The duplication of the canals is perhaps the most difficult problem which Mars offers to us for solution. Even if we admit that the canals themselves represent inlets or channels through which the melted polar snow makes its way across the equatorial continents, it is not easy to see how the duplicate canals can arise. This is especially true in those cases where the original channel seems to vanish and to be replaced by two quite new canals, each about the breadth of the English Channel, and lying one on each side of the course of the old one. The very obvious explanation that the whole duplication is an optical illusion has been brought forward more than once, but never in a conclusive manner. We must, perhaps, be content to let the solution of this matter rest for the present, in the hope that the extraordinary attention which this planet is now receiving will in due time explain the present enigma. The markings on the surface of this planet are, generally speaking, of a permanent character, so that when we compare drawings made one or two hundred years ago with drawings made more recently we can recognise in each the same features. This permanence is, however, not nearly so absolute as it is in the case of the moon. In addition to the canals which we have already considered, many other parts of the surface of Mars alter their outlines from time to time. This is particularly the case with those dark spots which we call oceans, the contours of which sometimes undergo modifications in matters of detail which are quite unmistakable. Changes of colour are often observed on parts of the planet, and though some of these observations may perhaps be attributed to the influence of our own atmosphere on the planet's appearance, they cannot be all thus accounted for. Some of the phenomena must certainly be due to actual changes which have taken place on the surface of Mars. As an example of such changes, we may refer to the north-western part of the notable feature, to which Schiaparelli has given the name of _Syrtis major_.[18] This has at various times been recorded as grey, green, blue, brown, and even violet. When this region (about the time of the autumnal equinox of the northern hemisphere) is situated in the middle of the visible disc, the eastern part is distinctly greener than the western. As the season progresses this characteristic colour gets feebler, until the green tint is to be perceived only on the shores of the Syrtis. The atmosphere of Mars is usually very transparent, and fortunately allows us to scrutinise the surface of the planet without putting obstacles in the way m the shape of Martian clouds. Such clouds, however, are not invariably absent. Our view of the surface is occasionally obstructed in such a manner as to make it certain that clouds or mist in the atmosphere of Mars must be the cause of the trouble.