² Fo. 28^r, II. 1–5. On this codex in general see J. Quicherat, ‘Fragment inédit d'un versificateur latin ancien sur les figures de rhetorique’, Bibliothèque de l'ecole des Chartes 1 (1839/40), 51–78, Th. Mommsen ap. Th., Bergk, Zeitschr. f d. Alt. 2. 3 (1845), 81–8 ( = Gesammelte Schriften vii (Berlin, 1909), 217–18), H. Keil ap. F. A. Eckstein, Anecdota Parisina rhetorica (Halle, 1852 (Programm der Lateinischen Hauptschule zu Halle für das Schuljahr 1851–2)), pp. Grammatici Latini iv (Leipzig, 1864), p. xli note, Usener, H., ‘Vier lateinische Grammatiker: IV. Paulus von Constantinopel’, RhM 23 (1868), 497–503,Google Scholar 543–4 (= Kl. Schr. ii. 179–92), Lejay, P., ‘Notes Latines V: B. N. Lat. 7530. VI Paulus Constantinopolitanus’, RPh 18 (1894), 42–59,Google Scholar E. A. Lowe, Die ältesten Kalendarien aus Monte Cassino (Munich, 1908 (Quellen und Untersuchungen zur lateinischen Philologie des Mittelalters, iii. 3)), pp. 4–6, Scriptura Beneventana (Oxford, 1929), plate IX, W. M. Lindsay, Notae Latinae (Cambridge, 1915), pp. xv, 12, 41, 46, 47, 51, 53, 55, 56, 69, 77, 100, ‘Palaeographical Notes’, App. II to H. J. Lawlor, ‘The Cathach of St. Columba’, Proc. Royal Ir. Ac. 33 (1916), 399 n. 2, C. H. Beeson, ‘Paris, Lat. 7530. A Study in Insular Symptoms’, Raccolta di scritti in onore di Felice Ramorino (Milan, 1927), pp. 199–211, Lehmann, P., ‘Fulda und die antike Literatur’, Aus Fuldas Geistesleben, ed. Theele, J. (Fulda, 1928), pp. 9–23,Google Scholar ‘Die alte Klosterbibliothek Fulda und ihre Bedeutung’, Aus der Landesbibliothek Fulda 2 (1928), 5 (= Erforschung des Mittelalters i (Stuttgart, 1941), 220), Campana, A., ‘Per la storia della Biblioteca della Cattedrale di Benevento’, Bull. Arch. Pal. It. N.S. 2–3 (1956–1957), part I, 152–63,Google ScholarBrunhOlzl, F., Zum Problem der Casineser Klassikeriiberlieferung (Munich, 1971 (Abhandlungen der Marburger Gelehrten Gesellschaft, 1971, no. 3)), p. 114,Google ScholarBischoff, B., ‘Die Bibliothek im Dienste der Schule’, La scuola nell' Occidente latino dell' alto medioevo, Settimane di Studio del Centro Italiano di Studi sull' Alto Medioevo xix (Spoleto, 1972), 396,Google ScholarCavallo, G., ‘La trasmissione dei testi nell' area Beneventano-Cassinese’, La cultura antica nell' Occidente Latino dal VII all' XI secolo, Settimane di Studio xxii (Spoleto, 1975), 363–5,Google ScholarHoltz, L., ‘Le Parisinus Latinus 7530, synthèse cassinienne des arts libéraux’, Studi Medievali, 3 ser. 16 (1975), 97–152.Google Scholar

³ Fo. 64^v, col. b, 17. 14–18. See on this codex Morelli, C., ‘I trattati di grammatica e retorica del cod. casanatense 1086’, Rendiconti della Reale Accad. dei Lincei, Cl. Sc. Mor. Stor. e Filol., Ser. V, vol 19 (1910), 287–328, E. A. Lowe, Scriptura Beneventana, plate XVI, A. Campana, op. cit., pp. 152–63, G. Cavallo, op. cit., pp. 363–5, 367–8.Google Scholar

⁴ Cod. Rome, Bibl. Vat. Lat. 3226 fos. 5^r/5^v TERENTI ANDRIA FINITA/INCIPIT EVNVCHVS TERENTI, 29^r/29^v, 52^v/53^r, 76^r/76^v, 96^r/96^v and cod. Milan, Bibl. Ambros. G. 82 ord. sup. fos. 224^v T. MACCI PLAVTI CASINA EXPLICIT INC. CISTELLARIA FELICITER, 280^r, 432^v are not exactly similar.

⁵ P. Oxy. 2256, frs. 1–7.

⁶ P. Bodmer 4.

