white areas in the latter picture belong to the second class, that of partly bright regions.
The question now arises: Is this white material really snow? The fact that it gathers at the poles, on mountain peaks and about the rims of craters would lead us to suspect that such might be the case; but there are still other facts bearing upon the question which should first be considered. It will be convenient for this purpose to again refer to the crater Linné, 7A [1.4, 4.9], which has perhaps been more carefully studied than any other feature of the Moon's surface and about which our early knowledge is more reliable. This crater is surrounded by a halo of partly bright material, which becomes visible one day after sunrise. The diameter of the halo was measured a number of times by ten different astronomers during the years 1866, 1867 and 1868. For the most part, these measures lie between five and a half and nine and a half miles—a perplexingly wide range of variation. In 1897 and 1898 another series of measures of Linné was made by the author, who fotind that his measures, too, varied through a wide range, extending from two and a half to five miles. The result seemed inexplicable at first, until it occurred to him to compare the diameters of the area in question with the number of hours that it had been exposed, in each case, to the Sun. The whole matter then became clear. When the white spot first became visible, one and a half of our days after lunar sunrise, it was five miles in diameter. As the Sun rose, the spot rapidly diminished in size, until, one day after the lunar noon, it was only two and a half miles in diameter. From then on till one and a half days before sunset, when it disappeared, it steadily increased in size, reaching a diameter of four miles. During the lunar night it must have continued to increase, until after sunrise it again became, as before, five miles in diameter. We thus see not only that the spot was then permanently smaller than it had been thirty years before, but also that it was subject to a change in size dependent on the altitude of the Sun. The latter phenomenon is evidently analogous to that of the changing size of the polar caps of Mars and of our own Earth. The size of Linné may be compared on Plates 7A, 5B, 5C and 5E.
As the Sun rises higher and higher upon a lunar formation there are two diametrically opposite effects produced. One is a real, but comparatively inconspicuous, diminution in size of the white area, due to melting. The other is a very conspicuous increase, which, however, is really only apparent, and is due to the shining of the Sun into the hollows and crevices of the surface, and thus illuminating white areas which had not
