ec Seopr ase phos ese,

essa An nr an oer pea

peer γί το Se

δα Fare Berra re

=

BERS

ae

css 32 pies c : : Z : : τε tas was nae sere pte τ : ΣΝ ΠΣ ΩΣ 3 Lege

We te Ae ee me

τς ς22. ὅπ. ΧΙΣ-Σ

~

3 ae “2 ΣΟ oes / oo ae recog eet pereneoorpe ssa

τ᾿

THH JOURNAL

HELLENIC STUDIES

TAWVHUOL BHT

a0 a

a Oe ΡΠ ΠΤ ΓΑΙ LA

Che Society for the Promotion of Wellenic Studies

THE JOURNAL

meee ΝΟ SU DTS

VOLUME XVII. (1897)

1897

KRAUS REPRINT Nendeln/Liechtenstein 1971

Reprinted by permission of THE SOCIETY FOR PROMOTION OF HELLENIC STUDIES

KRAUS REPRINT A Dwision of KRAUS-THOMSON ORGANIZATION LIMITED Nendeln/Liechtenstein 1971 Printed in Germany

CONTENTS

PAGE Kules of the Society ... ... .. 1s ie pe ee 1X εὐρν of Aomvers and, Mem betieccs. κα very yeh ais: Pwo 1 xv Proceedings of the Society, 1896-7 . (ods) Na ες ees Sie es XXXill Proceedings of the Cambridge Branch of the Society... ΕΞ xl δ πον που νήσοις, xis seb Une eee dees wes 5A xlv List of Periodical Publications in the Library Pe ey ore l warooue Or Lantern Slides τ. eke Tye τὸν τρις 9) eee lili ALLEN (T. W.) ...... The Text of the Homeric Hymns. Part If... 45 » 55 ... «The Text of the Homeric Hymns. PartIV. ... 241 ANDERSON (J. G. C.) .... The Road-System of Kastern Asia Minor with the Evidence of Byzantine Campaigns (Plate L.) ... 22 ve ᾿ ot .. A Stuunmer in Phrygia: [. (Plate XT.) Ἄν 396 Bury (J. b,) ee LG NBER INTOD, a: e- toate eee ρου : υ3 CALVERT (FRANK) ... On the Tumulus of Choban ἀηρροι in the 'Troad ... 319 Crowroot (J. W.) ... A Thracian Portrait (Plate XI.) ... ... ... ... 952] Kpear (C. E.) ... ... Two Stelae from Kynosarges (Plate IV.) ...- .... 174 Evans (A. J.) .... 0... Further Discoveries of Cretan and Aegean Seript : with Libyan and Proto-Egyptian Comparisons (Plates TX. 20.) uo." cee! Wi A ase Gea” \oacee 2 δ Garpnen (EK. Α.)... ... Caeneus and the Centaurs: A Vase at Harrow {ΘΙ ΒΥ 1), ἃς ee ae δου ee oe eis πο see GARDNER (Percy) ... The Mantinean Basis: a Note... ... ... ... 120 Grunpy (G. B.) ... ... Artemisium PREY Anerey. ΜΝ oh tet 9.5 i nd .. .. The Account of Salamis in Herodotus... ... 230 Hon (Gb. ) ... ... Notes on Additions to the Greek Coins in the British Museum, 1857-1896 (Plate IL.) ... ... 78 Hurron (C. A.) ... ... Votive Reliefs in the Acropolis Museum (Plates Δ ΠΣ: αν ΜΆΑΟΘΚΕΝΖΙΕ (1}.)... ... Excavations of the British School at Melos: the Site of the ‘Three Churches’ (Plate V.)... ... 122 Munro (J. AntHur R.) Inscriptions from Mysia Pate Oe dar, CERT, Tae χοῦ Mees (0 t.), dona)» τὸ | Exeavations in Cyprisin 1894 ..... ...5 ca ss (LSS hoperts (W. Rirys) ... The Greek Treatise on the Sublime: its Modern Interest τ Naethe Weds, Sue (eeeetl sinc” Van _ my ny ... The Greek Treatise on the Sublime: its Author- ite Leable) iis). Sear a? Ss. τ 1995 Peete (ORCL)... 29. Inscriptions from Melos... ὦ «en > <8 aw Ι Watters (H. B.) ... On some Antiquities of the Mycenaean Age

recently acquired by the British Museum PP iets ee, tein, Sees Rh eee τ 65

CONTENTS.

LIST OF PLATES.

Map of Eastern Asia Minor illustrating the Road System. Greek Coins added to the Collection of the British Museum. Engraved Gems added to the Collection of the British Museum. Reliefs from Kynosarges.

Melos: Site of the Three Churches

A Vase at Harrow: Caeneus and the Centaurs.

Votive Rehefs from the Acropohs, Athens.

Primitive Cretan Vrism-seals of Steatite.

A Thracian Portrait.

Map of the District of Laodiceia.

CONTENTS

LIST OF TLLUSTRATIONS

Rock-Tomb new Klimatoboun: Stelae from Melos Two Gold Fibulae Bronze Knite ; Bronze Chisel Marble Pyxis from: Aegina Seal of Rock Crystal from Cyprus Amethyst from Mycenae Karly Pottery from Cyprus Bowl from Saqqara Vase from Erment

7 » Calymna Kylix ,, Terracotta Model of ἀπήνη... Plan of the lmperial Quarter of Constantinople Field of ‘the Three Churches, Melos Statues found at Melos Pottery from Ag. Paraskevi The Neighbourhood of Kalopsida (Map) ... Various Objects from Kalopsida The Neighbourhood of Larnaka (Map) Laksha tu Riu: part of contents of Tomb 4 .. Engraved Chalcedony, Larnaka... Larnaka, Turabi. ‘Tomb 56 Haematite Scaraboid, Larnaka ... Graeco-Phoenician Pottery from Larnaka Lypes of Wine Amphorae from Larnaka... Amphorae from Larnaka

Votive Figures, Kamelarga

IN

vil

THE TEXT 4 ‘ 4 44 ᾽

s] le “1 ὡς

125, 126, 127, 128, 129 137

139

140, 141

149

150, 151

152

157

vill CONTENTS.

PAGt Plan of the Battle of Sakumis ; Original Position . 237 = ἐς 2 ts 3 Advance 238

-- -" ἘΞ 5: Ea Contact ... 239

A Vase at Harrow 295 Votive Terracotta Reliefs from Athens ... 309, 311, 312, 313, 314, 315, 316, 317 Tamutus of Choban Tepeh m the Proad οι, 320 Coin of Kotys 321 Three-sided Seal of Steatite from Kalokhorio ... 329 Steatite Gems, Mallia and Elunda 334 Steatite Gems, Crete .. 335 Steatite Seal from Siteta 336 White Steatite Seal, Crete... 37 Pictographie Symbols =... 25 5. Se τ <2 ull Eine paty oD Oy OLDE Grae Convoluted Agate Seal-stone from Gortyna 541 Signets from Crete —. wee he OP EE ag et ce ee Lentoid Bead of Steatite, Knosos 346 Black Steatite Whorl, Knosos 347 Comparative Examples of Monogrammatie Signs B48 Yellow Steatite Seal, Kalamatka “48 Inscribed Marble Vase from Cerigo ... 349 Inseribed Libation Table from Diktaean Cave... .. ere ἦν nf 002s GOD LoDy Sketch Plan of Atrium of Cave of Psychro (Diktaean Cave) δῦ Inseription on Diktaean Libation Table ... bo Signs from Diktaean Cave, with Parallel Forms 360 Prism-seal of Steatite from Karnak.. $62 Three Inscribed Cylinders ... 364 Libyan Bead-Seal from near Constantine... 368 Comparative Table of Human and Animal Types ... 360 Proto-Egyptian Clay Figure. Marble ‘Idol,’ Amorgos ... 3381 Signs on Clay Vessels from Abydos. . 383 Comparative Table of Sign-Groups 384 π᾿ ᾿ Signs 386 Sketch Section of Levels near Liaodiceia ... 407

INSCRIPTIONS FROM MELOS.

Tue following inscriptions were cither copied or excavated by the members of the British School during the spring of 1896, some during our preliminary visit in January, but the greater part during March—May, when we were living at Trypete. The majority of the inscriptions of the island are the product of the promiscuous amateur digging which has been going on probably almost continuously for nearly a century: they are consequently usually in the hands of peasants, who in most cases can only give one vague traditions regarding their original provenance. The personal interest of their present owners in them is naturally very small: it awakens to a languid existence when from time to time a foreign visitor makes them the subject of notebooks and (occasionally) of drachmas: but for the most part no sort of care is taken for their preservation, and if an inscription is to be employed as a paving-stone, it is usually the inscription side which meets the foot and the weather. Of course this state of things is not confined to Melos: I only mention it because there is at last serious talk of collecting the inscriptions and perhaps some of the other more important antiquities of Melos into a Jocal museum: whether as the result of our urgent representations or of those of the German visitors of last year, or as the result of years of suggestions, it inatters very little. It is greatly to be hoped that the good intentions of the demarch of Castri nay soon be carried out.

The inscriptions of this island have been the subject of numerous publi- cations ; a list of the most important is given below.! In the summer of 1895 Melos was visited by Messrs. Hiller von Gaertringen and Schiff, who were vccupied in collecting the inscriptions of this group of islands for the forth- coming volume of Js/vnd Inscriptions. Thanks to their kind cooperation I was enabled to climinate from our series those which had been previously published, or seen by them, and the list now printed, with one or two excep- tions (here included because the previous publications were defective), con- sists, so far as we are aware, only of examples hitherto unknown.

1 Besides the collections in C. 1. G. 2424 Mitth. i, 246; ii. 223; xi. 111; xxi. 220; Ricci ete., 7. G. A, 412 ete. Ross in Jaser, Gir. ned. in Mon. Ant. Lincei, ii. (1894), 276 ; and a Latin iii. 226 ete. and Peisen, iii. 19, see Ann, dell’ — inser. in C. J. LZ. iii, 490. Vesides these there Inst. 1829 p. 343; 1843, p. 332; Bull. dell’ is a small series published in Trans. Loy. See. Tast. ii, (1830) p. 195; Rangabe, Ant. Well. ii, Lit. second series, vol. v. p. 29 fiom Lieut, (1855), 1193 ete. ; ἘΦ. "Apx. 1859, 3507 etc. ; — Leycester’s copies, but the copies do not seem Bull. de Corr. Hell. ii. 521; iii. 256; Ath. — to have been preserved.

H.S.—VOL. XVII. Lb

~

2 INSCRIPTIONS FROM MELOS.

Melian inscriptions naturally fall into three main divisions, correspond- ing to the chict political vicissitudes which the isliund underwent: the first class covers all those in the Mechan character, which appear to belong to a period previous to the destruction of the town in 416 B.c.; the second class consists of those attributable to the brief period of the Athenian occupation ; and the third includes all those subsequent to Bc. +06, when, after the battle of Acgospotami, Lysander reinstated the Mchans (Xenophon, 7701]. i. 2.9; Plutarch, Lysand. xiv. 441) in the possession of the island. Of this latter class it is a curious fact that by far the largest proportion belong to Imperial times; indeed, it is extremely rare to find examples (written at least in other than Melian characters) which can with certainty be referred to a period be- tween 400 Bc. and the time of the Emperors. Of course it may be that the returning Meliaus went on for some period using their own style of lettering —even this is not as yet proved, and is at best only a probability—but it is most improbable that they can have resisted the imtroduction of the κοινόν for more than a century at the most. Possibly further excavation may con- tribute material which will supply the gap: but meanwhile the coimeidence seems worthy of notice.

A general discussion of Melian epigraphy will be more suitably left to the publishers of the Island Corpus, when the whole material will be brought together. I shall confine myself mercly to recording the few notes and obser- vations on the inscriptions of the first class which were jotted down in our notebooks at the time when these inscriptions were copied or discovered.

The inscriptions in the Mclian character, as is well known, consist almost wholly of epitaphs, cut on slabs of the rich reddish-black trachyte which is the characteristic building material of the prac-Roman architecture of the island. With the disappearance of the Mehan lettering, this material also goes out of use, and the iuscriptions (like most of the architecture) of the later period are invariably inimarble. Judging from those examples which are more complete, the form is also invariable, presenting an oblong face which is surmounted at the upper extremity by an obtuse-angled triangle, roughly suggesting a pediment. lnimediately below this pediment follows the in- scription, on a surface which has been carefully prepared by tooling; as a rule « considerable space remains below, unoccupied by the inscription, and this is usually left with a somewhat rougher surface. In No, 20 this lower portion projects about 3 cm. beyond the inscribed surface, but the lapidary, being pressed for space above, has cut on it the final € of his inscription. This in- scription, it will be observed, is couched in a formula different from that of the ordinary Melian epitaph; possibly this fact is accountable for the difti- culties which the lapidary seems in this instance to have found.

As the back of the stone is also as a rule left rough, it would seem that this class of inscriptions was intended to be partially sunk in the ground, probably against the eutrance to the tomb, in such a way that only the upper portion with the lettering was visible: in this respect they seem to have been followed by the stele of class 11., except that these terminate in a shaft specially prepared for insertion in the ground or a socket (see fig. 3). So far

INSCRIPTIONS FROM MELOS. 3

as can be ascertained, the inscriptions with Melian lettering have been chiefly found in the neighbourhood of Klimatobouni, where the tomb was usually in the form of a chamber, cither cut horizontally into the rock beside the ancient road, or (as the one in fig. 1) sunk into the ground.

As regards date, the evidence is unfortunately extremely scanty. Of the four periods given in Roberts (Gh. ELpigr. p. 36), those which can be assigned to period i, at present known are very rare: we have found no examples. Between periods ii, and iii. there is a stage of transition, when the sigma is found on the same stone in both forms, M and = (see nos. 12, 14, 20). In the best period the desire seems to be felt of avoiding straightness of line in the formation of the letters, which are here composed of a series of firm sweeping curves (see no. 2). In the latest examples there is a tendency to omit the horizontal lines between the rows of letters.!. If it be true, as the peasants assert, that no. 3 was found in the tomb-chamber which contained the red-figured vase with a Gigantomachia now in the Louvre, this would point to a date of about 430 νον for the inseriptions of the best Melian period ; but of course it does not necessarily follow that the vase and inscrip- tion were. actually contemporary.

As was doubtless the case with most Greek inscriptions, these also appear to have had the letters usually painted ; but whereas the ordinary custom was to colour the lines alternately red and blue, in the Melian inscriptions of class 1 one colour alone seems to have been employed: wherever such colouring could be proved to exist, this was invariably a rich vermilion, which in some cases was traceable in consecutive lines.

The inscriptions of this class are all reproduced to a scale of 1 of the actual size, except nos, 23 and 46 which are 4'5.

ip ᾿Εχετίμα Λυσι[δίκου On a slab of red trachyte in the courtyard of the house of Nicolaos Tsoulios at Trypete, where it is used as a paving-stone ; complete at sides,

1 Pollak in Ath. Mitth. xxi. p. 222 regards — against the third period ; but he gives no reason the absence of horizontal lines as evidence for for this view, which seems to me highly im- attributing an inscription to the second, as probable.

B 2

Ι INSCRIPTIONS FROM METLOS.

broken above and below, surface in excellent condition. Ht. 4a; 1m. W.°32 m. Published Rohl, Zaser7. Gr. no, 422; Roberts, Gh. Epigr. p. 33, no. A, and elsewhere.

2. EvOv?|xpitos . ay. λε[{|δα()

On a slab of red trachyte in a wall at Trypete, belonging to Peros Antonis Kyritsis. The owner took it out for us. Complete on r. side: the edge on l. side appears to be complete at the back, but is broken away on the face. Ht. 37m. W.:295. Thickness ΤΟΥ m. Space between ruled lines ‘09 m.

No. 4. 3. Nixoda ?|uos ᾿Ιμα[ράτου ¢ On a slab of red trachyte in the house of Nicolas Antonis Kyritsis at Klimatobouni. Ht. 86m. W. 32m. Space between ruled lines “12. τη. Said to have been found in a family cave-tomb, which also contained the great

red-figure Gigantomachia vase now in the Louvre. For ‘Iwapatos ef. CLG. lll. 547 -

a

ἘΞ: ορος Δικαφ[ίλου ?

On a slab of red trachyte, used as a paving-stone in the small church of St. Georgios at Klimatobouni: it lies near the centre of the nave, near the wooden screen. The surface has suffered considerably from the feet and candle-grease of the pious followers of the saint. Complete at sides, broken above and below. Ht.:27 τὼ. W.°23 in. Space between ruled lines ‘077 in.

et PS

CM

0, 5.

EM Νικοφ]ύλη (2) [ΕἸσθῆτος (ἢ) Ou τι slab of red trachyte, used as a paving-stoue in the courtyard of the house of Andreas Joanunes Babouni at Klimatobouni, where it was placed by

INSCRIPTIONS FROM MELOS. 5

his father about twenty years ago, Broken on all sides except on the r., where itis complete. Ht. “33 τὼ. W.-:27 m. Space between ruled lines “102 m.

6. ... op[ak 1] ΤΠΙραξ[εἸκλέους.

On a slab of red trachyte, in the house of Michael Joannes Bechos at Trypete. Ht. “34. m, W. “31 m. by ‘10 m. thick. Complete on both sides : the face worked away on Τὶ upper side. The letters in the lowest line are somewhat contracted into the space on the r. It is noticeable that p is here quite clearly P, instead of the usual R.

7. Op ]εψ[εἸκλῆς......

On a slab of red trachyte, in the wall of the courtyard of Emmanuel Bechos at Trypete. Complete on r. side only. Surface in good condition, Ht. “21 τη, W. 26 ἢ. I have read Θρεψικλῆς and not Δεξικλῆς, because the form of &

£) is perfectly well proved for Melian inscriptions of this period (see nos. 6, 15; I.G.A. 431, &c.). It does not seem probable then, as has been recently asserted (Ath. Mitth. xxi. pp. 221, 254, 432), that the alphabet of Melos at any period followed that of Thera in the use of V = ξ. In the inscription which gave rise to this theory (Ath. Mitth. xxi. p. 221), ΤΠΠραξικύδης seems at first sight certainly the more likely form; but considering the strangeness of Melian names in general, we need not reject Tpawixvdns as impossible. I may here remark that according to my copy of that inscription the initial letter cannot possibly be aM. In the companion inscription from the same house, the patronymic appears to be Evpudvax[tos: cf. Bull. de Corr. Hell. ii. p. 521, 1, Εὐρνανακτιδᾶν; the 6th letter is wrongly given there as M.

8. . στωσία Lwroxp...? From a house in Trypete: upper part much worn. Ηΐ. “30 τη. W. 91.

fi INSCRIPTIONS FROM MELOS.

9: ᾿Αρχ]ήἷἶδαμος ᾽Λρσι.

From a house in Trypete: upper part of surface much injured. Com- plete on both sides. Ht. 39 in. W. 24. τη.

10. Nuxora Kad... ..

On a slab of red trachyte in the courtyard of the house of Andreas Joannes Babouni at Klimatobouni. It was found with a quantity of other uninseribed fragments in a cave on the street of tombs near Klimatobouni. The two sides are complete, and the upper edge from the 1. to near the centre, showing that the top was of pediment-form. Ht. 33m. W.°30 in, Space between ruled lines 10 χη.

No. 11. ΝΟ. 19.

11: _... Πλλαγόρα.

On ἃ 5140 of red trachyte, in the house of Joannes Mourachis at Klima- tobouni, where it is built into the front wall, on the 1. of the door, laid on its r. side. It is apparently complete at the sides and lower edge, though only the r. side has a good edge. Surface in good condition. Ht.°32m. ΝΥ, “29 m.

Here it may be noted as a sign of transition that though the y is Γ, the o is C: it seems therefore to belong to the third period.

12. .... actos Broken above and below: 1. side and part of r. complete. Ht. 19 τη. ες a No. 13 Now 14: 13. Ἥρα ?]«Alwv Φασί .. ...

On a slab of red trachyte found by us in the wall of the garden of Nicolas Tsoulios at Klima, and now removed to his house there. Complete only on the upper part of the 1. side, and perhaps on the right edge. Ht. ‘27 m, W, *29 m,

INSCRIPTIONS FROM MELOS. 7

1. Ix [μ|]Ἱνὸς (7)

On a slab of red trachyte, found on Mareh 28th in the exeavation at Klima on site B (see the plan ΠΝ xvi. p. 848), beside the base of a column belonging to the lite Roman stoa: it lay on the pavement, ΑἹ ΗῚ depth of about & ft. below the level of the soil. The inscription is complete below, and the stone is complete on the 1. side. Ht. 92 χὰ, W.°:23 1m.) The surface is in excellent preservation, and the rea colour in the letters was vers brilliant when if was first taken out of the ground. From the position in which it was found it must have been used as building material for one of the late walls of this site: from which it would appear that the destruction of the Melian chamber tombs must have begun at an early period of our era.

ἘΝ

Nw. 15. No. 16.

1: .... Bevapyou.

On a slab of red trachyte, in the house of Basilios Michael Kyritsis at Trypete. Ht. 905 m. W. 265 m. by ‘09 m. thick. Complete edge on each side : surface in good condition: traces of red colour in the letters.

16. Βουλίας M....

On a slab of red trachyte, brought by one of our workmen to Trypete. Complete edge on the 1. and at the apex of the pediment. Ht. 305 τη.

W. 26 m.

No. 17. No. 18. 17. Kidpis ὐρυφι[λίδου ? On a slab of red trachyte, used as the doorstep in the house of Francescos Lilis in Klimatobouni. Complete edge on each side. Ht. 36m. δ. 99 χη. Lines ‘105 m. apart. Letters 10. m. Jugh.

8 INSCRIPTIONS FROM MELOS.

18. “Αγνοσ[θένης 1.

On a fragment of red trachyte, taken for us from the wall of a field out- side Klimatobouni: broken on all sides. Ht.-16m. W.-10 m. by 17 m. thick.

No. 19. ΝΟ, 20:

ΠΩΣ επ. {LO

On a fragment of red trachyte, used as ἃ paving-stone in the path leading down from Trypete to the house of Nicolaos Tsoulios: removed by us to Trypete. Broken on all sides, but surface in good condition: no ruled lines between the rows of letters. Ηὖ, 1ὅ m. Greatest width ‘17 m.

20. II 1]ασιθέα ἀδελφεοῖς.

On a slab of red trachyte, in the house of the widow Marina Constantina Kyritsis at Trypete. Said to have been found in her field on the farther side of Klimatobouni, two or three years ago: probably in a cave-tomb such as are of frequent occurrence in that neighbourhood. Complete edge on both sides. Ht. 45m. W. °32 m. by ‘15 m. thick. Below the ruled lines of the main inscription, the surface is raised about 3 em. and left rough, and on this surface the final € is inscribed. Evidently the lapidary had found the space prepared for him insufficient when he got to the end of the first word, and not even the diminution in the size of letters of the second word gave him room enough. The occurrence of the Ionic form ἀδελφεοῖς in a purely Doric series of inscriptions is interesting.

iB ATA

21, Διὸς Κατα͵ιβάτα

Engraved on the West side of a mass of rock which projects from the soil at the S, E. corner of the ‘Three Churches field:’ (see plan /oe, cit, site E).

INSCRIPTIONS FROM MELOS. 9

Judging from the form of the rock, it probably represented a natural open air altar; unfortunately the greater part has been hacked or blasted away to furnish material for walls. We dug all round it, but the results seemed only to show that it was natural outcrop. The surface is considerably weathered, but the letters are perfectly clear. Ht. of letters ‘07 m. Width from B to the final Α “27 τὰ, This inscription may be restored from the

MOSK Ay AIBAT-

Engraved ona rock at the top of the hill Perianti (see plan Joc. cit., there erroneously named Bereadi), near the S. edge of the hill; the rock has been cut roughly into the form of an altar, but is now half broken away. The inscription is on the upper surface, and reads from the W. side. Published incompletely in Ath. Mitth. i. (1876), p. 248, No. 6. Total length of letters in upper line ‘265 m. Ht. of A 034. For other instances of this formula, see the article by M. Delamarre in the Revue de Phil. 1895, p. 129. M. Perdrizet has kindly called my attention to yet another instance—Plutarch Demetrius x. describing the entry of Demetrius into Athens, says καὶ τὸν τόπον, ὅπου πρῶτον ἀπέβη τοῦ ἅρματος, καθιερώσαντες καὶ βωμὸν ἐπιθέντες Δημητρίου Καταιβάτου προσηγόρευσαν."

The exposed and prominent position of these two rocks is appropriate to the divinity ‘that descends in thunder and lightning’—The altar of Zeus Kataibates at Olympia was protected by a fence. Τοῦ δὲ Καταιβάτου Διὸς προβέβληται μὲν πανταχόθεν πρὸ τοῦ βωμοῦ φράγμα (Paus. v., 14, 10): possibly it was a natural rock similar to ours, and being thus likely to have its sacred character overlooked, required some such special protection.

aN EAITIA SR ONEA

No. 23.

23. Μελιτίας Πωνεα (7)

In the hill outside Klimatobouni to the 8.W. is a series of underground tomb chambers excavated in the sandstone: on the wall of one of the largest of these the above inscription is engraved.

The chamber measures 19 ft. 10 in. by 11 ft. 5 in., and is entered by a doorway at the bottom of a deep pit, from which two steps lead down to the

1 For other instances of the same formula see ©1672b. ‘The first of these is a steld marking an C.I.A. vol. iv. (suppl.) pt. ii., nos. 16595 and ἄβατον of Zeus Kataibates.

10 INSCRIPTIONS FROM MELOS.

Hoor of the tomb. which lies about 13 ft. below the level of the soil. On the side opposite the doorway a single recess has been hewn out (marked A in fig. 1) at a height of about 2 ft. from the floor, sufficiently large to contain a body: and three similar recesses occur in each of the side walls. In the recess marked B in the plan, an opening has been made (apparently for the purpose of rifling the tomb) communicating with an excavated passage from

the soil level.

Section through Cc.

Fic. 1.—RockK-TOMB NEAR KLIMATOBOUNI.

Above A the inscription is cut in deep letters of careful workmanship but diminishing in size from 1. tor. The height of the M is no less than 8S inches: that of the final A is 53 in. Published by Ross, Reisen iii. 16, and Jaser, Gr. Inca, 235, but incompletely.

INSCRIPTIONS FROM MELOs. 11

The following inscriptions seem to belong to a period Jater than the fifth century BC. when the Melian character was no longer in use,

XAPH ZATTOAAOPANOYS XAIPE

24. odes Oey

On a fragment of bluish marble, used as a paving-stone in the courtyard of the house of Nicolaos Chronis at Trypete: the owner says that it was broken from the same stone as the inscription beginning ᾿Ακροπόλεως (Schiff No. 52) which belongs to the same owner, and was found at Tramy- thia, but the scale of the letters in this fragment seems too large. Ht. 35 τη.

W. °33 m.

~ ὡς > ΄ - 25. Χάρης ᾿Απολλοφάνους χαῖρε.

On a stele of yellowish marble surmounted by a pediment with a palmette crowning the apex. Nearly complete, but broken away below the inscription. In the house of Georgios Geouroukes at Klimatobouni. Ht. ‘61 m. Greatest width "25 m. tapering slightly towards the top. Thickness ‘085 m. The inscription occupies the extreme lower part of the stele.

TIOOEINONOAPFIIYY AHS FN ENEAL XA|PEIN

No. 26. No. 27. 26. Ποθεινὸν Θαργηλίου χαίρειν (Fig. 2).

On ἃ stelé of yellowish-white marble in the house of Georgios Geouroukes at Klimatobouni. Complete, with vivid traces of colouring. Above, a pediment with broad lower moulding; at the base of the stelé, a broad moulding on which has been painted egg and dart pattern. The colour on the main surface seems to have been chiefly vermilion ; but in this, as in

12 INSCRIPTIONS FROM MELOS.

the preceding case, the position of the inscription at the base of the stele may be due to the fact that the upper portion was painted with a design : but if so, no traces of it beyond the red colour already described can now be identified. The lower part under the moulding tapers rapidly into a squared shaft. Ht. from lower moulding to apex of pediment ‘505 τὰ. W. °225 τη. by 10 m. thick.

The purpose of the squared shaft is seen from Fig. 3, which represents a third marble stelé (uninscribed) which is in the same house. This stelé,

TIO@EINON@APrHAICY ΧΑΊΡΕΙΝ

Fic, 2.—Sre_t rrom MELOs. Fia. 3.—Srevtt FROM MELOs,

which measures ‘635 m. high by “38 m. wide, has within a sunk panel at the top (unfortunately partly broken away) a relief representing ἃ man in an himation holding in his 1. hand a roll, presumably a poet or philosopher beside whom stands a boy looking up at him: both figures stand facing the spectator. Here also the lower part terminates in a squared shaft, which is evidently intended to fit into the socket of a rectangular base, found with it, so that the stelé may stand upright. As the back{of the stelé is left rough it probably stood in this way against the door of the tomb.

INSCRIPTIONS FROM MELOS. 13

Nos. 25 and 26 (and probably also the sculptured stelé) were found in 1866 by the side of an ancient road running from Klima to Adamas, which seems to have had tombs on both sides: it is now buried several metres below the present level of the soil, but its course can clearly be made out skirting the hill near Klimatobouni somewhat higher up than the present road. From inquiries we learnt that most of the tombs were excavated in the sixties chiefly by, or on behalf of, a certain Nostrakis.

The character of the sculpture and of the inscriptions points to the first half of the fourth century B.c. for these stelae: they have a peculiar historical interest as probable relics of the 500 Athenian ἄποικοι sent to colonise Melos after the destruction of the town in 416 B.c. (Thucydides v. 116).

27. .... Tos ὑπὲρ ἀνδρῶν Ans σωενεας (?)

On a fragment of marble moulding. Ht. ‘075 m. ΥΥ͂. 14 τὰ. (Ht. of moulding ‘04 m.) The letters, which are much worn, and in the second line

difficult to decipher, appear to belong to the first half of the fourth century B.C.

28. Ἣράκλ ἴ]ειτος Εὐ. . . σε Ἰ]αυ[σανίας "Ay ?|nvopos ᾿ΑΥὙ[7 .... 6 Εὔφρων

σι

. πάτρου Ape

...9 New

ΤῸ. 4

On a fragment of a marble slab. Ht.:15 m. W.°09m. Broken onall sides but the top. Brought to me by a villager in Trypete.

29. 6

αὐτοκ]ράτ[ορα ? ov.

On a fragment of bluish marble. Ht. 13 m. W. ‘09 m. by ‘06 νη. thick. The lower edge is complete about ‘015 m. below the inscription.

14 INSCRIPTIONS FROM MELOS. 30, HPQOF: “Hpa@os.

On a cylindrical marble base or altar in a field at Tramythia, below the tower of polygonal masonry. The upper part of the back portion is broken away. Diam, ‘55 1m. Letters ‘04 1. high,

« AIONY= @ TPIEFTHELS @

4 lal Διονύσῳ Τριετηρικῴ.

On a cylindrical marble base or altar, in the field of Manolis Galanos at Tramythia; the property of the Kallergis family. It was discovered in the course of an unauthorised private excavation four years ago, and when found was fitting into a socket in the pavement of a room, 4 ft. below ground level. Ht. 1:17 m. Diam. of base, 12 m. Height of letters, 045 m. The upper surface has a circular sinking. 4 centimetres deep, at a distance of “12 τὴ. from the circumference, probably intended to receive the dedicated object. The former excavators overturned it, and when re-cxcavated by us it was lying on its side under 3 ft. of earth.

The epithet Τριετηρικός occurs as an epithet of Dionysos Bassareus in the Orphic Hymns; hymn no. 45 has the title ὕμνος Διονύσου Baccapéws Tpretnpixod. It is possible that this dedication may be associated with the inscription which follows (No, 82) and which was found in the same field with it.

5:5: <M>M APION M. Μάριον TPODIMON | tii TONIEPODAN τίν.

TRIN P > OIMYZ TAI οἱ μύσται.

Inscribed on the front of the rectangular shaft of a marble iconic herm, found in the same field with the preceding, lying on the mosaic pavement.

INSCRIPTIONS FROM MELOS. 15

The shaft is ‘28 m. wide: letters ‘03 m. high. The figure, of which unfortunately the head has not yet been found, wears a short chiton, over which is a nebris, knotted over the 1. shoulder, The soft rounded forms and the feminine appearance of the dress would appear to suit the special character of a priest of Dionysos Bassareus. The mosaic, which, with the herm, will be published in a forthcoming number of the Hellenic Journal, has its most important panel occupied with spreading vines. It may be that the base, herm, and mosaic are all to be referred to the same cult; the hall, of which the mosaic forms the floor, is from its size unsuited to the requirement of a temple or private house ; it was possibly the τελεστήριον, the hall of initiation into the religious association of which M. Marius Trophimus was the hierophant. The fact that on the herm the name of the cult is not mentioned, is an argument for the existence near it of monuments or buildings which would leave the matter beyond doubt. This inscription proves the existence of a Melian branch of those religious associations of mystae which, under the Empire, were so much in vogue, especially throughout the islands and cities of Asia Minor. We know from other examples (see Foucart, Assuc. Hel. p. 114) that members of the most considerable families commonly took part or held office in these colleges. At Crete a festival in connection with one such association was celebrated every third year, when the sufferings of Dionysos were dramatically Ὁ represented (7bid. p. 111): the principal fete of the Dionysiastae at Rhodes also took place every third year, so that the epithet Τριετηρικός was probably characteristic of the type of Dionysos worshipped in this class of cult.

33.

κα...

On a fragment of white marble, found in a wall in the ‘ Three Churches’ field. Complete on the 1. side and at the top. Ht. ‘15 τὰ. W. ‘30 τη. Letters ‘02 m. high. ;

34, ἐσ ἀμ Καίσαρα. .OTaS...

θε]οῖς ?

On a fragment of white marble, excavated in the ‘ Three Churches’ field ; broken on all sides, but possibly complete below the lowest line. Ht. "12 τὴ.

16 INSCRIPTIONS FROM MELOS.

W. 095 m. Letters 028 τὴ. high. From the writing and size of the letters it evidently does not belong to the same inscription as No. 33.

Μνασικριτ...

κ]αὶ φιλίζας ἕνεκεν 7

On a fragment of white marble, excavated in the ‘Three Churches’ field. Broken on all sides, but apparently nothing has been inscribed above the top line. Ht. 08 m. W. ‘03 m.

36. Οἱ περιβώμιοι τὴν φίλανδρον Αὐρηλί-

τι Δ " tal av Εὐποσίαν ἐν τῷ ἰδί ΄ cal ΕΣ ἐδίῳ αὑτῆς ἔργῳ.

OLTTEPIBWMIOITHN PIAANAPONAY PHAL AN EYTIOCIANENTW IAIN AYTHEEPCW

On a marble base of a herm: described Ath. Mitth. 1890, p. 246, note 1 by Wolters, who gives previous notices. Ht. ‘06m. W. 14 m.

An Aurelia Euposia is mentioned in an inscription from Cyzicus, C. I. Gr. 3690, the wife of a certain Aelius Chrestion. The expression περιβώμιοι may signify the fellow members of some religious association, to which the lady here honoured belonged. For the phrase ἐν τῷ ἰδέῳ αὑτῆς ἔργῳ cf. Ath. Miith. 1896, p. 113.

37. AN TPATANONKAIZ APALE BAL TON

OYAYIONTONEYEPCE THNKAIZQTHPA EYZAY TOY TI TOL@AABIOL AAMEINOL

TITOZ4AABIOY API TOAAMOSEKAIAOHKHE 5 ¥ TANSAABIOYA A MEINOY

Avtoxpatopa Nepovjav Tpaiavov Kaicapa Σεβαστὸν

Θεοῦ Neplova υἱὸν εὐεργέτην καὶ σωτῆρα

ὁ ἀρχίερϊΠεὺς αὐτοῦ Τίτος Φλάβιος Δάμεινος

... Titos Φλάβιος ᾿Αριστόδαμος ἐκ διαθήκης δ..... αἸὐτῶν Φλαβίου Δαμείνου.

On a slab of yellowish marble complete on all sides but 1, used as a support for wine casks in the house of the tenant of site A (see plan Joc. cit.) at Klima. Ht. “ὅθ τὰ, W. 1:20 m. by 13 τὰ. thick. Surface a good deal worn. Said to have been found in the adjoining garden together with the Poseidon statue now in'the National Museum at Athens and other marbles: among these are a marble equestrian figure presumably of an emperor which still lies in the garden, but which I understand will shortly be published in the Ath. Mitth. As this inscription apparently refers to the erection of a

INSCRIPTIONS FROM MELOS. 17

statue of Trajan set up by the high priest of his cult in Melos, it is possible that the equestrian group is the one in question: the size of the slab would be suitable for such a purpose. About ‘20 m. from the lower edge three small rectangular sinkings. are wrought in the surface parallel to the inscrip- tions at regular intervals “90 τὴ, apart. At the extremity of ll. 3,5 a vertical line has been drawn, from the upper end of which a horizontal line is drawn to the r. edge of the stone: and on the upper edge are engraved four char- acters FA Η — Il, but it is not certain that any of these marks are contem- porary with the inscription,

~t, QPEINIANOEL LAPXOANMIAIQN TOP EKTANIAIONENTOIAINTOTQITO2PO AO ION ~ OHKEN

Tos Σωρεινιᾶνος Co Re vtos (1) ἄρχων Μηλίων τὸ γ ἐκ τῶν ἰδίων, ἐν τῷ ἰδίῳ τόπῳ τὸ ὧρο- λόγιον " ἀνέθηκεν.

Inscribed on the face of a Roman Doric capital in grey marble, found in our excavations on site D; when found it was in an inverted position, serving as the base for a column in coarse poros, in the corner outside the door of the cave in the field of Panagiouli Vichos. The abacus, on which is the inscrip- tion, measures ‘42 m. in length by ‘095 m. in height: the 1. upper portion is broken away: the lower member is ‘33 m. in length on the under side, and ‘068 τη. in height. The office of ἄρχων as held in imperial times in this island, is probably referred to on the coin of Melos (Br. Mus. Cat. of Coins, Crete, &c. p. 107), where a certain Ti. Pankles is named_as eponymous magis- trate for the third time, τὸ γ'

᾿ OAXMOX ἈΠΕΙΠΙΝΧΝΓΕΡΜΧΝΙΚΟῪ ΚΑΙΣΧΡΟΣΓΎΝΔΙΚΧΘΕΟΙΣ

Ὃ δᾶμος ᾿Αγριππείναν Γερμανικοῦ Καίσαρος γυναῖκα θεοῖς.

On ἃ rectangular block of whitish grey marble, found laid in a bed of cement in our excavations on the ‘Three Churches Site,’ slightly to the n. of no. 40, The base measures ht. ‘59 m., w. ‘54 τη. by ‘37 m. thick: the traces on the upper surface show that it supported a statue. The first half of the

H.S.—VOL. XVII. σ

18 INSCRIPTIONS FROM MELOS.

word Agrippina seems to have been twice inscribed, as if the lapidary had blundered in his first attempt. For a similar dedication of a statue of Agrippina see C.J. Gr. i. 1301, perhaps from Messene. The cult of Agrippina Major in Lesbos is probably referred to in C. L Gr, 11. 2183.

YPALONYM HX TATRA EO ee ΙΓ ΠΕΣ FH |

5. TATPAAKA EITA MEISE. TPO

Κλεωνύμης πατέρα ΤΙαγκλῆν ὃ[- αβεβιωκότα τὰ μέν mplo[s το]ὺς [θεο]ὺς [ε]ὐσεβῶς, τὰ [δὲ ὅ πρὸς τὴν] πατρίδα Kali πρὸς τοὺς πολ]εῖτα[ς] ὁσίως mpo[s] ἐμὲ δὲ φιλο]στόργ[ως.

On a drum of white marble, found by us lying on its side in a foundation wall in the field of the ‘Three Churches’: surface much weathered. The height of the drum is ‘83 m.: its upper diameter is ‘60 m., its lower ‘68 m. : the letters are ‘(02 m. high. The references in this inscription to the services Pankles rendered to his country and fellow citizens seem to imply that he was a public character: if so, we may perhaps identify him with the eponym- ous magistrate (archon 7) named on the coins of Melos already referred to (see ante, no. 38).

The name, which appears to be of Spartan origin (see C. I. Gr. 1260, 1. 7), also recurs in another Melian inscription, C. J. Gr. 2433, which is worth quoting here as it seems to throw light on the family relationships.

Δαμ]α[ἤνετος [ΦἸερεκύδου καὶ ἘΠ" αγόρα ΤΙ αγ[κλέ]ους τὰν α[ὑτῶν θυγ]ατέρα, [ΚΊλε..... SN sesh. κ]α[ὶ Φ]ερεκύδης oft] Δα[μαινέτου τὰν ἀδελ[φ]ὰ[ν ..... θεοῖς.]

By a comparison of this with our inscription, it looks as if we may restore in 1, 3 [Κ]χε[ωνύμη]ς. Now in C. 7. Gr. 2439 (an inscription from Melos of about the same period), we have the metrical epitaph of Κυδίλα Δαμαινέτου, whose name certainly sounds as if she had belonged to the same

INSCRIPTIONS FROM MELOS. 19

family. The name of the mother of this Κυδέλα is read in the impossible form Κλεισφύσσα, which I would suggest is an error for Κλεισαγόρα :' if so, it admits of a restoration of this name in 1, 2, and we then obtain from a combination of the three inscriptions the following pedigree for the family ;—

Pherekydes Pankles

Damainetos m. Kleisagora Kleonymes (the dedicator of No, 46) ae cre ae rig Kleonymes Sais): Pherekydes Kydila

41,

εὐ ANQE . ANQIKAIET ONIXAAKEAIAtE

[Ὁ δᾶμος τῶν Μαλίων 1] ἐστεφ]άνωσ[εν χρυσῷ 1 / \ > atep lave καὶ ἐτ[ἐμησεν ? > ‘ / > a e[ix]ove χαλκέᾳ ἀρετᾶς “ ‘ > / \ \ ὅ ἕνεκεν καὶ εὐνοίας κ[αὶ διὰ ? \ ᾽ , ἃ »Μ ὃ tal 7 τὰς εὐερ]γεσίας ἃς ἔχων [διατελεῖ 7 Great hue OT san

On a block of white marble found by us built into a foundation wall in the same field as the two preceding. The letters are 03 m. high.

42. TPO $ IMO = < Tpodipos KAAYA A= MAY ; Κλαυδίας Mas iATO PA mA oe TON ἀγοράσας τὸν TOTONK A'T E o> τόπον κατεσ- KEYAZE NHAYT2 5. κεύασεν ἡαυτῷ ΚΑΙΤΉΓΎΝ ΑἸΚΙΔΥῪ αἱ τῇ γυναικὶ αὐ- TOY KAITOI ~~ ¥ τοῦ ae τοῖς TEKNOIX : τέκνοις.

1 The recurrence of the first compound KAe- _— Kydila similarly suggests the second compound in the names of Kleisagora and Kleonymes is in ἴῃ the names of her brother and paternal grand- favour of their being brother and sister: father.

Cc 2

20 INSCRIPTIONS FROM MELOS.

On a thin slab of yellowish marble, in the house of Kati Isanna Bechos at Trypete. Ht. “39 τη. W. “20 m. by ‘02 τη. thick: complete on all sides: said to have been found in a cave below Trypete. It seems to have been let into a wall. and bears the mark of an iron clamp on the centre of the upper and lower front edges. L. 4 τὸν τόπον probably refers to the site itself and κατεσκεύασεν to the excavation of the cave in it.

43. ; ARON ED. τι:

On a fragment of white marble excavated in the same field as nos, 38-40. Ht. 22am. Wl;

44,

TW ἐμένα tn ΟΙΓΠ ...00ls me...

1E OC’ νιν λίε οστ΄... «ΝΟ avoy

On a fragment of white marble, brought by a villager to Trypete, said to have been found at Klima. Broken on all sides. Ηἴΐ, "22 m. Ὑ. "14 τη. Height of letters ‘025.

αει Toca tAEWVE

λεξα

On a fragment of a white marble slab, brought by a villager to Trypete. Ht. ‘08 m. W. ‘07 m. by ‘04 m. thick.

INSCRIPTIONS FROM MELOS. 21 " ;

ῬΔΓΙΕΘΕολιΡΕΦΡΟΝΤΙΖΕΗΜ ΟΝ

ΤΊΑγεε Θεόδωρε φρόντιξε ἡμῶν +.

Inscribed on a slab of a marble ambon in the northernmost of the two churches at Kepos (see Br. School Annual, vol. ii. pp. 161, 168 These

churches are half sunk in the earth and nearly ruined; they appear to be of great antiquity.

CECIL SMITH.

bo lo

THE ROAD-SYSTEM OF EASTERN ASIA MINOR

THE ROAD-SYSTEM OF EASTERN ASIA MINOR WITH THE EVIDENCE OF BYZANTINE CAMPAIGNS.

[PiaTE 1]

Pann 2

OF late years a good deal of discussion has been devoted to the Road- System of Cappadocia and the Tauros region in ancient times, and it might seem at first sight superfluous to discuss the subject over again. But con- clusions already reached must always be tested in the light of new facts; and in the case before us several new facts have come to hand, which illuminate our subject and enable us to introduce into it a considerable amount of simplification. I propose, therefore, in the following paper to describe the roads which traversed this part of the country and then to prove their direction as well as their importance from the evidence of Byzantine cam- paigns. This is the simplest order to follow, because one campaign generally covers several routes and it would involve a sacrifice of clearness to break up the campaigns into a series of disjecta membra.

At every period in the history of Asia Minor the most important roads from the west converged towards. Caesareia-Mazaka (Kaisariye), which in later times became the metropolis of Cappadocia, and radiated thence towards east and south. Sebasteia-Sivas forms another centre only second in import- ance to Caesareia; and the entire road-system of Eastern Asia Minor is most easily described and most clearly understood by taking these two cities as the starting-points. I shall therefore begin with the roads leading East

and South from Caesareia and afterwards go on to those radiating from Sebasteia-Sivas.

I. ROADS FROM CAESAREIA TO THE EAST.

These are two in number: (1) what may be called the great Eastern route by Herpa, Ariarathia, Tzamandos (Azizie), and Gurun to Melitene and the east; and (2) the Roman road over Anti-Tauros by the Kuru Tchai pass and thence by Kokusos (Geuksun) and Arabissos (Yarpuz) to Melitene.

(1) The former of these two routes has been almost entirely over- looked. Yet it was at all times the great route to the east. It is

/ J.H.S.VOL. XVII (1897) ρὲ. 1.

: EASTERN ASIA MINOR

ILLUSTRATING THE ROAD SYSTEM.

English Miles . ο 20 20 0 60 80 200

J-H.S.VOL. XVI (1897) PLL.

}Satala Ts (Sadaghs |

Arnuracag7

(Ghat ay, < »- Is

Ne κ' | = Ὁ a off | Ἢ fiver ε © = πὶ Ε er \ o féimara = τ

GS. μ᾿

(Romanopolis)

Arsamosata (Shamshag)s (~~

We

MAP OF

EASTERN ASIA MINOR

ILLUSTRATING THE ROAD SYSTEM.

THE ROAD-SYSTEM OF EASTERN ASTA MINOR. 23

the Persian Royal road!: it existed in Roman times: and it is the roal to the east throughout the Byzantine period. The course of the road is as follows. From Caesareia it goes over the plain to Arasaxa (Zerezck) and after crossing the River Karmalas (Zamanti Su) proceeds by Larissa ® to Herpa (Yere Getchen) on the main stream of the river which it follows as far as Tzamandos (Azizie). The fortress Tzamandos (Tfayavdds), which is mentioned several times during the tenth and eleventh centuries after Christ,’ is placed by Prof. Ramsay (Jist. Geog. pp. 289 ff.) with the greatest probability beside the modern Azizic, and the name is regarded as a native Anatolian word, which survives in the modern name_of the river (Zamanti), At Azizie there is a “magnificent series of fountains which rise from the hills that fringe the Karmalas-Zamanti” and flow down into that river: and Prof. Ramsay supposes that the modern name Zamanti Su is derived from the city beside these fountains, “the river being called ‘the water that comes from Tzamandos” just as the Hermos is now called Gediz Su, ‘the water that comes from Kadoi’ (Kdéous, accus.), though both Tzamandos and Kadoi were situated some distance below the actual source of the river.” * While Tzamandos is frequently referred to in the late centuries, no mention is made of Ariarathia, which was situated at an important point in the upper Karmalas valley on the Sebasteia-Kokusos road. In order to account for this strange fact, Prof. Ramsay formerly conjectured (H. @. pp. 310, 289 f.) that Tzamandos and Ariarathia were to be identified, Tzamandos being the native name which had been preserved in popular usage and passed into official use about the ninth century of our οἵα. ὃ He would now, however, modify this suggestion in view of a new piece of evidence. In an Armenian Notitia Episcopatwum (a translation of a Greek original of ca. 1200), published by Mr, Conybeare in Byz. Z/t. V. p. 127, we find Tchamanton (obviously Tzamandos)*® and Ararathias “quae est in Dauthn (i.e. ‘the warm’)” given as two distinct bishoprics under Caesareia. Now Dauthn (see infra) is probably the pass leading by Kuru Tchai and Kokusos-Geuksun into Kommagene; and consequently Ariarathia should be brought lower down the Karmalas valley and located at, or very near, Herpa.’”_ The

1 See Ramsay, Citics and Bish. of Phrygia, vol. i. p. xiv. n., and the Excursus at the end of this paper.

® Larissa and Herpa must have been near each other. Herpa (Strabo, pp. 537, 539) or Herpha (p. 663) was on the road from Caesareia to Melitene at the point where it crossed the Karmalas (see Hist. Geog. pp. 289, 272-3). Larissa cannot be located with certainty, but lay on the direct road to Melitene, not far east of Arasaxa (H. α΄. pp. 272-3, and campaign of 1069 infra). It was given, along with Komana, Tzamandos, ete., to the Armenian prince Gagik in 1064.

% By Const. Porph. De Them. p. 32 and De Adm. Imp. p. 228; in 976 (Kedr. ii. p. 423)

and 1068 (Mich. Att. pp. 121-2, Skylit. 678), see infra. The Armenian name is Dzamentay or Dzamentou, Arab. Samandou (St. Martin, Mém., sur V Arménie, I. p. 191).

4 The quotation is from MS. notes of Prof. Ramsay’s, to whose unfailing kindness I owe far more than can be actually specified in the preparation of this paper.

5 For similar cases see pp. 279 n., 280 n.

8 TCapayrds in Mich. Att. 121.

7 It is quite likely that Herpa is the older name of the town, which was renamed Aria- rathia after one of the Cappadocian kings, Ariarathes (ca. 350-36 B.c.): Herpa is not mentioned after beginning of first century B,c,

24 THE ROADSYSTEM OF EASTERN ASIA MINOR

Dazmentos of this Notitia is probably the same place as the fortress Dasmenda mentioned by Strabo (φρούριον ἀπότομον Δασμένδα, p. 540) as situated in Chamanene, “at the western extremity of the ridge which bounds Cappadocia on the north” (ZZ. G. p. 290).

After passing Tzamandos-Azizie, our road goes over the hills eastwards to Gurun. The section Gurun-Caesareia just described was traversed by the late Col. Stewart, and it will be useful to give his statistics (for which I am indebted to Prof. Ramsay).

Miles Gurun. 19% Keupek Euren, alt. 5994 ft. Commenced ascent of Gédilli Dagh. Crest of Pass, about 6,700 or 6,800 ft, due W. of Azizie. Borandere vill. Karagoz vill. Ford of Zamanti. Kara Boghaz. Azizie. Road over Godilli 1). is bad ; the araba road goes round North end of Dagh. Ekrek. Karadai. Kaisariye.

»: "ἡ ὦ OW os CO Ro bo τον --

τῷ μ- ao ϑ' CO he hoe tot

At or near Gurun the road passes through Lapara-Lykandos, which Prof. Ramsay now places here and identifies with the Paulician city Lokana, mentioned in Basil’s march, 872 a.D.1_ This localization is convincing and suits admirably the description of the march of Bardas Skleros in 976 (infra). The κλεισοῦρα of which Tzamandos and Lykandos are the limiting points (Const. De Adm. Jmp., p. 228) will then be the pass over Gédilli Dagh. Leaving Gurun, the road descends the Tokhma Su (Melas) to Taranta, Pliny’s Daranda-Dalanda, mod. Derende?; and thence to Melitene (Malatia). Somewhere in this vicinity was the pass (στενοχωρία) Boukoulithos (βυύκου λίθος) mentioned by Kedr. IL. p. 421. From Melitene the road goes to the Euphrates which it crosses at Tomisa, situated on the left bank at the

1 See Class. Review, April 1896, p. 140 and 137, and Map accompanying this paper. * Cl. Rev. 1.6. p. 137. The comparison of the

Malatia was taken by the Greeks ca. 698, the Moslems settled at Taranda after it had been captured in 702, It lies three marches distant

Arabic Taranda (Biladhuri, p. 186, ed. M. de Goeje), Armenian Daranda, Syriac Turanda (St. Martin /.c. p. 190), makes assurance doubly sure. Taranta was a place of importance, especially in the first half of the eighth century, when the district of Melitene was in Saracen hands. It is mentioned by Theoph. p. 312 De door (see infra) ; in 701 A.D. it was beseiged in vain by Abd Allah (’AfdeAGs), p. 372. Some further information about the fortress is given by Biladhuri, for all references to whom I am indebted to the kindness of Mr. Le Strange who has taken the trouble to translate for me his chap. on Mesopotamian Fortresses in the ‘ Book of the Conquests’ (ca. 869 a.p.). When

from Malatiyya, deep in the Greek country ; and was held by a garrison of about 2,000 troops from the Mesopotamian army, during the summer only (Biladh. p. 185). When Omar II. became Caliph (end of 717 Α.}.}, ‘he removed the Moslem population from Taranda, for he feared for their safety, the enemy being 30 near ; the people however objected, and when they were perforce removed, they would leave nothing for the enemy, breaking even their oil and vinegar jars. The Caliph transported them to Malatiyya, leaving Taranda to ruin’ (p. 186). In the following century it was a Paulician stronghold and surrendered to Basil I, in 872 a.p..(Theoph, Cont. 267).

THE ROAD-SYSTEM OF EASTERN ASTA MINOR. 25

extremity of Sophene, and then enters Anzitene, called in Byzantine times Xavéir, the military centre of which was the fortress Hanzit, frequently mentioned by Armenian writers as Handzith and by Arabic geographers as Hanzit (Hinzit), one of the Greek frontier fortresses near the Euphrates, between Melitene and Samosata. Anzitene-Hanzit is generally placed further east, but the evidence seems clearly to show that it denoted the country between the Euphrates (starting from about Tomisa where Sophene ends), the Murad Tchai (as far at least as Palu-Romanopolis), and the sources of the Tigris! A little to the south-west of Kharput (Χάρποτε, Kedr. IT. 419) the road forks. One section goes to Kharput and thence by Arsamosata (Arab. Shamshat or Shimshat) on the Murad Tchai to Palu (Romanopolis) : but no doubt there was an alternative route by mod. Kizin to Palu. Arsamosata-Shamshat has been located by Mr. Le Strange from the de- scription of Ibn Serapion (see p. 57) who says, ‘the Nahr Salkit (= Peri Tchai) falls into the Arsanas (Murad Tchai) one mile below the city of Shamshat, near a mountain that is over the city and closes it in’ (p. 314, ef. pp, 45, 63). Combine this with Ptolemy’s μεταξὺ τοῦ Evdp. καὶ τῶν τοῦ Τίγριδος πηγῶν .... ᾿Αρσαμόσατα x.t.r. (v. 13, 18-19) and the inference is clear that the city is to be placed one mile from the junction of Peri Tchai and Murad Tchai on the left (south) bank of the latter river.? It formed at one time a Theme in the Byzantine Empire ᾿Ασμοσάτου 3 θέμα, Const. De Adm. Imp., p. 226). Romanopolis, which derives its name from Romanus I. Lecapenus, has been identified with the highest probability by Prof. H. Gelzer (Georg. Cypr., pp. 176-7) with Palu, Armen. Balu. The κλεισοῦρα, therefore, mentioned by Const. lc. p. 226 (τὸ δὲ Xavtir καὶ ἡ 'Ῥωμανοπ. κλεισοῦρα), lies on the road between Palu and Kharput

1 Only the most important evidence can be given here. Ptolemy (v. 13, 19) places ᾿Ανζιτηνὴ μεταξὺ τοῦ Εὐφράτους καὶ τῶν τοῦ Τίγριδος πηγῶν, including amongst its towns ”Av¢nra and Arsamosata (below). In Byzantine and Arabic times it clearly denotes the district indicated above. It is always connected with the κλεισοῦρᾳαᾳ Romanopolis-Palu: before Romanus I. it was attached to Melitene (τὸ XavQr καὶ ἡ Ῥωμανοπ. κλεισοῦρα τῶν Μελιτηνια- τῶν ὑπῆρχον, Const. De Adm. Imp. p. 226), and was assigned by him along with Kamacha, Keltzine (Acilisene), etc. to the newly-constituted Theme of Mesopotamia (p. 227). Nicephorus, De Velit. Bell. p. 250 (quoted at end of Pt. 1.), is quite explicit: the trans-Euphrates passes into Saracen territory are those crossing ‘ the (Tauros) mountains which separate Chanzit from the enemy’s country as far as Romanopolis.’ All the passages from Syriac and Arabic writers collected by Gelzer, Geo. Cyp. p. 178 f., confirm this localization: 6.6. Faustus Byzant. v. 16 gives as conquered in succession Arzanene, Sophanene, Ingilene (about Egil), and Anzitene,

preserving the geographical order from south- east to north-west. The Arabic writers all agree in placing the fortress Hanzit close to the Euphrates between Malatya and Sumaysit (Samosata), on a tributary of the Euphrates, says Ibn Serapion, which ‘passes the city of Hanzit and the province thereof’ and then falls into the Euphrates (ed. Le Strange, p. 54, cf. nm. on p. 49), The Euphrates dépasse la ville de Hanzyt, puis tourne vers Vowest, arrive a Sumaysat...(Ibn Khordadbeh, Trans. p. 177). Space forbids further quotations. The fort then, should apparently be looked for near the Euphrates, west of Kizin: the position of the ‘ province’ Anzitene-Hanzit seems clear.

2 Ibn Serapion’s description is confirmed by Ibn Khordadbeh and Yakut (cf. Le Strange, Zc. p. 57).

3 This is a reproduction of the Armenian form ASmusat, as Χανῶτ is of Handzith (Prof. Gelzer on Geo. Cyp. p. 172). The Theme of which Arsamosata was the central fortress apparently extended north of Murad Tchai (Arsanias).

26 THE ROADSYSTEM OF EASTERN ASTA MINOR.

or Kizin. Crossing the river at Palu, the road then follows the right bank of the Murad Tchai to Akhlat (XAva7, see campaign of 1069 infra) on the lake of Van. The other section of this Eastern road passes to the south of Kharput by way of Kizin to Amida (Diarbekr) whence it follows the valley (left bank) of the Tigris to Nineveh, then crosses the R. Zab and proceeds to Arbela (Erbil). This I believe to be the line followed by the Persian Royal Road from 'Tomisa (see Excursus).

This great and direct line of communication between West and East is the route generally taken in Byzantine Expeditions against Persia, and the section Tomisa-Caesareia will be seen to be the favourite route for Turkish raids into Asia Minor. Its direction is fixed by Theoph. p. 312, ed. De Boor, where the return of Heraclius from his second expedition against the Persians is described. On March 1, 626 Α.Ρ., before leaving the Lake of Van on his homeward journey, Heraclius held a consultation with his troops as to the route which he should take. The choice lay between two routes, (1) one leading ἐπὶ TdpavToy, 1... Taranta-Derende, and (2) another ἐπὶ τὴν τῶν Συρίων γῆν. The latter, which was unanimously chosen because it was better supplied with provisions, although the more difficult of the two, is the route leading over the Eastern Tauros (near Van), across the Tigris, and then by Martyropolis (Meiafarkin) and Amida (Diarbekr) to Samosata.! The alternative route ‘by way of Derende’ went along the right bank of the Murad Tchai (Arsanias) to Palu where it crossed the river and descended by Arsamosata-Shamshat to Kharput, and then joined the ‘Eastern road’ through Tomisa, Melitene, Derende, &c., to Caesareia.

Some other marches which concern this route alone may be added here. In starting for his second expedition (624 A.D.), Heraclius probably took this same road. It has been generally supposed that he went to Armenia by sea; but it is pointed out by Εἰ, Gerland (Die Pers. Feldziige des Kaisers Herakleios in Byz. Zt. 111. p. 345 ff.) that while Byzantine authors are silent on the point, the Armenian historian Sebéos states that Heraclius marched from Constantinople to Caesareia in Cappadocia and thence to Armenia. This is obviously the correct account. Caesareia would be the most con- venient ἄπληκτον at which his forces could concentrate for an expedition to the East; and from Caesareia he then marched to Erzerim and the Araxes valley. He thus chose the same route as Philippicus, the general of Maurice, had done in 585-6: for it is stated that he also marched to Armenia by way of Caesareia. Finally, it is most probable that Heraclius returned by this way in 628 A.D. after his third expedition.

A very important march is that of Bardas Skleros in 976 a.p. (Kedr. II. pp. 419—423). Skleros, who was appointed governor of Mesopotamia by Basil II., revolted against the king and proceeded to invade Asia Minor. After laying up stores at Kharput and obtaining assistance from the Emirs of Amida and Martyropolis, he began to march towards Caesareia. <A detachment sent to reconnoitre fell in with a division of the Imperial troops

+ See the description in Theoph. p. 313.

THE ROADSYSTEM OF EASTERN ASIA MINOR. 27

at the pass Boukou-lithos (ἔν tue στενοχωρίᾳ), and suffered defeat. After some delay Skleros started himself and in three days! reached Lapara, ‘ now called Lykandos, where he met and defeated the Emperor’s forces, and proceeded thence to Tzamandos, a populous and wealthy city situated on a steep rock (ἀποκρήμνῳ πέτρᾳ). Lykandos and Tzamandos, therefore, are both on the direct route to Caesarcia,

For other campaigns see Part II.

(2) The Roman Road.—The other route from Caesareia to the East is that followed by the Roman military road, viz. Kuru Tchai—Sirica (Kemer) —Kokusos (Geuksun)—Arabissos (Yarpuz)—Melitene. The direction of this road has been established by the discovery of a series of milestones (several of them i sitw). A large number of these was found by Mr. 1), G. Hogarth and Mr, J. A. R. Munro in 1891, and the whole subject is treated in a complete and admirable paper by Mr. Hogarth in Mod. and Ane. Roads in East. Asia Minor (R. G. 8. vol. iii), part ii, pp. 38---78, Only a brief description, therefore, is required here. From Caesareia the road follows the route just described [no. (1)] to Arasaxa-Zerezek where it branches off to Muhajir on the Karmalas-Zamanti and then crosses Anti-Tauros by the Kuru Tchai pass through the modern village Tass and Coduzabala to Sirica- Kemer.’ Coduzabala, which the Antonine Itinerary gives as a station both on the Caesareia-Kokusos and the Sebasteia-Kokusos roads, should probably be placed on the Kuru Tchai pass at the junction of these two roads (see Map). Sirica, placed by Prof. Ramsay on the Saros six* miles east from Komana-Shahr (#7. @. p. 312), ve. at Kemer, probably corresponds to the Serikha of the Armenian Notitia following Tchamanton (Tzamandos): for, as Prof. Ramsay remarks, it naturally follows Tzamandos which was situated in the same region. From Sirica the road goes nearly due South along the base of Bimboa Dagh to Kokusos-Geuksun, and then strikes North-East along the Geuk Su to Arabissos-Yarpuz, after which it crosses the Khurman Su at Izgin and the Sogutli Irmak near Ahazli and thence passes over the hills in a nearly direct line by Osdara, Dandaxina, and Arga-Arca® to Melitene. The latter section of this road from Arabissos is fully described in H. G, pp. 273-4.

1 Measured apparently from about the Euphrates, though this is not precisely stated.

2 Tzamandos was situated on the hill above the modern Azizie, which occupies the lower slopes.

ΠΟΥ Id. aD. 2.1,

4 *One too many,’ Hogarth 1.6. p. 51.

6 With Arga it would be possible to identify the Paulician fortress Argaous, which occurs in the marches of Basil I. in 872 (Apyaové, Theoph. Cont. p. 270) and of Romanus IY. in 1068 (’Apyaod, Ioan. Skylit. 670). Kedr, IJ. p. 154 tells us that Argaous and Amara were the first cities founded by the Paulicians with the aid of the Emir of Melitene ; and that when

their numbers increased Tephrike was added. The first city would be in, or close to, Saracen territory : but as the sect grew in strength and became to a certain extent an independent state, they would have to find sites for their new cities outside Saracen territory, 1,6. further north. Now Amara (see infra iv. (2) δ) is north of Argaous and Tephrike is north of Amara. Argaous therefore might be Arga. But it is far more probable that it should be identified with modern Argovan, about twenty-five miles almost due north of Malatia, This suits the line of both marches (infra) far better and is in itselfa more natural position for the first Paulician city.

28 THE ROAD-SYSTEM OF EASTERN ASIA MINOR.

The evidence of the milestones shows that the military road was built or reconstructed (restitwit) by Sept. Severus, 1.6. not earlier than the end of the second century after Christ, but a road of some kind may have previously existed along this line. That there was a trade route from Ephesus to the East as early as 100 B.c. is certain. This κοινὴ ὁδὸς is described by Strabo (p. 668) on the authority of Artemidorus. Up to Caesareia the description is full and clear. But what line did the section Caesareia-Euphrates take ? Strabo merely says ἐντεῦθεν δ᾽ ἐπὶ τὸν Εὐφράτην μέχρι Τομίσων χωρίου τῆς Σωφηνῆς διὰ ΗἩρφῶν πολίχνης χίλιοι τετρακόσιοι τετταράκοντα. Ἥρφαι, elsewhere Ἧρπα (pp. 537, 539), is Herpa on the Karmalas-Zamanti (see H. G. p. 289). The route indicated therefore is evidently that by Herpa- Azizie-Gurun-Derende-Malatia-Tomisa (opposite Isoghli): for the Roman Road did not go by Herpa but branched off at Arasaxa, and the other route is the natural line for a trade-route to the East and the easiest way to the Euphrates. I take this, then, as a proof of the importance of the Caesareia- Derende-Tomisa route in the early Roman period. The line of the Roman military road was probably determined by different considerations—viz. to connect with Germaniceia-Marash and the Syrian frontier, as well as with the frontier at Melitene. In the Byzantine period an army marching from the West towards Melitene never takes this route.

II. ROADS FROM CAESAREIA TO THE SOUTH.

(1) To Germaniceia-Marash. An army marching to Germaniceia would follow the Roman road by Arasaxa-Zerezek and over Anti-Tauros by the Kuru Tchai pass to Sirica-Kemer. At Sirica, the road to Germaniceia forks, and two routes were possible : (a) by Maroga (Maragos) and Tanadaris (Tanir) to Arabissos and thence by the well-known pass to Germaniceia and Adata. See H. G. pp. 271-2. This was the route almost always taken by Byzantine armies, and it is hardly necessary to quote campaigns in proof.

(Ὁ) The alternative route still followed the Roman road to Kokusos- Geuksun. From that point there are three modern routes to Germaniceia, but only two of these are known to have been used in ancient times: (a) the Ayer Bel pass, which crosses the Geuk Su, ascends Ayer Bel, and passing Kalli- polis and Padasia (at Temelilik) crosses the Pyramos-Jihun to Germaniceia. This road is marked in the map in H. G. p. 266, and mentioned on p. 276. It was taken by Basil in 877, and Romanus IV. in 1068 (infra). (8) The other route, the most difficult of all, is the pass by Geben and thence along the Kursulu Su, round Dolaman Dagh to the Jihun (see Map). This was the road followed by the Crusaders in 1097 (below).

Germaniceia is a great centre from which roads radiate in all directions, and it is just this fact which accounts for its strategical importance. Leading towards the East there is a road over the Ak Su past Adata (which lay to

1 For the importance of Arabissos, see H, G. pp. 277, 311, 280, ete. From Tanadaris—Tanir there is also a direct road to Sebasteia (iv, (2) ὃ infra),

THE ROAD-SYSTEM OF EASTERN ASIA MINOR. 29

the south of Inekli!), Katamana, Nisus, and Tharse to the Euphrates which it crosses at Samosata. See H. G. p. 279. Towards the South there are two roads. One leads straight to Antioch along Mt. Amanos, and was followed eg. by the Crusaders in 1097. The other goes by Doliche-Duluk to Aintab and thence to the Euphrates at Zeugma-Birejik (see ΓΤ, @. p. 279) or south- wards to Aleppo (Χάλεπ). This latter route was frequently taken in Byzan- tine marches into Syria (infra).

This route by Kuru Tchai and Kokusos to Kommagene was called τὸ Δαουθᾶ, or at least it passed through the district which went by that name. In the difficult but important passage of Niceph. de Velitat, Belli p. 250 (quoted below, p. 32), which summarises the Passes from Byzantine territory over Tauros into the Saracen country, the Anti-Tauros region is denoted by to Καησοῦν καὶ τοῦ (?) Δαουθᾶ. By τὸ Καησοῦν is probably meant the district traversed by the passes over Anti-Tauros to Sis (see (2) infra). The word Aaov@a seems to have been at last explained by Mr. Conybeare’s Armenian WNotitia, which gives as separate bishoprics under Caesareia Tchamanton-Tzamandos and Ararathias in Dauthn. As already mentioned, therefore, Ariarathia must be placed at or near Herpa, and the pass crossing Aaov@a4a—Dauthn is the road from Caesareia by Kuru Tchai.

(2) To Sision-Sis. There are two roads to Sis, both indicated in the map in H. G. p. 2667: (a) from Caesareia by mod. Tomarze to Sebagena- Seuagen (or Suwagen) on the Karmalas-Zamanti and thence by the Gez Bel pass over Anti-Tauros to Hadjin. Between the point where this road leaves the Karmalas valley and Hadjin, probably near the mod. village Urumlu, is to be placed the Kaisos mentioned in Basil’s march 877 A.D. (Theoph. Cont. 279, Kedr. II. p. 214, infra). Kaisos should probably be connected with Kabissos (8 = fF, cf. H. G. p. 312 n.) given in Not. I. as a bishopric of Cilicia Secunda, and by Ptolemy as Kabassos in Kataonia,® and also with the τὸ Καησοῦν of Nicephorus, 0, Καησοῦν is the district of which the fort Kaisos-Kabissos is the centre. The name of the fort would be extended to cover a district beyond its actual vicinity, just as Sebasteia, or Koloneia, gives its name to the whole Theme. The passes therefore which cross τὸ Καησοῦν are the pass which we are describing and the following more westerly pass to Sis; and the whole expression τὸ Kay. καὶ τοῦ (?) Δαουθᾶ will denote ‘the Anti-Tauros region crossed by the passes leading over Tauros, From Hadjin the road leads across the Saros-Sihun (here called the Geuk Su) and thence over Mt. Tauros to Sis.

(Ὁ) The alternative route branches off from (a) on the north side of Mt. Argaios to Ferakhtin or Frakhtin on the Karmalas, thence to Kiskisos- Kisken and across Anti-Tauros by Enderessi Yaila to the Saros, after which it crosses Mt. Tauros to Sis.

(3) The two passes on the west of mount Argaios leading from Caesareia to the south are of ped pos sig See H. G. PP. 350 ff.

1 Cl. Review, l.c. pp. 138 sf. β ἮΝ pp. 271, 281, 291. 3 See H. G@. pp. 386, 451.

30 THE ROAD-SYSTEM OF EASTERN ASIA MINOR.

(a) The less important of the two is the difficult road which runs nearly due south by Develi-Kara-Hissar to Podandos-Bozanti and through the Cilician Gates to Tarsus. This pass was called “ Karydion” (77. G. p. 351). (Ὁ) The other pass “ Maurianon” by way of Tyana and Loulon was the regular route across Tauros into Cilicia. It coincides with (a) nearly as far as Develi Kara Hissar and then branches off to the place now called Zengibar Kalesi, half an hour west of Develi, “a striking mediaeval castle on a lofty two-peaked hill.” This is the absolutely impregnable fort which the Crusaders in 1097 passed by without attempting to take (see infra). It is not named by the historians of the first Crusade, but Prof. Ramsay points out to me that it was Kyzistra, as is proved beyond all doubt by a passage in Chamich’s history. In 1079 Gagik, the exiled king of Armenia, marched from Tarsus in the direction of Caesareia to annoy the Greeks and on arriving “on the plains of Arzias, near the fort of Kyzistra,” allowed himself to be led into an ambush and was imprisoned in the fort, which was impregnable. The Armenian chiefs laid siege to the place but could not take it, and when the body of the murdered King was suspended from the walls before their eyes, they retired, convinced that nothing could be done against his murderers. From Kyzistra the road proceeds to Tyana (Kizli-Hissar, three miles south of the mod. village Bor) and thence by Loulon to Podandos where it rejoins (a) and passes through the Pylae Ciliciae to Tarsus.

From Tyana there is another route to Herakleia-Kybistra (Eregli) and thence either through the Cilician Gates or westwards to Barata, where roads diverge to Iconium and over the Isaurian mountains. These routes occur in the marches of Romanus and the Turks in 1069 and of the Crusaders in 1097 (infra).

III. PASSES FROM MELITENE INTO KOMMAGENE.

The consideration of these will complete the list of Tauros-passes. There are at least two, and probably three, roads over Tauros from Melitene into Kommagene, indicated by Niceph. /.c.as those crossing τὰ (sc. ὄρη) παρακεί- μενα Μελιτήνην τε καὶ τὰ Καλούδια. The word Καλούδια is explained by a reference in Biladhuri (/.c. p. 187) who says that the fortress Kalaudhiyya was destroyed by the Grecks under Constantine Copron. in 751 A.D., after the capture and sack of Malatiyya (Malatia). Καλούδια is therefore the Graecized form of the Arabic name for Claudias. This fortress was situated on the Euphrates near Melitene and not south of Samosata, as is sometimes supposed. This is confirmed by Amm. Marcell. xviii. 7. The Persian King Sapor, marching into Asia Minor by way of Nisibis and Constantina, halted at the latter town where he learned that the Euphrates had risen high and could not be crossed by a ford; and consequently he decided to turn north- wards (flecti in deaterum latus) and, taking a more circuitous road through a fertile district, to make for the two fortresses Barzala and Laudias (Claudias), where the Euphrates “tenuis prope originem et angustus, nullisque adhuc aquis advenis adolescens, facile penetrari poterit, ut vadosus.” “ Prope

THE ROADSYSTEM OF EASTERN ASIA MINOR. 31

originem ” is of course an absurd exaggeration, but the passage indicates that Claudias was far up the river near Melitene, as is shown by the fact that the Roman troops on receiving intelligence of this movement prepared to hasten to Samosata and, after crossing the river there and breaking down the bridges at Zeugma and Capersana, to repel the Persian advance. Now it has already been mentioned (§ 6) that a detachment under the tribunes was engaged in fortifying the western bank of the Euphrates “castellis et praeacutis sudibus omnique praesidiorum genere.” All these facts, combined with the words of Niceph., seem to prove the existence of a pass leading south along the river by Claudias, Barzalo, and other places of uncertain situation to Samosata (as marked in the map in HW. G. p. 266).

The other two passes are better known. (1) One leads direct to Germaniceia up the course of the Sultan Su past Sozopetra-Zapetra (Arabic Zibatra, situated at Viran Sheher, four miles off the road towards the right’) and over Tauros by Surghi, Erkenek, Pavrelu, Inekli on the Ak Su, and past the ‘famous’ fortress Adata (Al-Hadath) to Germaniceia.

(2) The other pass follows this same route as far as Surghi and thence turns south-eastwards to Perre (Hisn Mansur, mod, Adiaman) and Samosata- Samsat (as in map in H. G.). It was traversed in 872 A.D. by,a detachment of Basil’s army despatched from his base camp on the Zarnuk south-west of Melitene. This column after passing through τὰ στενὰ τῆς ὁδοῦ captured Zapetra, laid waste the adjacent country, and took Samosata: whence they returned to the Zarnuk.

IV. ROADS RADIATING FROM SEBASTELA.

Almost all these roads join one or other of the routes already described. They are all Roman roads except Sebasteia-Tephrike (3), which is not known to be Roman.

_ (1) Sebasteia-Caesareia.—This road is of considerable importance as affording a direct route from Sivas (on the great military road of the Byzantine period) to Isauria or to the Cilician Gates; and as such it was used eg. by Romanus IV. in 1069 when he wished to reach the Cilician passes without loss of time in order to intercept the rapid retreat of the light Turkish horsemen. The road is clearly marked in Prof. Ramsay’s map (p. 266) and described on p. 270. It runs parallel to the course of the Halys through Malandara, Armaxa, and Aipolioi to Caesareia, Aipolioi is the Aepolion of the Armen. Not., and the name is preserved in mod. Palas.’

΄

(2) Sebasteia-Germaniceia.—There are two routes :— (a) Sebasteia-Tzamandos-Ariarathia and thence over Kuru Tchai by Kokusos to Germaniceia. The change in the position of Ariarathia necessitated by the Armen. Not. and the recognition of Kuru Tchai as the

1 See Cl. Rev. l.c. p. 188 f. as in modern Greek.’—Prof. Ramsay in MS. 2 * Palas=AlwoAlous, { being pronounced as y, —notes.

