,644 feet- Near the crossing of the C, M. & St. P. Railway, ten miles east, a low range of hills, the Antelope Moraine, is crossed. The landscape rises quite rapidly here toward the top of the great Plateau, the Coteau du Missouri. Low, long, undulating and rolling swells, the hills of the THE SOO LINE. 359 Gary Moraine, occur westward toward Kulm. The train toils heavily up the steep grade. The broad rolling hills rise twenty-five to forty feet above the hollows, and the crests stand 400 feet above the plain to the east. KULM. (LaMoure County), 118.8 miles. Altitude, 1,966 feet. Population, 463. Kulm is on the high prairie of the Missouri Plateau, a gently roll- ing, broad expanse of prairie. Northwest from Kulm the landscape is marked by broadly rolling swells. Extensive hay-meadows in the low, broad sloughs. Many small granite boulders lie upon the surface. Far- ther west, in southeastern Logan County, the landscape is more rough, hills steep, long hay-meadows in old glacial drainage channels. Railroad often cuts through the hills, showing gravel and drift clay. Boulders strewn upon surface. West of this belt of hills the surface is more smooth. Wave cliffs and terraces on the sides of sloughs show w r here have been lakes during the time of the melting of the ice-sheet. Frequent alkali lakes among the hills west. LEHR. (Mclntosh County), 139.3 miles. Altitude, 2,017 feet. The Altamont Moraine is crossed, with hills high and steep. Rail- road winds among the hills, with frequent cuts. Hills 150 to 200 feet high, with steep boulder-strewn sides. Railroad follows old drainage channel, with steep grade westward, rising onto the immense ridge or belt of the Altamont Moraine. WISHEK. 149.8 miles. Altitude, 2,010 feet. Between Lehr and Wishek is one of the most majestic developments of a terminal moraine crossed by any line of railway in the State. The height of the moraine is from 2,000 to 2,075 ^ ee ^ above sea level and rises 200 feet above the nearly level prairie on either side. The belt where crossed by the railroad east of Wishek is about eight miles in width, and is one of the most striking morainic regions in the State. The deep, well-marked valleys having no streams of importance in them indicate the action of glacial flood waters at the time these and later ranges of morainic hills were being formed along the edge of the Great Ice-Sheet. Many deep hollows containing lakes, having clearly cut ter- races on their shores showing the higher stage of the water at a former time, broad hay-meadows, which were once lake bottoms, and long chan- nels marked by high terrace flood-plains, tell of the great amount of water which once was here. The hills are very high and steep, and strewn with boulders. The landscape is one almost inaccessible to travel 360 THE STORY OF THE PRAIRIES. through overland, except on horseback. And then the traveler is very likely to lose his way, the high knob which was taken as a guide treacher- ously allowing another to be mistaken for it! A broad valley, an old drainage channel from the Altamont Moraine to the Missouri River, lies west of Wishek, along which the railroad runs. This valley is broad and deep, having flat bottom and well-defined terraces along its sides. A branch of the railroad south to Ashley follows this channel. NAPOLEON. (Logan County), 171.0 miles. Altitude, 1,951 feet. From Wishek the railroad runs nearly northwest to Napoleon, follow- ing the bottoms of glacial channels, broad and flat bottoms with small streams, or none at all. The sides of the valleys are steep, owing to the Fox Hills Sandstone, which forms the surface layer of rock, thinly over- laid with drift. Two large drainage channels meet at Berry Lake west of Napoleon. A terrace of gravel and sand fills the lower part of these valleys to a height of about twenty-five feet above the water of the lake. Traces of this terrace are seen along the valley running west. CAMPBELL. (Emmons County), 180.9 miles. Altitude, 1,896 feet. Terraces well shown along valley, which is followed by the railroad. A cut shows brown shaly sandstone in thin strata. BRADDOCK. 187.9 miles. Altitude, 1,860 feet. From Campbell west the railroad follows the deep, broad valley, marked terraces occurring along its sides. The valley is more than 100 feet below the general level west of Campbell. Morainic hills occur north, but these are largely concealed from view by the high valley walls. *********** ASHLEY. (Mclntosh County), 167.8 miles. Altitude, 1,998 feet. Ashley is eighteen miles southeast of Wishek. The railroad follows large glacial drainage channels through most of the distance. Many conspicuous channels of this character occur in this neighborhood. No part of the State offers a more interesting field for the study of glacial drainage than this region, in Mclntosh and Logan Counties. High boul- dery terraces and chains of lakes along these old lines of drainage give the landscape a unique appearance. The Altamont Moraine, which was crossed east of Wishek, extends nearly parallel with the line of the rail- road as far south as Ashley, when it extends eastward to about the edge of the great Missouri Plateau in southwestern Dickey County. The hills of this range rise from 100 to 150 feet in height between Wishek and Ashley, and in the southwestern township of Dickey County the higher knobs are nearly 200 feet high, their high crests standing conspicuously THE SOO LINE. 361 against the sky, as seen from the east, 500 to 600 feet above the plain to the east. Ashley stands upon a nearly level plain about six miles square, having a deep, fine, silt-like soil, as though it had been the bottom of a lake for a long time. A larger plain of similar character, though not as nearly level, extends over a large region west of the high morainic range. APPENDIX. TABLES SHOWING AVERAGE PRECIPITATION BY YEARS AND BY STATIONS, ALSO THE MONTHLY PRECIPITATION FOR APRIL, MAY, AND JUNE, AND THE MEAN ANNUAL PRECIPI- TATION FOR ALL STATIONS. Bismarck Fort Abercrombie Fort Buford Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1861 3-95 1.82 .04 45 4.20 3-72 45 83 2.16 3 2 1.36 1.50 2.OO .70 43 74 1.70 6.67 1.61 .87 38 83 .20 2.14 2.48 4-3 2 4.04 3 4.2O 2. 