¹ The two entries are lines 8 and 14 of col. II. Mention of the Feriale Duranum (D. P. 2) was made in the Fifth Dura Report (The Excavations at Dura-Europos, Prelim. Rep. of Fifth Season of Work (Oct. 1931-Mar. 1932), ed. M. I. Rostovtzeff, 1934, p. 295 f., and a small portion of it including line 14 reproduced (PI. XXXI, 2)). Its contents have been outlined by Professor Rostovtzeff in Münchener Beiträge z. Papyrusforschung, XIX (1934), pp. 364–7; and in Comptes-rendus de l'Acad. des Inscr. et Belles-lettres, 1933, p. 312 f. It will be published in full in Yale Classical Studies in 1937 under Professor Rostovtzeff's editorship by R. O. Fink, W. F. Snyder and myself.

² The question is not vital for interpreting the significance of the celebration. Even if the word rosaliae referred, for instance, to the actual decorations, the considerations which follow would not be invalidated, for in any case a decking of the signa with roses is clearly implied, and what is of interest is still to determine the reason for this act.

³ Tertull., Apol., 16; idem., ad nat., I, 12 (ad. fin.); Minuc. Felix, 29, 7. Their eagerness to demonstrate that all unwittingly the Roman soldiers had been worshipping the cross may well have led to some exaggeration of the cult's importance. The evidence relating to the cult of the signa has been collected by Domaszewski, A. von, Die Religion des römischen Heeres (Westd. Zeitschr., XIV, 1 (1895)), p. 12 f.Google Scholar; Renel, Ch., Cultes militaires de Rome: les enseignes, Lyon, 1903Google Scholar; W. Kubitschek in Pauly-Wissowa, s.v. signa, cols. 2342–4; Kruse, H., Studien zur offiziellen Geltung des Kaiserbildes im röm. Kaiserreiche, Paderborn, 1934Google Scholar.

⁴ Cf. such scenes on the Column of Trajan or of Marcus Aurelius, and Renel, op. cit., pp. 297–306.

⁵ Cf. e.g. Cichorius, Die Reliefs der Traiansäule, Pls. LXII, LXIII, LXXXV, and (from Dura itself) Cumont, Fouilles de Doura-Europos (1926), Atlas, Pl. L, and Dura Report V, Pl. XXXVI, 3.

⁶ CIL, III, 6224, 7591 (= Dessau, 2295); VII, 1030 f.

⁷ Josephus, B. J., 6, 6, κομίσαντεϛ τὰϛ σημαίαϛ εἰϛ τò ερòν … ἒθυσαν αὑταιῖϛ αὑτóθι. Herodian IV, 4, 5, us ὡϛ δὲ εἰσέπεσεν (sc. Caracalla) έϛ τò στρατóπεδον ἓϛ τε τòν νεών, ἓνθα τὰ σημεῖα καὶ τὰ ἀγάλματα τοῦ στρατοπέδου προσκυνεῖται, ῤίψϕας ἑαυτòν ἐς γῆν ὡμολóγει τε χαριστήρια ἓθυέ τε σωτήρια. Here the sacrifice is to the “imagines” (τὰ ἀγάλματα) as well, and in general the two cannot be separated (cf. Kruse, op. cit., p. 56). They are usually represented together, were present together at all ceremonies and are considered together as divine recipients of sacrifice. Hence they are probably to be understood as included with the signa in the rosaliae signorum.

⁸ Cf. e.g. an early IIIrd century tombstone from Ramleh by Alexandria, near the headquarters of the imperial army in Egypt (Comptes-rendus de l'Acad. des Inscr. et Belles-lettres, 1872, p. 208 ff.).

⁹ For this later type of supplicatio cf. Wissowa in Pauly-Wissowa, s.v. supplicationes, cols. 949–51. Some of the supplicationes represented on Trajan's column, for example, may well be to the signa grouped by the altar.

¹⁰ Cf. the anointing of aquilae and signa on festival days (Pliny, N. H., XIII, 23).