32 THE ROAD-SYSTEM OF EASTERN ASTA MINOR.

great Anti-Tauros pass will involve a modification of the route laid down in ΗΠ. G. p. 214. The road will now run by Tonosa-Tunus, Karmalis on the Karmalas-Zamanti (Viran Sheher, H. G. p. 289), Tzamandos-Azizie, Ariarathia-Herpa (at Yere Getchen), and thence by Kuru Tchai and Coduzabala to Sirica and Kokusos-Geuksun. At Tzamandos it joins the eastern route Caesareia-Azizie-Tomisa [I. (1)] which it follows to Ariarathia- Herpa and there leads it into route II. (1) to Germaniceia.

(Ὁ) Sebasteia-Gurun-Arabissos-Germaniceia. See AH. G. pp. 274-5. This was an important Roman road, used also in Byzantine times as affording a direct route north to Sebasteia from Germaniceia by the famous Arabissos pass [II. (1) a]. From Sebasteia it runs to Blandi (near Ulash) and thence to Euspoena-Ispa (at Deliklitash) on a branch of the Tokhma Su, the course of which it follows to Gauraina (Gurun) and Lykandos-Lokana where it crosses the eastern route. About midway between Euspoena and Gurun, near Manjilik, is probably to be placed the Paulician fortress Abara-Amara. The two names are obviously to be identified. Amara (Kedr. 11. 154) was one of the first two Paulician cities, the other being Arguous-Argovan (see note on Arga, supra, p. 27). Abara was one of the Paulician forts captured by Basil I. in 872 in his march from Tephrike by Taranta-Derende to Melitene (Cont. 267, Kedr. 207), and it was given along with Sebasteia, Larissa, and other cities to Senakerim, prince of Asprakania, by Basil 11. in 1021 (Kedr. Il. 464). It was a τοῦρμα of the Theme Sebasteia and became a κλεισοῦρα under Romanus IV. (Const. De Adm. Imp. p. 228). These statements seem to leave no doubt that it was situated on the pass between Sivas and Gurun, near Manjilik (see Map). From Gurun the road goes to Tanadaris-Tanir where it leads into the Arabissos-Germaniceia pass.

(3) Sebasteia-Tephrike (Devrik).—This road leading to Tephrike and thence to Zimara (Zimarra) on the Euphrates was of great importance during the Paulician revolt in the ninth century. It is probable that there was also a road of some kind from. Euspoena joining a road from Tephrike at mod. Kangal and thence following the course of the Kuru Tchai to Melitene. On this road Aranga-Arani was perhaps situated (7. G. p. 275). This is the direction in which Basil I. marched after withdrawing from Melitene in 872, capturing Argaous-Argovan and several other Paulician forts in this district.

Before I go on to give some proof of the lines laid down for these roads by an investigation of Byzantine campaigns which passed over them it will be useful to quote and endeavour to explain the passage of Niceph. De Vel, Bell. p. 250, which summarises the majority of the routes described above. The words are δι’ οἵας yap ὁδοῦ διελθεῖν βουληθῶσιν (sc. the Saracens), ἀπό te τῶν ἐν Σελευκείᾳ κλεισουρῶν καὶ τοῦ τῶν ᾿Ανατολικῶν

1 This might possibly be the ‘Aragines in βούκου λίθος near the Euphrates (Kedr. ii, 421, Pharakn’ of the Armen. Notitia. Is it possible supra, p. 24)? Prof. Ramsay, however, think that Pharakn (= ‘the sheepfold’) is the pass that Pharakn=Everek at base of Mt. Argaios.

THE ROAD-SYSTEM OF EASTERN ASIA MINOR. 33

θέματος, καθὰ τὰ Ταυρικὰ ὄρη τήν τε Κιλικίαν διορίζουσι Καππαδοκίαν τε καὶ Λυκανδόν' πρὸς τούτοις δὲ (sc. δι᾿ οἵας ὁδοῦ διελθεῖν βουλ.) καὶ τὰ (sc. T. ὄρη) παρακείμενα Γερμανίκειάν τε καὶ Ἄδαταν! καὶ τὸ Καησοῦν καὶ καὶ τοῦ (1) Δαουθᾶ Μελιτήνην τε καὶ τὰ Καλούδια' καὶ τὰ πέραθεν τοῦ Εὐφράτου ποταμοῦ διορίζοντα τήν τε τοῦ Χανζῆτι λεγομένην χώραν καὶ τὴν πολεμίαν ἄχρι Ρωμανουπόλεως" ἐν ὅλοις τοῖς τοιούτοις θέμασι, δι᾿ οἵας ἂν ὁδοῦ ὑποστρέφοντες διελθεῖν πρὸς τὴν ἰδίαν βουληθῶσιν, K.T.Xr.

The importance of this passage lies in the fact that Nicephorus is speaking from personal knowledge of the country gained during the wars against the Saracens. But the passage is exceedingly difficult, and especially the first clause (ἀπό... Λυκανδόν), which can hardly be right as it stands (if the words are taken strictly): for the passes between Seleukeia and the Anatolic Theme cannot be the passes leading from Cilicia over Tauros into the Themes of Cappadocia and Lykandos! The meaning must be (1) the passes from the Anatolic Theme over Isauria to Seleukeia, and (2) those on the west of Mount Argaios, especially the Cilician Gates, and probably also the passes to Sis; (3) the passes across the Tauros Moun- tains overlying the district of Germaniceia and Adata on the one hand, and the Anti-Tauros region (Kaésoun? and Daoutha) on the other, 7.2. the passes to Germaniceia; (4) the passes across the Tauros Mountains overlying Melitene and Kaloudia (Claudias), 1.6. the passes from Melitene into Komma- gene: and (5) the passes beyond the Euphrates leading from the district between Tomisa and Romanopolis-Palu (Xavéir) into Saracen territory. Cf. Const.’s words, τὸ δὲ Xavtir καὶ ἡ Ρωμανοπ. κλεισοῦρα (De Adm. Imp. p. 226).

Parr.

CAMPAIGNS IN THE CAPPADOCIAN DISTRICT SHOWING THE ROUTES DESCRIBED.

Heraclius’s March in 626 (Theoph. pp. 312-313). Routes traversed : Samosata—Germaniceia [under II. (1) ]; (?) Germaniceia- Arabissos—Sebas- teia [II. (1) a and IV. (2) a}.

After reaching Samosata by way of Martyropolis and Amida, Heraclius took the direct road to Germaniceia, passing Adata on the way. Theo- phanes’ description of his subsequent route is confused; and it seems best to accept Prof. Ramsay’s correction Ἄδατα (for ‘Adava) and the slight transposition which makes the sentence read περάσας τὴν Ἄδατα eis Γερμ.

1 Perhaps taken as a fem. sing., but ordin- arily rhy”Adara. ᾿Αδατᾶν in Bonn ed, isclearly wrong.

2 From this passage alone it would be natural to connect Kaésoun with modern Khesun in Kommagene, south of Besne : but see above II. (2) and campaign of 877 infra. The Sis passes

H.S.— VOL. XVII.

should strictly be included under τὸ Kay. καὶ τὸ Aaova but Niceph, is evidently thinking of the passes leading from the Anti-Tauros region generally across Tauros to Germaniceia and Adaia, The Sis passes ought to come under those leading from the Theme Lykandos into Cilicia. D

34 THE ROAD-SYSTEM OF EASTERN ASIA MINOR.

ἀφίκετο καὶ πάλιν τὸν Ταῦρον ὑπερβὰς ἦλθε πρὸς τὸν Xdpov.! The route taken by Heraclius will then be the Arabissos pass (which was the ordinary route) to the Saros which he crossed by a bridge, a solid structure with προπύργια capable of defence, such as we might expect to find on this road. While he lay encamped there, he was overtaken by the Persian general, Shahrbarfz, who had reached the Euphrates before him, and broken down the bridge of boats at Samosata, but had failed to intercept his retreat. The Persians succeeded in bringing on a battle but were defeated. Heraclius then continued his march to Sebasteia, when he went into winter

quarters.

Cont. p. 277 ff, Kedr. II.

Basil’s Campaign in 876-7? (Theoph. [Π-. (1) 4];

213 ff.). Routes traversed : Caesareia—Kokusos—Germaniceia. and the passes towards Sis [II. (2) a & 6].

To understand Basil’s movements in 877, it is necessary to observe that his march into Kommagene in that year was not a mere isolated expedition, but part of a concerted scheme to drive the Saracens out of the whole Tauros region. Operations were being simultaneously carried on in the north against the remnants of the Paulician community, in the south-west against Loulon and Tarsus, and by Basil himself against the country between Caesareia and Adata. The Arabs did not really conquer this region: they merely held it by strong garrisons in the various fortresses, levying imposts on the Greek inhabitants ; and consequently the capture of these strongholds would mean the recovery of the whole country. This was Basil’s object. In 876 the fortress Loulon, commanding the Tyana-Tarsus pass and therefore occupying a very important strategic position (Cont. 277), which ‘through the negligence of preceding Emperors had been captured by the Saracens and fortified and garrisoned by them on account of its natural strength, was recovered by Basil’s generals. This was followed by the surrender of the fort Melouos, on the Laranda-Kelenderis pass. About the same time the Paulician town Katabatala,? to which the Paulician refugees had retired after the fall of Tephrike (873), was taken and sacked. Next spring (877) these successes were followed up by an expedition against the inter- mediate country (between Caesareia and Kommagene) undertaken by Basil himself, while his generals continued the war in the vicinity of Tarsus and against the Paulicians in the north (see n. 37). Starting from Caesareia, Basil sent forth a detachment to pioneer the way and followed himself with the main body. The detachment captured the forts Psilo-kastellon (Cont.

most improbable that he would cross (1) Amanos, (2) Tauros by Cilician Gates, and thence by a most difficult route come round to Sivas.

1 See Class. Review, 1.6. p. 140 note, and H. G. p. 311. If the text of Theoph. is right, the description is obviously very bad. It may

be answered that he meant to say ‘on his way over Tauros (Amanos, cf. Mich. Att. 120, Sky). 677) he reached Germaniceia, and passing Adana came to the Saros.’ If so, the text requires much alteration, for the proper order is Ger- maniceia—Amanos—Saros—Adana ; and it is

2 The date is 877, not 880, for Sima, who submitted to Basil, was killed by Tulun of Egypt in 878 (Weil, Gesch. der Khal. ii 473 n.).

3 Kedr. calls it Kameia.

THE ROADSYSTEM OF EASTERN ASIA MINOR. 35

Xylo-kastron, Kedr.) and Paramo-kastellon (Phyro-kastron, Kedr.). Then the fortress Phalakron voluntarily surrendered. These forts lay between Caesareia and the Karmalas-Zamanti on the routes leading into the two Sis passes and the pass over Anti-Tauros towards Kokusos, Basil’s plan obviously being to secure all the passes as he advanced southwards.!__ Basil then crossed * the Onopniktes (Karmalas) and the Emir of Anazarbos (Anazarbe) along with the troops from Melitene fled before his advance (ἐπεὶ ἤγγιζε, Kedr.), thus leaving him free to secure the passes beyond the Karmalas. This was effected by the capture of Kaisos [or Katasama; Kasama, Kedr.], Robam [Kedr. gives Karba], Endelechéne* or Andala [Ardala, Kedr.], and Krémo-sykéa (or -sykaea, Cont.) ; and thereupon Simas ‘the son* of Taél,’ who held the passes of the Tauros (1.0, Anti-Tauros) and harassed the Roman borders, submitted to Basil, who then crossed the River Saros and continued his advance towards Koukousos (Kokusos).

These fortresses are again to be looked for on the passes leading to Sis.° Kaisos and Robam cannot be identified with Ibn Khordadbeh’s Kaisoum and Ra‘bin (De Goeje’s Trans. p. 70) which are frontier fortresses of Meso- potamia and identical with the Armenian Khesoun, modern Khesun (south of Besne), and the Armenian Rhaban, south-west of Khesun, between that town and Marash (St. Martin, Mém. sur lV Arménie, I. p. 194). But Kaisos should probably be connected with Kabissos of Not. I. and Ptolemy’s Kabassos in Kataonia, and Niceph’s. τὸ Καησοῦν (see above). τὸ Καησοῦν then is the district of Kaisos-Kabissos, which is itself to be placed on the more easterly of the two passes over Anti-Tauros towards Sis, nearly opposite to Kiskisos- Kisken on the western pass.

Arrived at Kokusos-Geuksun, Basil set fire to the woods round the town and then plunged into the defiles of Tauros, cutting his way through the pathless forests,® and cheering on his men by his personal exertions, past Kallipolis and Padasia to Germaniceia. The Arabs remained within . their walls, not venturing to offer him battle, but as the siege of the town was hopeless he passed on to Adata (Adapa in Kedr., 7 for τ), which he besieged in vain. He then devastated the adjacent country and captured the πολίχνιον Geronta (? Geron). After another attempt on Adata, he retired cautiously in fear of an ambush, and after receiving the submission of

1 Phalakron may be Frakhtin (Ferakhtin) on the western pass to Sis; the -tin is the Arabic word Din widely adopted in Turkish [W. M. R.]. Ῥ511ο-, Xylo-, ete. are all Graecized forms.

2 Cont. and Kedr. do not precisely say so ; after enumerating all the forts they say vaguely τὸν ᾿Ονοπνίκτην λεγόμενον ποταμὸν Kal τὸν Sdpov διαπεράσας, knowing only that the forts were in this district somewhere. The Saracen army would not take to flight, nor would Simas, ‘ who held the Tauros passes’ (infra), submit before Basil had reached the Karmalas, as their language would imply.

It might be suggested that the curious name Onopniktes is a popular word expressing the difficulty of fording the river (ὄνος, πνίγειν),

8 Cont. says τῆς Ῥοβὰμ ἤτοι ᾿Ενδελεχόνης ἡ πόρθησις γέγονεν, ἅμα δὲ καὶ ἡ τῆς ᾿Ανδάλου : probably to be changed to τῆς ᾿Ανδάλου ἤτοι Ἐν δελ.

4 Simas was not ‘son of Tael’ but his sur- name was Tawil, i.e. ‘the tall’ (Weil, le. ii. p. 473 n.).

5 Endelechone— Andala may perhaps be Enderessi on the western pass.

§ Which shows that this was not the ordinary route to Germaniceia,

D 2

36 THE ROAD-SYSTEM OF EASTERN ASIA MINOR.

Abdelomel, ὃς τῶν ἐκεῖσε τόπων ἐκυρίευσε, returned across Mount Argaeos to Caesareia. Here he received the news of his generals’ victories, which were soon confirmed by the arrival of the prisoners from the district of Koloneia and from Loulon,—they are said to be Saracens and Kurds (Kovprot) from Tarsus and the Paulician fortresses,—and after slaughtering them all he returned by Midaion, where his army went into winter quarters, to Constantinople.

Campaign of Romanus IV. in 1068 (Mich. Attal. pp. 104 ff, Skyl. pp. 668 ff.). Routes traversed: Caesareia-Lykandos [I. (1)] Sebasteia-Kokusos- Germaniceia-Aleppo [IV. (2) @ and II. (1) 0]; through Cilicia by Cilician Gates to Constantinople [II. (3) 6]; the Turks traverse I. (1) to Amorion.

This is the first of a series of campaigns against the marauding bands of Seljuks, whose ever-increasing raids made Asia Minor insecure from end to end. In the spring of 1068 Romanus set out from Constantinople with the intention of operating first of all against the Turks on the northern frontier. He advanced through Bithynia and Phrygia, 1.6. by the military road passing through Dorylaion and Sebasteia, and when he had got as far (apparently) as Basilika Therma, the Turks made a feint of retiring before him, and he resolved to march southwards into Syria against the Saracens of Aleppo (Χάλεπ) who, in concert with the Turks, were constantly attacking Antioch as the first step in a scheme for the reconquest of the whole of Syria. He therefore left the road leading ‘straight to Sebasteia and Koloneia’ and marched southwards obviously by the road leading to Caesareia and thence by the eastern route [I. (1)] to Lykandos, where he intended to remain during the hot season and then advance into Syria in the autumn. While encamped here, he received intelligence that the Turks had made a sudden raid on Neocaesareia-Niksar and were returning again loaded with their -spoils. Without losing a moment, he marched rapidly northwards again διὰ ἀτραπῶν δυσβάτων towards Sebasteia, and as he approached the town, he ordered the main body of his army under Andronikos to proceed thither, while he himself with thé cavalry hurried over the hills between Argaous and Tephrike? in pursuit of the rapidly retreating Turks. This means that he marched along the Gurun-Sivas road [IV. (2) Ὁ] about as far as Abara- Amara and then struck right across the hill-country towards Tephrike and the north-east. By this cross-march (κατὰ τὸ ἐγκάρσιον) he succeeded in overtaking the marauders and compelled them to relinquish their plunder and prisoners. He then rejoined his army at Sebasteia (beginning of October) and after a halt of three days marched south again by the defiles of Kokusos (διὰ τῶν τῆς Κουκουσοῦ αὐλώνων) to Germaniceia. Evidently, therefore, he

1 This means that the war begun the year the order of the words does not prove that before in the south-west and north was being Argaous is north of Tephrike: he has just said

carried on at the same time as Basil’s expedition: εὐθὺ KoAwvelas καὶ SeBaorelas. The site assigned next year (878) Abdallan, Emir of Tarsus, was to Argaous (supra) at Argovan suits this passage decisively defeated at Podandos. well.

2 τῆς τε Teppixns καὶ τῆς ᾿Αργαοῦ (Skyl. 670) :

THE ROAD-SYSTEM OF EASTERN ASIA MINOR. 37

took the route by Tzamandos and the Kuru Tchai pass. Before reaching Germaniceia, {.6. probably from Kokusos, he despatched a large division to Melitene to guard the frontier [route I. (2)] and prevent Apsinalios or Ausinalios [= Afschin (Weil, 1.6. iii. p. 112, 2. 2)], the Turkish commander in these parts, from raiding across the Euphrates. So inefficiently was the command executed that a band of Turks actually passed Melitene before the very eyes of the garrison and fell wpon Romanus’ foraging parties, some of which they cut off. They must therefore have crossed into Kommagene by the Melitene-Germaniceia pass [III. (1)]. Romanus, after leaving German- iceia entered the district (θέμα) called by the Armenians Τελοὺχ [Doliche,} the Dolouk of Ibn Khordadkeh (p. 70), mod. Duluk, two hours north-west of Aintab], and thence passed on to Aleppo, at this time under the Emir Mahmud (Μαχμούτιος). After ravaging the country around without attacking the town itself, Romanus marched against Hierapolis (Membid)) which he captured and fortified as an outpost to guard the Syrian frontier. While he was engaged in besieging the Acropolis, Mahmud made an unexpected advance from Aleppo and inflicted a serious defeat on the troops set to guard the part of the town already captured: but the disaster was avenged after the final capture of the town by a night attack on the Saracen camp, which freed the Emperor from further molestation. Placing Membidj under a στρατηγός, he advanced to Azas,? which he failed to take, then entered the country of Ausonitis where he burned Katma® (a fort. of the Emir of Aleppo) and encamped at Terch&la (Tarchéla). Shortly after this he,entered Byzantine territory and marching in the direction of Antioch captured by the way a town Artach (near Antioch), which was evacuated by its Saracen inhabitants. At this point he determined, in consideration of the exhaustion of his troops, not to proceed to Antioch but turned towards north- west and crossing Mt. Amanos by the Syrian Gates (Beilan pass, ai κλεισοῦραι δι’ ὧν ἡ Κοίλη Συρία τῆς Κιλικίας χωρίζεται) reached Alexandros (Alexandretta), Thence he marched by the road which skirts the Amanos range (τὸν Ταῦρον), until he emerged into the plain of Issos; whence he marched through Cilicia and the Cilician Gates to Podandos. Just as he was entering Typsarion or Gytarion (Skyl.) which Prof. Ramsay with great probability locates at the point where the Tarsus-Tyana and Tarsus-Caesareia (‘Maurianon’ and ‘Karydion’) passes forked, he received reports of the mismanagement of the general sent to Melitene to guard the frontier, who had allowed the Turks to cross the Euphrates and pass along the ‘Eastern road’ [I. (1)] by Caesareia to Amorion, which they took and plundered. They had left their camp at a place called Ghalceus (τῇ τοῦ Χαλκέως τοποθεσίᾳ) near Tzamandos, where the Roman general had his troops stationed; but so far from suffering any inconvenience from his

1 Δολιχὴ becomes Dolouk and then againin Chesney. It is called “Αζάζιον, two days’ march Greek Τελούχ ! It is mentioned both as πόλις from Berroia (Aleppo),”” in Kedr. ii, 492. and as θέμα in Kedr. ii. 494. 3 Modern Kutma, nearer Antioch. The 2 “ Azaz is about twenty miles north by west description of the march is very accurate. of Aleppo,’ Finlay, i. 472 quoting from Col,

38 THE ROAD-SYSTEM OF EASTERN ASIA MINOR.

presence there, the Turks on their return had actually defeated him and shut him up in the fort. Consequently the Emperor finding himself unable to pursue them returned direct to Constantinople.

Romanus’ campaign in 1069 (Mich. Attal. 122 ff., Sky]. 678 ff.). Routes traversed: Caesareia—Melitene, over Euphrates, [1 (1)] and thence north to Acilisene ; Koloneia — Sebasteia — Caesareia — Herakleia [IV. (1) and II. (3) b]; Melitene — Caesareia —Iconium [1. (1) and IT. (3) 4]; Icontum— Seleukeia—by Syrian Gates to Aleppo.

In 1069 Romanus undertook a second campaign against the Seljuks. After quelling the rebellion of Crispin, a Norman noble in his service, he arrived with a large force at Caesareia and continued his march eastwards to Larissa, where he heard that a Turkish horde was engaged in pillaging the country in the vicinity. A detachment despatched against them was driven back in rout, and Romanus then moved onwards towards Melitene. While he was engaged in pitching his camp, the Turks suddenly appeared and, occupying the higher ground, proceeded to attack the Byzantine army in the plain below, but were defeated. Romanus allowed them to retreat without molestation and when he followed them three days afterwards they crossed the Euphrates and encamped there, waiting till he should return home. When he had advanced within less than two days’ march of Melitene, he thought of returning again and abandoning a wearisome and _ fruitless pursuit, merely leaving a force to guard the frontier; but he finally determined to cross the Euphrates and march against Χλιῴτ, mod. Akhlat, on Lake Van, hoping by the capture of the town to secure the Armenian frontier and arrest the ruinous incursions of the Turks. Accord- ingly he advanced by Melitene and crossed the river (τῆς προσωτέρω φερούσης ἥψατο ἕως τὸν Εὐφρ. διαπεραιωθεὶς x.7.r.), compelling the Turks to retreat inland (εἰς τὰ σφέτερα). The line of march is thus the ‘ Εἰδβύθγῃ road’ [I. (1)]. The direct route from this point to Akhlat went by Kharput and Romanopolis—Palu and thence through difficult country to Van (supra). This route he followed for a short distance (ὡς yap εὐθὺς τῆς ‘Pwpavor. ἐλαύνων ἐφαίνετο, ἐξ ἧς ἡ πρὸς TO Χλιὰτ κάθοδος διὰ στενωπῶν ἐπιγίνεται, μεταστρέψας τὴν γνώμην...) and then suddenly halted ἐν βαθεῖ τόπῳ, where he divided his army and placed the stronger division under the command of Philaretos for the defence of the frontier, while he himself turned north- wards,! preferring a cooler climate. After passing over rough and mountain- ous country, he reached a place called Anthias, a fertile and well-watered spot amidst high mountains. It should be looked for in the watershed south of Mezur Dagh. Thence he proceeded to cross “ Mount Tauros, called by the inhabitants Movfoupos,” 1... Mezur Dagh (Arabic Jabal Mazur), and passing a second time over the Euphrates entered Κελεσίνη (Acilisene, Skyl. Κελτζηνή), which is accurately described as separated from Mezur Dagh by the river. While encamped here he received intelligence that Philaretos had

' The crossing of Murad Tchai is not mentioned, but must be assumed.

THE ROAD-SYSTEM OF EASTERN ASIA MINOR. 39

been defeated by the Turks and the routed troops soon arrived at his camp, fleeing by way of Anthias and Mezur Dagh. The Turks pursued for some distance, but finding the country impracticable for light horsemen, they turned back, crossed the Euphrates above Melitene, and over-ran Cappadocia in their usual manner, making for the populous but defenceless city of Iconium (Konia). Romanus rallied his forces and determined to intercept their return. His first plan was to “lead his army through the town of Keramon to the banks of the Euphrates as far as Melitene,” but it was pointed out that this route ran through a deserted and pillaged district where supplies would be hard to obtain, and that time would be wasted in traversing ground where it was necessary:to march in single file. The route indicated is not clear, but apparently it crossed Mezur Dagh towards Murad Tchai and Melitene! In any case he abandoned this idea and marched through Koloneia and the Armeniac Theme to Sebasteia, 7.e. by the road Satala— Koloneia—Nicopolis—Sebasteia. At Sebasteia he learned that the Turks were marching through Lycaonia and Pisidia on their way to Iconium? and so he advanced to Herakleia—Kybistra (Eregli), 1.2, by the road through Caesareia [ IV. (1)] and thence by Tyana to Herakleia [ II. (3) 6]. Hearing at this point that the Turks had sacked Iconium. and were returning, he despatched a detachment to Cilicia to effect a junction with Katatourios, the governor (“duke”) of Antioch, whom he requested to secure the passes east of Mopsuestia (Missis). The Turks marched διὰ τῶν τῆς Σελευκείας ὀρῶν and, as they emerged into the plain of Tarsus, they were attacked by the ΟΠ Armenian inhabitants but escaped, with the loss of their booty, through Cilicia. Being informed by Greek captives that a force was awaiting them at Mopsuestia, they avoided the town and after a short halt at Blatilibas (Baltolibas, Skyl.) hurriedly crossed Amanos (τὸ Σαρβανδικὸν ὄρος) by the Syrian Gates to Aleppo. Romanus learning of their escape at Claudiopolis, whither he had advanced to meet them, left a force to operate against other Turkish bands and returned to Constantinople.

The First Crusade in 10973 Routes : Nicaea—Dorylaion—Iconium— Herakleia (Eregli); thence to Tarsus—Adana—Syrian Gates—Antioch [under II. (3) ὃ ete.]; Herakleia—Caesareia—Kokusos—Germaniceia— Antioch [II. (3) ὃ and II. (1) Ὁ].

After the capture of Nicaea, the Crusaders proceeded by Dorylaion and Iconium to Herakleia—Kybistra, which was evacuated by the Turks on their approach. At this point the army divided. Baldwin and Tancred with their own following marched southward by Podandos and the Cilician Gates to Tarsus, which they captured without difficulty ; whereupon Adana (Addana,

1 Keramon can hardly be connected with τὸ Κεραμίσιον on the Zarnuk, the most easterly tributary of Tokhma Su (Melas) flowing past Melitene (Theoph. Cont. 268).

2 The Turks therefore took their favourite route by Caesareia [I. (1)] and thence to Iconium

[II. (3) 2].

3 I have followed the accounts of the Latin writers in Migne’s Patrol, Lat. vols. clvi. (Guibert) and οἷν, (Rob. Mon., Tudebodius, etc.).

40 THE ROAD-SYSTEM OF EASTERN ASIA MINOR.

Rad. Cadom.; Athena, Guibert 728 etc.) and Mopsuestia (called Mamistra, Mamysta, Manustra) voluntarily surrendered. From Mopsuestia Baldwin marched (through Amanus Gates) across the Euphrates to Edessa, while Tancred proceeded by the Syrian Gates to Antioch. Leaving Herakleia, the larger portion of the Frankish army under Raymond, Bohemond, and Godfrey took a longer route. They entered the ‘Armenian country, and marching by Tyana towards Caesareia reached a certain fortress on an impregnable site, which they made no attempt to take. This fort is not Tyana (in the plain) but Kyzistra (see on II. (8) δ), mod. Zengibar Kalesi, half an hour west of Develi Kara Hissar, “a striking mediaeval castle on a lofty two-peaked hill.... which has been (prob. not correctly) identified with Nora, where EKumenes defied Antigonus in 320 B. c.” (Hogarth in Sir C. Wilson’s Handbook p. 163). After passing Kyzistra they reached Caesareia, where they were welcomed by the inhabitants. They thus took the route traversed by Romanus in 1069 (in the opposite direction). At Caesareia they turned again towards Antioch. Marching no doubt by the Kuru Tchai pass, they arrived first at Plastentia, a town situated in a beautiful and fertile country (multae pulchritudinis et situs uberrimi civitatem), which had been besieged in vain by the Turks for three weeks before the arrival of the Crusaders, who were received with open gates. Plastentia! is evidently the Armen. Ablastha, Syriac Ablestin, which has usually been identified with Albistan, but should apparently be placed in the upper Saros valley. Thence they moved onwards to Coxon (Guibert, 730; Coxan, Tudebod. 776; Cosor, Rob. Mon. 695), 1.6. Geuksun— Kokusos, which was at that time in a very flourishing condition” From Geuksun they marched towards Marash by a route so exceedingly difficult that it calls fprth from the monk Tudebodius such choice epithets as diabolica, exsecrata montanea. It is described as a narrow path (arctus et nimis scrupeus calles praeruptus, Guibert), so broken and steep that everybody alike had to go on foot and it was impossible to pass by the man in front. It is evident, then, that they did not take the pass traversed by Basil in 877 and by Romanus in 1068, ὁ.6. the Ayer Bel pass by Kallipolis and Padasia, which by general testimony is by far the easiest road through the eastern Tauros (see the interesting account by Hogarth, Mod. and Anc. Roads in East. Asia Minor, p. 20), but the route by Geben along Kursulu Su and round Dolaman Dagh to the Jihun—Pyramos and thence to Marash. In the description of this pass in Sir Ο, Wilson’s Handbook, Mr. Hogarth says, “it crosses the spurs of Dolaman Dagh by a very difficult rocky path. The descent to the Kursulu Su which has run, with several falls, through a deep chasm, is very steep, and there is an equally bad ascent, the path being in places only a foot wide” (p. 271). Compare the words used by Robert, the monk of Reims, who gives a vivid account of the soldiers’ despair, ‘Semita non amplius quam unius pedis spatio dilatabatur’ (p. 695).

+ The name is given by Baldric, 3 Von Moltke, quoted by Hogarth in Mod.

2 ‘In qua erat maxima ubertas atque stipata and Anc. Roads, etc. p. 20, describes the road omnibus bonis quae nobis erant necessaria,’ from Marash to Geuksunas difficult. This may

Tudeb, 1.6. : so Guibert, etc, have been the route he took.

THE ROAD-SYSTEM OF EASTERN ASIA MINOR. 41

After emerging from this ‘exsecrata montanea,’ the Crusaders reached Marash (Marasim, G'wibert and Rob. Mon.; Marusim, 7udeb.), where they were hospitably received, and after a day’s rest proceeded towards Antioch.

Basil’s campaign of 8721 (Theoph. Cont. 267 ff., Kedr. 207 ff). Routes: Military Road to Tephrike [IV. (8)]; thence by IV. (2) b to Gurun and [by I. (1)] to Derende and over the hills to R. Zarnuk (west of Melitene).

In 872 Basil marched against the Paulicians by the Dorylaion— Sebasteia road to Tephrike. Failing to take the town, he captured several of their fortresses, the most important being Abara—Amara on the Sivas— Derende road (supra). As he marched southwards along this road, Taranta— Derende submitted and its submission was followed by the surrender of Lokana—Lykandos. From Derende he then crossed the hill-country between the Tokhma Su (Melas) and the Sultan Su (Arab. Karakis) to a position on the river Zarnuk (supra), south-west of Melitene, sending a detachment against Zapetra (Viran Sheher) and Samosata—Samsat. When this detach- ment returned, he marched on Melitene. The Emir’s forces sallied out to meet him but were defeated and shut up within their walls. It was hopeless, however, to attempt to besiege the strongly fortified town and Basil marched northwards again through the Paulician territory by way of Argaous— Argovan, which he captured. Several other forts were taken in the country between Argaous and Arauraca (which seems to be the place meant by Ararach—Rachat), and Basil then returned home. Tephrike was taken and the Paulician community crushed in the following year (873).

EXCURSUS. THE Roya Roap.

BEFORE discussing the line of the Royal Road from Caesareia eastwards, it is well to have realised the importance of the route by Herpa, Tzaman- dos-Azizie, Melitene, and over the Euphrates at Tomisa throughout the Byzantine period and apparently also in the last two centuries B.C., as reported by Strabo on the authority of Artemidorus (supra on I. 2). After passing Tomisa, the road to Persia would naturally turn south by Amida-Diarbekr and along the left (north) bank of the Tigris,—much in the line assigned to it in this part by Kiepert. The distance from the first crossing of the Halys to the Euphrates by this road will be found to cor- respond approximately to the 1194 parasangs (3585 stadia) which Herodotus’ Itinerary (V. 52) gives as the whole distance for Kappadokia and Kilikia (to the Euphrates).

Why then should this line for the Royal Road be doubted? Largely

1 Discussed in Class. Rev., l.c. pp. 136 ff., and only summarised here in the briefest possible manner,

42 THE ROAD-SYSTEM OF EASTERN ASTA MINOR.

because of the so-called ‘ Kilikian question’ in Hdt.’s account of the road, 1.6. the extraordinary fact that while Kappadokia is crossed only in twenty- eight stages (104 par. or 3120 stadia) the large district of ‘ Kilikia,’ extending to the Euphrates, requires only three stages (15} par. or 465 stadia). Now considering the large size of Hdt.’s ‘ Kilikia’ which extends ‘on one side to the Halys (1. 72), on another to Euphrates (V. 52), and also down to the Cyprian Sea (V. 49), the shortness of the distance across Kilikia reasonably excites suspicion. It is possible then that the distances are wrongly distributed between the two districts. This might be due (@) to corruption in the text; for it is admitted that the text of the Itinerary is corrupt at least in one place (de la Barre’s emendation) and probably in another (Stein’s transposition of the three Armenian rivers to Matiene). Or (Ὁ) it might be due to misconceptions on the part of Hdt. His knowledge of the Royal Road is derived not from ὄψες but from some unknown authority. But in one point Hdt. has misconceived his authority. The διξαὶ πύλαι and διξὰ φυλακτήρια passed by the traveller on the borders of Kappadokia and Kilikia must almost certainly refer to the guard at the Cilician Gates. Hdt. therefore conceived the road to pass through the Gates into maritime Kilikia either because he confused the guard at the Kilikian frontier with the guard at the Cilician Gates or because he has put ‘together two separate and un- connected facts: he has put the guard of the Cilician Gates on the Royal Road, and he has connected the “ Royal Road” therefore with maritime Cilicia (V. 49) whereas it crossed Cappadocian Kilikia (V. 52)’ (Ramsay, Cit. and Bish. of Phrygia, I. p.xiv.n.). Such an initial error would lead to other distortions of the facts before him, in order to bring them into harmony with the first misconception. We are familiar with the manner in which modern writers, more scientific than Hdt., often strain facts to make them fit into a theory. But apart from this supposition as to the διξαὶ πύλαι, Hadt., while very likely retaining the whole distance (1194 par.), may have modified the Kilikian distances! to suit his own ideas of ‘ Kilikia,’ which of course he would believe to be right! ‘Kilikia’ with Hdt. is no very definite region: it is the ‘land inhabited by the Kilikians’ as Assyria is that inhabited by the Assyrians, and Egypt by the Egyptians (II. 17),—a con- venient cloak for ignorance. Apparently it is made to extend to the Halys and Euphrates, just because these were the two great dividing lines in Eastern Asia Minor of which he knew, though his knowledge was vague enough. ‘But is ‘ Kilikia,’ after all, a large district in his conception? The distance between the Halys, the Euphrates, and the Cyprian Sea must have been for him exceedingly small. The source of the Halys must have been near the Euphrates, for it divides Lower Asia ἐκ θαλάσσης τῆς ἀντίον Κύπρου ἐς τὸν Εὔξεινον πόντον (1. 72); and it is only five days’ journey across this narrow isthmus (!). Need we be surprised then that, with conceptions like these to accommodate, ‘ Kilikia’ is crossed in three stages of 155 parasangs ? The Kilikia of Hdt.’s authority—if Kilikia was mentioned by him—may

1 i.e. if his authority mentioned Kilikia.

THE ROAD-SYSTEM OF EASTERN ASTA MINOR. 43

have extended to Halys and Euphrates, and he may have given 119} par. as the whole distance for Kappadokia and Kilikia from the first crossing of the Halys to the Euphrates. If so, his conception of Kilikia differed entirely from the Kilikia of the old well-marked and natural division between Kappadok (the country between Tauros and Euxine, Kuphrates and Halys), Kilik (the sea-board country south of Tauros and west of Amanos) and Kumukh (Kommagene) : according to which the Royal Road would not pass through Kilikia at all (on any theory now held).!| We may note that this older division is reflected in Hdt. e.g. v. 49, vii. 91, and that the inclusion of ‘ Posideion on the borders of Syria’ (iii. 91), ie. of the strip of coast fringing Amanos, is consistent with the older conception and constitutes no argument for the inclusion of Kommagene in Kilikia,

From all these considerations it would seem that an undue importance has been attached to the ‘ Kilikian question’ in discussions on the course of the Royal Road. Various solutions of this question are possible: and we must look outside Hdt. for evidence as to the line of the road.

An ingenious theory, which endeavours to explain the three Kilikian stages, has been lately put forward by Mr. Hagarth and accepted by others (see Macan’s Hdt. iv.—vi., vol. II. pp. 299 ff.). This theory brings the road from Pteria either in a direct line to the head of the Tokhma Su, and thence by Derende to Melitene, or by a détowr to Caesareia—Mazaka and east to Melitene (as advocated in this paper): but instead of crossing at Tomisa the road is made to turn south from Isoli and run up the basin of the Gerger Tchai by Kiakhta to Samosata, where it crosses the Euphrates and runs across the desert south of Mount Amasius to Nisibis and thence to Nineveh, &c. The difficulties of such a route over Tauros to Samosata and then through the desert to Nisibis, when an easier and more direct route is open, do not predispose one in its favour. What are its advantages? (1) It claims to solve the ‘ Kilikian question’ by making the distance between the spine of Tauros (the frontier of Kappadokia and Kilikia) to Samosata represent the three Kilikian stages of Hdt. Obviously this solution is reached only by an arbitrary interpretation and limitation of Hdt.’s ‘ Kilikia,’ which makes it include Kommagene while denying that it extends north of Tauros. But if Hdt. says that Kilikia extends to the Euphrates, he also says that it extends beyond the Halys, ὃς ῥέει διὰ Κιλικίας (i. 72): and the inclusion of Posideion (iii. 91) does not support the extension beyond Amanos_ to Kommagene. The reconciliation with Hdt., therefore, disappears. (2) It claims to be supported by Strabo’s account (p. 663) of the κοινὴ ὁδός to the east. The account, however, after Tomisa, is far from clear. At this point there is a break in the description, where Strabo cites the authority of Eratosthenes as confirming Artemidorus’ account of the subsequent route to India and refers to Polybius ; we note a vagueness and a lack of sequence in the following words as compared with the description of the Ephesus-

1 If Hadt.’s authority was an official document, should we not expect it to be based upon this division ?

44 THE ROAD-SYSTEM OF EASTERN ASIA MINOR.

Caesareia section; and it looks as if Strabo had mixed up or fused together two separate routes, one crossing the Euphrates at Tomisa and another ‘beginning at Samosata’ (which is not described and may simply have joined the former road at Amida, so that it would be possible to make a détowr by Samosata). Anyhow the description is not at all clear. The road goes ‘to the Euphrates as far as Tomisa in Sophene. Mr. Hogarth explains that the meaning is that the road ‘ touched Euphrates opposite to Tomisa but did not cross the river. But Strabo does not say this: for surely his words ought plainly to mean that the road crosses to Tomisa. Then he goes on: τὰ δ᾽ ἐπ᾽ εὐθείας τούτοις [Τομίσοις 1 and the dat.?] μέχρι τῆς ᾿Ινδικῆς τὰ αὐτὰ κεῖται καὶ παρὰ τῷ ᾿᾽Άρτεμ. ἅπερ καὶ παρὰ τῷ ᾿Ερατοσθένει... ἄρχεται δὲ [subject ?1] ἀπὸ Σαμοσάτων ... εἰς δὲ Σαμοσ. ἀπὸ τῶν ὅρων τῆς Καππαδοκίας τῶν περὶ Τόμισα ὑπερθέντι τὸν Ταῦρον σταδίους εἴρηκε 40. The last statement is incorrect : it is about 650 stadia. Mr. Hogarth explains the discrepancy by supposing that ‘ Strabo reckons from the spine of Taurus on the right bank lower down than Tomisa, which is not in Cappadocia at all’: it must at least be admitted that ἀπὸ τῶν ὅρων ths Karn. τῶν περὶ Τόμισα is a singular way of expressing it.

But in any case, even if the description were quite clear, we have to remember that this was a trade route and that the Royal Road was not a trade-route but a road for administrative purposes, a road for couriers. The line of a later trade-route would be determined by different considerations ; thus, for example, the Royal Road along the upper Hermos is so difficult that it could never have been chosen as a caravan-route. Lastly, the Roman bridge at Kiakhta need only show the importance of this district in a scheme of frontier defence and the road, if it existed, would be used for this purpose.2 It is hard therefore to see that this route affords any evidence for

the line of the Persian Royal Road. J. G. Ὁ. ANDERSON.

1 It ought to be τὰ δ᾽ ἐπ᾽ εὖθ. τούτοις, etc. cation of the west bank of the Euphrates in 2 Cf. the importance attached to the fortifi- Amm. Marcell. xviii. 7 (supra iii.).

THE TEXT OF THE HOMERIC HYMNS: III. 45

THE TEXT OF THE HOMERIC HYMNS.

Part III.

In the first two parts of this treatise, vol. xv. p. 136 sg. and 251 sg. I discussed the relation of the MSS. of the Homeric Hymns, and the history of the text so far as it could be inferred from them. Incidentally to this in Part II. a number of observations were offered on passages where the reading of the MSS. differed. I now, by way of completion, comment on lines where the MS. testimony is unanimous. My notes are critical, and do not aim at more than noticing passages where there is or has been doubt as to the reading or the sense. For historical and archaeological information I make constant reference to the excellent commentaries of Ilgen (1796), Baumeister (1860), and Gemoll (1886), to which, failing the discovery of new sources, it is unlikely that much will be added. Gemoll’s book is particularly valuable for the care with which the literature is summarised. I refer also frequently to the latest text of the Hymns, published at Oxford in 1896, for which, as Mr. Ὁ. B. Monro has been kind enough to declare (Classical Review, Dec. 1896), I am mostly responsible. It is but the truth to state that the edition owes a great part of what value it may have to Mr. Monro’s unrivalled judgment and feeling for Homer. To this edition the present treatise may serve as Pro- and Epilegomena. Other works upon the Hymns in general there have hardly been since 1886, if we except reviews of Mr. Goodwin's edition (18938), and of this last text; these together with other periodical literature are referred to below. The Index Homericus. Appendix Hymnorum Vocabula continens, 1895, by A. Gehring, is a very useful concordance, spoiled however to some extent by faulty method.

Before proceeding to the Hymns in detail, it may be well to mention two principles of criticism which have been lately advanced by authoritative scholars, Professor Tyrrell, in a brilliant review of Mr. Goodwin’s edition, (Hermathena, ix. p. 31) says: ‘we believe that the only theory on which we can account for the present condition of the text is the assumption that lacunae constantly interrupt the narrative.’ I am inclined to believe that such an assumption, if it is intended as a practical guide for our dealings with the text, can only be admitted under very precise conditions. In textual criticism, as elsewhere, our argument must proceed from the known to the unknown, and our inferences of what took place in the period before

»

46 THE TEXT OF THE HOMERIC HYMNS: III.

our MSS. ought to start from the phenomena that we can observe in them. In a list which will be found in vol. xv. p. 272-4, I have collected the omissions in the different MSS. of the Hymns. It is clear that among these the proportion that are due to obvious graphical causes is vastly greater than those for which no apparent reason can be assigned. I could easily increase the field of my induction. The inference seems sound, and is confirmed by all that we know of uncials and papyrus, that the same holds of MSS. of all ages. Now the actual fact occurring in a particular MS. and the scope that is allowed a modern scholar in reconstituting one, are very different things. The critic is bound by probability, and a low percentage of proba- bility practically disappears for him. Thus M may omit Ap. 23-73 without apparent cause, but the critic cannot make any MS. have omitted fifty lines anywhere he pleases. That fifty lines may have been omitted anywhere cannot be disproved; but it is a possibility that the critic is debarred from making any overt use of. All that a critic may allow himself—a critic who presents suggestions for serious readers and wishes his conjectures to have a chance of permanence—is to assume a moderate omission conditioned by homoeoteleuton, homoearchon, or some similar external cause. And, to clinch his suggestion, he should be able to fill his own lacuna by a verse or verses which should in this way explain their omission. I have, with however little success, endeavoured to do this. Wider lacunae than this are incommensur- able; nor do I find that the interpretation of the Hymns would demand such assumptions could they be legitimately made. In one place only (Herm. 415 sq.) should I be inclined to have recourse to such an expedient—and this is only equivalent to a confession of impotence. The text does not explain itself, no alteration of the words is satisfactory, in despair we say ‘something is lost.’ But what, and how much, it is impossible to define.

Another expedient is proposed by Professor Arthur Ludwich (in the preface to his edition of the Hymn to Hermes. Regimont. 1890/91), that of transposition. This method appears to me to possess less justification than the last. In mediaeval MSS. the scribes, who passed over a corruption or an omission with serenity, seem to have been peculiarly awake to dislocations; I may refer to the Journal of Philology, xxii. p. 181, where it is shown with what care the diorthotes of Laur. xxxii. 9 rearranges the dislocated text of Apollonius Rhodius. I think, as in the other case, we are bound to infer equal care in the earlier centuries. A palliation is sometimes brought to cloak the naked act of permuting lines; it is said that lines often fell out of a MS., and being added in the margin, were, when the particular MS. in its turn came to be copied, inserted in a wrong place in the resulting text. This is at best special pleading, for if you want to transpose a given line or couplet, what ground have you to assume this line or couplet have been assigned by Providence to this accident? Really, there is no palaeographical justification whatever for such procedure; and it is better to call things by their names, and say that when we transpose we do so at the bidding of our personal judgment. On a small scale, and under the same conditions a8 omission, we do find transpositions in MSS.; e.g. Ap. 41 is read by the D family in the place

THE TEXT OF THE HOMERIC HYMNS: ITI. 47

of 36, because 35 and 40 both end in ὄρος αἰπύ, and therefore transposition of small passages, under these conditions, may occasionally be admitted. The wandering eye, however, was far more prone to omit than to exchange : the effect of homoeoteleuton, etc. in causing omission in prose is enormous, as anyone may see by reading the apparatus to Stein’s ‘ Herodotus’: in the liad MSS. omissions are so common as to be almost negligible. Transpositions on the other hand are comparatively rare, nor, to take another argument, did the Alexandrines employ the expedient on any large scale.

I will take this opportunity of noticing Dr. Hermann Hollander’s tract ‘Ueber die new bekannt gewordenen Handschriften der homerischen Hymnen, Osnabriick, 1895. The learned author, to whom belongs the indisputed credit of first classifying the MSS. of the Hymns, deals here with the subject that occupied part of Part I. of this treatise. Upon the general question of the place that these new MSS. are to hold I am glad to see that we are agreed. On some points however I find myself differing. These I will briefly discuss. P. 6 Hollander, speaking of Goodwin’s edition, says it would have been better if the editor had confined himself to a smaller number of MSS. ‘ Welchen Wert hat es, dass man die Lesarten von 13 Codices der 7-Klasse findet?’ he asks, and suggests that P (Vat. Pal. 179) might represent the family. To me it appears that in the case of late and admittedly corrupt MSS., their cumulative testimony is essential; by this means the accidental is separated from the material. If P, as Hollander suggests in his note, were the actual parent of the Paris family then naturally the others might be neglected; but this has to be proved. P has this peculiarity, that it can be dated approximately ; the year of Manetti’s death (1459) is a terminus ad quem before which it must have been written. Otherwise it seems to me one of a family. ‘There- fore in Mr. Gooidwin’s edition I printed in full the evidence that I had upon the Paris family ; in the text of 1896 I denoted their concurrence by the letter p. P.10, 11, it is maintained that S (Vat. 1880) is an apograph of the copy of the ed. pr. which is in Laurenziana. That the ed. pr. agrees in many places with S (and At D)I have pointed out in my first part, p. 157, 8; but I regarded the ed. pr. rather as composed with the help of S and similar MSS. than as a source for them. According to Hollander the Laurentian copy has marginal readings entered in writing and at least one important correction of the text (γενοίμην for γ᾽ ἐροίμην, Ap. 65), and these MS. additions agree with S. The coincidence is interesting, and that there is a connection between the two documents can hardly be denied. But it appears more probable to me that a possessor of the Laur. ed. pr. copied readings out of δα into it, than that the contrary process took place. For (1) Hollander admits S has various readings of its own which are not in the ed. pr. (2) of S’s peculiarities one at least, εὔβωλο σε, 44}. 54, is corrupt as it stands, and therefore was probably in the archetype of S; but it is not in the ed. pr. (3) if S were an apograph of the ed. pr. we should expect to find it a sixteenth century MS. like G, a real instance of a copy from the printed text. But S is well within the fifteenth century, and may have been written before 1488. P. 24 note. I regret that I misunderstood Hollander’s classification. I take the opportunity of

48 THE TEXT OF THE HOMERIC HYMNS: III.

stating my own, to which I hold; it will be elucidated by a stemma, accident- ally omitted in part II., which is now presented. I make 4 classes, 1, M; 2, x = ELIIT and the minor members At DS ed. pr. HTK; 3, y = marginalia of 2; 4. Par. or p, the entire Paris family. 2,3 and 4 are descended from a com- mon ancestor 2.

ἡ |

p Oo EO 7 £7 Ate σι Had ah ABCFEL,L,NOPQ RR, V Non.

FRAGMENT TO DIONYSUS.

I agree with most editors and Εἰ, Maass (Deutsche Litteraturzeitung, Aug. 23, 1898) that the lines quoted by Diog. Sic. III. 65 are part of this hymn.

2. ὡς δέ, Ta μὲν τριάσοι πάντως τριετηρίοιν aiel. A verb is evidently wanted in the protasis, and Mr. Goodwin’s ἅδεν is so far justified. It may be simpler to read ta μὲν as one word τώμεν τε ἐτάμησαν. Translate ‘as three cuts were made, so shall men celebrate you at three-yearly festivals.’ For the construction τέμνειν δύο τρία x.7.r. ‘ cut into two, three’ cf. Theocr. ix. 25 & κρέας αὐτος σιτήθην...... πέντε ταμὼν πέντ᾽ οὖσιν. On the dis- memberment of Dionysus-Zagreus see Preller, J.c. p. 0806 5ῳ., Roscher, Ausfiihrl. Lexicon, p. 1057. The number of parts into which Dionysus was divided is it is true given by the authorities as seven (Lobeck, Aglaophamus, p. 557), but a different tradition will have been followed or established by the Hymn-writer, who seems to have wished to account for the orgiastic τριετηρίς. ,

4—6, 7. It is impossible to deny, with Maass, /.c., that these verses are alternatives ; cf. vol. xv. p. 300.

10. ἐπιλαθόμενοι. Ruhnken emended ἐπιληθόμενον, comparing Dion. vii. ὅθ. A slighter alteration would be ἐπιληθομένῳ. Φ 18 the MSS. vary between ἐκγεγαῶτι and ἐκγεγαῶτα.

THE TEXT OF THE HOMERIC HYMNS: TTT. 49

DEMETER.

This Hymn had until the present year received but little recent attention. Since Biicheler’s separate edition in 1869, we have had Gemoll’s in 1886, but little magazine-literature, the most important being Ludwich’s article Newe Jahrb. f. Phil. 1879 p. 303 sg. to which may be added the contribution of Wackernagel Rhein. Mas. 1889 p. 631, and Bloch Philologus, 1892 p. 65. Wegener’s long study Philologus 1876 p. 227 sq. is a sterile attempt at reconstruction. The programmes, collected up to 1886 by Gemoll are not helpful. We have now the very useful edition by Vittorio Puntoni, Bologna 1896. From the editor's critical method and aims I have expressed my dissent in the Classical Review, Nov. 1896, but I recognise with gratitude his conservative text and exhaustive variorum commentary, and in the prolegomena I have profited by many acute observations. The latest account of the myth is by Leo Bloch, in Roscher’s Lexicon pp, 1284—1379 (Kora und Demeter). I should mention also a work of very different quality, Mr. Pater’s profound and beautiful essay on Demeter in his Greek Studies, Oxford 1895. Mr. Farnell’s recent Cults of Greek States deals with Hecate (vol. ii.).

The Hymn to Demeter is contained in a single document. Criticism of this Hymn therefore is on a different footing to that of the others, and at its freest ; if the Mosquensis contained nothing else, bounds could hardly be set to it. Fortunately we are able to control and ascertain the peculiarities of the MS. by comparison with the rest through the greater part of the other hymns. This has been done in Parts I. and II. to which I may refer the reader. or convenience I repeat here the conclusion there arrived at, The Mosquensis was seen to be a MS. of independent origin, containing a large percentage of peculiar readings. It had been deeply corrupted in the course of transmission and presented many voces nihili and impossible forms, some slightly corrected, but most in their native roughness. Omissions owing to homoeoteleuton were frequent. Instances of gross corrup- tion are such as δίνησι for χλαίνησι, ἐκθήσομωι for τεχνήσομαι, δυσκλέε᾽ for δυσηλεγέ᾽, ἐκ μὴ τοῦ δὲ for ἐκβῆτ᾽ οὐδέ, ἐπ᾿ ἀμήτων for ἐπιβήτορες, νεοθηλέαν ἀγκαλωρήν for νεοθηλέος ἄγκαλον ὕλης. See vol. xv. p. 143-5, That similar depravation had attacked M in the Demeter-hymn we know from the passage 419 sq., which exists nearly identically in Hesiod J’heog. 351 sg. In the Hymn the following corruptions are seen to have taken place: ῥόεια for podeda, μηλοβόστη for μηλόβοσις, ὠκύρθη for ὠκυρόη, ἀκατάστη for ἀκάστη, ταλαξαύρη for γαλαξαύρη. Moreover in lines that are unguaranteed by any external evidence there are impossible and unmetrical words, 12 «@édus 7’, 228 ἐπηλσίησι, 261 ποίησασα (for ποίησα), 267 συναυξήσουσ᾽, 299 πτᾶσα, 430 δρεπομένη. The character of M is therefore, as one might expect, the same in Demeter as in the other Hymns. Emendation therefore must proceed upon these lines and expect to find small omissions and gross unconcealed corruptions, to be healed with a free hand, ,

10. θαυμαστὸν γανίωσα σέβας τότε πᾶσιν ἰδέσθαι. Torte is retained

H.S.—VOL. XVII. ᾿ E

50 THE TEXT OF THE HOMERIC HYMNS: I

by Puntoni and no doubt can be translated. Goodwin’s τό ye however gives a far more natural and Homeric turn, and the alteration is insignificant.

12, 93. τοῦ καὶ ἀπὸ ῥίζης ἑκατὸν κάρα ἐξεπεφύκει a 295 a nr ᾿] ᾽ Ἂς > \ [4 κῶδις T ὀδμῆ πᾶς δ᾽ οὐρανὸς εὐρὺς ὕπερθε K.T.Xr.

κηώδει (Ruhnken), κηώεντ᾽ (Ludwich) κύκλῳ tes (Goodwin) are all too far from the text: κωδείας (Mitscherlich) might stand, if it were established that the word is used of the heads of other flowers beside the poppy. The solution however, is given by Tyrrell’s κὦξ᾽ ἥδιστ᾽ ὀδμή (Hermathena, |.c. p. 34); the emendation preserves every letter of the MS. even to the circumflex upon #. The syllable & fell out, as many syllables have in M, see vol. xv. p. 144. Crasis as Tyrrell observes is not to be tabooed in epos;

cf. 227 «ov, Hesiod, Theog, 447 ἐξ ὀλίγων βριάει καὶ ἐκ πολλῶν μείονα θῆκεν, Ar. Pax, 1282 καὐχένας ἵππων (Kinkel, Epic. Graec. Fragmenta, p. 70).

For the construction ὀδμὴ ὄζει, cf. ε 59 + 210. That M here ‘manum emendatricem experta est’ is really too much for Baumeister to ask us to believe.

17. Νύσιον ἂμ πεδίον. Iam sorry to see that the unhappy suggestion νείατον or μέσσατον for Νύσιον is retained in the last edition of Preller’s Griech. Mythologie, p. 758 n. 3; no support is given it in Bloch’s article, although it pleased Baumeister and is gravely registered by Biicheler and Gemoll. The professional critic is an insufferable creature; he avenges his own ignorance on the document. So this Hymn is to be deprived of its one geographical indication because there was more than one Nysa, and Gemoll ejects Εὐρώπη from Ap. 251 and 291, the earliest mention of the name in- Greek literature, because the connotation of the term is uncertain. Why not banish “EAXas from the J/iad, and in short bring all historical documents down to the level of the critic’s intelligence ? Topography is sacrificed with equally light heart at v. 99 by Wolf, who turned Παρθενίῳ φρέατι into πὰρ θείῳ φρέατι and the clumsy gash is accepted by most editors.

22,23 οὐδέ πη ἀθανάτων οὐδὲ θνητῶν ἀνθρώπων ἤκουσεν φωνῆς οὐδ᾽ ἀγλαόκαρποι ἐλαῖαι.

Neither gods nor men nor olives heard her cry, for the first two were far off, and the trees were deaf. I venture to differ from Professor Tyrrell and many critics, who consider the introduction of trees in this context impossible. The participation (or non-participation) of Nature in human feelings is frequently assumed in Latin poetry, see Eelogues, i. 38, x. 13, and many parallels given by Forbiger. Our‘present passage simply proves that the notion is earlier than one usually supposes. I find Ignarra was of this view and the English translator Robert Lucas, who painfully renders

Alas! nor god nor man would hear her cry, whilst e’en the grove itself denies reply.

THE TEXT OF THE HOMERIC HYMNS: III. 51

Of the conjectures for ἐλαῖαι, I can only say, with great respect for their learned authors, that one is worse than the other. Puntoni relieves me from mentioning them.

Olives are natural features of any Mediterranean landscape ; Proserpine’s flowers grew in the fields among them. The epithet ἀγλαόκαρποι needs no justification (7 115 ἃ 58 it is used of apple-trees) but it is peculiarly appro- priate to the glossy black olive-berry when ripe.

24 εἰ μὴ Περσαίου θυγατήρ atara φρονέουσα ἄιεν ἐξ ἄντρου.

Wackernagel’s οἴη for εἰ μὴ (Tthein, Mus, 44, p. 631) is most unnecessary and has not been accepted.

37. I agree with most editors in accepting Hermann’s lacuna here; the omission of the actual cry, and the necessary antecedent to δ᾽ put the case outside the limits of those to be noticed below. The sense required seems to be ‘when she saw the light leaving her, ἤχησε μὲν Περσεφύνη, ἤχησαν δ᾽ ὀρέων Kopudal.’

«40. οὔτ᾽ οἰωνῶν τις TH ἐτήτυμος ἄγγελος ἦλθεν. The repetition of τῇ from 44 and the somewhat unusual accent are effective. Hermann’s common- place οὐδέ. τις οἰωνῶν τῇ was preferred by editors until Gemoll. The principle of anomalia should play a far greater part than it actually does in metrical criticism.

51-61 ἀλλ’ ὅτε δὴ δεκάτη οἱ ἐπήλυθε φαινολὶς ᾿Ηώς, ἤντετό οἱ ‘Exarn, σέλας ἐν χείρεσσιν ἔχουσα. καὶ ῥά οἱ ἀγγελέουσα ἔπος φάτο φώνησέν τε' πύτνια Δημήτηρ, ὡρηφύόρε, ἀγλαόδωρε, τίς θεῶν οὐρανίων ἠὲ θνητῶν ἀνθρώπων ἥρπασε Iepoepovny καὶ σὸν φίλον ἤκαχε θυμόν ; φωνῆς γὰρ ἤκουσ᾽, ἀτὰρ οὐκ ἴδον ὀφθαλμοῖσιν ὅς τις env: σοὶ δ᾽ ὦκα λέγω νημερτέα πάντα. ὡς ἄρ᾽ ἔφη ‘Exdatn: τὴν δ᾽ οὐκ ἠμείβετο μύθῳ Ῥείης ἠύὔκομου θυγάτηρ, ἀλλ’ ὦκα σὺν αὐτῇ ἤϊξ᾽ αἰθομένας δαΐδας μετὰ χερσὶν ἔχουσα.

This passage is treated by Puntoni, introd. pp. 3-9. His discussion contains many acute observations by which I have profited. He analyses and refutes the suggestions of modern scholars with much success, but towards the author of the Hymn his attitude is too much that of the giudice istruttore inviting a prisoner to reconcile his inconsistent statements. Such legal criteria have no application to literature at all, much less to a composition of the nature and the age of this Hymn. The difficulties of this Hecate-episode are concerned principally with ἀγγελέουσα v. 53 and σοὶ δ᾽ ὦκα λέγω νημερτέα πάντα 58. The role that Hecate plays in this Hymn is as regards the development of the action, useless: this is justly observed by Puntoni, p. 5. Her introduction both here and 438 sg. is doubtless due to the formal con-

E 2

52 THE TEXT OF THE HOMERIC HYMNS: III.

secrated story as actually represented in the Mysteries, and here I will refer to the acute and subtle observations (p. 9) in which Puntoni shows how a con- secrated story may fetter a poet who puts it into literary form. Critics how- ever have not been willing to accept this negative view of Hecate, and have tried in this passage to emend or interpolate so as to make her offer practical advice to Demeter. The majority, following Hermann, assume an omission after 58 in which a reference to Helios should be contained; ‘I know nothing, but Helios will tell you’ :! again, ἀγγελέουσα in 53 is objected to because no ἀγγελία is actually given; therefore ἄγχε θέουσα, ἐγκονέουσα are proposed.

I will briefly state my own view. Hecate, for all that her introduction into the poem may be due to the myth, is given a certain ἦθος by the poet. She is an officious, well-meaning, nurse-like person, inefficacious, but eager to offer assistance: the dignified and serious Demeter ignores her entirely, Hecate now has an ἀγγελία, news, to offer: she is one of the two beings who heard Kora’s first shriek : and this important information she comes to give. Her natural volubility however induces her to put the cart before the horse, and satisfy her own curiosity before giving her information concisely. ‘Oh Demeter, who has carried off Persephone? I heard the cry, but that’s all I know: you are certain to have the best and latest information.’ To which the daughter of Rheia vouchsafed no answer, but tacitly allowed her to go with her on the rest of her journey.

To give this sense to v. 58 one must construe λέγω δέ σοι πάντα νημερ-᾿ téa [εἶναι], 1 say that to thee all things are immediately manifest [or, certain]. I think that there is no essential difficulty in so doing, though no doubt it may startle people familiar with the old view. Not the least advantage is that a genuine and indisputable value is put upon ®@ka—a word that has been a stumbling-block in most other interpretations.

To object that Hecate only found Demeter on the ninth day is surely to introduce considerations entirely foreign to this sort of literature; on the other hand it is a real objection to the usual lacuna that it involves the situation of Demeter, a superior goddess, depending for counsel upon an inferior, Hecate. The objection that Hecate disappears from the moment of finding Helios till the return of Persephone is again false in such a context. If we are to reduce the poem to the conditions of an ordinary tale, the situation is that Demeter having vainly sought hither and thither for eight days, on the ninth betakes herself to the last resort of the Homeric world, the all-seeing Helios, safeguard alike of husbands and mothers, and on her way thither is met by the eager and unnecessary Hecate.

I trust that these considerations may commend themselves to Signor Puntoni; I can hardly expect that they, or any other argument will avail to recall his dismemberment of the poem. On the construction ὅς tus ἔην ef. vol. xv. p. 288, Dem. 119.

70. καταδέρκεται, 71 ὄπωπεν. The second person alone is possible and Ruhnken emended καταδέρκεαι, ὄπωπας. The source of the corruption is

'E.g. σοὶ δ᾽ ὦκα λέγοι νημερτέα πάντα | ᾿Ηέλιος ὃς πάντ᾽ ἐφορᾷ καὶ πάντ᾽ ἐπακούει (Hermann). you νήμερ ρᾷ

THE TEXT OF THE HOMERIC HYMNS: III. 53

hardly an error of sense (as Hest. xxix. 3 vol. xv. p. 268), since here the subject is very close. One must look rather to a scribes-tendency to separate adjoining vowels, e.g. K 82 ἔρχεαι ἔρχεται, 115 νεμεσήσεαι νεμεσήσεται. Karaédépxeat then having become καταδέρκεται, ὄπωπας followed suit. Somewhat similarly Ap. 71 (ns, ἀτιμήσω in x (vol. xv., p. 269).

64, αἴδεσσαί με θεὰν σύ περ. This is Ludwich’s excellent conjecture (Newe Jahrb. f. Ph. 1879, p. 305) for θέας ὕπερ: the other conjectures, collected in Puntoni, are unsatisfying ; best in sense is Peerlkamp’s θεὰν θεός, but like the Dutch school generally it leaves entirely out of account the evidence of the document. This is amply recognised by Ludwich; θεᾶσύπερ gave θεᾶς ὕπερ.

85. ἀμφὶ δὲ τιμὴν | ἔλλαχεν ὡς τὰ πρῶτα διάτριχα δασμὸς ἐτύχθη. Τιμὴν has been altered into τιμῆς and τιμῇ--- ΘΟ ΟΘΒΒΆΡΙΥ it would seem, for the sense ‘about honour, he hath by lot even as the partition was made’ is very prosy. I would take ἀμφὶ as part of the following verb, whether ἔλλαχεν or some other (for ἔλλαχεν may have come from 87), and separated from it by tmesis : τιμὴν would then be the direct accusative, and ἀμφὶ possess the quasi-intensive sense ‘fast, tight.’ For the order of the words cf. Hes. Opp. 74 ἀμφὶ δὲ τήν γε | Ὧραι καλλίκομοι στέφον ἄνθεσιν εἰαρινοῖσι.

87. τοῖς μετάναίεται (sic) τῶν ἔλλαχε κοίρανος εἶναι. Puntoni ventures on one of his rare conjectures, μεταναιετάειν ; it is, 1 am sorry to say, wholly unnecessary, and the difficulties raised as to this part of the poem are pure moonshine. The purely clerical correction of Voss μεταναιετάει should satisfy anyone. ;

99. Παρθενίῳ φρέατι ὅθεν ὑδρεύοντο πολῖται. For the metre of φρὲατι οἵ. below, 101, παλαυγενέϊ ἐναλίγκιος, and La Roche’s qggection on Ὦ, 285. The difficulty that Puntoni (p. 85) feels at the dative is surely illusory; see later on Aphr.173 ἔστη ἄρα κλισίη. On Wolf's πὰρ θείῳ φρέατι I have spoken above: I find unexpected assistance in Baumeister, whose words deserve all publicity. ‘Neque omnino critici diligentis esse videtur, vocem difficilem explicatu Ilap@ev/@ pro corrupta vel interpolata habere.’ Si sic omnia.

115. οὐδὲ δόμοισι | πιλνᾶς ; ἔνθα γυναῖκες x.t.r. It being held as certain that this active form (Hes. Opp. 510 wiAvd 3rd pers. sing.) cannot be used in a neuter sense, Voss’s πίλνασαι, supported among other forms in the Lexx. by πέλναται, Ap. Rhod. iv. 592, and ὑποδάμνασαι π᾿ 95 seems better than Hermann’s 7Avd, which rests only on E 99 @ τε σὺ πάντας | δαμνᾷ ἀθανάτους and is graphically farther from πιλνᾶς ; the syllable ac (as ἐπιβήσεσθ᾽ 332) was omitted doubtless from the effect of the hiatus. More or less parallel are Theognis, 1388, δώμνας δ᾽ ἀνθρώπων πυκινὰς φρένας, where Bergk reads δώμνασαι, δαμνᾶς the reading of D on ἘΞ 199, the variants πειρᾷ πειρᾶς φ 459, Aristoph. Hg. 161, καταγελᾶι R for καταγελᾶς.

119. τέκνα φίλα τίνες ἐστε γυναικῶν θηλυτεράων M. By an iota Fontein healed this line. (Cf. Hermanin’s correction πῆμα μέγ᾽ αἱ θνητοῖσι for μέγα θνητοῖσι, Hes. Theog. 592.) For the parenthetical use of αἵτει!ες ef. vol. XV. p. 288, and Theocr. xxii. 54 χαῖρε ξεῖν᾽, ὅτις ἐσσί. I should be inclined

54 THE TEXT OF THE HOMERIC HYMNS: ΤΗ.

to stop τέκνα φίλ᾽, αἵτινες ἐστε, γυναικῶν θηλυτεράων ; Demeter does not know the girls’ names, and addresses them by such description as she can give: ‘dear children of women, whosoever ye be.’

122. δὼς ἔμοι γ᾽ ὄνομ᾽ ἐστί. Fontcin’s Anw has been widely accepted and is retained by Puntoni. It provides for the metre, and the corruption is not unlike that of ἀέδης for ἀδῆ of Ap. 76. Still as Demeter is called Anew already in line 47 it seems better that here she should invent an epithet which was not one of her real names; her story is false, her name should be false also. But the matter is settled by Hesiod’s line, Opp. 356, das ἀγαθή, ἅρπαξ δὲ κακή, θανάτοιο δότειρα, where das is plainly an adjective. The linguistic connection between the Hymns and Hesiod is palpable, cf. the statistics in Francke’s dissertation quoted below ; the list of Oceanides, 418 sq. ; and Herm. 36, a literal parody of Opp. 365. The necessary μὲν was supplied by Brunck.

ΟΡ « Ν ” 125 sq. οἱ μὲν ἔπειτα a , al νηὶ θοῇ Θορικόνδε κατέσχεθον, ἔνθα γυναῖκες , ͵ > 4 > / ἠδὲ Ν > U ἠπείρου ἐπέβησαν ἀολλέες, NOE καὶ αὐτοί. ΄ Ν ΄, ΄ δεῖπνον ἐπηρτύνοντο παρὰ πρυμνήσια νηός.

It is a difficult question whether Hermann’s lacuna after 127 is necessary. On the whole I think we may see in the passage only a compression of the usual formula (A 432 ¢ 150, 547, μ 6 & 346 Apollonius i. 1110) if we compare the general elliptical style of this Hymn; eg. here the other γυναῖκες are first mentioned when they disembark, 317 ὡς ἔφαθ᾽ is said of Zeus while his*éctual words are not given, 446 sg. we have the remarkable construction νεῦσε δέ of κούρην x.7.r. where both verbs ὑπὸ ζόφον [ἰέναι and παρὰ μητρὶ [μένειν] are saved.

137 ὡς ἐθέλουσι τοκῆες" ἐμὲ δ᾽ αὖτ᾽ οἰκτείρατε κοῦραι /, / / ’ὔ Ν , > ὦ προφρονέως φίλα τέκνα τέων πρὸς δώμαθ᾽ ἵκωμαι ἀνέρος ἠδὲ γυναικός.

Here on the other hand the MS. reading can hardly be construed, though Baum. and Puntoni print it. Οἰκτείρατε cannot by any stretch govern τέων ; Cobet’s ἐμοὶ δ᾽ αὖτ᾽ εἴπατε is mere patch-work, and Ruhnken’s τέως improbable in the construction, apart from the lateness of such a use of the word. A lacuna therefore containing the idea ‘and tell me, seems wanted e.g. τοῦτο δέ μοι σαφέως ὑποθήκατε ὄφρα πύθωμαι (cf. 149); in this case προφρονέως which hardly suits οἰκτείρατε would qualify ὑποθήκατε. Besides the homoeoteleuton of my supposed πύθωμαι it is possible that προφρονέως and πρόφρων had to do with the omission of a line. The omission of αὖτ᾽ has been proposed, but the apparent metrical difficulty of τοκῆες is in its favour.

144. καὶ κ᾽ ἔργα διαθήσαιμι γυναικός. .Voss’s διδασκήσαιμι γυναῖκας (Hes. Opp. 64 ἔργα διδασκῆσαι) is excellent in sense and not farther from

THE TEXT OF THE HOMERIC HYMNS: III. 55

the MS., than the ascertained corruptions of M. Of the other suggestions Bothe’s διαθρήσαιμι (with γυναικός) is perhaps the best, though ἀθρεῖν (Ilerm. 29, 414) and its compounds do not seem to contain the sense of ‘oversce.’

157. κατὰ πρώτιστον ὀπωπήν. Ignarra’s ὀπωπῆς accepted by Voss is well refuted by Franke. Gemoll, whose conjecture I will not quote, returns to the charge. That superlatives occasionally take two terminations is a well-known fact, see Kiihner-Blass, 1, p, 554, Anmerk., and πρώτιστον ὑὐπωπὴν is as well established as ὀλοώτατος ὀδμὴ ὃ 442.

205. ἣ δή of καὶ ἔπειτα μεθύστερον εὔαδεν ὀργαῖς : ‘who, 1.6. Iambe, pleased her humour even afterward, that is Demeter continued to find relief in Iambe’s company. Biicheler’s ὀργή is not only prosaic, but incorrect, for Demeter’s mood altered considerably before long. Of Iambe, who was Demeter’s companion as long as she remained in Celeus’ house it would be just to say ‘she pleased her afterwards also, not merely for the moment. It is easy to understand that Voss’s εὔαδ᾽ ἑορταῖς did not find favour with Ruhnken. ’Opy7 is post-Homeric,! but it occurs in Hesiod and frequently in the next age, see the Lexx., and ef. for the sense of ‘mood, humour’ Tyrtaeus 11, 8 εὖ ὃ ὀργὴν ἐδάητ᾽ ἀργαλέου πολέμου, Simonides, Amorg. 7, 11, ὀργὴν δ᾽ ἀλλότ᾽ ἀλλοίην ἔχει, 41 ταύτῃ μάλιστ᾽ ἔοικε τοιαύτη γυνή | ὀργήν: Theognis, 213, Κύρνε, φίλους κάτα πάντας ἐπίστρεφε ποικίλον ἦθος | ὀργὴν συμμίσγων ἥντιν᾽ ἕκαστος ἔχει, 215 Ιουλύπου ὀργὴν ἴσχε πολυπλόκου κιτιλ., 312 γινώσκων ὀργὴν ἥντιν᾽ ἕκαστος ἔχει, cl. 964, 1059, 1072, 1073. I find no difficulty in the double dative ; it is a clear case of the σχῆμα καθ᾽ ὅλον καὶ

μέρος, commoner no doubt with the accusative, but cf. @ 129 δίδου δέ οἱ ἡνία χερσίν (Jelf). A 24 ἀλλ᾽ οὐκ ᾿Ατρείδῃ ᾿Αγαμέμνονι ἥνδανε θυμῷ. cl. the variant. Ν 82 τήν σφιν θεὸς ἔμβαλε θυμῷ. Scut. 221 ὥμοισιν δέ μιν ἀμφὶ.

211 δεξαμένη δ᾽ ὁσίης ἕνεκεν πολψπότνια δηώ τῇσι δὲ μύθων ἦρχε ἐύζωνος μετάνειρα.

Mr. Tyrrell justly objects to Voss’s ἐπέβη as a preposterous substitute for ἕνεκεν; even Baumeister considered it ‘paullo longius a litterarum ductibus recedentem. ‘Ooins ἕνεκεν seems particularly sound, ‘to save the rite, that is primarily the θεμιτόν of 207 and secondly the ritual of the historical mysteries. The other expedient, Franke’s πίε for πολὺυ- is almost as violent, and the compound epithet is well-established, cf. Apollonius, i. 1125, 1151, iv. 1069. Mr. Tyrrell’s own ἐνέχεεν for ἕνεκεν is more than ingenious ; but why should we complicate the situation by making Demeter pour the κυκεών from one cup into another? She drank it simply; ἔπιε, as the various other accounts of the story have it (and cf. Hippocr. Acut. 39 οἱ δὲ καὶ κυκεῶνα ἔπιον); any substitute for ἕνεκεν must=éme, as Voss remarks, I am surprised, considering the soundness of the line that no one

1 As χλεύη (202) also. Useful lists of this De h. in Cer. Hom. comp. dict. aetate, Kiel, Hymn’s vocabulary are given by K. Francke, 1881.

56 THE TEXT OF THE HOMERIC HYMNS: III.

(except now Puntoni, whose discussion pp. 60, 61 is correct though needlessly wordy), has made a lacuna, ¢.g. :—

ἔκπιεν: ἡ δὲ λαβοῦσα δέπας θέτο ἔνθεν ἄειρε,

or some other line sufficient to set up homoeoteleuton.

227.-Opéxpw κοὔ μιν ἔολπα καλ. Mr. Monro and Sig. Puntoni inde- pendently keep θρέψω and κοῦ ; the abruptness is not intolerable, and for the crasis see on v.12. Mr. Agar’s (Classical Review, 1896, Nov. p. 388) θρεψόμεν is no better than any of the other alterations, and his palaeography is illusory.

228 sg. οὔτ᾽ ap’ ἐπηλσίησι δηλήσεται οὔθ᾽ ὑποταμνόν, οἶδα γὰρ ἀντίτομον μέγα φέρτερον ὑλοτόμοιο οἷδα δ᾽ ἐπηλυσίης πολυπήμονος ἐσθλὸν ἐρυσμόν.

