One fact became apparent from these investigations-namely, that the meteorites move in space at a speed which does not belong to planets, but is only attained by comets. The Milan astronomer, Schiaparelli, noticed at once this peculiarity, and by the end of 1866 he succeeded at last in calculating the orbit of one ring of meteorites -namely, the Perseids. He found that it is not exactly a ring, but has the shape of a cometary orbit. It is a very elongated ellipsewhich nearly touches the orbit of the Earth at its point which is nearest to the Sun (the perihelion), while its outer end (the aphelion) lies very far beyond the orbit of Neptune. And this orbit proved to be almost identical in all its elements with the orbit of the second comet of the year 1862. In four memorable letters addressed to the director of the Rome Observatory, Secchi, Schiaparelli brought into evidence the analogies which exist between comets and shooting stars, and he ventured to suggest that the swarms of meteorites must haveformerly been comets which, owing to the unequal attraction exercised upon their different parts by the Sun and the major planetsespecially by Jupiter and Saturn-lost their original globular form, and gradually were transformed into serpent-like agglomerations of meteorites which continue to move along the orbits of their parent comets. Within a few months the brilliant hypothesis of Schiaparelli found a new confirmation in the fact that the orbit of the November Leonids, recalculated by Leverrier, by himself, and independently by Professor J. C. Adams, proved to be identical with the orbit of Tempel's comet of 1866; and next month the April meteorites were identified by Dr. Galle of Breslau with the comet of 1861. The orbit of both the Leonids and the Tempel's comet was also found to be a very elongated ellipse, of which the perihelion lies slightly within the orbit of the Earth, while its aphelion lies beyond the path of Uranus; and with regard to this orbit Leverrier made the striking remark that the comet and the swarm of the Leonids must have taken to their present orbit only in the year 126 of our era, when they passed near to Uranus and were deflected by it from their previous route."

Quite a revelation was contained in this discovery of the commonorigin of comets and shooting stars, which was further confirmed when a grand shower of shooting stars took place quite unexpectedly on

The last, i.e. the fourth, letter of Schiaparelli was published on the 31st of December 1866, and was immediately translated in the French scientific weeklies. Leverrier brought before the Paris Academy of Sciences nearly the same theory on the 21st of January 1867. In his fourth letter Schiaparelli gave also a first approximate calculation of the orbit of the November Leonids. When Leverrier, in this com. munication, and Schiaparelli on the 2nd of February 1867, had recalculated it, it appeared to be identical with the orbit of Tempel's comet of 1866. In the meantime Professor J. C. Adams, taking up the work at the point to which it had been brought up by Professor Hubert Newton (who gave five possible periods), calculated the same orbit independently, and communicated its elements to the Cambridge Philosophical Society in March 1867.

the 27th of November 1872, just as the Earth was passing very near to the orbit of the Biela comet. Besides, Schiaparelli gave a most ingenious and complete explanation of certain facts which formerly stood in the way of a general acceptance of a cosmical theory of the shooting stars. At the present time, since several other meteorite swarms have been identified with comets, and the comets themselves have been studied in more detail, the theory of the cometary origin of the shooting stars stands upon a strong scientific basis.

5

It

Of all comets, the Biela comet offers the greatest interest in this connection. It was discovered by Biela in 1826, when it was found that the same comet must have appeared in 1772 and 1805, and perhaps also, as Berberich suggests, in 1639 and 1657.6 revolves round the sun in 6 years, and in consequence of its frequent journeys amidst the planets it is only the more exposed to have its orbit hampered and its inner structure affected by Jupiter and Saturn. Besides, its orbit crosses the orbit of the Earth so closely that on the 29th of October 1832 the two were only separated by a distance of 2 terrestrial diameters the earth being, however, on that day, within a 31 days' march (i.e. nearly 57,000,000 E. miles) from the crossing of the two highways. It is known that something extraordinary happened to the Biela comet during its appearance in 1845. Under the very eyes of the astronomers it divided into two comets, one of which lagged so much behind the other that when the front one had reached the perihelion, the other was already two hours-i.e. something like 160,000 E. miles behind. Then both were lost sight of, and when they returned in 1852 the distance between the two was still greater, reaching already nearly 1,300,000 E. miles. After that they were seen no more. No amount of calculation and searching with the aid of telescopes could detect them, until as already mentioned-on the 27th of November 1872, when the Earth was nearly crossing the orbit of the Biela comet, a most brilliant display of countless shooting stars, often moving in hundreds in an indecisive, pulsating way, and of numbers of comet-like rockets was seen instead from six o'clock in the evening till midnight. Most of them seemed to come as they ought to, if they belonged to the

