PREFACE TO THE THIRD EDITION

ALTHOUGH in outward shape this third edition exactly resembles the second, it will prove I trust in essential points to be more in advance of the second, than the second was of the first edition. Partly for other obvious reasons, and partly because the second impression was larger than the first and has now been exhausted for nearly a year, the interval given me for reflexion has been much longer; and as the author is now so familiar to me, I have been able to bring all my reading to bear upon the improvement of the text and the commentary. The critical notes are I believe improved in many important points, though their bulk of course has not been increased: compression was needed here rather than expansion. The explanatory notes however have been enlarged by the substance of at least fifty pages, chiefly through a somewhat closer printing. Nor does this at all represent the real amount of change, as many of the longer notes have been entirely rewritten and much that was superfluous or erroneous has been cut out. Many thousand fresh illustrations have been added.

For more precise conceptions on some points of the poet's philosophy, especially the motion of his atoms, I have been greatly indebted to the works of Professor Clerk Maxwell and Professor Tyndall, and to a thorough and excellent article in the 48th volume of the North British Review on 'the atomic theory of Lucretius': of Martha's brilliant work I have spoken elsewhere. For the general criticism

X

of

my author I owe much to the well-pondered remarks of Mr N. P. Howard, whose letter to me I have printed in the first number of the new Journal of Philology; and, especially in the third and fourth books, to the communications of my friend Professor J. E. B. Mayor, to whose notes I have appended his initials.

In the 25th and 26th volumes of the Philologus there is a long 'Jahresbericht' by Mr Fried. Polle on the Lucretius literature after Lachmann and Bernays; and some remarks of his occur in Jahn's Jahrbuecher. He hardly touches on the interpretation or philosophy of the poet nor have I been able to adopt any of his own conjectures which are not very numerous. The most valuable hint I have got from him is on v 312, though my own correction is very different from his. In several volumes of the same Philologus appear very prolix notes on the earlier books by Mr Susemihl and Mr Brieger. The former confines himself chiefly to rearranging paragraphs and to proposing numerous transpositions of verses, in neither I think with much success. Many of his new arrangements of paragraphs are I assert demonstrably wrong; and his violent transpositions would lead to the wildest confusion. Once however I have obeyed him in not making a new paragraph of IV 168-175: it was an accident that this was not done before, as my attention was absorbed in refuting Lachmann's errors there. Mr Brieger, who is the more combative of the two, indulges mainly in conjectural alterations of the text. Once or twice I have referred to Mr Holtze's 'Syntaxis Lucretianae lineamenta.'

On the whole my criticism is now I believe more conservative than it was. Again and again I have found that, seduced by the learning of Lachmann, I have followed him in changes which really corrupt the author. This must hold then in many other cases as well. If the text of Virgil rested, like that of Lucretius, on a single manuscript, how much there is in him we should refuse to accept as Latin! This 'must give

xi

us pause.' Yet I have not sinned I think in defending the indefensible. It is probable however that, if I should ever issue another edition, I should leave as manifestly corrupt some passages which defy anything like certain or even very specious correction.

My lamented friend Professor Conington published a lecture on the style of Lucretius and Catullus which has been reprinted among his Miscellaneous works. This lecture, written in a tone of the kindest courtesy, is for the most part a criticism of a single paragraph in the introduction to my explanatory notes; and, so far as Catullus is concerned, almost of a single sentence. This paragraph I have now omitted: justice could be done neither to him nor to myself within the limits that could be permitted here. If I should ever venture on any reply, some other place and opportunity must be found for it.

TRINITY COLLEGE, APRIL 1873.

PREFACE TO THE FOURTH EDITION.

THE present edition is divided into three volumes, the first containing the text and critical notes, the second the commentary and general index, the third the translation.

The text is practically the same as in the last edition: in one place only (1 442) a new reading is printed, and that was only omitted before through accident as was explained at the end of the index. There are a few additions, not more than half-a-dozen, to the critical notes; these have been inserted in square brackets. The translation too has undergone no change.

In the commentary there are few alterations but considerable additions amounting in all to more than twenty pages. These supplementary notes and illustrations have been taken from an interleaved copy of the last edition which was found among Mr Munro's books after his death in the spring of last year. Also on p. 333 of the commentary will be found some extracts from letters addressed by him to Professor Palmer of Trinity College, Dublin, discussing the reading of v 1010; Professor Palmer was kind enough to send me these letters. As the extracts deal with the criticism rather than the explanation of the passage, they should have been inserted in the first volume; but by an accident they were delayed until after that volume was printed. Nothing has been inserted from any other source whatever. Here too all that is new has been inclosed in square brackets. In the new examples the references have been verified. To the index large additions have been made by myself.

It remains to express my thanks to Dr Forbes, the relative and executor of Mr Munro, for the kindly feeling which led him to entrust me with the preparation of this edition, and also for the generous consideration he has shown throughout.

TRINITY COLLEGE, CAMBRIDGE,

FEB. 18, 1886.

J. D. DUFF.

LUCRETIUS.

NOTES I

ON THE FORMATION OF THE TEXT

IF Lucretius had come down to us with a text as uninjured as that of Virgil and a few other ancient writers, he could scarcely have been reckoned among the most difficult Latin poets. Certainly he would have been more easy to explain than Virgil for instance or Horace; for he tells what he has to tell simply and directly, and among his poetical merits is not included that of leaving his reader to guess which of many possible meanings was the one he intended to convey. Fortune however has not dealt so kindly with him. Not that the great mass of his poem is not in a sound and satisfactory state: in this respect he is better off than many others; but owing to the way in which it has been handed down, his text has suffered in some portions irreparable loss. It is now universally admitted that every existing copy of the poem has come from one original, which has itself long disappeared.

Of existing manuscripts a fuller account will presently be given: let it suffice for the moment to say that the two which Lachmann has mainly followed and which every future editor must follow, are now in the library of Leyden. One is a folio written in the ninth century, the other a quarto certainly not later than the tenth. Large fragments of one, if not of two others, of the same age as the quarto and very closely resembling it are also still preserved, partly in Copenhagen, partly in Vienna. These manuscripts and at least one more must have lain for centuries in the monasteries of France or Germany, where they found at different periods several correctors, more or less competent. It is to be presumed then that they had some readers, though few if any traces of them are to be met with in the voluminous literature of the middle ages. In my previous editions I said that my friend Professor Mayor had given me a reference to Honorius of Autun in the bibliotheca maxima patrum xx p. 1001, who is there made to quote II 888 in this way, Ex insensilibus me credas sensile gigni, the context proving that he meant to say ne, not me; and asked whether this writer who flourished in the