general system is, curiously, one for the whole planet, the particular character of different parts of the surface alters the action there to some extent.

For an appreciation of the meaning of the changes, it is to be borne in mind throughout that the vernal equinox of Mars' southern hemisphere occurred on April 7, 1894; the summer solstice of the same hemisphere on August 31; and its autumnal equinox on February 7, 1895.

On the 31st of May, therefore, it was toward the end of April on Mars. The south polar cap was, as we have seen, very large, and the polar sea in proportion. That the polar sea was the darkest and the bluest marking on the disk implies that it was, at least, the deepest body of water on the planet, whether the so-called seas were seas or not. But from the fact that it was quite wide,--350 miles,--and that it all eventually vanished, it can hardly have been very deep. Its relative appearance, therefore , casts a first doubt upon the fact that the others were seas at all. This polar sea plays deux ex machina to all that follows.

So soon as the melting of the snow was well under way, long straits, of deeper tint than their surroundings, made their appearance in the midst of the dark areas. I did not see them come, but as I afterward saw them go it is evident that they must have come. They were already there on the last day of May. The most conspicuous of them lay between Noachis and Hellas, in the Mare Australe. It began in the great polar bay, and thence traversed the Mare Erythraemum to the Hourglass Sea (Syrtis Major). The next most conspicuous one started in the other bay, and came down between Hellas and Ausonia. Although these straits were distinguishably darker than the seas through which they passed, the seas themselves were then at their darkest. The fact that these straits traversed the seas suffices to raise a second doubt as to the genuineness of seas; the first suspicion as to their character--coming from their being a little off color not so blue, that is, as what we practically know to be water, the polar sea--finding thus corroboration. It will appear later that in all probability the straits themselves were impostors, and that neither seas nor straits were water.

The appearance of things at this initial stage of the Martian Nile-like inundation last June was most destructive to modern maps of Mars, for all the markings between the south polar cap and the continental coast-line seemed with one consent to have, as nearly as might be, obliterated themselves.

It was impossible to fix any definite boundaries to the south temperate chain of islands, so indistinguishably did the light areas and the dark ones merge into each other. What was still more striking, the curious peninsulas which connect the continent with the chain of islands to the south of it, and form so singular a feature of the planet's geography, were invisible. One continuous belt of blue-green stretched from the Syrtis Major to the Columns of Hercules.

For some time the dark areas continued largely unchanged in appearance; that is, during the earlier and most extensive melting of the snow-cap. After this their history became one long chronicle of fading out. Their lighter parts grew lighter, and their darker ones less dark. For, to start with, they were made up of many tints; various shades of blue-green interspersed with glints of orange-yellow. The gulfs and bays bordering the continental coast were the darkest of these markings; the long straits between the polar sea and the Syrtis Major were the next deepest in tone.

The first marked sign of change was the reappearance of Hesperia. Whereas in June it had been practically non-existent, by August it had become perfectly visible and in the place where it is usually depicted. In connection with its reappearance two points are to be noted: first, the amount of the change, for Hesperia is a stretch of land over two hundred miles broad by six hundred miles long; and, secondly, the fact that its previous invisibility was not due to any sort of obscuration. The persistent clear-cut character of the neighboring coast-line during the whole transformation showed that nothing of the nature of mist or cloud had at any time hidden the peninsula from view. A something was actually there in August which had not been there in June.

As yet nothing could be seen of Atlantis. It was not until the 30th of October that I caught sight of it. About the same time, the straits between the islands, Xanthtus, Scamander, Psychrus, and Simois, came