2O 1-70 3-*7 56 2.06 1.85 95 .26 1.72 1.46 2.00 6.83 3-05 1.02 2.OI 4.10 IO.I5 3-65 8.16 2.96 50 23-34 11.38 13.40 16.85 17-52 12.54 19.66 19.47 22.73 21.37 15.20 27.82 8-59 1862 1863 1864 1865 ... 1866 1867 . . .06 33 65 .00 45 i-55 I.2S IS 1.83 .10 13 1-85 2-75 74 i-34 94 .48 1.30 1.71 2.25 1.09 .61 .60 .60 1.88 42 1.79 1.78 3-92 2-43 1. 12 6.60 1. 80 i-39 4.00 4-5 2.60 2.56 4.02 1. 00 1.61 59 .14 1.02 i-44 i-59 .96 2.69 i-59 1.49 1.27 3-21 .86 77 .90 J -75 2.69 1.02 3-46 J-75 2.04 3-i5 3-35 5-46 3-44 1.87 97 99 6.05 93 3- 2 5 6-75 1.03 5-23 7.08 11.50 9.41 9.90 8.19 16.80 7" 58 14-85 12.34 12.29 19.67 23-25 14.90 12.73 10.82 7-37 !5-5 6 10.24 J 5-43 14.74 8.46 14.24 18.98 1868 . 1869 ... 1870 . . . 1871 .. . 1872 .. . 1873 ... 1874 ... 1875 ... 1876 .. . 1877 ... 1878 ... 1879 ... 1880 .. . 1881 ... 1882 ... 1883 ... 884 ... 1889 '' ' 1890 .. . 1891 .. . Av. Annual for each Station 4.22 2-77 1.32 5-7i 2.60 3-65 i. 02 3-56 i-57 2.20 3 .2I 1.49 i-S 2 .11 .26 .68 2.40 3-40 5-74 4-15 3-15 3-67 2.76 2.27 3-46 1-15 2.56 .92 i-73 2.19 .70 3-35 57 2.92 5.02 1.24 4.60 2.78 4-97 3-32 4.11 3.88 3-84 2-63 2-39 2.03 85 5-77 1.03 8.40 4.19 27.52 30.92 17.68 20.23 20.61 19-75 !5-7 6 2i-33 15.66 23-3 6 13.08 13.26 l6 -33 16.51 11.03 15-75 20.50 17.68 362 363 TABLES SHOWING AVERAGE PRECIPITATION Continued. Fort Pembina Fort Totten Average Annual f<>r all Stations Apr. May June An'l. Apr. May June An'l. 1861 1862 1863 .. 1864 .. 1 86s 1866 1867 1868 I =; 4.8 1860 j.^ .40 16 . o? 1870 4.0 3 tr en i ^ 20 1^.20 1871 i 87 I 04 f QI I 7 O 3 i T. . 47 1872 .. . 2.OO 1.90 2.09 17.19 3-J5 I.IO 1.66 19.00 20. 20 1873 ... 39 2. II 2.91 14.05 57 3-75 4.68 17.30 15-72 1874 ... .20 !-55 3-4i 11.88 .62 i-43 2.52 16.71 12 .06 1875 ... 47 1.87 3-83 J3-53 i-54 2 -77 6.25 22.17 19.52 1876 . . . -49 6-55 3-43 25-75 .42 3-92 I.IO 14.40 18.40 1877 ... .68 4-15 9-85 21.67 47 4.41 5- 12 18.84 17 .62 1878 ... 5.78 2.52 3-57 ^3-83 5-i9 2-38 2.90 22.45 25-50 1879 ..- 34 i-54 3-9o J9-3 1 .14 3.80 3-9 *9-T-5 19.68 1880 . . . 59 7.98 4-98 27-35 .24 4-38 3- 1 ? 22.24 2 3- I 5 1881 .. . .70 3-94 3.10 19.26 30 2-05. 3-88 18.15 17 .02 1882 . . . 50 2.50 3-34 2.40 1.30 4-3 18.58 17-55 1883 ... 1.67 i. 80 1.26 1.71 1.79 1.42 J7-93 14.80 1884 ... 1.86 !- 2 5 2.38 1.20 2.50 17.36 16.03 1885 ... 5-90 1.40 J-95 *7-37 3-7 1.70 3-45 18.69 16.18 1886 .. . 2.85 i-55 3.60 29.24 85 2-75 2 -79 14.22 16.24 1887 . . . 2.80 3-73 3-94 23-3 6 63 i-57 5-57 I 9-3S 18.62 1888 .. . .69 56 8.41 17.99 -65 .60 7.41 16.13 16.34 1889 .. . 7i .10 1.42 n-75 78 .62 1.56 10.51 10.44 1890 . . . 1.41 1.07 5-84 1.97 -79 6.84 .... 14.99 1891 .. . 1x6 2.^8 6 34 2 ^ Q"? 21.77 Av. Annual for each Station 17.77 Amenia Ashley Beach Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1892 .. . 3.12 i-95 2.42 1.24 4-92 2.66 i-54 97 1.91 45 3-84 1. 21 2-34 1.27 2.14 1.30 .81 2.92 3-34 77 2-44 4.22 .30 .89 1.86 1.74 2.90 4.71 7-44 5-9 4.20 4-63 .68 4-33 .96 2.94 2.60 1.48 4.72 3.56 .... 1893 .. . 15.68 18.73 16.64 25-58 16.55 9.40 17.88 13-85 14.00 16.43 1894 .. . 1895 .. . 1896 ... 1897 ..- 1898 .. . l899 .. 6.41 1.49 i. 60 3.56 1. 12 3-20 4.13 74 .69 3- J 9 3-36 1.32 5-45 4.06 t. 3 8 2.25 6.85 2-51 3-27 1.68 s'-36 54 6.30 3-7 4-47 3-40 22.15 18.19 20.61 10.40 24-13 21.65 21.88 24.82 21. II 20.56 1900 . . . 1901 .. . 1902 .. . 1903 ..- 1904 ... 1905 .-. 1906 .. . 1907 Av. Ann ual for each Station 1.92 1.90 1.42 1.63 2.27 1.56 2-34 .80 .... 20.03 22.48 17.27 35 1.97 3-9 .... 364 TABLES SHOWING AVERAGE PRECIPITATION Continued. Berlin Berthold, Fort Bismarck Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1892 . . . 1 8 9 3 ... 1894 ... 1895 ... 1896 ... 1897... 1898 ... 1899... 1900 .. . 1901 .. . 1902 .. . 1903 .- 1904 .. . 1905 ... 1906 . . . 1907 . . . AT. Annual for each Station 4.07 2.88 4.21 1.58 5-12 .83 1.41 1.30 1.03 1. 00 3-3 1.69 2.47 I.OI 3-03 .66 7 1 2.0O 4.OI .72 3-3 1 35 4-5 1 1.90 2. II 4-56 6-37 3-23 4.86 1.49 5-46 4.61 3-22 6.30 1.8 3 95 6-37 4.01 i. 80 5-40 3.02 4-27 2.27 20.83 12. 02 I7-85 16.62 28.93 22.52 23-63 19.58 22.68 25-45 18.78 17.92 20.47 -69 2-53 2.23 1.40 1. 12 'Is .61 1.38 .07 .85 .67 1.26 3-8o 1.98 1. 10 2.65 4-3 1.26 .04 3-0 3.20 77 1.87 5-37 1.98 4-33 4.01 2.64 3-44 1.21 5-57 2.32 3*63 1.44 4-36 5.61 4-78 13-74 16.92 16.63 M-33 17^88 J 5-95 17.96 14.17 17.19 I 8'. 2 2 18.24 3-24 4.04 - .70 33 i. 20 .96 47 3 .07 1.77 3 3.02 2.76 7.71 .177 2.04 7-79 2.50 1. 1 8 4-78 59 2-15 4.19 7.08 i. 06 2.24 8.08 14-83 11.90 14.69 4.02 5-34 6-35 3-05 13.66 14.08 Bottineau Buford, Fort Buxton Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 *vy* . . . 1893 .-. 1894 ... 1895..- 1896 .10 2.OO .80 .89 63 4-05 4-35 I. 21 3-74 6.94 38 4-79 1.84 26.96 17.49 17.27 21.49 10.62 21. IO I9-I5 5.85 .90 2.17 1.49 2.27 2.12 .8 4 2-5 1 6.66 1.56 3-25 4-73 .46 .11 4-56 2-93 1.68 5.38 1.90 4-48 2.85 .61 2.04 .29 1897 ... 1898 ... 1899 .-. 1900 . . . 1901 .. . 1902 .. . 1903 1904 . . 1905 1906 .. . 1907 .