The question naturally arises, if the water of Mars is piled up at the poles as snow, how does it find its way back on its melting? A discovery made by Schiaparelli in 1877 revealed the existence of various lines marking the surface which he called _canali_, or channels.[3] These lines cover the face of the planet like a net, they are laid out with geodetic precision. "The lines start from points on the coast of the blue-green regions, commonly well-marked bays, and proceed directly to what seem centres in the middle of the continent, since, most surprisingly, they meet there other lines that have come to the same spot with apparently a like determinate intent." In other words these lines--fine, straight, dark, as if cut by an engraver, some of them running for hundreds of miles--converge at certain centres. They all start, as Schiaparelli first observed, from definite regions and terminate at definite points. Many of them follow the arcs of great circles. These lines may be thirty or more miles in width, apparently preserving the same width throughout, though slightly wider where they leave the dark bands. They run in every direction, a number often converging at a common centre, and, when they do so, a round, dark area appears which Lowell has called an oasis. In the clear and steady atmosphere of Flagstaff, Mr. Lowell, by the aid of his superb telescope, has added about four times as many canals as are shown on Schiaparelli's chart. These canals form an intricate network of lines, and no one can contemplate these curious features without being impressed by their artificial character. Schiaparelli, who first discovered them in 1877, continued his observations from year to year despite the fact that no one else could see them. In the course of a few years he discovered a still more remarkable condition, and this was that a number of the canals appeared double. This, indeed, seemed an optical illusion, and by no means strengthened his position, as the single canals proclaimed by him were supposed to be figments of the imagination. Undeterred by the general scepticism, Schiaparelli established, at each fresh opposition, his previous announcements. For nine years no one was able to confirm his marvellous discoveries. In the year 1886, however, Perrotin, at Nice, with his assistant, Thollon, managed to make out a number of the canals, single and double, which were carefully drawn. Reference to Perrotin's work will be made further on. The reason why so few have seen them is the lack of observers with acute eyesight and patient devotion to the work, coupled with unsteady air. Size of aperture seems to be of little importance. That Schiaparelli, with an 8-1/3 inch glass, discovered the canals, while with the twenty-six inch glass of the Naval Observatory at Washington they have never been seen, is emphatic evidence of what a clear and steady atmosphere means in the study of delicate planetary markings.