^{page 3 note a} Comment. Lud. Cervarii Tuberonis de origine et incremento Urbis Rhacusance. Ragusa, 1790.

^{page 3 note b} The hitherto known inscriptions from the site are collected in C. I. L. iii. p. 288 seqq. and Prof. Mommsen (s. v. epidaurum) gives a résumé of the earlier sources for the, epigraphy of the place.

^{page 3 note c} Notizie istorico-critiche sulle Antichità, Storia e Letteratura di Bagusei. Ragusa, 1802, t. i. lib. i. ii. The remains at Ragusa Vecchia have been touched on since Appendini's time by Stieglitz, Istrienund Datmazien, p. 264 (Stuttgart und Tübingen, 1845), Wilkinson, , Dalmatia i. 373 (London, 1848)Google Scholar, Kohl, , Reise nach Istrien, Dalmazien und Montenegio, ii. 33 seqq. (Dresden, 1856)Google Scholar, Lago, , Siemorin sulla Dalmazia (Venezia, 1870)Google Scholar, and others, but the notices are slight and add little to our knowledge.

^{page 4 note a} On a Privilegium Veteranorum of Vespasian found at Salona there is mention of a P. Vibius Maximus,—Epitavr. EQ. R. In the Tabula Peutingeriana the name appears as Epitauro: in the Geographer of Ravenna as Epitauron (379,14) and Epitaurum (208, 10). In St. Jerome (Vita S. Hilarionis) Epitaurum: in the sixth century Council-Acts of Salona, Epitaurensis Ecclesia. The town is alluded to by Constantine Porphyrogenitus (De Adm. Imp. c. 29) as τὸ κάστρον τὸ ἐπιλεγόμενον Ηίταυρα and its early Slavonic name was Starigrad Pitaur, still preserving the t in preference to d. The readings of Ptolemy (2, 16, 5), Pliny (23, 143), and Antonine (It. Mar. 520), cannot weigh against this consensus of local testimony; but we need not with Prof. Tomaschek (Die vorslawische Topographie, &c. p. 37) seek an Iliyrian derivation for the name.

^{page 4 note b} 7, 350.

^{page 4 note c} Accepting the correction of the distance Narona—Ad Turres (see p. 79).

^{page 4 note d} Hist. Nat. iii. 22, 143Google Scholar.

^{page 4 note e} A Melita Epidavros Stadia CC. It. Antonini, 520.

^{page 5 note a} C. I. L. iii. p. 287, s. v. EPIDAURUM. I do not know to what Prof. Mommsen refers as the remains of the Amphitheatre.

^{page 5 note b} C. I. L. iii. 1738.

^{page 5 note c} Dr. Ljubié, Viestnik hrvatskoga archeologichoga Družtva (Journal of the Croatian Archccological Society), iii. p. 52Google Scholar, and cf. ii. p. 102, completely corroborates my observations: “Na Prevlaki neostoje ni traga rimskoirra gradu, a rimski nadpis koji ondje stoji uzidan u crkvici bez dvojbe je iz Risna iii iz Kotora donesen.” (There is not a trace of a Roman town at Prevlaka, and the Roman inscription, which is there walled into the church, has been doubtless transported from Risano or Cattaro.) Dr. Ljubie is replying to Gelchich, G., who in his Memorie sulle Bocche di Cattaro (Zara, 1880), p. 7Google Scholar, asserts at random that remains of the city exist at Prevlaka.

^{page 6 note a} Vide Numismatic Chronicle, N.S. vol. xx. pl. XIII. fig. 2.

^{page 6 note b} This gem is now in the possession of Mr. W. J. Stillman. It greatly resembles that engraved by King, Antique Gems and Rings, pl. XV. fig. 8, and probably preserves the outlines of a celebrated statue.

^{page 6 note c} Lopud (It. Mezzo) in the thirteenth and fourteenth centuries, Dalafota, i.e. Da Lafota or D'Alafota, Of. Dr. Constantin Jireček, Die Handelstrassen und Bergwerke von Serbien und Bosnien wahrend des Mittelalters, Prag, 1879, p. 9. Pliny (H.N. iii. 30, 151), mentions the seven Elaphites Insulæ as lying south of Melita (Meleda).

^{page 7 note a} “I di cui vivacissimi colori con maraviglioso artificio fra loro disposti presentano all' occhio una serie luminosa di vaghisshne liste,” is Appendini's high-flown description of this mosaic in 1802. Storia di Ragusa, p. 50.

^{page 7 note b} The engraving which I here reproduce is taken from my work on Bosnia, in which I have already given a popular account of some of the Roman Antiquities of Ragusa Vecchia.

^{page 8 note a} As for instance Molunta (cf. Illyrian-Messapian suffix -untum, -ventum, &c), Vitaljina from Vitalis, Citippi, not to speak of the mediæval reminiscences of Epitaurum, as Starigrad Pitaur, and its modern local name, Cavtat=Civitate, cf. Rouman: Cetate, Citat, Albanian: Giutet, &c. (cf. p. 32). Excavations conducted by my friend Dr. Luschan and myself in mediæval cemeteries about Mrcine and Sokko, not far distant from the head of the Aqueduct, amply demonstrate the prevalence of non-Slavonic crania. For the survival of Roman local names in the territory of Ragusa, see Jirecek, op. cit. p. 8. Still more curious are the fragments of the Roman provincial dialect of Dalmatia existing in the Slavonic dialect of the Ragusans. Vide Prof. Zore, Luko, Dubrovnik, iii. p. 195Google Scholar, Naš jezik tijekom naše književnosti u Dubrovniku. (Our language in the course of our literature in Ragusa)

^{page 9 note a} “Sono stato con sommo contento in Canali per vedere gli avanzi dell' Acquedotto per cui i Romani dalla lontananza di trenta miglie avevano condotto l'acqua in Epidauro, e per maggior godere di quella veneranda antichità alia volta con cavallo mi cacciai in quel letto medesimo su cui un tempo scorreva l'acqua.” The correspondence of Alleti is in the possession of Don Paulovich of Ragusa, by whose kindness I am enabled to reproduce the parts bearing on the antiquities of Epitaurum. Cervarius Tubero, Commentaria suorum temporum, remarks, “Quod autem Canalensis ager territorii Epidaurii fuerit, argumentum est opus mirabilis structuræ effectum, quâ, a vigesimo prope milliario aqua in urbem perducta est, partim subterraneo rivo, partim opere arquato.”

^{page 12 note a} C. I. L. iii. 1741. In its perfect state the inscription ran: p. cornelio ‖ dolabellar cos ‖ vu.viro.epvloni ‖ sodali.titiensi ‖ leg.pro.pk.divi.avgvsti ‖ et.ti.caesris.avgvsti ‖ civitates svperioris ‖ provinciae hillyrici. This Dolabella is referred to by Vellejus Paterculus, who, after mentioning the good government of his Illyrian province by Junius Biæsus in A.D. 14, continues: “Cujus curam ac fidem Dolabella quoque, vir simplicitatis generosissinwe, in maritima parte Illyrici per omnia imitatus est.”

^{page 14 note a} Cf. Marquardt, Handbuch der römischen Alterthümer, pt, iii. sec. i. p. 360. Their financial functions seem to have been later on transferred to the Curatores.

^{page 14 note b} At Dyrrhachium (Durazzo), Ænona (Nona), and Apsorus (Ossero) on this coast, the titles of akdilis and iivir qvinqvennalis are coupled on inscriptions. (Cf. C. I. L. iii. 611, 2977, 3138.) aedilis iivir is common: but on the other hand there were Ædiles who were not Duumvirs, and Duumvirs who were not Ædiles. At Narona we read of aedilis iiiivir: at Salonae of a Curule Ædile. (C. I. L. iii. 2077.)

^{page 15 note a} P. VIBIVS. P. F. VRBICVS ‖ P. ANVLENVS. BASSVS ‖ II. VIR. I. D ‖ CISTERNAM . EX pecunia.publica. REFICIEN‖DAM.CVRAVERVNT. (C. I. L. iii. 1750.)

^{page 15 note b} Antonio Alleti, Segretario della Repubblica di Ragusa, al Rev^do Don Georgio Mattei, a Roma, Dec. 14, 1724: “Mi sono impossessato di un mezzo busto di marmo ed è la figura di m. pomentino figlio di m. pomentino tvrbone iiviro l. d.” The inscription lias been published by Aldus Manutius and others and is given by Mommsen, who had himself personally collated it, in C. I. L. iii. 1748; but the hitherto unpublished passage in Alleti's correspondence is, I believe, the only reference to the bust which formerly accompanied it. The inscription itself at present exists in the Casa Gozze at Ombla. Alleti adds, “Anche allo scoglio di Mercanna ho trovato frammenti di vari iscrizioni senza pero che abbia potuto cavare altro che un barlume indistinto.” (Mercanna is a rocky isle opposite the peninsula on which Epitaurum stood; personally I have been unable to find Roman remains there.) In a letter written from Ragusa in April 1714 he describes an urn found near Ragusa Vecchia with TIPANSIANAS stamped on the lid. The stamp of the Figlinœ Pansianœ is common on Dalmatian sites. (Cf. C. I. L. iii. 3213.)