⁷ Cf. H. Keil, Zeitschr. f. d. Alt. 2. 6 (1848), 551 (‘eine Art von Didaskalie’), Schanz, M., Hosius, C., Geschichte der rOmischen Literatur ii (Munich, 1935), 163,Google ScholarBardon, H., La litterature latine inconnue ii (Paris, 1956), 31, L. Holtz, op. cit., pp. 98, 113, 139.Google Scholar

⁸ Cf. F. W. Schneidewin, RhM 1 (1842), 108, Ribbeck, O., Tragicorum Latinorum Fragmenta (Leipzig, 1852), p. 347,Google ScholarKroll, W. and Skutsch, F., W. S. Teuffels Geschichte der römischen Literatur⁶ ii (Leipzig-Berlin, 1910), 21,Google ScholarLowe, E. A., The Beneventan Script (Oxford, 1914), p. 17,Google ScholarHelm, R., RE ii 8.i (1955), 413,Google Scholar A. Campana, op. cit., p. 161, G. Cavallo, op. cit., p. 364, Lefevre, E., Der Thyestes des Lucius Varius Rufus (Wiesbaden, 1976 (Abh. Mainz Ak. 9)), P. 38.Google Scholar

⁹ Arch. f. Stenogr. 53 (1901), 207 (= Vorlesungen and Abhandlungen iii (Munich, 1920), 271). W. M. Lindsay wrote (CQ 16 (1922), 180) of a ‘vita Varii’.

¹⁰ The entry at Ol. cxc 4 in Jerome's Chronica shows that Suetonius included Varius among his poets.

¹¹ Zeitschr. f. d. Alt. 2. 3 (1845),84 (= Ges. Schr. vii.218). Cf. A. E. Housman's I ‘prefatory note’ (CQ 11 (1917), 42 (= Classical Papers iii, Cambridge, 1972, 941)).

¹² There is no clear case of signifying ‘note of performance’ in recorded Greek. On the contents of Aristotle's the use made of this work by. students of particular Attic scripts and continuations of it see G. Jachmann, De Aristotelis didascaliis (Diss. Göttingen, 1909), A. Pickard-Cambridge, The Dramatic Festivals of Athens, ed.² John Gould and D. M. Lewis (Oxford, 1968), p. 71, Pfeiffer, R., History of Classical Scholarship from the Beginnings to the End of the Hellenistic Age (Oxford, 1968), p. 81. Didascalia does not occur in recorded Latin (ThLL V 1, 1015. 8 appears to have been misled by the format of Schlee's edition of the medieval Terence scholia).Google Scholar

¹³ On the use and misuse of the term sec G. Zuntz, Byzantion 14 (1939), 548 ff. (= Die Aristophanes-Scholien der Papyri (Berlin, 1975), 64 ff.).

¹⁴ See G. Zuntz, Byzantion 13 (1938), 631–90, 14 (1939), 545–614 (= Die Aristophanes-Scholien der Papyri (Berlin, 1975)), An Enquiry into the Transmission of the Plays of Euripides (Cambridge, 1965), pp. 272–5,Google ScholarWilson, N. G., ‘A Chapter in the History of Scholia’, CQ N.S. 17 (1967), 244–56,CrossRefGoogle ScholarTurner, E. G., Greek Papyri. An Introduction (Oxford, 1968), pp. 112–24,CrossRefGoogle ScholarZetzel, J. E. G., ‘On the History of Latin Scholia’, IISCPh 79 (1975), 335–54.Google Scholar

¹⁵ Unlike the other four surviving Bembine notes this one fails to give the actors, the musician, the type of music or the date of composition. The Calliopian note, on the other hand, is formally intact.

¹⁶ See Mountford, J. F., The Scholia Bembina (London, 1934), pp. 119 ff., on the question of whether the original commentary or the extant commentum Donati was the annotator's source.Google Scholar

¹⁷ See Ritschl, F., ‘Die Plautinischen Didaskalien’, RhM 1 (1842), 29–88Google Scholar (= Parerga zu Plautus und Terenz (Leipzig, 1845), pp. 249–336), W. Wilmanns, De didascaliis Terentianis (Diss. Berlin, 1864), Dziatzko, K., ‘Lieber die Terentianischen Didaskalien’, RhM 20 (1865), 570–98, 21 (1866), 64–92, U. von WilamowitzMoellendorff, Euripides: Herakles i: Einleitung in die Attische Tragödie (Berlin 1889), 145, G. Jachmann, op. cit., pp. 52–60. Bibliographies list D. Klose, Die Didaskalien und Prologe des Terenz (Diss. Freiburg, 1966).Google Scholar

¹⁸ Hence FACTAST TERTIA CN. CORNELIO MARCO IVVENIO (Bemb. Ter. Heaut.) and the like.

