the assumed and certainly unexaggerated estimate of the probability in question) is exactly parallel.

Now, consider in connection with the facts here dealt with, the observed facts that

First, the earth encounters two meteor systems which are certainly very eccentric, very widely extended, and considerably inclined -after the manner of cometic orbits.

And secondly, the earth encounters more than a hundred meteor systems, which are probably no less cometic in all their characteristics, and many of which are certainly cometic as respects one characteristic, being inclined considerably to the plane of the earth's

orbit.

The legitimate conclusion from the estimated antecedent improbability that a meteor system, supposed to be placed and shaped at random, would cross the earth's orbit, and the observed facts, is beyond all question this, that the meteor systems actually encountered by the earth form but a minnte proportion of the total number of meteor systems actually belonging to the solar domain. There is absolutely but one way in which this conclusion could be avoided, and that way, as we shall see, is barred. If we had reason to suppose that our earth had the power of forcing meteor systems to take up a position intersecting her orbit, we should no longer be compelled to believe that for each such system millions of others exist within the solar scheme. For in the course of long past æons, a large proportion of the meteor systems might have been swayed into a partial subjection to the earth. If the earth were as Jupiter, for instance, if she surpassed in bulk and weight the combined mass of all the planets, we might suppose that to her own attractive energies alone the fact was due that so many meteor systems cross her path. Jupiter has

swayed, and still continues to sway, comets in this way; and doubtless he has swayed, and still continues to sway, meteor systems into partial submission. But the earth's influence over meteor systems is relatively all but evanescent. She has had no part or share in swaying to their present position either of those two systems whose paths have been determined. Indeed Leverrier has shown that the November system has been swayed into its present course by the planet Uranus. Most probably too-or rather, almost certainly-not one of all the meteor systems crossed by the earth has been brought appreciably nearer to her orbit by her own influence.

But as a commentary on these relations, let us consider how the two meteor systems whose orbits have been determined stand related to other planets besides the earth. There is the August system, as an illustration of a system taken at random. It does not pass near the orbit of any known planet of the solar system save the earth alone. Jupiter, mighty as he is, has had no power to bring the August system close to his orbit; Saturn, Uranus, and Neptune have been equally ineffective. If these planets have inhabitants, the August meteor system is utterly unknown to them. The November meteor system, again, does not pass within millions on millions of miles of the orbits of any of the planets, save only the earth and Uranus.

Two, then, of the known meteoric systems-and two taken at random so far as the question we are upon is concerned-are so placed that their existence must remain wholly unknown to inhabitants of all the other planets, except (as respects one system) the planet Uranus. not the conclusion legitimate, or rather-for there is no real question of the justice of the conclusion

Is

is it not rendered perfectly clear, that an enormous proportion of the meteor systems recognised from

other planets are so situated that we can never become cognisant of their existence? A large proportion of the actually existent meteor systems also must be so situated that they cannot be recognised from any of the planets at all.

Here then we have, as a direct and legitimate conclusion from admitted facts, a view of the solar system which four years ago would have been justly regarded as too startling for belief. We see the vast gaps which separate planet from planet no longer untenanted, or only traversed by an occasional comet, but literally crowded with meteor systems. If a vast model of the solar system could be constructed, all the parts being justly proportioned on some such scale as is considered in the well-known description given in Herschel's Outlines of Astronomy, and if the orbit of every meteor system were represented in this model by an oval hoop made of the finest possible wire, justly placed and shaped, then the space around the sun to a distance far exceeding the radius of the earth's orbit, would be an absolute network of these wire orbits. Nor would it be easy to say how far from the two-feet globe representing the sun some of the closed orbits would extend; while myriads of parabolic and hyperbolic wires would have to be introduced to indicate the paths of those meteoric clusters which assuredly approach from outer space and return to the star-depths, never again to visit our sun's neighbourhood. Amidst these millions of orbitwires the paths of the planets, if similarly indicated, would be almost lost; though individually the least of the planets probably surpasses the combined mass of all the members of those meteor systems which belong specially to the solar domain.

It would, indeed, be difficult to determine, even approximately, the weight either of any meteor system

or of the system of such systems circling around the sun. We know that among those meteoric masses which actually reach the earth there are some few of great weight. One at least has fallen whose weight amounts to about fifteen tons. Nor again can we regard the masses which are found in the earth after the explosion of a bolide as more than the fragments of much larger masses. But on the other hand, the bodies which form such systems as the Leonides and the Perseides are for the most part exceedingly minute, insomuch that the weight of some of these bodies has been estimated at less than a single grain.

Between these limits lie the meteors which explode in the upper regions of air, but without casting their fragments to the earth's surface in a solid form.

It is worthy of notice, however, that even if we set on one side meteoric displays properly so called and the great aerolites which from time to time crash down upon the earth, we yet find abundant reason for believing that our earth alone, small as she is, grows yearly in weight by many tons through the downfall of meteoric matter. Professor Newton has calculated from perfectly reliable data that on an average in the course of a single day 7,500,000 meteors large enough to be visible to the naked eye are consumed in the earth's atmosphere, and about 400,000,000 meteors such as could be seen in a telescope of moderate power. Now, if we consider the latter set to be equivalent to the former, and assign a single grain as the weight of each meteor visible to the naked eye, we deduce fifteen millions of grains as the. earth's daily increase of weight. This is rather less than a ton. So that in the course of about three years the earth's weight must increase (even on the very low value here assigned to a meteor's weight) by a thousand tons; and in the

VIRG

TIRGIL can never be rightly translated into any modern tongue. 'He seems to have studied not to be translated.' So says John Dryden, and forthwith proceeds to translate him, and devotes barely three short years to the task. We shall examine in this article how far facts support this dictum; but first it is necessary to state our views on translation in general. In considering this subject three points have to be settledthe translator, his instrument, and his style.

