5. The average number of aerolitic falls in a year was estimated by Schreibers at 700. Baron Reichenbach, however, after a discussion of the data at hand, makes the number much larger. He regards the probable annual average for the entire surface of the earth as not less than 4500. This would give twelve daily falls. They are of every variety as to magnitude, from a weight of less than a single ounce to over fifteen tons. The baron even suspects the meteoric origin of large masses of dolerite which all former geologists had considered native to our planet. 6. An analysis of any extensive table of meteorites and fire-balls proves that a greater number of aerolitic falls have been observed during the months of June and July, when the earth is near its aphelion, than in December and January, when near its perihelion. It is found, however, that the reverse is true in regard to bolides, or fire-balls. These facts are susceptible of an obvious explanation. The fall of meteoric stones would be more likely to escape observation by night than by day, on account of the relatively small number of observers. But the days are shortest when the earth is in perihelion, and longest when in aphelion; the ratio of their lengths being nearly equal to that of the corresponding numbers of aerolitic falls. On the other hand, it is obvious that fire-balls, unless very large, would not be visible during the day. The _observed_ number will therefore be greatest when the nights are longest; that is, when the earth is near its perihelion. This, it will be found, is precisely in accordance with observation. CHAPTER VIII. SHOOTING-STARS.--METEORS OF NOVEMBER 14.