^{page 15 note c} p.aelio.p.f ‖ tro ‖ osilliano ‖ novia.bassilla ‖ mater.et.novia.ivs ‖ tilla.avia · posvervnt ‖ et.sportvlis.decvrio ‖ avgvstalibvs et sexvi‖ris datis item pvgilvm ‖ spectacvlo dedicave‖ rvnt hvic vniversvs ‖ ordo decvrionatvs ‖ honorem et locvm ‖ statvae deorevit. (C.I.L. iii. 1745.) Discovered in 1856 in the ruins of an ancient building on the shore.

^{page 16 note a} C.I.L iii. 1746, on the authority of Dr. Eitelberger (Jahrbuch der Central Commission, &c. v. 288), who makes the third line simply l d d d. The letters, however, as given in my copy, are perfectly clear.

^{page 16 note b} C. I. L. iii. 1820, and cf. Mommsen, op. cit. p. 291, s. v. narona.

^{page 17 note a} C. I. G. ii. add. 1837, b, c, d, e. All these Pharian inscriptions are now in the museum at Agram. Vide S. Ljubié, Inscriptiones quee Zagabriœ in museo nationali asservantur. Zagabriæ, 1876, p. 71 seqq.

^{page 17 note b} C. I. G. ii. add. 1837, c. The mainlanders with whom the Pharians seem to have been at war were the Jadasini, the inhabitants, that is, of the later Jadera (Zara) and their Liburnian allies.

^{page 17 note c} C. I. G. ii. 1834.

^{page 17 note d} In C. I. L. iii. 2074, are mentioned two decuriones of the Roman Municipium of Issa.

^{page 18 note a} S. Hieronymi Opera, lib. iii. ep. 2, Vita Sancti Hilarionis.

^{page 18 note b} “Draco miræ magnitudinis quas gentili sermone Boas vocant.” The word boa = huge serpent, was known to Pliny (8, 8, 14). It is remarkable that a large species of snake still found in this district is known to the present Slav-speaking inhabitants as Kravosciac, i. e. cow-sucker, as it is supposed to suck the milk of cows. As Coleti, however, judiciously remarks, it is hardly big enough to swallow a dove.

^{page 18 note c} The words of St. Jerome, who must have had opportunities of taking down the tale from the lips of the Epitauritans themselves, are worth notice: “Hoc Epidaurus et omnis ilia regio usque hodie prædicat matresque docent liberos suos ad memoriam in posteros transmittendam.”

^{page 18 note d} This earthquake is placed by the Chronicle of Idatius iD the year 385.

^{page 18 note e} In the sonorous words of Appendini (Storia di Ragusa, vol. i. p. 68Google Scholar): “Il culto verso questo Santo non è punto scemato appresso i Ragusei: anzi una parrochia di cui egli è il Titolare: il concorso nel di della sua festa ad una piccola capella vicina a Ragusa Vecchia (e cio per voto), e tre altre piccole chiese innalzate nel sobborgo di Ragusa in sua memoria perpetueranno in tutti secoli avvenire la tenera pietà e gratitudine dei Ragusei verso un si gran Santo e Protettore.”

^{page 18 note e} The existing popular tradition given by Appendini and others, that this and another cave on Mt. Sniesnitza (about five hours distant from Ragusa Vecchia) were sacred to Æsculapius or Cadmus, is of course of later engrafting, and is akin to the appearance of Dolabella in Ragusa-Vecchian folk-lore.

^{page 19 note a} Τὸν πετρογἐνη the epithet applied to Mithra by Johannes Lydus. So St. Jerome (Adv. Jovinianum, 247), “Narrant et gentilium fabulse Mithram et Ericthoninm de lapide vel in terra de solo libidinis æstu esse generatos;” and Commodianus (Liber Instructionum), “Invictus de petra natus … … deus.” At Carnuntum, in Pannonia, an inscription was found—petrae genetrici. It has been supposed that the idea took its origin from the fact that fire was produced by means of flint; but this method of ignition was apparently, at least among Aryan peoples, a late usage. The real origin of the connexion of Mithra with rocks and mountains should be sought in cloudland.

^{page 19 note b} Cf. Porphyrius, de Antro Nympharum, c. vi. &c.

^{page 19 note c} Das Mithræum von Kroisbach. Dr. F. Kenner (in Mittheilungen der K, K. Central Commission, 1867, p. 119 seqq.)

^{page 19 note d} Del Mitreo annesso alle terme Ostiensi di Antonino Pio. C. Visconti (Annali di Corr. Arch. 1864, p. 147 seqq.)

^{page 20 note a} “ΙΙρῶτα μὲν, ὡς ἔΦη Εὔβουλος Ζωροάστρου αὐτοΦυὲς σπήλαιον ἐν τοῖς πλησίον τῆς ΙΙερσίδος ἀνθηρὸν καὶ πηγὰς ἔχον ἀνιερώσαντος εἰς τιμὴν τοῦ πάντων ποιητοῦ καὶ πατρὸς Μίθρου εἰκόνα Φέροντας αὺτῷ τοῦ σπηλαίου τοῖ κόσμου ὂ ὁ Μίθρας ἐδημιούργησε.” Porphyrius, De Antro Nympharum, c. vi.

^{page 20 note b} In the Mithraic mysteries the initiated died fictitiously in order to be born again by the symbolic sacrifice of a bull. Taveobolio in abternvm renatvs occurs on a monument of a Mithraic votary in C. I. L. vi. 510. Darmesteter (Ormuzd et Ahriman, p. 329) observes that Mithra has usurped the part performed by Çaoshyant in the Mazdean religion, who according to the Bundehesh (75, 6) will give men an immortal body from the marrow of the immolated bull Hadhayaos.

^{page 21 note a} The soul was thought to descend from the moon through the “gate” of Cancer, and to ascend again through the “gate” of Capricorn to the sun. Plato had learned this Magian doctrine (cf. Porphyrius, op. cit. c. xxx.) On their return to their celestial abode the spirits of men were thought to pass through the seven planets (answering to the seven Mithraic grades on earth), by which they were purified and rendered worthy to enter the fixed heaven, the dwelling-place of Ormuzd.

^{page 22 note a} In Lajarde's translation of the passages in the Zendavesta referring to the Eorosh: “Éclatante de lumière “(Recherches sur le culte de Mithra, p. 355.) The elongation of the sun's rays is observable on another Mithraic monument, found at Eome in the Via di Borgo S. Agata (Annali di Gorr. Arch. 1864, p. 177). In this case a ray is made to shoot through a sacred cypress towards Mithra.

^{page 22 note b} Βίον τὸν κρείτονα—the words used by Himerius the Sophist (Orat. vii. 9) in describing the state of the initiated.

^{page 22 note c} See Hammer (Les Mithriaques, PL V. VI. VII.), and cf. Greg. Nazianz, Orat. 3, who describes several of the tortures.

^{page 22 note d} De Prœscriptionibus adv. Hœreticos, c. xl. “(Diabolus) ipsas res sacramentorum divinorum idolorum mysteriis æmulatur. Tingit et ipse quosdam utique credentes et fideles suos. Expiationem delictorum de lavacro reproinittit.”

^{page 22 note e} Cf. Augustine (in Johannis Evangelium, Tract, vii.): “Et magnum est hoc spectare per totum orbem terrarum victum Leonem sanguine Agni … ergo nescio quid simile imitatus est quidam Spiritus ut sanguine simulacrum suum emi vellet quia noverat pretioso sanguine quandocumque redimendum esse genus humanum.” The Spiritus quidam is Mithra, as appears from the succeeding paragraph, in which the Christian Father alludes to the honey mixed with the sacramental water of the Persian rite: King's inference (Gnostics and their Remains, p. 61), that by the simulacrum given to the initiated is betokened an engraved Mithraic gem, affords a reasonable explanation of the passage. It would even seem from St. Augustine's words that he had in view a representation such as the present one of a Mithraic sacrifice, which result gives special point to his parallel. Even as “the Lamb” slays “the roaring Lion,” the Devil, so the false Spirit, “the Capped One,” is represented by his worshippers as slaying the Bull, which, according to their creed, was to herald the resurrection.

^{page 23 note a} C. xv. and c. xviii.

^{page 23 note b} “ἃς (sc. μέλισσας) βουγενεῖς ςῖναι συμβέβηκεν.” Porph. op. cit. c. xv. Cf. Virgil, Georg. iv. v. 554:

“Hie vero subitum ac dictn mirabile monstrum

Aspiciunt liqnefacta bpum per viscera toto

Stridere apes utero et ruptis effervere costis.”

It is to be observed that this portent is obtained by sacrifices offered to the shades of Orpheus and Eurydice; an indication that Virgil was conscious of a mystic connexion between bees, the Magian bull, and the spirit-world.

