HE time was when mankind, igno

Trant of the laws of evaporation

and condensation, looked to Heaven above for their water, and piously attributed to their gods the direct dispensations of flood and drought.

They did not suspect that the pitiless flood, on wings of gloom, had, yesterday, murmured in their lakes and brooks; that the snow, this morning so pure, so gentle, so lovingly covering away earth's unsightliness, closed, last night, over a lonely ship and washed the light from despairing eyes.

The science of 1868, however, regards the evaporation and the precipitation of water over the whole earth's surface as equal, evaporation being in great excess over precipitation upon the ocean only, while precipitation is excessive upon mountain ranges and toward the poles.

A certain ingenious friend of ours recently threw out the startling sugges tion that water is, at times, formed suddenly and in vast quantities by a direct combination of its elements in some unknown and unaccountable manner; and pointed us, in proof, to water-spouts and those sudden avalanches of water which are known to inundate mountain defiles without warning, sweeping everything before them in demoniac fury— such, perhaps, as that which has recently devastated Switzerland, and which are not uncommon in the tropics. We were not convinced, although our friend instanced and described such a torrent, which he had himself witnessed in the Rocky Mountains, while the air was perfectly calm and clear, and was posi tive in his statement that there was not, nor had there been, for days, any storm either near or remote.

It may not be impossible-in contravention of the old aphorism that no matter has been added to the earth since its creation-that once or twice in a generation a cosmical or cometary mass of water, or the elements thereof, should be discharged upon us, coming, in very truth, from the stars. We do not see a greater absurdity in accepting the occasional advent of such a meteorite than of the nickeliferous iron stones which do often fall to the earth. The anxious watchers of the 14th of November star-showers may possibly some time secure a douche bath visitation to cool off their ardor, undistinguishable from an ordinary rain storm, which yet may be as truly superterrestrial in origin as those wonderful iron stones which descend to us through so fiery a baptism.

Our inquiry, however, "Where does the water come from?" pertains to a limited portion of the earth's surface. In its discussion, we propose to admit into our consideration only such deductive reasoning as the present status of knowledge warrants.

Generally, over large areas, the amount of evaporation and the discharge by springs, lakes and rivers to the sea are completely measured by the rain-fall of the region; but over limited areas this law does not always obtain, as is assuredly the case with those gardens in the African deserts, called oases. The waters of these springy spots, which bring light, life and beauty into the very heart of Sahara, come from distant and happier regions.

The existence in the desert of a considerable underflow of water was demonstrated during the recent invasion of Abyssinia by the English, when Arte

sian wells, now proving to be the nuclei of other oases, were sunk along the projected line of march. But for this happy expedient, it is said that it would have been impossible for the British army successfully to invade the country.

Most large rivers have their sources in mountain regions, where the amount of rain-fall is very considerable, owing to the condensing power of the cool summits of mountains upon all aqueous vapor which may be brought in contact therewith, the waters from which descend with too great rapidity to the plains below to be greatly diminished by evaporation. But there is a certain table-land, most of it within the boundaries of the United States, which eliminates more water than any equal portion of the earth's surface not characterized by a great chain of mountains.

Let the reader cast his eyes over the map of North America, and note a certain section of country, embracing nearly the whole of Minnesota, a portion of the Territory of Dakota, western Wisconsin, a part of Michigan, and a small portion of British America bordering Lake Superior.

He cannot fail to be struck with the fact that this country, with an area of about one hundred and seventy thousand square miles, gives origin to Lake Superior, the first and grandest of our great North-American chain of lakes; the Mississippi, one of the largest rivers of the globe, flowing to the South; the Red River of the North, whose waters find their way at last through Lake Winnepeg to Hudson's Bay; and two very considerable rivers, the Big Sioux and the St. Jacques, which flow into the Missouri.

This region we would indicate by a line drawn from the western extremity of the State of Michigan on Lake Superior to the foot of Lake Pepin on the Mississippi; thence westwardly, across Minnesota, to the Coteau des Prairies, or the divide between the headwaters of

the St. Peter's river and the Big Sioux; then northwestwardly, along the divide between the tributaries of the Missouri and the Red River of the North, to the twenty-third degree of longitude west from Washington, and the boundary line of the United States and British America; thence eastward, following the boundary line to the northern shore of Lake Superior. The area included is about equal to twice that of the State of Minnesota. This region is certainly the greatest fresh-water ooze in the world. The portion of Minnesota and western Wisconsin which we have indicated, in particular, received from Nicolet the somewhat fanciful name of "Undine Region," suggested by the great number of lakes therein. According to Schoolcraft and others, there are over ten thousand within the limits of Minnesota alone.

Many of these lakes are strung together, forming an almost labyrinthian maze. With but little portage, nearly the whole area above mentioned can be circumnavigated by boats of moderate size.