¹¹ Pap. Genav. lat. 1, l. 9 (von Premerstein, Klio, III (1903), p. 12). Cf., however, the interpretation of Cagnat, L’armée rom. d’Afrique, p. 390, n. 4.

¹² Plut., Marcell., 22: ὤσπερ ἐν τοῖς καθαρμοῖς τῶν στρατοπέδων εἰώθεσαν, δάϕνη πολλῆ καταστέψαντες τὰ ὅπλα καὶ τοὺς ἂυδρας. If the lustration of the arms implied their crowning, so doubtless did that of the signa (for a lustration of the signa, cf. Suet., Vitell., 9).

¹³ Suet., Claud., 13. The decorations missing from the “incompta signa” of the Annals (III, 2, 2) and the “inhonora signa” of the Histories (IV, 62, 4) are probably, however, the dona militaria.

¹⁴ Claudian, X, 186–8.

¹⁵ CIL, II, 2552 (= Dessau, 9125), 2553 (= Dessau, 9127) (natale signor(um)), 2554–6 (now proved genuine by Ann. Epig., 1910, nos. 3–6), 6183 (= Dessau, 2293); von Domaszewski, op. cit., pp. 12, 20.

¹⁶ For these vexillations cf. Rostovtzeff, M., Münchener Beiträge z. Papyrusforschung, XIX (1934), p. 357 f.Google Scholar; idem, Comptes-rendus de l’Acad. des Inscr. et Belleslettres, 1933, p. 310 f.

¹⁷ Cf. infra n. 27.

¹⁸ On this cult cf. esp. J. Gagé Rev. hist., 171 (1933), p. 1 ff.

¹⁹ v. Domaszewski, op. cit., pp. 37–40.

²⁰ Roschers Lexikon, s.v. victoria (Latte), col. 299 f.

²¹ For the crown as the common symbol of victory, cf. Deubner, L., ARW, XXX (1933), p. 80 fGoogle Scholar.

²² E.g. Iupiter Dolichenus, Haas, Bilderatlas, 9–11. Lief., 119, 122. To Mithra too as invictus belonged the crown (Cumont, Textes et Monuments rel. aux mystères de Mithra, I, p. 319 and n. 2).

²³ Victory crowning trophies is very common and crowning the aquila relatively so. For Victory crowning the signa, cf. e.g. Cohen², t. I, Titus 380, Domitien 635; the signa already crowned, t. II, Dide Julien, 1 ff. The legionary eagle with a corona in its beak, not always eschatological in significance, is a common motif. For Dura cf. Report V, Pl. XXXVI, 3, and many other examples. On the top of a shield from Dura is represented an aquila being crowned by two Victories (cf. Rostovtzeff, Città Carovaniere, 1934, p. 163), to be published by F. E. Brown in Report VI, ch. XIII. The coronatio of statues is very often represented in Duran art, cf. Dura Rep. V, Pl. XIV; Yale Classical Studies, V (1935), fig. 50; by Victories, Dura Rep. II, Frontispiece and Pl. I, 1; Dura Rep. V, Pl. XXXVI, 2, 3. A graffito from the wall of a house (Block L 5, 32), to be published by F. E. Brown in Dura Rep. VII, may actually be a crude rendering of the coronatio of a vexillwn at the rosaliae signorum. On the only complete vexillum extant (Rostovtzeff, Monuments of the Egyptian section of the Museum of Alexander III in Moscow, vol. III, Plate, (in Russian)) is painted a victory standing on a globe holding in her left hand a palm branch and in her right a corona.

²⁴ Valerius Maximus, I, 8, 6; Plutarch, Marius, 22, Sert., 22, 2, Pomp., 31, 2 f., Paul. Aemil., 22, 1 (tents crowned); Pliny, N. H., XV, 133 f. (cf. XXXV, 201).

²⁵ Cf. the portent received by Sulla's army (Plut., Sulla, 27, 7). The encouragement of the men was obviously due to their feeling already crowned as after a victory. Tac, Hist., II, 70: “pars viae quam Cremonenses lauru rosaque constraverant” (to greet the victorious Vitellius). Cf. the use of flowers at triumphs, Ovid, Tristia, IV, 2, 20.

