The Suletind.—There are few watersheds in the world better defined than the mountain range which forms the "back-bone" of Norway. I lately climbed a peak in this range, the Suletind. From its summit I could look down into the valleys of the Lära and the Bägna, flowing in opposite directions to opposite sides of the peninsula. To the north of the Suletind is a large double lake called the Sletningenvand. The maps show this lake to be one of the chief sources of the westward-flowing river Lära. This lake is in August swollen by the melting of the snows, and at the time of my visit it was visibly the source of both these rivers. From its southeastern side flowed a large brook into the valley of the Bägna, and from its southwestern corner, equally distinctly, came the waters which fed the Lära. This lake, like similar mountain ponds in all northern countries, abounds in trout; and these trout certainly have for part of the year an uninterrupted line of water communication from the Sognefjord on the west of Norway to the Christianiafjord on the southeast,—from the North Sea to the Baltic. Part of the year the lake has probably but a single outlet through the Lära. A higher temperature would entirely cut off the flow into the Bägna, and a still higher one might dry up the lake altogether. This Sletningenvand, with its two outlets on the summit of a sharp watershed, may serve to show us how other lakes, permanent or temporary, may elsewhere have acted as agencies for the transfer of fishes. We can also see how it might be that certain mountain fishes should be so transferred while the fishes of the upland waters may be left behind. In some such way as this we may imagine that various species of fishes have attained their present wide range in the Rocky Mountain region; and in similar manner perhaps the Eastern brook trout[117] and some other mountain species[118] may have been carried across the Alleghanies.
The Cassiquiare.—Professor John C. Branner calls my attention to a marshy upland which separates the valley of the La Plata from that of the Amazon, and which permits the free movement of fishes from the Paraguay River to the Tapajos. It is well known that through the Cassiquiare River the Rio Negro, another branch of the Amazon, is joined to the Orinoco River. It is thus evident that almost all the waters of eastern South America form a single basin, so far as the fishes are concerned.
As to the method of transfer of the trout from the Columbia to the Missouri, we are not now left in doubt.
Two-Ocean Pass.—To this day, as the present writer and later Evermann and Jenkins[119] have shown, the Yellowstone and Snake Rivers are connected by two streams crossing the main divide of the Rocky Mountains from the Yellowstone to the Snake across Two-Ocean Pass.
Prof. Evermann has described the locality as follows:
"Two-Ocean Pass is a high mountain meadow, about 8,200 feet above the sea and situated just south of the Yellowstone National Park, in longitude 110° 10' W., latitude 44° 3' N. It is surrounded on all sides by rather high mountains except where the narrow valleys of Atlantic and Pacific creeks open out from it. Running back among the mountains to the northward are two small canyons down which come two small streams. On the opposite is another canyon down which comes another small stream. The extreme length of the meadow from east to west is about a mile, while the width from north to south is not much less. The larger of the streams coming in from the north is Pacific Creek, which, after winding along the western side of the meadow, turns abruptly westward, leaving the meadow through a narrow gorge. Receiving numerous small affluents, Pacific Creek soon becomes a good-sized stream, which finally unites with Buffalo Creek a few miles above where the latter stream flows into Snake River.
"Atlantic Creek was found to have two forks entering the pass. At the north end of the meadow is a small wooded canyon down which flows the North Fork. This stream hugs the border of the flat very closely. The South Fork comes down the canyon on the south side, skirting the brow of the hill a little less closely than does the North Fork. The two, coming together near the middle of the eastern border of the meadow, form Atlantic Creek, which after a course of a few miles flows into the Upper Yellowstone. But the remarkable phenomena exhibited here remain to be described.
"Each fork of Atlantic Creek, just after entering the meadow, divides as if to flow around an island, but the stream toward the meadow, instead of returning to the portion from which it had parted, continues its westerly course across the meadow. Just before reaching the western border the two streams unite and then pour their combined waters into Pacific Creek; thus are Atlantic and Pacific creeks united and a continuous waterway from the Columbia via Two-Ocean Pass to the Gulf of Mexico is established.
"Pacific Creek is a stream of good size long before it enters the pass, and its course through the meadow is in a definite channel, but not so with Atlantic Creek. The west bank of each fork is low and the stream is liable to break through anywhere and thus send part of its water across to Pacific Creek. It is probably true that one or two branches always connect the two creeks under ordinary conditions, and that following heavy rains or when the snows are melting, a much greater portion of the water of Atlantic Creek crosses the meadow to the other side.
