After having pointed out this new and important coincidence, M. Schiaparelli observes: 'It is very worthy of remark that the two well-known meteor streams, those of August and November, have each their comet. Are we to suppose that it is the same with all the others? If so, we should be forced to regard these cosmical streams as the result of the dissolution of cometary bodies. But it would be at least premature to extend this conclusion to all shooting stars. It is possible, as I have shown, that the whole of these bodies, great and small, may form systems in space bound together solely by their own attraction, and afterwards destroyed by the action of the sun. Perhaps, also, that which we term a comet is not a single body, but a collection of very numerous and minute bodies, attached to a principal nucleus.'

We cannot fail to notice the connexion existing between these views and those which led M. Hoek to his study of the theory of cometary systems; we must also perceive how completely these new views on the subject of cometary physics accord with the facts of observation that have been mentioned in a preceding section respecting the duplication of Biela's comet and the division and shattering of ancient comets, phenomena which have been handed down to us by tradition, but which

had hitherto been generally denied and regarded as fables by

astronomers.

In conclusion, and before leaving so vast a domain, open alike to new researches and conjectures, we must not forget to mention two more cases of identity between meteoric swarms and comets. The first relates to the meteors of April 20. According to MM. Galle and Weiss the orbit of this swarm has the same elements as the orbit of the comet of 1861. D'Arrest, and Weiss have likewise found an accordance between Biela's comet and the shooting stars of the end of November and the first days of December. We have already said that the remarkable shower of shooting stars which distinguished the night of November 27, 1872, appears certainly to have been due to the rencontre of the earth either with one. of the two comets, fragments of that of Biela, or with a stream of matter which originally belonged to that comet, and which followed in space nearly the same course.*

If these views-which would have appeared so strange half a century ago-should be confirmed, we have a new and quite unexpected means of putting ourselves in direct communication with comets, since the earth every year-every night in the year, indeed-comes in contact with nebulosities which have been comets. A new light would be thrown upon the physical constitution of these bodies, and we might then consider as highly probable that granulated structure of cometary nuclei, formed of isolated particles, which Babinet was led to suspect upon very different grounds.

*[See note, p. 265.-ED.]

ON THE CONNEXION BETWEEN COMETS AND

METEORS.

BY THE EDITOR.

The intimate connexion now known to exist between comets and meteors is perhaps the most striking and novel discovery of a purely astronomical kind that has been made in our time. To those who are aware how few years have elapsed since the apparitions and tracks of meteors seemed to be so arbitrary and capricious that in the opinion of many it was scarcely worth while to record them, it cannot but be matter of wonder to consider how great has been the advance in our knowledge, and how rapid has been the progress of ideas on this subject. On account of the importance of the results found upon the nature of comets, I, therefore, add here several details, chiefly historical, which will serve to show more fully how remarkable is the connexion that has been established.

It is less than ten years since the orbit of the first stream of shooting stars-that of the middle of November--was calcu lated, the circumstances being as follows:

Professor H. A. Newton, of the United States, collected and discussed thirteen historic showers of the November meteors between the years 902 and 1833.

The following table exhibits these displays, and the earth's longitude at each date, together with the same particulars for the shower of 1866

From these data Professor Newton inferred that these displays recur in cycles of 331 years, and that during a period of two or three years at the end of each cycle a meteoric shower may be expected. He also concluded that the November meteors belong to a system of small bodies describing an elliptic orbit about the sun, and extending in the form of a stream along an arc of that orbit which is of such a length that the whole stream occupies about one-tenth or one-fifteenth of the periodic time in passing any particular point. He further showed that the periodic time must be 180 days, 185 days, 355 days, 377 days, or 334 years, and suggested that by calculating the secular motion of the node for each one of the five possible orbits, and by comparing the values with the observed motion (about 52" annually or 29' in 33 years) it would be possible to decide which of these five orbits was the correct one.

Soon after the remarkable display of the November meteors in 1866, Professor Adams, of Cambridge, undertook the examination of this question. Beginning with the orbit of 355 days, which Professor Newton considered to be the most probable one, Professor Adams found the motion of the node would only amount to 12' in 33 years; that for the orbit of

377 days the value would be nearly the same, while if the periodic time were a little greater or a little less than half a year, the motion of the node would be still smaller. It therefore only remained to examine the orbit of 33 years, and Professor Adams found that, for this orbit, the longitude of the node would be increased 20' by the action of Jupiter, nearly 7' by the action of Saturn, and about 1' by the action of Uranus. The other planets produce scarcely any sensible effects, so that the entire calculated increase of the longitude of the node in the above-mentioned period is about 29'. As already stated, the observed increase of longitude in the same time is 29', and this remarkable accordance between the results of theory and observation left no doubt as to the correctness of the period of 331 years.*

Subsequently, however, to the commencement, but before the publication of Professor Adams's results, M. Schiaparelli had been led, on totally different grounds, to conclusions which first suggested a probable connexion between meteors and comets. These related to the August meteors, or Perseids as they are called from the constellation which usually contains their radiant point. A comparison of the average hourly increase in the frequency of meteors, throughout the year, from evening until daybreak, with a mathematical formula for the same variation in terms of their velocity, led M. Schiaparelli to conclude that the real average velocity of shooting-stars in their orbits round the sun did not differ much from that of comets moving in parabolic orbits, which is greater than the earth's mean orbital velocity at the same distance from the sun in the proportion of 1414 to 1. Discussing, then, the origin of meteoric currents M. Schiaparelli remarked that in all respects shower-meteors resembled comets rather than planetary

*

Monthly Notices of the Roy. Ast. Soc., vol. xxvii., p. 250 (April 1867).