^{page 24 note a} “σελήνην τε οῡσαν γενεσεως προστάτιδα μέλισσαν ἐκάλουν ἄλλως τε ὲπεὶ ταῦρον μὲν σελήνη, καὶ ὕψωμα σελήνης ό ταῦρος, βουγενεῖς δὲ αἱ μέλισσαι.” Porph. op. cit. c. xviii. An allusion to the same idea will be found on a very interesting engraving on a gold ring from Kertch (in the Siemens Collection) representing a bee above a full-faced bust of Deus Lunus.

^{page 24 note b} Fragmenta (Dindorf. 693). Quoted by Porphyrius, op. cit. in this connexion. Bergk emends the ἒρχεταί τ ἄλλη of Porphyrius, as above.

^{page 24 note c} As for instance on one engraved by Hyde, Historia Religionis veterum Persarum eorumque Magorum, Oxonii. 1700, tab. I.

^{page 24 note d} Porph. op. cit. c. xv.

^{page 24 note e} Les Mithriaques, p. 252.

^{page 25 note a} Given in C. I. L. iii. 1759. I have been unable to find any trace of its present existence.

^{page 25 note b} See p. 23, note ^b.

^{page 25 note c} Æn. vi. 640.

^{page 26 note a} Prudentius, , Contra Symm. i. 467Google Scholar.

^{page 26 note b} In the case of Constantius possibly also of 351. As Vetranio was deposed in January of that year the design can have nothing to do with the appearance of a cross in the heavens recorded four months later in the Chronicon Alexandrinum and in a letter of Cyril, both which authorities fix the date of the meteor, or whatever it was, on May 7, 351. Still less can it have any reference to the Vision of Constantius, which, according to Philostorgius, took place on the eve of the battle of Mursa, in September or October 351.

^{page 27 note a} Cons, H.. La Province Bomaine de Dalmatie (Paris 1882, p. 285Google Scholar): “Les Goths avaient encore fait irruption au-delà du Danube, Ténétré de nouveau jusqu' à l'Adriatique et détruit la Colonie d'Epidaure (Ragusa Vecchia, 265). Les habitants de cette malheureuse ville se réfugièrent au fond de la baie cachée où bientôt s'éleva Raguse.” Now, although the Eastern provinces of Illyricum, including Macedonia and Greece, suffered fearfully at this time, there is no mention of Dalmatia being invaded, much less of Epitaurum having been destroyed.

^{page 27 note b} De Adm. Imp. C. 29: “Οἱ δὲ αὐτοὶ ‘Pαουσαῖοι τὸ παλαιὸν ἐκράτουν τὸ κάστρον τὸ ὲπιλεγόμενον ΙΙίταυρα’ καὶ ἐπειδὴ τὰ λοιπὰ ἐκρατήθησαν κάστρα παρὰ τῶν Σκλάβων τῶν ὄντων ἐν τῷ θέματι, ἐκρατήθη καὶ τὸ τοιοῦτον κάστρον, καὶ οὶ μὲν ὲσΦάγησαν οὶ δὲ ᾐχμαλωτίσθησαν οὶ δὲ δυνηθέντες ἐκΦυγεῖν καὶ διασωθῆναι εἰς τοὺς ὑποκρήμνους τόπους κατῴκησαν … … ὐΦ' οὗ δὲ άπὸ Σαλῶνα μετῴκησαν εἰς Ῥαούσιον εἰσὶν ἔτη Φ' μέχρι τῆς σήμερον, ἥτις ἰνδικτιῶνος ἑβδόμης ἕτους στυνζ´.”

^{page 28 note a} Procopius, de hello Gothico, lib. 1.

^{page 28 note b} Farlati, Illyricum Sacrum, t. ii. p. 163. The bishop of Epitaurum signs next to the bishop of Siscia, what Attila had left of that once great city being now in ecclesiastical subjection to Salonse.

^{page 28 note c} Farlati, op. cit. t. vi. p. 4 seqq.

^{page 28 note d} Gregorius Sabiniano Episcopo Jadertino (in Farlati, op. cit. t. ii. p. 269) ad fin.:—Prceterea habitatores Epidaurensis Civitatis Florentium quern suum dicunt esse Episcopum sibi a nobis restituendum studiosissime poposcerunt.”

^{page 28 note e} In the same way after the destruction of Salonse, the church of Spalato was still known as “Sancta Salonitana Ecclesia.”

^{page 28 note f} This important letter, formerly in the Ragusan archives, begins “Dilecto in Christo filio Andree arcliiepiscopo Sancte Epitauritane ecclesie. Constituimus te omnibus diebus vite tue esse pastorem te et successors tuos super istam provinciam. Imprimis Zachulmie regno et regno Servulie, Tribunieque regno. — Civitati namque Catarensi seu Eosa atque Buduanensi, Avarorum (Antivarorum ?), Lieiniatensi (Ulciniatensi), atque Scodrinensi, nec non Drivastinensi atque Polatensi cum ecclesiis atque parochiis eorum.” Owing to the insertion of the Archiepiscopal title doubts have been thrown on the genuineness of this letter. It is, however, accepted by Kukuljevié, who gives it in the Codex diplomaticus regni Croatiœ, &c. p. 35.

^{page 28 note g} Gregorius Maximo episcopo Salonitano … “Et quidem de Sclavorum gente quse vobis valde imminet affligor vehementer et conturbor. Affligor in his qua; jam in vobis patior; conturbor quia per Istriae aditum jam Italiam intrare cœperunt.”

^{page 29 note a} Mansi, Collectio Concil. t. ix. Gregory appoints the refugee bishop to the bishopric of Squillace. Should, however, his own city be liberated at any time from the enemy he is to return to it.

^{page 29 note b} Farlati, op. cit. t. iii. p. 22.

^{page 29 note c} Šafarik for example (Slawische Alterthumer, ii. 271Google Scholar) imagines Constantine's derivation of Canali to have been founded on some blundering reminiscence of “Kolnich” which appears as the Slavonic equivalent of Via Carri in a document of the year 1194 referred to by Lucius (De regno Dalmatics et Croatiœ, lib. vi.)

^{page 29 note d} His explanation for instance of the name of the neighbouring old Serbian district of Zachulmia, “ὸπίσω τοῦ βουνοῦ ” is a perfectly correct piece of Slavonic etymology. Equally exact is his rendering of the Croatian Primorje by “ή ΙΙαραθαλασσία ” His derivation for the river-name Bona contrasts favourably with Šafarik's.

^{page 30 note a} “τὸ δὲ Καναλή ἑρμηνεύεται τῇ τῶν Σκλάβων διαλέκτῳ άμαξιά, ἐπειδή το εῖναι τὸν τόπον ἐπίπεδον, πάσας αὐτῶν τὰς δουλείας διὰ άμαξῶν ἐκτελοῦσιν ” De Adm. Imp. C. 34.

^{page 30 note b} Be Cursu Publico, xv. “Mancipium, cursus publici dispositio Proconsulis formâ teneatur. Neque tamen sit cujusquam tarn insignis audacia qui parangarias aut paraveredos ad canalem audeat commovere quominus marrnora privatorum vehiculis provincialium transferantur.” Du Cange (s.v. Canalis) interprets this to mean that pack-horses, &c. destined for lanes and bye-ways are not to block the highway, but agrees in the important point that canalis = via publica.

^{page 30 note c} De Curiosis, ii. “Quippe sufficit duos (sc. agentes in rebus) tantummodo curas gerere et cursum publicum gubernare ut licet in canalibus publicis base necessitas explicetur.” (Law of Constantius and Julian, 347 A.D.) Gothofred (ad loc.) observes, “Illud satis constat hie non pertinere ad aquarum sen fluminum canales, quandoquidem in his rhedas, birotum, veredi, clabulse, moveri dicuntur.”

^{page 30 note d} Graudentius (Cone. Sardic. can, 20) speaks of “ἕκυστος ήμῶν τῶν; ὲν τοῖς παρόδοις ἥτοι καναλίῳ καθιστώτων.” In the Latin translation (Mansi, t. iii. p. 22): “Qui sumus prope vias publicas seu canales.” Ducange supposes that the word canalis in a charter of A.D. 1000, published by Ughellus (Episcopi Bergamenses), has the same meaning of “via publica.”

^{page 30 note e} Apol. i. 340Google Scholar. οἱ ἐν τῷ καναλίῳ τῆς Ιταλίας.

^{page 31 note a} a In Serbian it often appears in the plural form, Konavle = the channels, showing that the name took in the lateral system of irrigation which ramified across the plain from the main Aqueduct. The plain of Canali is still (as has already been noticed) one of the best irrigated regions in Dalmatia—the inhabitants having in this respect inherited their Roman traditions.

^{page 31 note b} Konô (i. e. konol).

^{page 31 note c} The earliest Dalmatian chronicler, the Presbyter of Dioclea, who wrote about the year 1150, expressly identifies this Rouman population with the descendants of the Roman provincials of Illyricum. After mentioning the conquest of Macedonia by the Bulgarians under their Khagan he continues: “post hæc ceperunt totam provinciam Latinorum qui illo tempore Romani yocabantur modo vero Morovlachi, hoc est nigri Latini, vocantur.” Regnum Slavorum, 4.

