Numerous explorers of this region, from Long and Keating in 1823, to Gen. G. K. Warren in 1868 and Prof. N. H. Winchell in 1872, recognized the lacustrine features of the valley; and the last named geologist first gave what is now generally accepted as the true explanation of the lake's existence, namely, that it was produced in the closing stage of the Glacial period by the dam of the continental ice-sheet at the time of its final melting away. As the border of the ice-sheet retreated northward along the Red River valley, drainage from that area could not flow as now freely to the north through Lake Winnipeg and into the ocean at Hudson bay, but was turned by the ice barrier to the south across the lowest place on the water-shed dividing this basin from that of the Mississippi. This lowest point is found, as before noted, at Brown's valley on the western boundary of Minnesota, where an ancient water-course, about 125 feet deep and one mile to one and a half miles wide, extends from Lake Traverse, at the head of the Bois de Sioux, a tributary of the Red river, to Big Stone lake, through which the head stream of the Minnesota river passes in its course to the Mississippi and the Gulf of Mexico.

Detailed exploration of the shore lines and area of this lake was begun by the present writer for the Minnesota Geological Survey in the years 1879 to 1881, under the direction of Professor Winchell, the state geologist. In subsequent years I was employed also in tracing the lake shores through North Dakota for the United States Geological Survey, and through southern Manitoba to the distance of 100 miles north from the international boundary to Riding mountain, for the Geological Survey of Canada. For the last-named survey, also, Mr. J. B. Tyrrell has extended the exploration of the shore lines more or less completely for 200 miles farther north, along the Riding and Duck mountains and the Porcupine and Pasquia hills, west of Lakes Manitoba and Winnipegosis, to the Saskatchewan river.

This glacial lake was named in the eighth annual report of the Minnesota Geological Survey, for the year 1879, in honor of Louis Agassiz, the first prominent advocate of the theory of the formation of the drift by land ice; and the outflowing river, whose channel is now occupied by Lakes Traverse and Big Stone and Brown's valley, was named, in a paper read before

the American Association for the Advancement of Science at its Minneapolis meeting in 1883, the river Warren, in commemoration of General Warren's admirable work in the United States Engineering Corps, in publishing maps and reports of the Minnesota and Mississippi river surveys. Descriptions of Lake Agassiz and the river Warren have been somewhat fully given in the eighth and eleventh annual reports of the Minnesota Geological Survey, and in the first and second volumes of its final report. Two other special reports of my explorations of Lake Agassiz have been also published, the first in 1887, by the Geological Survey of the United States, and the second in 1890, by that of Canada; and a more extensive monograph of this subject has been prepared for publication by the United States Geological Survey.

Several successive levels of Lake Agassiz are recorded by distinct and approximately parallel beaches, due to the gradual lowering of the outlet by the erosion of the channel at Brown's valley, and these are named principally from stations on the Breckenridge and Wahpeton line of the Great Northern railway, in their descending order, the Herman, Norcross, Tintah,. Campbell, and McCauleyville beaches, because they pass through or near these stations and towns. The highest, or Herman, beach is traced in Minnesota from the northern end of Lake Traverse eastward to Herman, and thence northward, passing a few miles east of Barnesville, through Muskoda,. on the Northern Pacific railroad, and around the west and north sides of Maple lake, which lies about twenty miles eastsoutheast of Crookston, beyond which it goes eastward to the south side of Red and Rainy lakes. In North Dakota the Herman shore lies about four miles west of Wheatland, on the Northern Pacific railroad, and the same distance west of Larimore, on the Pacific line of the Great Northern railway. On the international boundary, in passing from North Dakota into Manitoba, this shore coincides with the escarpment or front of the Pembina mountain plateau; and beyond passes northwest to Brandon on the Assiniboine, and thence northeast to the Riding mountain.

Leveling along this highest beach shows that Lake Agassiz, in its earliest and highest stage, was nearly 200 feet deep aboveMoorhead and Fargo; a little more than 300 feet deep above

Grand Forks and Crookston; about 450 feet above Pembina, St. Vincent, and Emerson; and about 500 and 600 feet, respectively, above Lakes Manitoba and Winnipeg. The length of Lake Agassiz is estimated to have been nearly 700 miles, and its area not less than 110,000 square miles, exceeding the combined areas of the five great lakes tributary to the St. Law

rence.

When the ice border was so far melted back as to give outlets northeastward lower than the river Warren, other beaches marking these lower levels of the glacial lake were formed; and finally, by the full departure of the ice, Lake Agassiz was drained away to its present representative, Lake Winnipeg. The entire duration of Lake Agassiz, estimated from the amount of its wave action in erosion and in the accumulation of beach gravel and sand, appears to have been only about 1,000 years, and the time of its existence is thought to have been somewhere from 6,000 to 10,000 years ago.