Dr. Schmidt at Athens, and Coulvier-Gravier at Paris, had been for many years most careful observers of shooting stars and fire-balls, and they had shown that there exists a diurnal and an annual periodicity in the numbers of shooting stars, which periodicity was invoked to prove the existence of a certain connection between them and the Earth, and consequently seemed to support the theory of their atmospheric origin. Schiaparelli most ingeniously demonstrated that if we see more shooting stars after midnight than in the earlier hours of the night, in the second half of the year than in its first half, and in the eastern parts of our horizon than in its western parts, this simply depends upon the position of the Earth during its movement round the Sun.

• Dr. Berberich, in Naturwissenschaftliche Rundschau, 1898, xiii. p. 601.

The period of its revolution varied from 6 years 220 days to 6 years 281 days, ccording to the perturbations it experiences on its way.

comet-from the constellation of Andromeda, and there could be no doubt that these tiny bodies followed in their myriads the path that the Biela comet used formerly to follow in space. So great an authority as Dr. Klinkerfues was even sure that the Earth had passed through the comet itself.

Since 1872 the orbit of this swarm of Andromedides, or the Bielid swarm, as it is often spoken of owing to its origin, has been slightly altered. In 1889 and 1890 Jupiter passed very near to it, and, owing to the powerful attraction of the great planet, the orbit of the meteorites was altered in such a way that the Earth crossed it in 1892 nearly three and a-half days before the date, i.e. on the 23rd of November; and it is on that day again, or on the 24th, that the Earth will cross it once more this year-without the terrible consequences that have been foreseen by imaginative minds.

The Biela comet is not the only one which has had the same fate. Suffice it to observe that the great comet of 1882, which passed very near to the Sun, has also subdivided into four distinct fragments, which took independent routes and went so far apart that the revolutions of the two extreme fragments will now differ from each other by nearly 300 years. The Brooks comet of 1889 also was accompanied by four separate fragments, formerly enveloped in one nebulosity, and when this comet reappeared seven years later, the four fragments were seen no more; while in the fourth comet of 1893 the tail divided into several portions, one of which became a separate comet. In short, the subdivision of comets is a not uncommon occurrence, and it may be said that we can actually see the transformation of comets into swarms of meteorites.

III

Humboldt laid stress in his Cosmos upon the fact that the aëroliths are the only medium through which we, the inhabitants of the Earth, are brought into a direct intercourse with interplanetary space. All that we know of this space is only based upon light-vibrations which strike our eye; while the aëroliths bring upon the Earth the actual matter which circulates in the space between the planets. Now, since relationship has been established between meteorites and comets, we are entitled to say that the meteorites bring down upon our globe the matter which circulates in interstellar space, as they come from the spheres which lie far beyond the utmost limits of our solar system.

Clusters of those little bodies, out of which both meteorite swarms and comets are composed perhaps, vapours which suddenly pass

They will be, according to the calculations of Kreutz, 670, 770, 880, and 960 years respectively.

from the gaseous state into the solid state, as Daubrée was inclined to think--circulate in the infinite space in which the Sun, with all the planets attached to it, is moving. When such clusters meet our solar system in their wanderings, they enter it in virtue of the attraction exercised upon them by the Sun, and they describe round our luminary a parabolic curve which carries them away, after this short visit, back to the unfathomable interstellar regions. We take notice of them during this short passage, and as the cluster approaches the Sun, and while it flies round it at a tremendous speed, and becomes luminous in this part of its course, we catch a glimpse of it, either in the shape of a small nebulosity which is only visible through the telescope, or under the aspect of an elegant, tailed comet in which men see the announcement of coming misfortunes. According to the calculations of a Russian astronomer, J. Kleiber, no fewer than 240 comets enter every year within the boundaries of our solar system. Most of them, after having described a curve round the Sun, never return to it; but some of them pass near enough to one of the outer planets to have their speed slackened and their orbit changed into an elliptical one. In such cases they become permanent members of our system, and will periodically make their appearance on our sky. The laborious researches of Tisserand, 10 continued by Schulhof, as well as the classical work of Callandreau,12 and the investigations of Professor H. Newton, 13 have now familiarised us with that capture' of comets by the great planets. Astronomers even attempt to classify the comets according to the times of their 'capture.'

