^{page 158 note 1} Dionysius VII, 72. ἵνα ϕανερὰ γίνοιτο τοῖς ξένοις ἡ μέλλουσα ἀνδροῦσθαι τῆς πόλϵως ἀκμὴ πλῆθός τϵ καὶ κάλλος οἴα τις ῆν. Piganiol's, recent discussion, Recherches sur les jeux romains (Strasbourg, 1923) pp. 15–31Google Scholar, has contributed much to the understanding of Dionysius's account, though his explanation of the origin of the games seems unnecessarily complicated. The youths whose fathers had the equestrian census were, as Gelzer has shown, the group from whom, even after the Licinian laws, the officers for the Roman legions and the Roman magistrates were drawn. See Die Nobilität der römischen Republik, Liepzig, Teubner, 1912Google Scholar.

^{page 158 note 2} Cicero, Pro Caelio II: “Nobis quidem olim annus erat unus ad cohibendum bracchium toga constitutus, et ut exercitatione ludoque campestri tunicati uteremur eademque erat, si statim merere stipendia coeperamus, castrensis ratio ac militaris.

^{page 158 note 3} Rostovtzeff, , ‘Römische Bleitesserae,’ Klio, Beiheft, iii (1905), pp. 59Google Scholar ff. It is impossible to record all my indebtedness to this epoch-making study. I should like also to express my personal obligation to Professor Rostovtzeff for calling my attention to monuments and to recent investigations which have been of value for this study.

^{page 159 note 1} Aen. vii, 162 ff; ix, 601 ft. See Rostovtzeff, op. cit. pp. 62 ff; Norden, , ‘Vergil's Aeneis im Lichte seiner Zeit,’ Neue Jahrbücher, vii (1901), p. 263, no. 1Google Scholar.

^{page 159 note 2} Cf. also Horace, C. iii, 12Google Scholar and iii, 24, 51 ff, with Rostovtzeff's comments, l.c.

^{page 159 note 3} On the significance of this title see Mommsen, , Staatsrecht, ii, pp. 826Google Scholar ff; Res Gestae Divi Aug. pp. 44 ff.

Monumentum Ancyranum, xiv, 5 : ‘Equites [a]utem Romani universi principem iuventutis utrumque eorum parm[is] et hastis argenteis donatum appellaverunt.’

^{page 160 note 1} Suet. Aug. 38: ‘Liberis senatorum quo celerius rei publicae assuescerent protinus a virili toga latum clavum induere et curiae interesse permisit.’

^{page 160 note 2} On the position of the young men of senatorial families see Mommsen, , Staatsrecht, iii, pp. 470, 486Google Scholar; Rostovtzeff, op. cit. p. 61. Mommsen held the view that the young senatorials had probably been included in the equestrian order only since the time of Sulla. But the evidence presented by Gelzer shows that the relationship between the two groups goes back to the earliest times. See also Soltau, , ‘Reiter, Ritter und Ritterstand’ in Zeitschr. für. oestr. Gymnasien, lxii (1911), pp. 384 ff, 481 ff, 577Google Scholar ff. Soltau, however, goes beyond the evidence on a number of points.

On the question of the relation between the equites and the iuvenes it is often asserted that the terms are practically equivalent as used in the Augustan Age. But a study of the use of the words iuventus and iuvenes in Augustan and pre-Augustan literature (for the occurrences of the words in Cicero's letters I am indebted to my colleague, Professor Catharine Saunders) shows that such is not the case. The cavalry was, however, from the earliest times regarded as the foremost group of the iuventus. On several occasions in Livy the entire cavalry force of citizens is referred to as principes iuventutis (ix, 14, 16; xlii, 61, 5), primores iuvenum (vii, 10, 1), proceres iuventutis (ii, 20, 11). Only after the equestrian order had long been established as the representatives of the able-bodied manhood could the princeps iuventutis be rendered in Greek as πρόκριτος τῆς ἱππάδος (Dio lxxi, 35).

^{page 160 note 3} The interpretation of these tesserae is perhaps the most important of Rostovtzeff's many contributions to this subject. See his Tesserarium urbis Romae et suburbi plumbearum Sylloge, St. Petersburg, 1903.

^{page 160 note 4} These exercises are never referred to in the original text of ancient writers as the ludus Troiae, though modern authorities usually refer to them under that title. On the lusus Troiae see Goebel, , De Troiae ludo, Düren, 1852Google Scholar; Marquardt-Friedländer, , Staatsverwaltung ², iii, pp. 525Google Scholar ff; Toutain, s.v. Troia, Troiae ludus, Daremberg et Saglio.

^{page 161 note 1} Benndorf, , Sitzungsher. d. Wiener Akad. cxxiii (1890), pp. 47Google Scholar ff. The vase is reproduced by Toutain, l.c.

^{page 161 note 2} Suet. Caesar 39; Aug. 43; Dio, liii, 1 4; lv. 10, 6. On the last occasion while Gaius and Lucius directed circensian games (probably of high born youths), Agrippa, their young brother, took part in the Troia.