ABORIGINAL PEOPLES

Coming onward from the foregoing description of the Red River valley, and from this review of the latest chapter in its geological history, to the consideration of its present settlement by white immigrants and the great development of its agricultural resources, especially in the cultivation of wheat, we may first bestow a n.oment's thought upon the red men who have been displaced. Fifty years ago almost countless herds of buffaloes roamed over this region of far-stretching prairie, and no one could have foretold that so soon the buffalo would be practically exterminated and the Indian's hunting ground changed to fields of waving grain.

The aboriginal tribes of Ojibways and Dakotas, living in the drainage basin of the Red River of the North had made little progress toward a system of agriculture which would provide their principal food during the whole year. Like the other tribes of hunting Indians who inhabited all the area of the United States, excepting its southwestern borders, their dependence was chiefly on the chase and entrapping of game and on fishing. But even their rude and very limited efforts in agriculture yielded an important and valued portion of their sustenance. In pre-Columbian times and onward to the present day, the Indians have cultivated small patches of land,

carefully tending their crops and storing up the harvest for gradual use during the rigors of winter and until the next harvest, supplementing thereby their principal diet of game and fish. Such aboriginal agriculture, untaught by white men, yet far from being despicable, I saw in September several years ago at the Ojibway village a mile southeast of the narrows of Red lake. This largest village of the Ojibways in Minnesota was designated on Nicollet's map in 1842, at so early a date that no white settlement was shown in this state. It now consists of thirty or forty permanent bark lodges, scattered on an area which reaches a half-mile from northwest to southeast and is forty to sixty rods wide. Adjoining the village were fields of ripening maize or Indian corn, amounting to about fifty acres, besides about five acres of potatoes, and probably an acre or more of squashes. These crops showed a luxuriant growth and abundant yield.

At a somewhat earlier time, of which no distinct tradition was preserved by the hunting tribes of Indians inhabiting this region, other tribes, who built the mounds and probably lived more by agriculture and less by the chase, overspread all the prairie districts of the Red River valley, extending also east in the wooded country to Rainy lake. The enduring earthworks erected by this people testify of their formerly wide extension throughout the Mississippi and Red River basins, and show that the sites of their villages were chosen usually on the banks and bluffs which overlook the food-giving rivers and lakes, often commanding an extensive and beautiful prospect. Most of the mounds in Minnesota, South and North Dakota, and Manitoba, are round and have the form of a dome, their height ranging from three to ten feet, or rarely more, above the general surface, with a diameter of thirty to 100 feet or more at their base. Nearly all of them were made by the people for the burial of their dead, and the relics found with their bones prove that they surpassed the present Indians of this region in having skill to make rude pottery; but the superiority was very slight, and there are no evidences of the development of handicrafts to a degree at all comparable with the aboriginal arts of Mexico and Peru. There was some commercial interchange from great distances, but it was probably limited to a few articles which were highly valued for beauty or regarded as mysterious and sacred. Thus in the mounds on the bluffs

2

of the Souris river and Antler creeks in southwestern Manitoba, tributary through the Assiniboine to the Red river, Prof. George Bryce has found orraments made of sea shells, others of copper from Lake Superior, and pipes from the sacred red pipestone quarry near the southwest corner of Minnesota, but no evidence of any intercourse with Europeans was found (Historical and Scientific Society of Manitoba, Transaction 24, 1886).

SETTLEMENT BY WHITE IMMIGRANTS.

The first immigration by white men to colonize the fertile basin of the Red River of the North, bringing the civilized arts and agriculture of Europe, was in the years 1812 to 1816, when, under Lord Selkirk's far-sighted and patriotic supervision, the early pioneers of the Selkirk settlements, coming by the way of Hudson bay and York Factory, reached Manitoba and established their homes along the river from the vicinity of Winnipeg to Pembina. In its beginning this colony experienced many hardships, but in the words of one of these immigrants, whose narrative was written down in his old age, in 1881, "by and by our troubles ended-war and famine and flood and poverty, all passed away, and now we think there is no such place to be found as the valley of the Red river."* ! Fifty to sixty years after the founding of the Selkirk colony, the margin of the advancing wave of immigration in the United States reached the Red River valley. In a few places on the Red river, the Wild Rice river of North Dakota, and the Sheyenne river, small bands of immigrant farmers had begun the settlement of this rich agricultural area a few years before the building of railroads across it; but the main tide of immigration came after the railroads had provided means of sending the staple product of the country, wheat, to the markets of St. Paul, Minneapolis, and Duluth. The Northern Pacific railroad was built from Duluth to Moorhead and Fargo during the years 1870 to 1872, and the next year it was extended to Bismarck. Within the next three years a line of the Great Northern railway (then the St. Paul & Pacific) was built to Breckenridge, and another line to Crockston and St. Vincent. From 1875 to 1885 the settlement of the Red River valley and of a large contiguous area of North and South Dakota went forward very

*Manitoba: Its Infancy, Growth, and Present Condition, by Prof. George Bryce, London, 1882, p. 166.