²⁶ Cf. supra n. 20.

²⁷ Lines 14–16:

“V k[a]l [feb]rarias ob u[ic]tori[as arabicam et adiabenicam et parthica]ṃ maxi ṃ[a]ṃ diui seụ[e]ẵ[i e]ṭ ob [imperium diui traiani uictoriae part]ḥic[a]ẹ

b’[f’ d]iuo traian[o b’ m’.”

There can be no doubt that the names of victories are to be restored.

²⁸ Cf. Gagé, Rev. arch., XXXII (1930), p. 1 ff.

²⁹ Cf. Gagé, Rev. hist., 171 (1933), p. 1 ff.

³⁰ There are, it is true, two examples of dedications to the Victoria of a legion (CIL, VII, 217 and CIL, III, 11082). Even in the latter, however, the Victoria of the emperors comes first — “victoriae Augg. nn. et leg. I Adi. p. f. Antoninianae.”

³¹ For rosalia in general cf. Pauly-Wissowa, s.v. rosalia (M. P. Nilsson), and besides the works quoted there (col. 1115), M. P. Nilsson, Beiträge zur Religionswiss. (Stockholm), II (1917), p. 134 ff.; Laum, Stiftungen in der gr. und röm. Antike, Leipzig und Berlin, 1914, I, pp. 85–7; Picard et Avezou, B. C. H., 38 (1914), p. 47 ff.; Collart, P., B. C. H., 55 (1931), p. 58 ff.Google Scholar; idem, B. C. H., 54 (1930), p. 376 ff., but on this inscription cf. J. Carcopino, in the volume In memoriam Lui Vasile Pârvan, Bucharest, 1934, p. 77 ff. The latter has remarks on rosalia in general, ibid., pp. 90–2.

³² Cf. Rostovtzeff, M., Münchener Beiträge, XIX (1934), p. 366Google Scholar. This paper was already written when I read the remarks of Professor Carcopino, in the course of which (op. cit., p. 91, n. 1) he expresses the view that the rosaliae signorum had no funerary connexion.

³³ Cf. e.g. the collections of literary evidence by Laum, op. cit., p. 83, n. 2; Frazer in Ovid's Fasti, vol. II, pp. 433 f., 437; Cook, J. H. S., XX (1900), pp. 11–13; Perdrizet, B. C. H., XXIV (1900), pp. 299–301. Special books on the rose are quoted in n. 49. The epigraphical and archaeological evidence is too common to make it worth while to collect it.

³⁴ Examples are Dessau, 6711, 7258, 7267, 8342, 8366, 8369–8374. Collections will also be found in Laum, op. cit., I, 85–6 (with references to II), Pauly-Wissowa, s.v. rosalia, and Waltzing, Étude historique sur les corporations professionelles chez les Romains, IV, p. 541. These inscriptions are especially numerous in Northern Italy, in the Danubian and Balkan lands and in Asia Minor. The celebration took the form of a banquet at which roses were strewn on the grave and distributed to the guests. The dead man's statue (CIL, V, 5907, 5272) or the grave-herm (CIL, V, 5878) was sometimes decorated with them.

³⁵ Cassius Dio, LXVIII, 8, 2 (ed. Boissevain); Cichorius, die Reliefs der Traiansäule, Pl. XCI. His interpretation (op. cit., III, p. 99 f.) of this scene as the performance by Trajan of the above-mentioned sacrifice to the dead on the first anniversary of the battle is generally accepted (cf. B. Götze, Jhb. d. arch. Inst, 50 (1935), Arch. Anz., 348).

³⁶ On this monument (distinct from the Tropaeum Traiani at the same place) cf. Studniczka, Abh. d. sächs. Ges. d. Wiss., 1904, Abh. IV, p. 9 ff. (with bibliography).

³⁷ Furtwängler (Abh. d. bayr. Akad., I. Cl., XXII (1903), p. 473) mentions several.

³⁸ Vegetius, Ep. rei mil. II, 20; Cagnat, L'armée rom. d’Afrique, I, pp. 390–2.

