inferior only to those of November in numbers and brilliancy. In the year 268, in 859, and in 1451, star showers are recorded by the Chinese historians as occurring in August. About the 9th of August, 1798, Dr. Noah Webster observed great numbers of meteors in Hartford during several nights. They moved from northeast to southwest, and succeeded each other so rapidly as to keep the eyes of the spectator constantly engaged. And, in almost every year since, about the same date, a similar phenomenon has been seen on a small scale, so that the August meteors are now as well known as those of November. The shower in August is not on a great scale, but several hundred meteors may usually be counted on the night of the 10th and 11th.

The November showers have lately been the object of special study, not only on account of their peculiar brilliancy, but because of the interest attaching to the recurrence of their period. The shower seen by Humboldt and Bonpland from the Andes, on November 12, 1799, and described in the narrative of Humboldt's travels, is well known. On the night of November 13, 1833, a display which, from all accounts, must have been one of the greatest on record, was seen in this country. Throughout the South the negroes, like the Europeans of a previous century, thought the end of the world had come at last. It was carefully observed at New Haven by Professor Olmsted. He was the first to elaborate a theory of the cause of the phenomenon; and though his ideas are now known to be fundamentally erroneous, they contained some elements of truth. '

The recurrence of the shooting-stars thirty-four years after they had been seen by Humboldt on almost the same date, suggested to Olbers the idea of a thirty-four year period, and led him to predict their return in 1867. The idea was elaborated, and its correctness proved by Professor H. A. Newton of Yale College. The thoroughness with which this gentleman has investigated the subject of shooting-stars, particularly those of November, has rendered him pre-eminent in this department of astronomy. Collating the accounts, he found a long series of recorded apparitions at intervals of one third of a century, extending, with many breaks, from A. D. 931 to 1833.

The exact date of appearance, however, instead of being uniform, changed with considerable regularity. The shower of 931 was described as occurring on the last day of Muharram, in the year 599, according to the Arab chronology, which corresponded to the 19th of October. The successive showers appeared at a later and later date until the present time, when they occur on November 13. This change indicated a secular variation in the orbits of the group of bodies causing the showers, and was the means of fixing the position of the orbit.

Before the observations and researches of Professor Olmsted, absolutely nothing was known of the origin and causes of these phenomena. It was not even decided whether they were of cosmical or terrestrial origin, whether they came from the planetary spaces or were caused by electricity or other agents in the atmosphere. It is a little singular that so great an explorer and lover of nature as Humboldt should have failed to decide this question by his own observations, since we now know that the data for that decision must have been plainly presented to his eyes. Careful observation would have shown him that the lines of motion of the meteors, when produced backward, all passed near the star y Leonis, and that the point of intersection seemed to follow this star as it approached the zenith, thus showing that the direction of the meteor-fall did not follow the diurnal rotation of the earth, as it would if the meteors originated in the atmosphere. But he did not appear to suspect that the phenomenon was anything more than a local one, and it was left to observers thirty-four years later to show that it was not a meteorological, but an astronomical one.

The great shower of 1833 was accounted for by Professor Olmsted on the hypothesis of a light nebulous body moving in an orbit within that of the earth, and sometimes coming into contact with the latter. He conceived it to consist of light combustible matter, which, when it entered the atmosphere, powerfully condensed the air before it, and thus elicited the heat which set it on fire. The fate of this hypothesis may serve as a warning to philosophers in forming suppositions relating to causation, to assign no cause which is not a real phenomenon, and susceptible in its own nature of being proved by other

evidence. The "nebulous body" of Professor Olmsted was as hard to account for as the meteors themselves, and he had no other evidence of its existence than that it explained the phenomena of meteoric showers. With the progress of science it has vanished entirely, and a cause has been discovered, which accounts not only for shooting-stars, but for fireballs, aerolites, and telescopic comets. It is only within the last year that the new theory has been perfected and elaborated, and until the recent publication of the works named at the head of this article, no complete and intelligible statement of it was accessible to the general reader. Such an explanation we shall now endeavor to give.

The fundamental idea of the theory is this: The planetary spaces are crowded with immense numbers of bodies, which move around the sun in all kinds of erratic orbits, and which are too minute to be seen with the most powerful telescopes.

If one of these bodies is so large and firm that it passes through the atmosphere and reaches the earth without being dissipated, we have an aerolite...

