Since closer acquaintance takes from the maria their character of seas, we are led to inquire again into their constitution. Now, when we set ourselves to consider to what such appearances could be due we note something besides sea, which forms a large part of our earth’s surface, and would have very much what we suppose the latter’s aspect from afar to be, not only in tone, but in tint. This something is vegetation. Seen from a height and mellowed by atmospheric distance, great forests lose their green to become themselves ultramarine.

To dispossess a previous conception is difficult, but so soon as we have put the idea of seas out of our heads a vegetal explanation proves to satisfy the phenomena, even at first glance, better than water surfaces. In their color, blue-green, the dark areas exactly typify the distant look of our own forests; whereas we are not at all sure that seas would. From color alone we are more justified in deeming them vegetal than marine. But the moment we go farther into the matter the more certain we become of being upon the right road. With increased detection the markings they reveal and the metamorphoses they undergo, while pointing away from water, point as directly to vegetation. All the inexplicabilities which the supposition of water involves find instant solution on the theory of vegetal growth. The non-balancing of the areas of shading in their shift from one part of the disk to another, no longer becomes a circumstance impossible to explain, but a necessary consequence of their new-found character, denoting the time necessary for vegetation to sprout. The change of hue of vast areas from blue-green to ochre no longer presupposes the bodily transference of thousands of tons of substance, but the quiet turning of the leaf under autumnal frosts. Even the fact that they occupy those regions most fitted by figure to contain oceans fits in with the same conception. For that the Martian equivalents of forest and moorland, tree and grass, should grow now in the lowest parts of the planet’s surface is what might not unreasonably be expected from the very fact of their being low, since what remained of the water would tend both on the surface and in the air to drain into them.

For the change in question to be vegetal it must occur at the proper season of the planet’s year. This we must now consider. We have said that Schiaparelli detected change in the blue-green regions and suspected this change of seasonal affiliation. He inferred this from piecing together the aspects of different seasons of different years as shown in consecutive Martian oppositions. To mark it actually take place in a single Martian year came later. In 1894, at Flagstaff, the southern hemisphere was presented during its late spring and early summer; it was observed, too, for many of our months in succession. During this time the planet was specially well circumstanced for study of the change in that hemisphere, both by reason of the appositeness of the season and of the unusual size of the disk. Advantage was taken of the double event to a recording of the consecutive appearances certain regions underwent, and the contrasted states thus exhibited were such as clearly to betoken the action of seasonal change. What Schiaparelli had thus ably inferred from diverse portions of different Martian years was in this case shown occurring in one and the same semestral cycle.

Usually the change of hue seems essentially one of tone; the blue-green fades out, getting less and less pronounced, until in extreme cases only ochre is left behind. It acts as if the darker color were superimposed upon the lighter and could be to a greater or less extent removed. This is what Schiaparelli noted and what was seen in 1894 at Flagstaff. Three views en suite of the chain of changes then observed are shown in Mars, the region known as Hesperia being central in each. Comparison of the three discloses a remarkable metamorphosis in that “promontory,” a rise into visibility by a paling of its complexion. Nor is the contrast confined to it; changes as salient will be noticed in the pictures over the other parts of the disk.

There have been instances, however, of a metamorphosis so much more strange as to deserve exposition in detail; one where not tone simply is involved, but where a quite new tint has surprisingly appeared.

On April 19, 1903, when, after being hidden for thirty days, owing to the different rotation periods of the two planets, the Mare Erythraeum, the largest blue-green region of the disk and lying in the southern hemisphere, rounded again into view, a startling transformation stood revealed in it. Instead of showing blue-green as usual, and as it had done six weeks before, it was now of a distinct chocolate-brown. It had been well seen at its previous presentation, so that no doubt existed of its then tint. At that time the Martian season corresponded to December 30 in our calendar. Eighteen Martian days had since elapsed, and it was now January 16 there. The metamorphosis had taken place, therefore, shortly after the winter solstice of that part of the planet. The color change that had supervened proved permanent. For the next night the region showed the same brown hue, and so it continued to appear throughout the days that it was visible. Two months passed, and then the chocolate hue had vanished,—gone as it had come,—and the mare had resumed its usual tint, except for being somewhat pale at the south. It had come to be February 21 on Mars. Timed and tabulated, the metamorphosis through which the mare passed stands out thus:—