- iv. Annual fo each Station 55 .20 .10 T 37 0.69 29 05 2.96 .70 4.10 4.02 1.07 6.00 4.08 .16 3- 11 4-63 16.06 7.08 3-3 12.98 2 48 15.90 i ^ 4.84 1.90 20.53 13.16 1.79 .61 3-9 1.71 8.81 3-40 22.53 13-55 Cando Church's Ferry Coal Harbor Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1892 5-25 3-42 2.90 4.68 2-97 4.20 3-74 53 7-45 3-45 99 7.17 4.72 3-24 1801 ,11 1.46 5-98 1.20 1.26 3-29 .32 .08 .49 1.72 .89 .07 79 1-75 i-95 4.66 55 2.21 3.00 .66 .00 3.87 2.17 1.20 i-53 16.64 26.69 I 3-S7 23-39 20. IO 22.52 18.91 19.08 I3-42 23-35 18.86 2.72 1.90 1. 12 1.88 42 T T T 1-34 .84 .98 30 3-67 79 99 6.06 .82 -38 2.84 ".78 1-83 4-85 3-55 3-55 2.28 2.49 - 1.98 3-4 4.60 4.40 6. 37 6.04 3-66 2-39 21.29 12.58 18.22 19.26 I S37 21. 8l 18.25 1 80 ? 1806 1807 1898 n ' l8OO .. IQOO .. 1901 .. . IQO2 .26 4.70 14.72 1903 .- 1904 ... 1906 .. . 1907 ... AT. Annual fo each Station i-37 .01 .84 54 3.82 1.49 J-75 3.01 13 1.78 9.84 3-88 3-3 5-58 z'6.83 13-24 365 TABLES SHOWING AVERAGE PRECIPITATION Continued. Cooperstown Devils Lake Dickinson Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 3.60 45 2.40 1.36 2-35 1.63 i. 02 1-83 65 .62 .16 3 1 .92 .09 I.IO 3 1.77 2.90 .90 1.69 5-i6 73 2.OO 3-51 .48 J 3 3.02 3-59 .90 2.74 7.11 1.36 2. 2O i-45 3*28 2-54 1.30 2-59 -i 2.56 .1.18 6.10 3-75 5-40 2.52 11.67 is'.is 11.92 17.27 11.78 12.92 !5- J 9 16.55 20.46 is. 14 ^! y i8cn , 5-65 1804. . 3-78 5.16 1.81 180* .. 1806 ^ ' 1807 I.IO .19 i.56 I.IO .24 .24 2.18 1-52 .10 i.*5 .70 .80 J .39 2.84 .22 .IO 3-3 4.06 2.47 1.67 4-05 35 4-43 3-5 2.21 2.06 8.2 7 3-99 77 5- 6 4 4.41 2.13 4.88 16.09 16.99 24.14 20.29 16.27 18.24 21.41 18.50 I S-49 18.60 1808 . 1899 ... IOOO .. IQOI . IOO2 1903 1904 .. . 1905 IOO6 ".89 i. 6 1 1.99 4.07 5.00 4.70 1907 .. . Av. Annual for each Station .08 .18 3-9 1 Donnybrook Dunseith Ellendale Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 .. 4-5o 2.63 2.07 77 4-i3 3.62 18.02 I S-77 180^ 1804. * "V 4 * 1 80* 1.81 2.14 2.40 2.70 34 -85 4.07 2.03 3.26 3-3i 5.12 94 3-20 36 3-25 2-75 1.14 5- 12 J 3-35 10.03 17.99 22.07 18.85 20.49 15.11 16.59 9 * 1806 . 1807 . 66 06 I 41 1898 .. 1.23 I. 21 2.O4 1. 14 I.I5 J-55 2.30 1899 ... 1900 . . . 1901 .. . 1902 .. . 1903 .- 1904 .. . 1905 .. 1906 .. . 1907 .. . Av. Annual for each Station 97 32 74 39 1.30 3 i-ol 36 1.70 3 2 2-58 6.52 1.96 1.88 5-i3 .20 2.90 65 5.78 3- 11 2-5 1 4.26 4.69 3-i7 i-34 13-25 12.86 22.79 2 3-99 15.20 19.01 19.83 1.40 2 3 .40 .40 .78 .66 7o i-75 27 4-37 2.07 5-8i 54 1-45 2-75 5-75 .48 4-3 1 4.00 1.16 4.40 3-77 211 18.62 16.06 16.44 2.20 18.13 17-03 16.83 Edgeley Edmore Fargo Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 2.42 2.63 3- 2 5 1.36 3-64 .89 .88 J -39 1.82 1.76 2.30 i-34 2.74 1.28 2.21 3-88 77 2.62 1.62 4.70 74 4.15 4.22 .81 .98 4-25 2.78 1.47 4.10 3-3 i-93 3.62 6.23 4.81 2.41 7.10 2 25 3-44 2. II 5-9 1 3-7 .06 3-83 5.00 3- 11 20.92 16.11 16.05 21.77 22.50 16.36 21.21 25-54 25.40 23.24 21.91 20.26 17.70 180^ 1894 ... 180^ i<-yo > 1806 1807 . 1898 .. . 1899 ... IOOO 1901 .. . 1902 .. . 1903 .. 1904 .. . 1905 1906 .. . 1907 .. . Av. Annual for each Station 2-47 2.85 3-3 1.41 1.58 35 *5 3 94 2.24 1.29 4-33 5-59 5-63 5-9 !-57 5-40 3-48 3- 6 9 1-73 27.45 21.25 17.36 19.96 21.51 1.30 .80 1.17 2-35 1.89 2'. 6 2 2.78 I S-39 20.69 366 TABLES SHOWING AVERAGE PRECIPITATION Continued. Flasher Forman Fullerton Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 1801 3-9 95 3-49 18.94 J -35 1.19 1.74 1.65 2.09 3-4 3-45 1.40 2.82 .41 1804 ... 1895 .. . *.-33 5-32 I -39 1.19 2.58 2.72 35 3-3 1 4.46 3-3 1 5-56 i. 02 15-87 21.09 19.42 17-74 2.24 3-03 44 36 4-97 2.06 2.52 6. 02 8.08 3- J 7 1896 .. . 1897 .. . V. 5 6 4.04 .66 3.61 3-29 1.62 4.60 3-37 4-43 4-25 14.36 23.68 22.04 18.81 22.97 24.03 20.88 21.08 33-87 22.40 1898 ... 1899 ... 1900 . . . IOOI 2.28 5i i-75 2.22 2-59 1.97 2.52 3 41 .66 4.04 2-95 3-03 5.10 7.16 i-54 59 5-56 3-i6 2-99 5-93 3-44 3-92 3.21 21.15 23-74 20.86 2 3-52 22.36 25-52 20.93 IQO2 IQO3 I9O4 IQOt> 1906 .. . 1907 ... Av. Annual for each Station 1.58 5-76 i-55 6-45 2. 2O Gallatin Gladys Glen Ullin Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 . 1893 ... 1894 ... 1895 ... 1896 .. . 1897 ... 1808 2.24 3.88 i-43 6.26 77 83 1.42 3-07 5-43 79 2.50 4.26 93 73 3-52 2.76 2.66 5-49 2.18 3-25 2.80 2.22 1.8 4 6-35 2-45 12.67 15.66 16.13 27.26 13.28 13.02 15.02 17.19 20.33 16.80 .... 2.10 1.17 93 57 43 .42 43 47 1.41 .14 i-i3 3.16 84 1.42 1.30 1.03 44 2.83 3-76 1-43 1.27 7-55 59 5-9 2 2.36 1. 12 V-SS 1899 ... 1900 .. . 1901 .. . 1902 .. . IQO^ i-34 1.65 1.78 i-i3 1.46 15.81 5-49 2.69 7-23 6.05 4-44 i. 80 19.30 16.50 18.74 15-59 IQO4 IQO^ IOO6 I OO7 55 i. 08 2.99 Av. Annual for each Station 16.74 Grafton Hamilton Hannah Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1892 .. . 1893 ... 1894 ... 1895... 1896 .. . 1897 ... 1898 ... 1899 ... 1900 .. . 1901 .. . 1902 .. . IQO3 2.36 J -73 2.28 3.22 4-75 .40 1. 10 2.14 75 45 1.25 =;6 2.85 1.46 1.25 .90 6.72 .20 x '3 1.76 i. 20 T 2-75 3-54 2.IO 4.70 3-3 1 .20 4-45 2.44 7-45 2.40 00 16.47 10.84 16.76 25.12 10.28 12.80 20.02 .... 73 1.71 2.40 .18 .87 2 t8 1.84 1.26 3.10 .20 .00 3-85 1.78 6.16 3-72 i. 06 6.08 4.10 24.46 23-05 20.61 18.44 I 7-73 .... 1904 . . 1905 ... 1906 .. . 1907 . Av. Annual for each Station 1.20 05 97 5 2.50 V x" 6.65 2.91 1 6. 1 8 16.1 2 2.o6 .41 97 1.70 2.81 4.00 2.52 *-iS 8-55 5-95 3- 2 7 28.83 2 5-59 .64 2.65 1. 10 5.60 % 1.40 22.67 TABLES SHOWING AVERAGE PRECIPITATION Continued. 367 Hannaford Hillsboro Jamestown Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1892 .. 42 A 3OI .01 71A. 1893 .. 1804 . 2.78 2.48 1.23 6.20 .62 I. II 8s 1.67 .87 !-52 9-58 1.0 3 4.02 212 34 2.08 4-93 3-92 2.04 7.29 1.94 2 I O 14.56 12.64 33-09 18.73 1895 1896 ... 1807 1898 1899 .. . 1900 . . . 1901 .. . 1902 .. . 1903 . . . 