Ruhnken’s ἐπηλυσίη in 228 seems certain, cl. Herm. 37. The conjectures for ὑλοτόμοιο introduce words which do not exist, and moreover give no satisfactory sense. In the Classical Review, 1895, February, p. 13, I suggested that ὑποταμνόν and ὑλοτόμοιο the ‘under-cutter’ and the ‘ woodcutter’ were periphrases for ἕλμενς or σκώληξ, the worm, and that Demeter guaranteed Metaneira’s child against this malady. Cf. Aratus 959 σκώληκες | κεῖνοι τοὺς καλέουσι μελαίνης ἔντερα γαίης, Hesiod’s φερέοικος ‘snail, ἀνόστεος ‘cuttle’ ἕδρες ‘ant, Theocr. ΧΙ. 35 βούτομον ‘rush, and the epithets βοὸς ὑλοφάγοιο, Hes. Op. 591, γειοτόμοι μύρμηκες, Apollonius, iv. 1453. The forms of the words ὑποταμνόν, ἀντιτόμοιο, ὑλοτόμοιο seem guaranteed by the jingle. ᾿Αντίτομον, ἀντιτέμνω etc. are known from other places (v. Lexx.) in the sense of ‘antidote, but ἀντέτομον was chosen here doubt- less for the verbal antithesis to ὑλοτόμοιο. “1 know a counter-cutter to the woodcutter. ὑποτάμνον is the part., and must be so accented. Adjectives in -νός doubtless τά in the scribe’s head. For substantive and adjective cf. ἀμφιφῶν (Monro, Homeric Grammar, § 243, 1). |

236. οὔτ᾽ οὖν σῖτον ἔδων ov Onodpevos. Δημήτηρ. Hermann’s lacuna, and his supplement of γάλα μήτρος to 236 are generally accepted, and are an excellent instance of the method.

240. λάθρα φίλων γονέων. AadOpa φίλων weathered Ruhnken and Hermann, and seems first to have roused the suspicion of Spitzner. Zeno- dotus at I’ 244 read ἑῇ for φίλῃ, but I am disposed to think λάθρα or λάθρη ἑῶν too violent an alteration here, and much more so κρύβδα φίλων. Is λάθρᾶ impossible? It is true that it occurs nowhere else but in a doubtful fragment of Euripides (1117 Dind. v. 28), but why may it not be formed on the analogy (perhaps false) of σέγῃ σῖγα, aud ἅμα, κρυφῇ κρύφα, διχᾷ δίχα, τριχῇ τρίχα (Kiihner-Blass, ii. p. 306) ?

258. μήκιστον ἀάθης. The probability is certainly strong in favour of Voss’s νήκεστον, cf. Hes. Opp. 283, but I should like to feel certain that the

writer did not intend μήκιστον ἀάσθης as a superlative of μέγ᾽ ἀάσθη and ἀάσσατο δὲ μέγα.

THE TEXT OF THE HOMERIC HYMNS: III. 57

267. συναυξήσουσ᾽. συνάξουσ᾽ Ignarra, Ilgen. The word must result ove no from συνάξησ᾽, i.e. συνάξησ᾽ corrected; cf. 261 momocaca=olaca, +22 ar n

ἀκατάστη = ἀκατη, 431 δρεπομένη = δρεπόμεν, and vol. xv. p. 263. ἃ became av in obedience to a common law, ef. vol. xv. p. 289, and Herod. ii. 111 συναυ- λίσαντας for cvvadicavta. The meaning of 265—267 is not likely, failing new materials, to be satisfactorily made out; Matthiae’s alteration of ἐν ἀλλήλοισι into ᾿Αθηναίοισι will be admitted to merely heighten the confusion.

269 ἀθανάτοις θνητοῖσιν ὄνειαρ Kal χάρμα τέτυκται.

The line as it stands labours under two defects: (1) the absence of copula or other connection between ἀθανάτοις and θνητοῖσιν, (2) hyper- metry, unless ὄνειαρ be read as a dissyllable. The former difficulty has been met in two ways: (a) by altering ἀθανάτοις to ἀθανάτων (Stoll); this allows the genitive to depend on μέγιστον of the line before, but it gives but a mediocre sense. Demeter’s magnificent boast that she is the help and joy of gods and men, Olympus and earth (as the Hymn proves) equally dependent on her, is watered down to the undignified contention that she is the most useful divinity to men. (ὦ) θνητοῖσιν is altered to θνητοῖς τ᾽, and this I incline to accept, both for the sake of the sense, and as being a far slighter palaeographical change than the alteration of dative to genitive.

“Ovevap remains. With θνητοῖς τ᾽ preceding we have to scan it JC _; whether this is best done by leaving ὄνειαρ tel quel, or by writing it dveap (Ilgen’s proposal), is a matter perhaps best left to etymologists. In the Oxford text Mr. 1). B. Monro, following Schulze, Quaestiones Epicae, p. 228, wrote dveap, and in this I should acquiesce.

Two attempts to give ὄνειαρ its natural full value of J _ _, viz. Mr. Tyrrell’s ὄνειαρ kappa τέτυκται and Mr. Agar’s ὄνειαρ καὶ πολὺ χάρμα, have raised an interesting controversy upon the legitimacy of the lengthening of the 4th thesis by position in the Classical Review for Dec. 1896, Feb. and April, 1897. The question it is evident does not arise directly if θνητοῖς τ᾽ be accepted, and so far as these two emendations are concerned the evidence is not sufficiently strong to rule them out. The examples given by Mr. Platt from Hesiod alone (/.c. April, 1597, p. 154) amply cover them, and between the greater Hymns and Hesiod there is, as I have noticed on v. 122, con- siderable connection. As it is maintained that this lengthening of the 4th thesis is totally absent from post-Homeric writers, I may quote Matro, 35, ἣ μόνη ἰχθὺς οὖσα τὸ λευκὸν καὶ μέλαν οἷδε. For the rest Mr. Tyrrell’s κἄρμα, though ingenious and explicable palaeographically, is an improbable substitute for the familiar χάρμα; Mr. Agar’s hemistich does not require consideration.

279. ξανθαὶ δὲ κόμαι κατενήνοθεν w@movs. Ruhnken substituted the singular, which no doubt is correct, and is followed by Biicheler and Gemoll, The latter however quotes a remark from Franke that to the writer of the

58 THE TEXT OF THE HOMERIC HYMNS: It.

hymn κατενήνοθεν may have scemed a plural; with this I entirely agree, and would refer to v. 351.

289. ᾿Ηλούεον is not ‘impossible’ (Gemoll) but only a false formation. Cf. Smyth, Zonie Dialect, p. 535. Somewhat similarly Apoll. 120, the scribes give us the unmetrical λοῦον. The washing was not unnecessary, as Ludwich (NV. J. f. Ph. 1879 p. 307) thinks; the child must have been covered with wood-ash. The sisters only performed Demeter’s usual functions.

296. πολυπείρονα λαόν. Ἰ]ολυπείρονα again is a false-epic formation ; as ἀπείρων = boundless, πολυπείρων = many-bounded, wide. (So Voss). IloXv7apova which Gemoll prints lacks any probability.

328. τιμάς 8 ἅς κ᾽ ἐθέλοιτο μετ’ ἀθανάτοισιν ἑλέσθαι. ᾿Εἰθέλοιτο survived Ruhnken and Ilgen, but moved the just indignation of Hermann, who substituted κεν ἕλοιτο, and θεοῖσι at the end of the line from 444. This arrangement though since accepted is by no means binding ; ἐθέλουτο may be more naturally explained as having supplanted an original βόλοιτο. A 319 for δὴ βόλεται Ly, 45, 99 Ven. Vat... Mg have δὴ ἐθέλει; on ἐβόλοντο a 234 βόλεσθε π᾿ 387 there is no such gloss. ᾿ὡθέλοιτο then may result from

ἐθέλοι κε βόλοιτο.

337. ἀπὸ ζόφου ἠερόεντος. ΑΒ Voss pointed out, ὑπὸ is the preposition in the same phrase ¢ 56 Hes. Theog. 653, but it docs not seem therefore obligatory here. Where a change is so slight it is more prudent to hold to the tradition. Hes. Theog. 652, 659, the MSS. vary between ἀπό and ὑπό: ib. 669 ὑπὸ without variant.

344 ἠδ᾽ ἐπ᾿ ἀτλήτων ἔργοις θεῶν μακάρων μητίσετο βουλῇ.

In face of ἠδ᾽ ἐπ᾽ ἀτλήτων I must confess to being helpless: the metre as well as the sense of ἐπ᾿ ἀτλήτων defy explanation ; probably in ἐπ᾽ ἀτλήτων we have one of M.’s characteristic corruptions, but in the absence of another source it is hopeless to expect one conjecture to outweigh another. The earlier attempts are collected by Ilgen.!. In the next line however we may see some light; the line is unmetrical, and as θεοὺς in 325 has caused the loss of U_, so here the same synizesis has removed οὐ ~; in 325 Valckenar restored πατήρ, here after μακάρων insert στυγερόν, χαλεπόν or a similar word; ‘she was devising in her mind an evil thing against the deeds of the blessed Gods.’ The ingenuity of Ignarra’s ὀργισθεῖσα for ἔργοις θεῶν deserves recognition. As ‘to the reference in ἥδ᾽ (so corrected out of ἠδ᾽) I incline against Matthiae and Baum., to give it to Demeter, cl. v.27. Persephone’s innocent character was not given to revenge.

349. ἐρέβευσφι. This spelling is the vulgate at I 572 where however the’ MSS, La... 15) τῶ ¥CDaio ea See Mag Meee Ea, Moveore τὰ

Δ Ignarra’s ἀποτηλοῦ is one of the best. To γὰρ ἀτλήτων | μῶμον ὑπερπροφυγὼν, which how- the exx. of τὰ short in the commentators I add ever Hauvette rightly considers late. the epigram ap. Steph. Byz. 8.0υ. Θούριοι : τῶν

THE TEXT OF THE HOMERIC HYMNS: III. 59

ἐρέβεσφι; Hes. Theog. 669 only ‘V,’ has épéBeode. With Franke | have restored the correct form.

351 παύσειεν. I have little doubt that this word is original. <A correction such as παύσαιτ᾽: ἐπειὴ or Hermann’s λήξειε would be casy, but such a method ends by divesting documents of their linguistic value: cf. v. 279 above, and Aphr. 280 (vol. xv. p. 298). The writer may have justified himself by the epic examples usually quoted, Hes, Scut. 449 παῦε μάχης, ὃ 659 μνηστῆρες (as) δ᾽ ἄμυδις κάθισαν καὶ παῦσαν ἀέθλων, Where the MS. evidence is all but entirely in favour of the nominative.

366. σχήσησθα. The earlicr editors kept this form, the later follow Hermann (quoted from Franke) in reading σχήσεισθα. It is safer to leave what the MS. gives us. Cf. in gencral Bergk’s note on Sappho fr. 21 (P. LZ. G. ed. 4, vol. iii. p. 96).

371 αὐτὰρ Gy’ αὐτός ῥοιῆς κόκκον ἔδωκε φαγεῖν μελιηδέα λάθρη ἀμφὶ νωμήσας.

These words have tortured and been tortured by every interpreter since Ruhnken confessed his ignorance. The use of the word νωμᾶν is well put out in Voss’s long note, but his conclusion that νωμήσας here = προσαγαγόμενος ‘drawing her to him’ is singularly inconsistent. The word as we know it has two senses; (1) to distribute (2) to handle, whether literally, the bow, etc., or figuratively, of the mind, to ‘turn over, examine.’ The sense of distribution, with or without the alteration of ἀμφὶς for ἀμφὶ € is proposed by Santon, Hermann, Franke, and acquiesced in by Baum. and Gemoll. The Greek is possible, but the interpretation introduces an idea quite foreign to the story, that Hades shared the pomegranate with her. We have further a view dating from Matthiae and which is suggested by Liddell and Scott and apparently approved by Prof. Tyrrell (/. 6. p. 39) that νωμήσας is absolute, and -- παπτήνας. This rests upon Herod. iv. 128 νωμεῦντες οὖν σῖτα avatpeo- μένους τοὺς Δαρείου. When Plato, Cratylus, 411 1), says εἰ δὲ βούλει ἡ γνώμη παντάπασι δηλοῖ γονῆς σκέψιν καὶ vOunow τὸ γὰρ νωμᾶν καὶ σκοπεῖν ταὐτόν, he means by σκοπεῖν ‘search, turn over,’ in the primary sense of vopav. <A third view assumes a τμῆσις, and that the verb ἀμφινωμᾶν (Aeschyl. fr. 297, Nauck) = surround, clasp. The word would suit a nurse ‘handling’ a baby, but hardly applies to a grown-up woman.

For myself I can only explain νωμήσας as governing κόκκον or pony; ‘handling it about her.’ Aidoneus no sooner receives the news of the will of Zeus, than he thinks of the pomegranate which will serve his end (and ‘smiles with his eyebrows’ ; Wegener followed by Gemoll is plainly wrong in putting Persephone’s eating the pips into the past, for why should the circumstance be mentioned here 7); Persephone jumps up in a transport, while he, ‘ handling’ (0.6. stealthily reaching for it and opening it) a pome- granate ‘about,’ that is virtually behind, ‘ her, privily,’ gives her a few pips to eat. Λάθρη again is of course Περσεφόνης, so as not to rouse her suspicion,

60 THE TEXT OF THE HOMERIC HYMNS: TII.

not “Ἑρμαίαο, who had no personal interest in the ménage. Puntoni’s discussion (p. 40, 41) is interesting and clear, but goes off into the vague as soon as it leaves the actual interpretation of νωμᾶν. Persephone’s account of the incident (411 sg.) is naturally and properly different to that given here. Actually, in her exaltation, she paid no attention to the pomegranate pip: questioned by Demeter she remembers that Aides did put it into her mouth, and not at her wish ; but there was no sort of struggle. Any other criticism is forensic and inapplicable.

386. Ruhnken’s objection to the propriety of comparing Demeter to a μαινάς is a lapse of perception commoner in other critics. His contemporaries quoted X 460 of Andromache.

399. ef δὲ πτᾶσα. I may be allowed to point out the superiority of Mr. Goodwin’s τὸ πάσσαο over the vulgate δ᾽ ἐπάσω, inasmuch as it takes account of the τ in πτᾶσα. How does Puntoni get his ἢ δή τοι πάλιν adres toda’ out of the line? It is a curious but of course perfectly natural circumstance that in this damaged passage the perfect MS. had several gross corruptions. I see no reason, with Dr. O. Crusius (Literarisches Centralblatt, 1895, 5 Januar, p. 21) to plead for the retention of παομένη v. 393.

404. Ruhnken is evidently right in supposing an omission here; a question is required for the answer at 415. Supply as Hermann λέξον δ᾽ ὅππως ἦλθες ὑπὸ ζόφον ἠερόεντα, or perhaps εἰπὲ δὲ πῶς σ᾽ ἤρπαξεν Vv. ζ. ἡ. The homoeoteleuton with 403 explains the omission.

412. αὐτὰρ ἐγὼν ἀνόρουσ᾽ ὑπὸ yappatos. Apparently the first particle (which would naturally be ἤτοι) has been superseded by the second, as Σ 208 αὐτὰρ ᾿Αχιλλεὺς ὦρτο διίφιλος" ἀμφὶ δ᾽ ᾿Αθήνη several MSS. have αὐτὰρ ᾿Αθήνη. At the end of the line ὃ λάθρη is certain, as Mr. Goodwin’s note shows, and the propriety of the adverb cannot be a moment in doubt.

419 sq. The names of the Oceanids agree in the main with Hesiod’s catalogue, Thevg. 349 sg. Hesiod names 41, of whom the Hymn-writer has 16, and adds of his own Leucippe, Phaeno, Melite, Iache, Rhodope. The passage has this interest among others, that it supplies an authority to which to compare this part of the Hymn, otherwise dependent entirely on our fourteenth-century scribe. Ταλαξαύρη for γαλαξαύρη is trivial, but poeca for ῥόδεια, μηλοβόστη for μηλόβοσις, ἀκατάστη for ἀκάστη are remarkable, and illustrate the result of solitary tradition. ‘There are many fifteenth century MSS. of the Theogony, but the possibility of comparison and the existence of an external standard, has prevented these graphical corruptions. The text of Pausanias also, who, iv. 30, 4, quotes 417—42(), seems to be correct. This evidence confirms the conclusion (vol. xv. p. 307) that the Hymn to Demeter was lost at a very early period from the z corpus and that M was reproduced for centuries without contact with z. On the other hand the name χρυσηΐς in 422 has been thought the original of the corruptions κρυσίη, xpnonis, κρυσηίς, ete. Hes. Theog. 359. Pallas and Artemis (425) are companions of Persephone in all versions of the legend: in Claudian they even show fight. That they are mentioned last is surely a trivial objection ; the writer could hardly have mixed them with the Oceanidae, and the tail of a procession

THE TEXT OF THE HOMERIC HYMNS: III. 61

is under certain circumstances more honorific than the head. Puntoni, p. 96, retains the line.

429. νάρκισσόν θ᾽ ὃν ἔφυσ᾽ ὥσπερ κρόκον εὐρεῖα χθών. The substitu- tions for ὥσπερ κρόκον are all extremely violent: I need only mention Voss’s ὑπείροχον, commonly received, and Hermann’s αἰπὺν δόλον! I agree entirely with Ilgen’s second thoughts, that ὥσπερ κρόκον means not ‘as also the crocus,’ in which sense Ruhnken seems to have taken it, but ‘as abundantly as the crocus.’ The narcissus which tempted Persephone was no ordinary flower ; it was designed as a δόλος, and had a hundred flowers on one stalk, a marvel to Gods and men (8 sq.). Persephone illustrates this extraordinary abundance by saying ‘it grew like a crocus,’ In our ignorance of where the Νυσήιον πεδίον was, it would be rash to commit oneself to a definite botanical statement, but in Italy, and therefore probably in other Mediterranean countries, the narcissus is abundant only locally in damp places. (Anth. Pal. v. 143 θάλλει δὲ φίλομβρος | νάρκισσος). The crocus on the other hand covers the driest hillsides, and the mention of olives vy. 23 is enough to give this character to the field where Persephone was picking flowers.

438. γηθόσυναι δὲ δέχοντο trap’ ἀλλήλων ἐδίδοντο. Τηθοσύνας and ἔδιδόν τε Ruhnken, but γηθοσυν---ἰη the plural is always adjectival in Homer, and the phrase γηθοσύνας δέχεσθαι καὶ διδόναι is somewhat strange, even with Ruhnken’s parallel v. 8. Perhaps a line has fallen out to this effect, μῦθον τῶν ὅσ᾽ ἑκάστη ἐποίησάν 7 ἔπαθόν τε. Cf. δὸς δὲ δίκην καὶ δέξο Herm. 312, and the similar expression, though in another sense, v. 217 καὶ χαλεπόν περ ἐόντα δεχώμεθα μῦθον ᾿Αχαιοί.

441. ἐκ τοῦ is abrupt, but it has not roused any commentator’s suspicion.

443. ἣν μητέρα κυανόπεπλον. Fontein’s Δημήτερα for ἣν μητέρα is attractive and has commanded universal acceptance. I am not sure however if it is indispensable: ἣν μητέρα obviously in itself has much force (Rhea, Hes. Theog. 454, was mother to both Zeus and Demeter), and as to the construction, in the terse style of this hymn αὐτάς can be supplied to ἀξέμεναι easily out of ταῖς, and a nominative to ἕλοιτο from the general sense,

446 sq. I have noticed this extraordinarily pregnant construction above v. 126. Hermann originally made a lacuna after 448 (in which Biicheler follows him), Voss conjectured νεῖσθαι for νεῦσε ‘quod’ says Baumeister in one of his rare judicial moments ‘fidem prope excedit.’

452 sq. I apply Baumeister’s words to Gemoll’s astonishing pronounce- ment on this passage ‘Die Verse sind verdorben.’ On the contrary every- thing is in perfect order and ἕκηλον which Ruhnken, with the occasional aberration of a great man, condemned is an epithet peculiarly applicable to the idle and resting earth; the same idea of profitless waste is conveyed in Apollonius’ fine lines, iv. 1245 s9.,

οἱ δ᾽ ἀπὸ νηὸς ὄρουσαν, ἄχος δ᾽ ἕλεν εἰσορόωντας ἠέρα καὶ μεγάλας νῶτα χθονὸς ἠέρι ἶσα

τηλοῦ ὑπερτείνοντα διηνεκές" οὐδέ τιν᾽ ἀρδμόν οὐ πότον οὐκ ἀπάνευθε κατηυγάσσαντο βοτήρων αὔλιον, εὐκήλῳ δὲ κατείχετο πάντα γαλήνῃ.

62 THE TEXT OF THE HOMERIC HYMNS: IIL.

454 sq. The picture is consistent though elliptical. The field is suddenly to bloom with long ears [blades with ears], and then (dpa) on the ground the fat furrows are to be heavy with ears [.6. the ears are to be cut and falling on the ground cover the furrows] and other corn is to be being bound up in sheaves. So Franke. There are two times of year, the growing time and the harvest, and the harvest has two moments, reaping and binding into sheaves. These two moments are the same as in Σ 552.

δράγματα δ᾽ ἄλλα pet’ ὄγμον ἐπήτριμα πῖπτον ἔραζε [= πέζῳ] ἄλλα δ᾽ ἀμαλλοδετῆρες ἐν ἐλλεδανοῖσι δέοντο,

a place which and, with Gemoll’s leave, not Hes. Seut. 288 sy. was the hymn-writer’s model. Cf. Pseudophocylid. 165, ὁππότ᾽ ἄρουραι---λήια κειρώ- μεναι καρπῶν βρίθωσιν ἁλωάς. Hes. Opp. 473.

474 sq. Of the variants in Pausanias, 11. 14, 3, δεῖξε appears from Mr. Goodwin’s collation to have been the original in M; the sixteenth century corrector gave εἶπε. The explicit statement in Paus. that Homer mentioned the daughters of Celeus, Diogencia, Pammerope and Saisara, led Ruhnken to suspect a lacuna after 477, which Voss filled with the two lines. My con- tribution would be ΤΓαμμερόπῃ καὶ Σαισάρᾳ καὶ Acoyeveia, and the resem- blance of the last word to Διοκλεῖ would account for the omission. I do not however guarantee the quantity of Σαισάρα. Δρησμοσύνην has been unani- mously preferred to χρησμοσύνην ; the change from X to A is slight (ef. vol. xy. p. 143), but it is not certain that ypnop. is to be rejected, especially as δρησμοσύνη is itself an ἅπαξ λεγόμενον. The Lexx. quote Herod. ix. 33, χρησμοσύνης μετέεσαν, Which might be interpreted to support M’s reading.

479. ἀχέειν is still unscttled, but there is the stronger reason for leaving the word unaltered till itis. The meaning required is surely ‘divulge, reveal.’ Mr. Agar’s κοεῖν is not likely. He revives in 478 Ruhnken’s παρεξέμεν, which cértainly gives a good sense, but if ἀχέειν means ‘give out’ is hardly wanted.

490, ᾿Αλλ᾽ ἄγ᾽ (Ruhnken) seems the simplest correction for ἀλλὰ 6- ; ἄγ᾽ or ἄγε has often produced small confusions, e.g. H 299 & 314 (δέ γ᾽). In 494 Voss’s ὀπάξειν is easy for ὄπαζε; the infinitive as imperative is often corrected in Homeric MSS., e.g. ® 217 ῥέξε ῥέζειν.

THomas W. ALLEN.

J.H.S. VOL.XVII.(1897) PL.I

ENGRAVED GEMS, RECENTLY ADDED TO THE COLLECTION OF THE BRITISH MUSEUM.

Ἐς ae L* 7 δ ᾿ -- Ἃ ay ?

=

our aa pe eeu are Σὐυνᾶς ΗΝ μή νϑ νὸν tig avalide Se, Pees

> nee ake Με" peal 5 ad. By sep)

oe eek. ὙΜΟ allen 208 Dente κυ Ἁ opcg ἡ tthe ery athaila, ie ‘ ΓΝ ει ay 3 ΝΟΣ CY Wi. ῥτπεϑ ἀμ" ais Ἵ ΑΝ ἃ nt iv ποι Rattles Ne ine visto

ἐὼν νυν ι ΔΊ» OM. we δ Fray, i uf at : ον na raphy) / aad abana, ub 5 γ ἢ τ : ible at ce Fer Folie, eee er συσίῳ, Ra

" : wit! ᾿ Δ ἤν ΩΝ "ἡ " ἐλ : Σι Rupa ny ι νιν tin. ἢ ha 5 μη } Ἢ “ὌἌάΠοΕΥΨου υἱὲ

ἊΣ Ά, } WA Ss a" ty ae. etd

bind <i ree Ly. Beye, ἀνα

wi ier ee Ἰὼ ων ΠῚ apps ΨΩ

ἢ Lj rl, ἢ ne αὐ reads

) iin ὟΝ cd. ae μόμυν ἮΝ vee i ovale 4 es ΓΑ lions ; 49 fe ἴω oe ying ἡ ! yw ee We" Us Poa rials ποφυόζυνῃ, tog: a 44 id win ey a were iy, ὧν We

j ΜΔ “μιν UY

4a

τ G3G0A YWITHRORR 290 OSVARONE “MU32UM H2ITING 3HT 40 Μοῦ δου ᾿ φϑις --

ON SOME ANTIQUITIES OF THE MYCENAEAN AGE. 63

ON SOME ANTIQUITIES OF THE MYCENAEAN AGE RECENTLY ACQUIRED BY THE BRITISH MUSEUM.

[Piate 111}

THE present paper is intended as a bricf summary of acquisitions of the British Museum during the past ten years or so, which may be assigned to the Mycenaean period, and which have not as yet been published. The results of the Museum excavations at Curium and Salamis (Cyprus) in 1895-6 are not included, being reserved for publication elsewhere, and the gold treasure acquired in 1892 has been fully described by Mr, Arthur Evans in the Journal, vol. xiii. p. 195 ff. In the case of the vases the terminus post quem is afforded by the publication of Furtwaengler and Loeschcke’s Mykenische Vasen in 1886; for the gems, by the issue of the Museum Catalogue in 1888. ᾿

The most convenient classification for a description of this kind is perhaps a geographical one, but as in some cases the provenience of the objects is unknown or indefinite, I have thought it better to group them under the heading of material, with a geographical sub-classification, so far as such is practicable.

I.—GoLp OBJECTS.

Cyprus. Two fibulae from Moni near Amathus (Figs. 1,2). These are of a type not usually associated with Bronze-Age tombs in Cyprus, but rather with

Ric, 1. ΕἸα. 2)

those of the 7th—6th centuries B.c. (see O.-Richter, Kypros, the Bille, and Homer, pp. 355, 466, and Athen. Mittheil. xii. 1886, p. 19; other examples from the Brit. Mus. excavations at Amathus and Curium, 1894-5). But although no evidence on this point has reached the Museum, I am inclined to think that they may have come from a Bronze-Age tomb, for two reasons: (1) that

64 ON SOME ANTIQUITIES OF THE MYCENAEAN AGE

together with them was acquired a seal of undoubted Mycenaean character, which is described below ; (2) that a fragmentary bronze fibula of the same shape was found in a tomb with Mycenaean vases and gems at Curium in 1895. This tomb no doubt belongs to the latter end of the Mycenaean period, but all the other objects contained in it were purely Mycenacan. There is however much presumptive evidence to be drawn from recent excavations that the Mycenaean civilisation lingered on in Cyprus for many years after it had disappeared from other parts of the Greek world, and there need therefore be no cause for surprise that fibulae of this type should be associated with ‘ Bronze-Age’ tombs. Mr. Arthur Evans (Jowrnal, xiii. p. 228, note 52) records the finding of two gold fibulae at Paphos in 1888, together with a ‘sub-Mycenaean’ pscudamphora. These are now in the Ashmolean Museum (Nos. 1197, 1198), and resemble in shape the examples from Moni. A similar fibula, again, was found by Mr. Paton in his excavations at Assarlik in Caria (Journal, vii. p. 74, fig. 17), and is attributed by him to the transition period between the Mycenaean and Geometrical styles; this would accord very well with the evidence of our examples from Cyprus.

II.— BRONZE.

1. Suria or Suria, an island north of Karpathos, supposed to be the ancient Nisyros. From this site come three bronze implements presented by Mr. W. R. Paton in 1889. The first (Fig. 3) is a knife of a common Bronze-

Fic. 3.

Age type (length 17-2 cm.), the handle of which is lost, but there remain three rivets at the broad end by which it was attached.! The second object (Fig. 4) may be described as a chisel, and is also of a form familiar in the

Fig, 4.

Bronze-Age ;* similar chisels have been found in Kythnos, and there are two specimens in the Prehistoric Saloon of the Brit. Mus. from that island.’

1 For similar knife-blades from Ialysos, see Britain, p. 165 ff. P.-L. Mykea. Vasen, Pl. D., nos. 18 and 19. 3 Proc. Soc. Ant. 2nd Ser.iii. p. 437. * Evans, Ancient Bronze Implements of Gt.

RECENTLY ACQUIRED BY THE BRITISH MUSEUM. 65

The form is that of a narrow bar of metal brought to an edge at one end and left blunt at the other to receive the blows of the hammer or mallet, like the ordinary chisel of the modern stone-mason or mechanic. One such chisel Sir J. Evans gives as found at Plymstock in Devonshire ; others are known from Hungary, Switzerland, and Hissarlik.!. The third is a flat celt, length 19 cm.,? nearly oblong in shape, with slightly convex edge, and narrowing towards the head, where is a diamond-shaped opening through which passed a rivet for fixing it in the handle. This variety of celt is also represented in the Ethnographical Department of the British Museum by one found in Kythnos and illustrated in Prov. Soc, Ant, 2nd Ser. iii. p. 437. Sir J. Evans distinguishes four types of celt: (1) the flat oblong or axe-head-shaped celt, associated with the Bronze-Age in Southern Europe; (2) the flanged celt, 1.6. with a projecting ledge to each of the long sides; (3) the winged celt or palstave, 1.6. with short high flanges and a stop-ridge across the middle to prevent the blade from being driven too far into the head; (4) the socketed celt, with a hole for insertion of a wooden handle. The three latter classes are associated with the Iron-Age in classical countries.

2. Aegina. A bronze knife,? length 26:4 cm., in two pieces. The blade is of unusual size and resembles in form the κοπίς which is frequently repre- sented on later vase-paintings. The handle has probably been covered on either side with a piece of ivory, and was joined to the blade by a series of rivets, some of which still remain. The bronze is in bad condition, being much corroded. It was probably found together with the marble pyxis and the four Mycenaean vases mentioned below. A knife of the same shape, but with the blade and handle in one piece, is given in F.-L. Myken. Vasen, pl. D, ey.

11.-- MARBLE.

Aegina, A pyxis (Fig. 5) with cover on which is a knob; round the cover, a design of heart-shaped leaves; round the side, a band of chevrons. Ht. 13°5 em.; diam. 19cm. <A stone pyxis, but of a different (spheroidal) shape, with elaborate decoration, is given in ’E¢. Apy. 1888, pl. 7, 1, as found in a tomb at Mycenae.

1V.—ENGRAVED Gems.

1. Cyprus. A hemispherical seal of rock-crystal (Plate III. 1, and fig. 6) set in a gold swivel, found at Moni near Amathus and acquired with the two fibulae described above. On the base of the seal is an engraved design composed of an animal with circular body and six legs, which may be intended to represent a sea-urchin, between two trees with stiff branches ;

1 Schliemann, 7'roy and its Remains, p. 332. Catalogue of Bronzes. 2 Length 15°8 cm. An illustration of this 3 An illustration of this object will be given celt will be given in the forthcoming Museum in the forthcoming Catalogue of Bronzes.

HS ΞΞΞ ΘΙ. ΧΥ ΤΙ. ᾽ν

66 ON SOME ANTIQUITIES OF THE MYCENAEAN AGE

above is a tunny-fish. The swivel is formed of a double piece of gold wire passed through a hole at the base of the seal, the two ends of which are twisted together above and end off in a loop for attachment. The hole through the seal is lined with gold, and there is a ring of gold round the base.

Fig. 5:

The Mycenaean character of this seal is clearly shown by the design on the base, both in style and subject. The work 1s careful, but the trees are of a very conventional type. The tunny-fish is of course a common object on Mycenaean gems, and the sea-urchin (if that be the creature intended) is another instance of the fondness of these people for marine objects.

These are the first examples of the Mycenaean period that have come from the site of Moni. It is about six miles from another early site, that of Mari, which appears to be of considerable extent, as Mycenaean remains have appeared in several villages of the neigh- bourhood. Until two years ago no extensive Bronze-Age necropoleis were

RECENTLY ACQUIRED BY THE BRITISH MUSEUM. 67

known on the south coast of the island, but since the unexpected and im- portant discoveries at Curium, to say nothing of those at Salamis, it is impossible to say where Mycenaean remains may turn up in Cyprus.

2. Crete. Our knowledge of Island-gems has been increased by many accessions of late years, but no district has been more generous in this respect than the island of Crete. Mr. Arthur Evans’ discoveries need no further reference here, and it may suffice to say that it was largely the richness of Crete in Island-gems that led Milchhoefer! to regard the island as the centre of Mycenaean civilisation. The British Museum has indeed only acquired one gem during the past ten years from Crete, but it may be fairly said that it is one of the finest existing examples, and may well be compared with the basalt scaraboid from Giiriane 2 or the gem illustrated on p. 81 of Milchhoefer’s work. It is an agate chaleedony (Plate III. fig. 2) of a colour varying from rich brown to pale yellow, and measures 2°3 cm. in diameter. The design consists of a bull walking to the right, guided by a man who stands on the further side of it and holds a cord in both hands which is fastened round the bull’s horns. The figure of the man forms a curious contrast to that of the bull, with its rudely-drawn features and pinched-in waist, which gives a triangular form to the upper part of the body. He appears to wear a loin- cloth wrapped round his waist. ‘The anatomy of the legs is well rendered, and the arms fairly so, but the head is as usual quite incorrectly portrayed, though free from any Oriental conventionality.

This gem naturally recalls to our mind the Vaphio cups, with their life- like bulls and vivid naturalism, the most perfect production of this phase of Mycenaean art. And here I must express myself as fully in accordance with the views put forward by Mr. Arthur Evans (Journal, xiii. p. 220), with reference to the entire absence of Oriental influence, not only on the Vaphio cup but on the Island-gems of all kinds. Whatever theories we may hold of the date of the Mycenaean civilisation, or of the origin of the nation to whom it belongs, it cannot be denied that on a large class of objects no outside influences can be traced, and that whether we term them ‘Greek’ or not, they can only be products of some indigenous fabric, the sphere of which is confined to the Greek islands and the Peloponnese. ΤῸ quote Mr. Evans’ words: ‘In spite of every effort to bring it ready-made from Northern Syria or elsewhere, Mycenaean art has an obstinate way of clinging to the mainland and islands of Greece . . . . The noble representation of the bull-catching on the Vaphio cups, which we are asked to regard as of Syrian manufacture, cannot be separated from the fine animal figures, some representing parallel subjects, on the contemporary lentoid gems. But, unfortunately, amongst the many gems found on the Syrian coast and the neighbouring tracts of Asia Minor, this Mycenaean class is conspicuous by its absence, and the animal representations by their coarseness. On the other hand, the inexhaustible source of the gems which reproduce the Vaphio style in glyptic art is Greece and its islands, in a principal degree Crete and the Peloponnese.’

1 Antange der Kunst, p. 125 ff. 2 Times, 6 Jan., 1896.

68 ON SOME ANTIQUITIES OF THE MYCENAEAN AGE

)

3. Calymnu. A haematite or basalt lenticular gem (Plate III. 3), 2 cm. in diameter, with a very rudely executed Cretan goat standing to the left with its head thrown back over its shoulder. In the field are two branches. This gem was presented by Mr. W. R. Paton in 1889, together with the vase A 296, to be described below, and three terra-cotta whorls of the common Bronze-Age type (frequently found at Hissarlik and in Cyprus), all being from Calymna.

4. Melos. No less than Crete this island has been a fruitful source of Mycenaean gems, although here again we have only one example to discuss among the recent acquisitions. This is a small lenticular sea-green steatite (Plate III. fig. 4), 1°8 cm. in diameter, engraved with a cuttle-fish, above which is a tunny-fish to the right. The workmanship is somewhat inferior, as if the soft nature of the stone had tempted the graver to careless work, whereas the best work is often seen on the harder materials which called forth the best energies of the craftsman, like the example from Crete just described. The lines are lightly cut and ill-defined, while the tentacles of the cuttle-fish are merely indicated by a succession of shallow drilled holes.

5. Hydra. <A carnelian glandular gem (Plate 111. 5), 2 cm. in length, the surface partly striated with white. This gem bears a design of peculiar interest, as belonging to a class discussed by Milchhoefer’ and A. B. Cook,? with reference to animal-worship in the Mycenaean age. In the centre stands a figure to right wearing a horse’s skin over the head and body which is tied in at the waist and hangs down to the feet behind ; a ridge of upright hair extends all down the back. On either side of this figure is a man with whom he appears to be conversing; their hands are raised with animated gestures, and each wears the διάζωμα or loin-cloth characteristic of the period. The work is very careful, and the figures of the men closely re- semble that on the gem from Crete (supra). ,

As to the interpretation of this subject, it may be taken in conjunction with the representations figured by Milchhoefer and Cook (loce. citt.), which are all very similar. Both agree in regarding the figures as evidence of a horse-cult, but differ slightly in their exposition. The cult is that of the Chthonian horse as described by Pausanias in connection with Phigaleia (where one of the other gems with similar subject was actually found). The figure in the horse’s skin denotes a worshipper, a common feature of ancient cults, for which it is only necessary to quote the instance of the ἄρκτοι or maidens who went in bear-skins in the processions in honour of Artemis jrauronia. The exact signification of the scene before us is not easy to indicate; but possibly it is intended like the others to imply the subjugation of the death-deity by a mortal, according to Mr. Cook’s interpretation (speak- ing of a coin of Nicaea): ‘We have here a collateral relic of the Phigaleian rite, in which men dressed in horse’s skins and furnished forth with the emblems of death were overpowered by the celebrant—the purpose of the performance being to secure by mimetic magic immunity from danger.’

' Anfinge der Kunst, p. 54 ff. * Journal, xiv. p. 81 ff.

RECENTLY ACQUIRED BY THE BRITISH MUSEUM. 69

M. Pottier (Cat. des Vases Ant. du Louvre, i. p. 190) casts doubts ou Mr. Cook’s theories on the ground that they imply a too highly-developed system of religion for the period. It is however impossible to believe that these peculiar figures with their ceremonial garb had no significance, religious or otherwise, and were merely products of artistic fancy.

6. Sparta. Rock-crystal lenticular gem (Plate III. 6), 2°2 em. in diameter. On it is engraved a Cretan goat to left, with head twisted back over its left shoulder; in the two vacant spaces in the field are two crosses, which may possibly be characters like those of the Cretan script, but are more probably of a merely decorative character, and due to a horror vacwi.. On the left of the scene is an upright object, probably intended for a tree-stem. Except for the joints of the legs, the eye, and the nostril, the drill has not been employed. The ingenuity with which the animal’s limbs and horns are disposed to fill up the space is worth noting; but the design is of no great merit.

7. Galaxvidi. A small steatite lenticular gem (Plate ITI. 7) of a dull green colour, 1:6 cm. in diameter. The design consists of a stork to r., with addorsed wings, holding an cel by the middle in its beak; above is a dolphin to the right. The wings of the stork are treated in the usual manner, by a row of deep parallel vertical lines.

8. Mycenae. An amethyst in the form of a crouching lion (Fig. 7), the head resting on the fore-paws, the body curled round. Length 1°5 em.

The under side is flat, with a characteristic spiral pattern (Plate III. 8). The style is naturalistic, but the proportions are too thin, and the whole effect is meagre. Amethyst is a somewhat rare material for gems of My- cenaean Age.

9. Calabria. Mycenaean antiquities from Italy though rare are not quite unknown. Furtwaengler and Loeschcke (p. 48) give some examples of pottery from various sites, but no ‘island-gems’ have been hitherto recorded. The example about to be described (Plate HI. 9) is a lenticular sard (?), of a reddish-pink colour, with bands of greasy white, 2°5 cm. in diameter. The design consists of a deer to right with head bent round’ towards a fawn which she is suckling. The fawn approximately fills the space below the deer. In the field above is a dog extended at full length to the left. The drill is largely employed for the legs of the deer and fawn.

1 Journal, xiv. p. 272.

70 ON SOME ANTIQUITIES OF THE MYCENAEAN AGE

GEMS OF UNCERTAIN PROVENIENCE.

10. A small dise of sea-green steatite (Plate III. 10), acquired in Athens, and said to be from Amorgos, 11 cm. in diameter. On it is engraved a man on horseback to right, holding the bridle in his left hand, and_bran- dishing a club or short spear in his right. In front of the horse is a bird with head thrown back. The work is rather inferior, the forms being very thin and angular.

11. A circular steatite gem (Plate III. 11) engraved on both sides, acquired in Athens, but the place where it was found is unknown; diameter 2 cm. On the obverse is a bearded man running to the right and looking back; his attitude is that of the ‘Knielaufschema’ so familiar on archaic vases. The design is somewhat obscure, but he appears to brandish a sword in the right hand over his head, while his other hand is held up in a menacing fashion, In the field on the left is a tree. On the reverse is a Pegasos to right, with recurved wings; below, a bough with long sharp leaves. Only two of his legs are indicated.

12. A haematite lenticular gem (Plate III. 12) of a metallic indigo colour, 2.9. cm. in diameter. This gem has already been illustrated more than once,’ and a detailed description of it is therefore unnecessary, but it is included here to complete the series, and in order to ensure a more satis- factory publication. The design consists of a bull led by two men, one at its head, the other apparently over its back, but probably meant to be on the further side of it. Mr. Murray has pointed out the interest and im- portance of this gem for the points of comparison that it affords with tlie Tiryns fresco-painting. It may also be compared with the gem from Crete described above (no. 2), to which in style it is vastly inferior.

13. <A lenticular burnt sard,? said to be from the Greek islands, 2 cm. in diameter (Plate III. 13. On it is engraved a Hippocamp to right, with one fore-leg extended; the scales of the body are indicated by a double row of indentations, and the markings of the pectoral fin by a series of hatched lines. In the field are two water-plants in the form of thick stems marked all the way up with a row of indentations.

14. Α steatite cone or pyramid (Plate III. 14) with four sides, height 2°5 cm., with a hole pierced through the top, engraved on the base with two rude figures of a lion and a smaller animal running to the left, merely rendered in outline.

15. A pyramid or cone of black basalt (Plate III. 15), with four sides, acquired with No. 10, and also said to be from Amorgos ; the height is 23cm. A hole is pierced through the upper part of the pyramid, indicating that it has been strung on a necklace, On the base is engraved a very rude face, the eyes filled in with white; above it is an unintelligible mark.

‘ Arch. Anzeiger, 1890, p. 69; Perrot and and Journal, xiv. p. 127. Chipiez, vi. p. 851, Fig. 482, no. 12; Murray, 2 An illustration given in Perrot and Chipiez, Handbook of Gk. Archaeology, p. 45. Also vi. p. 851, fig. 482, no. 18. referred to in Class. Review, iv. (1890), p. 282,

RECENTLY ACQUIRED BY THE BRITISH MUSEUM. 71

10. A nearly hemispherical steatite gem (Plate IIL. 16), engraved with a Gorgoneion surrounded by a double ring. The type is that of the archaic Greek Gorgon, with tusks and tongue protruding.

PorTTeRY.

1. Cyprus. In 1888 a collection of thirty-four vases from the necropolis of Ayia Paraskevi near Nicosia was presented to the Museum by Col. Falkland Warren. This site is well known as a hunting-ground for early Cypriote pottery and Bronze-Age antiquities, and has been the scene of several different excavations, for a record of which it may be sufficient to refer to the forthcoming Cyprus Muscuwm Catalogue, Myres and O,-Richter, Oxford 1897, p.1. An account of the most recent exploration of the site is given by Mr. Myres in the current volume of the Jowrnal. The necropolis is wholly of the Bronze-Age period, and the finds closely analogous to those of Alambra, Curium, and Phoenikiais.

The pottery of the Bronze-Age tombs in Cyprus may be roughly divided into two periods, the earlier of which corresponds to the remains of the second city at Hissarlik, and is represented in Cyprus by red or black glaze hand-made vases with geometrical patterns incised with a hard tool and filled in with white. The second period is that in which we find imported Mycenaean vases in conjunction with local pottery; the latter takes the form of bowls, still hand-made, covered with a white slip, on which patterns are painted; or jugs of a thin gritty clay of metallic appearance on which patterns are painted in matt white or laid on in relief in the form of snakes or of cable-patterns. The group of vases above-mentioned belongs almost entirely to the earlier class. A few typical specimens may be described in detail.

(1) A 5 in forthcoming Cat. of Vases (vol.i.). Ht. 45cm. ; diam. 10°2 em. Plain bowl covered with a lustrous red glaze ; it has a small handle in the form of a thick ear pierced with a small hole.

(2) A 10. Ht.17-8cem. Jar with high looped handle and a small ear as A 5. It is of red ware, partly glazed and imperfectly fired.

(3) A 19. Ht. 12:1em. Jug or lekythos, of very rough clay, partially glazed. The vase is covered with patterns of bands, wavy lines, chevrons, and a sort of chain. pattern, all of which have been incised with a blunt. tool while the clay was still soft, and then fired. It is interesting to note that a similar practice obtains to this day in Cyprus, in the decoration employed on the pottery manufactured at Famagusta.

(4) A 30. Ht. 20 em. This vase may perhaps be described as a stamnos; it 15 ἃ form very common in Cypriote pottery of all dates, with more or less bulging body, straight wide neck, and horizontal side-handles. The colour of the clay varies from red to black; it is not lustrous and is imperfectly baked. The decoration consists of raised wavy lines and rings all round the neck and body, and wavy lines on the handles.

(5) A 33. Ht. 16 cm. Funnel-shaped vase, of rather rude execution,

72 ON SOME ANTIQUITIES OF THE MYCENAEAN AGE

and made of a red unglazed clay. In the rim are two holes for suspension, which have been pierced through the clay while soft. The whole of the vase is decorated with rows of short incised lines, forming in their general effect a chequer pattern.

(6) A 34. Ht.45cem. Diam.13cm. A bowl (Fig. 8) broken in fragments and pieced together. Itis covered with a black glaze, on which the patterns are

incised (on the exterior) and filled in with white. On the rim is a small ear. The patterns consist of a wavy line round the edge, three bands of hatched lines, and on the base a cross formed of hatched lines. .

(7) A 35. Ht. 57 cm. Diam. 10°8 cm. A similar bowl, but with different decoration; round the rim are two bands with groups of hatched lines at intervals ; below are triangles of hatched lines placed apex to apex, alternating with broad zigzags vertically placed and filled in with cross lines.

(8) A 36. Ht.64cm. Diam. 12:8 cm. A bowl covered with lustrous red glaze, with a small ear on the rim; round the top are bands of straight ἡ and wavy lines (incised and filled in with white), and below are parallel bands of hatchings.

(9) A 37. Ht. 58cm. Diam. 121 cm. Bowl covered with bright red glaze. Round the upper part are six quadrilateral figures filled with vertical incised lines and divided into four by diagonals; round the bottom are concentric rings with parallel lines drawn diagonally across them so as to form a star of four points.

(10) A 38. Ht. 43°83 em. Large jug with long narrow neck (the ordinary Cypriote shape), covered with a lustrous red glaze. On the body are incised chevrons, vertically placed ; on the neck, chevrons and plain bands of lines.

(11) A 40. Ht. 15:9 em. Similar jug; round the neck, bands; on the body, rows of parallel lies going in different directions, interspersed with concentric circles.

(12) Α 41. Ht. 14cm. Jug as before, with rather flat base; the glaze is of a yellow-brown colour; the patterns consist of lozenges filled with hatched lines, and bands.

(13) Α 42. Ht.14cm. Jug as before; lustrous deep red glaze. On the neck, a vertical row of hatched lines; on the body, chevrons between parallel vertical bands.

(14) A 44. Ht. 146 cm. Jug as before; very lustrous red glaze; patterns on body as the last example.

RECENTLY ACQUIRED BY THE BRITISH MUSEUM. 73

(15) A 47. Ht. 307 οἴη. ‘Stamnos’ (Fig. 9) like A 30 described above, but the neck is larger. This vase is very richly decorated with incised patterns, consisting of lozenges, chevrons, and concentric circles. On one hanglle deep

Fic. 9.

incisions have been made while the clay was soft, ason A 19. At the base of the neck are two small thick ears, and on the rim are upright projections, two low and solid, the other two (one broken away) pierced with three holes for suspension.

(16) A 48. Ht. 331 em. ‘Stamnos’ as A 30 but slimmer, of dull red clay. It is richly decorated with incised lozenge patterns and chevrons.

(17) A 66. Ht.13:°2cm. A jug or lekythos (Fig. 10), shaped as No.10, but belonging to the later class of Bronze-Age pottery, that found with white slip ware and Mycenaean vases. It is of the ‘ base-ring’ type, and of a thin

hard clay with dull black slip. The handle is flat and attached to the neck half-way up; from the base of it on either side springs a serpent in relief. Their heads are flat and lozenge-shaped, the eyes being indicated by large dots, and they have beak-like mouths, slightly open. Between the snakes are two vertical raised lines.

Paphos. During the excavations conducted by the Cyprus Exploration Fund in 1887 by Messrs. Hogarth and James on the site of the temple of Aphrodite at Paphos, a few Mycenaean antiquities were brought to light. Among them were the two gold fibulae mentioned above, now at Oxford. The share of antiquities that fell to the British Museum included a fragment of Mycenaean ware (height 7:7 cm.) of the technique of the later period,

74 ON SOME ANTIQUITIES OF THE MYCENAEAN AGE

decorated with a scale-pattern. These evidences of a Mycenaean settlement at Paphos have been strangely overlooked in the published report of the excavations (Jowrnal, 1x. p. 216 ff).

2. Hgypt. (1) Though not strictly speaking a vase of Mycenaean fabric, the specimen from Saqqara here illustrated (Fig. 11) must be included owing to its close conection therewith.! It is a bowl of a class familiar in

Cyprus, and undoubtedly manufactured in that island. It is hand-made, 11 cm. in height and 17-5 to 20 cm. in diameter, and is numbered C 4 in the forthcoming Vol. i. of the Cataiogue of Vases. The class to which it belongs is known as the ‘White slip ware’? and in Cyprus is always found in conjunction with Mycenaean ware and intombs of the Bronze-Age period, especially at Ayia Paraskevi and Curium. The example under consideration is of a rather flattened and elliptical shape, and the handle is broken off. It is ornamented with various linear patterns in a black colour which tends to fade to brown; the lines are mostly vertical, and combined in threes with triple cross-lines at intervals; alternating with these are pairs of vertical wavy lines with double lines of dots down each side of them,

Several other specimens of this ware have been found outside Cyprus, mostly in very early strata of pottery. Thera ὃ. has yielded one bowl complete, and a fragment has been found on the Acropolis of Athens; Brueckner 4 records a fragment from the more recent excavations at Troy, and mentions another as having been found by Dr. Schliemann.® Specimens have also been found at Tell-el-Amarna and Tell-Hesy.© For further discussion of this class see Myres, Cyprus Mus, Cat. p. 39, Duemmler in Athen. Mittheil. xi. p. 233, and Furtwaengler, Bronzef. aus Olympia, p. 8.

(2) A very remarkable Mycenaean vase from Erment (A 349) was acquired in 1890 through the late Rev. G. Chester. It is mentioned by Mr. A.S. Murray in Amer. Journ. of Arch. vi. p. 437 ff, ἃ propos of an almost identical vase in New York, there published, and is published by Perrot and Chipiez, vi. p. 625, fig. 485, but repeated here for the sake of completeness. The New York vase is a jug with spreading, rather squat, body, and straight

1 For Mycenaean vases from Saqqara, see Myken. Vasen, xii. 80, p. 22.

F.-L. Myken. Vasen, pp. 31, 82. 4 Troja, 1893, p. 101. 2 Myres, Cyprus Mus. Cat. p. 39, q.v. for § Schliemann collection 8125. classification of Cypriote pottery generally. 6 Petrie, 7'ell-cl-Amarna, p. 17 ; Bliss, Mound

8 Fouqué, Santorin, Pl. 42, Fig. 6; F.-L. of Many Cities, pl. 4, 181.

RECENTLY ACQUIRED BY THE BRITISH MUSEUM. 75

up-turned lip, but the Museum one (fig. 12) is of a shape not uncommon in Mycenaean fabrics, but for which there is no exact name. It is of a very flat shape, the sides being curved over to the mouth, and there is practically no

neck ; there are three small ear-handles.!. The colour of the clay is of a deep greenish-yellow, and the varnish is laid on in no great thickness, but otherwise the technique is quite ordinary. The height of the vase is 11 em. The decoration consists of a representation of the Argonaut or paper Nautilus? (not the chambered Nautilus with which we are more familiar) repeated in each space between the handles, while all the remaining surface of the vase is filled in with sca-weed patterns. It will be seen that the design is almost the same as that of the New York vase, and that the treatment of the Argonaut is absolutely identical.

3. Calymna. Together with the gem described above (p. 68) was acquired a vase of great interest and artistic merit, a pseudamphora (A 296; fig. 13) 26 em. in height, with figures in bright red on a deep buff ground, Perrot (vi. p. 929) gives a very similar vase from Pitane in Aeolis, but the design is there somewhat rougher though fuller in detail. On the front of the vase, below the spout, is the body of an octopus mouth downwards, from which extend nine tentacles, of which eight meet in pairs at the back of the vase ; the remaining one falls vertically and ends in a leaf-shaped sucker. The other tentacles end in spirals, and each pair is united by an oval radiated object, the meaning of which is not quite clear. Between the tentacles is a curious sort of webbing, formed of striated bands which are interlaced, and hold the tentacles together for about a third of their total length. In the field of the vase and between the tentacles are various animals: on either side of the spout, two birds; on the left side, in three rows between the tentacles, (1) two horses confronted, (2) a bird, porcupine, and sea-urehin, (3) three birds to the right ; on the right side, (1) two Cretan goats back to back (2) two birds, (3) a porcupine, sea-urchin, bird, and crab. On the shoulder on the reverse are two circles marked with crosses, and on the top of the handle is a large rosette.

1 For the shape see F.-L. Myken. Vasen, Mycenaean and other vases, see Perrot and Pl. 44, no. 82. Chipiez, vi. p. 926 and F,-L. Myken. Vasen, For other representations of the Nautilus on _ p. 80.

76 ON SOME ANTIQUITIES OF THE MYCENAEAN AGE

For a curious interpretation of the subject of the Pitane vase which may be regarded as applying also to the one under consideration, it may be sufficient to refer to an article by M. Houssay in Rev. Arch. 3rd Ser. xxvi. p. 1 ff.; the writer’s views are carried out further in another article in Rev. Arch.

aie ἵ Vin

μ' :

y) A me ἽΝ

ἐν δὲ Pha Se J Ἷ

7 \

xxx, p. 81 ff. MM. Perrot and Reinach have expressed themselves as in accord with the writer’s zoological theories, but the criticisms of M. Pottier in Rev. Arch. xxviii. p. 24 ff. appear to give a sufficient refutation of them without further discussion here.

Another series of Mycenaean vases obtained by Mr. Paton from Calymna has been described in vol. viii. of the Journal, p. 456; Pl. 83, but one vase that was not then published may be regarded as of sufficient interest to include in the present paper (fig. 14). ΤΌ is a kylix of the familiar Ialysos

type (ht. 18:5. em.); the shape is exceptionally graceful and the workmanship distinguished by care and delicacy. It is in excellent preservation except that the foot is restored, and the design, which consists of a cuttle-fish, is painted in red on a deep buff ground. The reverse is free from all decoration It will be seen from the illustration that the cuttle-fish is entirely conven-

RECENTLY ACQUIRED BY THE BRITISH MUSEUM. 77

tionalised and approaches nearly to the phase of a mere decorative pattern, though it is still clearly recognisable what animal is intended.

Aegina. Your vases obtained together with the marble pyxis and bronze knife described above. None of these are of special interest or importance ; they consist of (1) a jug, numbered A 350, ht. 11°5 cm., with straight-rimmed tip and a pattern of four lozenges on the shoulder, end to end, the angles filled in with curved lines; (2) pseudamphora, ht. 25°4 em., with net-work and parallel bands; (3) ‘stamnos’ (ht. 165 cm.) with ivy-wreath on shoulder; (4) ‘stamnos’ (ht. 13°3 cm.) with band of dots and wavy lines round shoulder.

Tiryns. A series of fragments of Mycenaean pottery, thirty-eight in all, obtained by Mr. A. H. Smith, in 1890, on the Acropolis of Tiryns, and presented by him. ‘Together with them was a fragment found on the Acropolis of Athens.

Summary. It will be seen from a comparison of the monuments described above that, although here and there special points of interest and importance may have arisen, there is nothing among them that throws any really new light on the problem of Mycenaean civilisation. In any case it would be beyond the province of a paper of this kind to enter into a discussion of this question ; it is a question the solution of which archaeologists must be content to forego for the present, and indeed nothing is to be gained by hasty generalisations from isolated pieces of evidence, to which there may have been some tendency hitherto.

It might have been hoped that Cyprus, with its close connection with Egypt and Phoenicia, would have furnished satisfactory and decisive evidence, if not for dating Mycenaean objects, at any rate for deducing the origin and ethnological affinities of the race. But though the recent excavations at Salamis and Curium have been fruitful beyond all expectations, it cannot be said that they have availed to settle the question except in so far as they have shown that the Mycenaean civilisation lingered on in Cyprus to a remarkably late date; a fact which will hardly surprise any one conversant with Cypriote archaeology and the circumstances of early Cypriote history. What holds true of Cyprus does not therefore necessarily hold true of other Mycenacan centres; and perhaps we must still look to Egypt to afford us, out of her wonderful treasure-house of things new and old, the key to this most

perplexing problem of classical archaeology. Η. B. WALrERs.

78 NOTES ON ADDITIONS TO THE GREEK COINS

NOTES ON ADDITIONS TO THE GREEK COINS IN THE BRITISH MUSEUM, 1887—1896.

[πα IT.]

DurtnG the last ten years no less than 4,361 coins of various parts of the Greek world have been added to the national collection.1 A certain number of these have been published year by year by Mr. Warwick Wroth in the Nwmismatie Chronicle, under the title ‘Greek Coins acquired by the British Museum.’ A still larger number are described by Mr. Barclay Head, the Keeper of Coins, in the annual Parliamentary Return of the Accounts &e. of the British Museum. As the former publication is not seen by all who are gencrally interested in classical studies, while the latter suffers the fate of most Blue Books, it has been suggested that a short paper on the subject of these additions might be interesting to readers of this journal. I need hardly say that the selection here given is not meant to be representative. Had I attempted to give a full report of the acquisitions during the last ten years, these notes would have reduced themselves to a mere catalogue. It has been necessary therefore to choose out a very few coins from among the more important acquisitions. I have omitted in the first place coins which have be- come well known through publications not primarily confined to numismatics ; in the second, pieces of purely numismatic interest ; and in the third, many rare and interesting coins, such as the coin of Nabis, tyrant of Lacedaemon, or the gold stater struck at Athens in the name of Mithradates, simply because it would be difficult to add anything to the information already collected by my predecessors. Even in the case of the coins selected, I shall be largely going over old ground; but my excuse must be that some readers will be glad to be reminded that they will find fuller and more valuable information on these subjects in the publications to which I have referred.

The date 1887 has been taken as a starting-point, for the reason that that year saw the publication of Head’s Historia Numorwm, to which every archaeologist naturally first refers for information on Greek numismatics. In the arrangement of places I have followed the geographical order of that

work. The sizes of the coins are given in inches and tenths and the weights in

grammes.

1 See the statistics given by W. Wroth in Chronicle, 1897, part ii. (‘Greek Coins acquired his forthcoming article in the MNuwmismatic by the British Museum in 1896’).

J-H.S. VOL. XVII. (1897) PL.II.

Wo

NY

te ; > ἐς ; een, δ δι τὴ" ZNO

ut & Ue) Ϊ

avTorrre

GREEK COINS ADDED TO THE COLLECTION OF THE BRITISH MUSEUM 1887 - [896

et CR) VION 2.4, es

“ἢ 77 yen ΜΝ AD

PM A) OG WE DWP Tea σευ; Poth) gh (ie lar, FA a

a γι ial oF vy, Rate ΡΩΝ

Vode

‘oy, (8 wA db. i ow bg (@ nih been Ὁ

) 301 ΟΥ̓ g8a0e 2HIDS Keane i3.0M HEITIAd ST 36 ϑ98ι ~ Teal

IN THE BRITISH MUSEUM, 1887—1896. 79

ΜΆΑΘΕΡΟΝ, Philip 1].

1. Head of Zeus 1., laureate.

Border of dots.

Rev. PIAIP POY. Naked jockey, holding palm-branch in r., on horse to

r. In field r., bee. MR °95.

Wt. 14°39 ers.

Pl: ilk,

[Wroth, Num. Chr. 1894 p. 2, Pl. 1. 1]. This coin is worthy of illustration for the sake of the unusually fine

treatment of the head of Zeus.

It should be compared with the head on the

Lampsacene Stater described below (No. 10), with which it has much in

common, and which it helps us to date.

The symbol in the field of the

reverse probably indicates the mint of Melitaea in Thessaly.

CHALCIDICE 2

2. Bunch of Grapes.

Border of dots.

Rev. Nude male figure running to r.; he has curved wings on his shoulders and wears boots with large tags; in each hand he holds a wreath.

The hair is long, and is represented by dotted lines.

square within incuse square.

AR 0};

Wt. 16°93 grs.

The whole in dotted

og iy SR

[Head, Num. Chr. 1891 p. 1, Pl. I. 3; Wroth, Num. Chr. 1892 p. 19.] This is one of a small find of archaic coins from the island of Cos. Mr. Head has attributed this piece to Cyrene, in company with three other

staters from the same find.!

1 These other pieces have a similar obverse type, but on the reverse either a head of Heracles or a helmet. The dies are inter- changed, which proves that all these cvins be- long to the same place and period. The hoard con- tained also an archaic tetradrachm of Mende of the usual type, and the presence of this coin at first

‘suggested Chalcidice to Mr. Head as the district to which the coins might belong. He however rejected this suggestion, mainly for the following reasons: (1) ‘there seems to be no city of Chalcidice to which such a type as a bunch of grapes would be appropriate’ ; (2) ‘ the reverses of all these Chalcidic coins consist merely of incuse squares, either quartered or subdivided into triangles. None of them in the earliest period exhibit a device upon the reverse, and at a later period, when reverse-types first make their appearance in Chalcidice, they are never enclosed, as on two out of the three coins now before us, in a dotted square.’ In the Museum is another coin which should be considered in this connection [B.M. Cat. Macedon ete. p. 136

no. 2: Figure running to r.; 1. arm raised, holding wreath; wings attached to waist by belt, from which hangs a short skirt ; on feet, winged talaria ; in front a flower r.: border of dots. Rev. Quadripartite incuse square. Note : ‘This coin was procured by Mr. Borrell at Salonica ?, to the neighbourhood of which town he attributed it.’]. Babelon (Rev. Nwm. 1885 p. 397) has attributed this coin to Cyrene, but its probable provenance is in favour of a Macedonian origin, and Dr. H. Weber possesses a specimen which was certainly found in Thasos. The small flower may be compared with that which figures on the coins of Acanthus [e.g. Berlin Beschreibung der ant. Miinzen Ba. ii Pl. ii 17]. With regard to Mr. Head’s objections to a Chalcidic origin, it may be pointed out that the bunch of grapes might well be the type of some wine city of which we have no other coins ; and that the helmet on one of the series is most suitable to Macedonia. Mr. Wroth (l.c) has suggested some objections to the attribution to Cyrene.

80 NOTES ON ADDITIONS TO THE GREEK COINS

Whatever may be the origin of these coins, the one before us gives a fine example of the early type of winged figure. Mr. Head describes the figure as a wind-god, comparing those winged divinities on cups of Cyrenaic origin which have been explained by Studniczka as Boreades. We may compare also the little winged figure which runs along the outstretched arm of Apollo, usually holding one or two branches, on the coins of Caulonia." The true interpretation of the winged figure seems, however, to be suggested by the words of Studniczka,? which Mr. Head himself quotes: ‘Similar... figures fulfil on Cyrenean bowls with victorious riders the same function as Nike does elsewhere.’ With regard to these figures it has been suggested ὃ that they are personifications of Agon; and that the same or a similar interpretation applies to the figure on our coin seems to me hardly to admit of doubt. The meaning of the wreaths is quite obscure, if the figure is a wind-god; but obvious, if it is a personification of Agon or something similar. If so, it belongs to the earliest representations of the kind, as its date is in the first years of the fifth century. The correctness of this interpretation of course does not depend on the attribution of the coins to

Cyrene.

EUBOEA ?

3. On a raised circular field, a horseman riding to front. His head and that of his horse are turned to r.; his r. hand comes round in front of the horse’s neck. With his 1. he leads a second horse, the head of which is to L., by a rope passing round its neck. The tails of the horses are seen in the space between their legs.

Rev. Incuse square divided diagonally.

AR Ὁ. Wt. 5°56 gr.

[Wroth W. C. 1890, p. 328, Pl. XIX. 20.]

PI ὙΠῸ Ὁ:

4. Horseman riding to front as on previous coin, but without a second horse.

fev. Incuse square, with traces of diagonal division.

AR 65. Wt. 2°64 or. PETES

These two coins belong to an interesting series, the most remarkable of which are tetradrachms representing a whole quadriga seen from the front. 'The series falls into two classes, an earlier and a later, the dividing line being about 500-480 B.c.

(a) Earlier class; lumpy fabric.

Tetradrachm (16° 77grs.) with quadriga. 1892, Pl. XV. 8.

1 Gardner, Types, Pl. 1. 1. For other parallels

Weber collection, Num. Chron.

s.v. Agon. The figure on this coin is clearly

see Babelon, Rev. Num. 1885, p. 295 ff.

2 Kyrene, p. 24

3. Knapp, Nike in der Vascnmalerei, p. 64, ἃ propos of the British Museum Vase, B 1. I owe the reference to Mr. Walters.

* See Reisch in Pauly-Wissowa, Real Encyel.

not to be classed with the ‘ Eros-figures of later times with agonistic attributes, which have been explained as Agon, on Athenian tetradrachms (Beulé, Monn. d’ Athénes, 222), intaglios (Arch. Zeit. VII. Pl. 2. 2) and sarcophagi (O. Miiller, Hdbch. d. Arch, 2. ed. p. 668)’

IN THE BRITISH MUSEUM, 1887—1896. 81

Octobol (5°60-5°56 gers.) with rider leading a second horse. Weber collection, l.c. Pl. XV. 9; Brit. Mus. (above no. 3); Sambon, Cat. d’wne prec. coll. etc., 1889, No. 833, Pl. IV.

Tetrobol (2°79 grs.) with single rider. Weber collection, /.c. Pl. XV. 10.

On the coins of this class the thighs of the riders seem to be held far out, almost at right angles to the body ; they are, however, really meant to be seen in profile, owing to the difficulty of foreshortening.

(0) Later class ; flat fabric.

Tetradrachm (15°68 grs.) Obv. Diademed beardless head to 1., archaic curls on forehead, and bunch of long hair on back of head.

Rev. Quadriga facing, in incuse square. Weber collection, 1.6. Pl. my, ey.

Tetrobol (2°63-2°46 grs.). Single rider, the legs foreshortened. Paris, Bibliotheque Nationale (Rev. Num, 1883, p. 66, No. 6, Pl. 11, 6) and B. M. (above No. 4).

To these must be added a coin described by Mionnet (PI. LI. 7) from the Allier cabinet: Obv. Horse standing facing with a man beside it: Rev. Eagle? flying 1. in shallow incuse square. [Weber in Num. Chron. 1892, p. 191.]

The scheme of arrangement on the tetradrachms is strictly symmetrical. The heads of the two inside horses (ζύγιοι) are turned towards each other, while the trace-horses (σειροφόροι) look outwards. This again is a method of avoiding the difficulty of foreshortening. As to the parallels from other ancient objects, the Selinuntine metope, which naturally occurs to the mind, does not offer a very close analogy. The high relief in which the metope is worked has allowed the artist to represent the heads of the horses in their natural position.

The closest parallels are to be found on the black-figured vases and the so-called Argivo-Corinthian bronze reliefs. For the latter I may refer to the monograph of de Ridder. Of the vases with a facing quadriga it is perhaps

worth while to give the following list, which of course does not pretend to be exhaustive :—

1. Hydria of Chalcidian style. B. M. Cat. B 76.

2. Corinthian? Crater. 4 Β΄ 15.

8. Chalcidian Amphora. Gerhard, A. V. II. Pl. CVI.

4, Corinthian Oenochoe. Pottier, Vases Ant. du Louvre, KE. 648, p. 59, Pk Τῇ:

5. Attic Cylix. Gerhard, A. V. I. Pl. LXII., Berlin Cat. 1799.

6. Attic Amphora.’ B. M. Cat. B 207. :

7. Crater, Attic, or Italian imitation, from Gela. Gardner, Ashmolean Vases, no. 190.

1 De Ectypis quibusdam aéneis quae falso scribed as a spear, but a goad would be more in

vocantur Argivo-Corinthiaca, p. 68. keeping with his function and with other 2 According to Lischcke (Ath. Mitth. xix. p. similar representations, 6.9. no. 1 and the tetra- 516) this also should be classed as Chalcidian. drachm of series (a).

8 The object which the charioteer holds is de- H.S.—VOL. XVII. G

82 NOTES ON ADDITIONS TO THE GREEK COINS

8. Attic Amphora. Gardner, l.c. no. 208. Sa = és no. 210. i ee , + By Andocides. Burlington Fine Arts Club Cat. (1888), no. 108 (Northampton collection), 11. Attic Oenochoe (style of Andocides). Β. M. Cat. B 524. 12. Attic Cotyle. Berlin Cat. 2085. 13. Eretrian Lecythus. Ath. Mitth. xi. p. 94.

The representations of a single horseman riding to the front are less common on vases.1 On a gold coin of Cyrene of later date (period B.c. 431—321) there is a fine representation of a facing quadriga conducted by Nike. On the whole, the evidence of the vases is not very conclusive as regards the distribution of the coins; but they all point to Central Greece, and some of them to Euboea.

The fact that the tetrobol in the French collection was found at, Histiaea is also in favour of a Euboic origin. Olynthus in Chalcidice has been suggested; and Dr. Weber (Num. Chron. 1892, p. 191) adduces in favour of that suggestion the later coins of Olynthus with an eagle.? But the eagle occurs also on the coins of Chalcis in Euboea.* On the whole therefore the evidence seems to me to be in favour of Euboea,

The series of coins described above are further interesting for the evident attempt to express denomination by means of the type.

AEGIUM in Achaia.

5. AIT] EQN Head of Zeus, r., laureate.

Rev. HMI OBE AIN (the last three letters retrograde). The infant Zeus suckled by the goat Amalthea, who stands to r. between two trees, with head raised towards an eagle above with expanded wings.

LE 18: Pl. II. Fig. 16.

The type has been published by Imhoof-Blumer and Gardner,* who say that ‘the proper home of Amalthea was in Crete; but there was probably at Aegium a local legend which in some way connected the name of the city with her, Aiyov with αἴξ But, as Head points out,® there is more direct evidence. Strabo says of Aegium δ:

ἱστοροῦσι δ᾽ ἐνταῦθα τὸν Δία ὑπ᾽ αἰγὸς ἀνατραφῆναι, καθάπερ φησὶ καὶ "Apatos,

al€ ἱερή, τὴν μέν τε λόγος Aci μαζὸν ἐπισχεῖν'

ἐπιλέγει δὲ καὶ ὅτι

᾿Ὡλενέην δέ μιν alya Διὸς καλέουσ᾽ ὑποφῆται"

δηλῶν τὸν τόπον, διότι πλησίον ᾿᾽Ωλένη.

1 Cf. Gerhard, A.V. iv. Ρ]. CCXLVIII. 4 Num. Comm. p. 85, R XIV. 2 Head. H.N. p 184 ; Berlin cat. Pl. V 48. 5 Brit. Mus, Return 1895, p 85. 3 B.M. Cat. Central Greece, Pl. XX. 7 ff. 6 VIII. p 387.

IN THE BRITISH MUSEUM, 1887—1896. 83

| The form HMIOBEAIN occurs on other coins of Aegium, but nowhere else. It may be compared, for its termination, with such a word as FPAMMATIN, found at Patara,! and other similar forms collected in J.H.S. 1895, p. 120.

PHENEUS in Arcadia.

6. Hermes, nude, seated to 1. on basis of two steps; petasus hangs behind his neck, being fastened by a string; 1. hand rests on the basis, r. holds a caduceus which rests on his right thigh. Behind, on the blank space in the field, EV+A in graffito.

Rev. OF. Ram tol.

. A 45. Wt. 96 gr. Pl. II. Fig. 8.

[Wroth, Nwm. Chr. 1896, p. 90 no. 7.]

The characters in graffito, which Mr. Wroth describes as uncertain, and which are barely visible in the photograph, may, I believe, be read Evya. If this reading is correct, it adds interest to the coin, which would seem to have been devoted at the shrine of Hermes. A considerable amount of evidence as to the practice of dedicating coins in this way has been collected by F. Lenormant.? The nearest parallel to the present inscription is ANA@G (ἀνάθημα); which Lenormant has noted on a great number of coins.

Apart from the graffito the coin is remarkable for the style of the obverse, which, particularly in the square and massive treatment of the chest, recalls the style of the school of Polycleitus, Statues of Hermes by this master and by his pupil Naucydes are known ®; but neither seems to have had any connec- tion with Pheneus. Hermes was the chief god of Pheneus, as would be clear from the coinage even if Pausanias did not tell us so (viii. 14. 10).

SYBRITA in Crete.

7. Head of Dionysus r., bearded; wears ivy-wreath with berries; in front, bunch of grapes.

Rev. [<]YBP ITIQN. Head of Hermes r., wearing petasus tilted forward on head, and attached by cord passing round the head; drapery on neck. In front caduceus.

MR 85. Wt. 11:28 grs. Pl. 11. Fig. 10.

[Wroth, Num. Chr. 1890, p. 11, Pl. XIX. 11.]

This didrachm, a work of the first half of the fourth century, gives per- haps the finest representation of the head of Dionysus to be found on coins of this period. The work is of a much softer character than on the coins of

-Thasos with which it has been compared ;* but the expression nevertheless is

1 Herberdey u. Kalinka, Ber. uber zwei Reisen 8 Plin, V.H, xxxiv. 80: Naucydes Mercurio in 5. W. Kleinasien, 1896, p. 27, No. 26. et discobolo et immolante arietem censetur, 2 Rev. Num. 1874-77, p. 325 ff. ; see also P. 4 Gardner, Types Pl. VII. 8. Gardner, Num. Chron. 1873, p. 183 ff. J G

84 NOTES ON ADDITIONS TO THE GREEK COINS

by no means effeminate. The nearest parallel is perhaps to be found in the small electrum coins of Thebes issued about B.c. 395—387.1

The head of Hermes on the reverse is of somewhat inferior execution, and is chiefly remarkable for the way in which the petasus is worn. Usually, on coins, the petasus sits well on the back of the head; but on vases of all periods it is as common as not to find it tilted forward,? although it usually fits better than in the present case.

BITHYNIA.

8. AYTAOMITIANOSKAIS-APZSEBAFEPMA. Head of Domitian r. laureate.

Rev. SEBAZTH OMONOIA. Homonoia standing to 1., holding in r. olive branch, in 1. cornucopiae ; beside her, serpent to 1.

& 14. PL. IL Fig. 19.

9. AYTAOMITIANOSKAIZAPZSEBASTOS. Head of Domitian ΓΤ. laureate,

Δίου. OMONOIA SEBASTH. Homonoia, veiled, standing to 1., 1, resting on sceptre or long torch, in r. uncertain object.

fi 1:1.

These are two of a mysterious series of coins, the attribution of which is uncertain. They are all struck under either Domitian or Trajan. A coin of the former emperor with the monograms TP and ff is published by Imhoof- Blumer,? who sees in them the initials of Prusias ad Hypium. This coin, he adds, is Bithynian in style, and the same is true of the coins above described. The omission of the mint-name on the majority of specimens may point to the coins being meant for circulation throughout the province. In fact, these bronze coins seem to correspond more or less to the silver ‘medallions’ which replaced the cistophori in Asia Minor in imperial times.

For the various representations of Homonoia on Greek coins it is sufficient to refer to the article in Roscher’s Lexikon, which gives a very complete if somewhat undigested mass of material. It is ditticult to see what is the meaning of tlie serpent which occurs on No. 8.

Two other coins of Domitian, belonging to the same series, also have the serpent, but differ from our cvin in that the cornucopiae is replaced by along torch. But these seem to be the only instances of the serpent being given as an attribute of Homonoia. It is not impossible that we have here a blending of Homonoia with Hygieia.®

’ B.M. Cat. Central Greece, p. 77, 78, Pl. five by Augustus to Ianus, Salus, Concordia,

Vie: and Pax ; coin of Galba with the head of Salus * £.g. Berlin Cat. 2538, Gerhard, Auserl. on obv. and Concordia on rev. (Cohen. Jfonn.

Vasenb. Pl. 327. Imp.? I p. 342 No 357) ; ef. Lucan 1V. 190: 3 Gricch. Miinz. No. 813. mixti salus Concordia mundi. Cf. also Sospes + ΤΡ, 2701 ff. Drexler-Stoll. Concordia (Roscher I. p. 920).

° See Peter in Roscher’s Lex. I p. 916: sacri-

IN THE BRITISH MUSEUM, 1887—1896. 85

LAMPSACUS in Mysia.,

10. Head of Zeus 1., laureate ; fulmen showing behind shoulder.

fev, Forepart of winged sea-horse τ, Traces of incuse square.