^{page 32 note a} The materials relating to the Rouman population of Dalmatia, Herzegovina, &c. existing in the archives of Eagusa haye been collected by Dr Const. Jireček in his paper entitled Die Wlachen und Maurowlachen in den Denkmälern von Ragusa. (Sitzungsberichte der k. böhm. Gesellschaft der Wissenschaften, 1879).

^{page 32 note b} e. g. Dókes = decessus (of the tide), rekesa = recessus, plaker = placere, lukjérnar = lucernarius. (Prof. Luko Zore, Nas jezik tijekom naše knjizevnosti u Dubrovniku. (Our language in the course of our literature in Ragusa.) (Dubrovnik, iii. 1871Google Scholar.) The preservation of the k sound of the Latin c is also a characteristic of the Latin forms contained in Albanian. The discovery of a Roman-Christian glass bowl of sixth-century date among the ruins of Doklea (Dukle in Montenegro), presenting inscriptions in the local dialect, shows that this guttural survival was an early peculiarity of the Romance dialect of this part of Illyricum. On the Doclean vase under the figure of Jonah and the whale occurs the line “Diunan de ventre queti liberatus est,” where the “queti ” for “ceti ” is, as the Comm. di Rossi (Bull. di Arch. Crist. 1877, p. 77) points out, not a mere barbarism but an archaistic survival carrying us back to the “oquoltod” for “occulto,” “quom” for “cum,” &c. of the S. C. de Bacchanalibuss. On a Dalmatian inscription (C. I. L. iii. 2046) QVELIE occurs for COELIAE. In the matter of the survival of the k sound of the c Dalmatia showed itself more conservative than the West. The epigrammatic address of Ausonius to Venus,

“Orta salo, suscepta solo, patre edita cœlo “

loses its alliterative point unless the cœlo be pronounced as beginning with a sibilant: and the natural inference is that in fifth-century Gaul the guttural sound of the Roman c had been already softened down.

^{page 32 note c} E. g. Vergatto (SI. Brgat), mediæval Vergatum, from Latin Virgetum; Zonchetto, Latin Junchetum;Rogiatto (SI. Rožat) = Rosetum; Delubie, on the bank of the Ombla, = Diluvies. (Cf. Jireček, Die Handelstrassen, &c. p. 8.) Montebirt, the name of a pine-clad height near Ragusa, seems to me to be a Mons Viridis (cf. Brgat for Virgetum), though the derivation from a combination of the Latin and Slavonic name for mountain—brdo—has been suggested by Professor Zore. In the latter case it would find a parallel in “Mungibel.” The rocky promontory of Lave or Lavve on which the earliest city of Ragusa was built derives its name from a low Latin form labes — land-slip. Constantine Porph. (De Adm. Imp. c. 29) gives it under the form λαῦ and makes it = κρημνὸς.

^{page 32 note d} Cf. Jireček, Die Wlachen und Maurowlachen, &c. p. 6.

^{page 33 note a} Libri Rogatorum, 1427–32. The older name for Mrcine in the Ragusan records is Versigne. Cf. Jireček, Die Wlachen, &c. p. 6.

^{page 33 note b} E. g. Mǎrǎcini⊡u, = a place overgrown with thorns; Mǎrǎcinosu, = thorny.

^{page 33 note c} Archiva istorica a Romaniei, t. iii. Bucuresci, 1867. Besturile unei carti de donatiune de pe la annul, 1348, emanata de la Imperatul Serbesc Dušan, &c.

^{page 33 note d} This etymology, if admitted, would be a strong argument against the exclusively Thracian origin of the Wallachians, which at present finds so much favour.

^{page 33 note e} Similar medieeval megalithic cemeteries, of which I hope to say something on another occasion, are scattered over a large part of what is now Herzegovina, Bosnia, Northern Montenegro, and certain districts of Dalmatia, and are common to both old Serbian and old Ttouman districts. They are therefore not by themselves of ethnographical value. The inscriptions when found are always Serbian, and in Cyrillian characters; the “Vlachs “do not seem to have had a -written language. A rich “Vlach,” however, being bilingual, might put up an inscription in Serbian, which was to him the language of Church and State.

^{page 34 note a} The Ragusans early found a more convenient Romance language in Italian. Nor is it necessary to suppose that they ever spoke a Rouman dialect in the sense that the Dalmatian highlanders spoke it. The correspondence between Ragusa and the other Dalmatian coast-cities, Cattaro, Budua, Antivari, &c. was conducted in Latin.

^{page 34 note b} This fact had already struck Lucius (De regno Dalmatics et Croatia, lib. vi. Francofurti, 1666, p. 277), who instances “Petra” = SI. “Brus”; “Via Carri” = SI. “Colnich”; “Circiutus,” = Sl.”Zavod”; “Calamet” = SI. “Tarstenich.” Cf. “Cannosa,” near Ragusa, SI. “Trstenik.” In the same way Vlach personal names were early translated into Slavonic equivalents, so that in Ragusan records we hear again and again of “Vlachi” with Serbian names.

^{page 35 note a} De Adm. Imp. C. 33: “ἐν τῷ τοιούτῳ χωρίῳ βουνός μέγας, ἔχων ἄνωθɛν αὐτοῦ δύο kappa;άστρα, τὸ Bόνα καὶ τὸ Xλούμ ὄπισθɛν δὲ τοῦ τοιούτου βουνοῦ διέρχɛται π οταμὸς καλούμɛνος Bόνα, ὄ ἐρμηνɛύɛται καλόν” At present the castle on the peak is called Blagaj, the river which wells in full volume from its foot is still called Buna. This passage of Constantine affords valuable evidence of the existence in the tenth century cf an Illyro-Koman population among the interior ranges of what is now Herzegovina. Bona is a characteristic Rouman name for good, clear, streams (cf. SI. Dobravoda, &c), and re-appears in this sense in the North Albanian Alps, where the Val Bona indicates the former presence of Romance-speaking highlanders in a glen which so far as language is concerned is at present Albanian. In the same way we find forms like Alpʼbona in the Ladine or Romance districts of Tyrol.

^{page 35 note b} C. I. L. iii. 6418.

^{page 35 note c} The Ragusan records and old Serbian chrysobulls reveal a great extension of Rouman tribes in this part of Western Illyricum in the early Middle Ages. Amongst those in the present Herzegovina and Montenegro were the Vlachi Banjani, Nikšiéi, Miriloviéi, Pilatovci, and the Rigiani in the mountains that overlook the ruins of Risinium. Their Alpine villages were called Cantons, in Slav. Katun, from whence the Katunska Nahia of Montenegro has its name. Like the Dokleates, the Illyrian tribe that once occupied a considerable part of the same mountain region, and of whom they were in part the Romanized descendants, they were great cheese-makers. The foundation charter of the church of St. Michael and St. Gabriel at Prizrend (1348) presents us with a number of Wallachian personal names with the Rouman suffix -ul, showing the Illyro-Roman survival in the ancient Dardanian province and its border-lands.

^{page 36 note a} Die vorslawische Topographic, &c. p. 37.

^{page 38 note a} C. I. L. iii. 1739.

^{page 39 note a} “Πολισμάτιον ɛὖ πρὸς ὸχυρότητα κατασκɛυασμένον, άνακɛχωρηκὸς μὲν άπὸ τῆς θαλάττης, ἐπ᾽ αὐτῷ δὲ κɛίμɛνον τῷ Ῥίζωνι ποταμῷ.” Polybios, ii. 11Google Scholar.

^{page 39 note b} Dalmatia, vol. ii. p. 234Google Scholar.

^{page 40 note a} See “On some recent discoveries of Illyrian Coins,” Numismatic Chronicle, N.S. vol. xx. pp. 269–302Google Scholar.

^{page 41 note a} Justinus, lib. xvii. 3: “(Pyrrhus) defertur in Illyros et traditus est Beroæ uxori regis Glauciæ quæ et ipsa erat generis Æacidarum.”

^{page 41 note b} As edited by Mommsen in C. I. L. iii. p. 285.

^{page 42 note a} See Numismatic Chronicle, N.S. vol. xx. p. 269 seqq.

^{page 42 note b} Lib. xlv. c. 26.

^{page 43 note a} Pliny, H. N. lib. xxv. 34: “Gentianam invenit Gentius rex Illyriorum, ubique nascentem, in Illyrico tamen præstantissimam.”

^{page 44 note a} V. 420 seqq.

“Καί τινα μὲν αὐτῶν βουλικαῖς ἐξουσίαις

ὐπήκοʼ εἶναι, τινὰ δἐ καὶ μοναρχίαις,

ἃ δʼ αὐτονομεῖσθαι θεοσεβεῖς δʼ αὐτοὺς ἄγαν

καὶ σϕόδρα δικαίους, ϕασὶ, καὶ ϕιλοξένους,

κοινωνικὴν διάθεσιν ἠγαπῃκότας

εἶναι, βίον ςηλοῦν τε κοσμιώτατον. ”

His words have a special reference to the south Dalmatian coast, as he places opposite the region of these civilized mainlanders the Greek island colonies of Pharos (Lesina) and Corcyra Nigra (Curzola).