^{page 161 note 3} The date of these games is fixed by Dio. lx, 5, though in general editors have failed to realize that the games there recorded are identical with the ludi sevirales. The only other definite reference to the games is found in Julius Capitolinus, Vita Marci Antonini, 6. ‘Pius Marcum … sevirum turmis ducendis iam consulem designation creavit et edenti cum collegis ludos sevirales adsedit. These may have been, as Borghesi suggested, the ludi Martiales at which Augustus was willins to have Claudius curare triclinium sacerdotum (Suet. Claud, 4). There were other ludi Martiales, celebrated on May 12. Cf. Mommsen, , C.I.L. i, ed. 2Google Scholar, Comm. diurn. under May 12. The general assumption that the ludi sevirales for iuvenes corresponded to the Troia for pueri is hardly accurate. The ludi sevirales, as the name ludi indicates, were fixed games, while the Troia conlsisted of exercises performed on the occasion of various festivals.

^{page 161 note 4} Zonaras (from Dio. lv) x, 35.

^{page 162 note 1} C.I.L. vi, 3835, 3530. The first case represents a senatorial career; the second is the record of a man who was tribunus militum and praefectus fabrum, and perhaps died before holding any civil offices. He is listed in Dessau (1314) and in the Corpus as equestrian, but it is to be noted that in the Augustan Age the title praefectus fabrum sometimes belongs to a senatorial career.

^{page 162 note 2} An exception is C.I.L. xi, 1330. On the office see Mommsen, , Staatsrecht iii, pp. 523Google Scholar ff.

^{page 162 note 3} An exception to this is the case of Marcus Aurelius quoted above. For others see Staatsrecht, iii, p. 506, 3.

^{page 163 note 1} On the early history of the cavalry Mommsen's reconstruction seems by far the most convincing, and I have summarized it without discussing in detail the very controversial material. See Staatsrecht, iii, p. 107, no. 3, 253 f. On the doubling in early Roman history see Holzapfel, , ‘Die drei ältesten römischen Tribus,’ Klio, i (1901), pp. 228Google Scholar ff.

^{page 163 note 2} On the development of the Roman cavalry see Mommsen, op. cit. pp. 260. ff; Kübler s.v. Equites, Panly-Wissowa; Helbig, . ‘Zur Geschichte des römischen equitatus,’ in Abhandlungen d. bayr. Akad. I Kl. xxiii, Bd. ii, Munich (1905), pp. 265Google Scholar ff. Soltau's investigation, already referred to, contains much of importance, but goes far beyond the evidence in insisting on the persistence of cavalry units in active service. The continuance of the equestrian centuries under the Empire, in which he believes, is very doubtful, and the distinction which he makes between equo publico and eques Romanus at that time does not hold.

^{page 163 note 3} In view of the other evidence for the use of both turma and centuria in the exercises of knights it seems quite unnecessary to follow Mommsen in assuming that Dionysius is wrong here. See Staatsrecht, iii, p. 522, no. 1 and 3.

^{page 163 note 4} See Piganiol's suggestion for the make up of turmae and centuriae, Recherches sur les jeux romains, p. 18, no.1.

^{page 163 note 5} There is much evidence for the turmae in the lusus Troiae. For the fact that there were divisions named from the centuries of Ramnes, Tities, and Luceres, see Servius, on Aen. v, p. 560Google Scholar.

^{page 163 note 6} This suggestion has often been made before, recently by Kübler, s.v. Equites, Pauly-Wissowa, col. 294.

^{page 164 note 1} This suggestion, a natural development from the previous one, is also an old one. See for instance Gilbert, , Topographie der Stadt Rom. ii, pp. 404Google Scholar ff.

^{page 164 note 2} Compare also the sacrifices performed for the State διὰ τῶν μϵγίστων ίππέων which Dionysius mentions on the occasion of the transvectio equitum.

^{page 164 note 3} Cf. Varro, , L.L. v, 85 and 153Google Scholar.

^{page 164 note 4} Pseudo-Galen, De ther. ad Pisonem, XIV, p. 212Google Scholar ed. Kühn, ὁπότϵ γάρ σου τῶν παίδων ὁ φίλτατος τὴν πϵρὶ τὸ πϵριτόναιον διάθϵσιν ἔσχϵν ἔκ τινος τοῦ ἱππϵύϵιν ἀνάγκης, ἐπϵιδή τις καὶ δημοτϵὴς ἦν μυστηρίων ἱϵρονργὸς διὰ θρησκϵίαν τὴν ὑπὲρ 'Ρωμαίων θϵῶν ἀναγκαίως ἀγομένην τότϵ, ἐφ' ᾦ τϵ καὶ τοὺς ϵὐγϵνϵστάτους παῖδας ἱππϵύοντας ϵὐρύθμως καὶ χορϵύοντας <ὤσπϵρ> τοῖς ἵπποις ἔδϵι τινὰ τῶν μυστηρίων καὶ αὐτοὺς ἐπιτϵλϵῖν. That the passage refers to the lusus Troiae was first pointed out by Friedländer, , in Marquardt, , Römische Staatsverwaltung ¹, iii, p. 505, no. 6Google Scholar.

