We now approach one of the most curious and suggestive discoveries of recent astronomy, - a discovery resulting from such a series of independent and apparently disconnected observations, that no single individual can claim the credit of making it. We shall ask leave to tell the story from the beginning.

In December, 1865, M. Tempel, an astronomer of Marseilles, discovered a faint telescopic comet. It was afterward discovered independently by Mr. H. P. Tuttle, at the Naval Observatory, Washington. It passed its perihelion in January, and, receding from the sun, vanished from sight in March. It was soon found to move in an elliptic orbit with a period of something like thirty years. The process of reducing and publishing astronomical observations is, however, so slow and laborious that generally at least a year has to elapse before the material for the definitive determination of a cometary orbit can be collected. So it was not until January, 1867, that Dr. Oppalzer of Vienna was able to compute an accurate orbit of this comet. The number of the Astronomische Nachrichten which contains the details of his computation is dated on the 28th of that month.

Let us now return to our meteors. It is well known that a considerable meteoric shower was seen in Europe on the night of November 13-14, 1866. It being settled that this shower, like that of 1833, was caused by the earth encountering a group of small bodies, moving in a different orbit, astronomers were naturally anxious to determine this orbit. But the data for this determination were insufficient until the periodic time was known. This important element the researches of Professor Newton had left in doubt. The shower recurring at intervals of thirty-three years, it might, at first sight, seem that the time of revolution must be thirty-three years. But this conclusion would be hasty, because the group might have returned several times in the course of the thirty-three years, erossing the orbit of the earth at times when the latter was not near the point of intersection. Professor Newton was led to consider a period of 1 years rather more probable than the longer. At the same time he pointed out a fact which might lead to a definite solution of the problem. We have seen that as centuries elapsed the shower occurred on a later

and later day of the year, the date being October 19 in 902 and November 13 in 1866. This indicates a progressive motion of the node amounting to fifty-four seconds in a century. Now, what must be the periodic time in order that this change may be produced by the action of the planets? Professor John C. Adams solved this question, and gave thirty-three years for the answer.

The periodic time and the radiant point being known, the data for determining the orbit were completely given, a fact which seems to have first occurred to Le Verrier. His solution was read to the French Academy of Sciences on January 21, 1867. The following are the elements of the orbit to which he was led :

Dr. Oppalzer's elements of Tempel's comet, as published in the Astronomische Nachrichten of January 28, 1867, are:

The similarity of these two sets of numbers is too striking to be the result of chance. The "inclination" is the only element which differs sensibly, and this difference was afterward found by Le Verrier himself to proceed from his having adopted an erroneous position of the radiant point in his calculations. In estimating the similarity of the orbits, it must always be remembered that the two calculators were entirely ignorant of each other's results until they saw them in print, and that the coincidence was first detected by a third person, C. W. F. Peters of Altona, who compared the printed results in the two publications.

The inference to be drawn from these facts of observation is, that the November meteors are caused by a long stream of minute bodies following Tempel's comet in its orbit. Whenever the earth passes through this stream, all the atoms that it encounters are swept away as shooting-stars. Thus, millions of the particles of the stream are destroyed, or rather added to the earth every year. But so vast is the number of these bodies that ages will be required to make any appreciable diminution of their number. By rough estimates, the details of which we need not enter upon, it is supposed that the individual meteoroids of the November stream pass a given point at the rate of millions a second. Yet several years even are required for the passage of the thicker part of the stream, its entire length at perihelion not being less than a thousand million miles.

It is a curious fact that the orbit of the stream of meteoroids not only intersects the orbit of the earth, but passes in close proximity to that of Uranus. This fact led Le Verrier to the conclusion that the stream did not originally form part of our system, but was a wanderer through the stellar spaces, until, accidentally passing in close proximity to Uranus, the attraction of that planet threw it into an elliptic orbit round the sun. He even attempted, by calculating back the revolutions of the meteoric group and of Uranus, to fix the date of this event, and gave the year 126 of the Christian era as the most probable epoch of its occurrence.