³⁹ It is at least possible, however, that some collegia celebrated rosalia for all their dead. Of such a kind may have been the rose-festival of the Collegium Aesculapi et Hygiae (CIL, VI, 2, 10234 = Bruns, Fontes⁷, p. 391, no. 176, l. 15 f.), though there can be no certainty that it had any connexion with the cult of the dead. The rose festival of the Collegium Silvani (CIL, X, 444 = Bruns, Fontes7, p. 398, no. 180, l. 11 ff.) apparently did not, since the same proceedings are prescribed for a number of other festival days which had no such connexion. ‘Ροδισμοí were celebrated by the Pergamene college of ὑμνῳδοí (Fraenkel, Inschr. v. Perg., II, no. 374, B 8 f., C 6 f., D 8 f.). These were probably celebrations in the cult of the dead (cf. B 21 f.—provision of incense for οἱ άναπυóμενοι, and Fraenkel, op. cit., p. 265 f.; Poland, Gr. Vereinswesen, p. 513).

⁴⁰ Ovid, Fasti, II, 563 (Feralia); V, 485 f. (Lemuria).

⁴¹ Cf. n. 9. I hope to treat of this change and its significance elsewhere.

⁴² Phil., I, 6, 13; II, 43, 110.

⁴³ Cf. the libation in Apul., Met., IV, 22, “poculis aureis memoriae defunctorum commilitonum vino mero libant.”

⁴⁴ Cf. esp. an inscription from Bulgaria (Seyrig, Rev. de l’hist. des relig., XCIV (1928), pp. 275–7):

ἐν δὲ ῤóδοισι [ῤ]óδον ἐπιχεύσατε

τῷ περιβώμῳ ….

ὁ βíος ἒνθα τóτε μ’ ὁρῶντα αὒξει ὡς τò πρóτερον.

⁴⁵ Cf. supra n. 35.

⁴⁶ CIL, I, 1², p. 264, “macellus [sic] rosa(s) sumat.”

⁴⁷ Dessau, 4918, l. 7, “III Idus Mai. rosaria ampiteatri.” The official character of this select list of festivals for Campania is stressed, cf. l. 1 “feriale domnorum” (i.e. Valentian II and Theodosius) and l. 12 “iussione domnorum.” On this document cf. Mommsen, Ges. Schr., VIII, pp. 14–24.

⁴⁸ Nilsson in Pauly-Wissowa, s.v. rosalia, col. 1114.

⁴⁹ The banquet is the time “cum regnat rosa” (Mart., Ep., X, 19, 21). Numerous examples of the ancient use of roses in this and many other connexions will be found in Schleiden, M. J., Die Rose, Geschichte und Symbolik in ethnographischer und kulturhistorischer Beziehung, Leipzig, 1873CrossRefGoogle Scholar, chs. II and III, and in Joret, C., La rose dans l'antiquité et au moyen âge, Paris, 1892Google Scholar, chs. II and III.

⁵⁰ Cf. Athen., IV, 29, 147 f. (Cleopatra's banquet for Antony in Cilicia); Suet., Nero, 27.

⁵¹ vit. Hel., 19, 7; vit. Gall., 16, 2; vit. Car., 18, 4.

⁵² E.g. Plut., Pomp., 57; Herodian, I, 7, 3 and 6 (Commodus’ entry into Rome); Cassius Dio, LXXV, 1, 4 (ed. Boissevain) (Septimius Severus’ entry into Rome); Herodian, IV, 8, 8 (Caracalla's entry into Alexandria).

⁵³ E.g. Lucr., II, 624–8 and Ovid, Fasti, IV, 346 (Magna Mater); Herodian, V, 6, 8 (procession of Elagabal's Syrian god).

⁵⁴ E.g. Tac, Hist., II, 55; cf. Tertull., De cor., 13, “coronatur et vulgus nunc ex principalium prosperitatum exultatione.”

