was this appearance, that, notwithstanding the fact that the supposition was distinctly opposed to what was at that time generally believed regarding the condition of the surface of our satellite, I determined to make a series of careful drawings of the apparent vapour column, in order to determine whether any variations in its outline might be detected from time to time, or whether, like a stain, it was immovably attached to the lunar surface. Since that time numerous drawings of it have been made, and of these, eight that were drawn in Arequipa, and eight that have been drawn since then in Cambridge, have been selected, and are arranged in Plate B, according to the portion of the lunar day in which they occurred. The scale is 11,000,000, or one inch equals sixteen miles. The drawings are oriented, as is customary in astronomical works, so that south shall be at the top and astronomical east at the right hand. Since the crater is near the Moon's eastern edge, the volume should be so turned that east will be at the top, in order to compare the sketches with analogous terrestrial phenomena.
To the east of the main crater and included in the area covered by the sketches are seven craterlets, which we will designate, beginning at the south, by the letters A, B, C, D, E, F and G. A, C and F are shown in the first figure, drawn October 14, 1891. B, C, D, E and G are well shown in the twelfth figure, drawn April 6, 1898. A casual examination of the sketches shows the great changes that are from time to time undergone by the vapour column, as we shall for convenience call it—changes that are readily detected by a six-inch telescope under ordinary atmospheric conditions.
The most marked of these changes depend for their existence upon the altitude of the Sun, for apparently no volcanic activity whatever is exhibited until about one day after sunrise. The activity then increases to a maximum, diminishes, and finally ceases a few days before sunset. The changes thus produced are naturally very marked, but, besides these, others are found to occur, which can be differentiated from them, and which seem to be quite as spasmodic, and quite as unamenable to any apparent law, as any volcanic phenomena that occur upon the surface of the Earth.
Thus comparing the appearance of the region during the two periods discussed, we find the most marked distinction perhaps lies in the appearance of the jet of vapour which issued from the main column in the direction of craterlet F during the earlier period and which is shown as a continuous white area in nearly all of the first eight drawings. In nearly all of the Cambridge drawings, on the other hand, a distinct break occurs in this jet. The next most striking difference between the two periods relates to