First as to the translator. Who will best interpret the works of a great original poet? It may seem to many (and the theory is no doubt plausible) that he alone can understand and reproduce the genius of a great poet who is himself also a poet of equal genius. But, on the other hand, it must be remembered that, translation being essentially imitative, originality is apt to be a hindrance rather than a help, and the result is likely to be of a hybrid description-not literal enough for a faithful translation, too close to permit the genius of the poet to expand itself without constraint. We may further doubt whether such a one would expend his full force merely upon the task of reproducing another's thought-whether, in fact, he would comply with the conditions without which even genius can never produce a masterpiece.

Even if so much be grantedif a poet of equal ability be willing to devote his entire powers to the work, and to spare himself no time and trouble to perform it in the best possible manner-is it desirable that he should do so? Will it not be a waste of power? Are originality and genius so common amongst the sons of men that they should be diverted from their proper purpose

of production to the inferior object of simple imitation? Whatever may be thought of Pope's Homer, for instance, or of Dryden's Virgil, can anyone fail to regret that their authors did not give us instead another Rape of the Lock or Absalom and Achitophel? Surely it was their duty not to revive the splendour of the past, but to shape the thought and purify the affections of the present. To take another example. Tennyson has shown, by his remarkable translation of the passage at the end of the Eighth Book of the Iliad, that, did he so choose, he might delight the minds of this and future generations with such a rendering of Homer as the world has never yet seen. But would this be a sufficient set-off against the loss of such a poem as In Memoriam, or Enoch Arden, or Lucretius?

Mr.

To return. Suppose it be conceded that the task of translating great poets in general, and Virgil in particular, should be reserved for versifiers-using the word in its most honourable sensewe are next plunged into the vexata quæstio of metres. Now, as to this terribly moot point, What is the ideally best metre in which to translate Virgil? we feel inclined to reply, By borrowing, if we may be allowed to do so, a famous expression used for a totally different purpose- Men, not measures.' We do not deny that there may be one metre, and one metre only, absolutely the most desirable, although grammatici certant; yet, inasmuch as no two critics seem to agree as to what that particular metre may be, it seems to us that the wisest plan is to leave the choice to the individual translator. For, as Mr. Conington observes, that which may be absolutely the best is by no means necessarily the best for the individual;' and of this he must of

course be the safest judge. Moreover, the real merit or demerit of a translation would seem to be only partially affected by a judicious choice in this respect. Skill and vigour will ennoble any metre, however ill-adapted for the required purpose; but no excellence of the medium employed can possibly compensate for lack of fire, faithfulness, and general ability. How much may plausibly be urged for two wholly dissimilar metres may be gathered by anyone who reads the singularly ingenious prefaces of Professor Conington and Mr. E. F. Taylor, a young aspirant to Virgilian honours, whose actual work at all events may convince us that the respective metres selected by them were wisely chosen. Neither will the late Lord Derby's contemptuous condemnation of that 'pestilent heresy' the English hexameter so atterly exclude the question, even as to this much-abused measure, when it is remembered that language wellhigh as trenchant has been used both of the so-called ballad measure of Scott and of the Spenserian stanza; and yet, when at length both metres have been put to the actual proof, there are now not wanting many firm friends of both. It is a remark frequently made of blank verse, as a vehicle of translating Virgil, that it is such an exceptionally difficult metre that two or three men at most in a generation can attempt the task with any chance of success; but obviously this is no argument against the metre itself, but merely against the lack of skill of its employers. We entirely endorse Mr. Taylor's remark on the whole subject-solvitur ambulando-let us condemn no measure until it has found a champion who has elicited all its capabilities.

The last point deserving attention is the style and general merits which we have a right to demand of a translation. No translation will,

A cer

we apprehend, be thought deserving of praise which does not faithfully represent the original. By faithfulness we do not mean that line should be rendered by line-a merely verbal correctness-but that no important thought or felicitous turn of expression should be omitted. Remembering the inferiority even of so flexible and copious a language as our own, we require that a translator shall at least thoroughly understand the capabili ties of that language, shall have a true poet's sense of the harmony of sounds, and shall be a scholar, but one who does not find it necessary to sacrifice sound to sense. tain subtlety, too, of perception will be indispensable to apprehend the nicer shades of meaning and the delicacies of expression which perpetually astonish the student. Above all must the translator possess a lively sympathy with that most remarkable perhaps of all Virgil's characteristics, his exquisite sense of pathos, without which, whatever may be the accuracy of his scholarship or the force of his style, the translator will never be able to touch our hearts as the mast touched them. We are inclined to believe that this especial beauty has been somewhat overlooked. We hear much indeed of the inimitable grace and finish Virgil, of the marvellous condensation of his style, of the admirable variety of his pauses, of his 'long majestic march,' and so forth; but in reality merits such as these, however undeniable, do not account for that singular charm by which he fascinates the mind. It is by his genuine, artistic it may be, but certainly not artificial, pathos that he goes straight to our hearts, appeals to the strong simple emotions of our common humanity, and makes us feel how, to use the wellknown words of our greatest poet, one touch of nature makes the whole world kin.'

has

of