The description given by Sir I. Richardson, in his narrative of an overland expedition in search of Sir John Franklin, of the "ridge," or divide, between the waters of Lake Superior and of Lake Winnepeg, would answer for much of the Minnesota country: "The surface of that tract is characterized by rounded and sometimes rugged knolls of granite, rising abruptly from lakes and swamps, but only to small hights, above the general level. The term ridge is used with reference to its being a hight separating two depressions, but its summit is a marshy plateau of some extent, across which narrow winding lakes afford a canoe navigation in a variety of directions." The State of Minnesota has certainly more lakes gemming its bosom than any other equal portion of the earth's surface of which we have any account.

This table-land, throughout its extent, unlike most heavily-watered districts, is characterized by a remarkable uniformity of elevation. It is only about fourteen hundred feet above the sea. The highest ridges or elevations of the country are the Hauteur des Terres and the Coteau des Prairies, the former being the appellation of the low ridge of drift accumulation dividing the basin of Lake Superior from the headwaters of the Mississippi river, with an elevation at its highest observed point of sixteen hundred and eighty feet above the sea, and only one hundred and thirty feet above Itasca lake, — the beautiful sheet in which the Mississippi has its origin. This moderate elevation of sixteen hundred and eighty feet, as determined by Nicolet, is the more surprising, when we consider that it is the highest continental elevation between the Gulf of Mexico and the northern seas, and is distant from the mouth of the Mississippi more than three thousand miles.

The "Coteau des Prairies," the peculiar ridge which divides the valley of the St. Peters or Minnesota river from that of the Missouri, has an elevation of only nineteen hundred feet above the sea.

The moderately elevated plateau of Minnesota, discharging its waters to the four quarters of the globe, is a wonderful mesh or net-work of lakes, and the fountain head of so many mighty waters, and yet the precipitation of moisture from the atmosphere, in the shape of rain and snow, is not great, indeed is far below the average in the Eastern, Middle, Western and Southern States.

We are, we think, prepared to show with an approximation to certainty, that a portion of the water discharged from this region is of subterranean origin, coming probably from a great distance, urged by some great hydraulic power, even from the Rocky Mountains.

There are few, perhaps, who do not understand the principle of the Artesian Well.

It is usually constructed by

boring downward, until a more or less perfect water-shed is reached, i. e., a rock which permits little or no water to percolate through it. It is necessary that this rock or water-shed should come to the surface in some elevated region, as for example, in a chain of mountains. In such localities, the surfacewater received from the atmosphere comes in contact with this water-shed, and follows the stratum along its dip, often to a great distance. If, then, we penetrate by boring to this stratum, although at a point hundreds of miles from its mountain outcrop, the water from the interior gushes up, and generally with an astonishing force and volume, and often from a great depth. The continental water-shed with us is undoubtedly the igneous and metamorphic rocks underlying the true sedementary formations. The Potsdam sandstone and calciferous sand-rock, above, furnish easily worn veins, reservoirs or passages for the water.

The formations of which we speak find a full development and exposure in the Rocky Mountains, and do not again revisit the surface in the interval until we reach Minnesota, where, at a moderate elevation above the sea, the subterranean waters must experience the full force of that wonderful hydraulic power, which we see illustrated in Artesian wells, and be discharged over the surface in the greatest profusion. Such is our theory.

We were led to indulge in the foregoing conjectures, many years ago, while living in Minnesota, and at a time when there had been no careful examinations of the aqueous discharge from Lake Superior through Sault St. Marie or of the Mississippi.

Quite recently the United States Engineer Corps, under the superintendence of General W. F. Reynolds and Assistant D. F. Henry, have made several careful water sections of the Sault St. Marie, the St. Clair river, and Niagara. The

Mississippi has also been carefully gauged, under the direction of General Humphreys of the Engineer Corps, in view of new ship-canals in process of construction at Keokuk, Iowa, and at Rock Island, under the enlightened superintendence of General I. H. Wilson of the Engineer Corps.

The general dryness of the atmosphere in Minnesota and the great average heat of the summer months in that country, added to the remarkable level uniformity of its surface, it being char acterized by nothing in the remotest degree resembling mountains or considerable elevations, such as distinguished most regions where great water-courses take their rise-impressed us with a strong conviction of error in ascribing all the water emanating from this region to immediate atmospheric influences.

The general absence of those irregularities of surface which insure the rapid drainage of a country, and the full scope given to surface evaporation by the retention of the water precipitated upon its area, as rain or snow, added an intensity to our convictions only to be hightened by the knowledge of the physical fact that the immense area occupied throughout this country by lakes and swamps must perforce yield up, by continual evaporation from their surface through the year, a very great excess over precipitation.

