1 Tusa, S. and Royal, J., “The landscape of the naval battle at the Egadi Islands (241 B.C.),” JRA 25 (2012) 7–48 Google Scholar.

2 Ancient account in Polyb. 1.59-61, cf. Diod. Sic. 24.11, Zonar. 8.17, Eutrop. 2.27, Oros. 4.10.5-8. For modern accounts, see Sanctis, G. De, Storia dei Romani vol. III Google Scholar. L'età delle guerre puniche. Parte I (Turin 1916) 184-87 and 264–67Google Scholar; Lazenby, J. F., The First Punic War: a military history (Stanford, CA 1996) 150-57Google Scholar. Eutropius (2.27.2) provides the date, on which see Walbank, F. W., A historical commentary on Polybius, vol. 1 (corr. repr., Oxford 1970) 124–25Google Scholar, and Morgan, M. G., “Calendars and chronology in the First Punic War,” Chiron 7 (1977) 109–12Google Scholar (both accept the near-equivalence of the Julian and Roman calendars at this point and the probable accuracy of the date). All dates in this article are B.C. unless otherwise stated.

3 Many of these aspects have already been debated in a workshop held at Oxford on April 8, 2013, and in an AIA-APA panel held at Chicago on January 3, 2014; a set of papers originating in those meetings is now being prepared for publication.

4 By P. Schmitz (Eastern Michigan University, Ypsilanti); for now, see Tusa and Royal (supra n.1) 43.

5 Tusa and Royal ibid. 36-39.

6 Ibid. 15-16.

7 Ibid. 24.

8 Ibid. 14 n.17.

9 Gnoli, T., Navalia. Guerre e commerci nel Mediterraneo romano (Rome 2012) 81–82 Google Scholar, also reports texts of Egadi 1, 3-4, 6-8, but without further detail, due to the limited information available at that time; a number of his readings and expansions can now be emended.

10 See id., “La battaglia delle Egadi. A proposito di ritrovamenti recenti,” RivStorAnt 41 (2011) 47-86 (historical); “Nuova iscrizione su un rostro proveniente dalla battaglia delle Egadi,” Epigraphica 74 (2012) 59–74 (epigraphic)Google Scholar; “Petronio e Nevio: nota a Petr. 30,1 s.,” Eikasmos 23 (2012) 249–58 (literary)Google Scholar. Further discussion now in M. Torelli, “Una trireme da Cosa. Il rostro scritto delle Egadi e il ruolo delle colonie latine nella flotta romana,” Ostraka 20 (2011 [2013]) 273-77.

11 Tusa and Royal (supra n.1) 45 n.92.

12 The “Acqualadroni ram” was found just off Capo Rasocolmo in September 2008. Radiocarbon dating of timbers preserved inside the ram, while not conclusive, suggests a date for the ship's activity of the mid-2nd c. B.C. or after: Buccellato, C. A. and Tusa, S., “The Acqualadroni ram recovered near the Strait of Messina, Sicily: dimensions, timbers, iconography and historical context,” IJNA 42 (2013) 76–86 CrossRefGoogle Scholar.

13 For a lucid summary of the casting process for such rams (by the ‘lost-wax’ method, in a single pour) see Murray, W. M., The age of Titans: the rise and fall of the great Hellenistic navies (Oxford 2012) 35–38 CrossRefGoogle Scholar.

14 The fact that such a size implies a smaller class of ship than the quinquiremes which dominate Polybius' account of the battle, and the war, is a problem that must remain outside the limits of this article.

15 The numbering of these groups is my own, as is the order in which I present them.

16 I am very grateful to J. Royal for a detailed report of the text on Egadi 10 immediately upon recovery, enabling its inclusion within this discussion.

17 For a contemporary example, see the engraved image on the Entella IV bronze inscription from W Sicily (SEG 30.1120; photograph in Ampolo, C., Da un'antica città di Sicilia. I decreti di Entella e Nakone [Pisa 2001] 18)Google Scholar.

18 See list of instances in ILLRP II, p. 508 Google Scholar, index, s.v. declinatio secunda.

19 The approximate form is described briefly as a cursive type by Hübner, E., Exempla scripturae epigraphicae latinae a Caesaris dictatoris morte ad aetatem Iustiniani (Berlin 1885) lxiv Google Scholar, but none of the 3rd-c. A.D. examples he offers bear much similarity to these instances; I have found no Republican period parallel for the form.