^{page 44 note b} Cf. G. Gelchich, Memorie storiche sulle Bocche di Cattaro, pp. 10, 11, and Ljubié, Viestnik hrvatskoga Arkeologičkoga Družtva, an. iii. p. 52. Most of these have been transported to Perasto.

^{page 45 note a} Fr. 140. Theopompos imagines that the vases must have reached the Naron by some underground river course forming a connexion between the Adriatic and the Ægean. Strabo, to whom the preservation of this notice is due, is justly sceptical as to the geological deduction of Theopompos: “Καὶ ἅλλα δʼ οὐ πιστὰ λέγει τό τε συντετρῆσθαι τὰ πελάγη ἀπὸ τοῦ εὐρίσκεσθαι κεραμόν τε Θάσιον καὶ Χῖον ἐν τῷ Νάρονι.” (vii. p. 488.)

^{page 45 note b} Περὶ θαυμασίων ἀκουσμάτων, c. civ.

^{page 45 note c} “εἶναι δὲ καί τινα τόπον ἐν τοῖς ἀνὰ μέσον διαστήμασιν εἰς ὂν ἀγορᾶς κοινῆς γενομένης πωλεῖσθαι παρἁ μἐν ἐκ τοῦ Πόντον ἐμπόρων ἀναβαινόντων τὰ Λέσβια καὶ Χῖα, καὶ Θάσια, παρὰ δὲ τῶν ὲκ τοῦ Ἀδρίς τοὺς Κερκυραϊκοὺς ἀμΦορεῖς.”

^{page 46 note a} Cf. Glavinich, Mittheilungen der k. h. Central Commission, 1878, xcii. In the museum at Ragusa is a Greek painted vase said to have been found on the site of Epitaurum.

^{page 46 note b} Since I took down these inscriptions copies of figs. 13, 14, 5, and 17 have been sent to the Croatian Archæological Society, and are given by Dr. Ljubié in Viestnik (an. i. p. 127; an. ii. p. 101), where my excavations are referred to. The examples in the Viestnik will he found to differ in some small details from mine, and do not represent the original lettering. Figs. 14 and 1G are at present in the Casa Mišetić. Fig. 13 was found in the campagna of Paprenica. Fig. 15 is from the left bank of the Fiumara; I have since deposited this stone in the museum at Ragusa.

^{page 46 note c} C. I. L. iii. 1730, as completed by Mommsen.

^{page 46 note d} Cf. C. I. L. iii. 2751, 2752, 2773, 2788, among inscriptions found at Verlikka and S. Danillo in Dalmatia; 3144 in the Isle of Cherso; 3804, 3825, at Igg near Laibach, herein a Celtic connexion:—“voltrex plaetoris”; in a Privilegium (C. I. L. iii. D. vii.) granted by Vespasian—platoei · veneti. f · centvrioni · maezeio; at Apulum and Alburnus Major (vicvs pirvstarvm) in Dacia where was a large Illyrian mining colony (1192, 1271)

^{page 48 note a} Cf. inscriptions found at Capo di Leuca, Πλατορας Παλεταος Ισαρετι, and at Ceglie beginning ΓΑΑΤΟΡΑΣ, given in Mommsen, Die unteritalienischen Dialekte, p. 51. The plebeian family name Plsetoria at Rome was derived from this source.

^{page 49 note a} Amongst other objects of Eoman jewelry obtained by myself from this site may be mentioned a part of a gold earring terminating in a lion's head, and two spiral snake bracelets of silver, much resembling a kind of bangle which has lately again become fashionable.

^{page 49 note b} On another Christian gem, obtained by me at Salona, the Good Shepherd stands at the side of a group of sheep and goats beneath a palm tree. The material is green jasper.

^{page 49 note c} Given in Farlati, Illyricum Sacrum, t. vi. pp. 411, 412. The letters are headed “Gregorius Sebastiano Episcopo Rhiziniensi.”

^{page 49 note d} “Quia ejus in nos malitia gladios Longobardorum vicit.”

^{page 50 note a} See Von Arneth, Monumente des k. k. Münz und Antiken Cabinettes, Wien, 1850, Pl. g. IV., g, v., g. xiv. &c.

^{page 51 note a} An account of the Torontal treasure will be found in Von Arneth, op. cit. p. 20 seqq.

^{page 51 note b} This inscription reads: ΒΟΥΗΛΛ · ΖΟΑΙΙΑΝ · ΤΕΣΗ · ΔΥΓΕΤΟΙΓΗ · ΒΟΥΤΑΟΥΛ · ΖΩΑΙΙΑΝ · ΤΑΓΡΟΓΗ · ΗΤΖΙΓΗ · ΤΑΙΣΗ. Von Hammer (Osmanische Geschichte, iii. 726) compares ΤΑΓΡΟΓΗ · ΙΙΤΖΙΓΗ with Δακριγοὶ Ιἄζυγες, a tribe of Jazyges mentioned by Dion (lxxi. 12). The Tagri are mentioned by Ptolemy (iii. c. 5). The inscription is cited by Ŝafarik (Slawische Alterthümer, i. 345Google Scholar) as a monument of the early connexion of Slavs and Sarmatians. cannot be other than the Slav Župan, the governor of the Žitpa or Mark.

^{page 54 note a} Archaeologia, vol. iii. p. 346Google Scholar.

^{page 54 note b} Prof. Ljubić in Viestnik hrvatshoga Arkeologiĉkoga Družtva, 1880, No. 1.

^{page 55 note a} C. I. L. iii. 2992, 2995.

^{page 55 note b} This is far from denying that there was an alternative road from Liburnia into Japygia by way of the Municipium that apparently occupied the site of the present Obbrovazzo. It stands to reason indeed that this line of communication was known to and used by the Romans. All that I have been maintaining is, that the natural route from Burnum towards Siseia and Senia wonld ran through the easier pass of the Zermanja, I am, personally, well acquainted with both routes.

^{page 56 note a} A copy of this inscription was sent by its present possessor to Dr. Kukuljević, and has been communicated by him to the Ephemeris Epigraphica (vol. iii. n. 570). The version given there, however, is misleading.

^{page 57 note a} Historia Salonitana, c. xxxix.: “Rex relictis stationibus Zagrabiensium partium cum omni comitatu suo ad mare descendit …, Rex vero et iotas flos reliquorum Ungarorum ad Spalati partes devenit.” Later he retreats to Traü, “cum uxore sua et cum omnibus gazis suis,”

^{page 57 note b} “Venit autem, non quasi iter faciens sed quasi per aerem volans loca invia et montes asperrimos supergrediens unde numquam exercitus ambulavit.” Op. cit. c. xl.

^{page 58 note c} The name Germeć covers a greater area to the South-East than that assigned to it in the Austrian General-Stabs Karte.

^{page 59 note a} Monatsbericht der k. preuss. Akad. der Wissenschaften, 1867, p. 741 seqq. Cf. La Via romana da Sirmio a Salona (in Bulettino di archeologia e storia Dalmata, 1882, p. 69). Hoernes, , Alterthümer der Hercegovina, ii. 131Google Scholarseqq., accepts Dr. Blan's conjecture as to the course of the way from Dobrinja across the Crnagora, and sees in the Boman remains found at Glavice, Glamoĉ, and Livno, an indication of its subsequent course. Tomaschek advocates the same general line (Die vorslawische Topographie der Bosna, &c. p. 16 seqq.), but his views on Dalmatian topography are not corrected by personal observation. A comparison, of the Tabula and the Itinerary seems to show that between Leusaba and Æquum there were two alternative routes. In the Tabula we, have Æquo, viii. In, Alperio, xiiii. Bariduo, —— Ionnaria, xiii. Sarute, vii. Indenea, v. Baloie, xii. Leusaba. In Antonine: Æquo, xvii. Pelva, xviii. Salvia, or Silviæ, xxiiii. Sarnacle (or Sarnade), xviii. Leusaba.

^{page 59 note b} “in, Ermangelung antiker Reste kann Leosaba nur im, allgemeinem in, der Hochebene Podraznica angegeben werden.”

^{page 60 note a} Cf. Blan, Reisen in Bosnien, &c, p. 110.

^{page 60 note b} Near Varcar, to the North of Banjaluka and Eastward of Kliuĉ, have been recently discovered. Roman remains, including a large hoard of denarii, mostly of the Emperors Alexander Severus, Gordian, Philip, Trajan Decius, Gallus, and Volusian, some sixty of which have passed through my hands. The discovery of Roman remains at this site establishes a link of connexion between the Sana Valley and the succession of Roman sites at Podlipci, Runić, Mosunj, Putaĉevo and Vitež, in the Valley of the Lasva, and points to an old line of communication between the Upper Bosna and the Sana, which, opens the moist, natural route towards Sisera.