¹⁹ The behaviour of Verrius Flaccus in citing one Plautine script under the titles Mostellaria (Fest. p. 166. 19) and Phasma (pp. 158. 33, 394. 18) and another under Cistellaria (p. 512. 10) and Synaristosae pp. 390. 8, 480. 23 - see E. Fraenkel, Philologus 87 (1932), 117 ff. (= Kleine Beiträge zur klassischen Philologie (Rome, 1964) ii. 33 ff.) - suggests that his texts were prefixed with notes naming the Greek originals.

²⁰ e.g. Zimmermann, J., Luciani quae feruntur Podagra et Ocypus praefatus edidit commentatus est (Leipzig, 1909); pp. 79 ff. Zimmermann also argues, p. 39, that the prose comes from a learned man who thought the and the to be by the one author.Google Scholar

²¹ Similar items to those which compose the Lucianic essay collect under the rubrics ‘ et sim. in P. Bodmer 4 and the medieval traditions of the dramatists. P. Oxy. 1235 (early second century AD.) carries a work perhaps the by Homerus Sellius, which provided for a number of Menandrian comedies information on the first performance, a summary of the plot, and a brief aesthetic judgement. A work on Euripides' tragedies carried by a number of first, second- and third-century-AD. papyri provided much longer plot summaries and perhaps also aesthetic judgements (see Coles, R. A., A New Oxyrrhynchus Papyrus: The Hypothesis of Euripides' Alexandros (Univ. of London, Inst. of Classical Studies, Bull. Suppl. 32 (1974)), pp. 69–70).Google Scholar Despite the access of new material Schneidewin, F. W., ‘De hypothesibus tragoediarum Graecarum Aristophani Byzantio uindicandis commentatio’, Abh. kön. Ges. Wiss. Göttingen 6 (1856),Google Scholar remains fundamental. See also Page, D. L., Euripides: Medea (Oxford, 1938),Google Scholar pp. liii-lvii, Zuntz, G., The Political Plays of Euripides (Manchester, 1955; corr. repr. 1963), pp. 129–52,Google ScholarBarrett, W. S., Euripides: Hippolytos (Oxford, 1964), pp. 153–4,Google Scholar R. Pfeiffer, History, pp. 192–6, Stevens, P. T., Euripides: Andromache (Oxford, 1971), pp. 26–8.Google Scholar

²² See E. G. Turner, Greek Papyri, p. 122.

²³ The situation is, of course, often rationalized; cf. P. Cair. 43227 (Menander: V cent.).

²⁴ These are usually said to have been ludi performed by Octavian after his triumph in 29 (Dio 51. 21. 7–9) but stage plays would have been odd at such ludi. The dedication of the Palatine temple of Apollo in 28 (Aug. Res gest. 19, Fast. Amit. (Inscr. It. 13. 2. 195), Fast. Ant. min. (Inscr. It. 13. 2. 209), Sueton. Aug. 29. 1, 3, Dio 53. 1. 3) offered a more likely occasion. Mimes by Publilius and Laberius had been performed at Caesar's dedication of the temple of Venus Genetrix in 46 (Cic. Fam. 12. 18. 2).

²⁵ P. Oxy. 2257, fr. 1 however, looks like part of an introductory note to Aeschylus' (see Fraenkel, E., Eranos 52 (1954), 61–75 (= KI. Beitr. i. 249–62)). It is tempting to speculate that a more specific version of Vit. Aesch. 9 stood in the part lost.Google Scholar

²⁶ Epist. 2. 1. 168–76.

²⁷ In the ‘ attached to Euripides’ ‘ occurs the statement

²⁸ For the overall literary quality of particular Terentian scripts see Eun. praef. 9 ‘in hac Terentius delectat facetiis, prodest exemplis et uitia hominum paulo mordacius quam in ceteris carpit. exempla autem hic morum trina praecipua proponuntur: urbani moris, parasitici, militaris’, Ad. Praef. i. 9 ‘in hac spectator, quid intersit inter rusticam uitam et urbanam, mitem et asperam, caelibis et mariti, ueri patris et per adoptationem facti. quibus propositis ad exemplum imitanda perinde fugiendaque Terentius monstrans artificis poetae per totam fabulam obtinet laudem', Hec. praef i.9 ‘in tots comoedia hoc agitur, ut res nouae fiant nec tamen abhorreant a consuetudine: inducuntur enim beniuolae socrus, uerecunda nurus, lenissimus in uxorem maritus et item deditus matri suae, meretrix bona.’ For the verbal skill shown in particular scripts see Ad. praef i.3 ‘prodest autem et delectat actu et stilo’, Hec. praef. i.3 ‘multumque sententiarum et figurarum continet in toto stilo. unde cum delectet plurimum, non minus utilitatis affert spectatoribus’, Phorm. praef. i.3 ‘et in affectibus constituta paene maioribus quam comicus stilus posceret, nisi quod arte poetae omnia moderata sunt.’

²⁹ This underlies the abuse at Aristoph. Pax 696–9.