20 See the brief observations by C. Hallett and T. Hölscher in Tusa and Royal (supra n.1) 42-43.

21 I am indebted to J. Royal for initial discussion.

22 For Paperius, see, e.g., ILLRP 473 (Aeclanum, 123 B.C.), 475 (Carthage, later copy of text of 121-119 B.C.); cf. Papeirius in ILLRP 720 (Capua, 84 B.C.), 759 (Delos, 113 B.C.); and see Schulze, W., Zur Geschichte lateinischer Eigennamen (Berlin 1933) 86 Google Scholar. For Populicius, see, e.g., ILLRP 35 (3rd c. B.C., Rome)Google Scholar = Degrassi, ILLRP Imagines no. 22a, and Schulze ibid. 414 n.1; Wiseman, T. P., Remembering the Roman people (Oxford 2009) 1 n.1CrossRefGoogle Scholar; cf. the intermediate later forms Poblicius (e.g., ILLRP 268 and 1272, 1st c. B.C.) and Poplicius (ILLRP 357 [Rome] and ILLRP 777 = Ariño, B. Díaz, Epigrafia latina republicana de Hispania [Barcelona 2008] C50 Google Scholar, Carthago Nova, both of the 2nd/1st c. B.C.).

23 See however (all from Italy, not Rome) ILLRP 466 (from Cereatae Marianae near Arpinum), 599 (from Frigento, Avellino); 605 (from Furfo in the territory of the Vestini); and 620 (Larinum).

24 See Gordon, J. S. and Gordon, A. E., Contributions to the palaeography of Latin inscriptions (Berkeley, CA 1957) 183 Google Scholar, for the rarity of interpunct at line-end.

25 See ILLRP 89, 304, 539, 543, 619, 641, 655, 666, 678, 683, 689 (based upon a survey of ILLRP nos. 10-695 [some of these inscriptions are now lost, so precise details of the layout are sometimes dependent upon earlier reports]).

26 In ILLRP 304 it apparently had a symbol comparable to a curly brace preceding (cf. CIL IX 5019); in ILLRP 619 it is preceded by an interpunct (photograph in CIL I² 1905); in ILLRP 641 it is in much larger letters ( Ritschl, F., Priscae latinitatis monumenta epigraphica [Berlin 1862] pl. 89.e)Google Scholar; in ILLRP 666 it is preceded by an interpunct (ibid. pl. 90.b).

27 ILLRP 539 (drawing in Brusin, G., Gli scavi di Aquileia; un quadriennio di attività dell'Associazione nazionale per Aquileia (1929-1932) [Udine 1934] 57)Google Scholar; ILLRP 641 (Ritschl ibid. pl. 89.e); ILLRP 655 (ibid. pl. 53.a).

28 On punctuation use, see Gordon and Gordon (supra n.24) 183-84; on the various types of interpuncts (round dots, as here, are both the earliest and most common in Republican times), see Zucca, R., “Sui tipi di interpunzione nelle iscrizioni latine dall'età più antica alla fine della Repubblica,” Miscellanea greca e romana 18 (1994) 123–50Google Scholar.

29 E.g., probave(re) in ILLRP 539; probaver(e) in CIL I² 2949a; probaver(unt) in ILLRP 562a.

30 See especially Gnoli, , Epigraphica 2012 Google Scholar and Eikasmos 2012 (both supra n.10).

31 A comparison of instances of N and M within the first two Scipionic elogia of the mid-3rd c. will suffice to make this clear: ILLRP 309 and 310, Ritschl (supra n.26) pls. 37-38.

32 P. Campbell (Univ. of Southampton) undertook highlight RTI capture of Egadi 1 in the summer of 2013 and kindly made the imagery available to me. The focus of that exercise was not the inscriptions but the ram as a whole, and the imagery does not permit one to enhance the readings presented here. Indeed, it is striking that on this occasion the best colour photograph serves better than the RTI data.

33 See Leumann, M., Lateinische Laut- und Formenlehre (Munich 1977) 65 para. 73Google Scholar: the form in -oe- is not attested before 113 B.C. (for which see ILLRP 705 of 112 or 111 B.C.).