While this date is somewhat problematical, the grouping of the meteoroids renders it almost certain that they have not been revolving in their present orbit during many centuries. For if we suppose a group of bodies, ever so close, to move round and round in a re-entering orbit, the slightly different velocities of the individuals will cause them gradually to spread out into a long stream, and this spreading will continue until the stream extends around the entire orbit. This may be illustrated by a race-course. If a number of horses start together and continue to run round and round the course, the fastest horse will leave the slowest behind until he is on the opposite side of the course. Then he will approach him from behind until he overtakes him.

The horses will then be scat

tered around the entire course. In the same way, when the swiftest meteoroid overtakes the slowest one the stream will be spread over the entire orbit. As this must happen in the course of centuries, and as the great body of the November meteoroids are included in one tenth the length of the orbit, it seems pretty certain that they have not had many centuries to scatter themselves.

Soon after the discovery of the connection of the November meteoroids with Tempel's comet was made known, an Italian astronomer, M. Schiaperelli, was fortunate enough to identify the orbit of the August meteoroids with that of a telescopic comet discovered in 1862. The orbits approach as closely as that of the November meteoroids to Tempel's comet. But the August meteoroids now seem to be spread around the entire orbit, the meteors of that month being about equally numerous every year.

It thus being rendered tolerably certain that the best known streams of meteoroids are composed of particles left behind by comets, the question naturally arises, What is the relation between these comets and the meteoroids? The answer which most readily suggests itself is, That the comets are themselves. composed of meteoroids, which, it will be remembered, are simply detached particles of solid matter, moving through the celestial spaces singly or in groups. It must also be remembered that the comets of which we speak have no tail, but seem like nebulous patches so thin and airy that any attempt to shake pestilence and war from their hair would simply result in their shaking themselves to pieces. They do indeed appear to the observer to form a continuous mass of matter of extreme tenuity. But we know that clouds, steam, smoke, and other forms of matter presenting this appearance at the surface of the earth are really formed of detached particles, and the presumption is strong that the comet is formed in the same way. It is true that an elastic gas would present the same appearBut it is difficult to conceive how a body of such a gas could escape being instantly dissipated by its own elasticity; so that, among known forms of matter, that of detached particles seems best to explain the appearance of telescopic

ance.

comets.

This view is sustained by the interesting fact, observed in numerous instances, of the gradual decay of known periodic comets. The two most notable cases are those of the comets of Halley and Biela. The former returns to the sun every seventy-six years. At its apparition in 1456 it was described as a terrible object, having a tail sixty degrees in length. It inspired Pope Calixtus III. with such terror that he ordered prayers throughout Christendom against its malign influence, thus giving rise to the wide-spread fiction of "the bull against the comet." But, at its last two returns in 1759, and again in 1835, its appearance was in no way remarkable.

Biela's comet has been known for nearly a century, being first seen in 1772. But its periodicity was not recognized until 1826, when it was found to return to its perihelion every six years and a half. In 1846 it became celebrated for an appearance altogether new in the history of astronomy, being separated into two parts. In the autumn of 1852 both parts were observed for the last time, and were very faint. The next return took place when the earth was on the opposite side of the sun, so that the comet could not be seen. In 1865, when its return in close proximity to the earth was to have been expected, not a trace of it could be seen, though looked for under the most favorable circumstances by the best observers, and with good telescopes. After showing itself for eighty years, it vanished from sight like a shadow. No doubt the individual particles of the comet are still revolving in their accustomed orbit; but in the course of successive revolutions they have become so widely dispersed as to be no longer visible.

Professor Newton describes the meteoroids as being apparently the material out of which worlds are forming. Although we cannot see how a world can be formed out of these materials, we do catch glimpses of a possible process by which they are being increased. Great clouds of diffused and finely divided matter are moving in all directions in the stellar spaces. Accidentally entering our system, some of them are thrown into elliptic orbits by the attraction of a planet. Such of these as come within range of our telescopes appear as comets of long period, and become permanent members of our system. In NO. 220.

VOL. CVII.

4