³⁰ It was much more appropriate in works like the of Callimachus, doubtless one of Aristophanes' sources (see Pfeiffer on fr. 456).

³¹ The scripts of Plautus and Terence thus have nothing to correspond with the prayer for victory which regularly concluded fourth- and third-century Athenian comic scripts (cf. Menan. Dysc. 968–9, Misum. 465–6, Sam. 736–7, Sicyon. 422–3, Posidippus, fr. 218 CGF, Com. anon. fr. 249. 20–1 CGF). The anecdote related at Macrob. Sat. 2. 7. 7–8 shows how unusual dramatic competitions were even at the end of the Republic.

³² Epist. 2. 1. 175. Also perhaps to the story about Plautus' financial activities related by Varro in his De comoediis Plautinis (Gell. 3. 3. 14).

³³ Donat. Vit. Ter. 3.

³⁴ The ‘didascalic’ notice to the Hecyra in the Bembine records three performances, with non est placita against the second and placuit against the third. The Calliopian notices have nothing of the sort. It is likely that Suetonius drew his information from a set of commentaries rather than a set of texts and remembered in passing an unusual text of the Eunuchus.

³⁵ Contrast the incipit … acta format of the Bembine and Calliopian notices.

³⁶ For the form Thuesta see Charisius, p. 85 Barwick. Tuesta would probably have been the Ennian spelling, despite the in Thyeste entries in the lexica of Festus and Nonius. The metre of Agam. 4 shows that Seneca formed the nominative as Thyestes.

³⁷ The -ii genitive does not begin to spread in inscriptions until towards the end of the reign of Augustus. The VARII of the extant titulus would have been introduced, by a copier with one eye on the subjoined note. If anyone before lunius Philargyrius attributed a Thyestes to Alfenus Varus (see below, p. 398), a titular VARI had surely assisted in the error. It is of interest that early first-century-BC scholars tried to interpret the PLAVTI of some comic tituli as pertaining to a Plautius rather than a Plautus (Varr. ap. Gell. 3. 3. 10). The problem of distinguishing persons who operated under the one name much exercised ancient scholars (see Demetrius Magn. ap. Diog. Laert. 1. 38 et alibi, Sueton. ap. Donat. Vit. Ter. 8).

³⁸ Porphyrio describes L. Varius as ‘et epici carminis et tragoediarum et elegiarum auctor’ (Hor. Carm. 1. 6. 1).

³⁹ A number of agrimensores are collected in the fifth-sixth-century cod. Wolfenbüttel, Herzog-Aug. Bibl. 2° 36. 23 (CLA IX 1374); a number of grammatici in the fifth-century cod. Naples, Bibl. Naz. Lat. 2, fos. 76–111, 140–56, 159 (CLA III 397a). A single codex contained Cicero's De inuentione, Quintilian and Fortunatianus in Cassiodorus' library (Inst. 2. 10). Where poets are concerned, the satirists Persius and Juvenal were already united in the sixth century (cod. Rome, Bibl. Vat. Lat. 5750, pp. 63–4, 77–8 (CLA I 30)). So too must have been the elegists Tibullus, Lygdamus, and Sulpicia.

⁴⁰ Not that we have an actual sentence from a chapter of the De uiris illustribus. The poet's full name would have come at the beginning and the account of a patron's munificence towards the end. Too much should not be made of the substance of the sentence. Comparison of Donat. Vit. Verg. 13 ‘possedit prope centiens sestertium ex liberalitatibus amicorum’ with [Prob.] Vit. Verg. p. 324. 14–16 Hagen ‘Aeneida ingressus … ab Augusto usque ad sestertium centies honestatus est' shows what could happen even in formal scholastic writing to talk of money.

⁴¹ Op. cit., p. 52. Cf. Schneidewin, F. W., RhM 1 (1842), 107:Google Scholar ‘Wer grollte nicht dem bösen Schreiber, dass er die durch sein INCIPIT erregte Hoffnung auf das kötliche Stück so bitter taüschen konnte! Hätte ihm doch gem Jedermann für nur 25 Verse des Varius seinen ganzen in alien Bibliotheken steckenden Isidorus so gem geschenkt’; Housman, A. E., CQ 11 (1917), 42CrossRefGoogle Scholar (Collected Papers iii.941): ‘One day towards the end of the eighth century the scribe … began to copy out for us … the Thyestes of Varius. He transcribed the title and the prefatory note. … Then he changed his mind; he proceeded with a list of the notae employed by Probus and Aristarchus, and the masterpiece of Roman tragedy has rejoined its author in the shades’; Helm, R., RE ii.8.i (1955), 413, s.v. Varius 20: ‘das incipit Thuestes Varii in dem Scholion des Parisinus deutet doch darauf hin, dass es bis in das 8 Jhdt. noch erhalten war.’Google Scholar

