cussed, (2) a general survey of their branch of literature, (3) brief notices of the authors in chronological order. The publication took place, according to Roth, 106-113 A.D.

3. Minor works, now lost (mentioned by Suidas), on Greek games, Roman games, the Roman year, on critical marks, on Cicero's Republic, on dress, on imprecations (Tepì δυσφήμων λέξεων ἤτοι βλασφημιῶν καὶ πόθεν ἑκάστη), on Roman laws and customs.. Some of these were probably only sections of the Prata, a miscellany in ten Books, which also treated of natural science and philology. The books on Greek games and on imprecations were almost certainly composed in Greek.

APPENDIX A.

ON SOME OF THE CHIEF ANCIENT AUTHORITIES FOR THE HISTORY OF ROMAN LITERATURE.

1. JEROME1 (HIERONYMUS) was born about A.D. 335 at Stridon, on the frontiers of Dalmatia and Pannonia, and died A.D. 420 at the monastery of Bethlehem. His contributions to the history of Roman literature are to be found in his translation of the Chronicle (Xpovikoì kavóves) of Eusebius, in which the dates are reckoned from the first year of Abraham (=B.C. 2016 according to his chronology), the point at which Eusebius commenced. On the period between the Trojan War and A.D. 325 Jerome not merely translated the remarks of Eusebius, as he had done in the earlier period, but also added numerous extracts from authorities on Roman history and literature. The source from which he derived nearly all his information on literature is universally admitted to have been the work of Suetonius De Viris Illustribus. With the state. ments in the surviving sections of that treatise the observations of Jerome agree, and there can be no reasonable doubt that he made a similar use of the parts no longer extant. It is a significant fact that the important authors on whom Jerome is silent, e.g. Tacitus, Juvenal, and the younger Pliny, are precisely those whom Suetonius, as a contemporary, naturally could not discuss.

1 See Quaestiones Suetonianae in Reifferscheid's Suetonius, pp. 363 sqq.

The statements of Jerome, based as they are on the high authority of Suetonius, may be regarded as in the main trustworthy. Some of them, however, are doubtful, and others manifestly wrong.

(a) Jerome's plan obliged him to fix every event to a definite year; and this, in many cases, can only be guess-work, for Suetonius, as may be seen from his extant writings, was often vague in his chronology.

(b) Comparison with the remains of Suetonius shows that Jerome's claim to have made his extracts with care was not always well grounded; e.g. his statement that Ennius was a native of Tarentum (see p. 27).

(c) In reckoning, according to his system of dates, events dated by one of the many confusing systems of chronology current in ancient times, many openings for error presented themselves; e.g. he sometimes erred through confusing consuls of the same or similar names, as in the case of Lucilius (p. 59); or through confusing similar events, as in the case of Livius Andronicus, although the mistake about the latter was of long standing (p. 2). Once at least he seems to have confused the date of an author's floruit and that of his death, making Plautus die in B.C. 200 instead of B.C. 184 (p. 8).

2. AULUS GELLIUS1 was born probably about A.D. 123, and studied under the most eminent teachers both at Rome and at Athens. Of his subsequent life nothing is known except that he held some judicial post at Rome. His work, the Noctes Atticae in twenty Books (of Book viii. only the headings of chapters are preserved), is a miscellany of information on philology, philosophy, rhetoric, history, biography, literary criticism, natural science, and antiquities. The title is due to the fact that the book was commenced in the winter evenings during the author's residence at Athens. The arrangement of

1 See H. Nettleship, Lectures and Essays (1885), p. 248 sqq.

the contents simply follows the haphazard order of the notes which Gellius made in the course of his reading of Greek and Roman authors. Those authors, and the conversation of contemporaries, are Gellius' professed sources, but in some cases the author he names is evidently quoted at second-hand, and many of the conversations are doubtless quite imaginary. Our obligations to Gellius are twofold.

(a) Innumerable extracts from ancient authors are preserved by him alone. (No quotations are given from post-Augustan writers a fact which accords with the affected archaism of his style.)

(6) His remarks on incidents in the lives of the Roman poets are in the main derived from Varro, whose work De Poetis is quoted for the epitaph of Plautus (see p. 9); elsewhere his source is indicated either vaguely or not at all, e.g. iii. 3, 15, 'accepimus'; xii. 4, 5, 'ferunt.' For literary criticism Varro is quoted: iii. 3, 9, sqq.; vi. 14, 6 (see pp. 10, 51).

3. NONIUS MARCELLUS,1 a Peripatetic, of Thubursicum in Numidia, is identified by Mommsen with the Nonius Marcellus Herculius of C.I.L. viii. 4878 (date A.D. 323); but nothing is known of his life. His work, De Compendiosa Doctrina ad Filium in twenty Books (of Book xvi. the title only is known; Book xx. is fragmentary), though modelled on that of Gellius, is immeasurably inferior in execution. According to the theory usually received Nonius borrowed largely from Gellius; but it is possible that both compilers made independent use of the same authorities, viz., scholars such as Verrius Flaccus, Valerius Probus, and Suetonius, whose works they knew either directly or through abridgments. The subjects with which Nonius deals are grammar, lexicography, and antiquities; and he is often our sole authority for the titles of works as well as for brief extracts.

L.A.

1See Nettleship, ibid. p. 277 sqq.

Z

4. AMBROSIUS THEODOSIUS MACROBIUS, doubtless identical with the Macrobius who held, among other high offices, the proconsulship of Africa A.D. 410, was probably, like Nonius, of African origin. Besides his commentary on the Somnium Scipionis of Cicero, Macrobius wrote a work in seven Books on Roman literature and antiquities with the title of Saturnalia. The imaginary conversations of which it consists are supposed to take place during the festival of the Saturnalia at Rome (hence the title); and the chief subject of discussion is the poetry of Virgil. A remarkable feature of the book is its wealth of quotation from Greek and Latin authors. Macrobius, like Gellius, bases his work on extracts from older authorities; but, unlike him, arranges his matter systematically.

5. AELIUS DONATUS, a grammarian who flourished at Rome about A.D. 350, and was one of Jerome's teachers, extracted from the lost work of Suetonius the Lives of Terence and Virgil, and prefixed them to his own commentaries on Terence and on the Georgics and Aeneid. The latter is lost, and the commentary on Terence contains much that is not from the hand of Donatus.

6. SERVIUS.-There are two versions of the Servian commentary on Virgil. The shorter is the work of Maurus Servius Honoratus, who was born about 350 A.D., and lived at Rome (Macrob. Saturn. i. 2, 15); his topographical references show that he composed his commentary there. Servius, whose notes are chiefly on the language of the poems, gives illustrative quotations from Roman authors, in some cases from memory and inaccurately. Donatus is the authority whom he mentions oftenest, but he undoubtedly made extensive use of Suetonius.

The longer version contains learned additions to the work of Servius by an anonymous Christian writer, who deals mainly with the subject-matter of Virgil.

7. ACRO and PORPHYRIO.-Helenius Acro (probably about 200 A.D.) was the author of commentaries on Horace and