^{page 150 note 1} References to Tacitus' Annals are to significance.

Book 4 unless otherwise stated.

^{page 150 note 2} Suet. Tib. 39–41. I have italicized those correspondences which seem of particular

^{page 150 note 3} maxime qualifies placuisse: cf. Gerber-Greef, Lex. Tac. i. 792, col. 2.

^{page 151 note 1} Tacitus does not make it clear that the episode at Spelunca (in (e) above) actually took place on the journey, but his introductory phraseology (illis diebus) closely resembles that of Suetonius, who does assign the incident to the journey (paucos post dies)— naturally, since Spelunca is in fact en route from Rome to Capua and Nola. The reason for Tacitus' obscurity on this point is doubtless to be attributed to his summary of the reasons for Tiberius' departure, an ‘insertion’ which somewhat disrupts the narrative (see Syme, Tacitus, 695).

^{page 151 note 2} Suetonius' slick comment at (c) above contrasts strongly with Tacitus' typically heavy philosophizing here, but the subjectivism of both authors would seem to have a common origin in some earlier source (see how their next statements both commence with nam (d)).

^{page 151 note 3} E. Koestermann's note on 67. 1 (p. 196 of vol. ii of his commentary) is the only helpful observation on the problem. On this point he says: ‘Tacitus begleitet nunmehr den Kaiser auf seiner Reise nach Capri und nimmt damit den cap. 58, 2 abgerissenen Faden wieder auf (he ought to have said 59. 2, cf. my n. 1 above).

^{page 151 note 4} Although Tiberius had spent a whole year in Campania before, A.D. 21–2 (cf. Tac. Ann. 3. 31. 1–64. 1). Koestermann (loc. cit) opts for an intermediate view: ‘Da sich die Erzählung in dem Zwischenstiick über zwei Jahre erstreckt, muB sich der Aufenthalt in Kampanien über langere Zeit am Ende des Jahres 26 ausgedehnt haben', a comment which, if I understand it correctly, would seem to be at variance with his next statement(quoted below, p. 152 n. 3).

^{page 152 note 1} Koestermann again: ‘Das cap. 67 Erzählte geht also zeitlich dem Einsturzungluck in Fidenae und dem Brand in Rom (cap. 62 ff.) voraus, wie denn auch mit dedicatis per Campaniam templis auf cap. 57, zuriickgegeiffen wird.’

^{page 152 note 2} e.g. at Ann. 12. 31–40, on which see (most conveniently) F. R. D. Goodyear, Tacitus, Greece & Rome, New Surveys in the Classics no. 4 (1970), 24, who refers to Kroll, W., Studien zum Verstdndnis der Literatur (Stuttgart, 1924), 371 ff.Google Scholar

^{page 152 note 3} ‘DaB aber der Historiker den zeitlichen Zusammenhang aus kompositionellen Gründen unterbrochen hat, geht aus Suet. Tib. 40 hervor, wo es heifit, Tiberius sei auf die Kunde von der Katastrophe in Fidenae nocheinmalaufdasFestlandzuruckgekehrt.’

^{page 152 note 4} This particular phrase strictly refers to Tiberius' previous sojourn in Rhodes, but we are clearly meant to understand it as referring to Capri also. Cf. next note.

^{page 152 note 5} This phrase, together with the one that immediately precedes it in our texts (cf. previous note), perhaps ought to follow on straight after 57. 1 saevitiam ac libidinem …locis occultantem; cf. Balsdon, J.P.V.D., CR lxi (1947)Google Scholar, 44 f., on the possibility of textual displacement.

^{page 152 note 6} See Lesky, A., History of Greek Literature(1966)Google Scholar, Index, s.v. ‘Ring composition'. rom.

^{page 152 note 7} For Latin instances see Williams, G., Tradition and Originality in Roman Poetry (1968)Google Scholar, Index, s.v. ‘Ring composition’.

^{page 152 note 8} See Adcock, F.E., Thucydides and his History (1963), 91Google Scholar f., who refers to R. Katicic, ‘Die Ringkomposition im ersten Buche des Thukydideischen Geschichtswerkes’, WS lxx (1957), 179–96Google Scholar. Cf. alsoHammond, N.G.L., ‘The Arrangement of Thought in the Proem and in other parts of Thucydides I’, CQ ii (1952), 127 ff.Google Scholar

^{page 153 note 1} Kleine Beiträge, ii. 594.