ΑΙ 75. Wt. 8°45 grs. Pl. II. Fig. 12.

[Wroth, Num. Chr. 1889 p. 257, Pl. XII. 12; B. M. Cat. Mysia, Pk AUX... 6],

The obverse of this coin should, as I have said, be compared with that of the silver tetradrachm of Philip II. of Macedon (no. 1). The present coin shows the finer work of the two, but Philip’s coin is much above the average. The treatment of both is more dignified and less florid than that of the heads on the coins of Alexander of Epirus! and of Metapontum 2 with which the Lampsacene stater has been compared. Both are of essen- tially the same epoch (the middle of the fourth century), but the coin of Lampsacus, judging from its style, is the earlier by a few years. As the tetradrachm of Philip comes between the dates 359 and 336 B.c., we are justified in placing the Lampsacene somewhere near the earlier of these two dates,

11. Head 1. wearing wreath (of myrtle ?); a small wing springs from the neck.

Rev, As on preceding coin.

A’ ‘7 Wt. 8°45 grs. Pl. II. Fig, 14.

[Wroth, Num. Chr. 1894, p. 11, Pl. 1. 11}.

Both Mr. Head? and Mr. Wroth have described this type as a head of Eros. The character of the features, however, has always seemed to me decidedly feminine. A comparison with two other heads on Lampsacene staters will I think show the probability of this view. Both the head of Actaeon (here no. 12) and that of a female satyr* show a strong stylistic resemblance to the present coin; and all three coins are justly attributed by Mr. Wroth to the same hand. On the neck of Actaeon the sterno-mastoid muscle is strongly and definitely marked; in the heads of the female satyr and of the ‘Eros,’ on the other hand, the roundness and the fold under the chin, characteristic of the well-developed female neck, are plainly visible. The strong development of the brow, which might be adduced as an argu- ment in favour of the male sex, is also found in the female satyr’s head. The hair is dressed in practically the same way as on the stater representing Persephone,® but is treated in a more florid manner, a tress being allowed to escape and hang down in front of the ear, as on the satyr’s head, There is no ear-ring, and this again finds a parallel in the head of Persephone.

For all these reasons I venture to differ from the authorities who have already described this type, and to suggest that the head is that of

1 Gardner, Types Pl. V 87. 4 B.M. Cat. Mysia, Pl. xix. 2; Head, Coins a ibid. Pl.V,) £0: of the Ancients III A 18. 3 Brit, Mus. Return, 1894, p. 87. 5 B.M. Cat. Mysia, Pl. xix. 1.

86 NOTES ON ADDITIONS TO THE GREEK COINS

Nike or Iris. The wreath has been described as myrtle, and is certainly not laurel; but it is treated in a somewhat different manner from the myrtle- wreath worn by Artemis at Massalia! or by Hekate at Pherae.? A coin of the latter place offers a parallel in another respect ; just as the wing on the coin of Lampsacus is absurdly inadequate in size, and is merely placed on the neck as it were to identify the type, so Hekate at Pherae is identified by a small hand holding a torch.

Of course the possibility of an androgynous Eros being represented on this coin is not to be excluded.

12. Male head 1., with sprouting stag’s horn. Rev. As on no. 10. Al Ἵ. Wt. 8°45 grs. Pl. Fl Big ΗΠ: [Wroth, Num. Chr. 1893 p. 9, Pl. I. 16]. This head is worth reproducing as a fine example of the rare represent- ations of Actaeon on coins. The head of Actaeon on the electrum coins of Cyzicus is of earlier date but of very much less artistic interest.

MYTILENE in Lesbos.

13. MYTI. Head of Apollo r. laureate, with short hair. Rev, ‘ Mill-sail’ incuse square. EL. "τό. Wt. 15°44 gr. PINIL Fig Ὁ;

[Wroth, Num. Chr. 1890, p. 15, Pl. XIX. 16; B. M. Cat. Troas, &c. PLEX XE ay

The later electrum coinage of Phocaea and Lesbos is represented solely by hectae, with the exception of this unique stater. The convention between Mytilene and Phocaea, according to which coins were to be issued by each city for a year alternately, is well known.’ It is noticeable that in fabric this stater is assimilated to the contemporary Cyzicene currency (although the ‘mill-sail’ form of the incuse square has not been so neatly produced): and the staters of Lesbos were evidently meant to compete with those of Cyzicus, although they were issued in much smaller numbers.

In style the head of Apollo does not closely resemble any other head occurring on Greek coins, being peculiar in its boyish expression.

Ionia.

14. Heraldic arrangement of two lions, heads reverted, standing each on one hind-leg; between them a column, against shaft of which each rests

1 Head, Coins of the Ancients, ΤΥ C 1. Cyzicus, P|. 1 25, 26 ; Babelon, Rev. Num. 1892 2 B.M. Cat. Thessaly, Pl. X 18. Pl Ves: 3 Gardner, Types, Pl. VII 36. 5 See Wroth in B.M. Cat. Troas. etc. p. Ixv.

* BM. Cat. Mysia, Pl. VI 6; Greenwell,

IN THE BRITISH MUSEUM, 1887—1896. 87

his other hind-leg ; on the capital each places one paw, the other being raised above.

Rev. Rude incuse square.

EL. °75. Wt. 14°00 grs. P). II. Fig. 3.

[Wroth, Num. Chr. 1896, p. 99. Pl. VII. 15.]

This coin, the importance of which as illustrating a certain class of architectural types need hardly be emphasized, was obtained through Mr. Lawson of Smyrna. Mr. Wroth says: ‘The coin before us can hardly be later than the middle of the seventh century B.c.; Mr. Head is even inclined to place it as early as 700 B.c.’ It is in any case distinctly earlier than the time of Croesus; but, apart from the opinions of the authorities quoted, I should have placed the coin, in judging purely on grounds of style, late in the second half of the seventh century, if not actually in the sixth century. In general appearance (fabric and colour) the coin bears a great resemblance to the stater attributed to Chios,! although the incuse square is shallower. And this and other similar coins are attributed by Head (/.c.) to about B.c. 500. The ruder, shallower incuse square of our coin, however, may permit us to place it earlier than this low date.

The details of the column are unfortunately not clear. The capital is represented by two dots; the base by a single line projecting from the shaft. Comparison with Phrygian architectural decorations is obvious; for the lions, see Perrot and Chipiez, vol. IV. pp. 111, 157, 180; for the form of the column, ibid. p. 136 (the two dots on the coin are probably meant to represent the Ionic volutes), The heads of the lions are represented as reverted, owing to the inability of the artist to foreshorten them. He doubtless thought of them as looking out of the relief, as once did the heads of the lions of the Lion-gate at Mycenae.

A much later representation of this type occurs on a coin of Tlos of the early part of the fourth century; but there the column is absent, its place being taken by a Lycian symbol, and the lions are seated.

The resemblance of this coin to those attributed to Chios, Clazomenae, Samos and other cities,? seems to point to the Asiatic coast of the Aegean as the district to which we must attribute it.

EPHESUS in Ionia.

15. Head and titles of Antoninus Pius. Rev. ἸΙΕΡΑΠΉΜΗ ΕΦΕΓΙΩΝ. Four-wheeled waggon with arched canopy supported on pillars, drawn by four mules to 1. ; within, driver seated. # 1°35. Pl. II. Fig. 17. The ἀπήμη or ἀπήμη ἱερά is a well-known object on the imperial coins of Ephesus, to which place the spelling ἀπήμη (for ἀπήνη) is, as far as I

1 Head B.M. Cat. Jonia, Pl. 119, III 19. 3 Head, B.M. Cat. Jonia, PI, I 19, ete. 2 B.M. Cat. Lycia, etc. Pl. VIII. 3,

88 NOTES ON ADDITIONS TO THE GREEK COINS

know, peculiar! On this coin the car is drawn by four, instead of, as usual, two animals, The compound word ἱεραπήμη is otherwise unknown, But for the fact that the adjective ἱερά, when used in this connection, is placed after the substantive, it would seem probable that an A has been accidentally omitted,

Of the two forms of ἀπήμη or ἀπήνη which are to be distinguished, that with two wheels is much commoner than the four-wheeled form. Homer is acquainted with the τετράκυκλος ἀπήνη (Jl. xxiv. 324); but in later Greek times the two-wheeled waggon was more generally used both for racing and for travelling purposes. Illustrations may be found on the coins of Rhegium and Messana (of which latter place a didrachm is illustrated, Pl. I. Fig. 7, for the sake of comparison).? In the ἀπήνη of mules, as opposed to the racing- chariot drawn by horses, the driver was seated. As regards the four-wheeled ἀπήνη, Mr. A. S. Murray has called my attention to a terracotta from Alexandria which is apparently an instance in point. (Fig. 1.) The waggon is seen from the side, but with the object of showing the driver the

opening of the tilt is brought round; while the back view shows a window in the side of the tilt. This is presumably the ordinary form; the sacred waggon at Ephesus was open at the sides, the tilt beimg replaced by a canopy supported on columns. It was doubtless used for the purpose of carrying the images of Artemis in procession from the pronaos of the temple to the theatre and back again, according to regulations such as those prescribed in the Salutaris inscription,’

For other illustrations of the car see Head, _ see the article ἀπήνη in Pauly-Wissowa, Real. 3.M. Cat. Jonia, Pl. xiii. 18, xiv. 11. Ene.

For references to the illustrations on vases 3 Gr, Inscriptions in B.M. No. 481, p. 132.

IN THE BRITISH MUSEUM, 1887—1896. 89

PHOCAEA in Ionia.

10, Seal to r. Below, Ὁ.

fev. Two incuse squares, one larger than the other.

ΑΙ ‘85. Wt. 16°516 grs. Pl. II. Fig. 1.

[Wroth, Num. Chir. 1894, p. 14, Pl. 1. 14].

This is one of the only two extant gold staters of Phocaea dating from the time of the thalassocracy of that state (B.c. 602-560). The coin, apart from its historical importance and rarity, is metrologically interesting, as a specimen of the Phocaic standard! The other specimen, which is at Munich, is a little lighter, its weight being given as 16°50 grs. The normal weight of the stater was probably 16°8 gr., 7.c. double the weight of the pure gold stater of Croesus. The Phocaic metal of this period is of very fine quality, con- taining much less silver than, for instance, the coins struck on the Milesian standard. Mr. Head infers that the coins struck on the Phocaic standard were meant to circulate as gold. The reproach which in antiquity was cast upon the Phocaic currency (Hesychius: Pwxaeis: τὸ κάκιστον χρυσίον) can hardly refer to anything but the later electrum.

Of the form © which occurs on this coin there is, so far as 1 know, only one other instance, viz. on an early coin of Phaselis.?

CN1DUS in Caria.

17. Head of Aphrodite 1., wearing ear-ring and necklace. Behind, in field, small prow 1.

Rev. KAAAIQPAN. In exergue, KN!. Head and r. foreleg of lion r.

M9, Wt. 1458 grs. Pl. II. Fig. 15.

[Montagu Sale Catalogue, Ist Portion, lot 599, Pl. VIII].

The head of Aphrodite on the obverse of this coin, which is dated by Head ® between B.c. 390 and 300, seems to me to bear more resemblance than is usual on Cnidian coins to the head of the Praxitelean Aphrodite. Un- fortunately the work is somewhat careless, particularly as regards the nose, The symbol behind the head presumably identifies the goddess with Euploia, which was the standing designation of the goddess at Cnidus, [1 is noticeable that none of the copies of the head on coins shows the fillet which confines the hair on the best replicas in marble. The other coins mostly represent the hair as rolled, or confined by an ampyx, a point in which again they are less faithful than the tetradrachm befure us to the Praxitelean type.

1 See Head in Num. Chr. 1875, p. 281 ff. ; 3 B.M. Cat. Caria, p. 272, no. 28 A. Cat. Ionia, p. xx. ff. 4 But it may be merely a magistrate’s signet. 2 B.M. Cat. Lycia, etc. Pl. XVI 5.

90 NOTES ON ADDITIONS TO THE GREEK COINS

LyYcIA.

18. Female head 1. ; three rows of snail-shell curls on forehead, the hair confined by a fillet which passes three times round it, and taken up behind. In the ear, pendant of the shape sf. Truncation of neck dotted, to repre- sent necklace.

Rev. 3[6]13Δ 33 ΔΤ (Taththivaibi). Lycian tetraskeles syinbol. ‘Serew’ border. The whole in incuse square.

A 8. Wt. 9°79 grs. Pl. 11. Fig. 4.

(Hill, Num. Chr. 1895, p. 26, Pl. I. 21; B. M. Cat. Lycia, p. 19, no. 89, PL Ne 81:

This stater belongs to an important series of coins struck by the Lycian dynasts Taththivaibi and SOMETPIP (Sppntaza), between about 480 and 450 B.c. Since the publication by Wroth? of two staters, one of each of these dynasts, the series of known coins has become fairly large.? The person represented is probably Aphrodite, and the mint to which the coins belong perhaps Antiphellus.? For our present purpose the head is chiefly interesting on account of its fine archaic style and its headdress. The latter is a good illustration of the crobylos, and the little spiral which is used to take up the hair behind, and which is of the kind identified by Studniczka with the tettix,t may be easily made out. The ear-pendant is of a form peculiar to this series, and the curious border on the reverse is, so far as I know, not to be found outside the Lycian series.

A similar head, of a more developed type, occurs on coins of Lycia towards the end of the fifth century.’

SELEUCIA ad Calycadnum in Cilicia.

19. Bust and titles of Macrinus.

Rev. CEAEVKEQ[N] ΤΩΝ TIP OC ΤΩ KAAV.

Naked child seated to front on throne; on either side one of the Corybantes beating shield with sword; behind, the upper part of a third Corybant.

f& 1°15. Pl. II. Fig. 18.

[Wroth, Num. Chr. 1895, p. 103, Pl. V. 16.]

This representation belongs to a class generally connected with the birth of Zeus; but as Sestini and Imhoof-Blumer point out, it is Dionysus about whom the Corybantes are dancing:

1 Num. Chr. 1893, pp. 15, 16. 3 B.M. Cat. Lycia, p. xxxiv. 2 See Babelon, les Perses Achéménides, PI. xiii. 4 Jahrb. 1896, p. 284 ff. 11; B.M. Cat. Lycia, Pl. V. 5 BM. Cat. Lycia, Pl. VI. 6, xliv. 9.

IN THE BRITISH MUSEUM, 1887—1896. 91

ἀσπιδιῶται παιδοκόμοι Κορύβαντες ἀεξομένου Διονύσου, “Ὁ 4 , Μ , , . ee / ot Φρύγα κόλπον ἔχοντες ὀρεσσιπόλῳ παρὰ ‘Pein, νήπιον εἰσέτι Βάκχον ἐκυκλώσαντο Θθοείαις. (Nonnus, Dion. xiii. 135.)

The proof of this lies in the fact that in similar representations on the coins of Magnesia on the Maeander, the cista mystica appears below the throne on which the infant is seated; while other coins of the same place show the infant Dionysus seated on the cista.1_ The whole question of the Corybantic cultus has been discussed by Immisch.2, With regard to the types with which we are concerned, it must be remembered that in Asia Minor, the play- ground of mythology, a hard and fast line cannot always be drawn between Zeus and Dionysus. ‘Sabazius-Dionysus, son of Zeus and Ariadne’ is Ramsay’s description of the infant round whom the Corybantes dance at Laodicea.®

1 Imhoof-Blumer, Griech. Miinzen, Nos. 314 ff. Pl, VIII 31-34. I take this opportunity of pointing out that the type of a coin of Adada in Pisidia, which I have described as an altar (B. M. Cat. Lycia, etc. p. 172, No. 6 Pl. xxx. 4) is probably a cista mystica with the dome-shaped

lid which sometimes occurs, as in Imbhoof- Blumer 1.6. Pl, VIII 30. 33. Huber’s reading BAXXEIA (Cat. Lycia, p. exviii.) is thus quite appropriate.

2 Roscher’s Lexikon, 5. v. Kuretes.

3 Cities and Bishoprics, J, p. 34.

G. Pam,

92 THE NIKA RIOT,

THE NIKA RIOT,

THE great popular insurrection which shook the throne of Justinian in the fifth year of his reign and laid in ashes the imperial quarter of Constan- tinople has been treated again and again by historians, but never in a com- pletely satisfactory way.! Its import has not been quite clearly grasped, owing to an imperfect apprehension of the meaning of the circus factions; the sources have not been systematically correlated; the chronology has not been finally fixed; and the topographical questions have caused much perplexity. It is not therefore superfluous to submit the material to a new investigation. I do not propose to enter upon the subject of the circus factions, as they have been well treated recently by the Russian scholar, Th. Uspenski;? but shall confine myself to problems relating to the authorities, the chronology, and the topography.

T.— AUTHORITIES.

The accounts of several contemporaries, some of whom were eye-witnesses of the event, have come down to us directly; two or three other contemporary notices have been preserved in the works of later writers.

§ 1. The Count Marcellinus was an Illyrian by birth, like Justinian him- self. He had been an official? in the service of Justinian when that Emperor was a Master of Soldiers in the first year of Justin. He retired from public life and embraced the clerical profession, before his patron came to the throne. The first edition of his Chronicle reached the year A.D. 518, but he sub- sequently re-edited it, bringing it down to A.D. 5384, His notice of the insurrection of A.D. 532 is brief’ but highly important, not so much for the

1 Gibbon, Decline and Fall, c. x).

618 sqq.

Lebeau, ed St. Martin, vol. 8, 184-196. J. B. Bury, Later Roman Empire, i. p. 340 W. A. Schmidt, Der Aufstand in Constan- «ηη.

tinopel unter Kaiser Justinian, 1854.

P. Kalligas, Περὶ τῆς στάσεως τοῦ Nika, in Μελέται καὶ λόγοι (1882), p. 329 sqq.

Labarte, Le Palais impérial de Constantinople, etc., p. 18-15.

A. Paspatés, The Great Palace of Constan- tinople, (transl. W. Metcalfe), p. 59 sqq.

L. von Ranke, Weltgeschichte, ii, 2, p. 28 894.

T. Hodgkin; Italy and her Invaders, iii. p.

2 In the Vizantiski Vremennik, i. p. 1 sqq.

3 Cancellarius. Justinian was mag. equituin ct peditum praescntalis, in A.D. 521. See C. J. L. 5, 8120, 3. Cp. Mommsen, Chron. Min., 2, p. 41.

4 Ed. Mommsen, Chron. Min., 2, p. 103. It is strange that M. Kalligas, whose study on the Nika revolt is fuller than any other (except Schmidt’s), should have entirely ignored the notice of Marcellinus,

THE NIKA RIOT. 93

details as for the colouring it gives to the event. The revolt is represented as a conspiracy organized by the nephews of Anastasius for their own personal ends, and not a hint is breathed of any other causes, This account is at variance with our other sources, in which the part played by Hypatius and his brothers is represented as merely an after-thought and quite unconnected with the origin of the tumult. When we remember the close personal connexion of Marcellinus with Justinian, we are justified in regarding the notice in his Chronicle as a quasi-official account. I do not mean to say that it was directly ‘inspired’; I mean only that Marcellinus, in sympathy with the existing régime, gave utterance to that interpretation of the revolt which Justinian and the court wished or feigned to believe,—namely, that it was not a genuine expression of popular feeling, but merely due to the machinations of Hypatius and his friends.

At the same time Marcellinus lets out a very significant fact. <A large number of the higher classes took part in the insurrection. This confirms the statements of other sources.”

§ 2. The narrative of Procopius* presents a marked contrast to that of Marcellinus ; it is full and circumstantial, it sets forth the causes of the revolt, and, though nothing disrespectful is said, we are permitted to read between the lines that the writer's sympathy is not with Justinian, but with the nephews of Anastasius. It is abundantly clear that in the Public History Pro- copius adopted the plan of placing his own hostile criticisms on the government in the mouths of the actors who appear on the stage of his story. He might thus defy censorship. If he were called to account for enumerating the evils which Justinian’s administration brought upon Italy, he had only to reply: ‘But I was only recording the lies uttered by the barbarian Totila.’ We are therefore justified in seeing a reflexion of the personal sympathies of Procopius in the last words of Hypatius: ‘We are innocent. We could not resist the people. It was from no illwill to the Emperor that we entered the Hippodrome.’ This is a blank denial of the view reflected in the notice of Marcellinus. The nephews of Anastasius are represented as innocent victims; the sentence of Justinian as unjust. And there is no doubt that it was the view of Procopius himself.

IT have said that the narrative of Procopius is circumstantial, but here it contrasts with the other circumstantial narrative which has been preserved, that of John Malalas. The historian leaves out no point essential for the com- prehension of the general course of the revolt and its political significance ; but he omits a great many details where the Chronicler is circumstantial, and

1 Tam plerisque nobilium coniuratis, Mommsen, Hermes, 6, p. 377, and in Theoph- 2 In the account of Procopius, we find the anes, p. 185, 1. 30 ed. de Boor). Cp. Panchenko, senator Origen among the rebels; and the 0 tainoi istorii Prokopiia, in Viz. Vrem., iil. property of the senators who supported Hypa- Ρ. 302. tius is confiscated. Malalas mentioned the 3 Bell. Pers., i. c. 24; vol. 1, p. 119 δα. ed. banishment of ‘eighteen illustres and senators’ Bonn. ἡ (omitted in the abridgement of the Baroccianus, * B. G., iii, 21, p. 340 ed. Comparetti. but preserved in the Escurial fragment ed. by

94 THE NIKA RIOT.

on the other hand he is circumstantial where the Chronicler is meagre. Pro- copius summarizes the tumults and conflagrations of the first days of the rebellion, in a few lines; he omits altogether the scene in the Hippodrome on Jan. 13; and he begins his circumstantial story on the evening of Saturday, Jan. 17. The great interest in his relation is that he describes what happened in the palace. Malalas only knows what went on in the city and the Hippodrome, but the secretary of Belisarius knew the doings and the deliberations of the court, nor can there be much doubt that he was in the palace with Belisarius during the last days of the insurrection. We may, I think, safely contrast the story of Procopius with that of Malalas by saying that: Procopius followed the revolt from the Palace, while in the account of Malalas the point of view is that of α spectator in the town.

§ 3. John the Lydian gives a brief account of the revolt in his treatise De Magistratibus (written after A.D. 551)... He does not relate its course, but enumerates some of the buildings which were burned down, and states that nigh fifty thousand of the populace were killed. The main interest of his notice lies in the fact that he ascribes it mainly to the rapacity and malad- ministration of John of Cappadocia. This is significant, when we remember that the writer, although disappointed, was loyal to Justinian and had still hopes from the court which would have prevented him from saying anything offensive.2 We may infer that, after the disgrace of John of Cappadocia in A.D, 541, Justinian was willing to let fall on that minister's administration part of the blame which, when Marcellinus wrote in A.D. 534, was imputed entirely to Hypatius and his adherents.

§ 4. The notice of Victor Tonnennensis, though very briefs supplies two points which we find in no other source. (a) Hypatius and Pompeius were slain at night. (b) The body of Hypatius was thrown in Rhewma; the Greek sources say simply into the sea. Victor was contemporary, though he wrote more than thirty years later, and might have heard from eye-witnesses. But it is probable that he took the notice straight from Italian Consularia.

§ 5. A summary account from the Ecclesiastical History of Theodore Lector (who carried his work down into the early part of Justinian’s reign) is

ὁ δὲ δῆμος οὐδὲν Bpadivas, καθ᾽ ἑαυτοῦ δὲ κινηθείς, αὐτὸς μὲν ἀπώλλυτο, ἡ δὲ πόλις πυρὶ πᾶσα διεφ-

1 Β, iii., c. 26, ν. 265-6 ed. Bonn. The author has been describing the decay of the

office of Praetorian Prefect, and, in connexion therewith, the misdeeds of John of Cappadocia. Lydus also notices the Nika in his treatise De Ostentis, § 8, p. 14 ed. Wachsmuth. He is enumerating portents of sedition and civil war ; among these he mentionsa kite hovering in the air over a crowded theatre, ὁποῖον ἱπποδρομίας ἐπιτελουμένης εἴδομεν ἐπὶ τῆς ἄρτι διελθούσης ἐνάτης ἐπινεμήσεως (A.D. 530-1) ἧς ἀγομένης ἰκτῖνος τὸ βέλος, τὴν λεγομένην σαγίτταν, τῷ ῥέμφει φέρων, ὅλον ἐπικυκλώσας τὸν δῆμον ἐπὶ τοῦ ὀβελοῦ ταύτην διωλύγιον συρίττων ἀπέθετο.

θείρετο, ws καὶ αὐτὴν τὴν βασιλείαν, εἰ μὴ θεὸς ἀντέπραττεν, οὐ πόρρω κινδύνων ἐλθεῖν.

2 For ἃ description of the career of John the Lydian I may refer to my Later Roman Empire, ii. p. 183-4.

8 Ed. Mommsen, Chron. Min., ii. p. 198, ad ann. 530. I suspect that the false date was due to the circumstance that 530 was Lampadio et Oreste conss. and 5382 py. c. Lampadi et Orestis ann. iii.

4 Cp. Mommsen, op. cit. p. 180.

THE NIKA RIOT. 95

preserved in Cramer, Anecd. Par. ii. 112, and with slight variations in Theo- phanes (see below § 9); and in a fuller form in Cedrénus (below § 10). It adds nothing to what we know from other sources.

§ 6. An interesting notice, though inaccurate and enveloped in verbiage, has been preserved in the Continuation of Zacharias of Mytilene.1 The cause of the riot is here imputed to the exactions of John of Cappadocia, who ‘favoured one of the factions.’* There were constant complaints ὃ against the Praefect and the Emperor; at length, the factions united for some days. The revolt is then briefly described with certain variations from, and additions to, the other accounts.

§ 7. Of the Chronicle of John Malalas (Rhetor) of Antioch, the first seventeen Books (with a few paragraphs which were then part of B. 17 but were afterwards prefixed to B. 18) appeared between A.D, 528 and 540. A second edition appeared after A.D. 565, bringing the work down to Justinian’s death (B. 18). The question is debated whether this revision and continuation was due to the author of the original work, John Malalas himself, or to some one else.* I believe that the second view is the true one; but in either case the chronicle of Justinian’s reign is due to a contemporary, and that is enough for our present purpose. Of the revised chronicle (published probably c. A.D. 566) the text which has been handed down in the unique Oxford MS.° is only an abbreviation. But we have material for approximating to the original shape in the works of other compilers who copied slavishly from the complete chronicle. So faras the eighteenth book is concerned, this subsidiary material consists of (1) the Paschal Chronicle, (2) Theophanes, (3) the Constantinian excerpts published by Mommsen,’ (4) excerpts published by Cramer’ from a ninth-century chronicler.’

This material does not, in the present case, enable us to restore with certainty the narrative of Malalas, though it enables us to see that this narrative was considerably longer in the original text than in the Oxford epitome. The difficulty is that the Paschal Chronicler and Theophanes used another source as well as Malalas ;° and it is impossible in certain passages to determine which of the two sources was responsible. It is however indispensable to make an attempt to distinguish these two sources,

1B. 9,14. Mr. E. W. Brooks most kindly 7 Anecd, Paris, vol. 2, 320.

supplied me with an epitome of the passage. It will be included in the forthcoming transla- tion of Zacharias and his Continuator, by Messrs. Hamilton and Brooks.

2 μέρη.

3 ἐκβοήσεις.

4 It is not necessary to go further into the ‘Malalasfrage’ here. See Krumbacher’s Gesch. der byz. Litteratur, § 140, and my review in the Classical Review, 1897, May.

5 Baroccianus 182.

8 Hermes, Bd. 6, p. 877 (fragment on Nika revolt).

8 I omit other chronicles which have to be taken into account in dealing generally with Malalas, but which do not help us for our present purpose (¢.g. John of Nikiu, George Monachus),.

® TI pointed out (Classical Review, loc. cit.) that Theophanes used three sources ; the proof being that he has three introductions, the third of which is the introduction of Malalas. The Paschal Chronicler also used the other unknown source. Gleye proved (against Patzig) that he used the Malalas chronicle.

96 THE NIKA RIOT.

To begin with; we must compare our text of Malalas with the Constantinian excerpt. This fragment is not an excerpt in the proper sense; it 1s not a verbal extract or series of extracts, but a brief summary in which the original phrases are not always retained.

(a) The fragment does not give the causes of the insurrection, but rushes in medias ves: ἀντῆραν αὐτῷ (lovativiave) ὁ δῆμος τῶν λεγομένων πρασινο- βενέτων καὶ πολλὴν ἀταξίαν καὶ ἅλωσιν ἐν τῇ Κωνσταντινουπόλει. These words are of course mercly a general introductory summary, and the only phrase which we can claim with security for Malalas is the compound πρασινοβένετοι, which the excerptor (as τῶν λεγομένων shows) did not make for himself2 The next clause notices the burning of a number of buildings :

καὶ ἔκαυσαν οἱ αὐτοὶ δημόται τοὺς ὑποτεταγμένους τόπους ἀπὸ τοῦ παλλατίου ἕως τοῦ φόρου καὶ τῆς ἄρκας δεξιὰ καὶ ἀριστερὰ μετὰ τῶν παρακειμένων πασῶν οἰκιῶν καὶ τὸ πραιτώριον τοῦ ἐπάρχου τῆς πόλεως καὶ τὸ λεγόμενον ὀκτάγωνον.

Now in our Malalas text conflagrations are mentioned at two points of the narrative: (1) on the night of the 13th Jan., and (2) after the conflict with Belisarius and his force of Goths. In the first case, the praetorium was fired and the following places were burned down :*

Ν , \ id »“ Lal / “ aA Lal \ e , TO πραιτώριον καὶ ἡ χαλκῆ TOU παλατίου ἕως τῶν σχολῶν καὶ ἢ μεγάλη ἐκκλησία καὶ ὁ δημόσιος ἔμβολος.

In the second case, it is merely stated generally: καὶ ἐν ἄλλοις τόποις ἔβαλον πῦρ At first sight the mention of the praetorium might seem to show that the excerptor had in view the first. conflagration, and ὁ δημόσιος ἔμβολος might be supposed to mean the porticoes along the Mesé, between the Augusteum and the Forum of Constantine. But (1) it is hardly conceivable that the excerptor would have omitted to enumerate the Great Church ; (2) ὁ δημόσιος ἔμβολος is, in the context, more naturally understood of the portico of the Augusteum than of the porticoes on either side of the Mesé; (3) the excerptor says nothing of the events which, in our Malalas-text, occurred between the two conflagrations, but goes on directly to the events after the second conflagration ; (4) it will be shown below that the praetorium was fired a second time.

I think we may therefore provisionally conclude that the words in our Malalas-text καὶ ἐν ἄλλοις τόποις ἔβαλον πῦρ are the epitomator’s summary of an enumeration of buildings, which is, wholly or partly, preserved in the Constantinian fragment ; the phrase ἐν ἄλλοις τόποις being suggested by the first words of the original τοὺς ὑποτεταγμένους τόπους (where

1 The date is given falsely eis τὸ δεύτερον ἔτος ᾿Ιουστινιανοῦ.

2. So George Monachus, i, p. 528, ed. Muralt. I have not devoted a special section to the notice of this chronicle (cp. the corre-

sponding notices in Leo Grammaticus and Theodosius Melitenus) ; it is abbreviated from OF ,—it contains nothing that is not in CF.

3 p. 474, 17, ed. Bonn,

4 p. 475, 10.

THE NIKA RIOT. 97

ὑποτεταγμένους suggests the position of the buildings between the first and second hill). (8) The fragment proceeds (without any hint of a time-interval) :

καὶ πάλιν μὲν ταῦτα κράζειν (sic) ὁ δῆμος, Ὕπάτιε αὔγουστε τούμνικας [id est, τοῦ Biyxas].

Our Malalas relates that on the 18th Jan. the Emperor appeared in the _ Hippodrome with the Gospels ; the people gathered together and the Emperor προσεφώνησεν αὐτοῖς μεθ᾽ ὅρκων μανδᾶτα (the Paschal Chronicle explains this clause, which in itself is not very clear): then

lal Ν καὶ πολλοὶ μὲν τοῦ δήμου ἔκραξον αὐτὸν βασιλέα ἕτεροι δὲ ἐστασίαζον κράζοντες Ὕπατιον.

We can infer with certainty that Ὕπάτιον of the epitomator is an abridgment of Ὕπάτιε αὔγουστε, τοῦ Biyxas. But we may infer more than this. The words πάλιν μὲν ταῦτα imply the repetition of a cry already mentioned, and have no meaning in the extract. It follows that they occurred in the original, where their meaning must have depended on a preceding account of cries uttered by the people. Hence we conclude that this passage was abridged by the epitomator of the Oxford text. Jn the original Malalas the words Ὑπάτιε αὔγουστε τοῦ Biyxas must have occurred twice.

This conclusion is confirmed by the Paschal Chronicler who supplies us with the material for restoring the passage which the Oxford epitomator has omitted. There we read, after the scene in the Hippodrome, that Justinian retired into the palace and dismissed the senators, that the people met Hypatius and Pompeius and cried Ὕπάτιε αὔγουστε τοῦ Biyxas. These incidents are omitted in our Malalas-text ; but this second cry is preserved in our Constantinian fragment. Thus we are justified in inferring that this passage in the Paschal Chronicle (καὶ éacev......rod βέγκας, p. 624, 1 sqq.) was derived from Malalas; to whom we may restore it with some such slight change as καὶ πάλιν ταὐτὰ; ἔκραζεν (for καὶ ἔκραξαν).

(y) The next section, describing the elevation of Hypatius, is much shortened in the Constantinian fragment=CF, but one or two points are preserved there which are lost in our Oxford Malalas text = OM.

CF. ΄ ΟΜ, καὶ ἀγαγόντες Ὑπάτιοντὸν στρατηλάτην καὶ λαβόντες οἱ δῆμοι τὸν αὐτὸν Ὕπάτιον ἀπή- ἐκ τοῦ οἴκου αὐτοῦ. γαγον αὐτὸν ἐν τῷ λεγομένῳ Φόρῳ Κωνσταντίνου.

Thus in the original Malalas, after the people met and saluted Hypatius in the street, he retired to his house, and the people fetched him thence to the Forum of Constantine. Moreover he was described by his title στρατηλάτης (magister militum). Another detail preserved in the CF, is his elevation ona shield, when he had been invested with the royal dress: καὶ ἀναγαγόντες αὐτὸν ἐν τῷ σκουταρίῳ.

1 So possibly for ταῦτα ; or perhaps simply πάλιν ἔκραζεν. ‘ Quellenforschung ᾿ has its limits. H.S.—VOL. XVII. H

98 THE NIKA RIOT.

(8). The next clause of CF. records the gathering of the people in the Hippodrome and is entirely omitted in ΟΜ. : πληρωθέντος δὲ καὶ ὅλου τοῦ ἱππικοῦ ἐκ τοῦ δήμου ὡς θελόντων θεωρῆσαι βασιλέα στεφόμενον. The sentence is incomplete; it probably fits in after the words ἐκ τοῦ παλατίου, Mal. p. 475, 22.

(c). The next sentence describing the orders of Justinian is also omitted in OM., and should come before the first words of p. 476, 1: ἔτε δὲ γέμοντος τοῦ αὐτοῦ ἱππικοῦ ἐκ τοῦ ὄχλου ἐκέλευσεν ὁ βασιλεὺς ἀπολυθῆναι τὸν στρατιώτην αὐτοῦ μετὰ καὶ φανερῶν ἐξάρχων. CF. also notes the places where Belisarius and Mundus respectively entered the Hippodrome: καὶ εἰσῆλθε Μοῦνδος μὲν ἀπὸ τοῦ καθίσματος ἐπάνω τῶν θυρῶν τοὺς ὄντας δήμους ἐν τῷ ἱππικῷ καὶ Βελισάριος ὑποκάτωθεν τοῦ καθίσματος.

(ζ). Passing over all the details of the scene in the Hippodrome CF. states the number of the slain less precisely than OM., but with a phrase which probably found a place in the original Malalas:

CF. OM.

καὶ ἀπέκτειναν ἐν φόνῳ waxalpas meph οἱ δὲ τὰς λ΄ χιλιάδας.

ἐν τῷ ἱπποδρομίῳ σφαγέντες ἦσαν

χιλιάδες τριάκοντα πέντε μικρῷ πλέον ἢ ἔλασσον. (n). CF. gives the reasons for the execution of Hypatius and Pompeius, and records the banishment of eighteen other senators (points omitted

by OM.):

καὶ συνελάβετο' ᾿Ιουστινιανὸς καὶ τὸν ὙὝπάτιον καὶ ἸΤομπήϊον καὶ ἀπέκτεινεν αὐτοὺς, τὸν μὲν ἕνα ὡς φορέσαντα βασιλικὴν φορεσίαν καὶ ἀντάραντα, τὸν δὲ ἕτερον ὡς συνευρεθέντα μετ᾽ αὐτοῦ, καὶ τή ἰλλουστρίους καὶ συγκλητικοὺς δημεύσας ἐξώρισεν διὰ τὸ κατ᾽ αὐτοὺς τῇ τυραννίδι Ὑπατίου προσθέσθαι.

(θ). CF. concludes with remarks which are left out in OM:

A 2 / ’ / b] a / \ / wv fel , Kal ἐγένετο εἰρήνη ἐν TH πόλει. Kal προεβάλετο ἔπαρχον τῆς πόλεως Τρύφωνα καὶ πολλοὺς τῶν δημοσιῶν ἐκόλασε. καὶ ἱππικὸν οὐκ ἦν ἐπὶ πολὺν χρόνον.

§ 8. We are now in a position to consider how far the Paschal Chronicler used Malalas. The short summary of the ἄκτα διὰ Καλαπόδιον---ἢ 6 recriminations between the Emperor and the Greens—which are reproduced in full by Theophanes, was derived from another source.?, The Chronicler seems then to pass abruptly to the middle of the narrative of Malalas. He omits the incident of the two criminals who fell from the gallows; and does not even state that the Blues and Greens reconciled their differences. Hence his story, taken by itself, is unintelligible; and it seems possible that our text is imperfect. It begins with the supersession of John, Tribonian, and Eudae-

1 συνελάβετο was probably not in the original, in this clause ; the excerptor took the word from the description of Belisarius seizing Hypatius and Pompeius (χειρὶ συνελάβετο, 476, 11).

2 Cp. above § 7, par 2, p. 95.

3 There is, at the transition, a curious inser- tion in P, not found in the other MSS. It does fit into the context ; some words must have fallen out :—

καὶ ἀπέστειλεν & βασιλεὺς ἰδεῖν τί κράζουσιν [P: ὡς ἔτυχεν: ἀλλ᾽ ὅτε πολλὴ γένηται ἀνάγκη

THE ‘NIKA RIOT. 99

mon; and it is clear from the following comparison that the Pasch. Chron. derived this passage from the work of Malalas.

CHRON. PASCH. (Ὁ. 620).

καὶ ἀπέστειλεν ὁ βασιλεὺς ἰδεῖν τί κράζονσιν. καὶ ἐξῆλθεν ἀπὸ τοῦ παλα- τίου & πατρίκιος Βασιλείδης ὁ ποιῶν τὸν τόπον τοῦ μαγίστρου Ἑρμογένους ἐν Κωνσταντινουπόλει, καὶ Κωνσταντίολος. καὶ στἠσαντεςτὰεἰσελαύνονταπλήθη ἔξω τοῦ παλατίου κατασιγήσαντες αὐτὰ προσεφώνησαν αὐτοῖς, ζητοῦντες γενέσθαι στασιάζετε; καὶ ἔκραξαν κατὰ τοῦ ἐπάρχου τῶν πραιτωρίων ᾿Ιωάν- νου τοῦ Καππάδοκος καὶ ‘Poudivou (sic) τοῦ κυαίσ- τορος καὶ τοῦ ἐπάρχου τῆς πόλεως Evdaluovos. καὶ ταῦτα «ἀκηκοότες ἀνήγαγον τῷ βασιλεῖ. καὶ εὐθέως διεδέξατο τὸν ἔπαρχον τῶν πραιτωρίων Ἰωάννην καὶ ἐποίησεν ἂντ᾽ αὐτοῦ κ.τ.λ. διεδέξατο δὲ καὶ Ῥουφῖνον τὸν kvalatopa καὶ κι τ. λ. καὶ τὸν ἔπαρχον δὲ τῆς πόλεως Εὐδαίμονα διεδέξατο καὶ κ. τ. λ.

λέγοντες, τί

[The words preserved in either printed in spaced type.]

ΟΜ. (p. 475).

καὶ ἐξελθόντες of περὶ Μοῦνδον καὶ Κων- σταντίολον καὶ Βασιλίδην μετὰ βοηθείας κατὰ κέλευσιν τοῦ Βασιλέως, }

βουλόμενοι κατασιγῆσαι τὰ στασιάζοντα πλήθη:

κατέκραζε γὰρ τὸ πλῆθος ᾿Ιωάννον τοῦ ἐπίκλην Καππάδοκος καὶ Τριβουνιανοῦ τοῦ κοιαί- στωρος καὶ τοῦ ἐπάρχου τῆς πόλεως Evdaiuovos: καὶ ταῦτα map’ αὐτῶν ἀκηκοότες οἱ ἐκπεμφ- θέντες συγκλητικοὶ ἀνήγαγον τῷ βασιλεῖ' καὶ εὐθέως διεδέχθησαν τῆς ἀρχῆς" ὅ τε ᾿Ιωάννης καὶ Τριβουνιανὸς

καὶ Εὐδαίμων.

source and omitted in the other are

We at once remark that the epitomator has here gone to work very

discreetly.

He has omitted those clauses, whose omission can best be spared,

and the only positive facts he has left out are the names of the ministers who were appointed to replace John, Tribonian, and Eudaemon.? The Paschal Chronicler was less discreet. While he “unnecessarily repeats the names of the offices of the deposed ministers (τὸν ἔ ἔπαρχοι τῶν TpaT. K.Tr.), he omits the important words μετὰ βοηθείας.

The next omission of the epitomator is less fortunate. Having mentioned the Emperor's concession in deposing the obnoxious ministers, he goes on to state that Belisarius issued forth with Gothic soldiers and fought with the mob. It was obviously necessary to say that the concession had failed to appease the people. The Paschal Chronicle preserves the requisite words :

CHRON. PASCH. OM.

ὁ δὲ δῆμος ἐπέμενεν ἔξω τοῦ παλα-

τίου εἰσελαύνων: καὶ τούτου γνωσ-

θέντος ἐξῆλθεν ὁ πατρίκιος Βηλισάρις ὃ

στρατηλάτη 5 μετὰ πλήθους Γότθων καὶ ἔκοψε πολλοὺς ἄχρις ἑσπέρας.

καὶ ἐξελθόντος BeAtoaplov μετὰ πλήθους Γοτ- θικοῦ καὶ συμβολῆς γενομένης πολλοὶ ἐκ τῶν δημοτῶν κατεσφάγησαν.

the original, but may be due to the epitomator who had to make up for the omission of the preceding sentence.

2 The Constantinian Fragment mentions the name of the new Praefect of the City, Tryphon ; but at the end of the episode.

τότε ποιεῖς ἃ ἐβουλεύσω. καὶ εἶπεν αὐτοῖς ὁ

βασιλεὺς ἐξέλθατε οὖν καὶ μάθετε τίνος χάριν στασιάζουσιν]. Kol ἐξῆλθεν κ.τ.λ.

Somebody seems to be remonstrating with Justinian in the Palace on his indecision, need not have been in

1 κατὰ κέλ. τ. Bac. H

100 THE NIKA RIOT.

I suspect that the first clause was dropped by the epitomator, because almost the same words had occurred before, (p. 474, 19) καὶ ἐπέμεινεν ὁ δῆμος εἰσελαύνων ἀτάκτως.

At this point the Paschal Chronicler leaves Malalas and copies his other source. From p. 621-623, we can find no trace of Malalas. The motive for thus changing sources doubtless lay in the circumstance that Malalas did not describe in detail the events of Thursday evening, Friday, and Saturday. In OM. we have nothing whatever corresponding to this period of time beyond the chronologically vague statement: θυμωθὲν δὲ τὸ πλῆθος Kal ἐν ἄλλοις τόποις ἔβαλον πῦρ καί τινας ἀτάκτως ἐφόνευον. As we have seen, this is supplemented by an enumeration of buildings which were burned, in CF. We may infer, I think, that the original Malalas did not contain much more than this enumeration.

On Sunday, Jan 18, the Paschal Chronicler returns to Malalas, and preserves more fully than OM. the scenes of Justinian’s appearance in the Hippodrome and the elevation of Hypatius. This is proved by the numerous verbal coincidences and especially by the argument which I brought forward above, in connexion with CF. Nor can there be much doubt that the incident of the sending of Ephraim to the Palace was related in the original Malalas. The epitomator merely gives the result of the mission, which was that Hypatius learned that Justinian had left the Palace.

CHRON. PASCH. OM.

(Ephraim says to Hypatius) ‘6 Ἰουστινιανὸς ἦν γὰρ μαθὼν ὁ Ὑπάτιος ὅτι ὁ βασιλεὺς ave- ἔφυγε καὶ οὐδείς ἐστιν «“ ἐνὶ τῷ παλατίῳ.᾽ καὶ χώρησε"

ταῦτα ἀκούσας ὃ Ὑπάτιος ἔδοξε θαρσαλεώτερος | καθέζεσθαι ἐν τῷ δεσποτικῷ καθίσματι τοῦ καὶ καθεσθεὶς ἐν τῷ καθίσματι μετὰ θράσους ἱππικοῦ. 3 | ἐτυράννει.

After this, the Paschal Chronicle has a sentence (ἦλθαν δὲ ἀπὸ Κωνσταν- τιανῶν K.T.r.), of which there is no trace in OM., and which may or may not have been in the original Malalas. It then goes on to relate that Justinian proceeded to the Cathisma, accompanied by Mundus, Belisarius, and others ; and there can be no question that in what follows the Paschal Chronicle copied Malalas.2 In OM. it is not stated that Justinian himself went to the Cathisma, but there is no reason to suppose that this detail was got by the Paschal Chronicler from a different source.

The notice of the burial and epitaph of Hypatius (p. 627-8) may have been derived either from Malalas or from the other source; but the notice of the confiscation of the property of the two brothers and a number of senators

1 Above § 7, (8). 2 καὶ ἀκούειν τὰς εἰς αὐτὸν εὐφημίας καὶ τὰς

ὑβριστικὰς φωνὰς ἃς ἔλεγον εἰς τὸν βασιλέα

Antioch (Theupolis), was slain in the Hippo- drome, must come from Malalas, and most distinctly points to Antiochene influence in the ᾿Ιουστινιανὸν καὶ εἰς τὴν αὔγουσταν Θεοδώραν. early part of the eighteenth Book. A purely Cp. Cramer, Anecd. Par., 2, 320 καὶ I. βασιλέω Constantinopolitan writer would never pick out ἀναθεματίζοντες ἐφύβριζον, Ὑ. δὲ πατρίκιον of 30,000, a person of purely local importance at εὐφήμησαν ἐν τῷ καθίσματι στέψαντες. Antioch ; whereas it is just what an Antio- 3 The notice preserved in the Paschal chene would do Chronicle that a certain Antipater, vindex of

THE NIKA RIOT, 101

was derived from Malalas. This is clear from a comparison with Theophanes and CF,

Cr, THEOPH. (p. 185, 30), CHRON, PASCH. (628), καὶ ἐδημεύθησαν of οἶκοι αὐτῶν, τὰ δὲ διαφέροντα αὐτοῖς πάντα καὶ ih ἰλλουστρίους καὶ συγκλη- μετὰ καὶ ἄλλων if πατρικίων καὶ ἐδημεύθη καὶ οἱ λυιποὶ πατρίκιοι τικοὺς δημεύσας ἐξώρισεν διὰ τὸ ἰλλουστρίων καὶ ὑπατικῶν δημευ- | οἱ ἅμα αὐτοῖς εὑρεθέντες ἔφυγον κατ᾽ αὐτοὺς τῇ τυραννίδι Ὑπατίου θέντων ὡς συνδρόμων Ὑπατίου. οἱ μὲν εἰς μοναστήρια οἱ δὲ εἰς προσθέσθαι. ᾿ καὶ ἐγένετο φόβος μέγας. εὐκτηρίους οἴκους καὶ ἐσφραγίσθη- σαν οἱ οἶκοι αὐτῶν. φανεροὶ δὲ

Kal ἐδημευθησαν καὶ ἐξωρίσθησαν.

§ 9. Theophanes! begins by (1) ἃ summary of the events of the sedition, derived from Theodore Lector. He then (2) copies in full the ἄκτα διὰ Καλαπόδιον, from some unknown source. (3) He passes to Malalas, and follows him mainly, though not altogether, for the details of the rebellion.

Adopting the same introductory formula as Malalas, Theophanes abbreviates and makes verbal alterations in the account of the incident of the two criminals rescued by the monks of St. Conon, Theophanes does not mention the four rioters who were beheaded, but only the three who were impaled ; on the other hand, he states that the two who escaped fell twice from the stake, while the epitomator of Malalas mentions only one fall. There is one discrepancy, which however need not be more than apparent. Theo- phanes states that the crowd, seeing the criminals lying on the ground, cried : τούτους TH ἐκκλησίᾳ, while OM says that they acclaimed Justinian (εὐφήμησαν τὸν βασιλέα). But there is no reason against supposing that the original text of Malalas, which both Theophanes and OM abbreviated, contained both statements.

And now we come to a remarkable point in the narrative of Theophanes. He states that the praefect sent soldiers to guard the rescued criminals in the asylum of St. Laurentius :

n , > \ ἀκούσας δὲ ὁ ἔπαρχος ἔπεμψε στρατιὼτας τοῦ φυλάττειν αὑτοὺς,

which corresponds to Malalas, 473, 19:

e

fa) fol \ ΄ καὶ γνοὺς ταῦτα ὁ τῆς πόλεως ἔπαρχος πέμψας στρατιωτικὴν βοήθειαν

Ε) ͵ > \ > wv ἐφύλαττεν αὐτοὺς ἐκεῖσε ὄντας.

But instead of continuing the narrative as it stands in Malalas, he omits the events described in 474, 1-14, and at once proceeds to the incident of the visit of the demes to the praetorium, to ask for an answer respecting the fate of the criminals,

om. (474, 14).

Bpabdelas δὲ γενομένης ὥρας ἦλθον ἐν τῷ πραι- τωρίῳ τοῦ ἐπάρχου τῆς πόλεως αἰτοῦντες ἀπόκρι-

THEOPHANES (184, 12). καὶ γνόντες of δῆμοι ἀπῆλθον εἰς τὸ πραιτώριον αἰτοῦντες τὸν ἔπαρχον ἀποστῆναι ἐκ τοῦ ἁγίου Λαυρεντίου τοὺς φυλάσσοντας στρατιώτας" καὶ σιν περὶ τῶν προφύγων τῶν ὄντων ἐν τῷ ἁγίῳ ἀποκρίσεως οὐκ ἔτυχον παρ’ αὐτοῦ" καὶ θυμωθέντες Λαυρεντίῳ. καὶ μὴ τυχόντες ἀποκρίσεως ὑφῆψαν ἔβαλον πῦρ εἰς τὸ πραιτώριον. | πῦρ ἐν τῷ αὐτῷ πραιτωρίῳ.

1 Theophanes is cited throughout from C. de Boor’s edition.

102 THE NIKA RIOT

Now the passage which Theophanes has here omitted is absolutely indispensable to the comprehension of the story, for it describes the union of the Blues and Greens. This union is the key of the whole episode, and the narrative of Theophanes is vitiated by its omission. The question arises : what was his motive for omitting.it? The answer is:

Theophanes thought that the scene in the Hippodrome described by Malalas (p. 474, 1-14) was the same as that in which the altercation between Justinian and the Greens respecting Calapodius had occurred, and which he had already tlescribed from another source. He therefore omitted it, to the detriment of his whole story.

Theophanes then states, with Malalas, that the people, receiving no answer, set fire to the Praetorium; but goes on, apparently deserting Malalas to group all the conflagrations of the riot together without distinction of the days on which they took place. His enumeration falls into three groups ; Ρ. 184, 1. 15-19; 2b. 1. 19-24; ib. 1. 24-27; corresponding respectively to Chron. Pasch. p. 623; p. 621-2; p. 622.

The Paschal Chronicler does not give the date of the first conflagration which he mentions, but he places it immediately after the sally of Belisarius and his Goths from the palace; the second took place on Friday: the third on Saturday. That he and Theophanes used the same source for their events is

clear from a comparison.

CHRON. PASCH.

καὶ λοιπὸν ἐνέπρησαν τὴν εἴσοδον τοῦ παλατίου τὴν χαλκόστεγον καὶ ἐκαύθη μετὰ τοῦ πορτίκου τῶν σχυλαρίων καὶ τῶν προτηκτόρων καὶ κανδιδά- των, καὶ γέγονε διακοπῆ. ὁμοίως δὲ ἐκαύθη καὶ τὸ σένατον ὕπου ἐστὶ το λεγόμενον Αὐγουσταῖον καὶ ἣ μεγάλη ἐκκλησία πᾶσα σὺν φοβεροῖς καὶ θαυμαστοῖς κίοσι πᾶσα ἐκ τετραέντου κατηνέχθη. κἀκεῖθεν κατῆλθεν ὃ δῆμος πάλιν εἰσελαύνων. ἐπὶ τὸν Ἰουλιανοῦ λιμένα εἰς τὸν οἶκον Πρόβυν: καὶ ἐζήτει παρ᾽ αὐτοῦ λαβεῖν ὅπλα καὶ ἔκραζον, Πρό- βον βασιλέα τῇ Ῥωμανίᾳ. οἶκον τοῦ αὐτοῦ πατρικίου Πρόβου: καὶ ὀλίγων καυθέντων ἐλείφθη τὸ πῦρ καὶ ἐσβέσθη. τῇ δὲ παρασκευῇ ἡ μέρᾳ τοῦ αὐτοῦ μηνὸς ιστ΄ ἦλθον οἱ δῆμοι εἰς τὸ πραιτώριον τῶν ἐπάρχων καὶ ἔβαλον ἐκεῖ πῦρ καὶ ἐκαύθησαν αἱ στέγαι τῶν

καὶ ἔβαλον πῦρ εἰς τὸν

βασιλικῶν οἴκων καὶ μόνον τοῦ αὐτοῦ πραιτωρίου ἐφύσησεν γὰρ ἄνεμος βορρᾶς καὶ τὸ πῦρ ἔξω τυῦ πραιτωρίου ἐδίωξεν καὶ ἐκαύθη τὸ βαλανεῖον τῶν ᾿Αλεξάνδρου καὶ ὁ ξενὼν τῶν Εὐβούλου ἐν μέρει καὶ ἣ ἁγία Εἰρήνη (ἥτις ἦν κτισθεῖσα K.T.A.) καὶ ὃ ξενὼν τοῦ Σαμψῶνος ὃ

g ‘ ὅπου Ta σκρίνια,

μέγας ἐκαύθη, καὶ ἀπώλοντο οἱ ἐν αὐτῷ ἀνακείμενοι |

ε

oaBBaty...... oi ὄχλοι ἦλθον αὐτοὶ εἰς τὸν ὀκτάγωνον τὸν ὄντα εἰς

ἄρρωστοι. τῷ δὲ

1 Om. codd., restituit de Boor. . Bieliaev is certainly wrong in suggesting as an alternative emendation τὸ στενὸν

ἕνατον codd,

THEOPH. (184, 19-24). καὶ ἐνεπύρισαν τὴν εἴσοδον Tod παλατίου τὴν χαλκόστεγον καὶ τὸν πόρτικον' τῶν προτικτόρω»

, 9 ~ > , καὶ τὸ σένατον" τοῦ Αὐγουστέω-.

καὶ κατῆλθεν ὃ δῆμος εἰς τὸν Ἰουλιανοῦ, τὸν Σοφίας λέγω, λιμένα, εἰς τὸν οἶκον Πρόβου ζητοῦντες ὕπλα καὶ κράζοντες ἄλλον βασιλέα τῇ

.« πόλει: καὶ ἔβαλον πῦρ εἰς τὰ Πρόβου καὶ κατη-

νέχθη ὁ οἶκος.

καὶ ἦλθον καὶ ἔκαυσαν τὸ βαλανεῖον τοῦ ᾿Αλε- ξάνδρου ;

καὶ τὸν ξενῶνα τῶν Σαμψὼν τὸν μέγαν καὶ ἀπώλοντο οἱ ἄρρωστοι, καὶ τὴν μεγάλην ἐκκλησίαν σὺν τοῖς ἀμφοτέροις κίοσιν: καὶ πᾶσα ἐκ τετραέντου κατηνέχθη.

(meaning the Golden Hand) ; Pyzantina, i. 132. σένατον is certain.

THE NIKA RIOT. 103

CHRON. VPASCH, THEOPH, (184, 19-24).

μέσον τῆς βασιλικῆς τῶν Tovvaplwy καὶ τοῦ δημο- | | |

σίου ἐμβόλου τῆς Ῥηγίας' καὶ. . . of στρατιῶται . ἔβαλον πῦρ καὶ ὑφῆψαν τὴν ὀκτάγωνον" | καὶ ἐξ αὐτοῦ τοῦ πυρὺς ἐκαύθη τὰ πέριξ τοῦ ἁγίου ] 1, 15-17, Θεοδώρου τῶν Spwpaxiov δίχα τοῦ σκενοφυλακίου καὶ ἐκάησαν οἱ ἔμβολοι awd τῆς καμάρας τοῦ

τοῦ Φούρνου τοῦ ἁγίου οἴκον. ὁ δὲ ἔμβολος ὅλος φόρου ἕως τῆς χαλκῆς τά τε ἀργυροπρατεῖα καὶ τὰ τῶν ἀργυροπρατείων καὶ ὃ οἶκος Συμμάχου τοῦ Λαύσου πάντα πυρὶ ἀνηλώθησαν.

ἀπὸ ὑπάτων dpdivaplwy καὶ ἡ ἀγία ᾿Ακυλῖνα ἕως |

τῆς καμάρας τοῦ ἄλλου ἐμβύλου τοῦ φόρου Κων- —

σταντίνου ἐκαύθη. }

Comparing these two accounts we see that :

1. Theophanes transposes the conflagrations which arose out of the burning of the octagon and which the Pasch. Chron. explicitly assigns to Saturday, and places them before the conflagrations which the Pasch. hron. assigns to Friday and the preceding days.

2. While Theophanes presents in the same order as the Pasch. Chron, the conflagrations which took place before Friday and those which took place on Friday, he exhibits one remarkable discrepancy. Instead of bringing the burning of St. Sophia into connexion with the burning of the Senate House and the palace porticoes, which the Pasch. Chron. places before Friday, he brings it into connexion with the burning of the Bath of Alexander, the Xenon of Sampson, &c. which the Pasch. Chron. places on Friday: and yet he describes the burning of St. Sophia in the identical words used by the Pasch. Chron.

3. While in the main Theophanes Ἴ the Paschal Chronicler were using the same source, there is one striking discrepancy as to a fact. They describe in almost the same words the rush to the house of Probus, but, while the Paschal Chronicler says that the fire was quenched ὀλέγων καυθέντων, Theophanes states that κατηνέχθη ὁ οἶκος. It seems clear that one of the two chroniclers must have here referred to a different source.

4. In the description of the conflagration of Friday, the text of Theophanes omits the important buildings, St. Irene and the Xenon of Eubulus, which are mentioned by Chron. Pasch., and of course by the common source. I do not believe that Theophanes intended to omit them. They easily fell out through homoioteleuton, and we should probably amend in the text of Theophanes (p. 184, 1. 25) :

καὶ τὸν Eevava τῶν «Εὐβούλου (ev μέρει) Kal τὴν ἁγίαν Εἰρήνην καὶ τὸν ξενῶνα τῶν; Σαμψὼν τὸν μέγαν.

5. In the conflagration of Saturday (Chron. Pasch.), Theophanes (184, 17) mentions the Palace of Lausus, which is not mentioned by Chron. Pasch. but he omits all mention of the Octagon and other buildings.

The main question which here arises is this: how is the remarkable inversion of the order of events in Theophanes, as compared with the Paschal Chronicle, to be explained? The answer must be postponed, till we come to consider the topographical difficulties connected with the riot

(see below § 21).

104 THE NIKA RIOT.

From the enumeration of the burned buildings, Theophanes passes to the resolution of Justinian to flee and gives us a unique notice as to the Emperor's plan of flight (184, 27-30). It is impossible to determine whether it comes from Malalas or not; but it seems to be out of its order, for the next sentence (185 1. 1—2=Chron. Pasch. 622, 18), concerns the events of Saturday.

The following account of the elevation of Hypatius and the final scenes is derived from Malalas, This can be seen without any difficulty by comparing it with the Oxford Malalas and Chron. Pasch,?

§ 10. George Cedrénus seems to have derived the first part of his brief account of the riot from Theodore Lector. It corresponds closely to the notice which Theophanes took from Theodore; only it is fuller, and therefore was not derived through Theophanes. It is fuller in two points: (a) the Xen6dn of Eubulus is mentioned; (ὦ) to ἡ μεγάλη ἐκκλησία are added the words καὶ τὰ χαρτῶα αὐτῆς δικαιώματα καὶ ἡ πρόσοδος πᾶσα." The second section is identical with the second part of the fragment of Cramer, Anecd. Par. ii. p. 320.3 This should conclude the notice ; but a statement is added that the Octagon and Zeuxippus were burnt; and there is a reference to the fire of A.D. 476. There is no trace here of the use of Theophanes,.

§ 11. Zonaras had before him, in writing his account of the Nika revolt,‘ a lost source which differed considerably from those that we possess.°

Starting with an introductory sentence suggested by a source which was also used by Cedrénus,® Zonaras comes at once to the main point, the union of the Blues and Greens. But the distinctive feature of his story is the prominence given to the battle between the barbarians and the demes, and the vain attempt of the clergy to pacify the tumult. He calls the barbarians Heruls,’ and his notice igs confirmed by Procopius, who mentions (at a different stage of the episode) that Mundus had a force of Heruls with him. From Zonaras alone do we learn of the part played by women in the riot.

1 This is so clear that it is unnecessary to show it in full But the comparison may be facilitated by the following references :—

Theophanes (de Boor) 185, 2-3 =cf. Malalas, 475, 22-23 ” », il = ” 476, 1 39 ᾽ν) 13 = 3} ” 3 f » 21 =ef. Mal., Hermes, 6; p.377. Ἢ », 22 =Malalas, 476, 9 37) 9?) 24 = 99 ” 10 τὸ EO OU aes », 19-22 ΩΣ ,, 80 =Mal., Hermes, 6, 377.

2 It may be noticed that Cedrénus gives τὸ προσκήνιον (τῆς βασιλικῆ5), as in Cramer, Anecd, Par. ii. 112; whereas Theophanes has προσκιόνιον,

3 καὶ τοὺς δύο ἐμβόλου----ἐν τῷ ἱπποδρόμῳ.

4 Ed. Dindorf, vol. 8, p. 271, 8-273, 22.

5 P, Sauerbrei, in his study on the sources of the middle part of the history of Zonaras, observed this, and describes the source as opti- mus ut videtur fons deperditus. De Font. Zon. quaestiones, in Comment. Philol. Jenenses, vol. I. p. 77. This (?) optimus fons deperditus was also used for the reigns of Leo and Zeno,

8 To the effect that all the beauty in the city left by the former great fire (of A.D, 476) was now consumed. Cp. Cedrénus p. 647. Probably both were using a common source. I cannot enter here upon the latest combinations of E. Patzig,—the Leoquelle and the Zwillings- quelle, &e.

7 Αἰλούρων.

THE NIKA RIOT. 105

But at this point! the value of the account of Zonaras ceases. He mentions that the fire was propagated by a violent wind—a fact known otherwise only from the Paschal Chronicler ; but of the buildings burnt he only mentions some in the region between St. Irene and the Palace. He has a peculiar statement? that Justinian wished to speak to the demes in the Hippodrome (‘theatre’), but that they, afraid of being caught (ὡς ἐν εἱρκτῇ), refused to run into the trap. This statement, referring to the appearance of Justinian in the Hippodrome on Sunday morning, is discordant with the account of Malalas (Chron. Pasch.), from which it appears that the

demes did assemble in the Hippodrome and reviled Justinian. The rest of the story however accords with the account of Malalas and may well have been derived either from Malalas or from a source dependent

on Malalas. Compare :

ZONARAS, 272, 31. καὶ ἐπὶ ἀσπίδος ἀναγορεύουσι βασιλέα.

αὐτὸν μετάρσιον ἄραντες

ZONARAS, 273, 4. ἴσχυσαν οἱ wep) τὸν βασιλέα πλείστων χρημά- τῶν διανομαῖς ὑποσῦραι τῶν Βενέτων πολλούς,

καὶ οὕτως τὴν τῶν δήμων ὁμόνοιαν διαστήσαντες χωρῆσαι κατ᾽ ἀλλήλων αἰντοὺς πε- ποιήκασι.

τότε τοίνυν ἀθρόον ἐκ τῶν βασι- λείων οἱ προκοιτοῦντες τῶν κρατούντων ἐξέθορον ἔνοπλοι.

Ib. 1. 12. μεστὸν γὰρ ἦν ἀνθρώπων τὸ θέατρον τῶν μὲν τῆς τῶν στασιαζόντων μοίρας, τῶν δέ γε πλειόνων κατὰ θέαν τῶν γιγομένων καὶ THS τοῦ Ὑπατίου ἀναρρήσεως ἠθροισμένων.

MALALAS. //ermes, 6, 377. καὶ ἀναγαγόντες ἐν τῷ σκουταρίῳ.

THEOPH., 185, 14 (MAL. 476, 4). ὑπέσυρέ τινας ἐκ τοῦ Βενέτου μέρους poyevoas χρήματα. MALALAS, 476, 6.

διχονοῆσαν δὲ τὸ πλῆθος ὥρμησαν κατ᾽ ἀλλήλων. THEOPH., 185, 2. ἐνόπλων στρατιωτῶν βοήθειαν καὶ κουβικουλα-

ρίους καὶ σπαθαρίου-.

MAL. HERMES, 6, 377.

πληρωθέντος δὲ καὶ ὅλου τοῦ ἱππικοῦ ἐκ τοῦ δήμου ὡς θελόντων θεωρῆσαι βασιλέα στεφόμενον.

Now I do not indeed regard these comparisons as conclusive ; it is quite

possible that they represent a different account, which agreed with, but was independent of, that of Malalas. Still it is remarkable that the account of. Zonaras would serve as a very accurate, brief summary of the account of Malalas. We must bear in mind the method of Zonaras, who was always con- cerned to change the words of his sources. If he found ἑἱππεκόν he was certain to substitute θέατρον ; if he found a part of πληρόω, he would use a part of γέμω or μεστός εἰμι; if he found ὑποκλέπτω one could predict that he would employ ὑποσύρω or something else? And he always of course avoided colloquialisms or Latinisms like σκουτάριν. There are only two points, one at the beginning, and one at the end, of this part of the episode, where the

1 p. 272, 5.

2 p, 272, 20-26.

3 Malal. and Chron. Pasch. have ὑποκλέπτω, Theophanes, ὑποσύρω. We must infer that ὑποκλέπτω was in the original Malalas. But it

would be improper to infer that Zonaras must have here used Theophanes ; for there it was quite natural that Theophanes and Zonaras should have hit independently on the same synonym.

106 THE NIKA RIOT.

influence of another source necd be assumed. The statement that Hypatius was proclaimed πῇ μὲν ἄκοντα, πῇ δὲ memecopévov,' which is in accordance with the story of Procopius, is not found in Malalas, so far as we can judge ; yet it would be a possible inference from the incident of the sending of Ephraim to the Palace. The number of the slain is set by Zonaras? at ‘about 40,000.’ According to Malalas, it was 35,000. It may be conjectured that in his other source Zonaras found 50,000 (the number given by John Lydus) and that he adopted 40,000 as somewhere between the two,

While I admit fully that the general coincidence may be accidental, and that Zonaras used throughout a different source, I cannot help thinking it more probable that the latter half of his narrative was derived, directly or indirectly, from Malalas.

IT. — CHRONOLOGY.

§ 12. Our data for determining the days on which the events of the riot took place are derived from Malalas and the Paschal Chronicler (who here pro- bably means Malalas), supplemented by two indications of Procopius. Theo- phanes has omitted all notes of time, except the most unimportant—namely that Hypatius and Pompeius were executed the day after they were arrested,

The first note of time is given to us by Malalas. The beginning of the tumult, the union of the Blues and Greens, the formal declaration of that union in the Hippodrome, took place on the 13th of January, A.D. 532.3 Their union was caused by the execution of members of both parties by the praefect; and that execution had taken place three days before,* that is on Sunday the 11th January. It follows that the celebration, at which the Greens pressed their complaints against Calapodius, took place not later than the 11th. Most historians have fallen into the error of confounding this first scene in the Hippodrome (described by the Paschal Chronicler and Theo- phanes) with the second scene on Jan. 13 (described by Malalas).°

On the night of the 13th (which fell on Tuesday),° the united demes, having got no satisfaction from the Emperor, proceeded to the Praetorium, and demanded whether the praefect had decided to pardon the escaped prisoners. Receiving no answer they set fire to the praetorium. Other places were burned, and the people remained in the streets, εἰσελαύνων ἀτάκτως (during the night).

In the morning (Jan, 14th, Wednesday),’ further outrages were com- mitted ; the complaints against John of Cappadocia, Tribonian, and Kudaemon were preferred, and they were deposed; Belisarius issued forth with a body of

1 p, 272, 29. distinguished them, and was followed by

2 p. 273, 19. Kalligas (so too Ranke). The assembly was % p. 474, 2-6. The date is also given by doubtless held on the eleventh, preliminary to

Marcellinus, the celebration of the thirteenth. Schmidt, p.47. 4 Ib. 1, μετὰ τρεῖς ἡμέρας. δ Bpadeias δὲ γενομένης Spas, p. 474, 14.

® Gibbon, Hedgkin, ete. Schmidt rightly 7 καὶ πρωΐας γενομένης, ἐν, 20,

THE NIKA RIOT. 107

Goths, and there was a conflict; then there were more conflagrations. ΑἹ] this is told in Malalas,' as if it took place on Wednesday, and in the corresponding part of the Chron, Pasch. there is no mark of time. But (1) it is improbable that all this occurred in one day; (2) the next events of which we hear belong to Friday; which leaves Thursday unaccounted for. Now in any case, something has been left out by the epitomator between the notice of the deposition of the obnoxious officers and the notice of the attack of Belisarius on the mob;? and this is probably the place where the night intervened. We may I think conclude, with great likelihood, that the main event of Wednesday, Jan. 14, was the deposition of the three ministers, and that the main event of Thursday, Jan. 15, was the sally of Belisarius from the Palace.®

At this stage Malalas, as represented in the Baroccian Epitome, deserts us; but fortunately the Paschal Chronicler, who up to this point has furnished no dates, now becomes precise, and fixes the events of Friday and Saturday.

On Friday, Jan. 16, the praetorium was burnt, according to Chron. Pasch.; the conflagration was carried by the wind from that building, and other buildings were burned, which are enumerated.

On Saturday, Jan. 17,5 there was a conflict between the soldiers and the mob. The soldiers set fire to the Octagon, and the fire spread to other build- ings. This was the fifth day of the riot, and here we get our first indication of time from Procopius.6 On the evening of this day, Justinian dismissed Hypatius and Pompeius from the Palace.

§ 13. The events of the following day, Sunday, Jan 18, are dated by Malalas, the Paschal Chronicle, Procopius, and Marcellinus. Malalas and the Paschal Chronicle give both the day of the week and the day of the month. Procopius names the day following the fifth day.’ Marcellinus is slightly in- accurate. While he assigns the beginning of the sedition to the 13th January, he states that the sedition lasted guingue continuos dies, and that Hypatius was elevated guinta huius nefandi facinoris die. It is clear that, if he counted the 13th, he should have said sex continwos dies, and placed the final scene (like Procopius) on the sixth. But the inaccuracy is a pure inadvertence. Every- one remembered two things: that the Hippodrome scene took place on the Ides, and that the riots lasted for five days. They began late at night on the 13th with the attack on the praetorium, and they were over before the night of the 18th. Thus the statements of Marcellinus, true separately, lead,

1 474, 20,-475, 10.

2 At 475, 8. 5 Ib, τῷ δὲ σαββάτῳ, τουτέστιν τῇ ιζ΄ τοῦ αὑτοῦ 3 Schmidt places the action of Belisarius on ἀυδυναίου μηνός. Wednesday, the burning of Chalké ete., on 5 p. 123; τῇ δὲ πέμπτῃ ἀπὸ τῆς στάσεως ἡμέρᾳ Wednesday night, the attack on the honse of περὶ δειλὴν ὀψίαν. Probus on Thursday (p. 60-1), Thus he follows 7 τῇ δὲ ὑστεραίᾳ (7d.), in reference to the date Chron. Pasch. cited in preceding note,

- -“ col , ~ , 4 γῇ δὲ παρασκευῇ ἡμέρᾳ τοῦ αὐτοῦ μηνὸς ιστ΄,

108 THE NIKA RIOT.

when combined, to a misconception. On the reckoning that Hypatius was elevated on the fifth day, the 14th of Jan. must be counted as the first day.

It would seem certain then that the appearance of Justinian in the Hippodrome with the Gospels in his hand, the elevation of Hypatius, and the bloody suppression of the riot, took place on Sunday, Jan. 18. But this is not the view generally accepted.

Although in Malalas (both in the Oxford epitome, and in the fuller account of the Paschal Chronicler) the elevation of Hypatius follows immedi- ately on the failure of Justinian’s solemn oaths to appease the revolt, historians have assumed that a night intervened between these two events! On this view, Justinian’s appearance in the Hippodrome takes place on the morning of Sunday, Jan. 18, and the coronation of Hypatius on the morning of Monday, Jan. 19. The motive for this arrangement of events is doubtless a wish to reconcile a slight discrepancy between Procopius and the Paschal Chronicler. According to Procopius, Hypatius and Pompeius were dismissed from the palace the night before the elevation of Hypatius. According to the Paschal Chronicler the senators were dismissed from the palace after Justinian’s appearance in the Hippodrome. It seems an easy and attractive way of reconciling these statements to suppose that the dismissal of Hypatius and his brother took place on Sunday night, and that ‘the fifth day’ meant by Procopius was Sunday, not Saturday. But there are serious objections to this combination.

(1) It is clear from the Paschal Chronicle and the Oxford epitome that the original chronicle of Malalas contained precise indications as to the days on which the various events fell. It is extremely hard to believe either that Sunday and Monday, the days (according to the received view) of the decisive events, would not have been distinguished in the original Malalas, or that both the Paschal Chronicler and the Epitomator let this distinction drop and so placed the final scene on the wrong day. The elaborate description in the Paschal Chronicle forbids the second supposition. (2) The tenor of the story deprecates the idea that a night intervened. According to Malalas (Chron. Pasch.), Justinian after he retires from the Hippodrome—it is still very early in the morning ?—dmmediately (εὐθέως) dismisses the senators; and, when they went forth from the Palace, ‘the people met Hypatius and Pompeius, and took Hypatius to the Forum. This

1S0 Gibbon and Mr. Hodgkin. So too Schmidt.

Kalligas does not seem to assume the interval of a night; but he follows Chron. Pasch, in placing the dismissal of Hypatius from the Palace on Sunday. He regards it as a conse- quence of Justinian’s fiasco in the Hippodrome ; but he does not seem to observe that he deserts the authority of Procopius, which in such a matter is decisive.

2 ἀπονύχιον (Chr.

Pasch.). This curious

word—émaét eip., I believe—must mean, as it is generally taken, in the early hours of the morning before the night is over. It is not given in the Lexicon of Ducange, but Sophocles compares Noy. Test. ; Mark i. 35, πρωΐ ἔννυχον λίαν. The closest analogy I can find to the use of ἀπὸ νυκτός, which the compound appears to presume, is in the phrase ἀφ᾽ ἡμέρας miveew=de die bibere. In the Wasps of Aristophanes ἀπὸ νυκτῶν μέσων means in the hours after midnight.

THE NIKA RIOT. 109

shows as clearly as possible that, according to Malalas, all happened on the same day. Only, Malalas has fallen into an error, easily explicable. Hypatius and Pompeius had left the Palace, as Procopius states, the evening before, and the people fetched them from their houses. Nothing would be more natural than a mistake of this kind,—the transference of the dismissal of the two brothers from the time at which it actually occurred to the time at which it assumed significance. (3) The statement of Marcellinus is decisive against the 19th as the day of the elevation of Hypatius. For in that case the sedition would have begun according to him on the 15th, which is, on no theory, possible. On Jan. 19th (Monday), Hijpstion and Pompeius were put to death.

IIJ.—ToroGrapuHy.

§ 14, It will be most convenient in the first place to determine as far as possible the sites of the buildings which were connected with the Nika riot, and then to examine, in the light of our conclusions, the evidence of the authorities, who in some respects conflict with one another.