1.30 1.17 .41 74 .12 3.36 1.18 7.91 2-57 14.92 23.64 16.62 41 i 6< .96 .80 i-33 1.85 1.28 2.07 59 *3 4-74 2.24 1.22 4.II 5-69 J -39 9.71 3-52 5i 7.09 5-09 2.20 25.66 16.89 19.99 22.12 20.46 1904 . . . 1905 . . . .18 1. 10 6.10 1006 2.90 1.29 5-78 i-37 3-49 3-87 24.70 1907 Av. Annual fur each Station 18.39 Kelso Kulm Lakota Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1892 .. 1893 .. 1894 ... 1895 ... 1896 .. . 1807 2.30 2.79 -65 7.19 97 *-73 2.41 6-35 4-03 3-46 4.71 74 16.38 18.76 3-38 I 70 I -3 I 3-90 5-4 3.82 4.90 1898 ... 1899 ... IQOO 1.36 1.84 4-56 4-3 6 5.86 IOOI 1902 I9CM I9O4 2.20 1.74 2.60 .67 2.36 5-Si 6.46 3.81 7-44 3-67 5-52 1.41 I9O ^ 24.00 25-31 y 2 1906 .. . 1907 . . . 2.08 33 4-45 Av. Annual for each Station Langdon Larimore Lisbon Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1892 1893 , 1894 . 2 71 I 37 i8ox 1.19 3-22 .8l 53 1.03 1.24 2.89 .98 .89 2.32 59 1-63 54 2.24 4-95 2.96 2.22 .16 J 7 4-45 J -75 2.08 2.38 3.80 52 6.17 3-9 3.62 1.82 .90 5-93 2.88 .70 6.97 4.01 3.16 3-92 14-25 ;> y 9 * ' * 1806 1897 ... 1898 ... 1800 . 52 I.OO 1.19 'IS 2-45 5-36 15.42 i5-5o 13-45 19.58 19.67 2 3-35 19.69 22.75 21.66 '8.53 1900 .. 1901 1902 .. 1903 .. 1.96 4.18 i.6 3 2.60 37 4-38 1.71 5-75 5-38 2.36 2.41 5-47 5-04 4.64 2.98 22.07 2I -73 27.08 23-62 1904 ... 1905 ... 1906 .. . 1907 .- Av. Annual for each Station .16 .91 77 3-03 3-54 55 5-99 7-58 4.48 !-39 368 TABLES SHOWING AVERAGE PRECIPITATION Continued. Mayville Manfred McKinney Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1892 1893 ... 1804. 2.06 i. 60 1.86 1.65 .87 .61 5 .21 T 5 1.46 3.56 4-65 32 .01 2.50 .81 T .78 4.68 4.71 4-25 6.82 10.40 3-i3 5-i8 43 5-9 1.50 1-85 3-35 4-03 2.58 1.41 14.29 15.62 20.06 19.36 iS-35 15-73 14- 5 1 6.58 15.28 13-75 i5-5 *uyf i8o< 1896 ... 1897 ... 1898 ... 1899 ... 1900 . . . 1901 . . . IOO2 5-85 49 2.25 i-37 8.61 78 2.84 4-15 .90 .40 4-57 1.17 i 69 2 '.6 7 .27 8.10 2-5 1 .04 20.44 1903 ... IQO4 05 14.84 J-59 33 .68 77 1.81 1.98 5-20 49 6.76 3.82 3-90 2-54 19.11 16.27 IQOS ... IOO6 1.22 3.66 5-34 .10 I.OO .80 2.83 2.81 2 5 1907 . . . AT. Annual for each Station .38 .40 5.08 Medora Melville Milton Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 1.85 2.10 4-75 i-75 8.51 2.25 i-39 2.10 3-97 8-55 1.94 3-64 4.28 5.80 5-07 M. 3 1 18-38 x 9-53 20.85 21 .56 1893 ... 1804. .18 3-84 .86 180=? 1896 ... 1897 ... 1899 .. 72 2.20 1-3^ 5- 1 ? 38 3-4o 4-97 1.19 2.80 9-9 .86 1. 21 .38 5 2 .80 1.46 J 5 i.37 1-25 15 2-33 4.18 T 49 4.32 2.4O 1.30 1-95 3-8i 1.24 3.38 2.02 2.92 6. 7 I 3 .66 .48 7-3 5-78 3.08 i-95 18.20 12.50 17.08 20.10 18.24 I3.6l 13.89 .88 3.00 .60 93 54 V. 7 8 .00 I.OO 2.70 .10 .20 4.98 2.04 1.94 3-70 3-35 J -35 6.10 3-*7 1-65 8.23 22.0O 18.21 17.94 1900 . . . 1901 .. . 1902 .. . 1903 1904 .- 1905 .- 1006 75 .06 .67 .IS 30 2 5 1.04 .07 2.64 1.07 .11 i-*3 6 22 1-13 5-44 1.65 2.32 2.01 4-3 1 T 3-57 14.21 14.00 2.52 1907 .- AT. Annual for each Station .06 .83 2.85 12.92 16.23 19.09 Minot Minto Minnewaukan Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 1893 , I 37 I OO 2 7 C 1894 . . 2.88 i-59 5-23 1. 10 i. 08 1.29 1.86 1. 12 .88 i-59 .41 1.52 .62 .96 1.91 6.30 83 1.94 25 15 3.68 !-53 2.70 3-67 2.68 53 i-75 4-32 2-93 1.81 4-97 3-9 1.86 5-58 3-94 65 5.80 4-55 2.82 4-13 12.71 J5-75 22.65 J3-97 21.09 19.89 18.97 20.50 28.26 T 9-3 T 74 I.OO 1.02 T 1.70 74 .00 ".46 i. 08 2.48 54 3' 7 6 2.08 i-37 3.02 5-05 4-39 2.49 I.OO 7.07 3-i8 1.20 5-77 6-53 16.62 *3-S9- 14.72 13.18 17.88 1895 ... 1896 .. . 1897 1898 .. . 1899 ... 1900 .. . 1901 .. . 1902 .. . 1903 ... 1904 .. - 1905 . 1906 . . . 1907 .- - Av. Annual for each Station 1.18 i-S5 1.40 ' ' T ' 2.46 2.90 1.03 36 .66 05 5-3i i.iS 5-56 1.50 5-3 .16 3.38 3.00 6.07 i-95 4-75 5.06 2.13 8'.8i 24.27 20.97 20.75 18.70 15.20 369 TABLES SHOWING AVERAGE PRECIPITATION Continued. Moyersville Napoleon New England City Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1892 .. . 1893 ... 1894 ... i8ox 5-5 i-7S 2.98 1.98 2.65 Z 1 3 2.67 .80 .80 1.26 .181 1.27 2-37 65 J-55 .10 2.67 1.78 1.23 2.82 2.63 .67 3-77 4-15 1.07 .20 4-39 2.41 2.06 2.69 3-29 1.32 4.89 3-56 3-25 4.22 3-44 8.96 2.07 5-85 3-45 6.22 4.18 1.94 4.64 3-79 4.01 2.80 26.60 17.66 16.29 16.73 21.56 20.67 19.83 20.47 22.22 19.79 20.79 17.86 !5'97 18.29 15.86 JQ-37 .... .70 -63 3.10 2.00 T 3 55 3 2 1.24 1.74 2.30 6.00 .12 J -59 2.50 I -3 I .06 2.68 3-92 1.40 1.87 3.10 2-44 .60 1.20 1.20 1.40 5-36 1. 00 1.24 4.20 3-97 .70 2.70 12.38 12.55 15.81 I3-5S 1896 1807 1898 1809 . . . 1900 . . . 1901 .. . 1902 .. . 1 003 1 004 1905 . 1906 . . . 1907 . ' .68 1.78 5.01 3-52 6.92 Av. Annual for each Station New Salem Oakdale Oriska Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 4-34 2.84 2 -37 6.70 .67 2.16 2-37 i. 60 4.48 4.81 5.08 2.60 4-49 3-70 16.54 16.78 21.47 14.91 *uy-* 1893 ... 1894 ... 1895 ... 1896 2. 2O i-7S 1.96 10.27 .62 2-35 1.92 2.40 i. 60 1.96 1.29 2 A 1897 ... 1898 1800 1900 . . . 1901 .. . 1902 .. . IQO3 94 .48 .58 85 1.90 5 .06 2.30 7-83 i-34 1.48 7.87 1.29 7.62 4.46 2.26 4.90 7.96 2-77 26.15 19.84 26.02 13-38 25.06 20.02 1904 1905 1906 . . . 1907 . . . Av. Annual for each Station .12 1.20 2.42 3-5 1 .88 4-56 5-57 !-33 4.06 4.16 3-3 25-32 Palermo Park River Pembina Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 **y< 1803 "yo 1804. iSoe; i 64 95 I 2Q 5-5 1 2 OO 1896 1897 ... 1898 .. . 1800 1900 1.79 i. ii 1. 01 1.41 1.09 39 .90 1.61 .41 .12 4-75 3-42 3-27 4.09 2.70 1.05 .90 4.60 3-79 2-54 8.15 4.46 4.19 2-39 21.06 16.50 21.64 27.81 18.70 2.12 20.67 IOOI 1902 1903 .. 1904 . . . TQOC 2.05 .26 95 77 1.71 3-4' 2.96 49 6.57 5-53 3-4o 4.01 20.38 25-32 1906 2 I C 1907 . . . Av. Annual for each Station .60 25 2-37 370 TABLES SHOWING AVERAGE PRECIPITATION Continued. Portal Power Pratt Apr. May June An'l. Apr. May June An'l, Apr. May June An'l. 1892 .. 