^{page 153 note 2} On this extended type of ring composition compare the remarks of Brink, C.O., Horace on Poetry, ii. 453 f.Google Scholar

^{page 153 note 3} Cf. W. A. A. van Otterlo, Untersuchungen über Begriff … der griech. Ringkomposition (1944), 3, ‘Der ganze Abschnitt … um- monorahmt … wird’, although the rest of his definition is more narrow than the type under discussion here. For sentence-framing in Latin prose (Livy), cf. Klingelhofer, H., Philol. Quart, iv (1925)Google Scholar, 321 ff.; in Latin poetry (Lucretius), Schrijvers, P.H., orror ac divina voluplas (1970), 154.Google Scholar Reference may also be made to Keaney, J.J., ‘Ring Composition in Aristotle's Athmaion Politeia’, AJPh xc (1969), 406–23Google Scholar, which I came across only after my paper had been Untersuchungen accepted for publication and which has Ringkomposition useful introductory remarks for those who cannot get hold of van Otterlo's monoragraph.

^{page 153 note 4} See the complicated internal chiastic structure explained by Fraenkel on Aesch. Ag. 205, and compare the structure of some Latin of Catullus' longer poems (especially Williams, Tradition and Originality).

^{page 153 note 5} procerum here picks up procerum at 63. largitio used of Tiberius here picks up verbally largitione at 63. 2, showing that the emperor's munificentia, like the activities of the proceres at 63. 2, is to be applauded as veterum institutis similis.

^{page 154 note 1} See Syme, Tacitus, 312 f., ‘Obituaries in Tacitus’, AJPh lxxix (1958)Google Scholar, 18 f., 30 f., (= Ten Studies in Tacitus (1970), 79^,89).

^{page 154 note 2} Nevertheless there is a sort of correspondence between chapters 61 and 66: each contrasts a pair of men, one of whom is claris maioribus, and in each case there is activities a distinguished nobilis who fails to live up to his family tradition. This surely confirms the care which Tacitus has put into the writing of this whole section.

^{page 154 note 3} The correspondences indicated on the left-hand side of the list denote thematic links; those on the right denots scale of treatment.

^{page 155 note 1} For patria carere in this sense cf., e.g., Cic. Att, 3. 26, Val. Max. 3. 8. 4. For adsidere see the remark of Koestermann here, ‘An unsere Stelle klingt also die Bedeutung ,, feindlich belagern” mit unter’, and to his fit examples of the verb in this sense now add Lyne, R.O.A.M., Latomus xxviii (1969), 6g4 ff.Google Scholar

^{page 155 note 2} See Walsh, P.G., Livy (1961),Google Scholar 191 ff., and compare Virg. Aen. 2. 746 quid in eversa vidi crudelius urbe ?

^{page 155 note 3} e.g. Cic. Dom. 37. 98 ea quae capta urbe accidunt victis, Sail. C. 52. 4 capta urbe nihil fit reliqui victis, H. 1. qospeciem captae urbis efficere, Catull. 62. 24 quid faciunt hostes capta crudelius urbe?, Ov. Met. 12. 225 captaeque erat urbis imago.

^{page 156 note 1} Quint. 8. 3. 67-70, cf. Hermogenes in Rhet. Grace, ed. Spengel, ii. 16, Rhet. ad Herenn. 4. 39. 51 (and the parallels there cited by H. Caplan in the Loeb edn., p. 358 n.).