There is no difficulty any longer as to the general position of the buildings around the Augusteum.? That place was bounded on the north by the southern side of St. Sophia, on the east by the Senate-house of the Augusteum (which must be carefully distinguished from the Senate-house of the-Forum of Constantine) * and part of the palace wall, on the south by Palace buildings, the Chalké or main entrance to the Palace, and the Baths of Zeuxippus. There is still room for doubt whether the west side of the Augusteum was partly closed by buildings or not. No doubt, an accurate line was drawn between the precincts of the Augusteum and the Mesé. The Mesé (‘ Middle Street’) led down from the Augusteum into the valley between the First and Second Hill, and ascended to the Forum of Constantine on the top of the Second Hill. Passing through the Forum it went on all the way to the Golden Gate; but with its course beyond the Forum of Constantine we have not to do here. The Milium, opposite to the S.W. corner of the atrium of St. Sophia, must have stood on the line which

1 Malalas, 476, 21. (So Theophanes).- 2 It is unnecessary to argue here against the

cepted fact. But then he is totally ignorant of the investigations of MM. Strzygovski and

untenable view of M. Paspatés, which has been universally condemned by competent critics. (His mistake was partly due to the confusion of the Augusteum with the Forum of Constantine ; a mistake partly derived from Labarte). See my paper in Scottish Review, April 1894, on the Great Palace of Constantinople ; Lethaby and Swainson, S. Sophia, p. 7 sqqg. Mr. Grosvenor in his recent work on ‘Constantinople’ (2 vols. 1895), a work which has very little archaeologi- cal value, adheres to the view of his master, M. Paspatés, as if it were an established and ac-

Forchheimer on the cisterns of Constantinople.

3 This mistake was made by Labarte, and before him by Ducange, who was guilty of a triangular confusion ; namely, the senaton of the Forum =the senaton of the Augusteum = the Basilica. This has been fully pointed out by Bieliaev in his important article, Chram Bogoroditzy Chalkopratiskoi v Konstantinopolie, in the Lietopis for 1892 of the Hist. Phil. Society of the University of Odessa (Viz. Otdiel. i.); p. 104-5.

110 THE NIKA RIOT.

divided the Augusteum from the Μοβῶ The Baths of Zeuxippus, which lay on the left of one issuing from the Chalké, occupied the south-west corner of the Augusteum. The position of these Baths in relation to the Chalké on one side and to the Kathisma of the Hippodrome on the other is represented with fair correctness on the plan of Labarte.?

"&, δ 2) 4 Chalhoprateia 2 » So,

Bas,)- Cungitiva

arig. δὰ andor

Plan of the imperial quarter of C ONSTANTINOPLE

Nore.—Buildings, ete., of which there are actual remains in situ are distinguished by BLOCK LETTERS underlined. RoMAN CAPiTALs are employed for sites which can be inferred with complete or approximate decision ; italics for those whose determination is only tentative.

Thus the general lie of the group of buildings around the Augusteum can be determined with sufficient certainty for our purpose, nor is there any doubt about the buildings north of St. Sophia—the Xendn of Hosios

1 There can be no doubt about the position of He shows from Constantine Porph., De Caer. the Milium. See Bieliaev’s thorough discussion 17, 106, 10, 84 (i. ed. Bonn) that in proceeding in Byzantina: Ocherki, etc., ii. p. $2-94; ep. from the Chalké to the Milium the Emperor had the Odessa Lietopis, 1892, op. cit., p. 102. This the Zeuxippus on his left, in procceding from involves a considerable change in the plan of | the Milium to the Chalké on his right ; and in Labarte which Biecliaev inserted in the first vol. one ceremony the Zeuxippus was a station be- of his Byzantina. tween the Milium and the Chalké.

2 Cp. Bieliaev, Byzantina, ii. p. 93, note 1.

THE NIKA RIOT. 111

Sampson, the Church of St. Irene, and the Xenén of Eubulos. and Sampson are still there.

The Irene

§ 15. It is different when we come to the buildings which were situated westward, on either side of the Mesé, between the Augusteum and the Forum.' Here we must be content with approximate and conjectural results, We have only a fixed line and a fixed point, in relation to which we have to attempt to group a number of edifices which have been destroyed. The fixed line is the direction of the Mesé; the fixed point is the position of the Basilica.

The site of the Imperial Stoa or Basilica, which contained the Library, is identified by the Cistern Basilica or Jeré batdn Serai. This identity, recognised by Gyllius, has been completely established by Strzygovski2 Procopius tells how Justinian built the cistern, and gives the valuable infor- mation that the cistern was laid on the south side of the great quadrilateral peristyle court of the Basilica.* We thay infer from this that the greater part of the Basilica buildings were to the north of Jeré batiin Serai, and that it did not reach down to the Mesé.

The Basilica is described as ‘behind the Milium’ by the anonymous author of the Patria,t and Zonaras® states that it was ‘very close to the Chalkoprateia, The proximity comes out in the fire of A.D 476. That fire began in the Chalkoprateia and destroyed the Basilica and ‘both the stoai.’ ® What are both the stoai? and was the quarter of the Chalkoprateia north of the Basilica, or between the Basilica and the Mesé ?

The position of the Church of the Virgin in Chalkoprateia, which was unportant in the court ceremonies, has received an elaborate discussion recently from Bieliaev.’ He shows clearly that it was on the north side of the Mesé, and that the Emperor when, in passing from the Forum of Constantine to the Palace, he visited this church, turned to the left from the Mesé in order to reach it. He also thinks that the church was quite close to the Portico which ran along the north side of the Mesé; but his arguments are not decisive.2 It cannot be determined from the data of the De Caeri- montis how far the Emperor had to proceed up the street to the left® before he reached the Chalkoprateia. Those data are uot inconsistent with another view which places the Chalkoprateia close to the north-west of St. Sophia,

11 shall use ‘Forum’ simpliciter, as Byzan- _ teum, whereas it means of course the approach to

tine writers did, for the Forum of Constantine ; ὁ φόρος is regularly used thus, 6.4. in Theophanes. For this use ep. Bieliaev in the Odessa Lietopis, 1894, p. 17.

2? Die byzantinischen Wasserbehiilter von Constantinopel, by P. Forchheimer and I, Stray govski (1893), p. 177-180. Cp. Mordtmann, Esquisse topographique, p. 66-7. Strzygovski falls into error in interpreting a passage of Con- stantine Porphyrogennetos (i. 165) concerning the second procession on the feast of the Annun- ciation. He places the Antiforumat the Augus-

the Forum (of Constantine), and he takes the Church of St. Constantine in the Forum for the Church of St. Sophia (p. 179).

3 See De Acdificiis, i. 2.

4 Banduri, 29. Codinus, 39.

5 xv. 3, (iii. p. 340, Dind.).

§ Cedrenus, i. 616 = Zonaras, xiv. 2 (p. 257),

7 Op. cit. in the Lictopis of Odessa, 1892.

ΒΊΩΙ 101.

9 ἀριστερὸν ἐκκλίνας, Const. Porph., i. p.

112 THE NIKA RIOT.

and thus north of the Basilica. This view is held by Mordtmann who identifies the Church of the Virgin with the Mosque of Zeineb Sultan,! and is also suggested on other grounds by Krasno-sel’tzev.2, To this question we shall return again.

§ 16. The anonymous author of the Patria, advancing from the Augus- teum to the Forum, speaks successively of the following buildings:* the Milium, the Church of St. John, the Apostle, the Church of St. Theodore Σφωρακίου, the Octagon, and the Palace of Lausus. It is left indeterminate which of these buildings is to the south and which to the north of the Mesé. The position of the Church of St. John (Diippius) can be pretty confidently placed south-west of St. Sophia, south-east of the Basilica, and not far from the Milium;* but it does not concern us at present. The Octagon he describes as close to the Basilica;® and this agrees with the notice of the Paschal Chronicler that it lay between the portico of the Regia (that is, the Basilica) and the basilica of the skindressers.6 The most probable inference is that it was west of the Basilica. It could hardly have been south, for then it would have been adjacent to the Mesé and there would hardly have been room for the basilica of the skindressers. Assuming then provisionally that it lay west of the Basilica, we might place the Church of St. Theodore pro- visionally south of the Octagon, that is, between the Octagon and the Mesé. This would suit the order of the Anonymus, quoted above, where St. Theodore is reached before the Octagon.’

§ 17. We now come to the Palace of Lausus, as to which new views have been recently put forward. It was close to the Mesé, but the question is, was it on the north or on the south side? The anonymous topographer leaves it open. Mordtmann places it on the south side; but Bieliaev and Strzygovski have independently argued that it was on the north side.

Bieliaev® has derived his view from passages in the De Caerimoniis. ‘Like the Chalkoprateian Church, the House of Lausos was on the right side of Middle Street, to one going along it from the Augusteum to the Forum of Constantine, and lay near the right-hand portico.’ The passages in the De

1 Esquisse top., p. 4. He quotes Antony of Novgorod, the Russian traveller, who says that ‘going towards the Hippodrome [from the Forum] under the covered portico of Eubulus, we meet the Church of the Mother of God, containing the marble table, on which our Lord celebrated the Sacrament.’ Is the portico of Eubulus the north portico of the Mesé? In any case, it is not necessary to conclude that the Church was adjacent to the Mesé. It is enough that the traveller reached it by a street off the Mesé.

* Zamietka po voprosu o miestopolozenti Chal- kopratiiskago chrama v Konstantinopolie, in the Odessa Lictopis of 1894, p. 309-316.

% Banduri, 27 sqq.

4 Mordtmann, p. 66, cf. his plan.

5 τὸ τετραδίσιον ὀκτάγωνον πλησίον τῆς βασιλικῆς».

6 p. 621.

7 The position is discussed by Mordtmann, p. 67. He places it north of the Octagon, but I fail to see the evidence. In any case Gyllius was wrong in seeking it on the western slope of the third hill, near the Véfa Meidani (i.11, p.58). Ducange’s notice does not help us (Deser. Urb. Const., p. 139-40). The Anonymous (Banduri 53) places the perfume market near the Octagon συνεγγὺς TOD ay. Θεοδώρου Tod Σφορακίου.

8 Op. cit. p. 103.

THE NIKA RIOT. 113

Cerimoniis describe the progress of the Emperor from the Forum to the Chalkoprateian Church. The Emperor having passed through the Anti- forum ‘enters the portico near the Lausus and from there goes’ to the Chalkoprateian Church.! If the weather is bad, he goes to the Forum (from the Milion) ‘by the portico, and ‘comes down again by the same portico and the Lausus, and turning to the left goes to the Chalkoprateia.’* It must be certainly admitted that prima facie it would be natural to understand the northern portico of the Mesé; and this would seem to imply that the Lausus was at the northern side of the Mesé, close to the street which turned north- ward to the Chalkoprateia.

We have however another totally different indication. The Church of St. Euphemia ἐν τῷ ἱπποδρόμῳ was west of the Hippodrome.* It was situated ἐν τοῖς ᾿Αντιόχου πλησίον τοῦ Λαύσου. It seems unlikely that the definition πλησίον τοῦ Λαύσου would be used, if the Lausus had been north of the Mesé. The Church of St. Euphemia was probably south-west of the Hippodrome.®

The indication of the proximity of the Palace of Lausus to the cistern of Philoxenus, which supplied it with water,® is unfortunately of no use, as the cistern of Philoxenus has not been found. It used to be identified with the Bin bir dirék,’ but this view has been upset by Strzygovski.* The cistern of Philoxenus was certainly close to the Forum, and adjoined the church of St. Aquilina; and Strzygovski concludes, by combining the Anonymus of Banduri-with statements of the Paschal Chronicler relating to the Nika riot, that the Lausus was on the north side of the Mesé. The force of the data in regard to the Nika riot will be appreciated below.

Certainly, the most important passages seem to be most satisfactorily explained by the view that the Lausus was on the north side of the Mesé ; and perhaps the passage of the Synaxarion may be brought into unison by supposing that part of the palace of Antiochos reached the Mesé and faced the palace of Lausus.

§ 18. One building still remains to be considered, the Praetorium. To reach the Praetorium from the Palace, one proceeded along the Mesé past the Palace of Lausus.® It was apparently on the Mesé, between the Lausus and the Forum.” Moreover it was close to the Church of the Forty Martyrs, which was in the Mes¢.1 But the Anonymus of Banduri seems to place this

1 Const. Porph., i. p. 165. καὶ εἰσέρχεται ἐν τὸ πραιτώριον ὑπήντησαν αὐτῷ of Πράσινοι eis τὸ τῷ ἐμβόλῳ πλησίον τοῦ Λαύσου, καὶ amd τῶν Λαῦσον.

ἐκεῖσε ἀπέρχεται κ.τ.λ. 10 Cp. the account of the fire in A.p. 603, in 2 p. 169. Chron. Pasch. p. 695: ἐκαύθη ἡ μέση ἀπὸ τῶν 3 Cp. Ducange, op. cit. p. 145. Λαύσου καὶ Td πραιτῶριν τοῦ ἐπάρχου τῆς πόλεως 4 Synaxaria, July 11. ἕως τῆς ἄρκας ἄντικρυ τοῦ φόρου Κωνσταντίνου. 5 Mordtmann, Joc. cit., and plan. This Arca comes in the Nika riot. In Const. 6 Cedrenus, i. 564. Porph., i., p. ὅθ, the Praetorium is a station 7 Gyllius, p. 127. So Mordtmann. between the Forum and the Milium.

8 Op. cit. p. 170 844. 11 Theoph. p. 267, 31.

9 Cp. Theophanes, p. 239, 9, ἀπερχομένου εἰς H.S.—VOL, XVII, I

114 THE NIKA RIOT.

church to the west of the Forum of Constantine and near the Forum Tauri.! This, however, is clearly a mistake. There is a passage in the Alexiad of Anna Komnéna which leaves no doubt that the Church of the XL Martyrs was east of the Forum. The Comnene ladies meet Alexius in the Forum and having taken leave of him made all haste ‘to the temple of the Great Sophia.’ Close to the precinct of the Forty Saints they were met by the tutor of Botaneiatés.2 Mordtmann places the church close to the Tiirbé of Mahmud, whose site marks the entrance to the Forum of Constantine.’

The site of the Praetorium has been discussed by M. Paspatés in his Bufavrivai Medérar.t He thought that he had found its ruins near the Church of St. Anastasia, which he successfully identified with the Mechmet Pasha Tzamii, south-west of the Hippodrome.’ But his arguments prove nothing. He points to several passages which show that to reach the Praetorium from the Sophian port one had to go up; but this datum would suit many sites. I have seldom seen a weaker piece of topographical identification.®

δ 19. From Procopius, John Lydus, and Malalas, one: would infer that there were two distinct conflagrations, of which the first consumed buildings around the Augusteum, and the second raged along the Mesé and especially among buildings north of the Mesé.

(1) ‘The city, says Procopius, ‘was invaded by fire. And the sanctuary of Sophia, and the Bath of Zeuxippus, and the parts of the Imperial Palace from the Propylaea to the so-called house of Ares, were burnt and destroyed.’ That is the first group. (2) ‘ And besides there were burnt at the same time [z.e. on the same occasion ; not “simultaneously ”’] the great porticoes (στοαί) reaching up to the Agora named from Constantine, and many houses of rich men, and large property.’ That is the second group.

1 p. 48: Forum ; Artopolion ; palace of Tox- aras ; Church of Forty Martyrs ; Anemodulion ; forum Tauri. Cp. p. 93.

2 Vol. i. p. 70 ed. Reifferscheid.

3 p. 69. ‘Une citerne ἃ l’ouest de Bin bir dirék, en face du tombeau de Sultan Mahmoud, parait marquer l’emplacement de l’église des

XL. martyrs de Nicopolis.’ Cp. plan. 4 p. 368 sqq. 5 p. 364 sqq. ὅν. 371-2. Nor is there any proof of his

statement that the house of Probus was near the Praetorium (p. 372). The texts which he cites—like so many of this antiquarian’s cita- tions—are irrelevant. This identification of Paspatés is, I observe, also rejected by G. Laskin, in his paper, Zamietki po drevnostyam Konstantinopolia, in the Vizantiski Vremennik, iii, p. 339. Laskin places the Praetorium between the Augusteum and the Forum, but 1

do not see how the passage in Chron. Pasch. (loc, cit.) proves that it was on the north side of the Mesé ; for this, I suppose, is what he means by saying that it was ‘on the other side of the street’ from the Great Palace.—Laskin thinks that the Great Embolos built by Arcadius oppo- site to the Praetorium (Theoph., p. 74, 23) is the Cherni Veliki Ubol of Anthony of Nov- gorod. It is noteworthy that Kondakov (in Vizantiskiia Tserkvi i Pamiatniki Konstantino- polia, 1886, p. 182) identifies this Black Embolos with the μακρὸς ἔμβολος τοῦ Mavpiavod. But the Embolos of Maurianus was at the other side of the Forum, as is proved by Const. Porph., de Cer., p. 156 (cf. Mordtmann, p. 7). The Black Embolos was near St. Anastasia, which was said to be in the region of Maurianus (S. W. of the Hippodrome), which must be clearly distinguished from the Portico of Maurianus,

THE NIKA RIOT 115

(1) ‘ The fire, says Lydus, ‘beginning with the Entrance to the Palace, spread from it to the chief Sanctuary [St. Sophia], thence to the senate- house in the Augusteum, and from it to the Zeuxippos’ [here an antiquarian digression]. This is the first group. (2) When these were consumed,! ‘ the porticoes up to the Agora of Constantine’ were ravaged, and ‘the adjacent buildings, north and south thereof, were naturally reduced to ashes,’ This closely corresponds to the second group of Procopius, and one suspects that Lydus had the work of Procopius before him, But he adds to the first group the senate-house, which Procopius omits,

(1) Malalas, as represented by his epitomator, notices the conflagration of the first group and connects it with the night of Jan. 13 :—7 χαλκῆ τοῦ παλατίου ἕως τῶν σχολῶν Kal ἡ μεγάλη ἐκκλησία καὶ ὁ δημόσιος EuBodkos— to which he adds on the following day μέρος τοῦ δημοσίου ἐμβόλου ἕως τοῦ Ζευξίππου. The original notice of Malalas, from which this is abbreviated, can, as we have seen, be made out with the help of Theophanes and the Paschal Chronicler. The important point is that the ‘public portico’ is the portico of the Augusteum, not of the Mesé. The phrase ἕως τῶν σχολῶν evidently comes to much the same thing as the ἄχρι és τὸν “Apews οἶκον of Procopius, and means the parts of the palace adjacent to the Chalké on the east side. (2) The second group, not preserved in the Oxford epitome, is preserved partly in the Escurial fragment—more fully in the Paschal Chronicler (see above § 9).

Now while our text of Malalas preserves the date of the first con- flagration (13—14 January), the Paschal Chronicler preserves the date and circumstances of the other conflagration (having, no doubt, derived these facts from Malalas). It took place on Saturday, Jan. 17. This fire spread from the Octagon, and was the work of the soldiers.

But there was yet another group of buildings consumed by fire, of which Procopius, Lydus, and our Malalas say nothing. We learn about this group from the Paschal Chronicler and Theophanes (who are confirmed by the enumerations of other writers) ; and there is reason to believe that we should have learned about it from the original Malalas. This group consists of the Church of St. Irene, the Xenodochia of Sampson and Eubulus, and the baths of Alexander; and according to the Paschal Chronicle this conflagration occurred on Friday, Jan. 16.

§ 20. We have thus three distinct conflagrations : (1) Jan. 13—14: Augusteum buildings, including St. Sophia ; (2) Jan. 16: buildings north of St. Sophia ;

(3) Jan. 17: Octagon, adjacent buildings, porticoes of Mesé, buildings south of Mesé.

1 rév δὲ τηλικούτων σωμάτων els πῦρ μεταβαλόντων, p. 266, 1. (Possibly σωμάτων should be

δωμάτων.) ἃ I

116 THE NIKA RIOT.

The order and the details are best preserved in the Paschal Chronicle (except in regard to the date of (1)). Procopius and Lydus preserve the order of (1) and (3), but omit (2). Theophanes falls into the curious mistake of changing the order to (3) (1) (2); and this mistake demands explanation.

Another building, passed over by Procopius! and Lydus, is stated by Malalas, the Paschal Chronicler, and Theophanes to have been burnt. I refer to the Praetorium. But, strange to say, its conflagration is connected with group (1) by Malalas,? with group (2) by the Paschal Chronicler, with group (3) by Theophanes. This is a very interesting question.

Now it is clear that topographically the Praetorium would belong to group (3); for we have seen that it was close to the Mesé and not far from the Forum. But, on the other hand, the burning of the Praetorium cannot be connected with the burning of group (3); for (a) it was not accidentally burned but deliberately fired by the people, and (Ὁ) Theophanes himself, following Malalas, places it as the first building burnt, whereas group (3) was burnt last. The circumstances of the outbreak of the riot do not permit us to doubt the statement of Malalas that the first outrage was the burning of the Praetorium. On the other hand it must not be imagined that the flames which consumed the Praetorium were continuous with those which consumed the buildings of group (1). The two fires were quite distinct. Having set fire to the Praetorium, near the Forum, the mob proceeded to the Augusteum and set fire to the Entrance of the Palace (cp. the words of Lydus).

We now come to the statement of the Paschal Chronicle, which, as it is generally read, is absurd. ‘On Friday the demes went to the Praetorium and set it on fire; and the roofs of the two imperial houses were burnt, and of the Praetorium only the archives (μόνον---ὅπου ta oxpiva). For a north wind blew and chased the fire out from the Praetorium, and the bath of Alexander was burnt, and the Xenon of Eubulus in part and St. Irene,’ &c.

As we have seen, the Praetorium was not near St. Irene and the other buildings mentioned.? These edifices were north-east of the Praetorium; the Praetorium was not north of them, as the sense of the passage, thus read, would require. The mistake lies in the division of the sentences; there should be a full period after the words ἔξω tod πραιτωρίου ἐδίωξεν. ‘Only a part of the Praetorium was burnt, for a north wind blew the flames away from it [down towards the harbour of Sophia]. And [a totally different con- flagration] the bath of Alexander, &c.

My interpretation may be supported by the notice of Zonaras, who is

1 He mentions the δεσμωτήριον, but not as Prefect of the East (Reichsjustizministerium),

burnt. See below.

* Kalligas on the strength of this, neglecting all other topographical data, represents the fire as spreading from the Praetorium to the adjacent Palace of Constantine (Great Palace) p. 340. But his study is useless so far as topography is concerned,

% Schmidt, op. cit., p. 62, thought that the πραιτώριον here meant was that of the Praet.

not of the Prefect of the city ; and he is silently followed by Kalligas, p. 344. There is no founda- tion for this view. It may be noted that Kalli- gas seems to have used the plan of Schmidt, which is hopelessly astray, but has two redeem- ing features—the distinction of the Forum from the Augusteum, and the position of the Prae- torium (of the Pref. of the city) near the Forum. In the latter point Kalligas deserts his guide,

THE NIKA RIQT. 117

using a different source. He mentions the detail that soldiers set fire to houses from which men and women were assailing them with stones, sherds and every missile that came to hand; that a strong wind blew, carried the flames, and burnt to ashes many fine buildings. He then mixes up groups (1) and (2). But I think we are justified in inferring that the conflagration of group (2) was due to the firing of houses north of the Xenodochion of Kubulus, the fire being propagated by the same north wind which averted the flames from the Praetorium.

The conclusion is that on the night of Jan. 18, the demes, wroth at receiving no answer from the praefect, set the Praetorium on fire. But it was only partly burnt ;! and on Friday Jan. 16, they again hurled brands into it; but this time the north wind hindered the attempt from being more than a partial success.

§ 20. It is manifest that Theophanes has here ventured to exercise a very unusual independence of judgment. On the strength of his own knowledge of the topography of Constantinople, he has permitted himself to alter what he found in his source. He found the burning of the Praetorium mentioned first in close connexion with that of the buildings of the Augusteum, and secondly in apparent connexion with that of the buildings north of St. Sophia. Rejecting these (only apparent) connexions as inconsistent with the facts of topography, he took upon himself to establish a juxtaposition between the Praetorium and the buildings of group (3) which are actually near it.

Theophanes has also taken another liberty with his source. St. Sophia was the connecting link between groups (1) and (2), since it formed the north side of the Augusteum and was next-door to the Sampson. Its conflagration (Malalas; Chron. Pasch.) was connected with the conflagration of group (1) ; but Theophanes has transferred it to group (2). He seems to have thought it more natural that the fire should have leapt from the Sampson to the Church, than from the Senate to the Church?

§ 21. A word may still be said on the third conflagration (Jan. 17) which begaa with the Octagon, reached the neighbouring church of St. Theodore, and spread to the Mesé, consuming among other buildings the Lausus-palace and St. Aquilina. It is to be presumed that the fire was spread by the same north wind which blew the day before. This suggests (a) that the Octagon was not north of the Basilica, otherwise the flames would have caught the

1 Procopius (p. 120) mentions that at the be- ginning of the revolt the rioters went to the δεσμωτήριον and loosed the prisoners. It was the δεσμωτήριον of the Praetorium. The fact that Procopius does not say that it was burnt down may be reconciled with Malalas by sup- posing that only a small part was burnt; so that it was food for flames again on Friday.

2 The Continuator of Zacharias of Mytilene has a curious notice, He places the burning of St. Sophia after the proclamation of Hypatius and says that it was set on fire (apparently by Justinian’s adherents) in order to disperse the people. There is clearly a confusion with the Octagon which was set on fire by the soldiers,

118 THE NIKA RIOT.

Basilica ; and (ὁ) that St. Theodore was south of the Octagon. Now if, as is probable (see above § 15) the quarter of the Chalkoprateia was north of the Basilica, it seems certain that the street along which the Emperor proceeded, when he turned to the left from the Mesé to reach the Chalkoprateian Church, ran between the Basilica (on the right) and the Octagon (on the left). As the Lausus marked the place where the street abutted on the Mesé, that palace would be south of the Octagon,—the Church of St. Theodore (con- iecturally) standing between them. If these inferences are right, the fire first reached St. Theodore, then Lausus, then ran along the northern portico of the Mesé, taking St. Aquilina in its course, and finally crossing the arch at the Antiforum; but meanwhile it might already have been blown across by the wind to the southern portico, directly from the Lausus.

§ 22. I ought to add that, so far as the notices of the Nika-riot are concerned, they seem to me to be reconcilable with the position of the Palace of Lausus either north or south of the Mesé; for we know (cp. especially Lydus) that part at least of the southern side of the Mesé was burned as well as the northern. I therefore do not agree with Strzygovski (see above § 17) that the facts of the Nika-riot taken along with the anonymous writer of the Patria prove that the Lausus was north of the Mesé. It is the arguments adduced by Bieliaev from the De Cerimoniis that seem to me to carry weight.

On the other hand, the facts seem rather to point to the conclusion that the Praetorium was on the south side of the Mesé; otherwise, lying in the line of the fire which swept continuously from the Lausus to the Forum, it would have been presumably mentioned in connexion with this conflagration. Theophanes confirms this by the form of his erroneous notice. “The Praetorium was burnt, and the porticoes from the Camara of the Forum of Constantine up to the Chalké.” The circumstance that the Praetorium was on the south side of the Mesé suggested the description “up to the Chalké” which would strictly be only appropriate to the southern ἔμβολος.

§ 23. The course of the events of the memorable 8 days, Jan. 11-19, A.D. 532 may then be arranged as follows:—

Sunday, Jan. 11. “Acta διὰ Καλαπόδιον in the Hippodrome. Altercation of Justinian with the Greens.

In the evening a number of criminals, both Blues and Greens, are executed by the Prefect of the City, clearly in consequence of the scene in the circus and with the political purpose of showing the Emperor’s impartiality to both Demes.

The rescue of a Blue and a Green to the Asylum of St. Laurence.

[The interval of a day gives the Demes time to concert joint action to obtain the pardon of the two condemned men.]

Tuesday, Jan. 13. Horse-races in the Hippodrome. Vain appeal to the Emperor for mercy and open declaration of the union of the Prasino- venetor.

THE NIKA RIOT 119

In the evening new demand for reprieve from the Prefect of the City. On receiving no answer, the Praetorium is attacked and set on fire, and the prisoners are let out of the Praetorium prison.

Then the rioters march tothe Augusteum to attack the Palace.

Conflagrations in that quarter during the night and following day. (For the buildings burnt see above § 19).

Wednesday, Jan. 14. The riot which had begun with a demand for a reprieve now develops into an insurrection against the oppression of the administration. There is an outery against John, Eudaemon, and Tribonian. Justinian yields, but it is too late. The insurgents are determined to depose him.

The rush to the house of Probus, which probably took place on this day, is significant. Hypatius and Pompeius who were in the Palace could not be got at; so the people sought Probus. This incident seems to mark the stage in the riot at which the overthrow of Justinian became the object of the rioters.

Thursday, Jan. 15. Belisarius and his Heruls and Goths issue from the Palace ; fighting in the streets. ΠΩ the intervention of the clergy mentioned by Zonaras.

Friday, Jan. 16. Second attack on the Praetorium. Fighting continued ; conflagration breaks out in streets north of the Xenon of Eubulus, ne is blown southward by north wind. (For buildings burnt see § 19).

Saturday, Jan. 17. Fighting continued. Rioters occupy the Octagon. Soldiers set fire to it, and the conflagration spreads south and south- west (see ὃ 19).

Evening, Hypatius and Pompeius leave Palace.

Sunday, Jan. 18. Before sunrise Justinian appears in the Hippodrome. His oath before the assembled populace. The solemnity is a failure, Hypatius is proclaimed, and Justinian thinks of fleeing. Council in the Palace, at which the view of Theodora prevails.

The suppression of the reyolt by the massacre in the Hippodrome.

Monday, Jan. 19. Execution of Hypatius and Pompeius, before day-break.! (According to the Continuator of Zacharias of Mytilene, Justinian wished to spare them but Theodora interfered; “ swearing by God and by him, she urged him to kill them ”).

J. B. BURY

ΝΥ night,’ Victor Tonn., ep. above § 4.

120 THE MANTINEAN BASIS: A NOTE.

THE MANTINEAN BASIS: A NOTE.

IN a paper published in this Journal last year (xvi. p. 280) I proposed an arrangement of the sculptures of this basis. Unfortunately I overlooked a very able paper on the same subject by Dr. Amelung, Die Basis des Praxiteles aus Mantinea, Miinchen 1895, in which certain observations of a technical kind in regard to the sculptures are made, observations which must be carefully con- sidered before any arrangement of the basis can be regarded as established. 1 give a translation of these observations.

‘ All three slabs are on their left end cut off by a straight line, and have there at bottom a vertical dowel-hole, to receive a dowel rising from below. Thus all on this side joined other slabs. Both the slabs with figures of Muses on their right end have similar flat surfaces. The Marsyas slab is different. At its right end the marble projected above and below so far as the mouldings are concerned, but these projecting mouldings are roughly cut off. The sur- face of the end between the projections is not perfectly flat, but follows a slight curve. It is evident then that this slab at its right end met neither of the Muse-slabs. Clearly the cornice and basis projected from the front of the basis to the side, and were roughly cut away when the slab was fitted into the pavement of the church. Looking next at the back of this slab, one sees that its whole surface is roughened, except by the top, and by the left end where the cornice protruded, where there is a smooth border of the breadth of 6 or 7 centimétres, that is, of the measure of the thickness of a slab. The Muse slabs shew this smooth border only at the top.’

‘We are thus compelled to place the Marsyas slab on the front of the basis, at its right end; a horizontal slab bordered it at top (at the back), on the right a slab ran from it at right angles (for the side of the basis). For the determination of the position on the basis of the two Muse-slabs two further facts are of importance.’ |

‘On careful examination of the moulded bases of the three slabs, we observe that that of the Marsyas slab differs somewhat from those of the others. In the former case the upper surface of the moulding projects further, and the channel beneath it is cut deeper, so that the shadow is stronger. So great is the difference that it seems impossible to place one of the Muse-slabs on the left of the Marsyas slab. The reason which suggests itself for the difference in the two profiles is tlat the Marsyas slab was meant for the front, the Muse slabs for the sides of the basis; in the latter situation refinement of the profile of the basis would scarcely be observable.’

THE MANTINEAN BASIS: A NOTE. 121

‘The second fact confirms this view, and shows the exact position of each of the Muse slabs. Any close observer must be struck with the fact, that on one of these slabs the standing Muse on the left, on the other the seated Muse on the right, is further distant from the end, in fact further distant from it than the corresponding Muses with cithara and flutes by about the measure of the thickness of a slab. So we cannot doubt that we should place the slab with the seated Muse at right angles from the right of the Marsyas slab, and we must suppose that a slab has been lost which contained the three missing Muses and occupied the left half of the front of the basis ; while the slab with the three standing Muses ran back at right angles from this.

‘Thus we reach a basis of 2°70 m. in width, of 143 m. in depth, and ‘96 m. in height, quite sufficiently large to support a group of three life-sized figures.’

Dr. Amelung’s observations show how dangerous it is to discuss monu- ments on the evidence of casts and photographs, and without a leisurely study of the originals on all sides. I argued fairly on the facts before me, but was not acquainted with other facts of material importance. I am unable from the casts to control the exactness of his statements, but presuming, as we are no doubt entitled to do, that they are accurate, it is very difficult to resist his conclusions. His arrangement, therefore, seems entitled to supersede those of M. Fougeres, Dr. Waldstein, and others, including my own suggested arrangement. I am quite ready therefore to withdraw such part of my paper as is concerned with the position of the slabs on the basis, as well as my contention that there is no ground to assume a slab to have been lost. The part of my paper which discusses the restoration of the group which stood on the basis is not affected.

Percy GARDNER.

122 EXCAVATIONS OF THE BRITISH SCHOOL AT. MELOS.

EXCAVATIONS OF THE BRITISH SCHOOL AT MELOS.

THE SITE OF THE ‘ THREE CHURCHES.’ [PLATE V.]

Tue field bearing the name of ᾿ς τὰς τρεῖς ἐκκλησίας is one of several small sites which Mr. R. C. Bosanquet and I examined in April 1890. It is marked E in the map, J.H.S. xvi. p. 348. We were afterwards joined by Mr. C. R. R. Clark, architect to the British School, to whose skill and diligence is due the plan which accompanics tlis paper. In

describing the course of the excavation I have had the free use of Mr, Bosanquet’s day-book. The inscriptions from our site are Nos. 21, 39, 40, 41. of Mr. Cecil Smith’s ‘Inscriptions from Melos, published in the present number of this Journal. Our inscription 4 will appear in a second series. Some remarks on the early cruciform font which came to light here will

| MELos. SITE OF THE THREE CHVRCHES. ee ΕΒ τα

BAPTISTERY.

39 B

GAZ Ze

ih See

Hi

SPACE Pe i iar Bi ss AS aa ZZ γ δὲ 7 Ν © i THRESHING \ PITHOS | \ FLOOR I \ | \ ἡ \ yi ὃν 7 S 4 A. SK ἘΞ a Ss. ξξέξξι "τὰ ςς ὁ —— =| METRES Tos Seb ene) Ιο Ι5 20 [ ἘΕΕΕΕΕΕΕΕΡΕΞ--- - -͵π - -π' 'ὖὁὃΘ 90’. ἢ FEET lo Si ο Ιο 20 30 40 5c

J. HOS. VOL. χνι". (1897) PL. Vv.

GROVE

EXCAVATIONS OF THE BRITISIT SCHOOL AT MELOS. 123

be found in the British School Annual ii. p. 161, 168, in a paper on the Churches of Melos, by Messrs. H. M. Fletcher and Τὸ, 1). Kitson. The ‘Three Churches’ field was carly included in our list of likely sites. Its prominent position, and ancient retaining walls, as well as the reports of frequent finds of ancient masonry and marble, all pomted to an important public place having existed on the site, which thus afforded exceptional inducements to excavation. Work began on the 10th and lasted till the 30th April. Owing to the confined nature of the plot only 12 or 15 men were employed.

The site in question occupies the south part of a kind of saddle lying between the cast and the west acropolcis of Mclos, and forming the highest point of the ancient city if we except the citadels. The narrow and stony road from Trypete passes westward through the ancient city wall near the east portal and just to left of its protecting bastion, skirts our ficld on the north, and then turns steeply southward in wide zig-zags past the theatre to Klima and the sea. Between our site and the road on the north is a little olive orchard. Immediately past it a road branches off northward from the main one along the east edge of the saddle and branchies again up to Plaka and Kastro and down to Tramythia.

The field itself descends gradually to westward and more slightly to the south, where it projects upon the valley, abutting on an ancient terrace wall overlooking the theatre. Our site commands a complete view of Klima valley, the gulf, and the southern reaches of Melos with the peak of Prophet Elias prominent in front. The west acropolis hides the prospect westward, with Eremomelos, from view. To northward we have again a glimpse of sea between the west citadel and the still loftier Kastro, the expanse of blue being broken by the red cliffs of Phourkovouni.

In the south-east corner of the field is the block of grey rock inscribed with a dedication to Zeus Kataibates (No. 21, p. 8 supra). The presence of this inscription led us to connect the tradition of three churches with the exist- ence of early temples on the site. On the other hand nothing was noticeable on the surface except a few fragments of marble pavement and of statue bases of late workmanship dug up and broken by the present proprictor while improving his field. The rock crops up here and there, and the barren patches in the growing corn which were visible in the spring time testify to scanty soil below.

Our excavation was begun at the altar with the BATA inscription and almost simultaneously at three points along the north dyke of the field.

While absolutely nothing of early date was found in the vicinity of the altar, and in other parts only poor Byzantine walls with foundations on virgin soil at not more than three or four feet down, the excavation in the north part of the field produced two bases with votive inscriptions.

The first, Inscr. No. 39 (p. 17 supra), a square marble base, dedicates to the gods a statue of Agrippina, the wife of Germanicus, the numerous marks of fixing on the top suggesting that the statue was of bronze. The

124 EXCAVATIONS OF THE BRITISH SCHOOL AT MELOS.

block, set in very hard cement, lay three feet down on its left side, the cement covering the left hand letters. The inscription face was turned to the north, the top of the base with the marks of fixing to the east.

Directly south from the position of the Agrippina base and immediately west from a long architrave block, A, we discovered at less than a foot from the surface a round pedestal, like the drum of a column, with the Kleonymes inscription (No. 40, p. 18). The base lay with its top to the north, the inscription being to the west.

The discovery of these bases and the fact that in the adjoining oliveyard there still hes the base of a signed statue of Roma (Loewy, Jaschr. Gr. Bildhauer, No. 217) seemed to mark the site as that of a public place in antiquity, where such statues as those referred to might suitably have stood.

Eastward from the Agrippina base and almost in a line with it excava- tion subsequently opened up a series of blocks forming part of a wall running east and west. West from it a number of marble slabs and one column, unfluted, lying in disorder, once formed part of the same wall, but were disturbed by a party of villagers who, setting to work in a manner natural to miners, made a tunnel from the neighbouring field under the road into ours, unearthing at the time the grey granite columns now built into the field- dyke further north.

East and west from the inscribed column and in line with it a second line of wall was next opened up, the first noticeable feature being a marble block, C, like part of a stylobate resting on a step projecting to north of it. East- ward from it were uncovered two very large blocks, A and B, of purple grey limestone, apparently parts of a former architrave. This wall at the west end turns south and then east again. All these lines of wall are similarly com- posed of the materials of previous buildings, the bases themselves having been employed for the same purpose.

The opening up of so many lines of wal] of miscellaneous materials was calculated to discourage, when all at once we came upon what proved to be the surprise of the excavation. Immediately to the west of the inscribed column and in a line with it and the architrave stones was uncovered a square marble block with cement on the east face of it. West of it were a small stone and a block of cement, and west of these two long slabs of shale lying side by side east and west. On raising these slabs we discovered below them a scrap of marble drapery and below this a long wedge-shaped frag- ment of drapery with deep cut folds. More marble having become visible projecting westward from under the square marble block, this block was raised and proved to be the upper part of a pedestal with mouldings. It was placed with the narrow part below. The upper surface is all grooved out hollow. The straight edges below the mouldings measure, front and back, ‘€3 m., the sides ‘58 m. The marble projecting westward from below it

* It is possible that some of the marks of be costly, and a block might often be re-used. fixing are due to an earlier use of the same Cf. the Samokles inscription at Katergari, which block : marble being unknown in Melos would _ seems to be a palimpsest.

EXCAVATIONS OF THE BRITISH SCHOOL AT MELOS., 125

turned out to be a round pedestal cap with marks of fixing for the feet and having a diameter of ‘63 m. below which exactly corresponds with that of the inscribed column. It was removed, and on further clearing the drapery a much larger fragment lying more to the west came into view. The trench was then lengthened to the west and widened, upon which a wall of cement marking the south side of a porous sarcophagus was disclosed.

The marble on further clearing proved to be (1.) a colossal torso (Fig. 2) in the type of Apollo leaning on a column. The head is broken off at

Fig. 2.

the neck, the break looking quite fresh as if the mutilation had just been perpetrated before depositing the marble where found. The arms and legs are also missing.

After the statue had been raised the sarcophagus, which was closed with slabs of slate, was opened. It had never previously been disturbed, but contained only a layer of fine dust half an inch deep all over the floor, which was pierced in three groups of holes for the escape of moisture. At the west end of the sarcophagus lay a curiously cut stone, evidently

126 EXCAVATIONS OF THE BRITISH SCHOOL AT MELOS.

placed in the tomb as a support for the head but certainly originally designed for another purpose. The stone is ‘47 m. long by “35 τη. broad. Its upper surface is fluted, and has a long slit in the middle, which widens below, communicating at each end with a low square opening grooved out of the lower edge of the stone. At each end is a dowel-like hole above the openings.

These articles are common in Melos, We found one very similar at the top of the descent to Tramythia while excavating a field belonging to M. Giclerakis, British consular agent at Melos. The two stones appear in Fig. 1, our example having the lower side up. The fragment of a sinular stone with the fluted upper surface was found in the site (D) excavated

Fic; 9.

below the theatre. A further fragment with the flutings but oval in shape and with the dowel-holes going up from below at each end I noticed on the road dyke just above Kepos, at the south-east end of the island, at a farm- house which has numerous stones from some Roman building. I have picked up two examples on the road to Klima of wedge shape without the flutings, without the slit in the middle, but with the dowel-holes at eaeh end, the stump of the iron cramp still adhering in one case.

Marble having become visible at the head of the tomb this part was cleared and gradually several fragments came into view.

(II.) Lying on its back the torso of a man, draped, from just below ἣν paneer to below the hips, the left hand holding the folds of a mantle

ig. 3). (IIT.) Beside (11.) west of the north-west angle of the sarcophagus, the

EXCAVATIONS OF THE BRITISH SCHOOL AT MELOS. 127

colossal right leg of a statue, draped, from above the knee to the ankle (Fig. 4).

The colossal right foot of a statue, sandalled, broken off at the ankle, found wedged in between (II.) and (III.) and the head of the tomb.

Below these, virgin soil was reached without anything further having been found.

On further digging through a few inches of soil below the spot where torso (I.) was discovered, marble again became visible.

- (IV.) The lower part of the torso of a woman, fully draped, from below

the waist to near the ankles (Fig. 5).

(V.) Lying beside (IV.) the torso of a man from the neck to the hips, the upper part undraped, the left hand holding the folds of a mantle (Fig. 6).

The manner in which all these mutilated statues were packed round what is evidently a Christian tomb and below the foundation walls next it, clearly reveals the intention to bury them out of sight, and suggests at once that we have here the remains of a very early Christian Church dating back to an age when those statues were still held in honour by part of the community.

The quest was now pursued further west between the stylobate block and the corner stone marking the turn south of the wall. Just west of the stylobate stone against the south-west angle of it and four feet down was found

(VI.) The lower half of a female statue (Fig. 7), over life size, and fully draped, with a fringed mantle, apparently knotted above the waist. This

128 tXCAVATIONS OF THE BRITISH SCHOOL AT MELOS

dress would seem to identify the figure as Isis, or more probably as a portrait of a priestess of Isis. On its back was cut a cross which, on the statue having been raised, appeared inverted. East of the stylobate stone was found fairly in the line of the wall a square marble block with a votive inscription, No. 41, which on being removed showed a similar cross below. These crosses point to a ceremony of consecration of the ‘idols’ before using them in a sacred building, one statue and one inscription being so marked as represen- tative of all. As both the statue and the inscription form an integral part

ἩΤΟ ἡ.

of the walls we have here additional evidence pointing to these having been part of a Christian building}

Immediately west of the corner-stone, D, and also at a depth of about four feet, after a fragment of drapery from the waist of a statue had been removed, appeared

Mr. Bosanquet points out to me that Pro- baptistery and saw a small Christian church kesch, Denkwiirdigkeiten, i. 537, describes the here as late as 1825.

EXCAVATIONS OF THE BRITISH SCHOOL AT MELOS. 129

(VII.) A large marble fragment, the torso of a female statue, draped, from below the waist to below the knee (Fig. 8).

Nothing further in the way of sculpture was found along this line of wall and the search was pursued east-ward from the sarcophagus.

After the whole system of walls had been planned the architrave block, A, was removed. The huge block so evidently formed part of the foundations that one could not have expected anything below it belonging to the same line of wall. Yet here was another surprise. At one foot down we first came upon the upper member of a square statue-base of marble, with mouldings, turned up side down, one side broken off. The straight edges below the mouldings measure, front and back “70 m., the sides ‘56 m

Fie. 8. Fic. 9.

It is very similar to the one previously found above the colossal statue, only it shows the hollows for fixing the feet on the top. On this having been removed there next came into sight almost directly below it,

(VIIL.) Part of the right leg of an over life size statue, draped, from the middle of the thigh to the ankle, there being also part of the left leg from above to below the knee (Fig. 9). The fragment had evidently been buried out of sight below these foundations at the time they were laid but, as the soft intervening earth shows, formed no constructive part of the wall itself.

The line of wall north of this was then examined and the square marble block, 4, to the east of the Agrippina base was found on having been raised to have on its lower face a dedication inscription— ... cov ᾿Αριστέα---

H.S.—VOL. XVII. K

130 EXCAVATIONS OF THE BRITISH SCHOOL AT MELOS.

4 (2) lines. The top of the inscription was to the east. The stone stands ‘72 m. high, measures front and back face ‘63 m., the sides “52 m.

Following and deepening the north and south trench in a line with the wall running south from the corner-stone, D, we encountered tough clay just west of the line of wall. Low down in this was a curious round drum, like the fragment of a coarse column, set in clay and stones, which gave it the look of standing on a stylobate. It was about four feet below the level of the adjacent blocks. Close by in the same trench was a Byzantine pilaster, upright. The clay looked natural, but the trench had been daubed on the sides and floor with it. Similar gullies were opened up further east. All these rifts seem to be nothing more than natural chasms in the hill-top.

To the south of the main building an early baptistery made its appear- ance a few inches under the soil. It is in shape like that discovered by us at Kepos,' only it has a rounded parapet at the top of each pair of steps. The outside is faced with narrow slips of marble arranged like wooden planks, following the curved wall in a polygonal line. The west lower step has a round opening below, which communicates with a tile-lined drain leading from the baptistery to the clay-lined gully mentioned above and explaining the same.

The proprietor remembers breaking up four large blocks of grey stone which steod one at each corner of the baptistery. They may have supported four columns carrying a baldachino.

Just at the close of the excavation season we cleared a line of wall running from the threshing floor, shown on the plan, northwards to the road- dyke. On the road side of the dyke a squared stone is visible in position exactly in line with this line of wall, and projecting in the same way from below the north dyke of the road are two similar squared stones, the one resting on the other.

GENERAL CONCLUSIONS.

To sum up, the results of our excavation were broadly what might have been expected from the indications previously existing on the surface of the soil. Very few traces came to light of early pottery, one fragment of Geometric style and one of Corinthian only being found: in fact nothing which can have belonged to a period antecedent to Roman Imperial times. The question now remains, what did the site at that period represent? As I have already remarked, the discovery of so many statues and bases would seem to indicate some important public place, if we may assume that these stood originally near the spot on which we found them? The size of the statues

* See Messrs. Fletcher and Kitson’s paper rocky virgin soil, filled with soft earth, and already referred to. much smaller than the block over it. ? Under block E was a square sinking in the

EXCAVATIONS OF THE BRITISH SCHOOL AT MELOS. 131

and the good preservation of their surfaces suggests that they were not brought far, and they could not have been conveyed any distance without having to be brought up hill. The loss of heads, arms, and legs is probably due to wanton mutilation prompted by religious fanaticism. Such portions of a statue were frequently carved in separate pieces of marble, especially in Melos after the second century B.c., and may very well have been built into the upper courses of the Church, to disappear when the walls were pulled down. It is quite possible that the walls were largely built of marble fragments, and that it was in order to burn these fragments into lime that the walls were demolished, the heavy foundations only being left in the ground. The proprietor has often pulled out and broken up large blocks, marble and other, which he has encountered while improving or cultivating his field. In this way whole lines of walls may have been obliterated. In particular, he took up the pavement of what he calls an entire Roman house in the north-west corner of the part which we dug, and this may quite well have been a porch or other public building facing the road which led from the east town-gate discovered by us.

Dedicatory inscriptions, portrait statues, traces of public buildings here and in adjoining fields, all point to the site as that of the ancient Agora; this opinion has already been expressed by Weil who in Athen. Mittheilungen, Vol. I., p. 247, publishes two more inscriptions recording the dedication of statues from the same site.

Its natural position favours the view. Firstly, it is a fairly large level space at the highest point of the ancient city which stretched down steeply on either side to south and north towards Klima and Tramythia. Secondly, it is next the east gate and the interior of the island whence came all the land produce which could be disposed of here without any further steep ascent or descent. There is every reason for believing that the prosperity of the Melians depended largely on the rich internal resources of the island, and if so we cannot conceive their Agora either at Klima or Tramythia, involving, as such a position would do, much steep descent and ascent. In the late Roman times when the varied natural resources of Melos were probably more exported, and when the harbour works were erected of which the remains still stand in the sea, there may have been, as the remains of a late Stoa there suggest, a second emporium at Klima, But this mart could not have entirely supplanted the Agora above, for all its most characteristic features are distinctly Roman.

There is further the evidence afforded by converging roads. Firstly, there is the great road from the east traceable for a long way outside the bounds of the ancient city, and passing through the town-gate excavated by us. Secondly, we have a cross road into the field of Antonios Anthroulakis skirting the so-called west or little acropolis on its east slope and going north- ward between the little acropolis and the square ‘ temple’ structure in Kallitza Komi’s field in the direction of. the descent to Tramythia. Traces of it are visible in the form of a reddish pavement on passing our site just’ as we begin to make the descent to Klima, projecting from under the dyke on the right

K 2

132 EXCAVATIONS OF THE BRITISH SCHOOL AT MELOS.

hand side. Traces of a precisely similar pavement in a line with it due north, project from below the south dyke of the first field on the Tramythia slope. It points upward and northward towards Komi’s ‘field in a direc- tion between the ‘temple’ structure and the little acropolis, and it points downwards towards Tramythia. Thirdly, there is the roadway, further east, laid bare for some yards in a field lower down towards Tramythia belonging to M. Gielerakis, built of smaller stones than the preceding.!

Marbles were found in the olive-yard north of the Three Churches of a character and era tending to show that they belonged to the same complex of buildings. It is further reported that a column with γρώμματα was found at the same time and sold. Near the same spot was found and now stands inverted against the road dyke on the field side a fine marble’ drum, unfluted in its main body but having a ring of fluting at its narrow end to which an upper fluted drum must have fitted. Such columns unfluted below were the fashion for stoas and other public buildings from Hellenistic times onwards. The unfluted part was carried to a height which placed the delicate fluting beyond the reach of ordinary contact, and our drum, 1:27 m. in length, accordingly requires another unfluted drum below it, for which it shows a large square dowel-hole. It thus immediately suggests a column from the colonnade of some stoa in connexion with a market-place. Thus the Agora must have extended northwards considerably beyond the range of our excavations, but its central part probably lay on both sides of the ancient road which led from the interior through the east gate.

With the lack of a Pausanias for the islands, we have particularly to regret that excavation has yielded no monumental evidence as to the nature of the site in Greek as distinguished from Roman times. We have to fall back on the following considerations. 1. The argument from the natural position of the site as occupying part of a large ievel space at the highest point of the ancient city next the east gate and the interior of the island has double force for those eras, preceding external invasion, when public prosperity in an island, historically known as emphatically self- sustained and independent, must have largely depended on the development of its internal resources.

2. There is, further, the probability of historic continuity in the site of the Agora, If all the monuments point to the site as the Agora in Roman times, we have no reason, without evidence, to assume the possibility of the Agora having been elsewhere in Greek times, since even without the evidence of monuments we have the probability that the Romans used an agora they found before them.

3. The position of our site entirely agrees with the one classical reference to the Melian agora—Thue. v. 115, 11—15. εἷλον δὲ καὶ of Μήλιοι τῶν ’AOn- ναίων τοῦ περιτειχίσματος TO κατὰ τὴν ἀγορὰν προσβαλόντες νυκτός, Kal ἄνδρας τε ἀπέκτειναν καὶ ἐσενεγκάμενοι σῖτόν τε καὶ ὅσα πλεῖστα ἐδύναντο

* This evidence from converging roads was collected by Mr. Bosanquet.

EXCAVATIONS OF THE BRITISH SCHOOL AT MELOS. 133

χρήσιμα ἀναχωρήσαντες ἡσύχαξον. Duker, Arnold, Krueger, and Classen, comparing i. 62, 1 and iii. 6, 2, all agree in interpreting the agora here men- tioned as a kind of military agora belonging to the camp of the Athenians. The following considerations make this interpretation doubtful. Firstly, the passages compared do not afford any real ground for assuming the existence of any such thing in the Greek army of the fifth century as a military agora, and this is in agreement with the silence of Xenophon and other historians. Secondly, in the passages cited an agora is established only where the army is on such terms with the inhabitants of a district or town that they can buy of them, as was the case with the Peloponnesians at Potidaea (Thue. i. 62, 1), and the motive for holding an agora outside the walls there was, as Arnold himself points out, to deprive the men of all excuse for neglecting their posts by straggling into the friendly town. In the case of the Athenians before the walls of Melos all motive for even this kind of agora is taken away, for not only have they no intercourse with the citizens, but they have the whole country in their own hands. Thirdly, though the expression ἐσενεγκάμενοι does not necessarily coincide with the ‘frumento et aliis rebus a Meliis raptis ’ by which Duker interprets it, the possibility is not excluded that part or whole of the corn and other provisions conveyed into the city by the Melians may have been plundered from the Athenian camp without involving the presupposition that the words of Thucydides refer to a military agora there. Fourthly, topographical considerations are all in harmony with the interpretation of the passage as referring to the agora of the Melians.! Between the east citadel and the declivities and cliffs that descend to the gorge below is the only fairly level and easily accessible space over which a road from the interior could be made into the city. The road and the gate we have. The Athenian watch must have been strongest just at that point where they thought it most probable the Melians would attempt a sally, namely, at the east gate. If it had been only a matter of getting out any way the Melians could possibly have managed an exit at the precipitous parts above or below, where the watch of the Athenians from the nature of the ground would have been less vigilant. But since it was a case of se- curing provisions for the famishing citizens they must make their desperate venture in the direction in which it was not only most possible to secure these but to convey them into the city. This was at the east gate, where the ground is level and in direct connection through the high road with the interior, where alone for Melians and Athenians alike provisions were to be had. Our Agora is next the gate, and τῶν ᾿Αθηναίων τοῦ περιτειχίσματος TO κατὰ τὴν ἀγοράν is just where the narrative of Thucydides requires it to be.

DuNCAN MACKENZIE.

i ὉὉΠθΘΦΦθθόόὁόθὖθῤ΄͵͵ Ξ aioe —— .--..-..-.--... ---- —_—— ocedl —_. i=

1 The view propounded is the same as that his discussion of the Melian fortifications in independently arrived at by Mr. Bosanyuet in the Brilésh School Annual, ii. p. 81.

134 EXCAVATIONS IN CYPRUS IN 1894.

EXCAVATIONS IN CYPRUS IN 1894.

EARLY in 1894 the Committee of the Cyprus Exploration Fund offered the small balance which remained from the excavations of 1891, to the British School of Archaeology in Athens, for use in Cyprus if possible. Asa student of the School was then watching the excavations which were being carried on at Amathus on behalf of the British Museum, this sum was applied to defray part of the cost of several small excavations, the principal object of which was to test certain theories current in Cypriote archacology; though some new ground was broken incidentally.

Five sites were examined in all; none of them exhaustively, but all with distinct and definite result.

I—AGIA PARASKEVI (Wicosta District): ΒΒΟΝΖΕ AGE NECROPOLIs.

The celebrated Bronze Age Necropolis which occupies the edge of the plateau §.W. of Nicosia seemed the most suitable site for making practical acquaintance with the Bronze Age of Cyprus, and for verifying previous observations, with a view to the re-organisation of the Cyprus Museum which took place in the course of the summer.

Fourteen tombs were opened along the northern edge of the plateau, half a mile north of the Church of Agia Paraskevi, to the west of the Larnaka road, and between it and the stone quarries in the direction of the village of Agii Omologitades. The tombs in the surface of this part of the plateau were found nearly exhausted by Dr. Ohnefalsch-Richter’s excavations in 1883-4: but enough evidence was collected to illustrate the general character of the site. Tomb 12, as explained below, was an intruder of Hellenistic or Graeco- Roman date. Similar tombs have been opencd on the low hills west of the road to Strobilo village.

The statistics of the finds may be expressed diagrammatically as follows :—

EXCAVATIONS IN CYPRUS IN 1894. 135

Gold, stone bead, figurine, &c.......

x

Register No. of ‘Tomb, δι, λιν θ vibe ld Dl Souk tele 18]

|

Red Ware plain ..... Ba ES a Pet A ee LP re eg ete . x | molio formants ὁ νον νἀ κα bX BC dak a ESE. δὲ

incised ornament, ............... See a x ¥ | Painted White Wave ...... Pate atte Sinn 56 TSC eee Bagorvings Wares ss fesices.. /seseapoeees| sueu% x ee Sp) ΉΥ | i: Painted White Slip Ware ............ 1 Χ ον aah OS beste Sy ct. xi concn] Me Ἃ RRO ZOM crete de oe Bi wich eins, eco ance ars cal x DT COLO WE ΡΥ" ΠΝ tt x x |

14 contained horse-teeth. 8 1 imitation.

As the table shows, the tombs fall into two classes: those in which only the red polished ware) is found, and those which contain, in addition, black slip ware, painted wares, and the numerous more or less fancy fabrics which accompany the latter. In these tombs the red ware is almost always of inferior style and fabric. It is also only in the second class that bronze—or rather copper—implements become at all frequent. The following tombs deserve more detailed description.

Tomb 1 was a small cave about 3 ft. square on the east slope of the plateau overlooking the new olive plantation towards the highroad to Larnaka. It was found collapsed through the weakening of the roof by surface-weathering. The pottery was all found, much broken, at a depth of about 8 ft. from the present surface: it comprised red ware of a coarse and degenerate kind, of inferior clay, ill-modelled, covered with a muddy dark red slip, and often almost unpolished: one fragmentary bottle of better fabric had incised ornamentation of concentric semi-circles applied to both sides of a zone of parallel lines: also one specimen of black slip ware; and a number of bowls, flasks and tubular-spouted bottles of painted white ware: two of these were fantastically formed in the shape of horned animals. There were no traces of bronze weapons, but a number of scraps of spiral rings of silver-lead were found, much corroded, and about the size of a finger ring. Ashmolean Museum.

Tombs 2, 3, 4, 6, 7, 9, 11, contained only red ware; with the exception of one fragment of basc-ring ware extracted from the layer of crushed pottery in Tomb 4. Tombs 9 and 13 had been already rifled.

Tomb 5 was a natural cave on the plateau, of which a large part of the

1 he classification of the native pottery and in the Cypriote collections of the Ashmolean throughout this paper is that adopted in the Museum, The publication of this Catalogue Cyprus Muscwmn Catalogue (Myres and Ohne- has been unexpectedly delayed ; but the refer- falsch-Richter, Oxford, 1897), where the fabrics ences to the figures in the text will identify the in question are described and discussed in detail; various fabries sufficiently well.

136 EXCAVATIONS IN CYPRUS IN 1894.

roof (12 ft. x 8 ft.) had fallen in irregularly, and lay in large blocks embedded in the débris. The tomb however was already partly full of earth when the collapse occurred, and some of the smaller vessels were unbroken. It contained many broken vessels of fine ved ware, with ornament incised and in relief, and much painted ware; a black clay spindlewhorl, two bronze pins of type a," and nine porcelain beads, all spherical except one which was narrow and spindle-shaped,

Tomb 10 was under the very edge of the plateau some distance further west again: it must have been a natural cave like 7ombs 5 and 14. The roof had collapsed, but the same layer of crushed pottery was struck at about 10 ft. from the original surface by two independent shafts. It contained all varieties of ved ware (Fabr. I. 1); several of the black slip ware (Fabr. I. 2) ; a number of small specimens of the base-ring ware (Fabr. I. 3), including one’ with punctured zig-zags, which perhaps shew the influence of the black punctured ware (Fabr. I. 5: οἵ, Kalopsida Fig. 4. 7. below) a small plate of a white basc- ring ware (Fabr. I. 4); and much painted white ware (Fabr. IL. 1), with frag- ments of the glossy variety (Fabr. IT. 2); and of the hemispherica) bowls with white slip (Fabr. 11. 3). No Mykenaean vases were found, though fragments of many are strewn all over the surface of this part of the plateau ; but the late date of the tomb 15 attested by the presence of one of the double-cone- shaped beads of soft stone? which are characteristic of the Mykenaean Age in Cyprus, and continue into the earlier Graeco-Phoenician Age.

The occurrence, on the other hand, of two porcelain beads of the usual spherical form shows the persistence in Cyprus of this class of imitations of the XII. Dynasty types.

Bronze was represented by the common dagger blades of type y (C.M.C., p. 53), with the midrib produced into a hooked tang; by a number of plain pins of type a@ (id.); an awl (C.M. 571); three ‘spiral earrings’ (C.M. 623 a.b.c.); and a number of the common spiral coils of thin bronze ribbon (C.M. 625). <A fragmentary object looked at first sight like part of a bow-fibula, but was more probably a pin with coiled eyelet head, (Type ὃ C.M.C. 598 p. 54). Of silver were a pin like those of bronze ; a ring of the usual unrefined silver-lead, cf. C.M. 611-614 and Lakshd tu Rit 1 and 2, below, and spiral- earrings of two close turns, exactly like those of the Graeco- Phoenician Age (C.M. 617, cf. 4119 ff). .The latter is noteworthy, and so far as I know unique from a tomb of the Bronze Age.

Of Gold there were a pair of funnel-shaped objects (C.M. 4502) with recurved lip at the wider end, which in spite of their unusual size, are almost certainly the setting of a cylindrical seal, like those of the Babylonian cylinder from Dr. Ohnefalsch-Richter’s excavations at Agia Paraskevi (1885, 1), which are also in the Cyprus Museum? A _ still more similar mounting 18

' Cyprus Musewm Catalogue, p. 53. Homer, Pl. clxxi. 14; cf. C.M.C. pp. 57, 134. * Apparently coarse steatite: v. below, Lak- For the cylinder itself νυ. K.B.H. Pl. xx. 4; sha-ltu-Rid 4 and Larnaka Twrabi 55. Bezold, Zeitschr. f. Keilinschr. 11, (1885) 191-

3. ©,.M. 4501. This tomb-group is published 193, in Ohnefalsch-Richter, Kypros, the Bible, and

EXCAVATIONS IN CYPRUS IN 1894. 137

published in Dr. Ohnefalsch-Richter’s Kypros, the Bible, and Homer ῬῚ. exlvi. 5. B; but in this latter case, as in the present instance, the cylinder itself was not found,

A Figurine of fairly fine clay, but of the rudest known Cypriote fabric! was found not far from the gold objects. It consists of a flat rectangular pellet of the shape and size of a ‘sponge-rusk’ biscuit, with the nose and ears indicated by slight projections; the hair by a row of small curls on the upper edge, the eyes and nostrils by punctured dots, and the arms by short projec- tions from the longer edges, folded forward on to the breast. Fig. 1.

Fic. 1.—PRIMITIVE FIGURINE Fic. 2,—REDWARE VASE FROM Ay. FROM Ag. Paraskevi, 1894. 10. Paraskevi, 1894. 11.

No recognisable human bones were found, except four well preserved molar teeth. This tomb fell to the share of the Cyprus Museum.

Tomb 11, a small square cave, about 5 ft. in diameter, on the slope of the N.E. spur of the site, contained only red ware of good quality, namely two large one-handled bottles, one with the serpent-ornament in relief on the neck; two large milk bowls, one with a tubular spout near the bottom, to drain off the skim-milk from under the cream; a smaller bowl, funnel-like, but with unperforated stem, and with projections on the rim; a pear-shaped jug pointed below, with two miniature jugs and a bowl set round the neck on the shoulder (Fig. 2); a two-handled vessel; three plain bowls; and an incised bottle of the black deoxidised variety of the red ware. Ashmolean Museum,

Tomb 12, the Graeco-Roman tomb above mentioned, was a surface grave, and contained three unpainted vessels of Hellenistic fabric and common forms. (C.M. 2159-61.) Cyprus Museum.

1 ΟΜ. 462 (Type I. C.M.C. pp. 27, 51). Cf. K.B.H. Ἰχχχνὶ, cxlvi. 3 B, clxxiii. 20 f.

᾿

138 EXCAVATIONS IN CYPRUS IN 1894.

Tomb 14 was a natural cave, some yards from the edge of the plateau, and south-west of Zomb 10: the entrance had been artificially enlarged, and the dromos was bath-shaped. About 5 ft. from the surface and 2 ft. from the roof of the cave was a layer of bone earth about a foot thick, with 1} ft. of clean earth between it and the rock floor. The bone earth yielded several pieces of red ware and painted ware ; and several horse teeth which had been used as burnishers, probably in the manufacture of the fine red ware.

II—KALOPSIDA (Famagusta District}: BRONZE AGE SITE AND NECROPOLIS.

While the excavations were still going on at Ag. Paraskevi, a report came of a new Bronze Age site between the villages of Kalopsida and Kuklia on the high road between Nicosia and Famagusta, about 14 miles from the latter. As this in any case lay far east of any Bronze Age site then known in Cyprus, leave was obtained at once to make trial excavations. The general results of a week’s work (April 17-23) were as follows. The site of a Bronze Age settlement was discovered, with indications of a pot factory; and the extensive necropolis was shown to represent at least two distinct periods of the Bronze Age; (A) the earlier, in which only the red polished ware was in use, when the polished surface was of fine texture and bright colour, and when bronze implements were, if not entirely absent, at all events so rare as not to come to light at all during the excavation: (B) the later, in which the red ware was very degenerate, either smeared with a loose dark red pigment, and unpolished, or, when polished, left of the natural brown colour of the clay; while painted pottery and bronze implements and ornaments were comparatively common; and foreign importations, and native imitations of Egyptian ornaments, supply a date mark, which is not demonstrably earlier than the XII. Dynasty, but on the other hand is certainly not later than the XVIII. Dynasty. (C) Finally, on the denuded surface of the necropolis, though not in any undisturbed tombs, were found one or two fragments of characteristic Mykenaean pottery.

The Bronze Age Scttlement.—The high road after leaving Kuklia village, runs nearly eastwards to Kalopsida, skirting the limestone plateau, which is here invaded by a southward bay of the marshes of the Pidias river. About a mile from Kuklia it passes a deserted farm (Daud Chiflik) on a low ridge between two small streams. The necropolis begins on the moor west of the first of these streams. The next ridge east of the chiflik is considerably higher, and is cut through by the road to a depth of 10-12 ft. In this cuttmg masses of broken pottery had been exposed; all hand-made, of a red or brown colour, like that from the tombs, and for the most part very rude. This pottery did not seem to come from tombs; but lay in a compact mass among loose earth and stones, some of which seem to retain a wall-like arrangement. What made it clear that a settlement was in question, was a well-defined layer of cockle shells, often broken, about 18 inches from the surface, and two or three inches thick: this layer could be traced for some

EXCAVATIONS IN CYPRUS IN 1894. 139

yards, and on both sides of the cutting. A little trenching on the north of the road, and close to the west brow of the ridge, revealed a fragment of wall built of unwrought stones of the size of a man’s head and under, bound together by a mud cement which was clearly recognisable, and itself full of scraps of pottery. Close to the wall were found (1) the upper stone of a ‘saddle quern, such as were commonly found all over this part of a site; two loomweights of baked clay, one (2) roughly conical, transversely perforated at the apex, and weighing 4 or 5 0z., (Fig. 4. 5.); the other (3) discoidal, with a small hole near the edge like one from Z'omb 32; (4) a small hand-made jug with pinched lip (Fig. 4. 9) 0:035 m. high, of light-coloured clay, but unpainted, (5) arude clay ladle like those found in the tombs (e.g. C.M. 26). (6-7) Two

uu nly

“any D\ was

----

VWKUKLIA

>

= | »

DAUD

se \

HANNA %

an WOM My, reg tll Oe, ty

παν τὖοι Ba Rakes, Cea ce πὸ Jon τος, cy | Seale ἢ: 400000(68 mitew ὁ inch) ! English Miles

: [9] Fic. 3.—THe NEIGHBOURHOOD oF KALoPsIDA: Based upon the trigonometrical Survey of Cyprus, and drawn by B. V. Darbishire.

massive but well-worked saucers of a hard crystalline rock (Fig. 4. 79); one (7) broken, and both still stained with a red pigment, exactly like that on the red pottery from the tombs: (8) a very rude saucer or crucible of coarse clay, (Fig. 4. 20) warped by excessive firing, with its rim pinched into two lips like those of the Graeco-Phoenician lamps, but wider, and at opposite sides of the bowl, (9) Fragments of furnace-slags.

Hitherto no lamp of the Graeco-Phoenician type has been found in any Bronze Age tomb or deposit in Cyprus, and though there is a temptation to regard this object (8) as a Bronze Age lamp, the opposite position of the lips, the absence of any smoke stain on them, the overfired look of the clay, and

140 KXCAVATIONS IN CYPRUS IN 1894.

its association with furnace-slags combine to suggest that it is more probably a crucible ; though whether for metallurgy or for porcelain-glazing cannot at present be determined. But the only Mykenaean lamp (if it be one) hitherto published? has the lips opposed in a somewhat similar fashion.

The red-stained saucers (6=Fig. 4. 19) and (7) meanwhile, and the unpainted jug (4) of a type which is with this exception, invariably painted, make it clear that we have to do with a local factory, probably both of the polished red ware, and of the painted white ware found in the tombs.

The Necropolis.— Tombs had been already opened in some numbers, and recently (though I could not discover by whom), on the ridge west of the

Fic, 4.—KALorsipA. VARIOUS OBJECTS FROM THE BRONZE AGE SETTLEMENT AND NECROPOLIS. Ashmolean Musewm.

chiflik, and on both sides of the high road: they exist also on the ridge where the chiflik stands, and are very frequent round the site of the Bronze Age settlement, East of the settlement the ground lies very low, and the limestone cap has been denuded away, leaving a great basin of the soft sandy underlying beds exposed. South of this depression the two table-topped hills between which the road goes appear to have been quite unoccupied at any time; but on the mainland of the plateau to the S.W. there are numerous well-preserved tombs, and these are the earliest part of the necropolis; which, as frequently happens, becomes later in character as it approaches the actual settlement.”

_——— - --. --

' Tsountas and Manatt. The Mykenacan 3 Cf. the necropolis of Idalion (K.B.H. Plate Age. 1897. Fig. 29, 30. ii.) and of Tamassos.

EXCAVATIONS IN CYPRUS IN 1894. 141

In this case, however, it is not quite clear that there was not an early shelter or building of some kind on the plateau itself to the S.W. of the furthest tombs noted in this direction: for close to the highest point of a well-worn short-cut from Kalopsida towards Vatili, and only a few yards south of site A, lie several large unhewn blocks of the cap-limestone, nearly in a straight line, running at a considerable angle with the nearest edge of the plateau-cap, and in positions into which they could hardly have fallen by merely breaking loose from the escarpment. The general impression was that of a very rude temenos, but though the site was nearly bare of soil, no pottery or other signs of occupation were visible within or without the enclosure.

Fic, 5,--KALorsipA. PART OF THE ConTENTS OF ΤῸΜΒ 11. Cyprus Musewm.

Site A. (Tombs 1-5.) The limestone cap, though not so thick as on the actual plateau eastwards, is firm and in good condition: the tombs are of the usual shaft-and-chamber type, fairly regular in form, and at a depth of 6-8 ft. froin floor to surface. They contained nothing but polished red ware, usually plain; some of the bowls from Tomb 3 had notched projections on the rim; some of the bottles and two-handled vessels had relief ornaments of snakes, crescents, etc.; and a few small bottles from 2. 3. 5. had incised ornaments. The clay was unusually soft and soapy, and the forms consequently rude and heavy: the polished surface also was of poorer quality and colour than at Ag. Paraskevi, and the incised ornament less sharply cut ; the white filling however was clearly traceable. A peculiar feature of the

142 EXCAVATIONS IN CYPRUS IN 1894.

ved ware, both here and on the later sites, was that the globular vessels almost all ended below in a blunt spike, as if to make them stand upright like wine amphorae in a bed of mud or sand. A similar nipple occurs very rarely elsewhere: e.g. Cypr. Mus. 59: Berl. Mus. (Tamassos (Lamberti) 1895. xxx ix. 741): Ashm. Mus. (Cypr. 79): the last-named specimen will be quoted again below (p. 145) for its peculiar punctured ornamentation. Compare also the vase from Tell-el-Hesy: Bliss, Mownd of Many Cities, Pl. 3, No. 83.

Tomb 5 produced two objects which deserve special mention. One was a small krater-like vessel which had formed part of a ring-vase like Cypr. Mus. 225-271: such ring-vases are not common, but seem to be confined, in the Bronze Age, to the earliest tombs: and consequently their correspondence with similar forms among the Libyan red-ware from Ballas and Naqada® is the more noteworthy, as it is not improbable that the very similar fabrics of Libya and Cyprus are closely related.*

The other object is a slab of polished red ware, of about the thickness of a Roman brick, bent at a right angle, so as to form a base and upright back: the latter is imperfect above, but appears to have been divided into four shallow panels. Whether this was a primitive shrine, or a copy of some piece of furniture, is not clear.

Site B (Tombs 6-19) lay about 50 yards north of A, in the direction of the road, and on the verge of the eroded hollow already mentioned. From the extreme east of this site (Tomb 11) westward and northward as far as the settlement in Site C the cornland area has been much denuded by the action of the weather and almost all the limestone cap has been removed; the surface was strewn with fragments of various kinds of Bronze Age pottery, spindle- whorls, and broken saddle-querns; all the tombs had collapsed, often shatter- ing the pottery, and many of them were found close to the surface of the eroded slope. ,

Site C (Tombs 20-28: 32) included the whole east and north brow of the ridge on which the settlement site was exposed. In the neighbourhood of the road the east face was much washed away by the rains, and the muddy slope was strewn with fragments of red ware and painted white ware: one small tomb (32), containing only ed ware and a discoidal loomweight (like No. 3 from the settlement but larger), was found close below the turf at the north end of this area. The tombs, as on Site B were all filled with earth, and had apparently collapsed, as most of the pottery was found broken and mixed with fragments of crushed limestone.

Site D was on the ridge west of the chiflik on the borders of a deserted vineyard: the tombs had mostly been rifled, and only one fresh one (29) was opened. It was about 7 ft. down, untouched, with the door in place. The lintel of the doorway was as usual level with the roof and the floor sunken some feet below the sill. The tomb was about half full of earth, but yielded

1 Cf. early Graeco-Phoenician sp. from Ama- 2 Petrie, Ballas Naqada, Pl. xxvi. and speci- thus, 1894. 286 (Brit. Mus.); A.B.H. cxlix. mens in Ashm, Mus. 15 6; Diimmler, Mitth. Ath, xi. 209, Beilage 3 Cf. C.M.C, Introd. pp. 16, 17. jii. 1,

EXCAVATIONS IN CYPRUS IN 1894. 143

only one plain bowl of ved ware, and a number of porcelain beads of the usual spherical type. ἶ Site EH was on the south side of the road, and a little south-west of the chiflik. Only two tombs were opened here (30-31) but nothing was got out, as the earth with which they were filled was compactly cemented together by infiltration of lime from above. They served however to indicate the ex- tension of the necropolis in this direction.

The tombs on Sites B C D E bore a totally different complexion from those of Site A as the analysis appended will show. None of the fine polished red ware was found at all; but in its place several coarse and degenerate

NECROPOLIS. A B σ B iy

Redware, plain..... x x x x | x relief ornt....... x x | x incised ornt. ... SS ΧΩ ἐς ὃς Χ Oa ae x x

brown yar. ......

x x x x x x x x x x x x

dull smear ... . YM oe see Black Punctured.. ex White Base-ring... Painted Pottery ..

Pottery. Xarx x

eT AK ς x x

Whetstones....... x x x

'§ Perf. stones..... .. x 3

ie Porcelain........... x x ) ΘΕ νοι 6000055220525.

Loomweight ...... x x

Stone grinder...... x *

* Horse-bones. + Human skulls. N.B.—C 26 has been intercalated in Site B to bring together the three examples of the Black Punetured Ware.

fabrics; of these the principal were (a) a fairly well modelled and polished brown or yellowish ware, often found by itself, and differing from the older red ware mainly in the poverty of its forms, and in the absence of a definitely red surface-pigment ; (Ὁ) a very coarse and rude fabric of unlevigated clay, covered with finger marks and bruises, and quite unpolished, but smeared or drenched with a muddy red slip!: the vessels of this ware were usually diminutive

—————

DE

a - - Fe eee) | I ee a a ae μα a ‘ Register No. of Tomb.| 1 2 3 5|6 1619 7 141018 8 13 26 1117 12 9 [af 22 25 23 28 24 21 32 29 31

144 EXCAVATIONS IN CYPRUS IN 1894.

probably be regarded as the last degenerate representatives of the polished red ware, and may be compared with the debased red ware of some of the tombs from Ag. Paraskevi (v. above p. 135).