2-97 1.22 3-i3 3-*5 3.60 1803 ., 1894 ... 1895 ... 1896 ... 1897 ... 1898 ... 1899 ... 1900 . . . 1901 .. . 1902 .. . 1903 .. . 1904 . . . IQO5 2.50 1.52 2.40 1-34 .82 .40 T I.IO 2.52 1.25 1.50 2.81 4.16 34 .82 1.26 i-52 .40 2.83 4-77 .70 3-98 3-67 4.24 2.03 3-4o 4-55 5 6.12 6.14 3-J3 3-16 13-65 19.48 16.24 9-5 6 14.17 12.89 2 3-45 2-75 "3 4.19 T 2.81 5 1.32 1.70 2.17 i-34 2.82 1.40 2.91 25 1.96 1.04 4.23 .88 2.31 2.63 .61 17 3.86 3-23 I.OI 6.09 3-9 1.98 3-5 1 3-26 3-52 6.20 2.36 4.14 2-39 5-78 2.66 74 7-59 3-95 4.60 4-73 16.13 18.89 24.38 20.36 17.96 26.11 22.88 18.37 22.94 28.31 21.63 ' '.80 .64 4-75 .26 3-42 3-94 2.08 18.27 1906 .. . 1907 ... Av. Annual for each Station 1.17 .80 3-3 65 7-95 2.25 20.97 16.30 Rolla Rugby Sentinel Butte Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 1893 ... 1894 ... 1895 ... 1896 ... 1897 ..- 1898 1800 IQOO 1901 1902 .. . 1903 . 1904 . . . 1905 .. . 1906 .. . 1907 . . . AT. Annual for each Station .98 5 1.04 1.76 5-34 5-44 6.20 6.65 5-9 3-39 i 06 8 is IS 2-35 5.67 2.06 6.21 3-8i 7-23 !-57 Steele St. John Sheyenne Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1802 180* 1.28 3- 11 2.20 4.14 54 .63 2.OI I.I7 4-22 5-28 I.OO .24 4.19 2.50 6.00 4-35 2.49 3-6i 18.31 17.64 22.78 14.36 1894 ... 1895 ... 1896 ... 1897..- 1898 ... 1899 ... 1900 . . . dl 3-87 1.62 1.46 57 71 I.IO 2.61 2.81 1.04 2.0& 3-67 1.1 A, 1.42 3-54 2.18 4-oo 3-96 L.ro. 12.77 J 7-59 1.58 8.61 77 .90 2.05 .06 2.25 6.28 .69 2-45 3-51 5-25 5-20 3.87 3". 1 8 1.87 19.06 31.21 16.69 1901 .. . 1902 .. . 1903 . 1904 . 1905 ... 1906 . . . 1907 Av. Annual for etch Station T *5 .70 75 .18 1.25 .21 T 1 2-54 2.06 i-95 2.31 6.68 1.38 4-80 4-9J: L35 6.56 3.16 6.10 2.91 18.14 18.91 15-51 I 9-73 .... 17.11 18.27 22.32 371 TABLES SHOWING AVERAGE PRECIPITATION Continued. University Valley City Wahpeton Apr. May June An'l. Apr. May June An'l. Apr. May June An'l. 1892 .. . 1893 ... 1894 ... 1895... 1896 .. . 1897 ... 1898 ... 1899 ... 1900 .. . 1901 .. . 1902 . . . 1903 . 1904 . . . 1905 ... 1906 . . . 1907 .. . iv. Annual for each Station 2.90 3-37 3-4 93 5-J9 .56 4.16 4.20 1. 21 2.06 2.46 75 '53 .96 1.98 .11 2.42 1.77 87 1.94 .96 1.76 2.32 6. ii 25 .28 3-88 1.85 1.70 3-89 3-5 .68 3.20 3-98 3-4i 1.40 5.10 1.84 i-45 1.28 X -JS 2-65 1.79 2-55 2.83 1-54 I 04. 5-36 1. 12 97 1.63 7-43 .87 2.09 5-66 43 .80 3.60 2.27 2.83 5-29 37O 3-09 2-45 2.63 6.60 5-76 6 -95 1.79 4-i5 3-09 6-31 2.48 19.04 I S-34 3 3' 1 8 28-55 22.44 23-03 22.95 29-75 3- J 9 3-75 6.74 3-i5 6.80 17.97 r 7-3 6 18.64 23.91 3-07 1. 01 1.82 2-33 .76 1.50 .42 3-94 6.32 3-29 17.88 17.63 .88 8.47 2.09 49 5-64 4-57 3.08 4-63 20.04 26.40 22.00 17.69 20.98 27.94 18.15 21.01 2.12 T 3-35 8.88 4-83 7.88 5-20 27.69 .91 1.50 2-43 1.41 23-55 Walhalla Washburn Wild Rice Apr. May June An'l. Apr. May June An'l. Apr, May June An'l. 1892 .. . 1893 ... 1894 ... 1895 ... 1896 ... 1897 .. . 1898 . . . 1800 3-73 4.06 2.17 3-79 7 1 2.31 83 1.82 i.6x 4.08 .62 4.24 3.06 4.84 5-54 2. 2O 5-74 1.70 22.29 20.98 20.23 25.06 24.48 4.09 .16 32 1.85 45 3-3 2.56 .09 I -5S 4-35 2.90 3.00 2-97 2.78 3-23 15-83 17.71 j. oyy . . . I9OO I9O2 IOO3 1904 .. . 1905 1906 . . . 1907 ... Av. Annual for each Station ' T - .69 30 3-42 3-67 3-36 8.05 6-43 5-36 21.66 .10 75 10 1. 12 6. 20 5-27 2.30 4.IO 22.61 White Earth Williston Willow City Apr. May June An'l. Apr. May June An'l. Apr. May June An'l 1892 i i 02 2.64 3-28 i-59 3-45 3.18 .70 3.68 6,38 S.ll 4.69 1.98 3-84 5.85 4-38 i. 80 16.20 18.03 18.45 15.24 18.03 17.19 1 80^ 5 1 2.66 I-3 1 3.60 .00 ".T .16 .20 53 i. 06 T i-37 24 3-37 5-7 1 58 2.60 94 .00 2.04 4.04 i.8 5 2-59 "f 1804 i jy 1806 . 1807 . 1898 ... 1899 ... I9OO TQO I .23 .65 4.71 IOO2 47 4.96 4-47 20.51 IOO3 1904 ... 1905 .- 1906 .. . 1907 .. . Av. Annual for each Station i. 60 1.29 >? 2-15 3-64 4-45 2.90 7.07 2.48 2.66 1.70 20.02 18.26 .12 26.03 3-73 16.72 16.70 Yule Average Annual for all Stations Apr. May June An'l I8(%2 19.32 16.47 15.81 24.06 23-43 16.47 16.16 J 7-56 18.60 18.66 19.64 18.97 19.27 20.78 20. 14 1892 .. . 1893 ... 1894 .. . 1 80 3 .61 3-71 2.18 1804 rfifve 1895 . . . io yo 1896.. 180*7 1896 ... 1897'.. . 1898 ... i&nn 1899 .. . 1 yy TOGO 1901 1 9O I 1902 .. . TOO 3 1903 .. . TOO A 1904 . . . 1905. 1906. M 1905 .. . I9O6 . ean Annual Precipi Stations having rec tation for all ;ords 18.59 I9O7 . Av. Annual for each Station AN INDEX. PAGE Advance and Retreat of the Ice-front 42, 58 Age of a Landscape 166 Age of Reptiles, in North Dakota 190 Ages of Rivers 237 Agriculture West of Missouri River 245 Alexandria, S. D 120 Alkali Flats 164 Alkali in Soils 275, 286 Alkali Lakes 48, 164 Altamont Moraine, The Described. . . .221 et seq Altamont Moraine 112,214, 244, 323, 359 Antelope Hills 119 Antelope Moraine ' 119 Archaean, The 209 Area of Lake Agassiz 81, 110 Area of Lake Agassiz in Canada, Minnesota, and North Dakota 86 Area of Lake Dakota. 120 Area of Lake Sargent . 120 Area o f Lake Souris 116 Artesian Springs 165 Artesian Water, Source of 165, 259 Artesian Wells 258 Artesian Wells, Depths of 260 Artesian Well, Grafton 261 Artesian Wells in Red River Valley 261 Artesian Wells, Minerals of 262 Artesian Wells, Not from Dakota Sandstone. . 263 Assiniboine Delta 115 Assiniboine River, Manitoba 115 Augites 25 Badger Creek, Manitoba 115, 124 Bad Lands, The 23, 71, 77, 173 Bad Lands, Mistaken Notions about 173 Bad Lands, Cause of 174, 187, 240 Bad Lands Not Bad 188 Bad Lands, Named 241 Balfour Ridge 119 Base-leveled Plain West of Missouri River . . . 187 Beaches of Lake Agassiz, Uplifted. . . '. 