34 E.g., ILLRP 187 for IIIvir, ILLRP 324 for triu(m)vir, both for Roman officials in the 3rd/2nd c.; also ILLRP 121 and 181 for duo(m)vir. The variation is widely attested in municipal epigraphy of the Republican period (see the numerous examples of both duoviri and IIviri in ILLRP 519-695).

35 Leumann (supra n.33) 266 notes Cato, ORF4 8.113 (si trium virum sim), Varro, RR 1.2.10 (Scrofam, vigintivirum qui fuit), Aul. Gell., NA 13.12.6 (ego trium-virum), Cic., Att. 2.6.1 (ubi quidem ego mallem duumvirum quam Romae fuisse).

36 Omission of final -m is commonplace – for example, specifically from the genitive plural in the legend ROMANO(m) on Rome's early coin issues of the 3rd c. (see many among RRC nos. 1-22).

37 Compare, e.g., the Gracchan termini, IIIvir(ei), in ILLRP 467-74, or the frequent municipal instances of IIvir(i) and IIIIvir(i) in ILLRP 519-695.

38 Pace Gnoli, Epigraphica 2012 (supra n.10) 67-68, it is entirely unnecessary to emend away the O on the basis that it forms an erroneous application of a dative ending to vir (suggested to be “frutto di una sorta di ‘attrazione’ esercitata dai nominativi in -o(s) che precedono”). Manganaro, G., ‘Tre note di storia e di epigrafia della Sicilia,” Epigraphica 75 (2013) 9–32 Google Scholar, at 27 n.66 offers the following reading of Egadi 1: C(aius) Sestio(s) P(ubli) f(ilius) [cum] ∣ Q(uinto) Salonio Q(uinti) [f(ilio)] ∣ sex. viro en[bol(um)] ∣ probave[re]. This is ingenious, but implausible.

39 Gnoli ibid. 69 with id. Eikasmos 2012 (supra n.10).

40 I repeat the suggestion I offered in “Cave navem,” CQ 56 (2006) 539 Google Scholar, that Petronius' choice of the unparalleled term embolum is surely deliberate, but not as a hypothetical Naevian allusion; rather, it is both an affected Grecism on the part of Encolpius and a deliberate echo of or response to the language of the freedmen in the Cena, where rostrum is instead used to mean chin or face.

41 For what it is worth, based principally upon the index of CIL I² ii.1-3, epigraphic instances of probatio by Roman officials are twice as likely to leave the object implicit as explict (9 implicit, 4 explicit); vice versa, across the rest of Italy (40 instances), explicit outnumber implicit by 4 to 1.

42 Prag in Tusa and Royal (supra n.1) 43-44 n.88.

43 Torelli (supra n.10) 275 (modelled on the duumviri navales classis ornandae reficiendaeque causa of Livy 9.30).

44 Mommsen, Th., Römisches Staatsrecht (3rd edn., Leipzig 1887–1888) vol. II.i, 662-74Google Scholar for ad hoc minor magistracies of this sort (at 667: “Ohne Zweifel kennen wir von diesen gewiss sehr zahlreichen und meist politisch bedeutungslosen Magistraturen nur einen verhältnissmässig sehr kleinen Theil”).

45 See Leumann (supra n.33) 227 section 229, and 423 section 352; R. Wachter, Altlateinische Inschriften. Sprachliche und epigraphische Untersuchungen zu den Dokumenten bis etwa 150 v.Chr (Bern 1987) 309 section 131 (CIL I² 7 for the Barbatus elogium with -us, followed chronologically by CIL I² 608/9 of 211 B.C.). For the methodological point, see Wachter ibid. 356-57, quoted below in n.48.

46 Leumann (supra n.33) 67 section 76; Wachter ibid. 357-58 section 159 (CIL I² 616 and 617 the earliest instances of -ae-).

47 Adams, J. N., Social variation and the Latin language (Cambridge 2013) 42 CrossRefGoogle Scholar (late use); Leumann (supra n.33) 606 section 443.3, and 513 section 394 (the transition from -ed, to -et, to -it); CIL I² 25 (ornavet) and Wachter ibid. 359-61 section 161 (on the Duilius text). I am grateful to R. Wachter for drawing my attention to this point.

