in crevices instead of on a smooth surface. The Sun must therefore necessarily attain a certain altitude before they can become visible. This explanation is further sustained by the fact that the bright streaks are well seen under favourable circumstances three or four days after new moon, upon the dark side of the Moon, where they are illuminated only by light reflected from the Earth. This shows that they must be there throughout the lunar night, and that the reason that we do not see them at sunrise and sunset is merely on account of the direction from which the light reaches them.
Before attempting to explain their peculiar shape we must state certain facts with regard to them. Hundreds of these streaks exist upon the Moon, but in the great majority of instances they are found to issue from minute, intensely white craterlets. These craterlets are seldom over one mile in diameter, and are usually much less. The streaks are very brilliant when they issue from the craterlet, but broaden and grow fainter as they recede from it. Their maximum breadth seldom exceeds five miles and their length usually Ues between ten and sixty miles. The nature of the individual streaks is seen very clearly in the telescope in the crater Pitatus, 10C [2.7, 4.1]. It is also well shown in Copernicus, 11C[1.7, 7.1], but does not become clearly visible there until the day before full moon.
Those streaks which do not issue from minute craterlets usually lie upon or across ridges, or in other similarly exposed situations. Frequently a number of the craterlets producing streaks are found to be arranged in the same direction in which the streaks lie. In such a case, before one streak comes to an end another will begin, thus forming a nearly continuous white band. A few of these white bands extend for more than six hundred miles. The combination of the streaks to form such a band is most readily studied in the conspicuous marking stretching from Tycho to Mare Nectaris. (Frontispiece [2.5, 5.7]; see also Plate 6C.) Under favourable atmospheric conditions it can be seen in the two parallel bands extending from Tycho toward Kepler, Frontispiece [4.5, 5.7], and in the band stretching from Menelaus across the Mare Serenitatis [2.5, 2.6].
The visibility of the streaks can be studied very satisfactorily from photographs. They become visible in general about twelve hours after sunrise and brighten for one or two days. They remain visible until about the same period before sunset. Latitude has little if any effect on the time of their appearance. It has been suggested by Professor N. S. Shaler that the existence of prominent craters on the other side of the Moon might be indicated to us by the presence of bright streaks radiating from them