The culmination of the transformation seems to have taken place about 60 days after the southern winter solstice, or in the depth of the Martian winter of that hemisphere. This is certainly just the time at which vegetation should be at its deadest.

The northern and southern portions of the mare did not behave alike in taking on the chocolate tint. From the notes made about them during the opposition it appears that the latter was later than the former in undergoing the metamorphosis, as will be seen from the following depth of the blue green estimated in percentages shown at different dates, calling the deepest tone ever exhibited by it unity.

From this table we may place the lowest point of the blue-green tint as reached about the 22d of January for the northern, the 5th of February for the southern, part. This would indicate that the wave of returning growth came from the north, not the south; an important fact, as we shall see later in studying the action of the canals.

At the next opposition, in 1905, a recurrence of the transformation was watched for, and not in vain. It occurred, however, somewhat later in the Martian season. On December 27 of the planet’s current year the Mare Erythraeum was still as usual, blue-green, nothing out of the ordinary being remarked in it; and so it was on its January 17, although the southern edge was darker than the northern. It looked certainly as if the metamorphosis were this year to be omitted. But such was not the case. When the region again came round, on February 1 of the Martian calendar, there the strange tint was as unmistakable as it had been on its original occurrence. Not only was the Mare Erythraeum so colored, but on February 5 (Martian) the northern portion of the Mare Cimmerium was observed to be similarly affected. In the Mare Erythraeum the anomalous chocolate hue was confined to a belt between the latitudes of 10° and 20° south of the equator; in the Mare Cimmerium it stretched a little higher, from 10° at the west to 25° at the east. It is noteworthy that the southern portion of the latter showed blue at the time the northern showed brown. Then the metamorphosis proceeded as shown in the following table:—

Here, as in 1903, a chromatic rise and fall is evident; the culmination of the change occurring in Martian early February about ninety days after the winter solstice. That it was not of long duration is also indicated. If we examine the evidence for the two portions of the mare separately, the northern and the southern, as in 1903, we find it as follows:—

Here again a slight retardation in the advent of the metamorphosis is observable in the southern portion.

There would seem to be a difference in the time of the change between the two years of fifteen days, 1905 being by that much the later. But with points of reference themselves thirty days apart, it is possible the two more nearly coincided than here appears.

Unlike the ochre of the light regions generally, which suggest desert pure and simple, the chocolate-brown precisely mimicked the complexion of fallow ground. When we consider the vegetal-like blue-green that it replaced, and remember further the time of year at which it occurred upon both these Martian years, we can hardly resist the conclusion that it was something very like fallow field that was there uncovered to our view.

From the recurrence of the phenomenon on two successive years, it is likely that it annually takes place. That it is seasonal can scarcely be doubted from the timeliness of its occurrence, and that different portions of its terrane successively underwent their metamorphosis shows further that it followed a law peculiar to the planet, to which we shall be introduced when we come to consider the phenomena of the canals.

Instances of relative hue in different dark patches corroboratory of seasonal variation, and therefore of vegetal constitution, might easily be adduced. Thus, in 1905 during the summer of the northern hemisphere, the Mare Acidalium was notably darker than the Mare Erythraeum to the north of it, which is what the law of seasonal variation would require, since it was June in the one, December in the other at the time. But we need not to add example to example or proof to proof, for there are no phenomena that contradict it. We conclude, therefore, that the blue-green areas of Mars are not seas, but areas of vegetation. Just as reasoning to a negative result drifts us to the first conclusion, so reasoning to a positive one lands us at the second.