48 Wachter ibid. 356-57 section 159a: “Die hier betrachteten Inschriften, die ein zufälliges Teilcorpus bilden und praktisch nur die stadtrömische Herkunft gemeinsam haben, zeigen in ihren Merkmalen zahlreiche Kreuzungen moderner und altertümlicher Merkmale. Darin verrät sich eindeutig die Erscheinung der archaisierenden Schreibung, Da wir mit solchen Verfälschungen aber auch in praktisch allen anderen Zeugnissen zu rechnen haben, ist es unmöglich, aus derartigen Merkmalen eine relative oder absolute Datierung der Inschriften zu erreichen”. Compare ibid. 350 for a Table comparing various texts of this period and the wholly random variation in forms that is visible.

49 Ibid. 323 n.767 on the variety within the elogia of Barbatus and his son (ILLRP 309-10 = CIL I² 6-9 = Ritschl [supra n.26] pls. 37-38).

50 See already the observations of Hübner (supra n.19) liii-liv on A, and lxiii-lxiv on P. Some examples: A broken bar Degrassi, ILLRP Imagines nos. 102 (4th c.), 187 (3rd c.), no. 189 (2nd c.); A unattached slanting, ibid. nos. 35, 37, 65 (all 3rd c.), 58, 100 (3rd-2nd c.); A straight bar, ibid. nos. 51, 81 (3rd c.), 22a (3rd/2nd c.); P square open, ibid. nos. 34, 41, 133 (3rd c.), 21, 63 (3rd/2nd c.), 54, 64 (2nd c.); P (partly) rounded and open, ibid. nos. 133, 93b (3rd c.), 22a (3rd/2nd c.); P rounded and closed, ibid. nos. 27, 65 (3rd c.).

51 The attempt to suggest dates within a period of 10-20 years and to construct a relative dating of these texts on the basis of letter-forms in Gnoli (supra n.9) 84-86 is unsustainable.

52 CIL I² 18, 22, 24 and 610.

53 Conveniently, Salway, B., “What's in a name? A survey of Roman onomastic practice from c. 700 B.C. to A.D. 700,” JRS 84 (1994) 127 Google Scholar.

54 Stemmata by Gündel, H. in RE XXIV, cols. 991 and 995Google Scholar.

55 Münzer, F., Roman aristocratic parties and families ((transl. Ridley, T.; Baltimore, MD 1999) 115 with Table 6 on 117Google Scholar; Gündel, (RE XXIV, col. 1039, s.v. Quinctius no. 40)Google Scholar had a similar view, but likewise chose to leave the man unattached. On the Livian list, see Briscoe, J., A commentary on Livy, Books 41-45 (Oxford 2012) 59 Google Scholar (arguing that there is little reason to doubt the first four names).

56 Badian, E., “The family and early career of T. Quinctius Flamininus,” JRS 61 (1971) 106 Google Scholar.

57 Brennan, T. C., The praetorship in the Roman Republic (Oxford 2000) 734 and 898 n.83Google Scholar (explicitly after Badian).

58 See conveniently Salomies, O., Die römischen Vornamen. Studien zur römischen Namengebung (Comm. Hum. Lit. 82; Helsinki 1987) 26–27 Google Scholar; texts such as CIL IX 4363 Google Scholar (Amiternum) containing both Kaeso and Gaius (C. Plotius C. f. K. n.) make the point very clear.

59 See the entries under ‘Publicius’ in RE and in the index of careers in Broughton, T. R. S., The magistrates of the Roman Republic (Atlanta, GA 1951–1986) [henceforth MRR]Google Scholar; the dedication is ILLRP 35.

60 Pliny (NH 18.286) dates him to 238 B.C., Velleius Paterculus (1.14.8) to 241 B.C.; the numeral in one must be corrupt, but which is unknowable.

61 Mommsen (supra n.44) vol. I, 524-26 (especially 526 n.2).

62 A. E. Astin, The Lex Annalis before Sulla (Coll. Latomus 32. 1958) for the codification of the Republican cursus.

63 Tusa and Royal (supra n.1) 45.

64 Münzer (supra n.55) 152.

65 Münzer (ibid. 105-9) has a brief sketch of the family, including stemma of the Papirii Masones; cf. RE s.v. Papirius, and Broughton, , MRR vol. II Google Scholar, Index of careers, s.v. Papirius.

66 The praenomen is clear on the stone of the Fasti Capitolini; Diod. Sic. 15.50.1 has T(itus), while Livy 6.27.2 has P(ublius); cf. Broughton, , MRR vol. I, 106 n.1Google Scholar.

