passes through several small lakes, and in one of these it becomes lost, reappearing again as springs at a lower level. According to Nicollet, this is the "infant Mississippi," the "cradled Hercules," whose power at maturity was sufficient to cause the continent to tremble, or to smile. Both he and Schoolcraft failed to observe another stream whose entrance into Itasca lake is constantly hid by rushes, but which leads to Elk lake. This stream was entered by Glazier. If the actual source of the Mississippi be pursued to higher levels than lake Itasca, the competition for the honor lies between these two streams. The essential facts are now well established by surveys. The Nicollet valley has been accepted as the chief tributary above Itasca lake by Nicollet and Brower, the latter being the surveyor who examined the whole region and reported, with maps and full data of all kinds, to the Minnesota Historical Society. The Elk lake valley, with its chief stream, Excelsior creek, is represented by Glazier as the principal tributary above Itasca lake.

The question may be relieved of all side issues and narrowed down to two propositions:

1. Which is the larger and longer valley?

2. Who discovered these valleys and water courses?

It is a singular fact, as appears from the representations of Glazier, that Elk lake was not seen either by Schoolcraft or by Nicollet, although they were both in pursuit of the source of the Mississippi under the guidance of the Indians, a fact which indicates the estimate put by the Indians on the relative importance of these streams. The actual measurement of these streams has been made at their mouths, by several persons. The Nicollet stream, which is in the continuation of the main valley of Itasca lake to the southwest, according to Glazier has a width of ten feet and a depth of two and a half feet. The Elk lake stream has a width, by the same authority, of seven feet, and a depth of three feet. The channels are, therefore, in point of capacity, as the numbers 25 to 21. If the velocity of the streams be considered the same, the Nicollet creek would carry nearly 20 per cent. more water than the Elk lake stream. But according to the descriptions, the Nicol

let creek is more rapid than the Elk lake creek, and may be estimated to carry twice as much water as the Elk lake stream.

In point of view of the length of the two valleys, or, more correctly, of the two streams, Nicollet and Brower trace Nicollet creek to a distance of several miles above Itasca lake, but Glazier allows this stream only a length of a mile and threeeighths. The valley which is drained by Excelsior creek, the chief tributary of Elk lake, Mr. Glazier followed to a distance beyond Itasca lake of 14,106 feet. From these data he decides that the length of "running water" is much greater in the Elk lake valley. There are, however, several facts bearing on the length of Nicollet creek which Mr. Glazier does not mention. He traces it up to a great spring. He is willing to suppose that a stream whose depth is two and a half feet, with a width of ten feet, may have its gathering area all embraced within a mile and three-eighths from its debouchure. Had an explorer, intent on finding the source of a stream, found it issuing apparently from the ground with such a volume, his own judgment would have driven him to search further up the valley, as Nicollet, Clarke and Brower did. He would there have found the same stream reappearing, and again disappearing. Sometimes in lakes, or in marshes, lost to sight as running water, like a "bashful maiden," as described by Nicollet, finally plunging under a screen of vegetable debris, bogs, peat, and floating driftwood, much overgrown with small trees, only to come to the light of day again at the "great spring," 7,307 feet from Itasca lake.

The length of this water course, thus included, is considerably more than the farthest traceable limit of Excelsior creek. It may not be in lake Hernando de Soto, as supposed by Brower, that the highest actual water of Nicollet creek can be identified, but it is certainly several thousand feet above the point adopted by Mr. Glazier. In northern Minnesota, where vegetation is rank and the materials in which it grows are loose, like the sandy soils about Itasca lake, it is no uncommon occurrence to find small streams blocked by such obstructions. They spread out, disappear in marshes, plunge under floating bogs or driftwood, and issue at lower levels. The St. Louis river, the principal stream entering at the head

of lake Superior, was permanently invisible for the distance of nearly a mile, near Cloquet, until within a few years. It flowed under a mass of floating driftwood on which grew small birches and aspens. Lumbermen finally cut the driftwood away for the purpose of floating logs to lower points. The celebrated raft of the Red river in Arkansas is a parallel case. The principle is the same as with the obstructions of Nicollet creek. Such interruptions of "running water" are not limitations of the valleys, nor of the streams that drain them. They are non-essential accidents, and cannot be considered as having any important bearing on the true size and length of Nicollet creek.

