Humboldt discussed the question a third of a century ago. "The natural highways of nations," said he, "will usually be along the great watercourses."

It impressed me deeply, as long ago as 1846, when the present enormous railway system of the continent had hardly begun to be developed. Spreading out a map of the Western Hemisphere, I then saw that from Cape Horn to Behring's Strait there was only one river-system that could be made available to commerce on the Pacific coast. In South America there is not a stream as large as the Merrimac flowing into the Pacific. The waves of the ocean break everywhere against the rocky wall of the Andes.

In North America the Colorado rises on the pinnacle of the continent, but it flows through a country upheaved by volcanic fires during the primeval years. Its chasms and cañons are the most stupendous on the globe. The course of the stream is southwest to the Gulf of California, out of the line of direction for commerce.

The only other great stream of the Pacific coast is the Columbia, whose head-waters are in a line with those of the Missouri, the Mississippi, the Red River of the North, and Lake Superior.

This one feature of the physical geography of the continent was sufficient to show me that the most feasible route for a great continental highway between the Atlantic and the Pacific must be from Lake Superior to the valley of the Columbia.

In childhood I had read the travels of Lewis and Clark over and over again, till I could almost repeat the entire volume, and, remembering their glowing accounts of the country,—the fertility of the valley of the Yellowstone, the easy passage from the Jefferson fork of the Missouri to the Columbia, and the mildness of the winters on the Western slope, the conviction was deepened that the best route for a railway from the lakes to the Pacific would be through one of the passes of the Rocky Mountains at the head-waters of the Missouri.

Doubtless, many others observant of the physical geography of the continent had arrived at the same natural conclusion. Seven years later the government surveys were made along several of the parallels, that from Lake Superior to the Columbia being under the direction of Governor I. I. Stevens. Jeff Davis was then Secretary of War, and his report set forth the northern route as being virtually impracticable. It was, according to his representation, incapable of sustaining population. A careful study of Governor Stevens's Report, and a comparison with the reports along the more southern lines, showed that the Secretary of War had deliberately falsified the statements of Governor Stevens and his assistants. While the surveys were being made, Mr. Edwin F. Johnson, of Middletown, Conn., the present chief engineer of the Pacific Railroad, published a pamphlet which set forth in a clear and forcible manner the natural advantages of the route by the Missouri.

In 1856 the British government sent out an exploring expedition under Captain Palliser, whose report upon the attractions of British America, the richness of the soil, the ease with which a road could be constructed to the Pacific through British territory, created great interest in Parliament.

"The accomplishment of such a scheme," said Mr. Roebuck, "would unite England with Vancouver Island and with China, and they would be enabled widely to extend the civilization of England, and he would boldly assert that the civilization of England was greater than that of America."

"Already," said the Colonial Secretary, Lord Lytton, better known to American readers as Bulwer, "in the large territory which extends west of the Rocky Mountains, from the American frontier and up to the skirts of the Russian dominions, we are laying the foundations of what may become hereafter a magnificent abode of the human race."