67 See RE s.v. Sestius; Broughton, , MRR vol. II Google Scholar, Index of careers, s.v. Sestius.

68 Torelli (supra n.10) 277; cf. D'Arms, J. H., Commerce and social standing in ancient Rome (Cambridge, MA 1981) 55–62 CrossRefGoogle Scholar with further references.

69 Sources in Broughton, , MRR vol. 1 Google Scholar; although listed in Pauly's first edition, the Salonii were omitted from the revised edition of RE (briefly noted in Der kleine Pauly).

70 Plut., Cat. Mai. 24.2 (ὑπογραμματεύς), cf. 27.7; Aul. Gell. 13.20.7-8 (cliens); Plin., NH 7.61-62 (filia clientis); Solin. 1.59; Aur. Vict., De vir. ill. 47.9.

71 Astin, A. E., Cato the Censor (Oxford 1978) 105 Google Scholar, assumes that “she was therefore probably the daughter of one of his freedmen”, although he later acknowledges that this is merely inference (263 n.67: “Salonius is not actually termed a freedman but his former occupation virtually guarantees that he was”); cf. Münzer (supra n.55) 301-2.

72 Torelli (supra n.10) 277 has also suggested a Cosan origin for P. Salonius Q. f. of Egadi 1 (to go with that claimed for Sestius), but he only offers later evidence for the name from Tarquinia and the Po valley and makes no mention of the Roman evidence for the family.

73 I discuss this problem in more detail in “The quaestorship in the third and second centuries BC,” in Dubouloz, J., Pittia, S. et Sabatini, G. (edd.), L Imperium Romanum en perspective. Les savoirs d'empire dans la République romaine et leur héritage dans l'Europe médiévale et moderne (Besançon 2014) 193–209 Google Scholar.

74 In 294, L. Opimius Pansa, killed fighting the Samnites (Liv. 10.32.9); in 249, 2 or more anonymous quaestors temporarily in charge of a fleet (Polyb. 1.52.4-54.8); in c.237, Q. Fabius Maximus Verrucosus held the office twice (Insc.It. 13.3.8); in c.230, Cn. Octavius Rufus (Suet., Aug. 2.2); in 229, anonymous quaestor killed in Illyria (Polyb. 2.11.13); in c.222, C. Terentius Varro (Liv. 22.26.3). There has been no attempt to collect the fasti of the quaestorship in the last 100 years, other than Broughton (supra n.59), and he did not include anonymous office-holders.

75 The most satisfactory discussion remains Harris, W. V., “The development of the quaestorship, 267-81 B.C.,” CQ 26 (1976) 92–106 CrossRefGoogle Scholar; cf. Loreto, I., “Sull'introduzione e la competenza originaria dei secondi quattro questori (ca. 267-10 a.C.),” Historia 62 (1993) 494–502 Google Scholar.

76 I argue (supra n.73) that there is no basis for assuming further increase, and show that the Roman state could have functioned quite satisfactorily with 8 throughout the 2nd c.

77 This is essentially the suggestion of Gnoli (supra n.9) 86-97, with earlier bibliography: he follows Harris and others in emending the number in Lydus from 12 to 2, while arguing that classici is merely a description of function and not a reflection of a formal title, and therefore that Lydus' text embodies a kernel of truth recalling the extensive delegation of naval duties to the new quaestors in the First Punic War (cf. Ferone, C., “Lido, De magistratibus, I.27 e la politica navale di Roma nel III sec. a.C.,” Klio 85 [2003] 70–81)CrossRefGoogle Scholar. This is quite plausible; my point is only that the rostra inscriptions cannot prove or disprove such a solution. See Mommsen (supra n.44) vol. II.i, 570-73 for the ‘Italian’ quaestors, and Prag (supra n.73) for the general weakness of the construction.

78 Duumviri navales (recent discussion in Dart, C., “ Duumuiri nauales and the navy of the Roman Republic,” Latomus 71 [2012] 1000–14)Google Scholar are explicitly attested in 311 (Liv. 9.30.3-4), in 282 (Liv. Per. 12), in 181 (Liv. 40.18.7-8, 40.26.8), in 180 (Liv. 40.42.8), in 178 (Liv. 41.1.2-3), and in 176 (Liv. 41.17.7). Consular oversight from the time of the First Punic War is regularly attested, as Day, S. (Fleets and prouinciae in the Roman Republic: institutions, administration and the conceptualisation of empire between 260 and 49 B.C. [D.Phil., University of Oxford 2014] chapts. 1-2)Google Scholar has now demonstrated, beginning with C. Duilius in 260 (CIL I² 25 with add. p. 861-62); in 254 (Polyb. 1.38.6-7); in 250 (Polyb. 1.39.15); in 249 (Zonar. 8.15); and in 218 (Liv. 21.17.2-4, Polyb. 3.41.1-3) (I owe these references to S. Day).

