We may now consider this ancient evidence that Andes lay three miles away from Mantua in connection with Conway's remaining arguments and with Virgil's ‘own statement’ in his Bucolics.

In the matter of the inscriptions, Conway's ‘impenitence’ does nothing to strengthen his case. All the points that he raises in an apparent refutation had been met by me. I had distinguished between public and private inscriptions, as Conway had not done in his earlier article, where he declared the period of the two monuments to be ‘Virgil's own,’ the inscriptions being ‘cut in the style which marks the best work between 50 B.C. and A.D. 50.’ I had not, as he asserts, failed to notice the point that caps the climax of his recent article, that ‘the P. Magius who wrote the inscription of Casalpoglio … was related to Virgil's mother.’ I was inclined to date both inscriptions, with the approval of Professor Egbert, somewhere in the third quarter of the first century of our era, but ‘even allowing for them both a date as early as the end of the first century B.C.,’ which is apparently as far back as Conway means to stretch their date, I clamored for an inscription of the time of Virgil's birth if it were to serve as evidence for his birthplace. This point is not met by Conway. If Virgil was born at Pietole, his fame and his family could certainly have extended as far north as Casalpoglio and Calvisano at the time of his death in 19 B.C. Supposing that we knew on other evidence that he was born at Calvisano, the presence of an inscription with the name of Magius or Vergilius would corroborate that evidence.

The twenty-four lines of this poem have been preserved only by the cod. Sangallensis 878 (variously assigned to saec. IX or IX-XI or XI-XII), whence it was edited in 1863 by K. SchenklSitzungsb. d. phil.-hist. Cl. d. kais. Akad. d. Wissensch. (Vienna) XLIII p. 71. A single line, the last, exists also in the cod. Bernensis 108 saec. IX. Fifteen survive in a MS of the 17th century now divided into two parts, Barberinus (or Barberinianus) XXXI 39 and Vaticanus 9135, the former containing the hexameters 3, 5, 15, 17, 19, 21, 23 and a spurious hexameter in lieu of 1, the latter the pentameters 2, 4,14,16,18, 20, 22, 24. The eight consecutive lines 6-13 are therefore missing from both ; and in this imperfect and interpolated form the poem was edited by Mommsen with Haupt's assistance inC.I.L. I p. 411.

This second visit to the place of Virgil's birth was made partly in actuality—for my wife and I, before taking part in the Virgilian Cruise of last summer, spent two delightful days at Pietole with our hosts, the Signori Prati, and our guest and friend Bruno Nardi—and partly in a renewed pondering of the arguments presented by my friend Professor Conway both in his earlier article and in his recent review of the question, to which, as he says, I had urged him to return. I promised him at the time that if he should not speak the last word on the subject, I would still further defend the view commonly accepted until he bestowed an extraordinary publicity on Calvisano and Carpenedolo. He declares that I maintain the traditional site at Pietole, ‘though not perhaps with very great confidence.’ He further implies that I ‘do not want to accept the evidence of Probus because “I prefer” to believe a mediaeval tradition.’ Let me assure him and the reader that I do not regard a mediaeval tradition per se as better proof than the certain statement of an ancient authority. I have been led, by studies in various fields, to respect tradition in general until it is disproved, and to lay the burden of the argument on those who would disprove it. But the mere sight of something hoary and mediaeval does not prompt me to exclaim, ‘Media Aetas locuta est; causa finita est’ I relish the attempts of an iconoclast to destroy rigid error, and accept his destruction if it is accomplished. In the present case, however, I have ‘very great confidence’ that Conway's assault on the tradition has come to naught. Possibly some new and unexpected evidence may yet be discovered, showing that Virgil was born at Calvisano, or Carpenedolo, or at some other site than Pietole—but nobody has presented it yet.

In the second book of his Epitoma rei militaris Vegetius sets himself the task of describing the organization of the antiqua legio of the Roman Army, the units into which it was divided, its officers, the arms of its soldiers, and its tactical employment on the field of battle. Interspersed in this account are frequent references to changes that had been subsequently effected and were in operation in the author's lifetime. But although these annotations destroy the synthesis of the book, and the transition from past to present is not always clearly indicated, the disentanglement of Vegetius' own additions from the material provided by his source or sources is not the chief difficulty that confronts the modern historian. For with few exceptions in chapters 4–14 Vegetius describes the antiqua legio in the imperfect, the legion of his own time in the present indicative, with the not infrequent addition of the word ‘nunc’ or ‘hodie.’ The great problem is to discover whether in his account of the antiqua legio the author is referring to any one period of Roman History, and, if so, whether its limits can be exactly defined. Many solutions have been attempted. Some scholars have thought that the period is earlier than Hadrian; the more generally accepted theory is that the antiqua legio can be identified with the legion as it was constituted in the reign of Diocletian. Recently two German historians have subjected the evidence to fresh consideration, and, while one of them agrees with the majority in selecting the age of Diocletian for his answer to the problem, the other believes that he can date the period to which Vegetius is referring to the late second century A.D.