⁵⁵ Cf. Cic., Verr., II, 5, 11, 27, “cum rosam viderat, turn incipere ver arbitrabatur.” In dealing with these rosalia, no attempt has been made to draw a rigid distinction between spring and summer. The Greeks often did not make a distinction at all (A. D. Nock, Gnomon, June 1934, p. 290 f.), and among Latin writers there was no agreement as to the exact date of the beginning of summer (cf. Frazer on Ovid's Fasti, V, 600, vol. IV, p. 72 f.). Our first festival coincides with the date given by Varro (R. R., I, 28) and Pliny (N. H., XVIII, 222), viz. May 9. Ovid (l.c.) dates it May 13, which is the date of the rose festival in the Feriale Capuanum. These coincidences are in all probability accidental, but it is possible that the rosalia were to some extent bound up with the transition from spring to summer, just as the Brumalia (cf. p. 29) were with that from autumn to winter. Although some rose-festivals were held later than this, they can all for our purposes legitimately be classed with spring festivals, since both celebrate the same general change — the reappearance of foliage and flowers.

⁵⁶ For some provincial representations cf. s.vv. Mois, Saisons, the indices to Inventaire des mosaïques de la Gaule etc., and the General Index (vol. X) to Espérandieu, Recueil général des bas-reliefs etc. A good example in a mosaic is Rostovtzeff, Social and Economic History of the Roman Empire, Pl. XLIII, 1. Cf. collection by Stengel in Pauly-Wissowa, s.v. Horai, col. 2311 ff. and Rapp in Roschers Lexikon, I, 2730–8. Cf. also the representations of the Fasti Philocaliani, Strzygowski, Jhb. d. arch. I., I. Ergänzungsh. (1888), pp. 67–9, with Pl. XXIII, and a Greek epigram on the Roman months (A.P. IX, 580, l. 4) in which May says εἰμì ῥóδων γενέτης.

⁵⁷ Roses at the Floralia, Philost., Ep. 55, p. 360 K; cf. Pauly-Wissowa, s.v. Floralia, col. 2750; at the spring festivals of Venus — after the sacred lavatio on April 1 (Ovid, Fasti, IV, 138), and at the Vinalia priora on April 23 (Ovid, Fasti, IV, 869 f.); at the May festivals of Dea Dia celebrated by the Fratres Arvales, Henzen, Acta Fratrum Arvalium, pp. 13, 27 (distributed among fratres after banquets), p. 36 (worn by magister for the ludi circenses of the second day), p. 27 (the goddess herself crowned, probably with roses).

⁵⁸ The spring festival-cycle of Magna Mater and Attis took up a greater section of the calendar than the festivals of any other god throughout the year (H. Graillot, Bibl. Écoles fr. d’Athènes et de Rome, CVII, Paris, 1912, p. 141).

⁵⁹ Rose in procession of Magna Mater, Lucr., II, 627 f.; at the navigium Isidis, Apul., Met., XI, 6, 9, 12; the rose and Bacchus, Daremberg et Saglio, s.v. Bacchus, p. 623, col. 2. The rose may even have been one of the sacred plants of Mithras, Cumont, Textes et Monuments, II, no. 246 (b) (Pl. VI), and remarks in I, p. 194 and n. 6, 195, 205 and n. 4.

⁶⁰ They certainly had an effect on Roman public and private religion (cf. in general Jullian in Daremberg et Saglio, s.v.feriae, p. 1062, col. 1). It was probably, for instance, the rites of the first day of the Magna Mater spring festival, the mourning for Attis (“XI Kal. Apr. Arbor intrat,” CIL, I, 1², p. 260), which gave rise to the celebration of the dies violae or dies violaris at graves by those who were not devotees of the goddess (Graillot, op. cit., p. 145 and n. 5).

⁶¹ Pettazoni, I misteri, Bologna, 1923, p. 283.

⁶² Nilsson calls them “ein ganz profanes Frühlingsgelage, das dem sentimentalen Naturgefühl einer überkultivierten Zeit entsprungen ist” (Beiträge zur Religionswiss., II (1917), p. 134).