^{page 60 note c} Interesting remains have been, lately discovered by Capt. Von. Handel in the Valley of the Unna about an hour to the south-east of Bihać. They consist of several inscriptions, one presenting the female Illyrian name-form ditveio and the Mazeian. name Andes, a Mithraic relief, a figure of a Faun or Sylvanus, and, other fragments. Prof. Tomaschek, who has published an, account of the discovery (Sitzungsberichte der k. k. Akademic der Wissenshaften, 1881, h. 2, p. 466 seqq.), is inclined to identify the site with, the ancient Rætininm. There is a height answering well enough to the description of the Acropolis of Rætinium, besieged by Germanicus.

^{page 60 note d} In one case a monogram appeared, Æ

^{page 60 note e} I have alluded to this discovery in, my Illyrian Letters, London, 1878, p. 37Google Scholar.

^{page 61 note a} Cf. Germ. Schanze.

^{page 61 note b} Viestnik hrvatskoga arkeologiĉkoga Družtva, 1880, p. 63: “jedan komad nadgrobne ploce na kojoj je u basirilifu ljuĉka slika skrstenima rukama izpod koje nadpis koj je zub vremena veoma iztroŝio, no vidi se ipak da je rimski.” In the same communication is mentioned the discovery of Koman coins of Constantine's time, together with other antiquities, at Kumiégrad, an hour's distance from Srb.

^{page 62 note a} C. I. L. iii. 6417.

^{page 62 note b} The monument (C. I. L. iii. 6418) is erected to a veteran of the 11th legion killed here, “finibvs varvarinorvm iu agello secvs titvm flymen ad petram longam.” It was found near the village of Puljane, at a spot still known as Duga Stina, “the long rock “(cf. p. 35).

^{page 62 note c} The other face of the monument when I saw it was built into the wall. Its height was about 2½ feet. The segment of this ornament (fig. 5^a) is taken from a sketch which the susceptibility of the Austrian authorities prevented me from completing and which is therefore imperfect.

^{page 63 note a} Engraved in Eitelberger, Die mittclalterlichen Kunstdenkmale Dalmaziens, p. 150.

^{page 63 note b} Arkiv za poviestnicu jugoslavenshu, vol. iv. p. 390Google Scholarseqq. The frontispiece to this volume contains a representation of the font.

^{page 63 note c} Codex diplomaticus Regni Croatiœ Dalmatiœ et Slavoniœ, xc. (t. i. p. 76). The Pope continues, “Quis enim specialis filius sanctæ Romanæ ecclesiæ, sicut vos estis, in barbara seu Sclavinica lingua Deo sacrificium offerre delectatur?”

^{page 63 note d} Codex diplomaticus, xcii. (t. i. p. 78).

^{page 64 note a} Farlati, , Illyricum Sacrum, t. iv. p. 280Google Scholar.

^{page 64 note b} On his coins, Baduila or Baduela. In this connexion I may mention that I have obtained from Bosnia a jacinth intaglio on which is engraved a monogram bearing the closest resemblance to that of Theodoric on his coins.

^{page 65 note a} Theophylact Simocatta, Hist. vii. 11, 12 (Ed. Bonn, p. 291.) Theophanes, Chronographia, p. 428.

^{page 65 note b} Tὰς λεωϕόρους.

^{page 65 note c} Slav. Alt. vol. ii. p. 238Google Scholar.

^{page 65 note d} Mon. Spec. hist. Slavorum Meridionalium, vol. vii. p. 254Google Scholar.

^{page 65 note e} I can see no reasonable grounds for accepting Prof. Tomaschek's conjecture (in the teeth of all the MSS.), that the word is a corruption of Salviis (Vorslawische Topographic, &c. p. 19), or the suggestion of Dr. Hoernes, (Alterthümer der Hercegovina, &c. vol. ii. p. 134)Google Scholar, that “Salviæ” (in most MSS. “Silviæ “) and “Balbeis “are alternative names for the same place.

^{page 66 note a} Perhaps the ad ladios of Antoninus.

^{page 66 note b} This monument is at present in the Museum at Spalato, and has been described by Dr. Glavinić.

^{page 67 note a} Cf. Thomas Archidiaconus, op. cit. c. xl. “Mortuæ sunt duæ puellæ virgines, scilicet filiæ regis Belæ et in ecclesia B. Domnis honorificè tumulatæ.”

Lucius, who gives this inscription in his notes to Thomas Archid. (in De Regno Dalmatiœ et Croatiœ, Frankfort, 1666, p. 473Google Scholar), adds, “Gulielmus quoque, Belæ ex filia nepos, in hac eadem fuga mortuus, Tragurii sepultus fuit.” The epitaph of this prince formerly existing at Traü is given by the same author in his Memorials of that city. It contained the lines,

“Arcente denique barbaro perverso

Infinitis Tartans marte sub adverso,

Quartum Belam prosequens ejus consobrinum

Ad mare pervenerat usque Dalmatinum.”

a imp · caes ∥ m. avrelivs ∥ commodvs ∥ antoninvs ∥ avg · pivs · sarm ∥ germ · maximvs ∥ brittannicus ∥ pont · max · trib ∥ pot · viiii · Imp · vi ∥ cos · IiiI · p · p ∥ pontem · hippi flvml∥nis · vetvstate cor∥rvptum restitvit ∥ svmptvm et operas ∥ svbministrantibvs ∥ novensibvs delmi∥nensibvs riditis · cv∥rante · et · dedicante ∥ L · ivnio · rvfino · procv∥liano · leg · pr · pr · (C. I. L. iii. 3202.) This inscription was discovered by Dr. Carrara and first published in the Bulletino dell' Inst. di Corr. Arch. 1815. The name of Commodus had been defaced in accordance with the orders of the Senate recorded by Lampridius.

^{page 68 note b} The form in which it appears in Ravennas, the only geographer who mentions it. He gives it (5, 14) as the last station before reaching Scardona, on the road from Tragurion (Träú). Its actual site was at St. Danilo near Sebenico. (Cf. C. I. L. iii. 2767, &c.)

^{page 69 note a} “Πόλιν Δελμίνιον ὅθεν ἄρα καὶ τὸ ὅνομα αὐτοῖς ἐς Δελματέας ετα Δαλμάτας ἐτράπη ” Appian, Illyr. ii. Cf. Strabo. vii. 5.

^{page 69 note b} The variant forms of the name occur: Delminum, Dalmis, Dalmion, Delmion.

^{page 69 note c} Cf. Prof.Tomaschek, W., Die vorslawische Topographie der Bosna, Herzegowina, Crnagora. und der angrenzenden Gebiete (Wien, 1880)Google Scholar. (Separat-abdruck aus den Mittheilungen der k. k. geographischen Gesellschaft), p. 9. The Catholic bishopric that existed here in the fourteenth century was still known as Ep. Delmensis or Dultncnsis.

^{page 69 note d} C. I. L. iii. p. 358, s. v. delminium.

^{page 69 note e} Die vorslawische Topographic der Bosna, Herzegowina, Crnagora und der angrenzeiiden Gebiete. (Separat-abdruck aus den Mittheilungen der k. k. geographischen Gesellschaft), p. 10.

^{page 69 note f} Bullettino di Archeologia e Storia Dalmata, 1878, p. 23.

^{page 69 note g} What is extremely pertinent in this regard, Constantine Porphyrogenitus mentions that the “Župa of Dalen,” the form given by him to the old Slavonic Dulmno (Duvno), belonged to the Pagani or Narentans: a fact which shows a certain facility of inter-communication between the inland plain of Duvno and the Narenta Valley. (Be Adm. Imp. c. 30.) Dr. Kukuljevié, Codex diplomaticus regni Croatiœ, Dalmatiœ et Slavoniœ, pt. I. p. 86, note, agrees in identifying the Zupa of “Dalen “with Duvno.

^{page 70 note a} Farlati, , Illyricum Sacrum, t. ii. p. 173Google Scholar.

^{page 70 note b} Loc. cit. The geographical details of Constantine regarding Dalmatia and its borderlands are peculiarly valuable, and seem to have been supplied by trustworthy native informants; not improbably Ragusan patricians, amongst whom was a Byzantine Protospatharius. Constantine's words are: “ἡ δὲ τοῦ Δαλενοῦ (ζουπανία) μηκόθεν ἐστὶ τῆς θαλάσσης καὶ ἐκ τἠς ἐργασίας ζῶσι τῆς γῆς ”

^{page 70 note c} “In planitie Dalmæ,” Diocleas, Regnum Slavorum (in Lucius de Begno Dalmatiœ, &c. Frankfort, 1666, p. 289.)

^{page 70 note d} Marci Maruli, Regum Dalmatiœ et Croatiœ gesta (in Lucius, op. cit. p. 306).

^{page 70 note e} Historia Salonitana, cap. xiii. “Istaque fuerunt Regni eorum (sc. regum Dalmatiæ et Croatiæ) confinia, ab Oriente Delmina ubi fuit civitas Delmis in qua est quædam Ecclesia quam B. Germanus Capuanus Episcopus consecravit sicut scriptum reperitur in ea.”