⁴² Quicherat's error, repeated by Schneidewin and Schneidewin's publisher, Welcker, F. G., Die griechischen Tragödien iii (Bonn, 1841), 1429 n. 24, was nailed by Th. Mommsen ap. Th. Bergk, Zeitschr. f. d. Alt. 2. 3 (1845), 85 (Ges. Schr. vii.218).Google Scholar

⁴³ Op. cit., p. 31. Cf. Jocelyn, H. D., The Tragedies of Ennius (Cambridge, 1967), p. 49 n. 1: ‘Varius' Thyestes may have reached the age of Charlemagne.’Google Scholar

⁴⁴ See Th. Bergk, Zeitschr. f. d. Alt. 2. 3 (1845), 84 n. (= Kleine Schriften i (Halle, 1884), 585) ‘wie übrigens Schneidewin … die Behauptung aussprechen kann das begreife, wer es kann; ich sollte meinen, das Blatt des Codex bewiese gar nichts oder vielmehr das Gegentheil.’ Cf. Keil, H., Zeitschr. f. d. Alt. 2. 6 (1848), 550–1,Google Scholar O. Ribbeck, op. cit., p. 347, Lindsay, W. M., CQ 16 (1922), 180, L. Holtz, op. cit., p. 114.CrossRefGoogle Scholar

⁴⁵ See C. Morelli, op. cit., pp. 321–2.

⁴⁶ The case of the text of Juvenal's sixth satire offered by cod. Oxford, Bodl. Lib. MS. Canon. 41 is famous. For the existence of a distinct South-Italian text of Servius' commentary on the poems of Virgil see Stocker, A. F., HSCPh 52 (1941), 65–97; for Priscian's Institutiones grammaticae see M. de Nonno, RFIC 105 (1977) 385–402.Google Scholar

⁴⁷ See the works of Brunhölz1 and Cavall cited on p. 387 n. 2. Northern sources could in theory have supplied exemplars and ‘insular’ symptoms would betray some of these. The very number however of texts unique to Southern Italy is significant and palaeographers are not as eager as they once were to detect ‘insular’ symptoms. On cod. Florence, Bibl. Laur. 68. 2 (Tacitus: XI cent.) see M. Zelzer, WSt 86 (1973), 185–95.

⁴⁸ On literary culture in Italy in the dark ages see B. Bischoff, ‘Scriptoria e manoscritti mediatori di civilta dal sesto secolo alla riforma di Carlo Magno’, Centri e vie di irradiazione della civiltà nell' alto medioevo, Settimane di studio del Centro italiano di studio sull' alto medioevo XI (Spoleto, 1964), 485–92 (= Mittelalterliche Studien ii (Stuttgart, 1967), 316–20); on the time of Charlemagne id. ‘Panorama der Handschriftenüberlieferung aus der Zeit Karls des Grossen’ in Braunfels, W., Karl der Grosse. Lebenswerk and Nachleben ii (Dusseldorf, 1965), 233–54;Google Scholar on Monte Cassino in particular Bloch, H., ‘Monte Cassino's Teachers and Library in the High Middle Ages’, La Scuola nell' Occidente latino dell' alto medioevo, Settimane di studio xix (Spoleto, 1972), 563–605;Google Scholar on South Italy Cavallo, op. cit., pp. 357–414. The classical works copied in Beneventan script can be found in Lowe, E. A., The Beneventan Script (Oxford, 1914), pp. 1618Google Scholar (Pal. Papers i.87–8), ‘A new list of Beneventan manuscripts’, Collectanea Vaticana in honorem Anselmi M. Card. Albareda, vol. 2 (Vatican City, 1962 (Studi e Testi 220)). The lists of Huglo, M., ‘Liste complementaire de manuscrits beneventains’, Scriptorium 18 (1964), 89–91,CrossRefGoogle Scholar and Finch, C. E., ‘Beneventan Writing in Codices Vat. Lat. 3032, 5951 and 7277’, AJPh 87 (1966), 455–7,Google Scholar and ‘More Beneventan Manuscripts’, Class. Bull. 52 (1975), 8–10, have not altered the appearance of the situation.Google Scholar

⁴⁹ C at one time possessed at least one other quaternion. On the order of items common to P and C in the exemplar see Holtz, op. cit., p. 141.

⁵⁰ On the design of the miscellany in P see Holtz, op. cit., pp. 142–5. The maker of that in C sought to supplement the exposition of Latin morphology which Ursus (elected Bishop of Benevento in 833) had based on books I-XVI of Priscian's Institutiones grammaticae.

