in place and therefore not attached to the soil. Now the only other thing capable of catching the light before it reached the surface would be something suspended in the air, that is, a cloud. Deduction, therefore, from the rarity of the phenomenon alone showed that the projections must be clouds. Their behavior in detail entirely corroborates this deduction from their intermittence. Such was shown by the action of the projection of December 6, 1900, as set forth in a paper before the American Philosophical Society and such again by that of the one of May 26, 1903, as we shall now note. To begin with, we notice that the projection seen on May 26 was not found either _in situ_ or in size on May 27 and had wholly vanished on May 28, though the seeing was substantially the same if not better on the two nights succeeding that of its original detection. Hence in its own instance this projection proved an alibi irreconcilable with the character of a mountain mass. But it did more. It not only was not on the second evening what and where it had been on the first, but the remains of it visible on the second occasion showed clearly that it had moved in the meantime. Furthermore it was disappearing as it went, for it was very much smaller after the lapse of twenty-four hours. The something that caused it was not only not attached to the soil, but was moving and dissipating as it moved. Only one class of bodies known to us can account for these metamorphoses and that is: cloud. But what kind of cloud are we to conceive it to be? Our ordinary vapor clouds are whitish and this would be still more their color could they be looked at from above. The Martian cloud was not white but tawny, of the tint exhibited by a cloud of dust. Nor could this color have very well been lent it by its sunrise position, for other places equally situated to be tinged by the hues of that time of day, Baltia to wit, showed distinctly white. So that we must suppose it to be what it looked, a something of the soil, not beholden to atmospheric tinting for its hue; a vast dust-cloud traveling slowly over the desert and settling slowly again to the ground. Precisely the same general course of drifting disappearance was taken by the projection of December 7 and 8, 1900. And this, too, stood an unique apparition in the annals of its opposition. Clouds, then, and not mountains are the explanation of the projections on Mars, differing thus completely from the lunar ones. CHAPTER X THE BLUE-GREEN AREAS