108 Beach -lines of Devils Lake 125 Bed-rock 28, 77 Beginnings of North Dakota, The. 202 Bell, Dr. Robert, Cited. 108 Big Bend, Sheyenne River 82, 94 Big Butte, see Mauvais Butte 123 Big Coulee Outlet of Lake Souris 115, 119, 122 Big Coulee, Pierce County 114, 116 Big Horn Mountains, Wyoming 74 Big Stone Lake, Minnesota 80,81 Black Hills, South Dakota .74 Blanchard Beaches 83, 104, 109, 327 Blanchard Stage of Lake Agassiz 104 Blue Hills 123, 124 Bois des Sioux River 328 Boulder Chains 83 Boulder-clay, or "till" : 90 Boulder Patches 59 Boulders, "Foreign" 62, 63 Boulders, Kinds and Sizes 62 Boulders, Large ; 22 Boulders, Scattered Over Bottom of Lake Souris 117 Boulders, Traveled 59 Boulder-strewn Prairies 62 Breaks, The, North of Dickinson 21, 178 Buffalo Boulders 63 Buffalo Lake, Pierce County 114 Buffalo River, Minnesota 86, 93 Buford 73, 74 Building Stone 357 Burnside Beach 109 Butte, in Pembina County 98 Butites 21, 77 Buttes, Forms of 176 Buttes, The Older 186 Buttes, The Structure of 180, 240 Buttes Standing on Shoulders of Buttes 186, 224 Caledonia, Depth of Lake Agassiz 85 Cambrian Formation, The 209 Camel's Hump 187, 243 Campbell Beach 96, 103, 104, 107, 315, 329, 332 Campbell Beach, Across Deltas 102 Campbell Beach, Multiple 107 Campbell Stage 104 Cannon Ball River, The 235, 238 Carboniferous Formations, The 209 Cass County, Sheyenne Delta in 94 Cause of Existing Lakes, The iol Cause of Lake Dakota 120 Chain-of -Lakes, The Alice 148 Channel Connecting Sheyenne and James Valleys 114 Clay, Formed 35 Clay, Stratified, on Bottom of Lake Agassiz 91 Cliff, Wave-cut 94, 96 Climate of Western North Dakota 246 Clyde Series of Soils, The 284 Coal Beds, How Old 210 Coal Beds, How Formed V 193 Coal in Bad Lands . . 201 373 374 AN INDEX. Coal in North Dakota, History of Formation of 190 et seq Coal Measures, The Western 196 Coal Mine in Cellar 201 Coal Seams, Elevations of 201 Common Wells 267 Conditions, Causing Lake Agassiz, The.... 79 Copper and Lead 298 Coteau des Prairies 72, 78, 79, 81, 120, 121, 214 Coteaus of the Missouri, The 213 et seq Coteau du Missouri 21, 73, 74, 78, 112, 244 Coulee, or Young Valley 17 Cretaceous Era 77 Cretaceous Inland Sea, The 77 Cretaceous Rocks 101 Cretaceous Shale in Delta Sand 98 Cretaceous Shales 96, 97, 124 Cretaceous, The, in North Dakota 202 et seq Crust of Earth, Changes in Form of. ... 105, 107 Custer's Trail, through the Bad Lands 183 Custer Trail Ranche 184 Cut-off at Lisbon 139 Cut-off at Valley City 142 Cycle of Erosion 15 Dakota Glacier 47, 48, 98, 115 Dakota Glacier and Lake Dakota 119 Dakota Glacier in South Dakota 52 Dakota Glacier, Moraine Formed by. ... 84, 112 Dakota Sandstone, The 203 Dam, Restraining Waters of Lakes Winnipeg, etc 85 De Groat Lake 125 Delta, Photograph of 95 Deltas and Beaches of Lake Agassiz, The . . 86, 92, 110 Depth of Drift about Devils Lake 124 Depth of Great Ice-sheet 70, 79, 80, 117 Depth of Lake Agassiz 82, 85, 111 Depth of Lake Dakota .- 120 Depth of Lake Sargent 121 Descent of Red River Valley 85 Des Lacs River 112, 119, 355 Devils Heart Hill 57,124 Devils Lake 21, 22, 30, 52, 75, 115, 122 Devils Lake, History of 123-128 Devils Lake, Ice on Hills South 93, 114 Devonian Formation, The 209 Dog Den Butte 21, 73, 112 Dovre Moraine 81, 103, 116, 120-122 Drainage West of Missouri River 234, 235 Drift Boulders 29 Drift Clay 151 Drift Clay, of Red River Valley '90 Drift, Depth of 29, 77 Drift, Mantle 76, 77 Drift on High Mountains 33 Drift Period, Denned 29 Drift West of Missouri River 225 Drift, The Older and the Newer 224 Dry Lake 125 Duck Mountain, Manitoba 110 Dunes. . . . . 94, 95, 118, 322, 329, 348, 350 PAGE Dunes, of Sheyenne Delta 94 Dune Tracts, Lake Soums 117 Eckelson Lake 159 . Edinburg 84, 98, 106 Elevation Above Sea -level of State 73 Elevation of Basin of Lake Agassiz. . . .108, 109 Elk-River 86, 92, 93, 98, 102, 316, 331 Elk River Valley, a Sound or Strait. 101 Elk Valley 84, 98, 316 Elk Valley Delta 315, 316 Elk Valley Delta, The. .53, 84, 92, 93,97,98, 100, 103 Emerado Beach 109, 316 Epeirogenic Movements of the Earth's Crust 108 Excursion among the Boulders, An ?23 Excursion to Some Glaciers, An 39 Extinct Lakes, Denned 161 Fargo 21, 73, 75, 81 Fargo, Section Across Valley 90 Feldspar 24 Fergus Falls Moraine, The. .82, 83, 100, 110, 133,320 Fergus Falls Moraine and Lake Agassiz 103 Fifth or Elysian Moraine 122 First or Altamont Moraine 50, 116 First Pembina Mountain 95, 96 Forests, Submerged 125 Formations, The Geologic in North Dakota. . 203 Fossils Stumps Associated with Coal 192 Fort Benton Formation, The 197, 203 Fort Pierre Formation The 97, 124, 197, 203 Fourth or Kiester Moraine 114, 122 Fox Hills Sandstone in Buttes 342 Fox Hills Formation, The 163, 197, 203 Future of North Dakota, The 303 Gary Moraine 112 Geology from Car Window Prefatory Note. . 313 Girard Lake 114 Glacial Channels 262 Glacial Elk River 53, 92, 100 Glacial Lake Agassiz 76, 79 Glacial Lake Dak6ta 119 Glacial Lakes. 49, 151 Glacial Lake Sargent 120 Glacial Lakes, Extinct 31 Glacial Lake Souris 112 Glacial Lake, Temporary ' 50 Glacial Period 30 Glacial Period, Denned 29 Glacial Period, Time Since 210 Glacial Rivers 49 Glacial Soils West of Missouri River 250 Glacier, Conditions for Forming of 35 Gladstone Beach 109, 110 Gold 296 Gold and Things not Gold 296 Gold, Mica Mistaken for 298 Gold in Hard Boulders 297 Gold in Shale 297 Golden Valley 84, 98 Golden Valley, Level Bottom of 102 Goose Rapids ' 83 Goose River 22, 75 Grafton. . . 