⁵¹ It is commonly held that this exemplar was made in South Italy but certainty seems unobtainable. C. H. Beeson, op. cit., pp. 202–3, claimed to find a small number of insular symptoms in the Paris manuscript's text of the SCEMATA DIANOEAS QVAE AD RHETORES PERTINENT and the DE LAVDIBVS VRBIVM. L. Holtz, op. cit., pp. 146–7, pointed out that the parts of the former dependent on Isidore relate to the ‘French’ rather than the ‘Italian’ tradition of the Origines but nevertheless maintained a South-Italian origin for the exemplar. See also Brunhölz1, op. cit., p. 114, and Cavallo, op. cit., pp. 363–5.

⁵² Fos. 53^v, col. a, 35–54^r, col. b, 15 (= P fos. 38^r, 16–38^v, 34 DE FIGVRIS FACTIS PER GENETIVVM CASVM).

⁵³ Fos. 54^r, col. b, 27–54'v, col. a, 11 (= P fos. 62^v, 1–17 without title but separated by blank space from previous summary of Pompeius' account of vices and figures of speech = Keil, Gramm. Lat. V. 133. 13–134. 2).

⁵⁴ Fos. 60^r, col. b, 27–63^r, col. b, 6 (= P fos. 221^r, 18–224^y, 3 SCEMATA DIANOEAS QVAE AD RHETORES PERTINENT = Halm, Rhet. Lat. 71–7).

⁵⁵ Fo. 63^r, col. b, 6–14 (= P fo. 224^v, 3–8 DE LAVDIBVS QVARVMQVE RERVM = Halm, Rhet. Lat. 587–8).

⁵⁶ Fos. 63^r, col. b, 15–63^v, col. a, 13 (= P fo. 224^v, 9–29 = Halm. Rhet. Lat. 588).

⁵⁷ Fos. 63^v, col. a, 13–64^v, col. b, 13 (= P fos. 224^v, 30–228^v, 5 = Halm, Rhet. Lat. 63–70 = Riese, Anth. Lat. 485 = Baehrens, Poet. Lat. Min. iii.273–85).

⁵⁸ Fo. 64^v, col. b, 14–18 + 18–37 (= P fo. 28^r, 1–5 + fos. 28^r, 5–29^r, 6 = Keil, Gramm. Lat. vii.533–6). A later hand has attempted to mark a division in P. The end of the item is missing from C.

⁵⁹ See Cic. Pis. 73, Fam. 9. 10. 1, Hor. A.P. 447–50, Quintil. Inst. 1. 4. 3, Front. p. 180. 3–5 Van den Hout, Auson. Epist. 13. 30, Opusc. 13. 1. 11–15.

⁶⁰ See Sueton. Gramm. 24. 3, DS Virg. Aen. 1. 21, Serv. Virg. Aen. 10. 444.

⁶¹ See Epist. 106. 7, 112. 19, Interpr. Pent. praef pp. 64 ff. ed. Rom., lob pp. 69 ff., 74 ff., Psalt. pp. 3 f., Salom. p. 6, Augustine in Jer. Epist. 104. 3.

⁶² See Inst. 1. 1. 8, 1. 9. 3. For other uses of notae see Inst. 1 praef. 9, 1. 26.

⁶³ Correctors seem to have been responsible for the sporadic signs visible in the margins of texts of Juvenal's satires (Roberts, C. H., Journ. Egypt. Arch. 21 (1935), 199–207,Google Scholar E. A. Lowe, CLA Suppl. 1710) and Gaius' Institutiones (P.S.I. 1182 (CLA III 292)) written in Constantinople in the late fifth or early sixth century and transported to Egypt, as of a text of Hilarius' theological treatises written some where in South Italy during the sixth century and preserved there (cod. Vienna, Nat. Bibl. MS. 2160 (CLA X 1507)). Many more annotated Greek texts survive (see E. G. Turner, op. cit., pp. 113–18, Greek Manuscripts of the Ancient World (Oxford, 1971), p. 17)Google Scholar than do annotated Latin. There may be nothing in this. The number of unannotated Greek texts surviving is also far greater; against, for example, P. Oxy. 3224 (Hesiod, Erg. and (?) associated works: ll cent.) with its obelos, diple, asteriskos, and chi (see West, M. L. in The Oxyrrhynchus Papyri. Vol. xlv (London, 1977), pp. 51–2) must be placed fifty-three other ancient copies of the same set of Hesiod's poems apparently unannotated. In some cases the signs seem to have been copied from the same exemplar as the text, in others to have been added by a corrector. An expert study of all the evidence would be useful.Google Scholar

⁶⁴ None of the signs used by Flavius lulius Tryfonianus in correcting a text of Persius in the year 402 (see cod. Montpellier, Bibl. Med. 212, fo. 79^r) has survived in the medieval tradition. Some of Jerome's signs however are preserved (see A. Rahlfs, ‘Der Text des Septuaginta-Psalters’, Septuaginta-Studien 2 Heft (Göttingen, 1907), pp. 124–34). Features of cod. Rome, Bibl. Vat. Pal. Lat. 1615 (Plautus: XI cent.) have been interpreted as remnants of an ancient system of signs (see Schoell, F., T. Macci Plauti Truculentus (Leipzig, 1881), pp. xxxv-vi,Google ScholarLindsay, W. M., Ancient Editions of Plautus (Oxford, 1904), pp. 82–3).Google Scholar