⁶³ The Floralia even spread to the municipia of Italy and to the provinces, cf. Pauly-Wissowa, s.v. Floralia (Wissowa), col. 2751. It is to be noticed, however, that the ludi Florales lost two days in the course of the Empire (Pauly-Wissowa, Suppl. V, col. 625).

⁶⁴ Wissowa, op. cit., p. 442. On the growth of the Kalendae cf. Nilsson, M. P., ARW, XIX, (1918), esp. pp. 51–4Google Scholar. It did not attain its greatest popularity until the IVth century. With it by then had become merged the Compitalia (ibid., p. 55).

⁶⁵ Wissowa, op. cit., p. 207, n. 8; von Premerstein, Klio III (1903), p. 11 f.

⁶⁶ Col. II, l. 1, “XIIIỊ Kaḷ apriles ob diem quinq[u]ạ[triorum] sụppḷị[c]atio in X ka[l. easdem.”

⁶⁷ For an interesting example (from Thessaly) of a spring festival which had become a mere carnival, cf. Robertson, D. S., J. H. S., XXXIX (1919), pp. 110–15Google Scholar (on Apul., Met., II, 31–III, 18), and for the origin of the modern carnival in nature rites, Clemen, C., ARW, XVII (1914), p. 139 ffGoogle Scholar.

⁶⁸ Stengel in Pauly-Wissowa, s.v. Horai, col. 2304. The large number of representations of these personifications in the art of the Empire (cf. supra note 56) is an illustration of the growth of interest in commemorating the seasons.

⁶⁹ Dessau, 4918: “lustratio ad flumen Casilino” (May 1) (festival of the standing crops), “rosaria ampiteatri” (May 13), “lustratio ad flumen ad iter Dianae” (July 25) (harvest-festival), “Vendemia Acerusae” (Oct. 15) (vintage-festival).

⁷⁰ Migne, PL., XVII, col. 360 f. In epist. ad Galat. IV, 10 (“dies observatis et menses et tempora et annos”): “tempora vero sunt ver, aestas, autumnus, hiems … Tempora vero sic observant, cum dicunt: Hodie veris initium est, festivitas est.” That the Christian festivals of the Quatre-Temps are a survival of such celebrations has been demonstrated by Dom Morin, Rev. bénédictine, XIV (1897), p. 377 ff.

⁷¹ On the Brumalia cf. Tomaschek, Sitzber. d. Wiener Akad. (Phil.-hist. Cl.), 60 (1868), p. 360 ff.; on the date, Pauly-Wissowa, s.v. bruma; E. Norden, Die Geburt des Kindes, p. 18, n. 4.

⁷² The others are the Saturnalia, the Ianuariae and the Matronales. The De idololatria was written probably in the first decade of the IIIrd century A.D. (Schanz: Ges. d. röm. Lit., VIII, 3³, 285).

⁷³ Pauly-Wissowa, s.v. Saturnalia (Nilsson), col. 210.

⁷⁴ Cf. supra n. 55.

⁷⁵ VIII Kal. Dec. Bruma (CIL, I, l², p. 276). “Bruma” signifies here not the winter solstice, but the first day of the Brumalia.

⁷⁶ For the survival of rosalia, cf. Pauly-Wissowa, s.v. rosalia, col. 1114 f. Add Lammerhirt, H., Neue Heidelb. Jhbb., VIII (1898), p. 1 ff.Google Scholar; Murko, H., Wörter und Sachen, II (1910), pp. 142–154Google Scholar; Nilsson, Neue Jhbb. f. d. klass. Alt. XXVII (1911), p. 682; idem, Beiträge z. Rel.-wiss. (Stockholm), II (1917), p. 149 ff.

⁷⁷ Ges. Schr., VIII, pp. 21–4.

⁷⁸ Cf. Nilsson, Beiträge, l.c., p. 137, “In den Rosalien fehlen von Haus aus die den Christen anstössigen Kultzeremonien.”

⁷⁹ Cod. Theod., XVI, 10, 4, l. 17 (A.D. 399).

⁸⁰ Dessau, 4918, 1. 6 f.