It is important to ascertain whether in the Actium campaign Antony really had only the small proportion of Italian legionaries sometimes assigned to him, and this can only be done by analyzing his legionary figures from Philippi onwards. Unfortunately, though Appian is usually quite clear, he omits one figure at the very start, which prevents one treating the matter synthetically; so one must analyze backwards from Antony's legionary coins.

It may be affirmed with some confidence that on this topic no generally accepted solution will be found in default of new evidence, for which we can only faintly hope. Against certainty on the matter it would seem that the Everlasting has fixed his canon: quis iustius induit arma scire nefas. Dogmatism is out of place; we must be content with whatever theory is least difficult to reconcile with the texts and with a reasonable interpretation of the course of events at the time and the comments on them of contemporary observers. The thesis advanced in this paper is that there are strong reasons for supposing that the Lex Pompeia Licinia contained a date by which Caesar's command ended, that this date was not the end of February of the year 50 or the year 49 or the end of December of the year 50, that it did not contain a clause forbidding the discussion of a successor to Caesar before March 1, 50, and that if the date ending his command lies between the end of February and the end of December of the year 50, it may have been in fact the Ides of November in that year.

The extent of the dependence of early Greek cosmogony on mythical conceptions has long been a prolific source of controversy. Views on the subject have varied from Professor Cornford's claim that ‘there is a real continuity between the earliest rational speculation and the religious representation that lay behind it’ to Professor Burnet's extreme statement, ‘it is quite wrong to look for the origins of Ionian science in mythological ideas of any kind.’ The solution of the problem that I wish to suggest is one which should satisfy those who insist on ‘the scientific character of the early Greek cosmology,’ while retaining a direct connection with pre-scientific beliefs—namely, that some, at least, of the earliest philosophers founded their doctrines of the beginning of the universe on a deliberate rationalization of earlier and contemporary mythical ideas. The appearance of notions like the ‘world-egg’ in a number of passages, too well known for quotation, shows that primitive cosmogony assumed an analogy between the generation of the world and the generation of animate creatures. I believe it can be proved that, so far from the ‘renunciation of sexual imagery’ which even Professor Cornford has recently attributed to Anaximander, he and other philosophers retained the habit of regarding cosmogony and anthropogony as parallel phenomena, only substituting for the mythical ‘world-egg’ a more scientific view of the embryo and its development. They looked at the world through the same spectacles as their predecessors, but after wiping from them the rose tint of mythological fancy.

In the course of his dispute with Conington on the comparative merits of Catullus and Horace, Munro taxed the Augustans with having made the lyric of the heart impossible in Latin by their virtual exclusion of diminutives from the language of poetry; and, whether that is the result or no, the general fact that diminutives are rare in the serious poetry of the Augustan age is well known. The details, however, are less easy to come by. Stolz (Hist. Gr. d. lat. Spr., p. 574) and Stolz-Schmalz (hat. Gr., p. 834) devote a few unilluminating lines to the Stilistik of diminutives: otherwise the grammars and the treatises on diminutives known to me concern themselves only with forms and meanings. Except for a note by Professor Housman which, at 4. 927, sets out Manilius's diminutives, I know of no collections for any Augustan poet, and it is perhaps worth while therefore to state the facts. I have not indeed read through Augustan poetry for the purpose, but for some time past I have been in the habit of noting such diminutives as I have come across in the course of reading, and these lists I have now checked and amplified from the indexes to the authors concerned. My lists are probably not complete, but I hope they are sufficiently near it to present a true picture of the position.

Mr. Shewring's recent articles on ‘Prose-rhythm and the Comparative Method’ (C.Q. XXIV, pp. 164 sqq.; XXV, pp. 12 sqq.) are gratifying in that they betoken a growing interest in the problems of a comparatively modern and fascinating study, and also an appreciation of the methods followed by different investigators. His estimate, however, of De Groot's services seems to me somewhat extravagant; his estimate of Zielinski's contributions unduly belittling (as I shall show below, he cannot have read Zielinski's Der constructive Rhythmus); while his references to my own work cause me to doubt whether he has grasped even the main contention of my book. I shall deal with all these points presently.

It is not without a certain feeling of surprise that I find the fragment preserved by Strabo VIII. 3, 20, and somewhat doubtfully ascribed by him to Stesichoros, still commonly attributed to that writer. As the purpose of this note is to give what seem to me cogent reasons for holding that no poem of such a metre and content could be by an author of any possible date earlier than Alexandrian times, I cite the passage of Strabo in full. It runs as follows, and is part of an argument for the existence of an ancient city called Samos in Elis.

The text of Seneca's Letters, despite the attention it has received from scholars in the last fifty years, still leaves much to be desired in a large number of places. It is a field in which emendations can be proposed (and judged) with rather more security than is often the case in classical Latin prose, because Seneca was a very prolific writer, exceeded only by Cicero and Livy in the bulk of his extant work. The absence of a special lexicon for this eminent author constitutes a genuine difficulty of course, but it can be overcome in part at least by a constant reading and re-reading of his works, not excluding the dramas. The following suggestions are based on a more intense application to Seneca than Quintilian would be likely to have approved.