^{page 71 note a} Farlati, , Illyricum Sacrum, t. iv. p. 169Google Scholar.

^{page 71 note b} Farlati, op. cit. t. iv. p. 168 seqq. From 1685 onwards the diocese was placed under Vicars Apostolic.

^{page 71 note c} Albanesische Studien, p. 232. Hahn is of opinion that Delminium answers to a Gheg Albanian form δελμίν-ευ = sheep-fold, or sheep-pasture. He further compares the name of the Dalmatian city with that of the two Epirote towns Delvino and Delvinaki.

^{page 72 note a} Geog. vii. 5: “Δάλμιον δὲ μεγάλη πόλις ἦς ἐπώνυμον τὸ ἕθνος μικράν δ ἐποιήσε Νασικᾶς καὶ τὸ πεδίον μηλόβοτον διὰ τήν πλεονεξίαν τῶν ἀνθρώπων ”

^{page 72 note b} Cf. Glavinić, Bullettino di Archeologia e Storia Dalmata, 1878, p. 54. A. K. Matas, Prinos za iztraživanje tragova rimskih puteva u Dalmaciji (“A contribution towards investigating the traces of the Roman roads in Dalinatia “), in the Viestnik hrvatskoga arkeologiĉkoga Družtva, 1880, p. 32, mentions an alternative route along the right bank of the Cettina, but omits to specify the evidence on which his statements rest.

^{page 72 note c} According to Prof. Glavinić. loc. cit. traces of a Roman road are to be seen running from Lovreć to the Western part of the plain of Duvno.

^{page 72 note d} C. I. L. iii. 1892, 1908, 1909, 1910.

^{page 72 note e} Acta Concilii ii. Salonitani, in Farlati, , Illyricum Sacrum, t. ii. p. 173Google Scholar.

^{page 73 note a} Dr. Glavinić traced its course in 1856 from Runović past the villages of Ploĉe and Drinovce to the Upper Tihaljina. Bullettino, loc. cit. Cf. Dr. Blau, , Reisen in Bosnien u. der Hertzegoviva, Berlin, 1877, c. 42CrossRefGoogle Scholar.

^{page 73 note b} Cf. Hoernes, , Römische Alterthümer in Bosnien u. der Hercegovina in Archäologisch-epigraphische Mittheilungen, vol. iv. p. 37Google Scholarseqq.

^{page 73 note c} C. I. L. iii. 6362, 6363, one of a.d. 173.

^{page 73 note d} Cf. the Diploma of Vespasian, C. I. L. iii. D. xi.

^{page 73 note e} ii lvoensivm, C. I. L. iii. D. xxi. in Mœsia A. 105: v. lvciensivm et callaecorvm. A. 60 in Illyricum. D. ii.: A. 85 in Pannonia D. xii.: in Pannonia Superior D. xxxix. In the Notitia Utriusque Imperii (Occ. xlii. 29) is mentioned the Tribunus Cohortis Lucensis, Luco.

^{page 73 note f} From the occurrence of Koman remains at a succession of localities (Vitina, Kreindvor, Studenci, Gradnić, Ĉerin, Kruŝka), between Ljubuŝki and the Vale of Mostar, Dr. Hoernes conjectures that on this side a road branched off from Bigeste to the valley of the Narenta. (Cf. Blau, Reisen in Bosnien, &c. p. 42).

^{page 74 note a} Glavinić, Mittheilungen der k. k. Commission, &c. 1880, p. xciii.

^{page 75 note a} C. I. L. iii. p. 291 seqq. and p. 1029.

^{page 75 note b} Cf. Glavinić, Bullettino di Archeologia e Storia Dalmata, &c. Ephemeris Epigraphica, vol. iv. p. 86 seqq.

^{page 75 note c} See p. 45.

^{page 75 note d} Cf. Glavinić, Mittheilungen, &c. 1880, p. xciv.

^{page 75 note e} “Vatinius Imp. Ciceroni ‥‥ ex castris Narona.” (Ad. Fam. v. ep. 9.) Vatinius complains of the Dalmatian winter.

^{page 75 note f} A specimen seen by me at Metcovich, and found at Viddo on the site of Narona, was precisely similar in form to tumblers found in Kent, in the Saxon cemetery at Fairford, in the Frankish graves at Selzen in Rhenish Hesse, in Normandy, and elsewhere. Cf. Roach Smith, Collectanea Antiqua, vol. ii. pl. li. Lindenschmidt, Die Alterthümer unserer heidnischen Vorzeit, vol. i. Heft xi. t. 7, &c.

^{page 76 note a} “Ergo regale unguentum appellatur quoniam regibus Parthorum ita temperatur …… Nihilque ejus rei causa in Italia victrice omnium, in Europa vero tota, praeter irim Illyricam et nardum Gallicum gignitur.” (H. N. lib. xiii. c. 2.)

^{page 76 note b} “Iris ….. laudatissima in Illyrieo et ibi quoque non in maritimis sed in silvestribus Drilonis et Narona.” (N. N. lib. xxi. c. 19.) Pliny here names the city Narona and not the river Naron.

^{page 76 note c} Ἵριν θʼ ἦν ἔθρεψε Δρίλων καὶ Νάρονος ὄχθη

^{page 76 note d} Hist. Plant, lib. ix. c. 9.

^{page 76 note e} Cf. the French word for Iris, Glaïeul.

^{page 76 note f} Also as Bogiša, from Bog = God.

^{page 76 note g} xii. 74, ‘Cum tibi Niliacus portet crystalla cataplus.’

^{page 77 note a} Luc. Faunus, de Antiquitatibus Urbis Romce, c. x. Cf. King, National History of Gems or semiprecious Stones, p. 105.

^{page 77 note b} C. I. L. iii. 1792, 1793.

^{page 77 note c} Cf. Mommsen, op. cit. p. 291.

^{page 77 note d} Ad. Virg. Æn. iv. 262. Festus' words are: “Secespitam esse Antistius Labeo ait cultrum ferreum oblongum, manubrio rotundo, eburneo, solido, vincto ad capulum auro argentoque, fixum clavis æneis, ære Cyprio: quo Flamines, Flaminicæ Virgines, Pontifieesque ad sacrificia utuntur.” On Consular coins the instrument of sacrifice generally appears as an axe.

^{page 78 note a} Adding on in the case of the Tabula the omitted distance of xiii. m. p.

^{page 79 note a} lxxxv. m. p.

^{page 79 note b} Alterthümer der Hercegovina und der südlichen Theile Bosniens, vol. ii. p. 146Google Scholar.

^{page 79 note c} Accepting the correction of the xxii. given, in order to square with the xxv. m.p. given by Antonine as the distance, Narona—Dallunto.

^{page 80 note a} Lib, iv. c. 16: “Attamen Dalmatian plurimas fuisse civitates legimus ex quibus aliquas designare volumus quæ ponuntur per litus maris, id est: Burzumi, Aleta, Saluntum, Butua, Decadoron, Buccinum, Rucinium, Epitaurum id est Ragusium, Asamon, Zidion, Pardua id est Stamnes, Turres, Narrona,” &c.

^{page 81 note a} Σταγνόν. It is difficult to understand why Professor Tomaschek, op. cit. p. 36, should go out of his way to suggest a derivation for the word “Entweder aus einem vorauszusetzendem illyr. Worte Stamen,- Maul, Bachen, Hals, oder aus Gr. στενόν,—Enge.” The mediæval Latin form Stamnum, like the Stamnes of Eavennas, is simply a corruption of Stagnum, and it is to be observed that these forms illustrate a Ronman characteristic, cf. Latin Signum, Wallachian Semnu, &c. The Slavonic abbreviation of the name is Ston.

^{page 82 note a} A little to the west of the Narenta mouth the Drina is made to run into the Adriatic, coalescing in some strange way with the Cettina. The promontory of Sabbioncello is not so much as indicated. On the other hand the outline of the coast and islands in the neighbourhood of Salonee has much greater pretensions to exactness.

^{page 83 note a} Revue Archéologique, N.S. t. xxiv. p. 1Google Scholar, engraved pl. xv.

^{page 83 note b} See Numismatic Chronicle, N.S. vol. xx. “On some recent discoveries of Illyrian Coins.”

^{page 83 note c} A fragment of the Alessio wall is engraved in Hahn, Albanesische Studien, p. 122.

^{page 84 note a} According to the Itinerary of Antonine this station is only xii. miles from Scodra–probably an error for xxii. In the same way the Itinerary increases the distance between Cinna and Berziminium by two miles = m. p. xviii., as against xvi. in the Tabula. With regard to the name of the place I adopt the reading of Antonine, as being generally more correct than those of the Tabula, and as giving the name of an Illyrian queen. In Ptolemy it appears as χιννα.