These sites were also clearly marked off from Site A by the presence of bronze implements in a number of the tombs; all, of the types usual in the Cypriote Bronze Age, with the exception of the long small-headed pins (Fig. 4. 72.) encased in a spiral of thin bronze ribbon, which apparently have not been noticed before.? On the other hand, the mushroom-headed pin, with a hole half-way down the shaft, is entirely absent; which confirms the impression that this type is comparatively late in Cyprus, and that the tombs in question are prior to the stage at which it appears. One ring of unrefined silver lead, of the usual type (cf. Ag. Par. 1894. 10, above) was found in Tomb 11, and the small soft whetstones (Fig. 4. 79) in Tombs 8, 11, 14.

Besides red ware and bronze implements, several subsidiary types of pottery were found on sites Band C. The common painted white ware (Fab. II. 1, Cypr. Mus. Cat. p. 381) was fairly frequent, but never of good quality : the character of the clay indicated that it was derived from a bed closely resembling that now worked by the potters of Varoshi (Famagusta); it is coarser and more gritty than that of Ag. Paraskevi, and was frequently underfired, so that it retains its natural greenish tint. The vessels are clumsily modelled, and the painted ornament is simpler, and more coarsely applied, in a very poor, dilute, and loosely adherent pigment of the usual native umber. A few frag- ments of the polished painted white ware (Fabr. II. 2, C.M.C. p. 38) were found on the surface, and in Tombs 9, 22, 27. It is here an importation, pro- bably from the neighbourhood of Agia Paraskevi, and indicates that these tombs are among the later of the series; which in the case of Tomb 9 is clear from other considerations also.

The leather-like black or brown base-ring ware (Fabr. I. 3, C.M.C. p. 37) does not occur at all, with the doubtful exception of one native imitation; (Fig. 4. 1.) which again indicates an early date for the whole necropolis, as this fabric does not seem to have been in use in Cyprus much before the introduction of the Mykenaean vases with which it is usually associated. But the group of fabrics which I have called white ware with base-ring (Fabr. 1, 4, C.M.C. p. 37) is represented by two characteristic varieties ; (a) plates, and deeper bowls with vertical sides, of a hard and gritty, very white, and often overfired clay, well turned, apparently by hand, but with close resemblance to wheel-made, and still more to bronze types; (from Tombs 11, 17) ;? (Ὁ) oenochoae with distinct foot, ovoid body with angular shoulder and slender neck like a lekythos, of a greyish or even blackish clay, with many micaceous particles ; wholly devoid of ornament, but covered originally with a thin, almost lustrous, slip of darker colour than the clay (Tombs 9, 11) ;° the one specimen which has

1 The spiral marks on the lower part of shaft 3 Fig. 5, 2. 4. 14.17; Fig. 5, 1-5. 10. 13-17. of the eyelet-pin published by Dr. Diimmler, 26-7.

Mitth. Ath. xi. 209, Beilage i. 15, are narrower, 3 Fig. 4. 16. 26.

and look more like the remains of the thread by 4 Fig. 4, 21; Fig. 5. 8,

which the pin was secured.

EXCAVATIONS IN CYPRUS IN 1894. 145

the neck perfect has a delicately modelled lip, and might easily be taken for a local imitation of an Attic vase of good period. This type suggested to the local potters an imitation in painted white ware, of which a specimen is pre- served in Tomb 11, ornamented with an eye on each side of the lip, and with panels, each enclosing a snake-like object, round the body: Fig. 4. 22. The only specimen which I have seen, which can be compared with this vase, is in the Turin Museum ; to which it was presented by Gen. L. P. di Cesnola, with the locality Throni, which remains to be identified, but is somewhere in this part of Cyprus.

The last-named fabric (Ὁ) cannot be entirely dissociated from the black punctured ware (Fabr. I. 5, C.M.C. p. 37-8) of which the examples from Tombs 11, 18, 26, now to be described,! are among the first found in Cyprus. The clay is quite black, and rather finely levigated; it is usually soft, with a smooth lustrous surface like that of the red ware; but when overfired it turns to a purple-brown colour and becomes harder; but there is always a tendency for a surface to flake off with long exposure. The commonest form is a small lekythos or aryballos on a narrow button-like foot, with depressed body, and a short neck with swollen rim, and without lip. The only ornament is composed of punctured dots, either irregularly all over the shoulder, or in oblique lines, or confined within triangular and other geometrical areas. This punctured ornamentation is occasionally imitated in ved ware (Ashm., Mus. Cypr. 79) and in a variety of base-ring ware (Ag. Paraskevi, 1894, 10, v. above p. 142 and C.JL.C. p. 57.) This fabric has been since found by Dr. Ohnefalsch-Richter at Nikolides near Dali (Excavations of 1894. Tomb 6; Berl. Mus.; to be published in a forthcoming work ‘Tamassos und Idalion’); and a specimen from Cyprus, acquired earlier, but without locality, is in the British Museum (A 73-4). But it was already known from a probably Bronze Age tomb at Beth Suwour in S. Palestine ;3 and as a rare type of intrusive pottery on the XII. Dynasty site at Kahun (now in Brit. Mus. &c.: Petrie, Kahun, Pl. XX VII. 199. I llahun, Pl. I. 202, J.H.S. XI. Pl. XIV. 9) and elsewhere in Egypt. A magnificent specimen with recurring spirals, and with all the incisions filled with a white chalky substance, acquired by Greville Chester, in Egypt but without locality, is in the Ashmolean Museum; and another from ‘Upper Egypt’ and the same collector, is in the British Museum (1891/6/23.) Others, more closely analogous to the Cypriote examples, were found in graves which appeared to be of XII XIII. Dynasty date, at Tell-el-Yahudiyeh,* Khat’aneh® and El Rotab.2 One from Khat’aneh (J.c. Pl. XIX. 15) is without punctured ornament, and recalls vases from Tell-el-Hesy,’ which appear, from the drawings published by Dr. Bliss, to be of the same fabric, and are of nearly the same shape as the example from Beth Saour, and one of those from Khat’aneh. Two other

1 Fiy, 4. 23. 24; Fig. 5. 6. 7. 5 Onias, Pl. xix. 1-9, 15-17. Cf. Goshen, 2 The sp. figured in Murray, Handbook of pp. 21.

Greek Archaeology, PI. i. 8. 6 Brit. Mus. Inv. 27471-3. Goshen, p. 21. 3 Brit. Mus. 1876/2/28/2, two specimens. 7 Bliss, Mound of Many Cities, Pl. 3, Nos. 4 Eg. Expl. Fund Memoirs. Onias, p. 56, 89, 90.

Eek, 5 25. 0<

H.S.—VOL. XVII. L

146 EXCAVATIONS IN CYPRUS IN 1894.

specimens in the British Museum! were acquired from Egypt, but without exact locality. Finally, in the National Museum at Athens,” is a globular vessel of greenish clay with black slip, ornamented with punctured areas

confined by incised lines thus 7 :E~ | |; the neck is unfortunately

broken, and the provenience is uncertain, Similar black fabrics with white- filled punctured ornaments have been found, as native manufactures, in Libyan graves at Ballas and Naqada in Upper Egypt 8 and at Ciempozuelos in Spain (Petrie, B-N. pp. 38, 63. Bol. R. Acad. Hist. Madrid XXV. p. 436-450. Pl. IV., V., XII); but these parallels are not very close, the clays are quite distinct, and the forms entirely different.

The fabric in question is probably not Cypriote, but belongs rather to the South-Palestinian area. As it has not been found at-all on XVIII. Dynasty sites in Egypt, or in association with Mykenaean importations there or elsewhere, and as all the dated specimens known are from XII. or XII.—XIII. Dynasty sites, it is probably fair to regard the specimens from Kalopsida as evidence for a pre-Mykenaean date for this necropolis, and as affording at least a presumption in favour of a XII. Dynasty date.

This presumption is confirmed by the character of the porcelain beads which were found in Tombs 11, 14, 26, 29, (twice, it will be noted, in 11 and 26, in company with specimens of this black punctured ware), and are of a type which is commonly found in a certain class of Bronze Age tombs at Ag. Paraskevi and elsewhere. This type is almost identical,.in its spherical form, friable white paste, and thin and very pale bluish or greenish glaze, with a characteristic XII. Dynasty type of Egyptian bead, and is frequently found in Cyprus associated with other types equally characteristic of the XII. Dynasty; but not with any types characteristic of any later Dynasty. This certainly points to intercourse between Cyprus and Egypt under the XII. Dynasty, for though many of the Cypriote specimens may be native imitations made in Cyprus at a later date, the types themselves can hardly have been introduced, unmixed with later forms, except under or immediately after the XII. Dynasty. And in this instance, at Kalopsida, the combined occurrences of the porcelain beads and the black punctured ware seem greatly to increase the probability that we are dealing with a stage of the Cypriote Bronze Age which must be dated well before the beginning of Mykenaean influence in Cyprus, (and this, as the results from Lakshad tu Rid show, was itself fairly early in the Mykenaean Age); and probably, on the commonly accepted reckoning, close to the beginning of the second thousand years B.c.

Bones, human, or other, were as usual very badly preserved at Kalopsida. Tomb 28 contained a thick layer of unburnt bones, apparently all human, from which three skulls were extracted ; one however fell to pieces at once, and the other two, being in too fragile a state for transport to England, were deposited

1 Inv. 4806 a, 20849. 4 Ag. Paraskevi, 1894, 10 (C.M.C. 630, pp. 2 No. 115 = Inv. 210]. 55, 57, cf. above, p. 186); Episkopi (Kurion) 5 Petrie, Ballas-Nagada, Pl. xxx. 1895, 35 (C.M.C. p. 181).

EXCAVATIONS IN CYPRUS IN 1894. 147

temporarily in the Cyprus Museum. Their general form resembled that of the early Graeco-Phoenician skulls from Amathus, now, after many mishaps, in the University Museum at Oxford. Tomb 9 contained part of a human jaw, also in very bad condition, and several horse bones; among them horre- teeth which were much worn on one of the long sides: they had been used apparently as polishers, probably for the red ware pottery.

IIL—LAKSHA TU RIU (Larnaka District): Laren Bronze AGE NEcro- POLIS WITH MYKENAEAN VASES.

During the excavation at Old Larnaka (IV. below) news was brought of a casual find of Bronze Age potsherds on the surface, about two miles from Larnaka. The site is on the Εἰ, edge of the marshy depression which drains towards the Salt Lake by the stream which passes under the eastern arches of the Larnaka aqueduct, and is diverted from the Salt Lake into the canal which passes the Church of St. George-in-the-Distance (Αγ. Γεώργιος ὁ Maxpis). This hollow extends as far as Kato Vlakhos Chiflik, where it is bounded by the corresponding plateau of the Pasha Chiflik and the village of Dromolaksha: a northern arm of the same depression nearly reaches the Nicosia road at the fourth milestone from Larnaka. The bridle path from Larnaka to Kalo Khorio descends sharply into this hollow about twenty minutes walk from its parting from the Nicosia road, and this part of the edge of the plateau seems to be known as the κατέβα (‘descent’) τοῦ καλοῦ χωρίου. But as Kalo Khorio is on the further side of the depression, which is itself known as the Aafia (‘hollow’) rod ‘P/ov and as this latter name was generally adopted by the men during and after the work, I have preferred to let it pass current as the name of the site on the E. slope? A peasant represented himself as owner of the site, and gave leave to excavate; but before the work was far advanced, the representative of a M. Zarifi living in Constantinople put in a counter claim, and prohibited excavation. By this time, however, the character of the site was evident, and it was enough to clear and close the tombs which were already open ; all on the strip of waste land along the actual brow of the hollow. So far as could then be seen, the ploughed land had never been disturbed ; but in the succeeding weeks the whole of that edge of the Zarifi estate was honey-combed with illicit digging, which the Government was apparently unable to prevent. By this time the site is probably ruined.

Tombs are apparently abundant also near Kato Vlakhos Chiflik and again at Agi Anna, which lies about half way between this and Alambra, and commands the valley route from the lowlands of Larnaka, Kiti, and Zarukas, to the neighbourhood of Dali. But neither of these necropoleis has been

1 In Kalymnos, where a rough kind of polished _one side polished by such usage, were common,

red ware is still produced, smooth pebbles are on the surface, at Kalopsida, used for this purpose ; a similar ware is made in 5 It is marked L on the map of Larnaka and Khios ; pebbles, also, of crystalline rock, with the neighbourhood, Fig. 6, L 2

148 EXCAVATIONS IN CYPRUS IN 1894.

properly examined, so far as I know. The only earlier hint of a Bronze Age necropolis at or near Larnaka is afforded by a number of hemispherical bowls of white slip ware (II. 4) labelled ‘Kition’ in the collection presented by General L. P. di Cesnola to the Turin Museum. These bowls, as will be seen below, were abundant in one of the five tombs opened in 1894, at Laksha tu Rit. A fragment of another such bowl, and one scrap of Mykenaean ware were found, almost on the surface, on the Turabi site, and may have come from earlier operations at Laksha.

At Laksha tu Rit the limestone cap is about three feet thick, brecciated, and in good condition. The tombs closely resemble those of Ag. Paraskevi ; Tombs 1 and 5 were dome-shaped with the opening at one side of the apex (cf. K.B.H. clxxii. 17, 18), and the dromos short and bath-shaped as at Ag, Paraskevi.

Tomb 1 showed clearly, on the sides and roof, the marks ot the hewing tool of its maker; which had “two cutting edges, the one adze-like, the other a blunt point; exactly like those of the modern κοῦσπο. This tomb had also a series of long shallow niches arranged round the sides, in which a number of the smaller vases were found lying. The floor was covered by a large talus of earth which had crept in between the decayed door-jambs and the door-stone, which was still in situ. It contained a large quantity of plain red ware of fair quality, and a few specimens of incised red ware of a dark-tinted, probably local fabric}; a number of well made vessels of painted white ware (C.M. 360 and Ashm. Mus.), and an unusually large series of bronze implements and ornaments; two axeheads (C.M. 503. and Ashm. Mus.); four daggers with handle-rivets (ΟΜ. Type a 5. 21-2. and Ashm. Mus.); one dagger with hooked tang (Ashm. Mus. cf. ΟΜ. Type y) ; one of the eyelet pins with large head (Ashm. Mus. cf. ΟΜ, 594-8); four beads of bronze ribbon, spirally coiled (ΟΜ. 626. and Ashm. Mus.) ; four spiral rings of bronze (C.M. 624. and Ashm. Mus.); and a concave dise or shallow bowl of thin sheet bronze, ornamented by four small circles stamped from the outside (Ashm. Mus.) ; the purpose of which is not clear, as nothing of the kind has been published. Several rings of unrefined silver-lead (C.M. 615), like the bronze rings, and an unrecognisable fragment of silver, complete the list of metallic objects. Three clay spindlewhorls of incised red ware were also found; and two perforated stone mace-heads or spindle- whorls like those from Kalopsida (p. 142). This tomb group is in Ashm. Mus, except the spp. noted in C.M.

Tomb 2 lay on a small spur about a quarter of a mile south of the path, and faced north. It measured 6 ft. x 4 ft. x 4 ft., was cut obliquely to: the right of the door, and was full of earth. Its contents resembled those of Lomb 1—bowls of red ware; several vases of painted ware (C.M. 880); a dagger (C.M. 523), two pairs of tweezers (C.M. 602-3); three silver rings

1 These forms were characteristic of the black but the clay in this instance was dark red all slip ware, which frequently oxidises and turns _ through. red with ill-regulated firing. Cf, C.M. 203-5:

EXCAVATIONS IN CYPRUS IN 1894. 149

- Ἢ

— SS {ΠῚ

Mo

‘ile = aes a a £ a

sain Whe 3 Teale sani Ὄς

|

Te alll, oy. \—

οἱ

Fic, 6.—THE NEIGHBOURHOOD OF LARNAKA. Based upon the Trigonometrical Survey of Cyprus and Drawn by Ὁ. V. Darbishire,

150 EXCAVATIONS IN CYPRUS IN 1894.

like those from Zomb 1, but found linked into a chain (ΟΜ. 616); and a similar perforated stone (C.M. 663). The occurrence of a chain of rings in a Bronze Age tomb is, I believe, unprecedented, though the abundance of the rude rings in many cases had long made it probable that such chains were in use. Cyprus Museum. C.M.C. p. 58.

Tomb 8, near the last named, was a large, perhaps partly natural cave; which had fallen in. It yielded very similar plain and incised red ware, and painted ware ; and a bronze spike or awl about 7 inches long, (C.M. 565). Cyprus Museum. C.M.C. p. 58.

Tomb 5, a small collapsed tomb, near the last named, produced red ware and painted ware (e.g. C.M. 345), and two perforated stones. (C.M. 651, 660). Cyprus Museum. C.M.C. p. 58.

Tomb 4 was of ‘beehive’ shape, like Tomb 1, and of about the same size but without the niches: but its contents were almost wholly different, and so abundant that it was thought desirable to depart from the usual practice, and

Fic. 7.—LAKsHA TU Riv: PART OF THE CONTENTS OF Zomb 4; IN THE Ashmolean Museum.

to divide the group between the Ashmolean and the Cyprus Museum: some duplicates also are in the Pitt-Rivers Museum, Oxford. Red ware was absent, and the ordinary painted white ware was only represented by a single bowl of unique form, obviously an imitation of the following type, of which 25 examples were found. This type is the painted ware with white slip (C.M.C. Fabr. II. 4. p. 39=the ‘ Agia Paraskevi ware’ of Mr. H. B. Walters in his report on Kwrion ; id. p. 181) which, though probably made in Cyprus, has a wide range of distribution, from Hissarlik, Athens, and Thera, to Egypt and §. Palestine.' Only the common hemispherical bowls however were represented, and none with any specially elaborate ornament.

The base-ring ware (Fabr. I. 3. C.M.C. p. 37 reff. Fig. 7. 3-6. 14: 8. 1-7), which has an even wider distribution, was abundant, and mostly of the white

i Fig. 7. 12. 13. 18: 8. 8. 9, 11-18. Cf. references in C,M.C. p. 39,

EXCAVATIONS IN CYPRUS IN 1894. 151

painted variety;! but one jug had the characteristic relief ornament of divergent spiral scrolls on the body,? and all showed clearly the double rims and bases, and the indented seam patterns which suggest that this ware derives from a leathern prototype. The common conical bowls, and both types of the one-handled jug were represented: a hemispherical cup on a high foot, with white binding pattern outside,’ is a new form, so far as I know.

Four Mykenaean vases confirm the impression of date conveyed by the preceding fabrics: three‘ are ‘ pyriform vases’ = ‘stamnoi,’ with three small handles and a simple lattice ornament (replacing the characteristic scale

Fig. 8.—LAxksHA TU Riv ; PART OF THE CONTENTS OF Tomb 4 ; IN THE Cyprus Museum.

pattern) on the shoulder; the fourth ° is a fine pyriform ‘biigelkanne’ ‘= pseudo- amphora’ with a characteristic flower-pattern of the finest style. Two large unpainted kraters® of coarse white ware (whether hand- or wheel-made, is not clear) also show Mykenaean influence: a coarse jug with slightly pinched lip’ corresponds with a type which is common in the coeval necropolis at Nikolides near Dali; and a small pear-shaped jug of yellow clay, hand-made but shaped outside with a knife to a point below,® resembles a vessel from Nikolides (1894 V. 161, Berlin Museum), and others from Enkomi (1896, 4, C.M.C. p. 188, ef.

4 Fig. 7, 1.9: 8. 10. 5 Fig. 7. 11. 6 Fig. 7. 8. 8. Fig. 7. 10,

1 Cf. Brit. Mus, A. 121.

2 Fig. 7. 6. 8 Ash. Mus. (Cypr. 114), shattered in transit, 7 Fig. 7. 7,

and not included in the photograph.

152 EXCAVATIONS IN CYPRUS IN 1894.

Brit. Mus.); and, still more closely, a small vase from Tell-el-Hesy, figured by Bliss, Mound of Many Cities, p. 87, Fig. 174.

No bronze or silver objects were eed from this tomb; and, of ornaments, only three small stone beads (ΟΜ. 709-10 = Fig. 8 ee 9 and 10) of the double cone type, which, though it begins in the Mykenaean Age of Cyprus (cf. Kurion 1895, Brit. Mus. 96/2/1/76—7), is equally charac- teristic of the earliest stages of the Graeco-Phoenician with well-bowed fibulae: 6... Amathus 1894, 201, 286 (Brit. Mus.): Larnaka Turabt 55 below, Of these ΟΜ. 709 is ornamented, like the specimens from Kuwrion, with drilled circles with a central point. Similar beads were found in Kalymnos with the sub-Mykenaean vases Brit. Mus. A 291 ff.

IV.—_LARNAKA: (Turabt Teké): GRAECO-PHOENICIAN AND HELLENISTIC TOMBS.

The land on both sides of the high road to Nicosia, and immediately south of the last houses of old Larnaka, which just reach it, belongs to the Turabt Teké, a Mohammedan shrine which lies on the east side of the road a little south of the houses aforesaid. The wall of old KITION can be traced as a nearly continuous escarpment 8-12 feet high in the fields S.W. of the Teké, and at about a quarter of a mile distance ; and the necropolis begins almost immediately outside the wall, though only with very late tombs. This part of the necropolis seemed to have been left almost wholly undis- turbed in recent times, though the traces were frequently found of ancient

Fic. 9.—ENGRAVED CHALCEDONY, Larnaka, Hassan 4. Ashin. Mus.

τυμβωρυχοί: and as the earliest parts of the necropolis of Kition were known to lie elsewhere, to the N. and to the S.W. of the ancient town, it appeared probable that the later Graeco-Phoenician and early Hellenistic Tombs might be found in this direction; more especially as a fine stele with a fourth century Phoenician inscription, now in the British Museum,! had been found im situ, at the point marked on the map (Fig. 6) with an asterisk, in making the highroad soon after the British occupation.

1 Brit. Mus. No. 47.=C.4.8. viii. 44.

EXCAVATIONS IN CYPRUS IN 1894. 153

As the site proposed was Vakuf land, leave to excavate had to be obtained from the Commission of Evkaf in Cyprus, and was eventually granted on condition that the Evkaf third of the finds should be deposited, with the Government third, in the Cyprus Museum.

While the negotiation was going on, a few shafts were made in the field of Hassan Effendi, north of the last houses on the E. side of the Nicosia road : but only late Hellenistic or Graeco-Roman tombs were found. Unfortunately, though a number of bronze coins emerged, they were al] quite illegible, and for the most part corroded through and through. Zomls 1 and 7 contained gold earrings of late types: (1) Ashm.: (7) C.M. 8049. Tomb 4 contained a plain alabaster sarcophagus, in which were found, together with five of the coins and five late lamps, two small ornaments of poor light-coloured porcelain, a glass counter, and a seal of bluish chalcedony which is represented in Fig. 9 (Ashm.) and gives an approximate date to the group.!

Scattered about the tomb were 44 late lamps, 14 coins, and the usual apparatus of glass bottles (including fragments of purple glass), coarse pottery, and bronze pins, mirrors and rings: four bronze sockets, like those in Perrot- Chipiez Hist. del’ Art. II. Fig. 356 (New York), may have formed the supports of a casket.

Tomb 6, with two long chambers in series, had certainly been opened before, though the door stone was in place, and no cippi or debris were in the shaft: for its sarcophagi had been slightly shifted, and only a Cypriote lamp, a small clay bottle, and a clay alabastron remained of the contents.2 In the side of the dromos was a small ‘ cupboard-grave’ (ἀρμάρι : = Tomb 2) with a spiral ring of bronze, and a few ‘bottle jugs,’ late lamps, ὅσ.

The object of the Zwrabé excavation was to test the conclusion drawn from the previous excavations of the Cyprus Exploration Fund, at Kuklia and at Poli tis Khrysokhou, that no accurate chronology of the native Cypriote pottery can be attempted; that the various styles occur in inextricable confusion in the tombs; and in particular that the characteristic Cypriote fabrics which are found with Attic black-figured and red-figured vases, are also found with Roman Imperial coins and the clear blown glass, which certainly does not begin until late Ptolemaic times, and most probably not until after the Roman annexation of the Island. The earlier excavations at Poli, which formed the basis of Dr. P. Herrmann’s ‘G'rdberfeld von Marion’ led, it is true, to a different conclusion, which was vigorously criticised by English excavators at the time; but so late as 1896 the Catalogue des Vases de Terre cwite du Louvre of M. Pottier reckons the fine native ved ware, (which actually begins with the earliest fibulae, and disappears in the eighth or early sixth century), as an imitation vf Attic importations; and the Cypriote bucchero, or ribbed ware with black slip, (which begins in the Mykenaean Age, and disap- pears earlier than the fibulae), as a Hellenistic fabric of the late fourth or the third century.

1 Cf. Brit. Mus, (Semitic Room, 1022, 1026, Commissioner of Larnaka, and placed in his 1039). garden. 2 One of the sarcophagi was extracted for the

154 EXCAVATIONS IN CYPRUS IN 1894.

Accordingly it seemed worth while to excavate carefully a site in which late Graeco-Phoenician, Hellenistic, and Roman tombs were likely to be found together and to classify and register even the commonest objects, in the hope of finding, among some sixty tombs, at least one crucial instance. And the result seems to have fully justified the experiment. The tabulation of eight characteristic types of Graeco-Phoenician pottery, and of ten kinds of objects which are commonly found in Hellenistic or Graeco-Roman tombs elsewhere, shows without need of further comment, the absolute gulf which divides the two styles on this site. The remarkable absence of the reburials which

Tombs marked thus x 27 δῦ 26 17 48 21 41 57 12 Tombs marked thus x | 23 62 47 42 61 58 11 27 59 56 34 60 39 35 4 24 44 1 45 22 20151816 7 2

Native Cups.:40t4). ἘΠῚ x Seige 3 Flat-lipped Jugs............... x x x x Flat-necked Amphorae ..... x REI? AE ΟΝ CORTE Round-bottomed Amphorac % PRUE LG ede oP tee ex Native: BowiSi;-1.---.sceoe x Χο oa Amphorae..,. 3401. tisecch ΞΕ τς x x a ἀφ δι) εἰ cox x

| Oenochoas . ΠΝ kee EPL ΤΣ CE: ΧΑ,

| Cypriote lamps................ 6.428) K O28 % Se 5 Dd ΡΧΗῸΣ x Hellenistic lamps ........... | χει GxXDh xh SX, Yi KBxX x Clear Glass .ox.. τ κα | x JPGEK? XE δὲν Be x Mirrors. ΠΡ ΠῚ Χο ΟΣ x x Coins. ........<cgean nee, posers x) χριὸς Sep hoes AXE *°'Tear-bottlesiivn..cjcs ee ee Be Kiar Χ Common jugs ..... ...... ... Χωρ Ἄλεος x Amphorae, late types........ 2G BK Χ δὲ Gold ornaments .............. Χ ὃς εκ Χ Bone ornaments .............. x x Alabaster’ τος πον πο * 4M x x x ‘* Bottle-jugs 70.0. ΜῊ ΠΕΤΣ x x PET ie: FIER OOD Ὁ x ΟΝ x x Bronze rings, &¢. ............. | * x x 14 * x x ΔΗ Χ

usually confuse the tomb record of a Cypriote necropolis is probably to be accounted for by the fact that the site lay remote from any considerable thoroughfare, and that consequently burials were at all periods infrequent, so that the site never became over-crowded, like the eastern necropolis of Amathus or many parts of those at Poli, where reburials are almost the rule. In one instance (7 ταδί 31-87) four tombs were found in the same shaft, each excavated in the earthen filling of its choked or collapsed predecessor, and therefore in any case at some considerable intervals of time. Here indeed Graeco-Phoenician pottery, a silver earring of a fourth century type, a Hellenistic terra-cotta statuette and a number of coins and late lamps, were

EXCAVATIONS IN CYPRUS IN 1894. 155

found together; but an exception of this kind, with so clear evidence of mixture, may be fairly claimed to ‘prove the rule’; and in any case it is worth noting that clear blown glass was absent, and that the coins, though illegible, were of Ptolemaic, rather than of Roman Imperial fabric.

In another case, (Z'urabt 40) a μνημεῖον or surface grave, presumably of late date, and with the bones well preserved, had been sunk too near to the crown of the roof of a Graeco-Phoenician chambered tomb with a sarcophagus in the middle; the floor of the μνημεῖον had subsequently collapsed, and most of the contents had fallen on to the lid of the sarcophagus; but here there was no difficulty in distinguishing the two sets of remains, and fortunately also, nothing appeared to have been buried in the μνημεῖον except the body; a strong confirmation of its very late date. f

Even the frequent evidence of previous disturbance of the tombs cannot be held to invalidate the argument from the pottery and glass ; since, in the most patent cases (e.g. Twrabt 35) only gold ornaments had been sought, and the pottery, glass, and other objects, and even the coins, had been left undisturbed ; and as it was in this very tomb (35) that the one instance occurred of a Cypriote lamp associated with Hellenistic lamps, the explanation is obvious that the Cypriote lamp was introduced by the tomb-robbers, and was probably used by them during their search ; just as my own men used such lamps, here and at Amathus, when we ran short of candles.

The form of the tombs themselves also gradually changes, and was found to supply an approximate indication of date. The tombs which contained Cypriote pottery of sixth and fifth century styles had uniformly flat or nearly flat roofs, never gable-shaped or rounded. In the fourth and third century, to judge again by the degenerate character of the pottery, the chambers become larger, and proportionately longer, and the roof becomes more or less definitely rounded at its junction with the walls, but does not yet rise to a semi-circular section. On the other hand, those with distinctly Roman coins, much glass, and late lamps, have uniformly a well rounded barrel-shaped roof, and frequently have a second chamber behind the first: both chambers also are much longer than they are broad; and the dromos is regularly provided with steps, which in the later examples (e.g. Hassan 1 and 5) are made of thick slabs of gypsum ; in the earlier tombs they are simply cut in the soft rock ; and in the earlier Graeco-Phoenician tombs no steps could be traced at all.

Taking the principal tombs in roughly chronological order :—

Tomb 58 represented an early stage in the purely geometrical period which succeeds the Mykenaean. It was a very small tomb, and contained only a small plate or saucer of a local fabric of Graeco-Phoenician red ware (Fabr. II. 3, C.M.C. p. 60) with black lines; and a one-handled jug with a spout on one side of the shoulder, of white ware (= C.M. 1028a, Fabr. II. 1, C.M.C. p. 59) painted with a fully geometrical design of chequers, triangles, lozenges, and swastikas: this was of the regular Kition clay,’ which is easily ΝΠ ae on gl, he eee ae ee

1 The local fabrics of Graeco-Phoenician pot- careful observation than they have received tery are often well marked, and deserve more hitherto, Typical specimens of this fabric of

156 EXCAVATIONS IN CYPRUS IN 1894.

distinguishable : it has a fine dusty texture, rather yellowish colour, and low hardness, and absorbs the paint with unusual readiness. Cyprus Museum (C.M.C. p. 178).

Tomb 55 cannot be very much later. It is dated by a stone bead or spindlewhorl of double conical form, ornamented with drilled circles, like those from Laksha tu Rit 4 (v. above) and the early geometrical tombs at Amathus quoted there, and contained also a clay horseman of the usual rude type,! and an oenochoe and another jug of common local fabric.

Tomb 34 contained a plain bronze bowl (C.M. 3513), two oenochoae of sixth or late seventh century form with concentric circles and * on the neck, an amphora with groups of fine black bands on the body, and a few painted plates. It might very well be of the sixth century, and the same date would suit the following tombs. Cyprus Museum.

Tomb 11 contained a wine-amphora of type 3 (Fig. 13), round-bottomed eyathi (‘bottle jugs’), a large oenochoe with bands of the purple-red paint which is characteristic of the fabric of Kition, two Cypriote lamps, and some bronze earrings and beads. Cyprus Musewm.

Tomb 12 contained wine amphorae of types 2 and 3 (Fig. 13), a round bottomed cyathus, and some bronze armlets: three undisturbed interments, but no other personal ornaments.

Tomb 25 contained wine-amphorae of types 2 and 3 (Fig. 13), one of the former mended with gypsum, and one of the latter bearing broad red bands edged with black : one Graeco-Phoenician amphora,” with groups of lines on the shoulder, and a wavy line round the greatest diameter (a sixth to fifth century type elsewhere in Cyprus), anciently rivetted and full of calcined bones; an oenochoe of sixth century form, also anciently rivetted ; another oenochoe with concentric circles; two pieces of the red ware, which becomes degraded in the fifth century ; and several commoner vases. Cyprus Museum.

Tomb 26 contained amphorae of types 1 and 2 (Fig. 13), and two varieties of cyathus.

Tomb 37, the lowest layer in the confused shaft already mentioned, (p. 154) contained a krater-amphora with concentric circles, a painted oenochoe, and a flat-lipped jug, which can hardly be later than the sixth century. Cyprus Museum.

Tomb 60, in the furthest part of the site W. of the highroad, contained an elaborate and more artistic equipment than most ofthe early tombs ; many of the vessels bore the concentric circle ornament, and two the wavy line on the neck, which is a Mykenaean survival, and disappears almost wholly before the period of Hellenic importations. The band of small black triangles on the shoulder of the angular vessel, the occurrence of red ware, even though not of the finest quality, and the frequency of the flat-rimmed globular jugs,

Kition are easily accessible at the British 4 Cf, ΟΜ. 3293-97 (Poli) ; 8299-3305 (Ama- Museum, South Kensington (2071/1876), and thus); Lowvre T.C. Cyprus 48 (Heuzey, Pl. x. 3); Ashmolean Museums. I regret that I have not Brit. Mus. (spp. from Amathus and Kurion). been through any German collection since my 2 Fig. 14a gives the form approximately. return from Cyprus.

EXCAVATIONS IN CYPRUS IN 1894. 157

also point to an early date. The ornament, however, of the krater with vertical handles in Fig. 12. 6. (= Ashm. Cypr. 501), consisting of a carelessly drawn lotus with buds, in the central panel, flanked by eight or nine leaved rosettes on a black ground, cannot be much earlier than the revival of Egyptian com- merce under the XXVI, Dynasty ; and an Egyptian eye-charm in pale blue glazed porcelain belongs to a class which seems to have entered Cyprus about that time, and continues to be very popular at Amathus until the middle or end of the fifth century. The tomb accordingly must be dated in the middle or latter part of the seventh century. Among the other contents were an amphora of type 1, and a painted one of type 2 (Fig. 13); a cylindrical-sided pot with diminutive handles and very low neck (Fig. 12. 4) which closely resembles one from the Cesnola collection (S. Kens. 2109/1876); a flask or pilgrim bottle (damaged when found, and shattered on the voyage to England) with one face nearly flat, the other almost conical, and recalling, by the arrangement of its four marginal handles, both its wooden prototype, and a Sardinian vessel in the British Museum (A 1680?), which, from the character

Fig. 10.—LARNAKA, TURAB{. Tomb 56, Ground Plan.

of its concomitants (A 1680!-%, cf. the earlier tombs on the hill of St. Louis at Carthage’), might well be of the same approximate date; and a model of a circular shield like those from the Kamelargd site below (p. 168), in local clay, with traces of blue or green circles on the convex surface, and the usual single κάνων across the concave interior. All the vessels cited, except the amphorae, are in the Ashmolean Museum.

Tomb 56 was in every way the most noteworthy in the whole series. It lay in the easternmost section, behind the Turabi orchard, in a range of very similar tombs nearly all of which had been rifled. The door faced E., and was furnished with two thick gypsum slabs, set one behind the other, and separated by door-jambs of masonry. The chamber itself measured 9 ft.x 9 ft.x 54 ft., and had a quite flat roof, only rounded an inch or two where it joined the sides (Fig. 10). Along each side of the chamber lay a pile

1 Delattre, Tombeaux Puniques 1890. Nécropole Punique 1896.

158 EXCAVATIONS IN CYPRUS IN 1894.

of the common gypsum slabs, about 6 ft. x 2 ft., forming couches: on the top of each pile, and between each slab of each pile were the flattened remains of an interment; apparently a fresh slab had been introduced at each burial to avoid the disturbance of the previous occupants. Under the top slab of the couch on the left of the door was found at the point marked with a star (Fig. 10), and among the bones of the hand, a siiver ring of type d (C.ILC. p. 127), with a swivel mount of electrum containing a haematite searaboid engraved in pseudo-Egyptian style (Fig. 11), with a representation, (such as is commonest on monuments of the XVIIIth Egyptian Dynasty, and especially on those of Thothmes IIT.1), of a male figure, (the king), as Horus tended, perhaps even suckled by a goddess, Isis, and overshadowed by the winged serpent of Uazi? twined round a papyrus in allusion to the refuge of Horus in the marshes of Buto. The ankh behind the goddess depicts the ‘long-life’ which she bestows. The seal is assigned by Prof. Petrie to the XXVIth. Dynasty. The form of the ring itself derives from a common XVITI.—XIXth

Fic. 11.—HAEMATITE SCARABOID : Drawn from an impression, Larnaka Turali 56; Ashm. Mus.

Dynasty type; but it is normal in Cyprus in tombs of the sixth—fourth centuries. Two beads of decomposed glass were found elsewhere among the interments.

Across the further end of the tomb was laid another large gypsum slab, on which were arranged, in the order shown in the drawing (Fig. 10), a number of earthenware utensils which we may regard as part of the furniture of a Graeco-Phoenician dinner-table. The concave-sided rings are of course the stands for the round-bottomed amphorae of type 2 (Fig. 13), four of which, with painted bands of red or yellow, edged with black and with black zigzag, together with one plain one, and four conical amphorae of type 1, were stacked together in the further left-hand corner of the chamber. The ring-shaped amphora-stands closely resemble Egyptian examples of XII.-XIII. and subsequent Dynasties. I know of no other example from Cyprus, but two examples of a rather deeper and narrower type have been found with

1 £.g. Lepsius, Denkmdler III. 35. Ὁ. Khat’aneh. (Thothmes III.). I am indebted to Prof. 3 E.g, ΟΜ. 4186-9 (Idalion 1894. 26 Flinders Petrie for the references and the iden- Amathus 1894. 80).

tification of the subject. 4 E.E.F. Onias, Pl. xix. 18, 19. 2 Cf. E.E.F. Goshen Pl. 9 Inscr. from

EXCAVATIONS IN CYPRUS IN 1894. 159

geometrically ornamented amphorae of a form like Fig. 12. 14. 15. at Carthage. One of the last-named was carefully set upright with its mouth sealed with clay ; the others seemed to have been arranged regularly round it, but the collapse of one of the painted ones had disturbed the remainder, An oenochoe with black-edged red band like that of the amphorae, three ‘ bottle-jugs’ to measure the wine, three plain bowls holding three to four times the content of a ‘ bottle- jug,’ five plates (including one of the characteristic red ones), and a saucer, were scattered about the chamber and completed the suite.

The cup-and-saucer-like utensils, which are of local clay, and some- what overfired, with a red band on the bevelled rims, are probably torch holders. Two similar vessels, but unpainted, are in the Cyprus Museum: one, C.M. 963, from Poli (C.E.F. excavations) 25, is exactly of the same clay as a deep dish-cover (C.M. 962) which has unfortunately lost its tomb- mark, but was found in 1894 lying with objects from Puli C.K.F., and may have come from the same suite. The other (ΟΜ. 964) bears the old label

Fic. 12.—GRAECO-PHOENICIAN POTTERY FROM LARNAKA TuRABf 56 (10—15) AnD 60 (1—9), Ashm, Mus.

[844]: it consequently belongs to an excavation not later than 1885: and as the vase (ΟΜ. 2148) which bears the old label [843] came from Episcopt (KURION) in 1884, it is possible that [844] may have come from the same site, There is a diminutive model of the same utensil in the Rugby School Museum, presented by Mr. C. D. Cobham some years ago: without locality, but apparently of the local fabric of Kirion. The only other examples with which I am acquainted are those from Tell-el-Hesy figured in Dr, Bliss’ Mound of Many Cities, p. 87, Fig. 174: cf. 238. These reproduce every detail of the outline of the vessels from Twrabt 56, adding only a very slight spout upon the outer rim. In the same photograph, and from the same stratum, are (a) a number of ‘cockleshell’ lamps of a type which in Cyprus would correspond with an early (seventh—ninth century) date, but which in Egypt would go

1 Delattre. In tombs which cannot be earlier than the seventh century.

160 EXCAVATIONS IN CYPRUS IN 1894.

back to the XVIII. dynasty; (@) the small jug with pointed base which has already been compared with a vessel from Laksha-tu-Riw 4 Fig. 7.10. and with others from the new Hnkomi site ; and is itself presumably of Mykenaean date- But the forms of the amphorae, oenochoe, and other vessels in Turabt 56, not to mention the ring and its engraved stone, point unmistakably to the sixth century or very little earlier. The persistence of the form of the torch holders over so long a period, is remarkable, if Dr. Bliss’s dates are accurate; but this is not the only instance in which discrepancy occurs between his dating of individual objects by their position in the Tell, and the presumable dating of the same objects by the correspondence of style with objects in the Cypriote series; and it is not improbable that whether from original unevenness of the layers in the Tell, or from whatever cause, some of the pottery at Tell-el-Hesy has been assigned to a lower layer than that to which it would more conform- ably belong. For the present therefore Twrabt 56 may be assigned to the sixth century. Its contents are in the Ashmolean Museum, with the exception of broken vessels, the common plates, and one of the painted amphorae (with yellow band and black lines and zig-zags), which was presented to the Cyprus Museum (ΟΜ. 2007 a). . .

t ia θ é 4 Fig. 13,—TyprEs oF WINE AMPHORAE FROM GRAECO-PHOEFNICIAN TOMBS AT LARNAKA.

(1) Larnaka 26. 41. 42. ὅθ. 59. Amathus 1894. 80 and 84. cf. Naukratite forms and E.E F. Tanis 11. xxxiii. 4. (Defenneh): variants with acute shoulder Zarnaka 17. 60: (1a) variant from Larnaka 59. cf. Naukratite form: (2) Larnaka 11. 12. 17. 56. 60; variants from 34: frequently painted : cf. Naukratite form : (3) Larnaka 25 cf. Amathus 1894. 251, and E.E.F. Tanis 11. xxxiii. 6. (4) Larnaka 62; cf. late one-handled form from Larnaka 8.

Tomb 58 contained an oenochoe of fine red ware with vertical circles, and groups of concentric circles: the only specimen of the fine red ware from the whole series of tombs. With it were two jugs, one with red bands, of common local type, a Cypriote lamp, a red plate like that from Tomb 56, two bowls, and some ‘bottle jugs.’ The fine red ware does not seem to come lower than the early sixth century; but the lamp was of the flat rimmed type which elsewhere in Cyprus seems to be rather later. Cyprus Musewm.

Tomb 59 contained two pairs of amphorae of types 1 and 2 (Fig. 13), a dish-cover and an amphora-base like those from Jomb 56; and lamps, bowls, plates and ‘bottle jugs,’ and an oenochoe, which also corresponded with the equipment of that tomb. The only new feature was a flat-rimmed jug like those from Tomb 60 above. A conical seal of green porcelain, found in the

EXCAVATIONS IN CYPRUS IN 1894. 161

dromos, would date, like the tomb, from the sixth century. The tomb itself had collapsed.

Tomb 17 contained two pairs of amphorae of the same types as those in Tomb 56, besides eight bottle jugs and a Cypriote lamp: there was no slab in the doorway, and the tomb had probably been cleared of its fine pottery. The amphorae however are enough to give it an approximate date in the sixth to fifth centuries.

Tombs 42, 43, 47, 47a, lay side by side, and were found to communicate with one another, and-to have been rifled. Zomb 42 had a nearly flat roof like the sixth to fourth century tombs at Amathus, and contained, besides amphorae of types 1 and 2 (Fig. 13), an unusually elaborate red ware amphora (C.M. 1157) the ornament of which is approximately reproduced in Fig. 14. The fabric appears to be local, but the style, like that of the krater from Jomb 60, recalls that of Ormidhia;+ though I do not know of any red ware from that site. The apparent collocation of ‘black-figured’ and ‘red-figured’ bands of ornament is no argument against a sixth century date for this vase; for the ‘red-figured’ treatment of the rosette, and other ornaments here illustrated is

Fie. 14.—a. GRAECO-PHOENICIAN AMPHORA OF V.—VI. CEenTurRY (Turabt): ὃ. DEGENERATE FORM OF IIJ.—II. Century DATE (Turabi 43): c. ANOTHER DEGENERATE PTOLEMAIC FORM INFLUENCED BY HELLENIC FORMS (T7wrabt),

well established in Cyprus at an early period, and derives not from a Hellenic, but from an Egyptian prototype. Another amphora, with ‘tree ornament,’ found in 43, might well be of early fifth century date.

Tomb 23 contained one of the flat lipped jugs (cf. 60), a common plate, and a bowl. The fragment of Bronze Age pottery already mentioned Ρ. 148), came from the shaft of this tomb; and another from the abortive

shaft 63. Degenerate Gracco-Phoenician Tombs. IV.—III. Century.

Tombs 26, 27, 28 and 61, 62, represent the decadence of the Graeco- Phoenician Age: the equipment becomes very meagre, and the forms of the vessels lose their character (Fig. 14): the tombs themselves also begin to

1 Perrot-Chipiez, Histoire de l' Art, iii. figs. 507, 623. H.S.—VOL. XVII. M

162 EXCAVATIONS IN CYPRUS IN 1894.

change their form, and acquire barrel-shaped roofs; and are approached by a flight of steps.

Tomb 26 contained only an amphora of type 1 (Fig. 13), two smaller amphorae of kindred type, and three ‘bottle jugs.’ The door was in place, and the tomb full of earth and undisturbed, so that the tomb has almost certainly not been plundered. The equipment of this tomb, and the even scantier furniture of Tombs 27, 28, resemble closely that of the majority of the earlier Punic tombs at Carthage."

Tomb 61 had a slightly barrel-shaped roof, and contained an oenochoe, a degenerate flat-lipped jug like those of Carthage, a plate of red slip ware, two bowls, and a Cypriote lamp.

Tomb 62 had-a fully rounded barrel roof, and a second chamber behind the first. A distinct flight of rock-cut steps descended right into the tomb: the door-lintel being almost level with the roof, and the door-slab resting on the second step from the bottom. Though the door-slab was in place, and the tomb apparently undisturbed, there were found only a degenerate ‘tree pattern’ amphora, six ‘bottle jugs, and a one-handled wine-jar of type 4 (Fig. 13).

Late Graeco-Phoenician Tombs with Hellenic Importations. 111.---ΤΊ, Century.

Tombs 18, 29-80, 32-3, 39, are characterised by the presence of Hellenic black-glazed ware, all of poor quality and apparently of late date. Tomb 13, the best of these, is further assigned to a late date by the use, as its door-stone, of a limestone stele with the Phoenician inscription No. 2 (p.172), which itself cannot be earlier than the end of the fourth century. Other fragments of Hellenistic masonry were built into the sides of the doorway. The tomb itself contained a late lamp, some ‘tear bottles,’ and a black-glazed kantharos of the debased form which occurs also at Amathus, (e.g. (1894) 113, 119, 197, 211) and there seems to be uniformly late.

Tomb 39, which was found collapsed, yielded only a black-glazed phiale with stamped palmettes &c., and an unpainted vessel of very late Graeco- Phoenician type.

Tombs 29, 30, 32, 33, mere trial shafts nearer the line of the wall of ΚΊΤΙΟΝ, produced only fragments of black-glazed, and late red-figured ware, such as are strewn all over the surface of this part of the site. A fragment from Tomb 30 bore the graffito AHA////, No. 14 below: and the inscription No. 7 came from Tomb 33.

Hellenistic and Graeco-Roman Tombs with late Lamps, Glass, and Coins.

In this group, Graeco-Phoenician pottery is wholly absent, with the exception of the Cypriote lamp in Tomb 35. The tombs are uniformly long and barrel-roofed, and often have a second chamber behind the first; though

1 Delattre, Tombeaux Puniques 1890 ; Nécropole Punique 1896.

EXCAVATIONS IN CYPRUS IN 1894. 163

the partition is in many cases almost wholly decayed. As no detailed chrono- logy can as yet be attempted of this group, the tombs are described in the order of their discovery.

Lomb 18 contained many lamps and glass bottles, a poor mirror, and other fragments of bronze: one valve of a Pectunculus shell may have served as an ornament, or as a spoon. Cippi with inscriptions Nos. 11, 12 were found in the dromos.

Tomb 20, barrel-roofed, with a second chamber behind the first, contained sarcophagi of roughly-hewn alabaster: the contents were in confusion, but among a numerous late equipment there remained a fine late lamp with moulded top and triangular handle-plate; and a small stone altar with a human face between two trees, rudely incised on the front.

Tomb 22 closely resembled 20, both in plan and in contents: a large slab of gypsum in the floor of the first chamber, near the left-hand wall, may have served as a couch like those in Tomb 56 above. A number of the very late ribbed amphorae of gritty red earth were set upright in the corners on either side of the door. Among the glass was a painted lid, like ΟΜ. 2861 ff.; but the colours, and even the outline, were almost wholly defaced.

Tomb 35 was a long barrel-roofed chamber containing four plain sarcophagi, with room for two more towards the door. Its equipment was very elaborate, and particularly rich in glass, including an unusually large and thin glass plate, a small bottle of blue glass, and another of the blue and white streaked glass. It was in this tomb that the heterogeneous Cypriote lamp above men- tioned (p. 155) was found lying by an opened and rifled sarcophagus, all the jewellery of which had disappeared except one late gold earring of type e (C.M.C. p. 122).

Tomb 44, with two long barrel-roofed chambers, was entirely filled with earth, but a tunnel through its whole length yielded an elaborate late equip- ment, and fragments of a terracotta sarcophagus, with outward-turned rim, and square projections at the corners: nothing was recovered of the cover. A large plain alabaster sarcophagus stood on the left side of the hinder chamber. One of the lamps was some ten inches long, with triangular handle-plate, and stamped central medallion, unfortunately blurred by over- firing and vitrification of the surface, but apparently representing Eros struggling under a heavy tree trunk; perhaps masquerading as Herakles. This lamp was stolen from the store-room of the Teké before it had been photographed, but may reappear some day in a private collection. One gold earring of late type had two pearls, or perhaps decomposed glass beads, as pendants.

Tomb 45, of similar form to the preceding, yielded a Rhodian wine- amphora (C.M. 2024) bearing a rectangular stamp with a caduceus and illegible inscription: the caduceus reappears in the stamps C.M. 2313-5; and in C.M. 2321-2 associated with a grape-cluster and the name MIAA. The glass from this tomb was peculiarly rich, and included a saucer of millefiore glass (C.M. 2850) with a medley of composite coloured rods, and shreds of white and yellow glass, in a dark blue transparent ground. The tomb con-

M 2

164 EXCAVATIONS IN CYPRUS IN 1894.

tained also a gold ring of late type, with the gem missing (C.M. 4217), three late earrings (C.M. 4097, 8058-8072), and two nearly flat bronze dises with small handles in the middle (C.M. 3557,3559), which might be either miniature cymbals, or the covers of small toilet vessels.

The fragments of a marble stele, with the Phoenician inscription No. 3, was found in the shaft: a fragmentary akroterion is of the same marble, and probably from the same stele.

Tombs 14-16, 19, 24, 36, 38, 41, 46, 48-52, 54, 57 produced nothing of importance: 48-52 had been already rifled : 36 struck no tomb, but the large marble stele with the Phoenician inscription No. 1 was found face upwards about three feet from the surface: its position was about 100 ft. E. of the high road, and 20 feet S. of the lane leading into Old Larnaka on the north of the site.

V.—LARNAKA: KAMELARGA: GRAECO-PHOENICIAN SANCTUARY WITH VoTIVE TERRACOTTAS.

The circuit wall of ΚΙΤΙΟΝ, which, as above mentioned, forms a conspicuous escarpment in the open ground to the S.W. of Old Larnaka, has never been properly examined; and during the progress of the Turabi excavation, it seemed worth while to open a trial trench through the escarpment, in its highest part, where its total height is some 12-15 feet. The Commissioner of Larnaka, to whose hospitality and constant help I owe more than I could easily express, had put at my disposal the piece of ground marked K in the map (p. 149) lying behind his own garden, abutting south- wards on an unenclosed cart track leading from the 8. end of the main street of Old Larnaka towards the Turabt Teké, and northwards along the escarpment itself, on a narrower path, which at its junction with the main street aforesaid is lined with small houses, and dignified with the name of ‘ Leopold Street.’ This piece of ground goes by the name of the Kamelarga (Καμηλαργὰ : LL. camelaria), from a camel stable, which occupied it formerly but was pulled down a few years ago. The walls of this camel stable were remembered to have contained large stones which had been found on the site, so that there was reason to hope that some traces of the foundations of the wall of ΚΊΤΙΟΝ might still be traced at this point. As the ‘ Leopold Street’ footpath pre- vented a clear trench being cut from the actual foot of the slope, a beginning was made by a shaft above the path, near the E. boundary of the site, and at a point where two large stones appeared to be in situ. These proved, however, to be only part of the foundation of the camel stable, or of some other building, mediaeval at earliest, and neither here, nor on any other part of the site, was the Graeco-Phoenician masonry discovered. The whole of the wall itself in this quarter seems to have been destroyed, but the difference of level within and without its course indicates that its destruction here was compara- tively recent. Several large and well-squared blocks of compact limestone

EXCAVATIONS IN CYPRUS IN 1894. 165

which must be from it were found on the site and in neighbouring walls and enclosures ; one on the N.W. corner of the site measured 24’ x 20” x 14", and others, less perfect, by estimation about the same.

Attention, however, was wholly diverted from the town wall to a compact layer of votive terracottas which appeared in the original shaft A, about five feet from the surface, and proved to be about two feet thick. To test the extension of the layer a second shaft B was begun about six feet further east, and on the very edge of the site. Here also the layer of terracottas was found, but at a slightly lower level, and on tunnelling inwards the surface line was traced, until it joined that in shaft A. The west side of A was similarly enlarged by tunnelling as far as was safe, and the layer was found as far as the excavation extended ; but further progress in this direction was obstructed by the compact concrete foundation of the surface walls, which it would have been necessary to break up in order to go further. Towards the path also a series of the thick gypsum slabs marking old floors or thresholds at various levels above the terracotta layer made digging very difficult. It seemed clear, however, from a small trench beyond the road, that the heap did not exist undisturbed so near the surface of the slope. Probably it was formed against, or immediately within the circuit walls, and was a principal contributor to the rise of the ground-level within the town. Enough, however, of the layer was examined to make clear its general character, and the outline of its history.

The votive terracottas were all, with the few exceptions noted below, of the same rude fabric, and of approximately the same dimensions ; the height varying from five to eight inches. They were made by the simple method of turning, on the potter’s wheel, a deep funnel with slightly trumpet-shaped rim, and truncated point. Sometimes the funnel bulged at the point ; some- times it narrowed evenly, and in the latter class the paint was usually more elaborate, and the style of the head somewhat more advanced. This, when partly dry, was reversed and set upon its wide mouth: and into its narrow end was thrust the long stalk or neck of a solid clay head, the face of which had been impressed in a shallow mould. Several varieties of mould can be recognised ; some distinctly negroid (6... ΟΜ, 5549), and the majority of the mixed oriental style which passes for Phoenician, but none Hellenic: all are beardless; but if a male figure was to be indicated, a pointed beard, often of considerable length (Fig. 15. 14. 19.), was added in soft clay to the freshly moulded chin; consequently if the chin was already somewhat dry, the beard failed to adhere, wholly or in part. The joint between head and body was superficially welded with wet clay, but remained usually, like the junction of chin and beard, a very weak spot.

Similarly the arms were expressed by hand-modelled pellets of clay, and subordinate types of offering were distinguished by the gesture or by the addition of musical instruments, weapons, or various sacrificial offerings. Finally, the figures were dipped in a fine slip of paler colour than the clay, and painted; the hair, and outlines of the drapery in black; the face and sometimes the hands, in dark red, with the eyes in white outlined with black ;

166 EXCAVATIONS IN CYPRUS IN 1894.

and the garments in brighter red, bright dusty yellow ochre, and occasionally in a very powdery perishable blue.

Besides this normal funnel-bodied type, a number of simpler figures were found, with the body thinner, more columnar, and solid. These also had the head less distinctly moulded, and occasionally showed signs of hand-

a:

Fic, 15.—KAMELARGA. VoTIVE Figures. Cyprus Museum.

modelling. They therefore probably represent a survival of the earlier ‘snow-man technique’ which is common in Cypriote tombs of the ninth and earlier centuries, and is normal on the sites of the sanctuaries excavated by Dr, Ohnefalsch-Richter ! at Khytroi and Soloi.

2 Kuyrrol, K.B.H. p. 13, Pl. xl. xli, C. M.C.5201 ff. Sotor, K.B.H. p. 20, C.M.C. 5401 ff.

EXCAVATIONS IN CYPRUS IN 1894. 167

It does not follow, however, that they are wholly of earlier date, as they were found in approximately the same proportions at various depths and parts of the mass.

A further development in the other direction from the funnel-bodied figures, leads to a class made all in one piece, hollow throughout, and pressed in a full length mould. These were represented as fully draped, with a rounded or peaked headdress. They were all female and all carried tam- bourines. Fig. 15. 4.

The following principal types were distinguished among the funnel-

bodied figures :—

A. Tambourine players: either male or female; the tambourine being held either upright between the hands and at right angles to the body, or flat against the breast. Fig. 15, 3. 5.

B. Harp players: very rare and all female; of advanced style aud elaborate painting: the harp is of the oriental three cornered type, held on the left arm and played with the right hand. Fig. 15. 76.

C. Suppliants: fairly common and all female; the hands are pressed together, fingers upwards, in front of the breast. Careless workmanship con- fuses this type with ill-formed tambourine-players on the one hand, and with the ‘Oriental Goddess’ type (1) with arms crossed, or folded on the breast, on the other.

D. Mother and Child: a rare variant of the Suppliant type, connecting it with the types which follow. In a specimen in. the Ashmolean Museum the child extends its arms in an attitude of adoration; so it is clearly the votary with her infant, not a Madonna-Goddess, who is represented in this, as in the other types of this series.

E. Votaries bringing offerings; very common except at the top of the heap, and either male or female: the offering is frequently indistinct, and is either a flower, a wreath with crossed ends, a dish of cakes, a deep bowl with incurved rim, (in one instance, C.M. 5539, flower and bowl are combined; in another the bowl becomes a tall cup, and one hand is laid, in consecration- gesture, on the rim), a bird, (either swan, or dove); or a horned animal, calf orkid. Fig. 15. 4. 10. 11. 12. 14. 16. 17.

F. Lamp bearers: rare and all female: the figure is that of a votary, usually carrying a bowl; on the head is a Cypriote saucer-lamp, of the late (fifth—third century) type with flat rim, usually gaily painted. Fig. 15. 9.

G. Warriors: fairly common, especially in the lower part of the heap: of course all bearded and presumably male.2_ They wear a pointed Assyrian helmet of seventh century type, like that found by Prof. Petrie at Thebes in

1 For details v. C.M.C. pp. 6. 153-7. dite’ possibly suggested by warrior-statuettes 2 But was the story of the ‘bearded Aphro- with beards omitted or defaced !

168 EXCAVATIONS IN CYPRUS IN 1894.

1896 ; ashort sword girt high under the left arm, and very rudely represented ; and often also a round shield running out to a sharp central point ; to minimise the chances of breakage the front and back edges of the shield are usually pressed into the body, as though the shield were a soft petasos. Frequently the warriors carry also a sacrificial offering of type #. Fig. 15. 14. 19.

H. ‘Oriental Goddess’ .type: rare and always female: the arms are crossed in front of the body, and the breasts are prominently indicated. It is difficult to believe that this type alone of the series represents the Deity and not the worshippers: and it may probably be taken as representing a mode of self-devotion, of which in this instance we have no precise account.

How the last named type came into currency here is probably explained by the discovery of a number of examples of the well known solid moulded, flat-backed, strongly Egyptizing nude female figures, in a slightly different, and Nilotic-looking clay, but similar black and red paint. These came mostly from the middle and lower part of the heap, but are unfortunately not sufficiently characterized to serve as a datemark: their nearest analogues are from Cypriote tombs of XX VI. dynasty date. Fig. 15. 2. 6. |

Other miscellaneous objects worth mentioning are a number of fragments of larger hand-modelled statues, as large as quarter life-size, of local clay, but in a style resembling that of the ‘Toumba’ site at Salamis,! but exhibiting a wider range of influences. Some of them were strongly Egyptizing; others more purely Cypriote: one of the former showed well the spiral ornaments which are frequently found in sixth—fifth century tombs in the upper as well as in the lower lobe of the ear.2_ One face had been modelled separately as a mask for suspension, with perforated eyes (C.M. 5560).

The pointed caps of some of the little figures are illustrated on a larger scale by a detached male head C.M. 5555, and by a separate votive cap (C.M. 5556), which has never been part of a statuette: and the shields of the ‘warriors’ (type G.) by a round convex shield, with single κάνων behind, and red and black bands and rim-pattern of triangles outside? Fragments of horses and horsemen (C.M. 5562-4) and of votive doves, and a bull’s head modelled hollow for suspension, and painted black, complete the series of terracottas.

A few stone figures were found, which are important for the dating of the whole find, A female figure (C.M. 5571) in a Cypriote armless chiton and stole, and holding a tambourine,‘ seems to belong to the later sixth cen- tury, and corresponds with the tambourine playing terracottas. A male torso (Fig. 15. 13.), of very long, narrow proportions, with arms by the sides and painted indications of skin-tight striped vest and red loin-cloth or drawers, indicates a similar date. The early fifth century is represented by a small

1 J.H.S. xii. 116 ff. 96

2 Cf. K.B.H. xlviii. 2, lv. 7; C.M. 5981-2 1895 (2.1).

(Limniti ἢ). ; 131 3 Cf. that from Larnaka Turabi 60 (in Ashmo- * Fig. 15. 18. ΟΝ, Brit. Mus. A 9, 10, 15,

lean Museum). and one from Kurion (Brit. Mus. 18; K.B.H. Ixviii. 1, 13.

EXCAVATIONS IN CYPRUS IN 1894. 169

youthful male head (ΟΜ, 5575 = Fig. 15. 7.) of the rare archaic Cypriote school under Hellenic influences, which is represented by the marble torso from Poli and sundry heads from Dali and Akhna in the British Museum ; by other heads from Dali and Limniti at Berlin; and by three heads from Voni, and another from Dali and Tamassos in the Cyprus Museum!

On the other hand, a genre group (C.M. 5576= Fig. 15 7.) of a seated boy playing with a dog, though rudely cut, and gaily painted like the little figurines, cannot be dated earlier than the middle of the fourth century, and might be much later: but the total absence of any other object of Hellenistic style, coupled with the fact that this was found completely engaged in the heap, though not low down in it, precludes the reference of it to a much later date than the Ptolemaic conquest. A few charms and fragments of Egyptian glazed porcelain (C.M. 5577-8, 4712, 4766), a small stone incense altar C.M. 5579, and a single rude spindle whorl complete the tale of the discoveries.

From the date given above, it will be seen that the extreme chronological limits of the deposit are given by the character of the rudest figurines, which cannot be put higher than the seventh century; and by the group of the boy and dog just mentioned, which probably belongs to the end of the fourth: intermediate dates have been already indicated in the sixth and early fifth centuries; and some evidence has been stated to show that the top of the heap at all events is of slightly different, and presumably later character than the remainder.

The attributes, and offerings, of the votaries do not give a decisive clue to the presiding deity of the shrine. The mould-pressed nude female figures, and analogous types of tambourine players (but associated with flute-players, ΟΜ. 5302-3) and of votaries with drink offerings, occur on the site of the sanctuary of the ‘ Paphian Goddess’ at KuytTRo1:? and flowers, doves, cakes, lyres, and tambourines on that of Aphrodite in the lower town of IDALion.3 The preponderance of female statuettes also argues in favour of a female deity. But unfortunately no inscriptions came to light in the trial shafts; and the mass of superincumbent walls and floors prevented the excavation of any further parts of the early layer with the means which remained at my disposal.

A few specimens, for the most part uncatalogued, in the Salle des Origines of the Louvre, are so closely analogous in every particular that they may be regarded as certainly from this site, though their date of entry into the Louvre shows that they are not from the excavation of 1894. From the excavators’ share of the latter, duplicate sets of figurines have been offered to the British Museum, the Ashmolean and Fitzwilliam Museums, the Louvre, and the Berlin Antiquarium: and a number of examples of the commoner types are still available for distribution.

1 Voni C.M. 5005-7 ; Dali, C.M. 5642; Ta- 2 O0.M C. pp. 149 ff massos, 6083. Cf. C.M.C. p. 30, n. 2, for 3 C.M.C pp. 157 ff. further references.

170 EXCAVATIONS IN CYPRUS IN 1894.

VI. LARNAKA: BATSALOS.

On my first visit to Larnaka, Mr. Cobham called my attention to a passage of Colonna Ceccaldi’s Monwments Antiques de Chypre &c. (p. 19, reprinted from Rev. Arch. 1870 p. 26), in which he describes one of General L. P. di Cesnola’s excavations on the borders of the Salt Lake behind Larnaka. At the suggestion of Colonna Ceccaldi, workmen were employed on the conspicu- ous mound known as Batsalos, and indicated in the map (p. 149) at the east end of the causeway which crosses the lagoon. According to Colonna Ceccaldi,! who unfortunately gives neither plan nor further details :—

Ils mirent ἃ découvert des restes de magonnerie, substructions en petits moellons, enfouies peu profondément, et récolterent, éparpillés en divers endroits, un assez grand nombre de fragments de marbre portant tous sur leur bord, gravées en caractéres lapidaires assez nets, des inscriptions phéniciennes.

Un tertre de meme genre existe sur la rive opposée du marais, a un mille environ du Tekieh de la Sultane,? sur une langue de terre qui s’avance assez loin dans l'eau. Des travaux suivis améneraient peut-étre en cette endroit des découvertes d’antiquités égale- ment phéniciennes.

Acting on this suggestion I spent some time in examining, and eventually in testing by trial shafts, all the promontories on the west side of the lagoon north of the causeway. South of the causeway there is no promontory, nor any mound on the shore of the lagoon which would attract the attention of an observer standing on the Batsalos hill. In all cases we struck virgin soil— mere decomposed rock, immediately below the turf; and the only object of antiquity which was found was a fragment of a ‘hemispherical bowl’ of the painted white slip ware of Lakshd tu Rid: but this obviously proved no more than did the corresponding fragment found on the Turabt site.

Mr. Cobham told me that in spite of the difficulties of examination, he had assured himself that the ‘holy place’ enclosed by the Halé Sultana Teké was a megalithic monument like the chapel of the Phaneroméne nearer Larnaka.? But I was not able to come within sight of it.

Disappointed of a new site, I spent some days trenching the top of the Batsalos mound, which is a genuine hill of soft limestone, with two or three feet of barren soil on the top. A number of ill constructed chambers were traced, with walls of undressed rubble in strong cement. Most of the stones were small, but in some cases the foot of the wall was composed of blocks 2-3 feet long, but still unhewn. The only detail of the construction which could be traced was a cement-lined gutter draining the eastern side of the building

and falling into a shallow pit or

1 Colonna Ceccaldi gives a map in which Cesnola’s site is identified with Batsalos, but Batsalos projects northwards into the lake from the southern margin of its main basin. A misreading of the map has given rise to new errors in that given in the Corpus Incriptionum Semiticarum I. p. 85. General di Cesnola

cess-pool lined with rubble, a few

himself (Cyprus, p. 55 ff.) describes the site in terms which suit the Batsalos hill fairly well, but he places the scene of his operations on the South- West of the Salt Lake.

2 Halé Sultana Teké,

8 J0H.S...iv. ps 111.

EXCAVATIONS IN CYPRUS IN 1894. 171

feet from the N.E. corner. No architectural fragments were found at all: but on the surface one fragment of a white marble bowl with inscription No. 4 (Ashm. Mus.); the upper half of an Egyptian glazed porcelain statuette of XX VI. dynasty or later date (Ashm. Mus.) ; a fragment of a terra- cotta horse of snow-man technique (C.M. 5591); and two fragments of a fragment of a Hellenistic Draped female statuette (C.M. 5590) Attic black- glazed bowls with Phoenician graffiti (C.M. 1996, 1997). Cesnola’s men, or

the stone-hunters of whom he too complains (l.c.) had done their work only too well.

VII. ZARUKAS.

From this late Bronze Age site, I acquired, from a peasant, the following objects now in the Ashmolean Museum; they were said, and with probability, to have been found together :—

(1) a small flat saucer of green serpentine.

(2) an oval crucible (7) of greenstone, well worked, with a spout at one end, and a rudimentary handle at the other.

(3) a conical grinder or pestle of hard greenish limestone (?) ; too large to have been used with (1), which moreover shows no sign of grinding.

(4) two small oxen of ‘base ring ware’ attached to fragments of the upper surface of a vessel of undeterminable form. These give a fair date mark for the whole group.

(5) a very small conical bowl: pointed below, and only an inch across the rim: of quite rude hand-made fabric, and only noteworthy on account of a distinct flame-stain at one point of the rim: from which it is clear that the vessel has been used asa lamp. As there was no trace of grease or oil, this use cannot have been recent; consequently it may be presumed that we have here, so far as I am aware, the unique example of a Bronze Age lamp from a Cypriote site, and the only survivor of the predecessors of the Graeco- Phoenician saucer-lamp which was introduced, probably from Egypt, at the beginning of the sub-Mykenaean Age.

INSCRIPTIONS FROM THE EXCAVATIONS AT LARNAKA.

As above stated p. 152 the marble stele, now in the British Museum (No. 47), and published in the Corpus Inscriptionum Semiticarum (Vol. I. No. 44: Pl. viii. 44, 44a) was found at the point marked 7' in the map (Fig. 6. ef. the map in 6.1.8. p. 35). The following were found in the excavations of 1894 above described.

1, A stele of white marble, apparently Pentelic, in perfect condition ; in the form of a four sided obelisk, narrowing slightly towards the top, which is gable-shaped. The inscription is on the front, and about halfway down; in five lines of small clearly cut letters of the third century, perfect except one

172 EXCAVATIONS IN CYPRUS IN 1894.

which was bruised by the pick of the workman who cleared the face. The stele is exhibited in the British Museum (No. 31) outside the entrance to the Cypriote Room. The inscription was published, shortly, in the Academy No. 1238 (Jan. 25, 1896) by Rev. G. A. Cooke, Fellow of Magdalen College, Oxford, who hopes to discuss it at greater length elsewhere: and by Dr. Néldeke, from an impression, in the Zeitschrift fur Assyriologie 1894, p. 400. It runs as follows :—

This is the pillar which Arish, chief of the Stewards (?) erected to his father, to Parsi, chief of the Stewards (?), son of Arish, chief of the Stewards (?), son of Menahem, chief of the Stewards (?), son of Mashal, chief of the Stewards (?) son of Parsi, chief of the

Stewards (?) ; and to his mother, to Shenizabal, daughter of Baal-ram, son of Milk-jathan, son of ‘Azar, chief of the Treasurers, over the bed of their rest for ever.

The stele was found lying face upwards, but no trace remained either of its base, or of the ‘bed of their rest, the tomb of those whom it commemor- ates.

2. A stele of local limestone, very shelly and now rough on the surface from exposure ; of flat form with a low pediment. The inscription is in two lines close below the cornice, in large bold letters of the late fourth or early third century. Below the inscription, incised on the face of the slab, is an outline which appears to be intended for a plough or a chariot ; but only the upper part is preserved, as the stele has been broken obliquely across, and the lower part is missing. There is also a break on the left edge which damages two letters or so of the inscription. Published by Rev. G. A. Cooke, Academy no. 1287 (Jan. 16. 1896). The inscription reads as follows :—

To ‘Abd-ashtar, son of Eshmun, the chariot-smith : he made this. [May (the deity) bless him].

The allusion to chariot-making, and, probably, to a guild of chariot-smiths, is new. The stone was deposited with the Commissioner of Larnaka, and is No. 6231 of the Cyprus Museum Catalogue.

3. Two fragments of a stele of white marble, not unlike Parian, were found in the shaft leading to Tomb 45. One was part of the left hand akroterion, with a boldly worked palmette on the front face. The other was a fragment of the left edge, apparently of the same stele, with part of an inscription in large and coarse but quite clear letters of the same period as the preceding.

To Shamar-,fibaal, .. τς this pillar [j,,..,. . to them. [

It is published with the preceding in Academy no. 1237, deposited with it at Larnaka, and numbered 6232 in the Cyprus Museum Catalogue.

4. On part of the rim of a bowl of slightly bluish, perhaps Syrian marble, found on the surface of the atsalos site, is the following inscription, in small clear letters of the late fourth or early third century, on the flat upper surface of the rim.

‘.... his lord, the son of Melgarth, the son of Mikal... .’

EXCAVATIONS IN CYPRUS IN 1894. 173

The stone is in the Ashmolean Museum, and is published, with the pre- ceding in Academy no. 1237: cf. a note by Mr. ἃ. Β, Gray in Academy no. 1239, (Feb. 1, 1896).

General L. P. di Cesnola states (Cyprus 55 ff.) that he found a number of such inscribed bowls in his excavation by the Salt Lake (CLS. Cyprus 14 ff.); and it is possible that this fragment may belong to one of these, but it has not yet been possible to assign a place to it. Another fragment of a similar bowl, but uninscribed, was found on the Batsalos site in 1894.

5. Graffito on a fragment of black glazed ware, found close to the surface of the Batsalos site. =C.M. 5592/1996.

6. Graffito on a fragment of black glazed ware, found on the surface of the Batsalos site. = C.M. 5593/1997.

Greek Inscriptions. All from the T’wrabi and Hassan Effendi sites.

7. Fragment of a bluish marble like No. 4: found about 10 feet down, in Tomb 33, a trial shaft which led to a collapsed and rifled tomb: the inscribed face is slightly concave: the back is roughly dressed and has a large rectangular boss: small clear letters of the fourth century.

ATAOOKAHE MAKPQNOE ΠΕΡΞΗΞ

Compare a painted stele with the ethnic Βαβυλώνι[ος] from Amathus (1894), 109.

8.-13. All the remaining inscriptions were on the usual roughly turned cippt of local limestone. None presented any noteworthy feature, and for brevity the personal names only are given here, in the vocative as on the original, and in cursive text, as no squeezes were taken, The numbers in brackets are those of the tomb-shafts in which they were found: other cippi from (1) (5) had the name mutilated, or illegible.

8. (1) Προτώκτητε (sic.) 9. (5) Εὐτυχε."

10. (5) ᾿Αρέστων.

11. (18) ᾿Ολυμπίανη (fem.).

12. (18) Λεονέδη (masc.). |

13. (44) Mapov.

14. Graffito on a fragmentary black glazed bowl from the trial shaft Tomb 30.

ΔΗΛ

1 Cf, Εὐτύχη χρηστὲ χέρε copied together Munro, in the Turab{ Τοκό itself, J. H.S. xii. with ᾿Ονασκ[ρ]άτη χρηστὲ χαῖρε by Mr. J. A. RR, 522.

JoHN L, Myrgs,

TWO STELAE FROM KYNOSARGES. [Prats IV.)

THE two fragmentary stelae here published were discovered in the course of the excavations of the British School at Kynosarges in the winters of 1895 and 1896. One had been built into a Roman wall, the other into a late Hellenic watercourse. Within the area excavated there was no trace of any tombs of the period to which these reliefs may be referred, nor was any such trace to be expected. But the masons may very well have picked them up from close at hand; originally they may have stood by the side of a neighbouring road; indeed, they may be two of the actual tombstones desecrated by the soldiers of Philip."

The first of these two stelae attaches itself to an early group of grave- stones described and dated by Kohler in the Athenische Mittheilungen of 1885. The letters which remain of the inscription run ...ryov«... (perhaps Σ]τησικίλεια or Κ]τησικ[λεια) and end with a second σ᾽ on the other side of the break. The H is evidently the Ionic H, and the sigma has four limbs ; but this need not prevent the conclusion suggested by the style, that the stele was erected some time, but not a long time, before the outbreak of the Peloponnesian War. |

The field is surmounted by a pediment with acroteria in low relief. An advance towards the free pediment of later times may be seen in those stelae which end in an angle at the top of the central acroterion; like many of the earlier tombstones, such as that of Xanthippus in the British Museum (no. 618), our stele retains the original rectangular form, the surface only being cut away so that the pediment itself stands out in relief. It seems, however, to have been bounded at the sides by pilasters.

All that is left of the actual sculpture is the upper part of the figure of a woman. Her hair is confined by a band wound over it several times—a favourite arrangement in the second half of the fifth century. Between the circuits of the band the hair lies in many lines, exactly like that of Hegeso on the well-known relief. She wears a solid circular earring. With her left hand she holds her mantle out in front of her. The straightness and stiffness of the fingers are marks of the early date of the stele, yet the difficulty of representing in relief a hand so held is still evident in one work at least

1 Livy xxxi. 24, 18.

Te igeee ay ) 4 νυ .

ἢ ae "a. Ἢ ἐὺ " ἡ» “ A, Bi. he Ry ' : +, ΩΣ μ ed ie fy, Sar) ᾿ ; yi

) aes, ᾿ τ jan ων io a : a ea Ὁ" ἢ of, a

Fe ia * ‘y 4: Mm ' ar ᾽ an ; ἂν ᾿ f ? By A

‘aad ὶ wy top eee pee | PS ἃ + : ϊ

[Pes er es Ase ς aa δον ae ea ac ΕΣ

j ἝΞ. τῇ + Υ 5B ἡ P 1? Ἢ d ; Ν᾿. ͵ i 4} x ᾿ ( ihe a) eat on MLO Ρ ἢ ἢ

f κι

Seve ae

“ pas . " : a mere ᾿ 7 Ὗ 5 ‘ f wy See Me See clk ae τὸς eee , Joa T Ἂ ις Ζ ᾿ x rh ve « eS. mee = ae as Ψ ΔῊΝ, = ᾿ «εἷς ᾿ - , es ᾿ se elon set ἢ ἢ “we ~~ γὦ Εἰ \ 5 i ‘ ae αὐ - ὦ -_ Pe.) ἢ ᾿ a } "» νὰ 47 ‘ a 1 Ι Ν ; + _ Y rr. ; " Ἢ ee Cee ᾿ ee ἡ ἫΝ ' Uh 7 wes ~ τ, x τ ἐς “ “ΒΥ ' BA ἂν P 9 - Ὺ a ' é . ν᾿ 4 é Π 1.’ ae) ἢ 1 μι" ὡἧςς » - ΜΒ -. Ὁ". ἐ 4 ᾿ ᾿ νὺν “ἢ ᾽ν -” { ᾿ ΤΡ ων Σ Ὶ .» *y ’ ῥ ~~, hits Ι ties , » a © ᾿ oF \ i= A oi ᾿ uit * 4 isa * > ae AW? | ᾿ a) νι d ἦν “* re Ἴ ν΄ a RPS τ πον th a?