At the end of Republic VI. Socrates compares the Good with the sun as a cause both of existence and intelligibility. Afterwards, when he continues and expands this comparison, the symbolism becomes so complex that the interpretation of almost every part of it is in dispute. We start with the contrast of light and darkness; to this is next added the contrast of image and original, and also of up and down along a vertical line; in the allegory of the Cave these three sets of contrasted terms are worked in together so as to explain the effect on the soul of the intellectual education by which we are converted from what is dark and imitative and ascend finally to knowledge of the Good.

InC.Q., January, 1928, pp. 16 sqq., I examined afresh the two discussions of poetry as imitation which are found in Plato's Republic. I pointed out that Plato used the term ‘imitation’ (μίμησɩς and cognates) in two senses, a good and a bad. The only kind of poetry which Plato excludes from his ideal state is that which is imitative in the bad sense of the term. He admits, and indeed welcomes, that kind of poetry which is imitative in the good sense (without discussing the question whether any such poetry is actually extant), and which he calls either imitative or non-imitative according as he is using the term ‘imitative’ in a good sense or a bad. The kind of poetry (briefly summed up in Rep. 607a as consisting of hymns to the gods and eulogies of good men) which is admitted into the ideal state is certainly imitative; and (which is the important point) it is in fact called imitative by Plato himself (e.g. in 397d τòν τοῦ ἐπɩεɩκοῦς μɩμŋτὴν ἄκρατον)—of course in the good sense of the word. Now the tenth book begins by stating that the result of the earlier discussion had been the decision to exclude ‘so much of poetry as is imitative.’ The implication is obviously that the remainder will not be excluded. (Plainly it is assumed that such a remainder either exists or will be successfully called into existence by the ideal state.) But what are we to call that remainder, unless it be non-imitative in the sense (the bad sense) in which the word is here used?

Cassandra speaking.

The first of these lines is not in dispute; the three which follow are notorious; they are subjoined as in the manuscripts, with punctuation to mark the ostensible construction:

ε δ ἔπαρχος Ίλίοʊ τ άναστάτης

ούκ οἶδεν οἶδ уλσσα μισητς κʊѵòς

λέξασα καì κτείνασα ϕαιδρόνοʊς, δίκην

ἄτης λαθραίοʋ τεύξεται κακῇ τύχῃ.

The first of these passages states some simple principles of mechanics. The second uses one of these principles to prove that a finite mover cannot cause a motion that will occupy unlimited time. The argument there has given much trouble to commentators because the principle in question was not understood, owing to the choice of a false reading in the earlier passage.

One of Sulla's first acts on assuming the dictatorship in 81 B.C. was to fill up the numbers of the Senate by the addition of some 300 new members. Tradition is divided on the question of the rank of these men before their promotion, and no unanimity has yet been reached in the matter. There are two distinct versions in the ancient authorities, both equally well attested. Appian and the Epitomator of Livy state that the new members were equites, while Sallust and Dionysius of Halicarnassus assert that they included men of the lowest rank (gregarii milites is Sallust's phrase). When we turn to modern historians we find opinion similarly divided. Writers such as Lange, Zumpt and Herzog accept the first version, whereas Niebuhr, Botsford, Willems and Gelzer incline to the view of Sallust. It is tempting to do as some writers have done and to combine the two versions, saying that some of the new Senators were equites and some common soldiers. But this is a very arbitrary cutting of the knot. Herzog seems to play with the idea that Sulla promoted ordinary soldiers via the equites when he says ‘wobei noch dazu mancher war, der seine Ritterstellung erst eben errungen hatte.’ This, however, is mere guesswork.

There are four problems to be faced in 710–12. Is πρσει correct? Who was Orestes? What is the construction of 'ρστῃ? What is the meaning of ἵνα μῥιλν ποδὐῃ? A brief review of the evidence collected on the subject of Orestes is the first essential.

For about a hundred years there has been an intermittent but sometimes vigorous debate1 on the question whether Quintus Smyrnaeus and Tryphiodorus directly used the Second Aeneid as a source for their epic descriptions “of the capture and destruction of Troy. Heyne thought that they did not; but towards the end of the nineteenth century it appeared more likely that they did. Heinze opposed the general belief: but it was reaffirmed for Quintus by Paschal and Becker4 and for Tryphiodorus by Dr. E. Cesareo on internal evidence. Meanwhile Professor Samuel E. Bassett had concluded that Quintus was not after all dependent on Vergil: and still more recently Dr. J. W. Mackail, reviewing Cesareo's monograph, has maintained with emphasis that the Aeneid is certainly not the source of the late epic. Accordingly the controversy cannot even now be considered closed.

Editors of Pliny's Naturalis Historia have not had to deplore the paucity of the MS. tradition, but rather its value; while MSS. belonging to the ordo recentiorum are numerous and fairly complete, those of the ordo uetustiorum are very few, and never contain more than a few books, often with considerable gaps. They are A ii 196–vi 51, M xi–xv, P and H parts of xviii, I xxiii, xxv, B xxxii–xxxvii (end). There are also some scattered fragments. Detlefsen indeed claimed that the restoration of the original text of the N.H. is more difficult than that of any other Latin author owing to the mutilated and defective transmission of considerable parts.