⁶⁵ See L. Traube, Textgeschichte der Regula S. Benedicti, ed. 2 H. Plenkers (Munich, 1910 (Abh. d. hist. Cl. d. K. Bayer. Ak. d. Wiss. 25. 2)), pp. 65–6, 121, Lindsay, W. M., ‘Collectanea Varia II. Correction of MSS’, Palaeografica Latina Part II (Oxford, 1923), pp. 10–15.Google Scholar

⁶⁶ Isidore, Orig. 1. 21 (de notis sententiarum) was excerpted for cod. Paris, Bibl. Nat. Lat. 7530, fos. 154^v col. 2, 1–155v col. 2, 11, and for cod. Cava, Arch. della Badia di S. Trin. 3 (misc. XI-XII cent.), fos. 2471–248^v. On other accounts of signs in the latter codex see A. Reifferscheid, RhM 23 (1968), 127–33. On cod. Munich, Bayer. Staatsbibl. CLM 14429 (misc. X cent. from S. Emmeram in Regensburg) see Kettner, H., ‘Kritische Bemerkungen zu Varro and lateinischen Glossaren’, Progr. d. Klosterschule Rossleben 1868, pp. 33 ff.,Google ScholarWeber, P., Quaestionum Suetoniarum capita duo (Diss. Halle, 1903), pp. 8–13. The Bobbio library had a volume containing among various grammatical works a treatise De emendatione et notis ueterum librorum (item 102 of the 1461 inventory: see A. Peyron, M. Tullii Ciceronis Orationum pro Scauro, pro Tullio, et in Clodium fragmenta inedita … idem praefatus est de Bibliotheca Bobiensi, cuius inuentorium … edidit (Stuttgart and Tübingen, 1824), pp. 29–30).Google Scholar

⁶⁷ See above, p. 394 n. 46. One of the works excerpted for cod. Paris, Bibl. Nat. Lat. 7530 was Servius' commentary on Donatus' Ars grammatica (fos. 156^v-183^v).

⁶⁸ See in P itself fo. 40^r, 10–12 (at end of brief treatise on Horace's metres) Seruii grammatici scripsit d(e)o propitius papulus cons theyderichi indic. ii mensis februarii. xxv dies saturni hora iii die (i.e. 25 Feb. 674; see P. Lejay, RPb 18 (1894), 53–9); fo. 46^r, 3–4 EXPLICIT. Feliciter luliano scolastico sardiano Seruii grammatici maximae et antiquae Romae. On subscriptions copied in the traditions of Latin poets and prose-writers see Jahn, O., Ber. Verb. Sächs. Ges. d. Wiss. z. Leipzig, Phil.- hist. KI. 3 (1851), 327–72,Google ScholarBrink, C. O., Horace on Poetry. The ‘Ars Poetica’ (Cambridge, 1971), pp. 30–1Google Scholar, Zetzel, J. E. G., HSCPh 77 (1973), 225–43Google Scholar. For the same phenomenon in the Greek and Syriac bible traditions see Zuntz, G., The Ancestry of the Harklean New Testament (London, 1945Google Scholar (The British Academy. Suppl. Pap. 7)), pp. 11–33, ‘Die Subscriptionen der Syra Harclensis’, Zeitschr. d. Deutsch. Morgenl. Ges 101 (1951), 174–96Google Scholar, The Text of the Epistles (London, 1953), p. 72.Google Scholar

⁶⁹ On Philargyrius (or Filagrius) see Barwick, K., ‘De lunio Filargyrio Vergilii interprete’, Comm. philol. len. 8. 2 (1909), 1–66Google Scholar, Wessner, P., BPhW 29 (1909), 1366–71Google Scholar, 47 (1927), 454–60, 51 (1931), 206–9, Funaioli, G., Esegesi Virgiliana antica (Milan, 1930 (reprinting articles published between 1915 and 1920)).Google Scholar

⁷⁰ For Milan as the place in which Philargyrius taught see Schol. Bern. Virg. Georg. p. 169 Hagen, Funaioli, Esegesi, pp. 400–1.

⁷¹ So Schol. Bern. p. 145 Hagen.