^{page 164 note 5} Premerstein, Von, ‘Das Troiaspiel und die Tribuni Celerum,’ in Festschrift für Otto Benndorf, Vienna (1898), pp. 261–266Google Scholar.

^{page 165 note 1} Fronto, , ad Marcum Ep. v, 21, 23Google Scholar: ‘Consul populi Romani posita praetexta manicam induit, leonem inter iuvenes quinquatribus percussit, spectante populo Romano.’ Cf. Suet. Domit. 4 ‘Celebrabat et in Albano quotannis quinquatria Minervae cui collegium instituerat ex quo sorte ducti magisterio fungerentur redderentque eximias venationes et scaenicos ludos superque oratorum ac poetarum certamina.’

^{page 165 note 2} On the meaning of the word lusus see Rostovtzeff, , Rev. numismatique, ii (1898), 458–460Google Scholar.

^{page 165 note 3} Rosenberg, , Der Staat der alten Italiker (1913), pp. 93 ffGoogle Scholar.

^{page 165 note 4} Della Corte, , Iuventus, Arpino, 1924Google Scholar.

^{page 166 note 1} C.I.L. iv, 1595.

^{page 166 note 2} C.I.L. iv, 2437.

^{page 166 note 3} The word covehriu of the Volscian tablet from Velitrae and the Latin Quirites are perhaps both to be associated with this word. See Kretschmer, , Glotta, x (1920), pp. 147Google Scholar ff.

^{page 166 note 4} Egger, , Jahreshefte des oester. Arch. Inst. xviii (1915), pp. 115–129Google Scholar.

^{page 166 note 5} Recently discussed by Piganiol op. cit. ch. iv. Piganiol's assumption that the figures in these paintings are adults in the guise of children is unlikely.

^{page 166 note 6} On these offices see Rosenberg, , Hermes, xlix (1914), pp. 267 ffGoogle Scholar; Wissowa, , Hermes, 1 (1915), pp. 11 ffGoogle Scholar.

^{page 167 note 1} The evidence for the municipal institution is fully given in Rostovtzeff's investigation which has already been referred to a number of times.

^{page 167 note 2} N.d.S. (1916), 448, 449, fig. 14.

^{page 167 note 3} On the Augustales see my study in Trans. of the Am. Philol. Assoc. xlv (1914), pp. 231 ffGoogle Scholar.

^{page 167 note 4} See Liebenam s.v. collegia, Pauly-Wissowa.

^{page 167 note 5} See Cagant, s.v. militiae municipales in Daremberg and Saglio.

^{page 167 note 6} See the discussion of this inscription in Rosenberg, op. cit. p. 93 f.

^{page 168 note 1} Space does not permit me to give full references for all statements in the next few pages. What I have to say is based on a study of all inscriptions of Augustales and seviri. A list of the dated inscriptions will be found in Trans, of the Am. Phil. Assoc. xlv (1914), pp. 245–253Google Scholar. For Augustales and seviri Augustales see von Premerstein's article ‘Augustales,’ in Ruggiero, Diz. Ep.

^{page 168 note 2} The simple term sevir is also found here both for libertini and ingenui. On the titles used here and see Mommsen's, discussion, C.I.L. v, p. 635Google Scholar.

^{page 169 note 1} C.I.L. xi, 5278, 5287; cf. iii, 8261.

^{page 169 note 2} That Soltau is wrong in maintaining that a distinction can be made between equo publico and eques Romanus as a designation of Roman knights in the empire is clear from the case of Q. Fabius Hermogenes of Ostia, who in two different inscriptions receives the two titles. Cf. C.I.L. xiv, 6148; N.d.S. (1910), p. 13.

^{page 170 note 1} See the indices to C.I.L. v.

^{page 170 note 2} On the identity of the Genius Augusti and Numen Augusti see Trans. Am. Phil. Assoc. li (1920), p. 132, no. 59Google Scholar.

^{page 170 note 3} See Hatsfeld, , ‘Les Italiens resident à Delos,’ in Bull, de Corr. Hell, xxxv (1912), pp. 102 ffGoogle Scholar; Boak, A. E. R., Class. Phil. xi (1916), pp. 25 ffCrossRefGoogle Scholar.

^{page 170 note 4} Such seems to have been the case at Tibur, where the Herculanei Augustales are both libertini and ingenui. When they are free-born men their title, like the title sevir Augustalis at Praeneste, comes early in the municipal cursus. Lead tesserae from Tibur with the inscription Herc. Aug. indicate that these officers, like the seviri, had a share in games. They were doubtless associated with the local iuvenes. The title Augustalis is applied to the iuvenes in Capua and Ameria. The aedilis Augustalis of Neapolis is probably an officer of the municipal iuvenes who in other towns had aediles as magistrates.

^{page 171 note 1} Rosenberg's suggestion that the iuvenes kept traces of old city organization has already been mentioned. His belief that iuvenes and equites were identical in the inscription of Nepet discussed above, is however unlikely. The equites of that inscription comprised only a portion of the iuvenes, and the sevir was an officer associated with games and ceremonies performed by that portion.

^{page 171 note 2} Usener, Vorträge und Aufsätze, pp. 125 ff.