11

However, a comet, once it has become a member of our system, will not remain what it was. It is bound to disaggregate. As has already been mentioned, the unequal force with which the attraction of the Sun and the planets which it passes by acts upon the different parts of the loose aggregation tears it to pieces.1 And once the disaggregation has begun, it is bound, as Callandreau has shown, to proceed at an always accelerated rate. Several comets which formerly were considered as independent bodies have already been shown to be mere fragments of larger comets. Others have already been transformed into rings of meteorites; and the older the ring, the more equally the meteorites are distributed along its whole length. Such is the case with the August Perseids, while in

• Astronomische Nachrichten, 1892, No. 3104. No fewer than 5,900 comets, he added, must already circulate round the sun within the orbit of Neptune.

10 Bulletin Astronomique, June and July 1889; Revue Générale des Sciences, 1889, i. p. 68; Richard A. Gregory, in Nature, 1890, vol. xlii. p. 31.

"Bulletin Astronomique, November and December 1889, and April, May, and

June 1891.

12 Annales de l'Observatoire de Paris, 1891, t. xx.; Nature, 1891, vol. xliii. p. 474. 13 American Journal of Science, September and December 1891, vol. xlii. 14 The general reader will find an excellent article on the disaggregation of comets, by Dr. Berberich, in Naturwissenschaftliche Rundschau, 1893, vol. viii. p. 221.

the Leonids we still see those dense clusters which the Earth pierces each thirty-three years, and which remind us still of the parent

comet.

The Leonid ring, in its turn, will not remain what it is now. It is disaggregating under our very eyes. Already during the passage of the great cluster of 1866 through the node where its orbit meets the orbit of the Earth, it could be concluded from the numbers of meteorites seen in the two subsequent years that the Earth met with more than one single compact cluster.15 It appears now, from a careful discussion by A. S. Herschel of the appearance of the Leonid swarm in the years 1894-6, that the main swarm must be accompanied by two much longer, although less dense, swarms containing 'sensible sparse gaps,' one of which precedes and the other follows the main agglomeration. Besides, there are a number of erratic shooting stars which seem to indicate the existence of branch streams accompanying the main one.16 Further detailed investigation by G. Johnstone Stoney and A. M. Downing shows also that the main cluster of the Leonids (to which the name of Ortho-leonids has been given) is a serpent-like cluster of such a length that it takes three years to pass through the perihelion of its orbit, but so narrow at the same time that the Earth pierces it in a few hours. But besides this stream, which remains true to its parent orbit, there are myriads of other Leonids (which were named Clino-leonids') whose orbits already differ to a noticeable extent from the former. There are already side streams accompanying the main one, which itself, in its turn, has subdivided into two distinct portions. One of them-the part A-was pierced by the Earth in 1866, while the other part, B, is the one which we have met this year. Besides, as it results from the laborious researches of W. F. Denning upon the displacement of radiant points during each great shower, the Earth herself must deviate the meteorites from their regular paths—so much so that part of the cluster A, deviated by the Earth in 1866, must have become now Clino-leonids,' moving independently from the main stream.18 The ring of meteorites has thus its own life and undergoes continually a series of destructive changes.

The displacement of the radiants which thus indicates the existence of a considerable number of side streams of meteorites accompanying the main ones gives fresh support to the idea that some inner forces must have been at work in the parent comet before it was transformed into a meteorite ring. In fact a comet,

15 British Association Reports for 1868 and 1869.

16 A. S. Herschel in Nature, the 24th of December 1896, vol. Iv. p. 174.

17 G. Johnstone Stoney and A. M. W. Downing, 'Perturbation of the Leonids,' in Proceedings of the Royal Society, the 2nd of March 1899; Nature, the 23rd of March 1899, vol. lix. p. 497; also vol. Ixi. p. 28.

19 Astronomische Nachrichten, 1898, No. 3513, where a summary of the positions of the radiants of smaller showers is given; and a series of articles on Meteorites in Nature.