TWO STELAE FROM KYNOSARGES. 175

which is a full generation later (Acropolis Museum; ’Ed. ’Apy. 1893, PI. VIII.) ; there also the little finger has the same impossible bend.

The total breadth of the stele has been 48 centimetres. There is room for a second figure on the right and perhaps the Σ᾽ over this portion may be the last of a second name. The marble is Pentelic.

The second stele, also of Pentelic marble, is a considerably later specimen of the same class of work. The relief is low and is not framed at the sides by pilasters. The height of the fragment is 53) centimetres, its width 29.

The principal figure on the part which is preserved is a woman seated in an attitude of mourning. In the background beside her chair stands a bearded man who clasps hands with another figure of which only a slight fragment remains, probably the end of a himation wrapped round an upraised left arm. It is natural to infer that this lost figure represented the person in remem- brance of whom the tombstone was set up; for it is towards him that the gaze of the mourning woman is directed, while the second figure, which is placed in the background, can scarcely be meant for the chief member of the group. The arrangement of the figures in this scene (one of them seated in front and not taking part in the clasping of hands) is unusual.

The attitude of the seated lady is to be observed. One frequent type of mourning in Greek art is a figure with bent head, the right hand supporting the chin or cheek, the right elbow resting on the left hand. Here this mourning attitude is combined with another typical expression of bereave- ment, the upward look into the face of the lost friend: this latter motive is common enough in the fourth century but not until then. One conventional touch, very characteristic of Greek reliefs about the close of the fifth century, is the way in which the mantle hangs from the lady’s shoulder. There is the most exact similarity in this particular between our stele and the tombstone of Tynnias in the National Museum in Athens. (Conze, Gr. Gr. vol. ii. pl. 118, Gardner, Sculpt. Tombs of Hellas, pl. x.).

C. E. EpGAR.

1 See sketch of development of this type by Furtwangler, Coll. Sab. Pl. xv. text.

176 THE GREEK TREATISE ON THE SUBLIME.

THE GREEK TREATISE ON THE SUBLIME.

ITS MODERN INTEREST.

A FEW months ago the authorities of the British Museum announced the discovery of some of the lost odes of Bacchylides, the con- temporary of Pindar. Hitherto Bacchylides has been known only in fragments, none of them exceeding a dozen lines in length. Now we are promised over a thousand lines, of which as many as two hundred belong to a single ode. The discovery is, thus, one of considerable importance. Directly, it will reveal Bacchylides himself more fully; indirectly, it may be expected to throw light on many points of collateral interest. One such point is the relation of Pindar to the poetry of his time, another is the value of the literary judgments of some of the ancient critics. By both the links just mentioned the subject of this paper associates itself with the discovery.

The author of the Treatise on the Sublime frames an estimate of Bacchy- lides which seems likely to be confirmed by a fuller knowledge of his poems. He ranks him below Pindar for the same reason that he ranks Hyperides below Demosthenes. Correctness is not to be compared, says he, with genius; flawlessness is no match for inspiration. It is not a little remarkable that, in the case of Hyperides no less than in that of Bacchylides, the preservation of papyrus manuscripts in the sands of Egypt has enabled the modern world to test and verify this estimate. The result has been the enhancement of the already high reputation of the De Sublimitate. The treatise is, in point of fact, one of the best pieces of literary criticism be- queathed to us by classical antiquity, and any circumstance is welcome which helps to preserve it from the oblivion with which it has been often threatened.

The old saying that books have fortunes of their own is eminently true of the Treatise on the Sublime, with its many vicissitudes of neglect and celebrity. Of the early history of the work little is positively known. The tradition which ascribes it to Longinus, the celebrated minister of Queen Zenobia, has long been disputed ; and in the unsettled state of critical opinion upon the subject it seems better to treat the author as anonymous. Written at a date which within the limits of the first three centuries of the Christian era has been most variously assigned, the treatise appears to have remained almost unknown until it was printed by Francis Robortello at Basle in 1554 A.D., and by Paulus Manutius at Venice in the following year. Since that time it has been edited again and again. Dutch and French, as well as

ITS MODERN INTEREST. 177

Italian, scholars have done much for it. In the eighteenth century it reached the climax of its authority, and was regarded by Boileau and Pope almost as a final court of appeal. In the earlier part of the present century its popularity declined owing to various causes. A wider outlook over the world of literature and man reduced to their right proportions the extravagant claims of some of its admirers. Less legitimately, the ultra-scientifie tendency of classical scholarship in Germany led that country to devote more attention to the vexed question of the authorship of the treatise than to the elucidation of its contents. ' Latterly, however, there have been, in Germany as elsewhere, signs of a reaction. Critical texts have been attempted, and many transla- tions have appeared. At one time or another the book has been rendered into almost every European language, and within the last few decades versions of it have appeared in Spain, where Castilian illustrations of its precepts are freely offered ; in Italy, where the traditional interest in literary criticism and in this book in particular has produced excellent fruit; and in Sweden, where the vigorous modern school of Scandinavian literature thus connects itself with the past. In England, too, though no adequate edition exists, and no edition at all has been published within the last fifty years, signs of renewed interest may be found in the issue or reissue of several translations. :

The treatise, of which about one-third has been lost, has probably often suffered misconception through its customary English title. It has been thought to be at once more ambitious in purpose, and more narrow in scope, than it really is. The Greek title Περὶ “Yous, ‘Concerning Height or Elevation,’ does not convey that idea of abnormal altitude which is usually associated with the word sublime. The object of the writer rather is to indicate broadly the essentials of a noble and impressive style. In fact, if we were to describe the treatise as one on style, or even on literary criticism generally, we should be nearer the mark than if we connected it solely with the idea of ‘sublimity’ in the narrower sense. The author's own words make this plain, for early in his book he remarks that the friend whom he is addressing is too well versed in literary studies to need the reminder that sublimity is a certain distinction and excellence in expression, and that it is from no other source than this, that the greatest authors have derived their eminence and gained an immortality of renown! The friend in question is Postumius Terentianus,a Roman, who though young has had some experience in public affairs, and who is, like the writer himself, much interested in Greek and Latin literature. A rapid glance at the actual contents of the book will show the width of its range and indicate its true character.

At the outset the author, after offering the definition of sublimity just given, proceeds to ask whether there is such a thing as an art of the sublime.

11, 8: γράφων δὲ πρὸς σέ, φίλτατε, τὸν mai- καὶ συγγραφέων οὐκ ἄλλοθεν ἢ ἐνθένδε ποθὲν δείας ἐπιστήμονα, σχεδὸν ἀπήλλαγμαι καὶ τοῦ διὰ ἐπρώτευσαν καὶ ταῖς ἑαυτῶν περιέβαλον εὐκλείαις πλειόνων προῦποτίθεσθαι, ὡς ἀκρότης καὶ ἐξοχή τὸν αἰῶνα. τις λόγων ἐστὶ τὰ ὕψη, καὶ ποϊητῶν τε οἱ μέγιστοι

Η.5.- ΟΣ, XVII. Ν

178 THE GREEK TREATISE ON THE SUBLIME.

His answer is, that, though elevation of tone is innate, yet art can regulate the use of natural gifts. It is, he says, with diction as with life. A man favoured by fortune ought to know how to use his advantages; a writer of genius ought to profit by the help of art. In order to show that a systematic treatise can effect much in the way of warning as well as by means of precept, he gives a short account of defects of style which are opposed to sublimity. With this purpose he describes and illustrates the vices of tumidity, puerility, misplaced passion, and frigidity. This done, he further characterises the true sublime, and shows how it may be distinguished from false imitations. Next he enumerates five sources of the sublime. The first and most important of these is grandeur of thought—the power of forming great conceptions. This power is founded on nobility of character. Elevated thoughts are also, we are told, the result of the imitation of great models, of imaginative power, and of the choice and grouping of the most striking circumstances. The second source is vehement and inspired passion. While affirming that there is no tone so lofty as that of genuine passion, the author does not treat of this topic in detail, but reserves it for a separate work. Third in order come figures of speech, such as adjuration (in illustration of which is given the famous oath which Demosthenes swore ‘by those who at Marathon stood in the forefront of the danger’), rhetorical question, asyndeton, and lastly hyperbaton or inverted order. The writer makes the general remark that a figure is at its best when the very fact that it is a figure escapes attention. The fourth source of sublimity is noble phrasing or diction. The chief element in this is the choice of proper and striking words, a choice which, he says, wonderfully attracts and enthralls the hearer, and breathes into dead things a kind of living voice. Other elements are metaphors, and similes, and hyperbole. Fifthly and finally comes elevation in the arrangement of words. Of this examples are given, and some remarks are added on such specific vices of style as arise from the use of too few words or too many, of too much rhythm or too little. The author concludes with a notable passage in which he endeavours to trace the causes of the dearth of great literature in his own day.

This short sketch of the contents of the treatise will indicate its relation to the general subject of style. When we come to particulars, this relation is seen to be still more intimate, and yet to imply no narrowness of view on the author’s part. His hints with regard to thought and expression are shrewd and helpful, all the more so that he is too broad-minded to have any super- stitious faith in such formal Rules of Style as used to be popular in England a generation or two ago under the shadow of his name. A few examples of his illuminative observations may be offered. Speaking of Demosthenes, he remarks how that orator shows us that even in the revels of the imagination sobriety is required.” His good sense is seen in his praise of familiar language when used in season. A homely expression, he says, is sometimes much more

‘xxx. 1: ἡ τῶν κυρίων καὶ μεγαλοπρεπῶν πράγμασι φωνητικὴν ἐντιθεῖσα. ὀνομάτων ἐκλογὴ θαυμαστῶς ἄγει καὶ κατακηλεῖ 3 xvi. 4: διδάσκων ὅτι κἀν βακχεύμασι νήφειν τοὺς ἀκούοντας... . .. οἱονεὶ ψυχήν τινα τοῖς ἀναγκαῖον.

ITS MODERN INTEREST. 179

telling than elegant diction, for it is understood at once since it is drawn from common life, and the fact that it is familiar makes it only the more con- vincing.! Of tumidity, or bombast, we are told that it seeks to transcend the sublime, and that it is a fault which seems particularly hard to avoid, but that if examined in the light of day, it fades away from the awe-inspiring into the contemptible.? An over-rhythmical style is condemned on the ground that it does not communicate to its hearers the emotion conveyed by the words but that conveyed by the rhythm. The author is the determined enemy of conceits and puerilities of all kinds, and he remarks that men fall into these errors because, while they aim at the uncommon and elaborate, and most of all at the attractive, they find that they have drifted into the tawdry and affected.* He expressly denounces that ‘pursuit of novelty in the expression of ideas which may be regarded as the fashionable craze of the day.’* ‘ Art is perfect,’ he says in one place, ‘ when it seems to be nature, and nature attains her end when she contains art hidden within her’; and again ‘ We should employ art as in every way an aid to nature, for the conjunction of the two may be held to constitute perfection.’ In this spirit he makes the remark, with reference to Demosthenes, that the tricks of rhetoric are hidden away in the blaze of the noontide splendour of sublimity and passion. ‘By what means,’ he asks, “has the orator here concealed the figure? Clearly, by the very excess of light. For just as all dim lights are extinguished in the glare of the sun, so do the artifices of rhetoric fade from view when bathed in the pervading splendour of sublimity.® Evidently with the critic who writes thus the judgment of style was, to quote his own words, ‘the last and crowning fruit of long experience.” Everywhere the man’s sincerity of purpose and clearness of vision are manifest, and a book written in this earnest and enlightened spirit does not soon fall out of date.

Furthermore, the treatise may be regarded as a disquisition not only on the formation of style, but on literary criticism generally. In proof of this, it is only necessary to add to the foregoing description of its contents the reminder that it is a perfect storehouse of quotations illustrating excellencies and defects both of manner and of matter, both of form and of spirit. Reference is made to as many as fifty Greek writers, whose dates range over something like athousand years. Some of these are quoted repeatedly, Homer oftenest of all, and after him Herodotus, Plato, and Demosthenes. The author’s quality as a critic is most decisively seen in his preference of the best. The second-rate writers of Alexandria, though nearer in time, are not suffered to eclipse the true classics of Greece ; they are quoted rather in illustration of defects than of merits.

1 χχχῖ, 1: ἔστιν ἄρ᾽ ὁ ἰδιωτισμὸς ἐνίοτε τοῦ κόσμου παρὰ πολὺ ἐμφανιστικώτερον' ἐπιγινώσκε- ται γὰρ αὐτόθεν ἐκ τοῦ κοινοῦ βίου, Td δὲ σύνηθες ἤδη πιστότερον.

7 iii, 1, 3, 4.

3 iii, 4 and iv.

4 γ΄, : τὸ περὶ τὰς νοήσεις καινόσπουδυν, περὶ ὃ δὴ μάλιστα κορυβαντιῶσιν οἱ νῦν.

YORU, 1, SERVI. 4.

6 xvii, 2: τίνι γὰρ ἐνταῦθ᾽ ὁ ῥήτωρ ἀπέκρυψε τὸ σχῆμα; δῆλον ὅτι τῷ φωτὶ αὐτῷ. σχεδὸν γὰρ ὥσπερ καὶ τἀμυδρὰ φέγγη ἐναφανίζεται τῷ ἡλίῳ περιαυγούμενα, οὕτω τὰ τῆς ῥητορικῆς σοφίσματα ἐξαμαυροῖ περιχυθὲν πάντοθεν τὸ μέγεθος.

7 vi.: ἡ γὰρ τῶν λόγων κρίσις πολλῆς ἐστι πείρας τελευταῖον ἐπιγέννημα.

180 THE GREEK TREATISE ON THE SUBLIME.

But in Homer we are bidden to admire such passages as speak of Ossa and Pelion; of Strife, ‘with her head in the skies and her feet on the earth’; of the Battle of the Gods; of the earth-shaking Posidon; of the ery of Ajax to Father Zeus ‘to slay, if slay he must, in the light; and of the yet more impress- ive silence of the same hero in the shades! Nowhere is the critic’s skilful touch better seen than where he treats of Homer. In drawing, for instance, a comparison between the Iliad and the Odyssey, he assigns the former poem to the poet’s vigorous manhood when he was at the height of his inspiration, the latter to his mellower age. ‘In the Odyssey Homer may be likened to a sinking sun, whose grandeur remains without its intensity.’ But he is careful to. add, ‘If I speak of old age, it is nevertheless the old age of Homer.’ ? Again, he has the rather happy remark that Homer ‘has made, as far as lay within his power, gods of the men concerned in the Siege of Troy, and men of the gods.” Altogether, it is refreshing to see how often and with what sympathy the latest of critics reverts to the earliest of poets, His admiration for noble literature has incidentally accomplished even more for Sappho than for Homer, though the former is but once men- tioned by him. In his tenth chapter, as an example of the proper choice and grouping of the most striking circumstances, he adduces, and in so doing has preserved for posterity, a fragment of Sappho’s poetry. The gist of his com- ment on the wonderful love-ode in question is that we see depicted in it not one passion only but a concourse of the passions. His critical acumen is, more- over, seen in the illustrations given, up and down his work, not only of sub- limity but of its opposite. One specific instance, in which the offender is Aratus, the Alexandrian poet, is worth a moment’s notice, as it seems to show that the writer of the treatise had at least a spark of humour in his composi- tion. Aratus is contrasted, to his disadvantage, with the great exemplar

Homer. When pourtraying a storm and threatened shipwreck, Homer speaks thus :— :

On some tall vessel, from beneath the clouds

A giant billow, tempest-nursed, descends :

The deck is drenched in foam ; the stormy wind

Howls in the shrouds ; the affrighted seamen quail

In fear, but little way from death removed.

Aratus, the author continues, tried to produce the same effect in the following line :— But one small plank their doom doth keep away.

He has thus described (so runs the criticism) the scene in terms that are neat and trivial rather than terrible. ‘Further, he has put bounds to the danger by saying A plank wards off doom. After all, it does ward it οὔ ἢ

1 vili., ix. Mb τ 7: 2 ix. 18 : ὅθεν ἐν τῇ ᾿Οδυσσείᾳ παρεικάσαι τις 4 Iliad xv. 624-8, Lord Derby’s translation. ἂν καταδυομένῳ τὸν Ὅμηρον ἡλίῳ, οὗ δίχα τῆς 5 x. 5, 6: πλὴν μικρὸν αὐτὸ καὶ γλαφυρὸν

σφοδρότητος παραμένει τὸ μέγεθος. ἶχ. 14: γῆρας ἐποίησεν ἀντὶ φοβεροῦ" ἔτι δὲ παρώρισε τὸν κίνδυ- διηγοῦμαι, γῆρας δ᾽ ὅμως μήοου. νον εἰπών ξύλον ἄϊδ᾽ ἀπείργει᾽. οὐκοῦν ἀπείργει.

ITS MODERN INTEREST. 181

Besides Aratus, other minor writers,such as Timaeus and Theopompus, are made to furnish examples of faults which should be shunned by those who wish to write in the elevated manner. But the author is of too fearless a nature to strike only at the lesser men. He assails the great writers, such as Herodotus and Aeschylus, where they seem to him to offend against the canons of good taste. He has the courage to say that Demosthenes is too austere to be graceful and witty, and that when he forces himself into jocularity, he does not excite laughter, but rather becomes the subject of it. And he makes bold to affirm with regard to Euripides, the idol of the rhetorician, that he is by nature anything but elevated, and that it is only by force put upon his natural disposition that he appears to rise to tragic heights.” In such comments as these, whether we agree with them or not, we recognise pieces of genuine literary criticism, and the literary critic stands revealed no less in the note of pleasant egotism which makes itself heard now and again in the course of the treatise, and in such general maxims as that the poet must himself see what he would have others see,—must, in fact, have his ‘eye upon the object.’

Nor are such well-known topics of criticism as correctness, the standard of taste,and the comparative method, neglected by the author. Upon the question of correctness he shows a breadth of view which is in marked contrast with the opinions commonly held (and by his admirers, strange to say) in England for a century or more from the time of the Restoration. He is no believer in what is faultily faultless ; he is a supreme believer in fervour and inspiration. Elevation with some flaws is, he cannot doubt, to be preferred to uniform correctness without elevation. The passage is a characteristic one and may be quoted at some little length :

‘I am well aware that lofty genius is far removed from flawlessness ; for invariable accuracy incurs the risk of pettiness, and in the sublime, as in great fortunes, there must be something which is overlooked. It may be necessarily the case that low and average natures do remain as a rule free from failing and in greater safety because they never run a risk nor aim at the sublime, while great endowments prove insecure because of their very greatness. Further, I am not ignorant that it naturally happens that all human things are always better known by their worse traits, and that the memory of errors remains indelible, while that of excellencies quickly dies away. I have myself noted not a few errors on the part of Homer and other writers of the greatest distinction, and I am anything but pleased with the slips they have made. But still I do not term them wilful errors, but rather oversights of a random and casual kind, due to neglect and introduced with all the heed- lessness of genius. Consequently I do not waver in my view that excellencies higher in quality, even if not sustained throughout, should always on a comparison be voted the first place, because of their sheer elevation of spirit if for no otber reason. Granted that Apollonius in his Argonautica shows himself a poet who does not trip, and that Theocritus in his pastorals is most happy, would you not, for all that, choose to be Homer rather than Apollonius ?...... Again, in lyric poetry would you prefer to be Bacchylides rather than Pindar? And in tragedy to be Ion of Chios rather than Sophoc'es?,,,... What need to add that each one of the great authors often redeems all his errors by a single sublime and

1 xxxiv, 3: ἔνθα μὲν γελοῖος εἶναι βιάζεται καὶ τὴν αὐτὸς αὑτοῦ φύσιν ἐν πολλοῖς γενέσθαι ἀστεῖος, οὐ γέλωτα κινεῖ μᾶλλον ἢ καταγελᾶται. τραγικὴν προσηνάγκασεν, 2 xv. 8: ἥκιστά γέ τοι μεγαλοφυὴς ὧν ὅμως

182 THE GREEK TREATISE ON THE SUBLIME.

happy touch, and (most important of all) that if one were to pick out and mass together the blunders of Homer, Demosthenes, Plato, and all the rest of the greatest writers, they would be found to be a very small part, nay an infinitesimal fraction, of the triumphs which those heroes can point to on every hand ?’?

Again, our author’s answer to the often-asked question whether there is any trustworthy test of the sublime—any sure standard of taste in literature— seems surprisingly modern because it is so permanently true:

‘When a thing is heard repeatedly by a man of intelligence, who is well versed in literature, and its effect is not to dispose the soul to high thoughts, and it does not leave in the mind more food for reflection than the words seem to convey, but falls, if examined carefully through and through, into disesteem, it cannot rank as true sublimity because it does not outlive a first hearing. For that is really great which bears a repeated examination, and which it is difficult or rather impossible to withstand, and the memory of which is strong and hard to efface. In general, consider those examples of sublimity to be fine and genuine which please all and always. For when men of different pursuits, lives, ambitions, ages, languages, hold identical views on one and the same subject, then that consensus of judgment, so to say, which results from discordant elements makes our faith in the object of admiration strong and unassailable.’ ?

No modern critic could formulate more precisely, in relation to literature, the quod semper, quod ubigue principle.

Modern in many ways, the author is in nothing more modern than in foreshadowing the application of the comparative method to the study of literature. It is easy to scoff at literary comparisons, and no doubt there is often much that is puerile and inept about them. But, as M. Brunetiére has pointed out, the ridicule comes with ill grace from those who celebrate so loudly the triumphs in our own day of comparative anatomy, comparative physiology, and comparative philology. In a sense science may be said to begin in comparison, in the effort to distinguish things that differ and thereby to bring out the true nature of each and all. At the same time it is well to remember the necessary limitations of the comparative method where literature is concerned. It is utterly out of place and futile, if its object is to place the great writers in an order of merit, and to establish a sort of literary hierarchy. And even where the aim is simply to bring out the distinctive points of contrasted authors, it should not be forgotten that the methods of the laboratory can never fully be applied to the analysis of the finest products of the human mind. In this matter it may not unfairly be claimed that the author assumes a judicious attitude. The comparison, already quoted, of a passage in Homer with a passage in Aratus is distinctly happy. And so, in its way, is the comparison between Homer in the Iliad and Homer in the Odyssey. And so, again, is the passage in which he compares, not the same poet in different works, but two orators of different countries, Demosthenes and Cicero, Speaking with due diffidence as a Greek addressing a Roman, he ventures the opinion that it is in profusion that Cicero chiefly differs from Demosthenes. The latter is like a thunderbolt or flash of lightning; the

former resembles a widespread conflagration which rolls on with all-devouring flames,’

ΦΧ ΚΠ, ΒΑ xxxvi, 2. 3. vii. 8, 4, 3 xii. 4,

ITS MODERN INTEREST. 183

In his use, however limited it may be, of the comparative method the author has the advantage over his great predecessors Plato and Aristotle, neither of whom knew any literature except his own. It is interesting to observe in what general features he agrees with, or differs from, these masters of literary criticism. With both he has this in common that he may often seem unduly verbal and philological,—may often seem to attach excessive importance to rhythm, to figures, and to questions of form generally. Not that it is so in reality. Rather, attention to such matters must be the backbone of criticism, and especially of early criticism. In other points the author resembles Plato much more nearly than he resembles Aristotle. He breathes the spirit of the Ion rather than of the Poetics. He is subjective rather than objective. He is an enthusiast rather than an analyst. He is better fitted to fire the young than to convince the maturely sceptical. He speaks rather of ‘transport’ or ‘inspiration’ than of ‘purgation’ or ‘the universal,’ He was not, to tell the truth, a man of deep and penetrating intellect like Aristotle, but he was nevertheless a critic of keen artistic sensibilities. His book does not offer the great luminous definitions contained in the Poetics, nor is it marked by the cool and searching scientific analysis by which that work is distinguished. Yet it may be that it supplies something of its own. +Aristotle but seldom makes us feel that there sometimes dwells in words a beauty which defies analysis because it is the direct expression of a human spirit and is charged with emotion as well as controlled by reason. Our author’s chief aim is, on the other hand, aesthetic rather than purely scientific. This difference in standpoint has had at least one noteworthy indirect effect. Let us suppose fora moment that every particle of ancient Greek literature had perished with the exception of the Poetics which is a fragment, or with the exception of the Treatise on the Sublime which is also incomplete. In the latter case we should at least possess the better anthology; we should be in a better position to form some conception of the supreme excellence of Homer, and Sappho, and other Greek poets. And this result would be due to the fact that the author's method is much less rigorous than that of Aristotle in the Poetics. He allows himself great liberty of quotation because he believes, like Mr. Matthew Arnold in our own age, that it is best to make free use of illustrations in order that the critic may help others, no less than himself, to feel their way in matters in which dogmatism is dangerous and advance must be tentative.

His catholicity has led him still further. Aristotle, notwithstanding his encyclopaedic learning, knew, as has been already said, no literature beyond his own. Our author refers not only to Latin literature but to Hebrew; and not the least interesting feature in his treatise is that we see in it the confluence of three literatures. Among the many literary critics from Aristo- phanes to the Alexandrians and Dionysius of Halicarnassus, and from Cicero to Quintilian and the author of the Dialogus de Oratoribus, he is distinguished by the account he takes of three several literatures. It is not impossible that he had been anticipated in this respect by the Caecilius to whom he so often refers, the Caecilius who is elsewhere described as ‘in faith a Jew.’

184 THE GREEK TREATISE ON THE SUBLIME.

But we cannot tell. All we know is that, when discoursing on noble thought as inspired by nobility of soul, our author writes: ‘The legislator of the Jews, no ordinary man, having formed and expressed a worthy conception of the might of the Godhead, writes at the very beginning of his Book of Laws, “And God said—what? Let light be, and it was; let earth be, and it was.” ’?

And here a word may fitly be said as to the connexion of sublimity, in the more restricted and more usual sense of the English term, with Hebrew influences. It has sometimes been maintained that sublimity, in this sense, is the peculiar possession of the Hebrew race and is unknown to the Greek classic writers. The contention is suggestive, but too absolute. The highest possible examples of sublimity, it may be urged, are to be found in such Hebrew writers as Ezekiel. Moderns like Milton, it may be further advanced, owe much of their sublimity, directly or indirectly, to Hebrew sources. But on the other hand we can hardly deny the quality, however rigorous may be our definition of it, to early Greek writers such as Homer and Aeschylus, and to the early phases of some of the more modern literatures. Are we, then, to look everywhere for Oriental influences, and not rather to seek the clue in the brooding wonder of primitive man wherever found? The whole question is too large and vague for summary treatment. In France, for instance, an eminent critic has suggested that the reason why the literature of his country is deficient in sublimity is that the I'rench translation of the Bible is a poor one and has never taken possession of the popular mind, while the English version is magnificent and has influenced English literary style for centuries. But surely the cause lies deeper than this. We must not forget that in French there is no essential difference between the vocabulary of prose and that of poetry. We cannot forget, either, Voltaire’s comment on the ‘darkness visible’ of Milton and on a similar expression in Spanish: ‘Ce n’est pas assez que l’on puisse excuser la licence de ces expressions, l’exactitude frangaise n’admet rien qui ait besoin d’excuse.’? That is quite an intelligible attitude to assume, but it is one which at once puts sublimity out of the question. We can imagine that Aristotle might have assumed it; so completely does he sometimes seem to regard poetry from the logician’s point of view. But such an attitude we should feel to be quite alien to the author of the Greek Treatise on the Sublime, and equally alien, we may add, to the author of the English treatise on the Sublime and Beautiful. Burke’s admirable work is notable, among many other things, for its striking quotations from the Old Testament and from Milton, and for its insistence upon the truth that sublimity is closely connected with a sense of uncertainty, obscurity, infinity. ‘A clear idea,’ he says, ‘is another name for a little idea, and then proceeds to quote from the Book of Job a passage whose amazing sublimity he considers to be principally due to the terrible uncertainty of the things described. Sublimity belongs, in fact, to the region of vastness and mystery. In a pregnant sentence Aristotle declares that a good style must be clear without being mean; lucidity is, from this point of view, the first essential. But when

1 ix.,9) 2 Voltaire, Oewvres, xiii. 441.

ITS MODERN INTEREST. 185

sublimity is in question, the order is reversed. First and foremost stands grandeur of conception, even if a certain obscurity of expression should follow in its train.

It has been seen that the word swblimity is, in its modern acceptation, too limited in scope to cover our author’s meaning. Shall we, then, do better to think of him as an exponent of what is sometimes called the grand style? This term has the advantage of possessing a wider range than the other. But it has also disadvantages of its own. It is not free from the suggestion of bombast and excessive elaboration. Against such vices our author strongly protests, and he wouid have been the last to eulogise a style whose brilliance may seem dazzling to one generation, but whose disappearance awakens satisfaction rather than regret in the mind of the next. His admiration is reserved for something much more permanent, a classic excellence. His attitude is that of one who cares little whether or no the grand style dis- appears if only the great style remains. And his view of the elements of a great style is at once a discriminating and a lofty one. He is too sound a verbal critic to overlook the importance of the more technical or scholastic side. But he is also too broad-minded to forget that greatness of style must ultimately repose on a much wider basis than that afforded by technical rules. His double standpoint is worthy of attention because it must have been rare in his own time and it cannot be said to be common in ours.

As a critic he sees that care and study are needed in the formation of a great style. And if proof of this fact were required, it would be necessary only to point to specific instances in ancient and in modern times. Writers hike Virgil and Tennyson perhaps bear the marks of elaboration upon them, and it would therefore be superfluous to refer to their known habits of work. But such carefulness has often characterised those authors whose seeming naturalness and spontaneity afford but little trace of it. Recent inquiries have shown what pains Burns and Keats lavished on their work. In antiquity there was a well-known story of the immense trouble taken by Plato in writing the exordium, so simple in appearance, of his Republic. In our own day the scholar who has endeavoured to make Plato an English classic is known to have given the greatest possible attention to the art of expression. The same thing might be shown to be true of writers like Cardinal Newman, and of more obviously self-conscious authors such as Mr. Matthew Arnold. Even where there is simplicity, it is usually a studied simplicity ; where there is ease, it is elaborate ease.

As to our author’s own style we sometimes feel, as perhaps might be expected from his theme, that he neither possesses, nor simulates the possession of, that business-like directness of exposition which is so effective when information or instruction is to be imparted, and which is so foreign to the atmosphere of a leisurely seclusion. Of succinct expression he has little to say in this treatise ; it does not belong directly to his present subject, and possibly he had already dealt with it elsewhere. But whether he had done so or not, we feel that he would not have desired to conceal any limitations or shortcomings which could fairly be alleged against himself. His book leaves

HS VOL, XVII. O

186 THE GREEK TREATISE ON THE SUBLIME.

upon the mind'the agreeable impression that he would have been quite ready to allow that there might well be defects in his own style and in his treatment of his subject. In his style he sometimes shows the faults of the late period at which he wrote, faults such as diffuseness and poetical phraseology. Similarly, in his treatment of his subject, he is apt to be too minute and to lose himself occasionally in technicalities. In fact, he does not escape the characteristic failing of the teacher who has to deal with pupils of all grades of intellectual apprehension ; now and then he appears to be unduly didactic and to verge upon tediousness.

But these are trifling blemishes, and we scarcely heed them in the presence of his deeply earnest purpose and his breadth of view. As his fourth chapter shows, no one could entertain less respect than he for mere bookishness. Nor could any one discern more clearly how mistaken is the view of those who regard style as an end in itself or talk glibly of ‘art for art’s sake’ Like the author of the Dialogue on Oratory, he sees in literature not a convention, not a matter of form, but the reflexion of a national life; a great style is evoked by great surroundings and great events. A few extracts will serve to illustrate his lofty conception of individual and of national morality, and his view of the relation of both to literature. ‘It is not possible,’ he says in a noble outburst, ‘it is not possible that men with mean and servile ideas and habits prevailing throughout their life should produce anything that is admirable and worthy of immortality.’1 Again he remarks, ‘I wonder, as no doubt do many others, how it happens that in our time there are men who have the gift of persuasion to the utmost extent and are well fitted for public life, and are keen and ready, and particularly rich in all the charms of language, yet there no longer arise really lofty and transcendent natures unless quite exceptionally. So great and worldwide a dearth of high utterance attends our age.’ The explanation he finds when he glances at the characteristic vices of the time: ‘The love of money (a disease from which - we all now suffer sorely) and the love of pleasure carry us away into bondage, or rather, as one may say, drown us body and soul in the depths, the love of riches being a malady which makes men petty and the love of pleasure one which makes them most ignoble.” Vast wealth leads to such vices as extravagance, insolence, shamelessness. The final result is that ‘men no longer lift up their eyes and there is no further regard for fame, but the ruin of all such lives is gradually consummated, and sublimities of soul fade and wither away and become contemptible when men are lost in admiration of their own mortal parts and omit to exalt that which is immortal.’? His own conception of man’s place in the universe is a lofty one: ‘Nature has

1 ix. 3. Probably no modern language can better reproduce the fine Miltonic roll of the

tutta la vita si danno pensiero ὁ sollecitudine di cose piccole ¢ servili, profferiscane alewna sentenza

author's style, with its long ear-satisfying words, than the Italian. Canna’s version of this passage, together with the sentence which immediately follows it in the original, runs thus: Perocché non é possibile, che womini, i quali per

mirabile ὁ degna del’ immortalita ; ma grandi sono, con’ ὁ naturale, le parole di coloro di cut siano serie le cogitaziont.

2 xliv. 1, 6, 8.

ITS MODERN INTEREST. 187

appointed us men to be no base nor ignoble animals; but since she ushers us into life and into the vast universe as into some great assembly, to be as it were spectators of her triumphs and the keenest aspirants for honour, straightway she implants in our souls the unconquerable love of whatever is elevated and more divine than we. Wherefore not even the entire universe suftices for the thought and contemplation within the reach of the human mind, but our imaginations often pass beyond the utmost bounds of space, and if we survey our life on every side and observe how surely the victory rests at every point with that which is striking, and great, and beautiful, we shall soon discern the purpose of our birth.’1 About a man who can write thus there is the profound moral gravity and the lofty eloquence which mark a Demosthenes or a Burke. The ethical fervour of the author’s style calls to mind his own saying that,‘sublimity is the echo of a great soul,’* He is himself a man of great moral endow- ments ; the misfortune was that he had fallen uponevil days. The heroic age was in the far past, and the present was, to him, a time of spiritual destitution, when men loved show and comfort, and were no longer earnest in the pursuit of perfection.

Beyond and above all other motives for high effort our author places the hope of literary immortality. It is strange to reflect that this motive should have counted for so much with him and should count for so little comparatively with the moderns. No doubt he remembered that the spell of Homer had been felt for a thousand years. But to us as we look back the precariousness of such immortality on the merely material side seems appalling. The preservation of a few manuscripts appears almost an accident when we think of such permanence as our author himself has attained. With the moderns, on the other hand, there is the security of the printing-press, and there is the wide diffusion rendered possible not only by this but by the spread of the English language, by the practice of translation into various tongues, and by improved means of communication generally. And yet the ideal is apt to be not higher but lower. Immediate popularity with its rewards is sought rather than the approval of the best judges in all ages. As a consequence, we find many passable imitations of elevated style, but few sustained efforts, few real works of art. The very language in which our author sets forth the other view strikes a modern ear as somewhat exaggerated and high-flown. Aim high, says he in effect ; match yourself with the great ; imagine that you are appearing before a tribunal of the finest writers of the past; take heed that you do not act an unseemly part before the bar of the future. ‘For if a man fears at the time that he will not utter anything to outlast his own life and age, the conceptions of his mind must be incomplete, blind, and as it were untimely born.’* When we remember his longing, so often expressed, for immortality, there is satisfaction in the thought that his book is still read, though probably by but few. Much of the literary criticism and the art criticism of his time has been lost or is ignored, but a niche is still, we may hope, reserved for the writer of the Treatise on the Sublime. No one would have recognised more readily than

REV, "2563. 2 ix. 2: ὕψος μεγαλοφροσύνης ἀπήχημα. 3 xiv, 3.

188 THE GREEK TREATISE ON THE SUBLIME. |

he that the author of a treatise on poetry or on rhetoric is not to be compared with a true poet or a true orator. But it is no mean thing to have upheld lofty ideals of artistic excellence, such ideals as need not shrink from the scrutiny of posterity. It is rather as the embodiment of a spirit, than as a formal system of rhetoric, that the treatise has continued to hold its own. It is not merely a code of laws ; it is an attempt to indicate and illustrate the noble temper of mind in which those who aim high should write. It is because of the spirit in which it is conceived that the book will not readily become obsolete; its rules are transient and will pass, its purpose is permanent and abides. W. Ruys RopBeErts.

RICHARD CLAY AND SONS, LIMITED, LONDON AND BUNGAY

THE GREEK TREATISE ON THE SUBLIME. ITS AUTHORSHIP.

WHEN Francis Robortello at Basle, in the year 1554, issued the editio princeps of the Greek Treatise on the Sublime, he attributed the work to ‘Dionysius Longinus. Διονυσίου Λογγένου ῥήτορος περὶ ὕψους βιβλίον are the words that are found upon his title-page. In this ascription he was followed by Paul Manutius, who in the next year (1555) published an edition at Venice. The fashion thus set by the earliest editors became universal. Edition followed edition in quick succession, and translations made the book known in almost every European country. But in all the editions and in all the translations, Longinus was assumed to be the author. It was the same with the foremost critics and writers of France and of England. Boileau was in this matter at one with the rest of the translators. His acquiescence in the general view was shared by Fénelon, Rollin, and Laharpe, and in England by Addison, Hume, Hurd, and Blair. Pope, in a well-known passage, speaks of the ‘ bold Longinus,’ whose ‘own example strengthens all his laws,’ And even the severely scientific Gibbon refers, with a touch of sarcasm, per- haps, in the adjective but with no touch of scepticism in the name, to the ‘sublime Longinus,’

An ascription so firmly rooted in the tradition of two centuries was not easily shaken, and even now it finds, here and there, unquestioning acceptance. But since the first doubt was raised at the commencement of the present century, the tendency of critical opinion has been, with some fluctuations, increasingly adverse to the old view. I propose to examine the evidence under the two heads, A.—EXTERNAL, and B.—INTERNAL; and in each case it will be convenient to treat first of the negative indications (ὖ.6. arguments drawn from silence, from omissions, etc.), and afterwards of the positive. And in order to clear the ground, I may anticipate so far as to say that an endeavour will be made to establish, in the light of the most recent research, two main propositions : (1) the external evidence in favour of the historical Longinus is of a dubious character; (2) the internal evidence seems to point to the first century rather than the third as the period during which the treatise was probably composed. Having said thus much, with no desire to prejudge the issue but merely in order to supply a guiding thread in a somewhat com-

H.S.—VOL. XVII. Ρ

190 THE GREEK TREATISE ON THE SUBLIME.

plicated discussion, I will now set forth, as impartially as may be, and under the headings already indicated, the considerations which may be advanced on the one side and on the other, !

A—EXTHRNAL EVIDENCE.

(a) NeGative. It is a remarkable fact that the Treatise on the Sublime is not quoted or mentioned by any writer of antiquity. So complete is the silence with regard to it that some have conjectured that 10 was written for private circulation only. Publication, they think, was deliberately avoided by its author, who was influenced either by modesty or by prudential motives. Its epistolary form may possibly be held to give some colour to this view.} At all events, the obscurity which surrounded it until it was printed was great, as great as its subsequent celebrity. The silence extends—and this brings us face to face with the problem before us—to those lists of the works of Longinus which we owe to Porphyry, Suidas, and others. The De Sublimitate is not by any of these authorities mentioned among the writings of Longinus, and the omission is the more striking that the treatise is no ordinary one. The seriousness of the difficulty has long been recognised by those who have regarded Longinus as the author. But the ingenuity of scholars has, as usual, proved equal to the occasion. They suggest that the περὶ ὕψους formed part of of φιλόλογοι (or ai φιλόλογοι ὁμιλίαι, as the title is also given), one of the attested works of Longinus. But while the possi- bility of this explanation cannot be denied, it should be remarked that it does not find any very obvious support in the character of the surviving fragments of οἱ φιλόλογοι, nor in the character of the περὶ ὕψους itsclf. The latter, to all appearance, occupies a position of its own as a polemical essay directed against the work of a writer who is named in its opening sentence. It may be added that in various passages (vill. 1, xxxix. 1, xliv. 12) of the De Sub- limitate the author seems to indicate that he had written, or intended to write, about Xenophon, about composition (σύνθεσις λόγων), and about the passions (τὰ στάθη); but these subject-headings, also, fail to appear in the lists of the works of Longinus.

(8) Positive, The absence of the treatise from the accredited lists of Longinus’ works, although it was felt to require explanation, caused no great uneasiness till the beginning of this century (1808 A.D.), when the Italian scholar Amati made an important discovery. He found that a Vatican MS, (no. 285) of the De Sublimitate contained the following inscrip- tion: Διονυσίου ἢ Λογγίνου περὶ ὕψους. Hitherto it had been taken for granted (by Robortello himself, no doubt, as well as by those who followed him) that all the manuscripts attributed the book to ‘Dionysius Longinus’ ;

1 Cp, G, Buchenau, De Scriptore Libri Περὶ Περὶ Ὕψου:λεγόμενον βιβλίον Κριτι- Ὕψους, p. 66, and A, Jannarakis, Els τὸ καὶ Enpetdoers, p. 8.

ITS AUTHORSHIP. 191

it was disconcerting, therefore, to find that one of them indicated ‘ Dionysius or Longinus’ as the author. But this was not all. Once curiosity had been aroused by Amati, another discovery followed. It was found that the same alternative was offered by the Paris MS. 2036, which dates from the tenth century and is by far the best of the existing codices of the De Sublimitate, if it is not their actual parent. True, the other title was also given in that MS.; but the new point to be noticed was that, just after the index of the ‘Physical Problems of Aristotle, the words Διονυσίου ἢ Λογγίνου occurred. Theyoccurred also, it was found, in MS. 985 of the Bibliothéque Nationale,’ And last of all, it was discovered (and for this final discovery we return from France to Italy) that a manuscript at Florence had, as the inscription on its cover, ἀνωνύμου περὶ ὕψους. The most surprising thing, perhaps, about all this new information, was that it was not obtained earlier, But the treatise had become so wedded to the name of Longinus that any hints to the contrary passed almost unheeded. Indeed, the variation in Codex Parisinus 2036 had been noted, a considerable time before Amati announced his discovery in the Vatican Library, by the German scholar Rostgaard ; but nothing came of Rostgaard’s observation.

But once it had been fairly opened, the question could not again be closed. A wide field for speculation was presented. The names of‘ Longinus’ and ‘Dionysius,’ without further specification, lent themselves to numerous conjectures. And even if, as seemed most probable, the names were to be understood of their two most famous bearers in the literary domain, the uncer- tainty became, in reality, not less but greater. For when a free choice is allowed between two men who stand inore than a couple of centuries apart, we feel justified in assuming that we have before us nothing more than the guess of some late authority who was himself in doubt and therefore named, alternatively, the two most distinguished critics he could call to mind. On this interpretation, the title might have run, as some one has suggested, Διονυσίου ἢ Aoyyivou ἢ ἄλλου τινός. It might, in fact, have been compressed into a single word, the ἀνωνύμου of the Codex Laurentianus,

1 During a recent visit to the Bibliotheque Nationale I have had an opportunity of examining P. 2036 and P. 985. In P. 2036 the περὶ ὕψους follows the Problems of Aristotle which occupy the greater part of the manuscript. The Problems are prefaced by an index or table of contents (forming fol. 1, 7. and v.). At the end of the index are added the words :

E + AIONYCIOY H AOFFINOY TI YYOYC +

At the beginning of the text of the treatise the heading is: + AIONYCIOY AOFFINOY TIEPIYYOYC: +

This title is distinguished from the other by the absence of the #, but it is also distinguished (and this appears to have escaped even Vahlen’s careful scrutiny) from it by the fact that a considerable space separates the first word from the second and the second from the third, while the third and the fourth are run together. It would almost seem as if (notwithstanding the absence of the ἤ) the reader were still offered his choice between Dionysius and Longinus, The same absence and presence of the %, and the same separation and non-separation, are to be observed in DP. 985, on f. 222 v. (beginning of the treatise) and f. 79 v. (index) respectively. ;

ΡΣ

192 THE GREEK TREATISE ON THE SUBLIME.

And here, while the question of the name or names found in the manu- scripts is under review, it may be pointed out that the traditional ascription of the treatise to Longinus had been felt to present a special difficulty on the score of nomenclature. But the difficulty, instead of encouraging a healthy scepticism, had led once more toa display of that ready ingenuity which is certainly no less characteristic of the conservative than of the inno- vator. The full name of Zenobia’s minister, as given by more than one ancient authority, was Cassius Longinus. How, then, account for ‘ Dionysius Longinus,’ which at best is a rather strange combination of a Greek and a-Latin name? The answer was ready to hand. Longinus in his youth had borne the Greek name of Dionysius, but later he adopted that of Cassius Longinus, in honour of some powerful Roman patron of that name: let us, therefore, designate him Dionysius Cassius Longinus. And so he was designated, until the discovery of the real inscription came to remind those interested in the matter that this elaborate theory was not only a baseless, but a perfectly gratuitous fabrication.

It has already been said that the Treatise on the Sublime is not quoted or mentioned by ‘any writer of antiquity.’ From that statement there is no occasion to recede; but before we leave the consideration of the external evidence, allusion should be made to certain passages from an external source which have sometimes been supposed to show a knowledge of the book. The source in question is the commentator John of Sicily (Ἰωάννης Σικελιώ- ts). The references which John of Sicily has been thought to make to the treatise are vague and disputable. But even if we were to assume for the sake of argument that they were definite and unmistakable, they would be of little importance; and for this reason. The date assigned by Walz to John of Sicily is the thirteenth century. Now, as we have seen, the Paris MS. 2036 of the De Sublimitate is supposed to belong to the tenth century. Accordingly John may have drawn any ideas he entertained with regard to the authorship of the treatise from that manuscript of it. He cannot,

therefore, be safely regarded as in any sense an original and independent authority.

B—INTERNAL EVIDENCE.

(a) NEGATIVE. The Treatise on the Sublime abounds in references to Greek authors and in quotations from them. Catholic alike in praise and blame, it ranges the centuries for its illustrations of good style or of bad. Bards of the prehistoric days of Greece, writers of its Attic prime, erudite

1 How precarious any arguments connected with John of Sicily are may be inferred from the fact that Emile Egger, who urged them in

des Savants (Mai 1884). Further details, if desired, may be found in Vaucher, Etudes Critiques sur le Traité du Sublime, pp. 57, 58,

the first edition of his Histoire de la Critique chez les Grecs (pp. 531-533), silently abandons them in his second edition and in the Journal

62, 63, and in Canna, Della Sublimita: libro attribuito a Cassio Longino, pp. 39, 40.

ITS AUTHORSHIP. 193

poets of the Alexandrian era, rhetoricians of the Augustan age,—all figure in its pages. But notwithstanding the great number of its references to writings of an earlier date, the Treatise (or so much of it as we now possess) makes no mention of any rhetorician, philosopher, or other writer belonging to the second or to the third century A.D. Here again the supporters of the traditional view that Cassius Longinus was the author are confronted by a grave difficulty. The gap is a truly remarkable one. How comes it that no reference is made to the rhetorician Hermogenes, who flourished during the reign of Marcus Aurelius Antoninus, and who (rather than Caecilius) might have been made the pivot of a book? How is it that Longinus, who was the centre of a wide circle, makes no mention of his companions in the schools or of his friends 4 How is it, lastly and above all, that he makes no mention of his enemies, some of whom presumably had written books? For granted that his taste may have been too fastidious to find examples of excellence in the writings of his contemporaries or of his more immediate predecessors, yet the task he set himself was the exemplification not only of the elevated manner but also of its opposite. And to go back for instances of defective style to Alex- andrian times or to a period earlier still, instead of attacking living offenders, would have, it is thought, meant magnanimity too great even for the man who, in the name of liberty, bade defiance to Aurelian and met his death unflinchingly.

(8) Positive. The internal evidence of a positive character is various in its nature and unequal in its value. It will be convenient to examine first that portion of it which relates to the names of persons. The evidential bearings of the prosopoyraphia, so to say, of the treatise are considerable.

I. PRosopoGRAPHIA. Under this head let us, following the example of the author in his book, start with Caecilius.

(1) Caecilius. The book opens thus: τὸ μὲν τοῦ Κεκιλίου συγγραμμά- τιον, ὃ περὶ ὕψους συνετάξατο, ἀνασκοπουμένοις ἡμῖν ὡς οἶσθα κοινῇ, Ποστούμιε Τερεντιανὲ φίλτατε, ταπεινότερον ἐφάνη τῆς ὅλης ὑποθέσεως, κιτιλ. Τὸ ἰβ clear from these words that Caecilius had composed an essay on the sublime, and that our author is dissatisfied with it. Now Caecilius was a rhetorician contemporary with Dionysius of Halicarnassus, of whom in fact he was a close friend! The question, therefore, arises whether it is probable ‘that in the third century a writer would follow, so closely as our author appears to do, the treatment which his chosen subject had met with in the reign of Augustus. To such a question, as to other similar questions propounded in this paper, one who entertains the most serious doubts as to the third-century authorship will nevertheless think it fair to reply that, though not likely, it is not impossible. For to borrow an illustration from

EE ὁ... στὸ. -ΘῬΟ---ςἘς-ς-.ς---ςς-ς---- ΎἬᾳζο τ

1 For some account of the life and literary οἵ Calacte: 8. contribution to the history of activities of Caecilius, reference may be made Greek Literary Criticism’ in the AMERICAN to an article by the present writer on ‘Caecilius JouRNAL o¥ PHILOLOGY, October 1897.

194 THE GREEK TREATISE ON THE SUBLIME.

another field, did not seventy years pass before a reply was made, by Origen, to the Zrue Word of Celsus? And on the fiery battle-ground of religious controversy one might expect that polemic would know no lengthy pause. A treatise need not, therefore, follow very closely in the train of one that suggested it. But on the other hand it must be admitted that this particular treatise is written with all the earnestness and ardour of a writer who is refuting the errors of a contemporary or a near predecessor. Hermogenes might have provoked a third-century antagonist to this display of zeal, but hardly Caecilius.

(2) Moses. Moses is not expressly named in the De Sublimitate, but he is unambiguously indicated in the well-known words of ὁ. ix.: ‘Thus it is also, that the Jewish lawgiver, no ordinary man, having formed and expressed a worthy conception of the might of the Godhead, writes at the very be- ginning of his Book of Laws, “God said” —what? “ Let light be, and it was: let earth be and it was,”’! It is sometimes contended that the mention of Moses tells in precisely the opposite direction to the mention of Caecilius ; it makes the third century more likely than the first. But even if this be admitted (and we can hardly admit any implication that such a reference to Genesis is out of the question in a Graeco-Roman author of the first century), there is still open to us the plausible suggestion that we should seek a connecting link in Caecilius himself. The author may possibly have had no direct knowledge of the Old Testament, but may have drawn this illustration from the tractate of Caecilius, who was ‘in faith a Jew. 2 The fact that the citation is not an exact one may be held, so far, to confirm the conjecture.

(3) Ammonius. At one time the occurrence in the treatise of this name seemed not only to supply a definite post-Augustan reference, but also to create a strong presumption that Longinus was the author. For it is recorded of Longinus that when a young man he had travelled widely, and that at Alexandria he had attended the classes of the leading Neoplatonists, and among them of Ammonius surnamed Saccas. But Ammonius, standing by itself, was, as F. A. Wolf cautiously observed, not an uncommon name, and identification must not be too hasty; further inquiry must be made before Ammonius Saccas, or any other Ammonius, was supposed necessarily to be meant. Some time after this useful word of warning and exhortation had been dropped, G. Roeper made an interesting discovery which he communi- cated in the year 1846 to the first volume of Schneidewin and Leutsch’s Philologus. Searching the Venice scholia to the Iliad, he found that an earlier Ammonius, a successor of Aristarchus at Alexandria, had written περὶ

τῶν ὑπὸ ἸΠ]λάτωνος μετενηνεγμένων ἐξ “‘Opunpov.? These words accord so

1 ix. 9: ταύτῃ καὶ ὃ τῶν ᾿Ιουδαίων θεσμοθέτης, οὐχ ὁ τυχὼν ἀνήρ, ἐπειδὴ τὴν τοῦ θείου. δύναμιν κατὰ τὴν ἀξίαν ἐχώρησε κἀξέφηνεν, εὐθὺς ἐν τῇ εἰσβολῇ γράψας τῶν νόμων “εἶπεν ὁ θεός᾽ φησί' τί; “γενέσθω φῶς, καὶ ἔγένετο' γενέσθω γῆ, καὶ ἐγένετο"

2 τὴν δόξαν ᾿Ιουδαῖος, Suidas.—A fuller dis- cussion of the passage translated above will be found in an article by the writer in the CrAs-

SICAL Review, December 1897.

3 Scuou, A HOMERI IL.{1x. 540: ἔρδεσ κεν" ᾿Αμμώνιος ev τῷ περὶ τῶν ὑπὸ Πλάτωνος per- ενηνεγμένων ἐξ Ὁμήρου διὰ τοῦ ¢ προφέρεται ἔρεζεν. Svumpas: ᾿Αμμώνιος ᾿Αμμωνίου ᾿Αλεξανδρεύς, ᾿Αλεξάνᾷρου γνώριμος, ὃς καὶ δι- εδέξατο τὴν σχολὴν ᾿Αριστάρχου πρὸ τοῦ μοναρ- χῆσαι τὸν Αὔγουστον,

ITS AUTHORSHIP. 195

well with the reference to Ammonius in the De Sublimitate (c. xiii.) that there can be little, if any, doubt that this is the Ammonius in question ‘Was Herodotus alone a devoted imitator of Homer? No, Stesichorus even before his time, and Archilochus, and above all Plato, who from the great Homeric source drew to himself innumerable tributary streams. And perhaps we should have found it necessary to prove this, point by point, had not Ammonius and his followers selected and recorded the particulars,’ ἢ

(4) Theodorus, Theodorus is mentioned in the third chapter: ‘A third, and closely allied, kind of defect in matters of passion is that which Theodorus used to call parenthyrsus.* Here the imperfect tense (ἐκάλει) may possibly imply that the writer had attended the lectures of this Theodorus, who can hardly be other than Theodorus of Gadara (or ‘of Rhodes,’ as he preferred to be called), who taught rhetoric to the emperor Tiberius, and who is often quoted by Quintilian.’ The way in which his name is introduced, without further preface or addition, seems to imply that its bearer was a recent, and (like Theodorus of Gadara) a well-known, authority.

(5) Cicero. The treatise contains a set comparison between Cicero and Demosthenes, introduced by the words: ‘And it is in these same respects, my dear Terentianus, that it seems to me (supposing always that we as Greeks are allowed to have an opinion upon the point) that Cicero differs from Demosthenes in elevated passages. For the latter is characterised by sublimity which is for the most part rugged, Cicero by profusion,’ etc. We are not concerned here with the substance of this comparison; its main interest for us lies in the fact that it was instituted at all. With regard to its bearing upon the date of composition, two considerations present them- selves: (1) references to Cicero in the Greek rhetoricians are excessively rare, and it would be hard to find a parallel to this passage of the De Sublimitate in any subsequent Greek work; (2) the passage had not only a parallel, but a precedent, in the lost dissertation (συγγραμμάτιον) of Caecilius. Plutarch is our authority for the statement that ‘the all- accomplished Caecilius...... had the youthful temerity to publish a comparison between Demosthenes and Cicero.’ ®

(6) Zerentianus, About the identification of the Terentianus to whom

1 xiii, 3: μόνος Ἡρόδοτος ‘Ounpixwtaros ἐγέ- vero; Στησίχορος ἔτι πρότερον ὕ τε Ἀρχίλοχος, πάντων δὲ τούτων μάλιστα ὁ Πλάτων, ἀπὸ τοῦ ὋὉμηρικοῦ κείνου νάματος εἰς αὑτὸν μυρίας ὅσας παρατροπὰς ἀποχετευσάμενος. καὶ ἴσως ἡμῖν» ἀποδείξεων ἔδει, εἰ μὴ τὰ ἐπ᾽ εἴδους καὶ οἱ περὶ ᾿Αμμώνιον ἐκλέξαντες ἀνέγραψαν.

2 iii. 5: τούτῳ παράκειται τρίτον τι κακίας εἶδος ἐν τοῖς παθητικοῖς, ὅπερ ὁ Θεόδωρος παρένθυρσον ἐκάλει.

3 Quintilian, Znst, Or., iii. 1, 17 : Theodorus Gadareus, qui se dici maluit Rhodium, quem studiose audisse, cum in eam insulam seces- sisset, dicitur Tiberius Caesar. Suetonius, 7%b., 57; saeva ac lenta natura ne in puero quidem

latuit: quam Theodorus Gadareus rhetoricae praeceptor et perspexisse primus sagaciter et assimilasse aptissime visus est, subinde in obiurgando appellans eum πηλὸν αἵματι repupa- μένον.

4 xii. 4: οὐ κατ᾽ ἄλλα δέ τινα ἢ ταῦτα, ἐμοὶ δοκεῖ, φίλτατε Τερεντιανέ, (λέγω δέ, εἰ καὶ ἡμῖν ὡς Ἕλλησιν ἐφεῖταί τι γινώσκειν) καὶ ὁ Κικέρων τοῦ Δημοσθένους ἐν τοῖς μεγέθεσι παραλλάττει. ὁ μὲν γὰρ ἐν ὕψει τὸ πλέον ἀποτόμῳ, ὁ δὲ Κικέρων ἐν χύσει, K.T.A.

5 Plutarch, Demosth. 3: 5 περιττὸς ἐν ἅπασι Καικίλιος...ἐνεανιεύσατο σύγκρισιν τοῦ Δημοσ- θένους καὶ Κικέρωνος ἐξενεγκεῖν,

196 THE GREEK TREATISE ON THE SUBLIME.

the treatise is addressed, and whose name occurs at its commencement, as well as in the passage just quoted and in several others, it will be convenient to inquire a little later.

(7) Πυγμαῖοι, Koroaads, Πυθέα. Lastly a few miscellaneous names may be added to the personal names already given. The Pygmies are referred to in a curious passage of 6. xliv.: ‘Just as, he proceeded, the cages (if what I hear is true) in which are kept the Pygmies, commonly called nani, not only prevent the growth of the creatures confined within them, but also lessen their original size through the bonds which hamper their bodies, so one may term all servitude (though it be most righteous) the cage of the soul and a public prison-house.!_ The point here is that the exhibition of Pygmies seems to be regarded by the author as a novelty (εἴ ye τοῦτο πιστὸν ἀκούω). This would, it appears, apply best to the period of the early Caesars ; afterwards the thing became more common.? But manifestly an argument of this nature cannot be pressed. The gaps in our information are too formidable to allow us to draw, without hesitation, such inferences as the one just suggested, or the allied one that the author must have been living at a distance from the capital when he wrote the passage. Still more precarious are any arguments based on ‘ the faulty Colossus’ (xxxvi. 3), or on the Pythia (xii. 2). It has been maintained that by ‘the faulty Colossus’ must be meant, not the Colossus of Rhodes, but that of Nero, which was renovated under Vespasian ; and it has been pointed out that the Pythian priestess ceased to give oracles under Domitian, resumed her activity under Hadrian, and became finally extinct under Caracalla. Pieces of evidence so indecisive as these are added rather in the hope of making the review complete than of proving any special point.

II. StyLE AND VocaBULARY. Arguments drawn from style and vocabulary are notoriously insecure, and to be of any value at all they must be based upon a long and minute analysis, for which there is here no space. I shall, therefore, pass lightly over this branch of the controversy, reserving a fuller statement for some future occasion. In general terms, however, it may here be said that the style of the De Sublimitate is thought to differ substan- tially, especially in its marked eloquence, from that of the fragments of

1 xliv. 5: “ὥσπερ οὖν, ef ye’ φησί “τοῦτο πιστὸν ἀκούω, τὰ γλωττόκομα, ἐν ols of Πυγμαῖοι, καλούμενοι δὲ νᾶνοι, τρέφονται, οὐ μόνον κωλύει τῶν ἐγκεκλεισμένων τὰς αὐξήσεις ἀλλὰ καὶ συνα- ραιοῖ διὰ τὸν περικείμενον τοῖς σώμασι δεσμόν, οὕτως ἅπασαν δουλείαν, κἂν ἢ δικαιοτάτη, ψυχῆς γλωττόκομον καὶ κοινὸν ἄν τις ἀποφήναιτο δεσμω- τήριον. συναραιοῖ (‘but actually attenwate them’) is a recent conjecture of W. Schmid, Rhcinisches Museum, 111. (1897), p. 446. MSS, συνάροι.

* It may be remembered that the presence of a nain (to use the French form) on the so- called Bayeux tapestry is interpreted as an indication of the date of its production, but

only because the addition of the dwarf’s name Turold may be taken to imply contemporary knowledge of the events portrayed. ‘Souvent, dans une discussion de ce genre, ce sont les moindres détails qui fournissent les meilleures inductions,’ as M. l’Abbé J. Laffetay remarks. —Much recondite information with regard to the Pygmies both in ancient and in modern times will be found in B. A. Windle’s edition of Edward Tyson’s ‘Philological Essay con- cerning the Pygmies of the Ancients,’ one of the volumes included in Nutt’s Bibliotheque de Carabas,

ITS AUTHORSHIP. 197

Longinus admitted to be genuine. This is the opinion of many critics, and among them of Vaucher, who has edited and translated all the remains of Longinus with the utmost care.' Vaucher has also brought out very clearly the many points of contact between the vocabulary of the author and that of Plato, of Plutarch, and of Philo. But in the matter of comparisons founded upon style and vocabulary there are, as has already been said, many uncer- tainties. There is the fact that critics disagree so widely in their judgments upon such matters. There is also the fact that an author's manner of writing may, at one period of his life or when he is writing upon one subject, differ altogether from that which characterises him at another period of his life or when writing upon another subject.’ There is, further, the danger of incom- plete investigation. To illustrate this last point, it may be mentioned that it was once urged, as evidence of late authorship, that the word ἀλληγορία, found in the treatise, did not occur before Plutarch’s time. This often- repeated statement was a rash one in any case, in view of the fact that we possess only a few fragments of the writings of antiquity, but it did not even take full account of the materials we actually possess, As a matter of fact, the word occurs twice in Cicero, by whom it was probably derived from Stoic sources.

III. GENERAL ConTENTS. This heading is still more vague than the last. But it may be useful to inquire whether the writer’s habits of thought and intellectual standpoint seem to be those of the first or of the third century, and with which of the two he has the closer literary and spiritual affinities.

His subject is elevation (ὕψος) of style, and this, he holds, depends ultimately upon elevation of character. ‘Sublimity is the echo of a great soul’ (ὕψος μεγαλοφροσύνης ἀπήχημα, ix. 2).3 The breadth of view, here evinced and elsewhere prominent, is a distinctive feature of his treatise, and seems, as we shall see presently, to ally him rather with the Roman writers of

the first century, than with any Greek writers whether of the first century, or the third.

1 Cp. Vaucher, op. cit., p. 50: la différence Treacherous always, such comparisons are

sensible que Von remarque entre le style simple et égal des fragments de Longin, et le style animé, véhément, figuré du Traité περὶ ὕψους, dont le sujet, quot qu'il en dise, ne prétait pas plus ἃ Véloquence que ceux des Fragments. Cp. also ib. 68-72, 383-442 Ruhnken, it is true, took another view, but he is not supported in it by his modern successor Cobet.

A signal instance of such variation in our own day is that afforded by the style of Thomas Carlyle. Let it be supposed that nearly two millenniums had passed since he wrote, and with what confidence we can imagine the position assumed and maintained that Carlyle the Edin- burgh reviewer and Carlyle the philosopher of Chelsea could not possibly be identical.

doubly treacherous when advanced concerning men of marked individuality who have been driven, more and more, into themselves by the circumstances of the times in which they live.

3 Itis convenient, asa rule, to adhere to ‘ sub- limity’ or ‘the sublime’ as the accepted ren- dering of ὕψος. But the English expressions are apt to mislead, by reason of the existence of Burke’s treatise particularly. It is perhaps regrettable that the earliest English titles of the De Sublimitate (‘Of the Height of Eloquence,’ John Hall’s Translation, 1662; ‘Of the Lofti- ness or Elegancy of Speech,’ John Pulteney’s Translation, 1680) have not held their ground in some slightly modified form,

198 THE GREEK TREATISE ON THE SUBLIME.

But a word must first be said about the narrower or more scholastic side of the treatise. This offers more obvious—we must again make every allowance for possible defects in our information—points of contact with the Greek and Roman rhetoricians of the first century than with those of the third. In his rhetorical terminology, and it may be added in his literary judgments, the author is distinctly at variance with the views implied in the surviving fragments of Longinus, whereas on a similar book by Caecilius our treatise is in a certain sense based and it would seem to follow that essay more closely than its combative tone might on a first reading suggest.’

Between the De Sublimitate and Quintilian, again, the points of resem- blance, especially where the rhetorical figures are concerned, are many and unmistakable.2 So remarkable, indeed, are they that some have thought that both the author and Quintilian must be drawing on Caecilius.? But the whole question of the exact relationship between Caecilius, Dionysius of Halicarnassus, Quintilian, and the De Sublimitate, though highly interesting, seems with our present data hopelessly insoluble. The important point at present to be observed is that there 7s a close affinity in these cases, and also in that of the Dialogus de Oratoribus to which reference will be made ‘immediately.

But besides its decided Roman affinities, the treatise sends out its roots in other directions also. That it has points of contact with the Jews has already appeared. But here direct reference may be made to passages in two first-century Graeco-Jewish writers, Josephus and Philo. The passage of Josephus (Antigg: Iud., ad. init.) is: ἤδη τοίνυν τοὺς ἐντευξομένους τοῖς βιβλίοις παρακαλῶ τὴν γνώμην θεῷ προσανέχειν, καὶ δοκιμάζειν τὸν ἡμέτερον νομοθέτην, εἰ τήν τε φύσιν αὐτοῦ ἀξίως κατενόησε καὶ τῇ δυνάμει πρεπούσας ἀεὶ τὰς πράξεις ἀνέθηκεν. That of Philo (De Hbrictate, 198 ; vol. ii., p. 208, in Cohn and Wendland’s edition, 1896-97) is: ἐγὼ δ᾽ οὐ τεθαύμακα, εἰ πεφορη- μένος καὶ μιγὰς ὄχλος, ἐθῶν Kal νόμων τῶν ὁπωσοῦν εἰσηγμένων ἀκλεὴς δοῦλος, ἀπ᾿ αὐτῶν ἔτι σπαργάνων ὑπακούειν ὡς ἂν δεσποτῶν ἢ τυράννων ἐκμαθών, κατακεκονδυλισμένος τὴν ψυχὴν καὶ μέγα καὶ νεανικὸν φρόνημα λαβεῖν μὴ δυνάμενος πιστεύει τοῖς ἅπαξ παραδοθεῖσι καὶ τὸν νοῦν ἐάσας ἀγύμναστον ἀδιερευνήτοις καὶ ἀνεξετάστοις συναινέσεσί τε καὶ ἀρνήσεσι χρῆται. If these two passages be compared, the first with De ϑιιδί. ix. 9, and the second with De Subl. xliv. 3, 4, the close parallelism will assuredly cause surprise. But of course such parallelisms do not furnish mathematical demonstration of a first-century authorship; on the contrary, they would be consistent with the claims of the historical Longinus. The same may be said of any resemblances between the treatise and the writings of Plutarch, resemblances which often have their origin in a common admiration of Plato.

1 Cp. Vaucher pp. 73 segg., and Canna pp. 3 Coblentz 54, 58, 59. 23-26, for Longinus ; for Caecilius, cp. AMERI- 4+ The more we investigate, the more certain CAN JOURNAL OF PHILOLOGY (as cited above), | weare as to the existence, and the less certain as and the dissertations by Martens and Coblentz to the particular origin, of a vast floating mass named in the note at the end of this article. of literary criticism contained in the rhetorical 2 Vaucher 45 n., 85, 201; Canna, 21, 22, writings of the first century.

ITS AUTHORSHIP. 199

Traces of Stoicism, also, or of Alexandrian influences, are in themselves little to go upon; nor yet can we safely build an argument upon the analogies drawn from the realm of art with which the treatise illustrates and enforces its literary precepts, though we are at liberty to point out that such analogies are very frequently employed by writers of the first century.

More is perhaps to be expected from an examination of those speculations with regard to the causes of the decline of eldquence which are found in c. xliv. of the treatise. Here are some extracts from the chapter: ‘It remains, however (as I will not hesitate to add, in recognition of your desire for learning), to clear up, my dear Terentianus, a question which a philosopher recently started in conversation with me. “1 wonder,” he said, “as no doubt do many others, how it happens that in our time there are men who have the gift of persuasion to the utmost extent, and are well fitted for public life, and are keen and ready, and particularly rich in all the charms of language, yet there no longer arise really lofty and transcendent natures unless quite exceptionally. So great and world-wide a dearth of high utterance attends our age. Can it be, he continued, that we are to accept the trite explanation (πιστευτέον ἐκείνῳ τῷ θρυλουμένῳ) that democracy is the kind nursing-mother of genius, and that literary skill may be said to share its rise and fall with democracy and democracy alone? For freedom, they say, has power to feed the imagina- tions of the-lofty-minded and inspire hope, and therewith there spreads the eagerness of mutual rivalry and the emulous pursuit of the foremost place . +». One may term all servitude, though it be most righteous (κἂν 7 δικαιοτάτη), the cage of the soul and a public prison-house.” I answered him thus: ‘‘It is easy, my good sir, and characteristic of human nature, to find fault with the age in which one lives. But ¢onsider whether it is not the world’s peace (ἡ τῆς οἰκουμένης εἰρήνη) that ruins great natures, but far rather this war illimitable which holds our desires in its grasp, aye, and further still those passions which occupy as with troops our present age and utterly harry and plunder it. For the love of money (a disease from which we all now suffer sorely, πρὸς ἣν ἅπαντες ἀπλήστως ἤδη νοσοῦμεν) and the love of pleasure make us their thralls, or rather, as one may say, plunge the ship of our lives in the depths with its human crew, the love of riches being a malady which makes men petty, and the love of pleasure one which makes them most ignoble..... The same is true where the entire life of each of us is ordered by bribes, and huntings after the death of others, and the laying of ambushes for legacies (Sexacpol καὶ ἀλλοτρίων θῆραι θανάτων καὶ ἐνέδραι διαθηκῶν), while gain from any and every source we purchase—each one of us—at the price of life itself, being the slaves of pleasure. Can it be that, in an age which is ravaged by plagues so dire, we think that there is still left an unbiassed and incorruptible judge of works that are great and likely to reach posterity, or is it not rather the case that all are influenced in their decisions by the passion

1 For these analogies reference may be made Brzoska’s dissertation De Canone Decem Ora- to Εἰ. Bertrand, De Pictwra et Sculptura apud torum Atticorum Quaestiones, Veteres Rhetores, and to the appendix to

200 THE GREEK TREATISE ON THE SUBLIME.

for gain? Nay, it is perhaps better for such as we are to be ruled than to be free (ἀλλὰ μήποτε τοιούτοις, οἷοί περ ἐσμὲν ἡμεῖς, ἄμεινον ἄρχεσθαι ἢ ἐλευθέροις εἶναι), since our appetites, if let loose ina body upon our neighbours like beasts from a cage, would overwhelm the world with calamity”’ (xliv. 1, 25,6, 9; 10).

The drift of these passages is plain. The decline of eloquence may be traced to the decay of liberty, or it may be traced to the spread of wealth and luxury. The lament of liberty seems to be uttered with some timidity, and to be placed in another’s mouth. It seems to be implied more than once that the servitude may be a just servitude. But the main point is that the lament should be made at all. Nothing of the kind, I think, is found in similar writings subsequent to the first century—in Lucian, or Aristeides, or Maximus of Tyre. In the first century, on the other hand, the topic was a commonplace (ἐκεῖνο τὸ θρυλούμενον) of Roman literature, and as such doubtless it is reflected in our treatise.’

Much the same may be said of the reference to the evil influence of riches. With "HAH vocodper in the passage translated above Cobet aptly compares Livy’s “NUPER divitiae avaritiam et abundantes voluptates desiderium per luxum atque libidinem pereundi perdendique omnia invexere.” And Cobet further asks how Longinus could have so written of his contem- poraries as the author does in the words which follow those just quoted: ‘num Longinus aut Graeci aut Syri accipiebant pecuniam ob rem iudicandam aut mortibus alienis inhiabant aut malis artibus heredipetarum utebantur ? Romana haec sunt vitia et flagitia.”

CONCLUSION.

We take it, then, that in the Treatise we hear the voice of a dying liberty, not of a liberty long since dead. We seem to catch the accents of a Tacitus. Those words ἅπασαν δουλείαν, κἂν ἢ δικαιοτάτη recall the bitter sarcasm of the Annals (vi. 8): tibi summum rerum iudicium di dedere ; nobis obsequii gloria relicta est. The phrase ἡ τῆς οἰκουμένης εἰρήνη reminds us of the Dialogus (xxxviil): postquam longa temporum quies et continuum populi otium et adsidua senatus tranquillitas et maxima principis disciplina ipsam quoque eloquentiam sicut omnia depacaverat.

The parallelism, seen not in the point just mentioned only but in many others, between the Dialogus and the De Sublimitate, might well form the subject of a separate paper. The opening sentence of the Dialogus breathes

1 For various references to the degeneracy lost, on the decay in prose composition, De and its causes, see Seneca, Hp. 114; Pliny, Causis Corruptae Eloquentiae: ep. A. Reuter, Hist, Nat. xiv. 1; Plin. iun. Zp. viii. 14; De Quintiliani libro qui fuit De Causis Cor- Tac. Dial. de Orat, xxix. xxxvi. xxxvii.; Vell. ruptae ELloquentiae, Vratislaviae, 1887.

Paterc. Hist. Rom. I. 17; Petronius, Satyr. 2 Mnemosyne, N.S., vii. 421. Ixxxviii. ; Quintil. Znst. Orat. ii. 10, 3 ete. 3 Compare also ce, xxxvi., xxxvii., ibid. Quintilian further wrote a separate treatise, now

ITS AUTHORSHIP. 201

the very tone and spirit of the Treatise on the Sublime: Saepe ex me requiris, Iuste Fabi, cur, cum priora saecula tot eminentium oratorum ingeniis gloriaque floruerint, nostra potissimum aetas deserta et laude eloquentiae orbata vix nomen ipsum oratoris retineat; neque enim ita appellamus nisi antiquos, horum autem temporum diserti causidici et advocati et patroni et quidvis potius quam oratores vocantur. Both inquirers—both the Roman and the Greek—agree in the answer they would give to this question: they hold that the literary decline is due to deep-seated moral causes. It is this elevation of view that raises their works so far above the standpoint of the ordinary handbooks of rhetoric.

Among minor and more accidental points of resemblance may be reckoned the fact that both books have been preserved in a more or less fragmentary form, and that both alike lay for centuries in complete obscurity without a hint, from any quarter, of their existence. Possibly both were intended for private (perhaps for secret) circulation rather than for publication in the ordinary way. Around both, again, an extensive controversy with regard to authorship has arisen, but with marked differences in its circum- stances and its results. The manuscript ascription of the Dialogus to Tacitus is definite and unimpeachable. The book was, therefore, naturally included in the editio princeps of Tacitus’ then known works, that issued by Vendelin de Spira at Venice in 1470. The great attack upon its authenticity was made by Justus Lipsius a century later, an attack resting principally (like those which have followed it) upon grounds of style. But although scepticism began much earlier in the case of the Dialogus than in that of the De Sublimitate, the Tacitean ascription has fared better than the Longinian. For while a few critics still suspend their judgment, the majority (and among them its latest editors in America and England, Gudeman and Peterson) hold that the Dialogue is an early work of Tacitus. With the De Sublimitate it is, as we have seen, otherwise. The claims of Longinus are upheld by few. And although the evidence is not absolutely conclusive, we must, I think, perforce admit that the balance inclines strongly in favour of the first century and against the third. The equivocal testimony of the MSS.; the absence of direct references in ancient authors; the names included in the treatise or absent from it; its affinities in style, in thought, and in general standpoint ; such considerations, when taken singly, cause hesitation, and when taken together raise the gravest doubts as to the truth of the traditional view.