⁸¹ Cf. Tert., De Cor., 13: “ceterum a saeculo coronantur … et ludus et ipsa amphitheatra.”

⁸² Doubtless also a show in it formed part of the celebration.

⁸³ CIL, I, l², p. 264. For the macella at Rome, cf. Platner-Ashby, A Topographical Diet. of Ancient Rome, p. 332 f.; for those in towns of Italy and the provinces, Pauly-Wissowa, s.v. macellum, cols. 130–32 (K. Schneider). They were not exclusively meatmarkets, but general provision markets (Pauly-Wissowa, l.c., col. 133). J. Carcopino (op. cit., p. 91) accordingly seems to give too restricted a significance to the rose festival of the macellum (“une fête corporative des bouchers, une liesse de quartier autour de leur macellum”).

⁸⁴ Cf. supra n. 14.

⁸⁵ Dura Rep. V (1934), ed. M. Rostovtzeff, pp. 211–13.

⁸⁶ No doubt also the sacred objects in the domus signorum and possibly in other temples were crowned with them (cf. the στέψις τῶν ἐν τῷ ἱερῷ πάντων in the records of the Arsinoite temple of Iupiter Capitolinus, Wilcken, Grundzüge, Hist. Teil, Chrest., no. 96, p. III, l. 26–p. IV, l. 2; p. IV, ll. 8 f., 13 f., etc.). They were besides probably hung on the temple-walls (for this common custom at Dura, cf. Cumont, Fouilles de Doura-Europos, Atlas, Pls. L and LV, and text, p. 134), and the altar for the supplicatio decked with them. Every flower festival is characterized by a profusion of flowers (cf. Claudian, X, 295–7, “nee signifer ullus/nec miles pluviae flores dispergere ritu/cessat….”).

⁸⁷ Cf. supra n. 16.

⁸⁸ Cf. supra n. 66.

⁸⁹ Hist., II, 80.

⁹⁰ SHA., vit. Avid. Cass., 5, 11, “ille sane … omnes Sores de capite collo et sinu militi excutiet.” This letter of the prefect of Syria to Marcus Aurelius will hardly be genuine, but we may accept the evidence that the Syrian army was addicted to flowers.

⁹¹ Cf. e.g., Suet., div. Iul., 67, “iactare solitus milites suos etiam unguentatos bene pugnare posse.”

⁹² Pliny, N. H., XIII, 23. Kruse (op. cit., p. 57, n. 3) is almost certainly wrong in holding that the anointing was merely for the purpose of cleaning the standards. That costly unguenta might have been wasted on dusty standards (“signa pulverulenta ilia”) for cult reasons is not so difficult to understand, but surely it is quite inconceivable that they could have been merely for purposes of cleaning. Moreover, the cleaning would have been done not on the festi dies themselves, but beforehand in preparation for them. The anointing was just as much a cult act as the decoration with flowers.

⁹³ For Severus as corrupter of the army's discipline, cf. Cassius Dio, LXXVI, 15, 2 (ed. Boissevain) and esp. Herodian, III, 8, 5.

⁹⁴ Cassius Dio, LXXIX, 28, 1–3 (ed. Boissevain).

⁹⁵ SHA., vit. Sev. Alex. 53 f. “Corrupta disciplina” is a stock topic of the Scriptores, but in this case the fact is attested by Dio, LXXX, 4 (ed. Boissevain, III, p. 475 f.).

⁹⁶ Cf. the fear of the praetorians at this time that a strict commander like Cassius Dio might be placed over them, Cassius Dio, LXXX, 4, 2 (ed. Boissevain, l.c.).

⁹⁷ Cassius Dio, LXXX, 4, 1 f. (ed. Boissevain, l.c.) (Mesopotamia); Herodian, VI, 4, 7.

⁹⁸ The bearing of the Feriale Duranum on military religion in general will be discussed at length in the main publication.

⁹⁹ von Domaszewski, op. cit., pp. 13–20; A. D. Nock in C. A. H., X (1934), p. 483.

¹⁰⁰ von Domaszewski, op. cit., p. 13.