This important omission of an essential fact in the investigation conducted by Glazier, seems to be fatal to the claims of Elk lake and Excelsior creek.

We next ask: Who discovered Elk lake, which has now been named "Glazier lake" by the recent travellers? It was thought, at one time, that Julius Chambers entered it in 1872, but Mr. Glazier shows that his description applies rather to one of the lakes of the Nicollet valley. Mr. Glazier found it in 1881. He hastily promulgated it as a new discovery, announcing this at various points on his way to the mouth of the Mississippi. In 1875, however, this region had been surveyed by the officers of the United States Land Survey, under Gen. James H. Baker. This lake was platted and reported, in the regular manner, to the Government at Washington, under the name which it seems to have borne among the Indians and early explorers, Elk Lake. As such it has gone into the official records. The Minnesota Historical Society has approved this nomenclature, and finally the Minnesota Legislature has passed a law declaring that in the public schools of the state no geography shall be used by the pupils which gives this lake any other name. The fact of the earlier naming of this lake is not disputed by Glazier. He claims priority on the ground that the business of the land surveyors was not to discover the source of the Mississippi, that they did not trace out its feeders, and that they did not make wide publication of their discovery. If these be considered fatal defects in the governmental discovery of this lake, it is probable that there will be no objection to admitting the priority of Glazier.

When a careful and dispassionate examination is made of the essential facts, as now known, the conclusion is forced that Mr. Glazier fails to substantiate his claim. A hasty examination of his last work, without a full knowledge of the facts brought out by the Minnesota Historical Society's survey, would lead to the favorable consideration of his claim, since he evades the adverse facts and dwells on the repeated assertions of his friends and followers, and on the favorable showing which he is able to make in respect to the lake which he found in 1881. History and geography cannot be promoted by such partial and interested advocacy.

If Itasca be not allowed to stand as the source of the Mississippi, the competition lies between Nicollet and Excelsior creeks, and the former has the greater length and volume. If it is necessary to choose a lake as its source, then some of the upper lakes of the Nicollet valley must be accepted. If, finally, it be necessary to accept Elk lake, that lake was first discovered and mapped by the United States Surveyors in 1875. Mr. Glazier's claims, in every respect and in any case, are thus annulled, on the basis of facts which, if he does not himself publish, he does not call in question.

PREHISTORIC MAN AT THE HEADWATERS OF

THE MISSISSIPPI RIVER.*

BY HON. J. V. BROWER.

I.

PRELIMINARY REFERENCES.

At pages 123 and 124, Vol. VII., Minnesota Historical Collections, prepared and submitted by me in 1889-93, the following appears:

Concerning the presumable fact, that, antedating the first known visit of white men at Lac La Biche, French voyageurs may have reached the basin, no reliable statement in writing is known to exist describing such visit. In the absence of any known record as to the movements of the French fur traders and voyageurs who first established themselves in lines of trade and traffic with the Indians across the northern portion of the territory which now constitutes the State of Minnesota, no definite record can be found concerning a mere probability that they may have reached Elk lake. To the writers of the future must be left the task of discovering the record of the manner in which "Lac La Biche" first became known to the French and of any visits they may have made to the locality, if any such record exists, which now seems doubtful. Certain it is that Mr. Morrison's letter is the only record of the first visit to the source of the Mississippi of which we have any knowledge.

Upon page 16 of my report to his Excellency the Governor of Minnesota, for the two years ending Dec. 1, 1894, the fol lowing tabulated historical record of the descent of title by possession appears:

*Abridged extract from the Journal of the Manchester (Eng.) Geographical Society, vol. XI., pp. 1-80, 1895; to which is appended an addendum, relating to the early visits of Mr. Julius Chambers and Rev. J. A. Gilfillan to Itasca lake, prepared for the Minnesota Historical Society by Mr. Brower.