79 Another inscription (CIL I² 808 = ILLRP 465 = Giovagnoli apud R. Friggeri, M. G. Granino Cecere and G. L. Gregori [edd.], Terme di Diocleziano. La collezione epigrafica [Milan 2012] 207-9) records the probatio of roadworks in the Late Republic by a quaestor urbanus, T. Vibius Temudinus, but this is in his distinct capacity as curator viarum, and the quaestorship is merely coincidental.

80 Mommsen (supra n.44) vol. II.i, 426 with n.5; also at 554-55 and 561 (in line with this he asserts [431] that the locatio of the Sicilian grain tithes by Roman quaestors in Sicily was wholly exceptional); cf. Trisciuoglio, A., “‘Sarta tecta, ultrotributa, opus publicum faciendum locare’”. Sugli appalti relativi alle opere pubbliche nell'età repubblicana e augustea (Naples 1998) 117–47, especially 137 n.95Google Scholar, where he concedes that quaestors could sometimes conduct locatio, but relegates this to a footnote within his final category of “Magistrati straordinari”, likewise basing such conclusions upon the lack of evidence.

81 Trisciuoglio (ibid. 117-31) discusses the extent to which even censorial locatio was to some degree consequent upon senatorial initative.

82 Mommsen (supra n.44), vol. II.i, 561, notes specifically the quaestor's place in this regard, not least in his rôle as military adjunct to the consul.

83 Trisciuoglio (supra n.80) 136-38; Mommsen ibid. II.i, 667-71.

84 Still pertinent are the comments of Badian, E. in Publicans and sinners: private enterprise in the service of the Roman Republic (Ithaca, NY 1983)Google Scholar, especially at the top of 21 on the centrality of contracting but also on the unsurprising lack of evidence prior to the Second Punic War. See CIL I² 18, 22, 24 and 610 for other 3rd-c. instances of probatio on the part of aediles and consuls.

85 Tusa and Royal (supra n.1) 18 n.21; cf. Murray (supra n.13) 31 and 35, with references to the evidence of the Athlit ram. In “The weight of trireme rams and the price of bronze in fourth-century Athens,” GRBS 26 (1985) 141–50Google Scholar, the same author discusses the Athenian evidence for cost and curation of trireme rams in the 4th c. B.C.

86 Polyb. 6.13.1, cf. 23.14.5, and Crawford, M. H., Roman Republican coinage (Cambridge 1974) vol. II, 616 Google Scholar.

87 Crawford (ibid. vol. II, 617) notes Livy 32.2.1-2 (199 B.C.), which suggests that the quaestors had responsibility for checking the purity of incoming bullion.

88 Crawford ibid. vol. II, 617, with n.1; Cicero, Flacc. 30, shows general fleet expenditure being made through the two urban quaestors in 61 B.C.

89 See Crawford ibid. vol. I, 313, for a single urban quaestor having responsibility for the aerarium.

90 Trisciuoglio (supra n.80) 155-56 brings out this point clearly; the naming of roads after the magistrate responsible is perhaps the most conspicuous example; but note especially the claim in the Duilius elogium of 260 B.C. (ILLRP 319, 11.5-7): Enque eodem mac[istratud bene r]em navebos marid consol primos c[eset copiasque c]lasesque navales primos ornavet pa[ravetque].

91 In the years 260 (Polyb. 1.20.9-21.3), 257/56 (1.25.5-7), 254 (1.38.5-7), 250 (1.39.15), 249 (1.52.4-8 with Zonar. 8.15), and 242 (1.59.6-8).

92 Cf. Tusa and Royal (supra n.1) 45.

93 Livy 26.39.5: postremo ipse (sc. Quinctius) a sociis Reginisque et a Velia et a Paesto debitas ex foedere exigendo classem viginti navium … effecit; cf. Torelli (supra n.10) 274 n.12.