We extract from the Surveyor General's Report: Total area of the State of Minnesota, 51,479,000 acres, and 32,000,000 acres arable land, leaving for lakes and swamps 19,479,000 acres, or considerably over one-third its whole surface.

A striking peculiarity of the climate, during the winter months, over this whole region, is also very suggestive of the truthfulness of our views. The average amount of precipitation in the form of rain and snow for the three winter months, in Minnesota, is two inches, and over west Wisconsin and the

Lake Superior region it is three inches. These averages are taken from Blodget's tables in his valuable work upon the Climatology of North America, and are founded upon careful observations continued over a period of more than thirty years, at various military posts. The above small precipitation for the winter months affords a striking contrast for the same period with the average precipitation over the Atlantic States, which is from ten to thirteen inches.

Notwithstanding the small amount of winter precipitation in the country we have been considering, and the additional fact that the severity of the cold during the winter months precludes the possibility of any considerable portion of the snow or water of precipitation finding its way to the Mississippi or to Lake Superior. Yet the Mississippi and the Sault St. Marie hold their volume of water in the interim very remarkably, or rather, the decline of water stage during the winter months is not unusual as compared with most rivers in the United States.

Another remarkable peculiarity which has not escaped popular attention in Minnesota is this: A very large proportion of the snow that falls is evaporated as snow from the surface, so that there is no particular rise of the river on the breaking up of winter. During the advance of winter we have repeatedly heard individuals remark, that the snows over the surface of the country in Minnesota had almost everywhere disappeared, although there had not been a single thaw during the winter. It is a popular notion, in which some meteorologists have indulged, that the evaporation is very inconsiderable at or below thirty-two degrees Fahrenheit, or rather during the winter. Any house-wife, however, could instruct us better, with her oft-repeated observation, that her clothes "are freezing dry," so that her weekly washing and drying need not be interrupted by any possible intensity of

cold.

The real fact is, that evaporation, or the drying process, goes on at all

known temperatures.

The great table-land which we have been considering varies, as we have said, but little in elevation, its lakes and streams being from fourteen hundred to sixteen hundred feet above the sea. The lowness of the divide between the head-waters of the Minnesota or St. Peters river and the Red River of the North is evidenced in the fact that with almost no portage boats can pass at all seasons between the sources of the two rivers. We have conversed with voyageurs who stated that they had often floated from the lake-like expanse at the head of St. Peters into Lake Traverse, the head of the other.

At the risk of some tediousness, we will now introduce our calculations concerning the amount of precipitation of water from the atmosphere, and the amount eliminated by evaporation and discharge for the greater portion of the areas we have been considering.

In Lake Superior, which drains a very limited area, and whose aqueous discharge through the Sault St. Marie has been, of late, carefully estimated, we find no very complicated problem in determining the ratio of evaporation and discharge to the annual precipitation it receives. The land area drained by the lake, excluding its own area, is about 70,000 square miles. The average annual rain-fall over this region is thirty inches, according to Blodget's Rain Chart and the Meteorological Tables. The annual average evaporation over the general land surface in England is estimated at twenty-three inches, which we will take as our standard in the Lake Superior region, although it is well known that, owing to the great natural dampness of the air, in England, the evaporation is much less than anywhere in the United States. By deducting the above amount of annual evaporation-twenty-three inches from

thirty inches, the observed annual rainfall for this region-we have seven inches as the annual excess of precipitation over evaporation. This will give for 70,000 square miles an annual yield of 40,833 square miles of water, one foot in depth.

The lake, being continuously exposed to evaporation, loses much more water thereby than is received from the rainfall upon its surface. The annual evaporation from an exposed reservoir of water protected from the rain, has been found, in the experiments of Dr. Holyoke, continued over a series of years, at Salem, Mass., to average fifty-six inches. Salem, however, has an average annual temperature considerably above that of Lake Superior. We will, accordingly, place the average annual evaporation at fifty inches-the amount observed at Syracuse, N. Y., where the annual average temperature is a little lower than at Fort Snelling, Minnesota. This gives an excess of twenty inches per annum of evaporation over precipitation.

The area of Lake Superior is 32,000 square miles; hence, 53,333 square miles of water one foot deep is the annual excess of evaporation for the lake. The discharge of water at Sault St. Marie is 90,900 cubic feet per second, which is 90,740 square miles of water one foot deep per annum. То this, add 53,333 square miles, the amount lost by evaporation from the lake surface, and the result is 144,573 square miles one foot deep, as the total discharged and evaporated from Lake Superior per annum. If we now deduct 40,833 square miles of water one foot deep-the excess of precipitation for the land area drained by the lake-we have 140,500 square miles of water one foot deep, which must be received from subterranean sources. Undoubtedly, should this underground supply be cut off from Lake Superior, the lake would be drained by evaporation alone.

We are positive that we underesti