^{page 84 note b} imp · caes · p · licinio · gallieno ∥ pio · felici · avg · pont · max | trib · pot · p · p cons · III · res| pvbl · docleativm · (C. I. L. iii. 1705). The best account of the ruins on the site of Dukle is in Koyalevski, Četyre mêsjaca v Černogorii. (Four months in Montenegro.) St. Petersburg, 1841, pp. 81–85, cited by Jireček, op. cit. There are massive remains of an aqueduct, town walls in the form of a parallelogram, columns and ruins of a temple or large building known as “Carski Dvor=the Emperor's palace,” sarcophagi with bas-reliefs and Latin inscriptions. Some new inscriptions from this site have been recently communicated by Dr. Bogišié to the Ephemeris Epigraphica. Doklea gave its name to the Slavonic region of Dioklia, from which in the early Middle Ages the Serbs extended the name More DioklitijsTco, “the Dioclitian sea,” to the Adriatic itself. The additional “i ” of the later form of the name, Dioclea, is said to have been due to an endeavour to justify its etymological connexion with the name of Diocletian. But the alternative name Dioclea appears too early to justify such an artificial origin. The authority for Diocletian's birth at Dioclea is the almost contemporary Aurelius Victor, whose statement on this head is clear: “Diocletianus Dalmata, Anulini Senatoris libertinus, matre pariter atque oppido nomine Dioclea, quorum vocabulis donee imperium sumeret Diodes appellatus, ubi orbis Romani potentiam cepit Grajum nomen in Romanum morem convertit.” (Epit. c. xxxix.) It is to be observed that Constantine Porphyrogenitus, while placing Diocletian's birth-place at Salona, makes Diocletian found Dioclea: “Τὸ κάστρον Διόκλεια τὸ νῦν παρὶ Διοκλητιανῶν κατεχόμενον ὁ αὐτὸς βασιλεῦς βασλεῦς Διοκλητιανὸς ᾠκοδόμησεν.” (De Adm. Imp. c. 29, and cf. c. 35, where he speaks of it as being then ἐρημόκαστρον, as we should say, “a waste Chester.”) Ptolemy mentions a Διοκλεὶα (al. Δό) in Phrygia; not unknown to ecclesiastical history.

^{page 85 note a} This vase is now in the Musee Basilewsky in Paris. It is described and illustrated by the Rossi, Cav. di in the Bullettino di Archeologia Cristiana (Rome, 1877, p. 77Google Scholar). The linguistic peculiarities of the inscriptions on it suggest interesting comparisons with the Romance survivals in the dialect of Ragusa. See p. 32, Note.

^{page 85 note b} It appears to me probable that the obscure “Diode,” placed between “Lissum “and “Codras,” or Scodra, in Guidonis Geographia (114), stands for “Dioclea” a hint that the name appeared under this form in some copy of the Tabula.

^{page 85 note c} The older Serbian name of Podgorica was Ribnica, still preserved by the small stream that flows beside its walls. (Cf. Jireˇek, op. cit. p. 20.) This place derived its importance as lying in the centre of the district of Zenta.

^{page 86 note a} Geog. Ravennas, p. 211 (ed. Pinder et Parthey): “Item juxta Burzwmon est Civitas qua dicitur Medione,” &c.

^{page 86 note b} Prof. Tomaschek neglects the abiding conditions of intercourse as fixed by the physical configuration of the country in seeking the site of Aleta out of the Zeta Valley: “Vielleicht östlich von Cettinje, bei Gradac oder Uljici,” op. cit. p. 42. The name Aleta itself he compares with the Albanian hel [pi. heljete (hejete)] = a point, as of a lance, &c.

^{page 88 note a} The attempt to identify Sallunto (ii.) with the Slansko Polje (Hoernes, , Alterthümer der Hercegovina, vol. ii. p. 149Google Scholar), on the ground of similarity of name, is too hazardous.; and the same applies to its comparison with either of the two Slanos. The Serbian form of the Illyro-Roman word, if directly adopted and preserved, would be Solunat: Tomaschek's suggested comparison with the name of the village of Zaljut (inadmissible on other grounds) must therefore be discarded. I would suggest the identification of this “Sallunto “with the ” Lontodocla “in the region of Dioclia, mentioned by Constantine Porphyrogenitus (op. cit. c. 25). It might be a “Sallunto-Docleatium,” to distinguish it from the other “Sallunto “on the same route further to the West.

^{page 88 note b} Lexicon, s.v. Vukova Megja.

^{page 88 note c} “Od jednoga kraja do drugoga ove megje prijekijem putem ima oko Četiri dana hoda; a kad bi se islo preko gudura i litica pored nje bilo bi mnogo više.” (“From one end to the other of this boundary-wall, as you go forward, is about four days' journey; but were one to go along it through glen and over ridge it would be much further.”) Vuk, loc. cit. This description recalls rather the up and down progress of a Roman frontier-wall, such as that from Tyne to Solway, than any Roman road.

^{page 89 note a} This, of course, is historically impossible, as St. Sava died at Tirnovo, in Bulgaria, and must therefore have been carried to Mileševo from the East.

^{page 89 note b} Jireček, Die Handelsslrassen, sect. 11. Von Cattaro nach Plevlje (p. 72).

^{page 90 note a} Römische Alterthümer in Bosnien und der Hercegovina, vol, ii. (in Arch. Epigr. Mitth. vol. iv. p. 47Google Scholar).

^{page 91 note a} C. I. L. iii. 1782, i · o · m ∥ chor ∥ tali. In the present inscription the H of chor(tali) is obliterated, hut doubtless was originally contained within the c.

^{page 91 note b} It seems to me probable that this line Nikšić—Gacko—Gorazda is indicated by the Geographer of Ravenna, who refers to a line of stations, “Sapua—Bersellum—Ibisua—Derva—Citua—Anderba.”

^{page 92 note a} Acta Concilii II. Salonitani, in Farlati, Illyricum Sacrum, t. ii. p. 173. The identification of Stantinum with Stagno, urged by Dr. Hoernes on the strength of the existence of the later Župa Stantania from Ston, the Slavonic form of Stagno, is hardly admissible, since the Acts of this Council of Salona show as yet no trace of Slavonic settlement or nomenclature in that part of Dalmatia which they concern.

^{page 92 note b} I have referred to these in my work on Bosnia (2nd ed. p. 361), where, however, Tassovčić is wrongly printed Tassorić”.

^{page 92 note a} The name Dabar suggests a connexion with the important tribe of the Daversi or Daorsi, who inhabited the ranges East of the Narenta at the time of the Roman Conquest. In the Romance dialect of Dalmatia (as exemplified by its surviving remnants in that of Ragusa), v is changed to b.

^{page 92 note b} Though the Itinerary of Antonine seems to give us authority for striking off 10 m. between Dilunto and Narona, see p. 79.

^{page 94 note a} C. I. L. iii. 1763, 1764, 1765.

^{page 95 note a} C. I. L. iii. 1761.

^{page 95 note b} The traces of the Eoman road above Plat are doubtless the same as those observed by Dr. Constantin Jireček in the neighbourhood of Ragusa Vecchia. (Die Handehstrassen und Bergwerke von Serbien und Bosnien während des Mittelalters, p. 8.) Dr. Jireček observes that the “via vetus quse vocatur via regis “is mentioned in the Ragusan Catasters of the fourteenth century, and supposes, with great probability, that its Slavonic name was “Carski Put,” “Cæsar's Way,” a name by which Roman roads were generally known to Serbs and Bulgars in the Middle Ages, and answering to the Byzantine ὁδὸς βασιλική. In 1880 I took Dr. Hoernes to visit the traces, and his impression of their appearance as recorded by him (Römische Alterthümer in Bosnien und der Hercegovina, vol. i. p. 2Google Scholar) agrees entirely with my own.

^{page 97 note a} C. I. L. iii. 3207, 3208, 3209, 3211. The title given to Julian on these is “Victor ac triumfator totiusque orbis Augustus, bono reipublicæ natus.”

^{page 99 note a} I have a denarius of the Empress Lucilla from this site, with the reverse legend ivnoni lvoinae.

^{page 102 note a} The construction of this road is recorded on a milliary column found at Feltria (C. I. L. v. 8002): ti · clavdivs · drvsi · f ∥ caesar · avg · germa∥nicvs · pontifex · maxv∥mvs · tribvnicia · potesta∥te · vi. cos. iv. imp xi p. p. ∥ censor · viam · clavdiam ∥ avgvstam · qvam · drvsvs ∥ pater · alpibvs bello pate ∥ factis · derexerat · mvnit · ab ∥ altino · vsqve · ad · flvmen ∥ banvvivm · m. p. cccl. Another similar was found at Meran (C. I. L. v. 8003).

^{page 105 note a} Ramberti, Delle cose de Turchi, Libri tre, Nel primo, il viaggio da Venetia à Costantinopoli, &c. p. 5, (In Vinezia, nell' anno m.d. xxxxi. In casa di Maestro Bernardin Milanese.)

^{page 105 note b} Mentioned already in 1380 as the site of a Ragusan customs station and small commercial colony. (Liber Reformationum Majoris, Minoris, et Rogatorum Consiliorum, Civitatis Bagusii. Cf. Jireček, op. cit. p. 75.)