^{page 156 note 2} Quintilian mentions: the crash of falling roofs (ruentium tectorum fragor), the confusion (ex diversis clamoribus unus quidam sonus, aliorumfuga incerta), the clinging to relatives (alii extremo complexu suorum cohaerentes), the wailing of women and children (infantium feminarumque ploratus), the cruelty of fate (male usque in ilium diem servati fato senes). Tacitus mentions: the falling structure (conferta mole, dein convulsa, dum ruit intus aut in exteriora effunditur, … praeceps trahit atque operit), the confusion (nequedum comperto quos ilia vis perculisset, latior ex incerto metus), the presence of women and children (virile ac muliebre secus [an impressive phrase, cf. Tränkle, H., WS lxxxi (1968), 128Google Scholar], omnis aetas, … immensamque vim mortalium), their wailing (per diem visu, per noctem ululatibus et gemitu coniuges aut liberos noscebant), the loss of relatives (iam ceteri fama exciti, hie fratrem, propinquum Me, alius parentes lamentari), the cruelty of fate (quos principium stragis in mortem adflixerat, ut tali sorte, cruciatum effugere: miserandi magis quos abrupta parte corporis nondum vita deseruerat).

^{page 156 note 3} Curt. 8. 3. 13 confuderat oris exsanguis notas pallor nee quis esset nosci satis poterat, Sen. Tro. 1114 ff. signa clari corporis et ora et Mas nobiles patris notas confudit imam pondus ad terram datum.

^{page 156 note 4} In the famous digression in this book, 32. 1 (expugnationes urbium).

^{page 156 note 5} This is a somewhat different appraisal from that of Koestermann, who says: ‘Mit the in quos wenigen Strichen hat Tacitus so ein aussergewohnlich lebendiges Bild der turbulenten Szenen entworfen. Die Erzahlung, obwohl nicht frei von rhetorischen Elementen ….’. Koestermann also invites us to compare Plin. Ep. 6. 20, on the eruption of Vesuvius. This letter was actually written to Tacitus, providing first-hand source-material for a part of his Histories, now lost. It would perhaps be attractive to imagine that when Tacitus came to write about Fidenae he re-utilized the account of Vesuvius which Pliny had sent on request a few years earlier. But this romantic view cannot be contemplated because Pliny, like his hero Cicero (Alt. 2. 1. 1–2), elaborates even the factual fundamentals to make it look like ‘real' (i.e. rhetorical) history (cf. H. W. Traub, ‘Pliny's Treatment of History', TAPA lxxxvi [1955], 213 ff., esp. 229–31). Both Pliny and Tacitus, despite their different genres, are working within a common tradition. [An analogous case, also involving Pliny and Tacitus, can be found elsewhere in this book of the Annals. At Ep. 1. 20 Pliny writes to Tacitus on the question of style, saying (12): plerumque parvae res maximas trahunt. At chapters 32–3 Tacitus has a famous digression on historical style, saying (32. 2) that: non tamen sine usu fuerit introspicere ilia primo aspectu levia, ex quis magnarum saepe rerum motus oriuntur. The sentiments are identical; but they constitute a commonplace occurring first in Aristotle, Politics 5. 3. 1 (not 5. 4. 1, as Furneaux says, nor even ‘Polyb. 5. 4. i’, as Koestermann curiously notes), and then in Caes. BC 3. 68. 1 and Liv. 27. 9. 1. Compare also Aeschin. Tim. 4 Jin.'].

^{page 157 note 1} On which cf. Peter, H., Die geschichtliche Litteratur über die röm. Kaiserzeit (Leipzig, 1897), iGoogle Scholar. 108 ff. For further cliche in this section (64. 1 fortuita ad culpam trahentes), cf. Cic. Verr. 5. 131, Leg. Man. 10, Pis. 43, Rab. Post. 29, Veil. 118. 4, Sen. Clem. 1. 2. 1.

^{page 157 note 2} praeter spent evasit, as Suetonius remarks (Tib. 39).

^{page 158 note 1} The pathos is driven home at 63. 1 where Tacitus tells us that the consequence of the collapse of the amphitheatre was even fewer games. This is no less ironical than pathetic: it almost vindicates the emperor's distaste for such spectacles.

^{page 158 note 2} See above, p. 153, and compare, e.g., A. 1. 72. 2 non tamen ideo faciebat fidem civilis animi; nam legem maiestatis reduxerat, another typically Tacitean link-sentence.

^{page 158 note 3} Tacitus, 624 n. 3.

^{page 158 note 4} I am extremely grateful to Mr. R. H. Martin and Dr. T. J. Saunders for their helpful comments on an earlier draft of this paper.