⁷² Cf. DS Buc. 6. 6 ‘… fugatoque Asinio Pollione, ab Augusto Alfenum Varum legatum substitutum, qui transpadanae prouinciae et agris diuidendis praeesset: qui curauit ne ager, qui Vergilio restitutus fuerat, a ueteranis auferretur’; 9. 11 ‘“carmina” autem nonnulli quibus sibi Pollionem intercessorem apud Augustum conciliauerat, accipiunt: quo fugato, rursus de praediis suis fuerat Vergilius expulsus’; 9. 27 ‘sane blanditur Alfeno Varo, qui, Pollione fugato, legatus transpadanis praepositus est ab Augusto.’

⁷³ It was conceivably relevant at 6. 6–7 (where Philargyrius also has it), 9. 27, and 9. 35.

⁷⁴ Cf. DS Buc. 9. 35 ‘nonnulli sane Alfenum Varum uolunt, qui, licet iuris consultus et successor Servii Sulpicii esset, etiam carmina aliqua composuisse dicitur; sed hoc teste Horatio falsum est, qui Varium poetam laudat.’

⁷⁵ See above, p. 393, on the ambiguity of this genitive.

⁷⁶ A tragedy Thyestes, a Varus and Virgil are associated in a strange story at Servius, Buc. 3. 20 (cf. DS Buc. 6. 3). A Quintus Varus figures in yet another at [Acro], Hor. Epist. 1. 4. 3 (‘Quintus Varius’ according to Porphyrio). The statement at [Acro], Hor. Carm. 1. 6. 8 - ‘tragoediam Varus scripsit’ - may be simply the result of palaeographical error.

⁷⁷ The attribution to this tragedy of an anapaestic fragment cited by Philargyrius on Virg. Buc. 2. 70 (sic auarus … inquit codd. sic Varus … inquit J. H. Vossius) made by Weichert, A., Commentatio III de L. Vario poeta (Grimma, 1831), p. 20Google Scholar (cf. Bergk, Th., Commentatio de fragmentis Sophoclis (Leipzig, 1833), pp. 12–13) and now commonly accepted ought to be reconsidered against the general background of what we know about the ancient commentators.Google Scholar

⁷⁸ Some books of the Annales of Ennius were still available in the fifth century or perhaps even later to provide marginal supplements for a copy of Orosius' Historiae (on cod. St Gall, Stiftsbibl. 621 (IX cent.) and the eleventh-century notes on pp. 108, 143, 157 see Norden, E., Ennius and Vergilius. Kriegsbilder aus Roms grosser Zeit (Leipzig-Berlin, 1915), pp. 78–86)Google Scholar. I should guess that the extracts from Cornelius Nepos' De historicis Latinis on fo. l^r of cod. Wolfenbüttel, Herzog-August Bibl. Gud. class-Lat. 278 (Cic. Phil. XII cent. - see Marshall, P. K., The Manuscript Tradition of Cornelius Nepos (London, 1977. BICSL Suppl. 37). pp. 8–9) were originally made around this period.Google Scholar

⁷⁹ Cod. Rome, Bibl. Vat. Pal. Lat. 24 used bits of Seneca, Lucan, Hyginus, Fronto, Gellius, Livy, and Cicero manuscripts (CLA 168–77) to carry the Old Testament; St. Gall, Stiftsbibl. 912 bits of grammatical, medical and biblical manuscripts and a Terence manuscript (CLA VII 967–75) to carry the ‘Abba, abauus’ glossary; Milan, Bibl. Ambros. E 147 ord. sup. + Rome, Bibl. Vat. Lat. 5750 bits of Fronto, Symmachus, Persius and Juvenal manuscripts, a commentary on Cicero's orations and two obsolete Christian works (CLA I 26a-31) to carry the proceedings of the Council of Chalcedon.

⁸⁰ Cf. the ‘incipit’s on fos. 224^v and 280^r of the Ambrosian Plautus (Milan, Bibl. Ambros. G 82 ord. sup.) and some of those of the Medicean and Roman codices of Virgil (Florence, Bibl. Laur. 39. 1; Rome, Bibl. Vat. Lat. 3867).

⁸¹ Cf. fo. 593^r of the Ambrosian Plautus, where the metrical arguments were set by a later hand in the space left blank after the titulus and ‘didascalic’ note.

⁸² It may have been because the original texts had been only lightly washed that parts of pp. 11, 12, and the whole of pp. 78, 166 of cod. Rome, Bibl. Vat. Lat. 5750 were not reused. For a different explanation, however, see Ehrle, F., M. Cornelii Frontonis aliorumque reliquiae quae codice Vaticano 5750 rescripto continentur (Milan, 1906 (Codices e Vaticanis selecti phototypice expressi, VII)), pp. 7, 21.Google Scholar

⁸³ The words notae xxi quae uersibus apponi consuerunt look like a defective sentence belonging to the account rather than a title. The words F. (? = finit) de notis Probianis preceding EXPL. NOTAE very likely once formed a title (cf. the way in which the list of signs in cod. London, Brit. Libr. Harley 5693 is titled ).