If the body is so small or so fusible as to be dissipated in the upper regions of the atmosphere, we have a shootingstar.

A crowd of such bodies sufficiently dense to be seen in the sunlight constitutes a comet.

A group less dense will be entirely invisible unless the earth happens to pass through it, when we shall have a meteoric shower.

In accordance with a proposal of Professor Newton, we shall call these bodies by the general name of "meteoroids."

Thus one simple hypothesis accounts for at least three seemingly diverse phenomena. To show this clearly, the mechanical theory of heat, with some of its attendant physical facts, must be brought to our aid. It is now established that heat is a certain form of motion, that hot air differs from cold air only in a more rapid vibration of its molecules, and that it communicates its heat to solid bodies simply by striking them with its molecules. If, then, a body moves rapidly through the air, the mere impact of the aerial molecules ought to warm it just as

hot air would. This result of theory has been proved correct by the researches of Professor William Thomson and others. A thermometer being placed in front of a rapidly moving body rose one degree when the body moved through the air at the rate of one hundred and twenty-five feet per second, and with higher velocities the rise was as the square of the velocity, so that a velocity of 250 feet produced a rise of 4°; of 375 feet, 9°; of 500 feet, 16°; and so on.

The earth moves in its orbit at the rate of 98,000 feet per second, which is the velocity with which the air would strike a body at rest in the planetary spaces. This velocity would produce a rise of temperature of 600,000 degrees. Such a body would therefore be suddenly exposed to a temperature far above any the chemist can produce by the most powerful agents. If, as will commonly be the case, the meteoroid is moving to meet the earth, the relative velocity, and therefore the temperature, will be yet higher. The November meteors, for instance, strike the atmosphere with a relative velocity of forty-four miles per second, which corresponds to a temperature of three million degrees Fahrenheit! Exposed to such a temperature, neither great size nor combustibility are necessary to account for both the brilliancy and brevity of their course. In fact, Professor Harkness, of the United States Naval Observatory, calculates that, if we suppose the ratio of light to heat to be the same as in the Drummond light, a meteoroid weighing but a single grain would give light enough to shine like a star of the first magnitude at the distance of one hundred and twenty miles.

We have alluded to the fact, that in a meteoric shower, if the paths of the individual meteors are produced backward, they are all found to pass through the same point of the heavens. This is called the "radiant point." The radiant point of the November meteors is in the constellation Leo, that of the August ones in Perseus. It appears in the same position wherever the observer is situated, and it does not partake of the diurnal motion of the earth. These two facts prove the theory that meteoric showers are caused by the earth encountering a group of particles moving independently in the planetary spaces. The meteors really move in parallel straight lines,

like drops of rain in a shower, and the radiant point is simply an effect of perspective, which makes these lines appear to converge toward a vanishing point, like the streets of a city in a perspective view. The best visible illustration of this appearance will perhaps be afforded by watching light flakes of snow fall during a calm. The flakes which are falling directly toward the observer do not seem to move at all. The surrounding flakes seem gradually to separate from these on all sides; those which are going to fall to the left seeming to move toward the left, and so with those which will fall toward the right, the front, or the rear. So with the meteoric showers. A meteor coming directly toward the observer does not seem to move at all, and the only point in which such a meteor can be seen is itself the radiant point. The surrounding meteors, though all falling in the same absolute direction, seem to diverge on all sides like the snow-flakes. If two other observers are situated at a considerable distance on either side of the first, a meteor falling directly toward the latter will to the lefthand observer seem to move off to the right, and to the righthand observer off to the left, so that the two observers see the same meteor moving in apparently opposite directions.

The radiant point being that of the direction of the meteorfall, it appears from actual observation that, when the earth arrives at a certain point of its orbit, we see an unusual number of meteors, falling in a direction which has never sensibly varied for at least a third of a century. This general fact proves our general proposition respecting the cause of meteoric showers in a manner both direct and indisputable.

The evidence that the sporadic shooting-stars, visible during every clear night, are caused by small bodies encountering the earth in its orbital motion is not of the same direct character, because these shooting-stars exhibit no definite radiant point. Still, the fact admits of no rational doubt. The appearance of sporadic meteors, and of those which fall in showers, are so exactly similar, that we cannot avoid attributing them to the same cause. Moreover, the existence of immense swarms of minute bodies moving in definite orbits through space being proved, it is highly probable a priori that many such bodies would be scattered at random.