21, 110, 165, 209 AN INDEX, 375 PAGE Grand Forks, Section Across Valley 90 Grand River, South Dakota 78 Granite, in Boulders 24, 25 Grass Lake 123 Gravel Pits 89 Great Divide or Height of Land 79 Great Ice-sheet, The 30, 35, 36, 53, 70, 79, 116 Great Ice-sheet, and the Missouri River 74 Great Ice-sheet, Dam to Northern Drainage. . 86 Great Ice-sheet, in North Dakota 47 Great Ice-sheet, Region not Covered by. ... 71 Great Ice-sheet, Western Limit 73 Great Northern Railway, Highest Point of. . 123 Great Plains, The 230 "Great Salt Lake" of North Dakota. 123 Great Stone Face, The 243 Ground Moraine, Defined 38 Gumbo Soils West of Missouri River 253, 285 Gypsum 300 Hanging Valley of the Maple, The 139 Hard Head Boulders, Origin of 209 Hawk's Nest 73 Heart? River, The 78, 235, 238 Heerman, Captain, Cited 128 Height of Land 79, 81 Herman Beach. . .82, 94, 104, 107, 108, 110, 315, 316 330, 335 Herman Beach, Across Deltas 110 Herman Beach , Beginning of 103 Herman Beach, Multiple 106 Herman Stage of Lake Agassiz.82, 102, 108, 110, 111 Hillsboro Beach 109, 326 Hills, East and West, Compared 20 Hills, Morainic 36 Hills, South Devils Lake 52 Hills, Types of 23 Horneblende 24 Hudson's Bay, Filled with Ice 86 Hurricane Lake 123 Ibsen Lake 123 Ice, Behavior of, Under Pressure 32, 35 Ice-cascade . . . ; 44 Ice-dam, Cause of Lake Souris 112 Ice-flow, Due to its Weight 34 Ice Movement, Direction of 31 Ice-sheet, Thickness of 33 Ice, Stratified .44, 54 Ice-water, From Melting Glacier 41, 44, 49 Inkster, Profile at 100 Iron Minerals 293 Iron Pyrites 293 Island Lake 123 Islands in Lake Agassiz 101 Islands, on East Shore of Lake Winnipeg. ... 85 Itasca Lake, Minnesota . 84 Itasca Moraine 101, 110, 316, 317, 333 Itasca 'Stage, Dakota Glacier 84 James River 73, 74, 79, 114, 115, 116, 121 James River, Dammed by Ice 120 James River, Enlarged to Form Lake Dakota 119 James River Valley 74, 114, 119, 121, 151, 213 Jura -Trias Formations, The 209 PAGE Jerusalem 124, 128 Joint Terminal Moraine 54, 68 Jotenheimen Mountains, Norway 39 Kettle Moraine 320 Kiester Moraine 114 Killdeer Mountains 187 Knife River, The 235, 238 Laccolite Mountains 231 Lac des Roches 115, 124 Lagoon, Formed Behind Sand-bar 89 Lake Agassiz 74, 119, 121, 122, 151 Lake Agassiz, Area of -. 81, 110 Lake Agassiz, Compared in Time with Lake Souris 116 Lake Agassiz, History of 109 Lake Agassiz, How Caused 102 Lake Agassiz, Increase in Size of 82 Lake Agassiz, in Relation to Deltas 102 Lake Agassiz Opened 148 Lake Agassiz, Outlet of 80 Lake Agassiz, West Shore of 75 Lake Bottom Lands 307 Lake Dakota 74, 119, 121, 122 Lake Dakota, Extent of 120 Lakes, Kinds of 151 Lakes, Map Studies of 157 Lakes, Meaning of on Landscape 166 Lake Sargent 120, 121, 122, 151 Lake Souris 74, 114-116, 151 Lake Superior, Outlet to 104 Lakes on Turtle Mountain 162 Lakota, Moraines Near 84 Lanona Plain, The 131, 140 Laramie Formations 197, 203, 207 Laramie Formations, Thickness of in North Dakota 200 Larimore 22, 30, 36, 53, 72, 78, 86, 98 Lateral Moraine 37, 43 Lead Copper and 298 Leaf Hills Moraine. 52, 83, 84, 100, 110, 316, 319, 320 Leaf Hills Stage of Lake Agassiz 101 Level of Lake Agassiz, Cause of 110 Level of Surface of Lake Agassiz, Deflection 107 Level -prairie Portion of State 71 Lightning's Nest 94 Lignite Coal in Bad Lands 181, 188 Lignite, Meaning of term 195 Lime Minerals 299 Limestone (Calcium Carbonate) 301 Limestone in Deep Well 206 Lisbon 30, 82, 94, 121, 122, 138 Little Missouri River 21, 74, 78, 178, 184 Little Missouri River, Danger of Piracy of ... 239 Little Missouri River, The, Described 238 Little Muddy Creek 325 Little Pembina River 97 Long Lake, McHenry County 112 Long Lake, Rolette County 123 Lost Creek 102 Lowland Plain, in Central Portion of State . . 78 Manitoba Escarpment, The .71 , 74, 75, 79, 205 et seq, 316, 328, 331, 336 376 AN INDEX. PAGE Manitoba Lake 58, 81, 85, 110, 111 Manitoba, Multiple Beaches in 106 Maple Lake, Minn., Beaches Near 106 Maple Ridge 335 Maple River, The History of 143 Marshall Series of Soils, The 283 Mauvais Butte 123, 124, 319 Mauvaise Coulee 115, 123, 318 Mayville 22, 83, 98 Mayville, Contour Line 87 McCauleyville Beach 96, 102-105, 107, 109, 315 329, 330 McCauleyville Stage of Lake Agassiz. .103, 105, 109 Medial Moraine 37, 49 Medora 183-4, 347 Mesabi Moraine 85, 110 Milnor Beach 82 Mine, A Burning 241 Minerals, Non-Metallic 299 Mineral Resources 308 Minerals in North Dakota 291 Minnesota Glacier 47, 48, 83, 98, 116 Minot 22, 59, 60, 73, 116, 118, 123 Missouri Plateau, The. .112, 116, 123, 186, 197, 207, 213, 214, 218, 229 et seq Missouri Plateau, The Eastern Slope of. .219,220 Missouri Plateau, East of Missouri River. . . . 244 Missouri Plateau, Structure of 241 Missouri River, Landscape East and West Compared 166, 217, 234, 284 Missouri River, The 23, 26, 30, 71, 73, 74 76, 78, 112, 224 Missouri River, The, at Bismarck 342 Missouri Slope, The 73, 224, 235 Monadnocks 233, 242 Moraines, Denned 36 Moraines Formed upon Bottom of Lake Souris 118 Moraine Leveled, in Red River Valley 83 Morainic Hills 76, 83 Morainic Lakes . 47, 76, 151 Morainic Lakes, Formed 38 Morainic Lakes, Typical Group of 163 Morainic Ridges 36, 37, 41, 43 "Mountains, The" 49, 84, 101, 332, 333 Mouse River 74, 79, 114, 117, 119 Mouse River Valley 71, 123, 151, 197, 213 Nelson River Outlet of Lake Agassiz. .85, 109, 111 Niobrara Formation 197, 203 Niverville Stage of Lake Agassiz \ll Norcross Beach.. 96, 102, 106, 107, 108, 110, 316, 330 Norcross Stage of Lake Agassiz 103, 104, 108, 109 Oakes 21, 30, 36, 71, 120 Ojftta Beach 315 Old Buttes 242 Old Drift 324 Old North Dakota 76, 77, 123 Old (Pre -glacial) Landscape of North Dakota 76 Old River Valley 123, 124 Older Rocks of Eastern North Dakota, The. . 209 Ossawa Beach 109 Outlet at Lake Traverse ... . . 108, 109 PAGE Outlet Lake Agassiz, South, Map Showing Beaches 105, 107 Outlet Lake Agassiz to Lake Superior 105 Outlet of Lake Agassiz, Northeast 86, 104, 109, 110 Outlet of Lake Agassiz, South 103, 104, 109, 110 Outlet of Lake Souris 112 Outlet of Lake Souris by Girard Lake and Big Coulee 114 Over- wash Plains 52 Ox-bow of Mouse River , 74, 112, 116 Ox-bows in River Courses 75 Pembina Delta, The 92, 95, 98, 103, 115 Pebbles Mistaken for Drift West of Missouri River 253 Pembina Mountain 72, 78, 79, 95, 96, 98, 115, 316, 328, 331, 333, 336 Pembina Mountain Highland 53, 71, 81 Pembina Mountain, Underlying Rock For- mations 96- Pembina River 75, 86, 92, 95, 97, 115 Pembina River Outlet of Lake Souris 116 Petrified "Butterfly" 67 Petrified Forests, The 188, 346 Petrified Wood, How Formed 188 Pictured Rock, at Fort Ransom 305 Plateau Region of North Dakota, The 229 Plateaus, Lesser 232 Pony Gulch 112, 158 Pyramid Park 346 Quartz 23 Quartz, in Granite 24, 25 Quartzite 27 Quartzite Boulders 23-25 Rainfall in North Dakota 310 Rainfall in Western North Dakota 247 also see Appendix. Red River of the North 76, 78, 79, 121, 123: Red River of the North, Axis of Lake Agassiz Bottom 105 Red River of the North, Crossed by Moraines 83, 84 Red River of the North, Early 110 Red River of the North, Elevations of 73 Red River of the North, Flood of 125 Red River of the North, Valley of 206, 209 Red River Valley, Bottom of Lake Agassiz. . 85 Red River Valley, Buried by Ice 116 Red River Valley, Depth of Drift in 80 Red River Valley, Fertility 57, 58, 86 Red River Valley, Leveled Drift Materials. . 83 Red River Valley, Low Lands of Lagoons. . 89 Red River Valley, Section at Wahpeton .... 89 Red River Valley, The 21, 22, 71, 75, 79, 88, 151 Ridges of Drift Forming Islands 122 Ridges, of Shore Sand and Gravel 88 "Ridge" The 49, 84, 100, 101, 316, 331, 333 River of Lakes, The 164 River Ransom, The 135, 348 Rivers, Work of in Western North Dakota. . 233 Rivers of Western North Dakota, The 235 Roosevelt Cabin, in Bad Lands 184 Salt Beds on Lake Bottoms 155 Salt Lakes from Artesian Springs 165 AN INDEX. 377 PAGE Salts and Alkalies, Sources of 155 Salts in Lake Waters 154 Salts in Soils 275 Sandstone Concretions 295 Sand Prairie Spillway, The 133 Sand Prairie 134 Sand-bars, Built by Waves 88 Sand Dunes, McHenry County 118 Sandstone South of Turtle Mountains 117 Scoria as Natural Brick 182 Sea Bottom, Ancient, in North Dakota 202 Second or Gary Moraine 112 Second Pembina Mountain 96, 97 Section, Generalized, in North Dakota 202 Sentinel Butte.. / 73, 241 Seventh or Dovre Moraine 116 Sheyenne Delta, Area 94 Sheyenne Delta, Beginning of 82 Sheyenne Delta, Section 94 Sheyenne Delta, Stage of the Lake 102 Sheyenne Delta, Structure 94 Sheyenne Delta, The . . 92, 93, 98, 103, 115, 206, 350 Sheyenne River. .21, 49, 73, 75, 79, 81, 86, 92-94, 115, 116, 124, 125 Sheyenne River, Beginning of Lake Agassiz . . 109 Shore Boulder Chains 61 Silurian Formation, The 209 Silt, Deposited in Lake Agassiz 86, 92 Sioux Quartzite 27, 120 Sixth, or Waconia Moraine 122 Soil, Composed of 28 Soil Classes, The 287 Soil, Denned 269 Soil Map, Making of a 277 Soil Map of North Dakota, The 289 Soil, Organic Matter in 276 Soils, A Study of The 269 Soils, Classification of 287 Soils, Derived from Different Kinds of Rock , 273 Soils of the Glacial Lakes 279 Soils of the Rolling Prairies 282 Soils, Their Study a Geological Problem 269 Soil Regions, The, of North Dakota 278 Soil Series, The 288 Soil Survey, How Made 277 Soils West of Missouri River : 249, 284 Stages of Lake Agassiz 102, 104 Stonewall Beach. . . . . 109 PAGE Stormy Lakes 120, 121 Strata, Defined 27 Stratified Gravel and Sand in Sand-pits .... 63 Stria- 53,55 Striated Boulder 53, 55 Stump Lake 115-128 Sully 's Hill 57, 124 Terminal Moraine, Defined 37, 59 Terminal Moraines 36, 40, 45, 46, 50 Terminal Moraine, How Formed 217 Terminal or Dump Moraine, Illustration of. . 316 Terraces of Sheyenne Valley 140, 339 Tewaukon Lake 121, 122 Third or Antelope Moraine 119 Till, or Boulder-clay 90 Till, Underlying Delta 97 Tintah Beaches 96, 102, 103, 107, 109, 315 Tongue River 75, 96 Traverse Lake 79-81, 85, 103, 121, 328 Turtle Mountains. . . 55, 57, 73, 74, 78, 114-116, 123, 321 Turtle Mountain Plateau, The 79, 163, 207, 233 Types of Landscape, Three Described. .218 et seq Upham, Warren, Cited 81 Valley City 21, 30, 36, 121 Veneered Hills 55, 57, 58, 123, 336 Wahpeton and Breckenridge, Depth of Lake Agassiz at - 82, 85 Wahpeton, Section Across Valley 89 Warren, Gen. G. K 81 Warren, River 80-82, 104, 105, 115 Washington Lakes 125 Watershed between Devils Lake and Mouse River 123, 319 Watershed, between Hudson Bay and Gulf of Mexico 219 Watershed between Little Missouri and Heart Rivers 184, 235, 345 Water Supply, The, in North Dakota 258 Water Supply, West of Missouri River 256' Wells County, Old Channel in. . . ." 114 Wheatland, Profile 89 Wild Rice River 75, 121, 122 Wild Rice River, Minnesota 93 Winnipeg Lake, Depth of Lake Agassiz Over , 85 Winnipegosis Lake 81, 111 Wintering Creek 119 Wood Changed to Coal, Manner of . . , , . 194 RETURN TO the circulation desk of any University of California Library or to the NORTHERN REGIONAL LIBRARY FACILITY Bldg. 400, Richmond Field Station University of California Richmond, CA 94804-4698 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 2-month loans may be renewed by calling (415)642-6233 1-year loans may be recharged by bringing books to NRLF Renewals and recharges may be made 4 days prior to due date DUE AS STAMPED BELOW AR 51990 7- 3979 f 808773 UNIVERSITY OF CALIFORNIA LIBRARY