The alternative—the highly probable alternative—is to regard the first century as the period of composition and an unknown author as the writer. An ‘unknown author,’ because the various attempts at identification have failed to carry conviction; they still remain conjectures, nothing more. As far as I am aware, no other Zonginus than the Longinus of history has been put forward at any time as a possible author of the treatise. But it is different with Dionysius, the optional name given in the manuscript inscrip- tion. This name has led to a plentiful crop of guesses: Dionysius of Hali- -carnassus, Aelius Dionysius of Halicarnassus, Dionysius Atticus of Pergamus, Dionysius of Miletus. But the claims advanced on behalf of these writers

202 THE GREEK TREATISE ON THE SUBLIME.

are advanced either without evidence or in the face of evidence. It is the same if the conjectures take a wider range. W. Christ suggests the name of Theon, who wrote a treatise περὶ συντάξεως λόγου But this is avowedly pure guesswork. Vaucher’s advocacy of Plutarch, on the other hand, is supported by much argument and a considerable array of facts. But the theory is surrounded by so many difficulties of its own that it is now practically abandoned.? On the whole, it seems best frankly to confess our ignorance, and while recognising the probability of a first-century authorship to think of the author himself as AvcTOR IGNoTVs. We had best inscribe the work ANWNYMOY, thus following the reading of the Florence manu- script. This may seem an inconclusive conclusion, but it is the only one at present within our reach.

But while it is good science to refuse to hazard any conjecture which our information does not warrant, it is good science also to decline to follow some critics in abandoning all hope of ever seeing a solution of this knotty problem. Let us rather recognise that we are confronted with one of those stimulating and fruitful uncertainties which classical research so often presents to its votaries,—uncertainties which are stimulating because there is some possibility of removing them, and fruitful because in any case they lead to the more thorough investigation of the obscurer by-ways of history and literature. Two directions from which light might possibly come in the present case may here be mentioned.

(1) Best of all would be the discovery of a fresh MS. of the De Sublimitate, free from the lacunae which at present disfigure the treatise. It is well-known that the gaps amount to something like one-third of the whole work, the approximate extent of the loss being ascertainable from the leaves missing in P. 2036. In these lost parts there may have been references which would help to fix more nearly the date of the book. An ounce of definite fact of this kind inspires more confidence than a ton of loose speculation upon supposed variations of style. It is men like Amati and Roeper that have really advanced matters, and this because they have kept their eyes open to hard facts within and without the treatise, and have recognised that even the most trivial fact may become luminous and instructive when duly correlated with others. Very welcome, in particular, would be the discovery of any such correspondence between the treatise and some other writing as one between the Dialogus de Oratoribus and Pliny’s Epistles which was first noticed by A.G. Lange. In c. ix. of the Dialogus occur the words: adice quod poetis.... in nemora et lucos, id est in solitudinem, secedendum est (cp. ibid. ο. xii. ad init.). Lange pointed out that Pliny (Zp. ix. 10), addressing Tacitus and referring to

1 W. Christ, Gesch. d. gr. Litt. (second edi- sunt, lvi. Dionys. Att. of Perg. Vaucher 46, tion, 1890), p. 630. 90; Canna 12-14; Pess. 292; Blass, Griech.

2 For Plutarch reference may be made to Bereds., 158. Dionys. of Miletus, Vaucher 91 ; Vaucher 93-119 ; Canna 15,16; Winkler19; Pess. 292. [Full titles of the books here in- Brigh. 37. ‘For Dionys. of Halic., see Vaucher dicated by the authors’ names will, where not 44, 45, 50, 54, 90; Cannall. Ael. Dionys. of already given, be found in the bibliographical Halic., Vaucher 91; Egger, Longini quae super- note at the end of this article.)

ITS AUTHORSHIP. 203

the pursuit of poetry, says poemata .... tu inter nemora et lucos commodissime perfict putas. This, though it may not be proof positive, is at the least a remarkable resemblance, and one cannot wonder that much is made of it by the supporters of the view that Tacitus wrote the Dialogus. Our own problem furnishes, as we have seen, some similar correspondences, but we could wish for something more precise and definite than we at present have. The missing portions of the treatise, should they be discovered, might possibly supply our want. And in view of some pleasant recent surprises, who shall venture to say that such a discovery is an impossibility ?

(2) The second possible side-light is the identification of the Terentianus to whom the treatise is addressed.! This question deserves, perhaps, a fuller consideration than it has hitherto received.

Let us first collect the particulars as to Terentianus which are provided, directly or indirectly, by the treatise itself. At its commencement he is addressed as Ποστούμιε Τερεντιανὲ φίλτατε, though there is here in the best MS. a different reading, to which we must return presently. The other forms of address have been classified as follows in the interesting Swedish edition by Elias Janzon (Upsaliae, 1894), where the references are to the pages of Tahn-Vahlen’s text: Τερεντιανὲ φίλτατε (44,7; 66,7), φίλτατε Τερεντιανέ (24, 20), Τερεντιανὲ ἥδιστε (3,1; 7,17), ὦ νεανία (27, 20: altered by the editors to ὦ Τερεντιανέ, against the best manuscript authority, and against the usage of the author, who elsewhere couples some endearing epithet with the name Tepevtiavé), ὦ φίλος (9,16), ὦ ἑταῖρε (41, 2), ἑταῖρε (2,11; 14,12; 16, 4), κρώτιστε (59, 12), φίλτατε (2, 14; 9, 22; 25, 25; 38, 22). It is clear from these expressions that a close friendship existed between the two men. By the form of allocution ὦ νεανία, and by such expressions as ἕνεκα τῆς σῆς χρηστομαθείας in xliv. 1 (cp. the didactic tone of rod μαθεῖν χάριν and ὅπως ἢ σοι γνώριμον in ix. 10 and 15, as well as the words ἀνεγνωκὼς τὰ ἐν τῇ Πολιτείᾳ τὸν τύπον οὐκ ἀγνοεῖς in xili.), it may or may not be implied that the two friends stood, or had stood, to one another in the relation of master to pupil; probably it is. Certainly they had examined the work of Caecilius together (i. 1), while if we follow the reading of one MS. (ὡρισάμεθον viii. 1) they had been even more closely associated in the study of Xenophon. It is, moreover, implied in the treatise that Terentianus was a cultured Roman with some experience of public life (xii. 4; i,2,3,4). The author seems to wish it to be understood that his book consists of jottings only (ὑπομνηματίσασθαι 1. 2, ὑπομνήματος xxxvi. 4),and that it is designed specially, if not exclusively, for the delectation of Terentianus (i. 2).

The particulars thus collected are interesting, but they cannot be said to

1 In continuation of a parallelism already mentioned, it may be noted that the Fabius Iustus to whom the Dialogus is addressed was probably Pliny the Younger’s friend, Consul Suffectus in 102 a.p. The person addressed is, therefore, in the one case as well as in the other, a factor in the determination of the date.

—Again, a question arises in both cases as to the precise signification of iwvenis or νεανίας. Tacitus (or whoever the author was) speaks of himself as ‘iuvenis admodum’ at the time of the Dialogue. In the De Sublimitate, on the other hand, it is Terentianus that is addressed in the words ὦ νεανία,

204 THE GREEK TREATISE ON THE SUBLIME.

be precise. If we chose to designate the author as the AVCTOR AD TERENTIANVM, that designation would not at present mean anything more than AVCTOR INCERTVS or AVCTOR IGNOTVS. Probably we need fresh material from within or from without the treatise before we can hope for an actual identification. But meanwhile we must make the most of every fragment of evidence we possess. And from this point of view it cannot be considered satisfactory that so little attention should have been paid to the reading of P. 2036 at the beginning of the treatise. P. gives Prwpevtiavé, for which the editors, following Manutius, have with one accord substituted Tepevtvavé, in order to bring the address into line with those found elsewhere in the treatise. Probably this change is right as far as it goes, but it does not go far enough to account for what, if unexplained, must seem a strange aberration in so excellent a manuscript as Ρ.1 I should like to suggest, though tentatively and with all reserve, a possible explanation. It is that, in its original form, the address ran thus: Ποστούμιε Mad pe Tepevtiavé φίλτατε. At a comparatively early stage of the manuscript transmission doubt may have arisen as to Madpe, it may have been changed into Φλῶρε, and finally a ‘conflation’ of Φλῶρε and Τερεντιανέ may have yielded DdAwpevtiavé? Madpe might well be doubted on grounds of: (1) rarity, (2) order, (3) superfluity. I will take the points one by one. (1) Rarity. ‘Maurus, as a personal name or affix, is not common in Latin, and still less common in Greek, where its transliterated form may have been none the more pleasing because of its close resemblance to μῶρος. But the form itself is, of course, well attested both in manuscripts and in inscriptions such as this :—

AM MIO49V TWZH +69

Μαῦρος Μηζώτρου. (Kaibel, Inserr. Gr. Sic. οὐ It., 2412, 31.)

(2) Order. The inversion in the order of Tepevtiavé and Matpe may have caused difficulty to a copyist. But this inversion is not uncommon, in writers of the imperial period at any rate. Incidentally an instance (‘Iuste Fabi’) has already been quoted from the Dialogus, and ‘Afro Domitio’ may be added from ¢. xiii. of the same book. In Greek we find instances as early as Dionys. Halic. (e.g. Βάρρων Tepévtios = Terentius Varro, Antigg. Rom., i. 14). The usage is rarer when the praenomen, as well as the nomen and cognomen, is used (the full array of the ‘tria nomina’ is itself rare); but I do not think

1 The exact reading of P. is φλωρεντιανὲ, served in the Cambridge University Library, There is, as I can testify from personal inspec- gives gAwpévriavé, with τερεντιανγὲ in the tion of the MS., no doubt about the presence of margin. the dot. ‘A puncto notatum ut suspectum,’ 3. Φλορεντιανέ is found in other MSS. as the editors say.—The Codex Eliensis, pre-

ITS AUTHORSHIP. 205

it is unexampled in the Latin of this period. Here, I take it, considerations of rhythm or euphony (to which our author pays great attention) would suggest the order Postumius Maurus Terentianus, the same explanation probably holding good in the case of the Tacitean Afro Domitio already quoted. (3) Superflwity, A long-suffering scribe would be prone to think that one of these names might easily be spared, and he may therefore have dropped the Madpe altogether as some of the MSS. have done, or preserved only a scanty vestige of it in PAwpevtiavée. But I would ‘suggest,’ as the lawyers say, that our author of set purpose gave the name in full at the commencement of his treatise, and there only; he wished to be specially | formal and honorific at the beginning. His first sentence, even as it stands, is of an astonishing amplitude, and he would probably have regarded an additional word as an advantage rather than the contrary. Whatever the name may be which has disappeared,—whether it be Μαῦρε, or Φλῶρε, or Φλώρηνς, or Prwpevtive, or Pr. = (Φλάβιε or PrAdove),—I feel confident that some name has been lost, and that this is the key to the reading of the best MSS. For it must be remembered that they show no variation when Tepevtiavé occurs, as it does occur five several times, in other passages of the treatise.

I will now go a step further, though still with the same diffidence, and suggest that the person actually addressed was Terentianus Maurus, the writer on prosody. I must begin by admitting frankly that we have, as far as I know, no evidence to show that this writer’s full designation was Postumius Terentianus Maurus. Consequently we can do no more than point out (a) that his name may, in the scanty notices we have of him, have come down to us in the abbreviated form in which authors are constantly mentioned, ‘ Terentianus’ simply being the usual designation in his own case ; and (6) that the combination does not seem an impossible one in itself. Wilmanns gives an inscription of late date (Inscriptiones Africae Latinae, 9016) which not only unites the names Postumius Maurus, but is followed by a blank space, one large enough (we may add, on our own account) to accommodate such a word as Terentianus, if we might for a moment assume that it had originally stood there.

If we were in a position (as we certainly at present are not) to establish this identification, the result would perhaps, after all, be interesting rather than important. We should hardly be able to fix the uncertain date of the De Sublimitate by means of the uncertain date of Terentianus Maurus. But though the date of Terentianus Maurus remains uncertain, yet the tendency of recent critical opinion has been to assign a much earlier floruit to him (as to Petronius Arbiter, whom he quotes) than the third century of our era. Teuffel (Gesch. ἃ, Rom. Litt.’ ii. 945) thinks that he lived about the close of the second century; A. Werth (‘De Terentiani Sermone et Aetate,’ in Fleckheisen’s Jahrbiicher fiir Classische Philologie, 1896) suggests that he was born in the reign of Hadrian (117—138 a.D.). It is not, however, impossible that he was writing as an old man in the reign of Hadrian or shortly after it, and that his youth fell well within the first century. I desire definitely to

H.S.—VOL. XVII. Q

206

THE GREEK TREATISE ON THE SUBLIME.

revive this view, which (we shall see presently) is not altogether a

new one.!

But if we waive the question of the date, Terentianus’ genial elderly disquisitions (couched in various verse) De Litteris Syllabis Metris might well have come from a pupil of the author of the De Sublimitate, who, it will be remembered, is not averse from discussing the minutiae of style, and who inculcates the use of homely phrase upon occasion. Terentianus seems, too, to hint that in his earlier days he also had essayed the grand, or elevated, style:—

Sic nostrum senium quoque, quia iam dicere GRANDIA maturum ingenium negat, nec spirant animas fibrae, angustam studii viam

et callem tenuem terit, tautum ne male desidi suescant ora silentio.

quid sit littera, quid duae, iunctae quid sibi syllabae, dumos inter et aspera scruposis sequimur vadis. fronte exile negotium

et dignum pueris putes ; adgressis labor arduus

nec tractabile pondus est. at mens tenditur acrius, ne contenta sit obviis, rimantemve recondita subtiles fugiant notae, neu discretio falsa sit rerum tam gracili modo. instat callida cautio,

ne sermo ambiguum sonet, ne priscum nimis aut leve, vocum ne series hiet,

neu compago fragosa sit, vel sit quod male luceat ; dum certo gradimur pede, ipsi ne trepident pedes. par examinis aestus est, ceu SUBLIMIA disseras:

1 It is doubtful whether sufficient attention has been paid to the line, ‘dulcia Septimius qui scripsit opuscula nuper’ (v. 1891). The points to be remembered are (1) that Terentianus was,

as appears in his Preface, an old man when he wrote the line, and (2) that Septimius Serenus is reckoned, by so high an authority as W. 5, Teuffel, among the poets of Hadrian’s time.

ITS AUTHORSHIP. 207

par est iudicii mora:

pompae gloria vilis est. Praefatio, 51-84."

The man who wrote these lines had surely a fine ideal of the dignity of the grammarian’s life; and in a different way the epilogue (1282— 1299) to his second section, that De Syllabis, is even more moving and even more instinct with the spirit of the ‘Grammarian’s Funeral’ :—

Forsitan hune aliquis verbosum dicere librum

, non dubitet; forsan multo praestantior alter pauca reperta putet, cum plura invenerit ipse ; deses et impatiens nimis haec obscura putabit : pro captu lectoris habent sua fata libelli. sed me iudicii non paenitet: haee bene vobis commisi, quibus est amor et prudentia iuxta, et labor in studiis semper celebratus inhaeret : vos sequar, in vestro satis est examine cautum. haec ego cum scripsi, bis quinis mensibus aeger pendebam ambiguum trutina sub iudice corpus, alternum nutans et neutro pondere sidens: nam neque mors avide nigros pandebat hiatus, nec vitam forti retinebant stamine Parcae. sic varios tam longa dies renovando dolores duxit ad hoc tempus semper sine fine minando. cum potui tamen, obrepens incepta peregi, quo vitae dubius vel sic vixisse viderer.

De Syill., 1282-1299.

One question may be asked and answered before we leave Terentianus’ grammatical discussions in verse. Did he know Greek, as the Terentianus of the De Sublimitate must have done? The answer is in the affirmative. He quotes the Greek technical terms proper to his subject, and he adduces Greek examples; and though he does not claim an extensive knowledge of Greek literature, he nevertheless feels the debt due to Greece (‘artium parens et altrix Graeca diligentia est,’ 342), especially the debt which the schoolboy owes (‘Graecus pueros ut docet insonans magister,’ 253).—Just one more question, the answer to which we shall hardly hope to find in these gram- matical treatises. Had Terentianus Maurus any part in public life, such as the Terentianus Maurus of the De Sublimitate seems to be taking at the time when he is addressed, a time which may of course be considerably later than the period of his instruction (if instruction he had) at the hands of the author ? It has sometimes been suggested, by those who advocate an early date for

1 It need hardly be pointed out that sublimia τὰ ὑψηλότερα, τὰ ὑπερφυᾶ, τὰ μεγέθη, τὰ μεγάλα, and grandia are the obvious Latin equivalents .7.A. of such expressions in the Treatise as τὰ ὕψη,

Q 2

208 THE GREEK TREATISE ON THE SUBLIME.

Terentianus Maurus, that he is to be identified with Terentianus, the governor of Syene in Egypt, mentioned by Martial in one of his Epigrams (I. lxxxvi. 6, 7: tam longe est mihi quam Terentianus, | Qui nunc Niliacam regit Syenen. Date, 85-86 a.D.: L. Friedlaender). With the view that Terentianus wrote his De Litteris as an old man in the reign of Hadrian, and with the view that he is no other than the Terentianus of the De Sublimitate such an identifica- tion would agree admirably, but I am bound to confess that there seems to be no positive evidence in support of it and that it postulates an earlier date for Terentianus Maurus than would readily be conceded by Teuffel-Schwabe. But I feel free to point out that the suggestion has been made, and made without any thought of the Terentianus of the De Sublimitate. Perhaps it is just worth mention that Wilmanns, in the collection already named, has the following mutilated inscription (8402) :—

MARTIALIS E MAVRO RO E SVO FECIT

No argument can be based on an unintelligible fragment of unknown date, one also which, it should in candour be added, was differently read by an earlier authority (Vincent in Revue Africaine, xxi. 315). But if the reading is correct (and the authority of the Corpus Inscriptionum Latinarum is not lightly to be set aside), we do at least find the names of a Martialis and a Maurus brought into some sort of connexion. For the sake of completeness, I will just add, as the names Terentianus and Maurus are both somewhat uncommon, that the latter is found in a two-line epigram (Anth. Pal. xi. 204) attributed to Palladas, which begins with the words 'Ῥήτορα Μαῦρον ἰδών. Flavius Terentianus occurs in Inserr, Afr. Lat. 8412 and 8932, in both of which inscriptions the man in question is described as ‘ praeses Mauretaniae Sitifensis’; the date may be given as 318-319 A.D. The inscriptions, like the epigram, are late; but as already said, they are added here simply for the sake of completeness.

Whether or no the Terentianus of the De Sublimitate has any direct connexion: with any Terentianus Maurus and with Africa, we shall do well, I think, to recognise that the writer of the Treatise has many points of contact with Alexandria. In some respects the Nile (to which he refers with admira- tion) seems to be nearer to him than Rome itself. He sometimes writes as if, when writing, he knew of things in the capital by hearsay rather than by actual experience. He can speak in general terms of Roman vices, but he does not appear (as has been already seen) to possess the knowledge of a resident with regard to definite, though perhaps trivial circumstances, such as the confinement of the Pygmies. But the very theme of his book, as well as its specific points of contact with Philo, with Josephus, with Caecilius, with the Hebrew scriptures, seems to associate him, in spirit if not in residence, with Alexandria, the great meeting-place of Jew and Greek.

The hypothesis that the book was produced at a distance from Rome, or sent to a friend at a distance from Rome, might help to account for the fact

ITS AUTHORSAITP. 209

that it seems to have been so little known in antiquity. If that friend was also in an official position, there might seem double reason for secrecy with regard to a work which might be held to embody seditious sentiments. A book designed for private circulation would naturally not be multiplied to any extent, and this would explain the paucity of independent copies of the treatise.

However, I have, I need not say, no intention of pressing any of these speculations, nor even that from which they started—the identification of Terentianus with Terentianus Maurus. If there were any truth in this suggestion, it would no doubt have been made before. Some, indeed, might go so far as to regard the Terentianus of the treatise as an entirely fictitious person, the offspring of the literary convention which conducted such discus- sions in the form of dialogue or epistle. But so extreme a view, though it might be put forward, could hardly be successfully defended. For apart from the fact that the general practice was to introduce real personages into such letters and dialogues, there is a special reality and intimacy about the references to Terentianus in the De Sublimitate. One of the chief impressions, in fact, which we form upon internal evidence with regard to our anonymous author is that, whatever else he may have been, he was at least a warm-hearted friend and an enthusiastic teacher. Internal evidence also assures us that he was a Greek, who had some acquaintance with Latin and even with Hebrew literature ; that he was conversant, to some extent, with art as well as with literature ; that in his general view of things, as well as in his diction, he had been influenced greatly by Plato; and that he had written on other subjects than his present one.!

This is all we can state about the author with any approach to certainty, and no doubt it is a meagre record when we compare it with our recollections of the historical Longinus of the third century, whose learning won him the curious designation of ‘a living library and a walking museum,’ and made him famous as the prince of critics; who at Alexandria had been the brilliant pupil of the Neoplatonists ; who at Athens gained celebrity as the teacher of young men ambitious of philosophical and literary culture ; who at Palmyra, as the minister of Zenobia, inspired the defiant reply sent by the queen to the letter of the emperor Aurelian which demanded her submission ; who met his death in the spirit of a hero. We lose much in losing the halo of romance which such a name throws about a book, and it is with a certain sadness that we see Longinus giving place to Pseudolonginus (as if the writer were an impostor) in the hands of every German graduand. The work has come to be regarded as a foundling, and to suffer the foundling’s fate. Its present

1 The following passages seem to contain references to other writings of his: viii. 1, ὡς Kay τοῖς περὶ Ἐενοφῶντος (if this is a refer- ence to a separate work) ὡρισάμεθα, or ὡρισά- μεθον. ix. 2, yéypapd που καὶ ἑτέρωθι τὸ τοιοῦ- τον" ὕψος μεγαλοφροσύνης ἀπήχημα. xxiii. 8, καὶ τὸ Πλατωνικόν, ὃ καὶ ἑτέρωθι παρατεθείμεθα,

ἐπὶ τῶν ᾿Αθηναίων: “οὐ γὰρ Πέλοπες,᾽ κ.τ.λ, xxxix. 1... ἡ διὰ τῶν λόγων αὐτὴ ποιὰ σύνθεσις. ὑπὲρ ἧς ἐν δυσὶν ἀποχρώντως ἀποδεδωκότες cur τάγμασιν... Xliv. 12... τὰ πάθη, περὶ ὧν [ἐν ἰδίῳ προηγουμένως ὑπεσχόμεθα γράψειν ὑπομνή- MATL...

210 THE GREEK TREATISE ON THE SUBLIME.

neglect in England may be due to some such hazy prejudice, as well as to a not unnatural reaction against the excessive claims at one time made on its behalf, in England and in France, as an infallible court of appeal.

In future the treatise must stand upon its own merits; and it can well afford to do so, for those merits are of no ordinary kind. After all, it is the most striking single piece of literary criticism produced by any Greek writer subsequent to Aristotle. It claims our respect and admiration by its noble tone; by its apt precepts; by its judicious attitude towards fundamental questions such as those of the errors of genius, the standard of taste, the relation of art to nature and of literature to life ; by its value as a treasury of extracts, and of happy appreciations destined to be confirmed by every fresh discovery of Hypereides or Bacchylides ; and lastly, by its historical interest as one of the earliest essays in comparative criticism, and as an aesthetic treatise which has had some degree of influence upon almost every European literature. Surely such a book deserves an English edition.

W. Ruys RosBERTs.

Bibliographical Note on Recent Publications.

I append a list of books and articles published in or after the year 1870, all of which deal with the De Sublimitate and most of which have some bearing upon the question of the authorship. These productions may be conveniently grouped, year by year, under two headings: A. Separate publications, B. Articles in Periodicals.

A.—SEPARATE PUBLICATIONS.

Aem. Winkler, De Longini qui fertur Libello ΠΈΡΙ Y¥OYS, Halis, 1870.—M. Haupt, Ind. lect. habend. in Univ. Frid, Guil., Berolini, 1870 (reprinted in Haupt’s Opuscula, ii. 428-433).—H. A. Giles, Longinus, an Essay on the Sublime : translated into English, London, 1870.—Giovanni Canna, Della Sublimita : libro attribuito a Cassio Longino, Firenze, 1871. —H. von Rohden, Commentat. in honorem F’. Buecheleri, Bonnae, 1873.—L. Martens, De Libello ΠΕΡῚ Y¥YOYS, Bonnae, 1877.—A. Reifferscheid, Indices lect. Vratislav., Vrat., 1879- 1880,—J. Vahlen, Ind. lect. in Univ. Frid. Guil. habend., Berolini, 1880.—A. Jannarakis, Eis τὸ Περὶ ὝΨους λεγόμενον βιβλίον Κριτικαὶ Σημειώσεις, Marburgi, 1880.—M. Hertz, Ind. lect. Vratislav.,Vrat., 1881.—M. J. Moreno, Tratado de la Sublimidad traducido fielmente del Griego de Dionisio Casio Longino: con notas histéricas, criticas y biogrdficas, y con ejemplos sublimes Castellanos comparados con los Grieyos citados por Longino, Sevilla, 1882. —H. Hersel, Qua in citandis scriptorum et poetarum locis auctor libelli περὶ ὝΨους usus sit ratione. Berlin, 1884.—AIONYZIOY H ΛΟΓΓΊΝΟΥ ΠΕΡῚ Y¥OYS: edidit Otto Iahn a. MDCCCLXVII : iterum edidit a. MDccocLXxxviI Ioannes Vahlen. Bonnae.—B. Coblentz De Libelli MEPL ὙΨΟΥΣ Auctore. Argentorati, 1888.—The Poetics of Aristotle, together with the Treatise on the Sublime by Longinus, London, 1889. (Cassell’s National Library, edited by Henry Morley).—H. L. Havell, Longinus on the Sublime translated into English : with an Introduction by Andrew Lang, London, 1890.—Elias Janzon, De Sublimitate Libellus in patrium sermonem conversus adnotationibusque instructus. Upsaliae, 1894.—Rhetores Graeci ex recognitione Leonardi Spengel. Vol. 1. Pars II. Edidit C. Hammer. Lipsiae, 1894, [Originally edited by Spengel in 1853.]—G. Meinel, Dionysius oder Longinus, Ueber das Erhabene: uebersetzt und mit kritischen und exegetischcn Bemerkungen versehen,

ITS AUTHORSHIP. 211

Kempten, 1895 —E. Brighentius, De Libelli Περὶ Ὕψους Auctore Dissertatio, Patavii, 1895.—F. Nicolini, Adnotationes in Longini Περὶ “Yous Libellum. Catinae, 1896.—J. Freytag, De Anonymi Περὶ Ὕψους Sublimi Genere Dicendi, Hildesheim, 1897.

B.—ARTICLES IN PERIODICALS.

M. Schmidt, Rhein. Mus., 1872, xxvii, p. 483.—U. von Wilamowitz-Méllendorff, Hermes, 1876, x, pp. 334-346.—H. Diels Hermes, 1878, xiii, pp. 5, 6.—E. Rohde, Rheinisches Museum, 1880, xxxv, pp. 309-312.C. G. Cobet, Mnemosyne, N. S., 1882, x, pp. 319-323.—R. Pessonneaux, Annales de la Faculté des Lettres de Bordeaur, 1883, v, pp. 291-303.—F. Buecheler, Rhein. Mus., 1884, xxxix, pp. 274, 5.—E. Egger, Journal des Savants, Mai, 1884, pp. 252-254,—M. Rothstein, Hermes, 1887, xxii, pp. 535 -546.—J. B. Bury, Classical Review, 1887, I, pp. 300-302.—M. Rothstein, Hermes, 1888, xxiii, pp. 1-20.—Robinson Ellis, Hermathena, 1896, ix, pp. 385-388.—W. Schmid, Rhein Mus., 1897, lii, p. 446.

212 ARTEMISIUM.,

ARTEMISIUM.

Ir has fallen to my lot in the course of the last few months to examine the text of Herodotus with a view to discovering, if possible, the extent to which the Greeks of the time of the Persian War were acquainted with the principles of strategy.

What may be called the ‘incidental’ nature of the historian’s narrative demands, of course, that the greatest care should be expended by any one who pretends to examine it, and the manifestly unprofessional character of the military portion of it, together with the evident inexperience of the author in matters connected with war, would render the task a hopeless one, were not the nature of the theatre of events so marked in character as to elucidate much that would otherwise be obscure or incomprehensible.

Among the war problems which Herodotus places before us, that of Artemisium is by no means the least interesting, a fact of which the amount of critical literature which has grown up about it, gives eloquent, and, perhaps, embarrassing proof. My only excuse for adding to its volume, if not to its weight, is that this literature has a tendency, which has developed noticeably of late, to rewrite the whole history of events, on the plea that the tale told by Herodotus is past understanding and will not bear examination. As Herodotus is practically our only authority, I cannot, I confess, regard as convincing such reconstructions as ignore the evidence of fact which Herodotus gives, nor yet those which are founded on a manifest omission to take into consideration the whole of the facts as recorded by him. I do not for one moment imply that such omission is in any case intentional; it results, no doubt, from the method, or want of method, employed by the historian.

In studying Artemisium we are faced by the same difficulty which we find in the case of Plataea. Herodotus has evidently been unable to obtain evidence as to the reasons which actuated those who were in responsible command on either side, save only where some large and generally recog- nizable question was involved, as, for instance, whether the great stand against the Persians should be made at the Isthmus or north of it.

But again, if a line may be taken from Plataea, it might, perhaps, predispose us to accept, as being, in the main, true, the plain statements of fact which he makes with respect to Artemisium.

Putting aside, however, any predisposition of the kind, it may be well to take the narrative in detail and to examine the validity of the objections which have been made to various parts of it.

The strategical interdependence between the army at Thermopylae and

ARTEMISIUM. 213

the fleet at Artemisium is, of course, the leading factor which must enter into any criticism of the narrative of either battle.

Thermopylae would have been untenable against the combined land and sea force of the Persians, had not the Greek fleet been at the movth of the Euripus, so as to

(1) Defend the sea flank of the defending army of Leonidas;

(2) Prevent the landing of ἃ Persian force in rear of the pass.

The last fact seems to be recognized by all commentators, but many of those who have written on the subject seem to be quite unaware that the depth of water close in shore at the narrowest part of the land passage was such as to admit of ships being brought sufficiently near in for them to be able to take a prominent part in the attack and defence of the position! This is, however, to be clearly seen from the history of another fierce battle in this pass almost exactly two centuries later.

In 279 B.c. Brennus and his Gauls, after invading Italy, Illyria, Pannonia, and Thrace, marched on South Greece. At Thermopylae they found them- selves faced by 25,000 Greeks. Brennus crossed the Spercheius with a force of 200,000, and attacked the pass with the utmost ferocity, but was unable to force it. This is, it need hardly be said, of the greatest importance as showing the immense strength of the defensive line of Mount Oeta ; but we are further told that in the attack the Athenian galleys, which were supporting the army in the strait, did considerable damage to the Gallic force by coming close in shore and attacking them with iissile weapons.

What followed is not apposite to our immediate purpose, but is peculiarly illustrative of the strategical geography of Greece. Baulked at Thermopylae, the Gauls made a diversion into Aetolia, but, defeated there, returned and attacked Thermopylae again. The very fact of their having to return to the scene of their former failure shows how very restricted is the line of commu- nication from the north at this point. Then happened what was practically a repetition of the events of two centuries before. The Gauls surprised the Pho- kians who were defending the path of Hydarnes, of the existence of which they had been informed by the Herakleots, and the defending force at Thermopylae was obliged to embark on the Athenian galleys and sail away down the Euripus.

The maintenance of the line of Mount Oeta against attack from the north was, then, absolutely dependent on the command of the Euripus, and not merely of the Euripus, but of the north end of it.

Some commentators say that Herodotus had no appreciation of the interdependence of the two positions. That he had no professional apprecia- tion of the fact is, of course, true, and is exactly what might be expected from him; but that he had a general appreciation of it his language on several occasions seems to prove.”

1 The rapidity with which this coast has 2 Cf, vii. 175: ταῦτα yap ἀγχοῦ τε ἀλλήλων advanced renders it probable that what was ἐστὶ ὥστε πυνθάνεσθαι τὰ κατὰ ἑκατέρους“ édvra...: done in 279 8,0. μόγις καὶ οὐκ ἄνευ κινδύνου and especially the passage in viii. 15 beginning would have been an easy exploit in 480, cf. ἣν δε was ὁ ἀγὼν....... to the end of the chapter : Pausanias x, 21, 4, ef, also viii. 21.

214 ARTEMISIUM.

But there is another general objection raised against Herodotus’ narra- tive, which is held to impugn the reliability of his account, and to render it necessary to reconstruct the whole history of events.

It is this :—

If,as must be admitted, there was this absolute interdependence between Artemisium and Thermopylae,

and if, as must also be admitted, this interdependence was plainly recog- nized by the Greek commanders,

how is it that we hear :—

(1) Of an actual retreat to Chalkis, 1,6., of an abandonment of the north Euripus after the capture of the Greek outpost vessels off the Macedonian coast (vii. 182) ?

(2) Of a contemplated retirement ἔσω és τὴν “EdAdéa (viii. 4), due to the panic which the sight of the magnitude of the Persian fleet at Aphetae caused among the Greeks?

(3) Of a contemplated movement south to meet the 200 Persian vessels which had been sent round Euboea (viii. 9.) ?

(4) Of a contemplated retirement ἔσω és τὴν Ἑλλάδα (viii. 18.) after the third day’s battle at Artemisium, in which the Greeks had been roughly handled ?

This objection seems hardly valid, when the general drift of the history of events which Herodotus has given us is taken into consideration.

If any general fact is brought into prominence in his account of the war of 480—479, it is that the Peloponnesian Greeks were ever hankering after the Isthmus as the line of defence against the huge Persian force. Drawing largely from Athenian sources, he also brings the selfishness of this policy into special prominence. Selfish, indeed, it must have seemed to the Greeks north of the Isthmus, who were probably more or less aware of the eminently defensible character of such a line as that formed by Oeta, though their acquaintance with the topography of the region north of the Boeotian plain seems to have been of an imperfect character! But is it so strange that the Peloponnesian Greeks should have preferred a defensive line at the Isthmus which they did know, and in which they believed, to one of which they can have had but imperfect knowledge, especially after they had just been involved in what they must have regarded as a fiasco—that expedition to Thessaly ? Was it, after all, in view of their then knowledge, so selfish a policy to fix the defence in a war in which, opposed by the enormous magni- tude of the Persian power, they must have faced the situation with the courage of despair, at a line in which they had some sort of confidence, rather than at one where the chances of success were impossible of calculation ?

Ὁ Cf. the mistake made as to the defensive Lamia-Delphi road through Oeta via Cytinium, nature of the position at Tempe: also,ignorance in the account of Thermopylae, and this, too, of the existence of the path of Hydarnes at although its use by the Persians immediately Thermopylae: also, the absence of all mention after the battle is almost certainly implied in Herodotus, and, inferentially, in Herodotus’ (viii. 81), . 1 gources of information, of the pass on the

ARTEMISIUM. 215

That selfish consideration of their own special interests contributed to the formation of their views on the strategical question is doubtless the case, but, at the same time, they had what must have seemed to them some very sound arguments in support of those views. With this policy, that of the northern Greeks, championed especially by the Athenians, was, of course, in conflict ; and though this latter policy in the end prevailed, yet Herodotus clearly shows that there were times throughout the whole period of the war when the contest between the two was doubtful, times, too, when it seemed as if the Peloponnesian policy must win the day. That they were all but in equi- librium at the time of Artemisium is evident. The northern policy had so far prevailed as to induce the Lacedaemonians to make a show, at any rate, for the defence of the northern Greeks, who, if they thought themselves aban- doned, were only too likely to refuse to join in the defence of the Isthmus, to remain at home, and be forced to medize ;1 whereas, if they had some practi- cal demonstration of the apparent impracticability of the defence north of the Isthmus, and of the apparent willingness of Sparta to make some sacrifice on their behalf, they might be induced to aid in the defence of the Isthmus. Is it possible to look on the defence of Thermopylae as having been in any way a serious effort on the part of the Peloponnesian Greeks? We shall never know with certainty the inner history of the policy which sent the force under Leonidas to Thermopylae, the instructions given to him and to Eurybiades; but, though the ways of the Spartan Ephorate were dark in more senses than one, and though the possession by them of a conscience in the modern acceptation of the term might be incapable of historical proof, it seems hardly likely that they deliberately and knowingly sent Leonidas and his band to meet their fate as a sacrifice to a policy of compromise, though this extreme view has been held by some who have sought to explain the half-heartedness or heartlessness of the policy which led to Thermopylae. Ignorant of the existence of the path of Hydarnes,’ it may well have seemed to them that in any case Leonidas and his force in combination with the fleet might, at worst, execute a safe withdrawal. But, be this as it may, it is im- possible to look upon Thermopylae as having been, or having ever been intended to be, a serious effort on the part of the Peloponnesian Greeks. It was necessary to propitiate the northern Greeks. Athens, without the co- operation of whose fleet even the Isthmus would be indefensible, had to be humoured ; and, further, a few days’ stand made at Thermopylae might, we can easily suppose, be of great value towards the completion of arrangements at the Isthmus.

However secret the policy of the Ephors may have been, the insincerity of the Peloponnesian effort at Thermopylae and Artemisium must have very soon become apparent to those who took part in it, and it would not be

1 Cf. vii. 206. when the decision to make a stand at Thermopy- 2 That the existence of the path was a inatter ae was arrived at, cf. viii, 208 especially Adyor- of purely local knowledge cf. vii. 214, ef τῇ res δι' ἀγγέλων. χώρῃ πολλὰ ὡμιληκὼς εἴη: and that those 3 Cf. vii. 206, possessing that knowledge were not present

216 ARTEMISIUM.

unnatural that those engaged in such an effort should snatch at every excuse for withdrawal from it. We even get a hint in the Herodotean narrative which is not without its instructive side when the state of the case is con- sidered. It is with the Korinthian admiral that Themistocles has apparently most difficulty when the question of retreat from Artemisium is discussed ; and, at the end, when the retreat actually takes place, it is the Korinthian who leads the way. Eurybiades had to play a part. He could countenance a proposition to retreat, but, as representing the sham policy of the Ephorate, he could hardly propose it.

Herodotus’ narrative of this part of the war is noticeable from its being in the form of a diary of events, or, rather, of a two-fold diary of events at Thermopylae and Artemisium respectively. Moreover we find two points of contact between the two diaries, viz. at their beginning (the departure from Therma) and at their end (the disaster at Thermopylae) with the singular result that there is a discrepancy of two days between the two, 1.6. the Artemisium diary covers a period two days less than that of Thermopylae. This inconsistency is held to discredit the narrative.’ But the inconsistency cuts both ways. A man who consciously invents a tale, and can do so at leisure, is not likely to let a manifest inconsistency be found therein.

The explanation lies possibly in the fact that we have in the parallel narrative of Artemisium and Thermopylae two stories of different origin, the former in the main Athenian, the latter in the main Laconian.

Busolt in his Greek History (vol. ii., p. 681, note 3), has drawn up from Herodotus a parallel journal of events, which reads as follows :—

Day. 1. Persian army leaves Therma. Le oa zi ae ἐῶν 5 Persian fleet leaves Therma and reaches Magnesian coast. LC ae Be "Σὰ si hrs -Storm begins in morning.

14. Army reaches Malis Storm continues.

15. ay bagi ΩΣ al. "ἢ. Storm continues.

eee “5: ae τς uh Storm ceases. Fleet moved to Aphetae. Dispatch of 200 vessels round Euboea. First sea-fight.

by One aia = ἦν ἢ Second sea-fight after the arrival of fifty-three Athenian ships.

18. First attack on Thermopylae. Third sea-fight,

the evening.

News of disaster at Thermopylae in

19. Second attack on Thermopylae. 20. Disaster at Thermopylae,

1 An ingenious explanation has been put for- ward for what is thought by some to be the inexplicable withdrawal to Chalkis (vii. 183). It is suggested that this refers really to the sending of fifty-three Athenian vessels to guard the south part of the strait, and that these vessels returned to Artemisium after the wreck of the 200 in the Hollows of Euboea.,

About this theory more need not nowybe said than that it demands a dislocation of the whole story. The result is a narrative consistent, indeed, with itself, but wholly at variance with Herodotus. It is only by a close examination of the Herodotean version of the story that we can judge whether so complete a reconstruction is demanded. ͵

ARTEMISIUM. 217

The discrepancy is evident.

Since, however, Herodotus is practically our only authority for the history, he alone can furnish us with the means of correcting himself, and it is only by a close examination of his account that we can hope to arrive at some sort of conclusion as to where the error lies.

On the twelfth day after Xerxes’ departure from Therma the fleet started from that place.

Ten fast sailers preceding the Persian fleet fell in with three Greek scouting vessels and took two of them (vii. 179—180). Now these three vessels must have been far north, north even of the Thessalian coast, for we are told that the one whose crew escaped was run ashore finally at the mouth of the Peneius river (vii., 182).

The Greek fleet was apparently at Artemisium on this day (vii. 182). We are then told (vii. 182) that the Greek fleet got news of the disaster by fire signals from Skiathos,

The mouth of the Peneius river is just 70 miles north of Skiathos, therefore the disaster could not possibly have been visible from that island, and, as far as we can judge, the first news of what had happened must have been conveyed to the watchers on Skiathos by the appearance of the ten Persian vessels with the Greek ships in their company. It is almost certain then that the news cannot have reached the Greek fleet until the evening of the day on which the disaster occurred, and the use of fire signals may, perhaps, be taken to confirm this.

On receipt of this news the Greek fleet retreated to Chalkis (vii. 182).

There is nothing in Herodotus which indicates the time at which the retreat to Chalkis was made, but Herodotus evidently understood that it began after the news of the capture of the three vessels reached the Greeks. As this can hardly have reached them until the evening, and as, if Herodotus’ statement of the cause for this retreat be taken as true, the decision to move cannot have been come to without discussion, and, probably, considerable opposition on the part of Themistocles and the Athenian contingent, it is improbable that the retreat was made immediately 1.6. during the night. It is more probable that it began next morning.

I would suggest that the retreat did take place next morning, apart from any consideration of the cause alleged by Herodotus.

I would further suggest that it is very possible that the storm was the real cause. It broke on the next morning. It was a gale from the E.N.E. blowing straight into the northern bend of the Euripus, and, if it caught the fleet moored off the coast, (as it very likely would be, since the signals from Skiathos made it clear that the arrival of the Persian fleet might be momen- tarily expected) the only course for the Greek fleet would be to run before the wind into the inner part of the strait. Once round the bend at the N.W. of Euboea the fleet would be in calm water as far as Chalkis, for the mountains of the island edge that coast of the Euripus very closely and fall into the sea in a long line of lofty and precipitous cliffs: once in the narrows at Chalkis it would be as good as in harbour.

218 ARTEMISTUM.

It may seem, at first sight, strange that Herodotus does not mention this cause, if it existed. On the general question it is only stating a commonplace to say that statements of fact and statements of cause in Herodotus’ history cannot be placed on the same level of credibility, and that a distortion of the latter often becomes apparent when the moral of the tale is in question. The suspicion is not perhaps unfair that in this case Herodotus has shirked the statement of the true cause, since it would have sadly detracted from the moral of this part of his tale, the divine intervention of heaven (in the form of the winds) in favour of Greece in her great struggle, had he related how the Greek fleet was driven from its chosen position by a storm. Moreover the reason he does give would serve admirably to heighten the effect of the picture he draws of the magnitude of the dread which the expedition of Xerxes and the Persian power inspired in Greece.!

It is noticeable that not only does this retreat never go beyond Chalkis, but there is nothing suggested of any intention to retreat beyond that point. We have not, as we have elsewhere, a contemplated withdrawal ἔσω és τὴν Ἑλλάδα. And yet withdrawal to Chalkis meant the sacrifice of the position at Thermopylae. The fleet at Chalkis might be almost as well at the Isthmus ' for all it was worth to the land army. And yet we hear of no withdrawal or contemplated withdrawal from Thermopylae, though the fleet must have been weather bound at Chalkis for two days at least after its arrival there. If there was any intention to remain at Chalkis, surely it must have been plain to the most limited intelligence in the army at Thermopylae that the Persian fleet would be able to land men in rear of their pass, and the army be caught in the veriest trap that ever an army ran into, And yet Leonidas did not move. He must have known that the Greek fleet was prepared to sail back to its position at the north end of the strait, the moment the weather permitted of its so doing.

And now for the movements of the main Persian fleet on this day. The details are interesting, because they afford us a very fair estimate of the sailing capacity of what were probably the best ships of those times.

The fleet performed the distance between Therma and the Sepiad strand, which is about 120 miles,in one day. At the time of year, the middle of summer (viii. 12) it would be daylight shortly after 4 a.m. and dark about 7.15 p.m. Of dawn and twilight there is but little in this latitude? That

1 I make this suggestion of cause with the greatest diffidence, but with the conviction that Herodotus’ own evidence suggests it. It seems to me that calculations from the statements of Herodotus render it improbable that the Greek fleet could have started from Artemisium before the storm broke. At the same time I do not wish to give the impression, which would be a wrong one, that I look upon the cause stated by Herodotus as being impossible. Considering the state of feeling then prevalent in a section of the fleet, it would be dangerous to assert that

any excuse, however specious, might not have been seized upon as an argument for a retreat. But here, as in the long account of Plataea, Herodotus’ statements of causes are not of the same credibility as his statements of facts, especi- ally when there is present anything that might enhance or detract from the moral which he in- tends to convey, f

2 At 4 a.m. in the first week in August it used to be pitch dark out on the Bay of Nava- rino. In the evening it would be bright day- light at 6.45 p.m., and quite dark at 7 p.m.

ARTEMISIUM. 219

would give fifteen hours of daylight. We can hardly suppose that an expedition of this magnitude could possibly start in the dark, Again it must have put in at the Sepiad strand at an hour which would allow of sufficient daylight for what must have been the long operation of mooring and drawing up on shore. Ata liberal computation the amount of available sailing time cannot have been more than twelve hours. The pace was therefore somewhere about ten miles an hour, and though this seems great, yet we know that in later times a pace some 60 per cent. greater could be maintained by a quin- quereme in a voyage from Carthage to Ostia.

If this was the pace of the fleet, it is exceedingly probable that there were vessels in it which could sail twelve miles an hour.

The ten scouting vessels accomplished even a longer voyage, apparently on this day, but, being a small number, they would not be hampered by the circumstances affecting the departure or arrival of a huge armada. It is, however, recorded that several of them came to grief on the rock Μύρμηξ in the channel between Skiathos and the Magnesian mainland.

On the thirteenth day the storm overtook the Persian fleet on the Sepiad strand,:and on the morning of this day probably the Greek fleet retired to Chalkis—by reason of the storm in all likelihood, if any calculation from the narrative may be made.

On the fourteenth and fifteenth days the storm continued, and the Greek fleet remained at Chalkis.

On the former of these two days the watchers on the Euboean heights reported to the Greek fleet at Chalkis the immense losses which the Persians had experienced in the storm (vii. 192).

It is when we come to the journal of the sixteenth day that we arrive at what is the crucial part of the story.

The storm had ceased.

On the morning of this day the Greek fleet must have moved back to the station at Artemisium.

The Persian fleet also moved from the Sepiad strand to Aphetae.

We do not know with exactness the position of Aphetae. The movement to it is described by Herodotus as having been:—és τὸν κόλπον τὸν ἐπὶ Παγασέων φέροντα: and again, ἐστὶ δὲ χῶρος ἐν τῷ κόλπῳ τούτῳ τῆς Μαγνησίης, ἐπὶ τούτῳ δὲ τῷ χώρῳ οὔνομα γέγονε ᾿Αφέται (vii. 193). It was therefore within the gulf, onthe Magnesian side, probably at the extreme end of that long narrow peninsula which shuts in the Pagasaean Gulf on the south.

Artemisium is placed in our classical atlases at the N.E. point of Euboea. But the position of the Greek fleet cannot have been less than ten miles from this point, for it must certainly have been west of the entrance to the gulf. One of two things must be the case,—either Herodotus describes the position

1 This fact, the fire signals on Skiathos inan Euboeareached the Greek fleet point, as it would earlier chapter, and the rapidity with which the 5661), to a regularly organised signalling system. news of the loss of the 200 ships in the Hollows of

220 ARTEMISIUM.

of the Greek fleet very loosely, as seems to be the case, or the promontory near Oreos, in the middle of the north coast of Euboea, is Artemisium. The Persian fleet is recorded to have arrived at Aphetae early in the afternoon (vil. 6).

The number and nature of the events which are crowded into the remainder of the day render it extremely probable that we must look here for part, at least, of the chronological error which Herodotus has made. Τὶ would seem as if he had attributed to one day the events of at least two. A very cursory consideration of the narrative might convince the reader that something of the kind had happened.

The Persian fleet, then, arrived at Aphetae early in the afternoon. Fifteen of their vessels, however, ὕσταται πολλὸν ἐξαναχθεῖσαι, fell into the hands of the Greek fleet under the impression that it was theirown. Unfor- tunately this does not give us any clue as to the time of the arrival of the Greek fleet at Artemisium. One thing, however, is plain. The main part of the Persian fleet must have already doubled the cape into the gulf before these fifteen got round the Sepiad promontory. It must have been well on in the afternoon before the capture took place.

That these events took place on the sixteenth day, there is plain evidence in the text, but it seems impossible to believe that what followed must all be attributed to the same day. On the mere question of lapse of time it is impossible.

In viui. 4, 5, Herodotus tells us that the Greeks on seeing the magnitude of the Persian force at Aphetae δρησμὸν ἐβουλεύοντο ἔσω ἐς τὴν ᾿Εἰλλάδα. We then have the tale of the bribery of Themistocles by the Euboeans, a tale which may be true, or may be merely an invention of his enemies at Athens which gained currency in later times. As to the contemplated retreat and the probability of the truth of the assertion that it was discussed, we have already spoken. But if we come to consider the time which must have been spent on the discussion, and the difficulty which, from Herodotus’ own statement, Themistocles must have had in reversing the decision to retire, we are practically bound to conclude that the matter could not all have been settled on what remained of that sixteenth day. There is, in fact, enough- material to account for the expenditure of a whole day, and, in any case, whenever the debate began, the determination to remain at Artemisium cannot have been arrived at until the seventeenth day was well advanced.

The next incident which Herodotus attributes to this day is even less likely to have occurred upon it. The Persians, he says (viii. 7), in order to let not one of the Greek fleet escape, dispatched a squadron (the number of which he states as 200) to circumnavigate Euboea. In the first place, before we enter pon the details given of the movement, it seems highly improbable that this squadron could have been dispatched from Aphetae on that day. The fleet must have been terribly disorganised by the storm and the disaster

1 This promontory is, if 1 remember, a low cation of it with Artemisium. one, and this perhaps tells against the identifi-

ARTEMISIUM. 221

which resulted from it, and, as it left the Sepiad strand on the very morning after the storm had ceased, there was no time for reorganisation before the arrival at Aphetae. The time of the departure of the 200 is fixed for us within certain limits. It evidently took place in daylight, otherwise there would have been no reason for the deception practised with regard to the course taken round Skiathos. It is also quite plain from what Herodotus tells us that it took place before the first engagement. But this does not, unfortunately, give us any means of judging the day on which the squadron started. Still consideration of the circumstances renders it almost certain that the start could not have been made before the morning of the seven- teenth day.

But there is another consideration in reference to this important point in the narrative.

The division of the Persian fleet was a step of such magnitude that it cannot have been taken without Xerxes’ direct orders. A reference to the journal of Thermopylae shows that on the sixteenth and seventeenth days, Xerxes wus before the pass, but had not as yet attacked. Communication with the fleet before it arrived at Aphetae would have been difficult. It has been suggested that the squadron of 200 was despatched from the Sepiad - strand, not from Aphetae, as Herodotus states: that they were despatched on the very morning the fleet arrived there, and were wrecked in the storm which arose next morning.

There is one objection which seems fatal to this hypothesis, and several others which militate strongly against it. This squadron was wrecked in the ‘Hollows of Euboea.’ The ancient authorities! tell us that these hollows were the chief bays on the S.W. coast of the island. But the wind which caused the havoc on the’ Sepiad strand is described by Herodotus as having been ἀπηλιώτης, called by the dwellers on the Magnesian coast the ‘ Hellespontian ’ wind. As nearly as we can calculate, it must have been a gale from the E.N.E., a wind which would blow off shore, not on shore, on the S.W. coast of Euboea. But when we turn to Herodotus we find a curious and, apparently, undesigned confirmation of his statement that the storm which destroyed the 200 ships was the same as that which broke on the Greek and Persian fleets at Artemisium and Aphetae respectively, after the first day’s engagement. He does not in this instance mention the direction of the wind; probably he did not know it; but he tells us (viii. 12). of δε νεκροὶ καὶ τὰ νανήγια ἐξεφέροντο és tas ᾿Αφέτας. The wind must have blown into the entrance of the Pagasaean Gulf, and, moreover, towards the Magnesian coast of it. It must in other words, have been a S. or 8.S.W. wind, the very wind which would blow vessels in the S. Euripus on to the Hollows of Euboea, Authorities seem to be, whether rightly or wrongly, in agreement as to the fact that Herodotus had no personal acquaintance with the region in which the events took place. This can only emphasize the undesigned confirmatory evidence which the

1 Cf, Strabo 445... τῆς Εὐβοίας τὰ Κοῖλα τὸν τόπων' κολποῦται γὰρ ἡ παραλία. λέγουσι τὰ μεταξὺ Αὐλίδος καὶ τῶν περὶ Γεραισ-

H.S.—VOL. XVII. R

222 ARTEMISIUM.

historian gives of the truth of his statement respecting the identity of the storms. ’

Some commentators make a difficulty as to the circuitous route which, so Herodotus tells us, this squadron took round Skiathos from Aphetae. I confess I cannot see the difficulty. The circuitous route was, it seems to me, evidently taken to deceive the Greek fleet at Artemisium, and was, moreover, admirably designed for so doing; in fact the Greeks do not seem to have suspected its intention until Skyllias informed them of its object. Nor is this surprising. ‘lhe squadron, as it doubled Cape Sepias, must have had all the appearance of sailing north, and might well give the impression that it had gone back to Therma or elsewhere for commissariat purposes. Moreover, a glance at the map will show that Skiathos shuts in the view at the end of the strait, and a squadron sailing round it, would be out of sight of the fleet at Artemisium.

Jf Herodotus had not seen this region, and 7f the story of the despatch of the squadron from Aphetae is not true, it is a very remarkable fact that the invention should be so topographically correct.

We next come to the tale of Skyllias.

After the departure of the squadron (vill. 7 ad fin.), the Persians reviewed their fleet at Aphetae. It is needless to say that this cannot have been done in a few minutes: it must have been an affair of some hours at least, and must, if carried out on that seventeenth day, have occupied a large part of what remained of it after the despatch of the squadron.

While this review was going on (vill. 8 ad init.), Skyllias deserted to the fleet at Artemisium. Herodotus thinks, as is indeed probable, that he went in a boat. We have no details of the adventure which might explain how he managed to get away in broad daylight without being pursued and brought back, for he must have had some ten miles to go. The hour of his arrival is, however, pretty well marked as the afternoon of a certain day (viii. 9), the seventeenth according to our calculation. The news he brought of the despatch of the vessels round Euboea was eminently calculated to create consternation in the Greek fleet. The dilemma was, indeed, a serious one. The detached squadron had got at least several hours’ start. Unless they sailed away promptly to oppose it, it might get through the narrows at Chalkis and land a force in rear of Thermopylae. But again, if they did this, the main fleet at Aphetae might sail into the Euripus from the north, and land a force behind the pass. There was the further possibility of their being caught in a trap between the two fleets unless they managed to defeat the 200 before the main fleet came up. It was absolutely necessary then to get a good start from Artemisium before the main fleet at Aphetae became aware of their departure. After long discussion they decided to remain where they were till after midnight, and then start for the inner strait. In the meantime they took the offensive against the fleet late in the afternoon (viii. 9), so Herodotus tells us.

He also gives a motive for their so doing, viz., ‘they wished to make trial of their mode of fighting, etc.’—here as elsewhere in Herodotus just the sort of

ARTEMISIUM.

223

motive which would suggest itself to some one unacquainted with the designs of those in command. The design of the latter was, no doubt, one which is common in the history of war at all times, viz., to cover a retreat by a previous attack, and to render the Persians less likely to suspect the movement which was about to be made to the inner strait.

We now come to a somewhat delicate calculation, but one which can be made with considerable probability from the data at our disposal. As far as the Persian fleet is concerned, the following events (all correction of Herod- otus’ chronology apart) had taken place on one and the self-same day :—

(1) The departure of the 200 vessels. (2) The review of the fleet. (3) In the late afternoon, an attack by the Greeks

It may be taken then as fairly certain that the first of these took place in the early morning, any time after daylight.

These 200 vessels were apparently picked vessels (viii. 7, ad. init.), We may be safe, then, in assuming that their sailing qualites were at least equal to, probably greater than, those of the fleet generally, i.c.,as we have seen, some ten miles an hour. It would require then some fifteen hours for them to accomplish the 150 miles from Aphetae round Skiathos to the south point of Euboea, which they must have reached about 8 or 9 o'clock that night. From what is necessarily implied by Herodotus’ account they must have been round this point before the storm broke upon them, otherwise they could not have weathered the headland. As they would then be within 70 miles of Chalkis, it is plain that the Greek fleet, starting after midnight, would not have been in time to stop them at the narrows, though it would have been in time to prevent their landing a force behind Thermopylae.

At the same time the Greek fleet had no alternative but to remain where they were, until they could withdraw without attracting the attention of the Persian fleet at Aphetae.

As a fact, had the Persian squadron ever arrived at Chalkis, it must have found there these fifty-three Athenian vessels which reached the fleet at Artemisium next day, to which fact the apparent immunity from disaster which the fifty-three enjoyed during the storm points. The Greeks at Artemisium seem to have had no actual information of these fifty-three vessels being on their way.’

1 A recent criticism of Herodotus’ story com- ments on the absence of ‘motive’ in Herodotus’ account of the arrival of the fifty-three Athenian vessels, It suggests that the retreat to Chalkis in the early part of the narrative is to be ex- plained as having been in reality the despatch of these fifty-three from Artemisium to that place with a view to defending the south entrance of the strait.

Does it seem probable that the Peloponnesian contingent would have consented to remain at Artemisium under such circumstances, and to entrust the re-opening of the seriously threat-

ened line of communicatians to a purely Athenian squadron? for there was, it must be remembered, according to this theory, no Per- sian fleet as yet at Aphetae to render retreat from Artemisium dangerous.

In the absence of ‘motive’ given, the most probable which can be suggested is that the Athenians, at this time novices in the fitting out of large fleets, had not been able to make more than 147 of their vessels ready for sea in time for the despatch to Artemisium, and sent on the remaining fifty-three when they were ready.

R2

224 ARTEMISIUM.

The events of the remainder of what we suggest as having been the seventeenth day are :—

(1) The attack of the Greek fleet on the Persians at Aphetae (viii. 10).

(2) The storm that night (viii. 12).

(3) The wreck of the 200 in that storm (viii. 13).

The next day, 1.6., the eighteenth of our calculation ; both sides remained inactive during the earlier half of the day (vii. 14).

At an unspecified hour, certainly earlier than the late afternoon! the fifty-three Attic vessels reinforced the fleet at Artemisium.

Herodotus then says:—atrai te δή σφεας ἐπέρρωσαν ἀπικόμενοι. καὶ ἅμα ἀγγελίη ἐλθοῦσα, etc. of the destruction of the 200 at the Hollows.

The arrival of this news coincided more or less with the arrival of the fifty-three, or may even have been~brought by them.

The fifty-three must have spent the night at Chalkis. The storm apparently did not come before daylight, otherwise we should have expected that the fleet at Artemisium would have started to meet the 200.

After daybreak such a movement in view of the Persians at Aphetae was not to be thought of.

Now we do not know

(1) the hour of the wreck at the Hollows,

but, since the Hollows are just round the south cape of Euboea,

and the 200 must have rounded that cape before the storm broke,

it must have taken place early in the storm,

1.6,, early in the night.

(2) the hour of the departure of the fifty-three from Chalkis.

Probably they did not start at daybreak, because they would in all likeli- hood wait till the storm had quite blown itself out.

They had 70 miles to go to Artemisium, at least seven hours’ voyage, possibly more, since the Greek vessels were not as good sailers as the Persian.

They certainly arrived there before the late afternoon, so the probability, if there be any, is that they started about 6 a.m.

If the wreck, as indeed must have been the case, took place early the night before, this would give eight or nine hours for the news to reach Chalkis. We may be certain that the course of the fleet had attracted the notice of the inhabitants of South Euboea, who must, too, have seen the Attic squadron pass up the strait earlier in the day. There is no impossibility that the tidings of the wreck were actually carried to Chalkis before the Attic squadron sailed thence, and were carried to the fleet at Artemisium by the latter.

The last event of this day was the attack by the now united Greek fleet on the Cilician contingent of the Persians (viii. 14).

1 Cf. φυλάξαντες τὴν αὐτὴν ὥρην, after the arrival of the fifty-three,

ARTEMISIUM. 225

viii. 15 is filled with important information. It is a journal of the nineteenth day. The first significant words are τὸ ἀπὸ Ξέρξεω δειμαίνοντες, It will however be well to have these aside until we come to a comparison with the journal of Thermopylae.

The Persians now for the first time took the offensive. The battle which ensued seems, from what Herodotus tells us, to have been somewhat of a Pyrrhic victory for the Greeks. In consequence ‘dpyopov δή ἐβούλευον ἔσω és τὴν ᾿Ελλάδα᾽ (viii. 18, ad fin.). And now the Peloponnesian policy prevailed. It needed indeed but a slight weight in the balance of the then state of feeling to make it do so, and we may gather from Herodotus that Themistocles, whatever he may have thought of its wisdom, gave up for the time being all idea of opposing it. This anxiety seems to have been solely to get the Greek fleet away before the Persians could become aware of its having moved and the tale in viii. 19, is doubtless rightly interpreted to mean that he advised the lighting of fires on land in order to give the Persians the impression that the fleet was still at its station. The guise in which Herodotus presents the story is probably due to his well-known tendency to convey a moral, in this case the danger of disobeying an oracle. It will be noticed that the action of the Greeks on this occasion singularly accords with that of the day but one before, when they had not ventured, even under the most pressing danger from the squadron of 200, to leave their post under circumstances which would render their withdrawal immediately obvious to the Persians at Aphetae.

These measures must have been taken late in the afternoon of the nineteenth day, for the fight took place at midday (viii. 15).

The arrival of the news of the disaster at Thermopylae (viii. 21) set the seal to the decision to retire and the retirement took place during the night.

It would seem, then, that, with the exception of the palpable crowding of the events of at least two days into one, the tale which Herodotus gives us of Artemisium is hardly deserving of that suspicion which some commentators have cast upon it.

It really contains two difficulties.

(1) The one to which I have just referred, viz., the manifest crowding of the events of two days into one, a mistake of the existence of which Herodotus himself gives us plain evidence.

(2) The mistake which still remains, though now reduced to one day, in the lack of chronological concord between the journals of Thermopylae and Artemisium.

I see no possibility of arriving at any evidence worth calling such as to the cause of the latter mistake. Herodotus gives us none, either directly or incidentally, and it is a pure conjecture on my part to say that I believe the miscalculation or mistake to have been made with respect to the number of days which intervened between the departure of the army and that of the fleet from Therma. It would, at any rate, not be unnatural for a man, speaking from recollection, to make a mistake of one day in a period of such

a length.

226 ARTEMISIUM.

Some would reconcile the chronology by subtracting two days from the time which Xerxes spent inactive before Thermopylae, on the plea that the delay is unaccountable.

But is it, after all, so strange that the experienced generals of Xerxes should have hesitated about advising a direct attack on a position of such strength ?

Of its strength we have ample evidence quite apart from Herodotus.

There is the case already mentioned of Brennus and his Gauls in B.C. 279.

There is also what seems to me the most instructive case of the methods employed by Philip of Macedon with a view to getting possession of it. For his Greek policy the possession of Thermopylae was of capital importance, and yet, though possessed of what was far and away the best army of his time, he dare not attack it directly and submitted to long postponement of his designs in order to avoid the hazard of so doing.

The suspicion with which Herodotus’ statement of the four days’ delay at Thermopylae has been received is apparently due to the absence in his account of any substantial motive for the delay. We are told (vii. 210) that after reconnoitring the pass Xerxes τέσσερας μὲν δὴ παρεξῆκε ἡμέρας, ἐλπίζων αἰεί σφεας ἀποδρήσεσθαι. The fact of the delay we may suppose Herodotus got from his source of information, but a very brief consideration of the circumstances may convince us of the improbability of his having been able to get any information worth calling such as to the real reason for the delay. The strength of the Greek position, ample proof of which subsequent history, as we have said, affords, was no doubt the first reason, but this only partially accounts for the course of events. In order to elucidate the point fully, let us consider the parallel journals of Artemisium and ‘Thermopylae, taking the former in its revised form 1.6. adding a conjectural day after the departure of the army from Therma, and dividing the events of Herodotus’ sixteenth day in accordance with what is demanded by his narrative.

No. otis Ho. of wi aay in A . a DS text ot coin. Events at Artemisium. eT ane 9 added. paper. 13 12 P. fleet leaves Therma. Early morning. P. fleet arrives at Sepiad strand. Evening. Capture of three Greek scouting vessels | Probably before by Persian advance squadron. mid-day. G. fleet at Artemisium. G. fleet receives news from Skiathos of | Evening, _ pro- the loss of the three vessels. bably late. 14 15. | P. army reaches | First storm begins. Malis. Wreck of part of P. fleet at Sepiad strand. G. fleet retires (or driven) to Chalkis. | Probably early morning.

No. of day with

conjec- tural day added.

15

16

17

18

19

20

No. of

day in

text of this

paper.

14

15

16

17

18

19

Events at Thermopylae.

P. army inac- tive before

Thermopylae.

P. army inac- tive before Thermopylae.

P. army still in- active.

Ῥ: army makes first attack on Thermopylae.

Second attack on ‘Thermo-

pylae.

Third and suc- cessful attack on Thermo-

pylae.

ARTEMISIUM.

Events at Artemisium.

Storm continues.

P. fleet at Sepiad strand,

G. fleet at Chalkis.

Report of P. disaster reaches G. fleet at Chalkis.

Storm continues. Fleets as on 15th day.

P. fleet moves to and arrives at Aphe- tae.

G. fleet moves back to Artemisium.

G. fleet captures fifteen P. vessels.

G. commanders consult about retire- ment from Artemisium.

G. commanders on Themistocles’ per- suasion determine to remain at Artemisium.

P. fleet despatches squadron of 200 to circummavi ate Euboea.

P. review their fleet.

Skyllias carries news to the G. fleet of the despatch of the 200.

G. fleet engages P. fleet at Aphetae. Second storm begins.

Wreck of the 200 Persian vessels in the Hollows of Euboea.

Fifty-three Attic vessels on their way to join the G. fleet spend night at Chalkis.

Both fleets inactive.

The fifty-three Attic vessels join the G. fleet.

G. fleet receives news of the wreck of the 200.

United G. fleet attacks Cilician con- tingent of Ῥ, fleet.

P. fleet takes offensive against αὶ, fleet.

General engagement.

G. commanders decide to retire ἔσω és τὴν Ἑλλάδα.

News οἵ disaster reaches α fleet.

G. fleet retires from Artemisium.

at Thermopylae

227

Bah Fg time

Arrival early in the afternoon,

In afternoon.

Probably even- ing.

Morning.

Probably early morning.

After despatch of the 200. During the re-

view. Arrived at Artemisium in afternoon. Late in afternoon, Early in the evenin

Early in the night.

Night

ae part of

Earli ier than the late afternoon. About same time

as above. Late afternoon.

228 ARTEMISIUM.

It now remains for us to consider briefly the relation between the two series of events at Thermopylae and Artemisium respectively.

A comparison of the two brings into immediate prominence the fact that the delay of four days at Thermopylae corresponds practically with the time during which the Persian fleet was delayed outside the strait by the storm. It is also noticeable that the first attack on Thermopylae was made on the very day after the fleet arrived at Aphetae, when it would be in a position to force the strait, or, at any rate, to keep the Greek fleet employed. It is not unreasonable so suppose that Xerxes reckoned on his fleet having no difficulty in forcing the strait immediately after its arrival. It was immensely superior in numbers, and was drawn largely from nations who had the reputation of being the best seamen of their time.

I venture to think, too, that Xerxes and the experienced generals with him may have been apprehensive lest the Greek fleet should sail close in shore and take part in the defence of the pass, an actual possibility which the events of two centuries later make quite evident. I do not say that the Greeks had any such intention, at least, there does not exist the slightest evidence of their having had it, but nevertheless it was plainly a contingency which might present itself to the consideration of the Persian council-of-war, and might decide them to defer the attack until their own fleet arrived to occupy the attention of the Greek naval force in the straits.

The despatch of the 200 ships round Euboea would seem to have had as its main object the capture of the whole Greek fleet, as, indeed, Herodotus tells us. We find, too, this design repeated at Salamis. Xerxes evidently thought that his fleet at Aphetae could deal with the Greek fleet at Artemisium without much difficulty, and there is reason to believe he was not far wrong. That last day’s fighting must have ended in something much more like a defeat than we might suppose from Herodotus’ story, and the absence of any mention of any opposition on the part of Themistocles to the plan of retreating from Artemisium is somewhat significant. It was one thing to face the immensely superior numbers of the Persian fleet in the narrow strait of Salamis, less than a mile broad: it was another to face that fleet in the broad northern bend of the Euripus where there was plenty of sea room.

That the landing of troops behind Thermopylae in case of the failure of a direct attack may have been a secondary object, is, of course, possible, or even probable, and had not the discovery and successful use of the path of Hydarnes rendered this unnecessary, we may suspect that the main fleet at Aphetae would have been employed, after the loss of the squadron of 200, in an attempt to effect such a landing by forcing the strait.

As far as can be seen from Herodotus’ narrative, Xerxes’ strategy after his arrival in Malis seems to have been :—

(1) To defer attack on the pass until his own fleet should have arrived within the channel to occupy the attention of the Greek fleet.

(2) To then attempt a direct attack on the pass.

(3) To take measures for the capture of the whole Greek fleet.

ARTEMISIUM. 229

(4) To provide for the contingency of failure of direct’ attack on Thermopylae in two ways :—

(a) By giving orders to the main fleet to force the strait and land troops in rear of the pass.!

(Ὁ) By giving orders to the squadron of 200 to land troops in rear of the pass.

This paper has not been written out of any spirit of conservatism. Anyone who, like myself, heartily accepts such historical reconstructions as Mr. Macan’s ‘Marathon’ or Professor Bury’s ‘ Aristides at Salamis’ can hardly be suspected of such a tendency. But after studying this part of Herodotus’ history, it seems to me that this section of his narrative does not demand anything of the nature of a forced explanation, still less calls for practical rejection, for to this some of the proposed reconstructions amount whether the writers intend it or not. That the narrative is defective in chronology, is, of course, quite plain, but calculations made from the bare statements in it, result, when the chronological defect is eradicated, in singular accord. So at least it seems to me. I should very much like to know how it seems to others who are acquainted with the scene of events.

G. B. Grunpy.

1 Cf. τὸ ἀπὸ Ξέρξεω δειμαίνοντες, viii. 15.

230 THE ACCOUNT OF SALAMIS IN HERODOTUS.

THE ACCOUNT OF SALAMIS IN HERODOTUS.

It is, I think, impossible to read Professor Goodwin’s article on ‘Salamis’ in the Journal of the Archacological Institute of America, 1882—83, without arriving at the conclusion that he has made out a very strong case for his main thesis, viz., that the plan of the battle was wholly different from that which has been constructed by modern historians, who misinterpret, as I suppose Professor Goodwin would hold, the account of Herodotus. I confess to cordial adherence to the main proposition contained in the article, but I find great difficulty in understanding the argument by which he endeavours to reconcile his view with the account given by Herodotus.

That there is some defect in Herodotus’ account is clear. At first reading it seems to consist in an absence of information as to the movements of the fleets in the battle itself.

My impression is, however, that Herodotus had at his disposal informa- tion with regard to those movements, but misunderstood it.

It will be necessary, in order to explain exactly what I mean, to go over much of the ground which Professor Goodwin traverses, and indeed to make use of some of his arguments.

It is generally agreed, in fact the evidence is unanimous on this point, that the Persians drew up their fleet in some way so as to block the eastern end of the Salamis strait, though the way in which they did this is disputed. But the main points in dispute are :—

(1) As to the locality of the part of the other end of the strait which they blocked so as to prevent the Greek fleet from escaping.

Viz., whether it was the narrow portion of the eastern strait at the point where it enters the Bay of Eleusis, or whether it was the strait between Salamis island and the Megarid coast.

(2) As to the position of the Persian fleet, especially at daybreak, on the morning of the battle.

The scheme of the battle given in nearly all, if not all, modern histories of Greece, represents the Persian fleet as drawn up on the morning of the battle along the Attic coast from the narrows at the entrance of the Bay of Eleusis almost to the mouth of Piraeus harbour, while the Greek fleet is opposite, extending from a point some way north of the Island of St. George almost to the end of Kynosura (v. Grote, etc.)

I notice that this scheme has been adhered to in histories of Greece

THE ACCOUNT OF SALAMIS IN HERODOTUS. 231

which are either new or have been re-edited since Professor Goodwin’s article was published.

The objections to it which Professor Goodwin urged seem to me 80 strong that I am surprised that the scheme is still adhered to by great authorities on Greek history. The reason for this adherence I have not seen stated in print, but I can only suppose that those who hold to the old view reject wholly the version of Diodorus, where it differs from that of Herodotus, and would hold that the latter is not contradicted in any essential respect by the, for historical purposes, imperfect account of Aeschylus. Professor Goodwin lays down the canon that on any detail he does mention, Aeschylus is the authority to be followed, because he was an eye-witness. He further seeks to reconcile Herodotus’ account with that of Aeschylus, with the result that he reproduces a history of the battle which is in nearly all essential respects that of Diodorus.

It is plain, of course, that the ‘ eye-witness’ argument may be carried too far, since it is possible, if not probable, that the accounts of Herodotus and Diodorus were drawn from the records of eye-witnesses.

The thesis which I propose to put forward is that we have in the tale of Salamis one of the rare cases in which Diodorus has either obtained better information or made better use of his information than Herodotus. I believe Professor Goodwin might possibly agree to this. But I would urge that those who have adopted the ‘old’ scheme of the battle have rightly interpreted Herodotus’ view, and that the mistake, if any, is Herodotus’, _ not theirs.

In this, I believe, Professor Goodwin would disagree with me.

The arguments against the old scheme, of which the most convincing have been already stated by Professor Goodwin, are :—

Since the passage between Attica and Psyttaleia is 1,300 yards wide ;

And that between Aegaleos and Salamis 1,500 yards;

And between Aegaleos and St. George Island 1,200 yards ;

ae. the whole channel is very narrow ;

(2) How could the Persian movement of cutting off be accomplished so secretly that the Greeks got no wind of it? (H. viii. 78, Plut. Them. 12, Arist. 8.)

How could the Persians have slipped along the other side of the narrow strait in the night unperceived? ‘Tradition says, moreover, that it was a moonlight night.

(Ὁ) Can we believe that the Greek fleet was allowed to form quietly in line of battle at the other side of this narrow strait, in the very face of the Persian fleet only a few hundred yards distant ?

Surely the Persian fleet would, being eager to capture the Greek fleet, have seized the ships while the crews were preparing to embark,

(c) Aeschylus, an eye-witness, testifies that it was only after the Greeks had rowed forward from their first position that they were fairly seen by the Persians (Aesch, Pers. 400).

(d) Aeschylus, Pers, 441-464, H. viii. 76, 95, Plut. Arist. 9 concur in the

232 THE ACCOUNT OF SALAMIS IN HERODOTUS.

statement that Xerxes landed a body of Persians on Psyttaleia because he thought that it would be a central point of the sea fight.

Such are Professor Goodwin’s objections to the old scheme. To the last I would add that Herodotus expressly describes the measures taken with regard to Psyttaleia as being synchronous with those for blocking the straits (viii. 76).

Of these objections :— ~

(a) is strong as being Herodotus’ own evidence, and it is on Herodotus that the old scheme must rely. The passages quoted from Plutarch are, how- ever, manifestly from the Herodotean source.

Objections (6), (6), (d@) seem to me unanswerable. As I read the narrative, the old scheme of Grote and others cannot stand in face of them.

Experience has convinced me of the fact that Herodotus is a most difficult and dangerous author to criticise. At the same time, in those parts of his Greek history which I have had occasion to examine minutely,— some on the actual scene of events,—I have never come across any statement of fact which could be suspected to be the pure invention of the author. His mistakes, in so far as can be judged, arise almost wholly from:

(1) Misreading of sources,

(2) Use of defective or mistaken sources, not from the invention of imaginary facts. His painful conscientiousness seems to be genuine, not fictitious. But, eminently unmilitary himself, he was peculiarly liable to misunderstand the information at his disposal with regard to military matters, and this, as it seems to me is exactly what has happened with regard to his account of Salamis, and in the following way :—

It is, of course, a commonplace of criticism to say that Herodotus gives us no account of the general movements or manceuvres of the two fleets on the actual day of battle, save that he mentions that the Aeginetan vessels fell on the Phoenician ships which the Athenians put to flight. What I may call the enunciation of my proposition is this :—

This failure of information in this part of his narrative is due to the fact that he had already in the previous part of it used up his information on this point.

He antedated a movement made on the night preceding the battle to the previous afternoon, and further antedated the movements in the battle itself to the night preceding the battle.

I will now attempt to prove this proposition. Unfortunately I do not see my way to doing so without going into the detail of the description of the fighting.

Apparently on the day but one before the battle (cf. H. viii. 64) Themistocles at the meeting of commanders urged the importance of fighting in the strait (cf. H. viii. 60, D. xi. 17). His idea was, of course, that in the narrow seaway the Greek fleet would not be outflanked, and the superior numbers of the Persians would not be of avail. It is noticeable (1) that his argument is dependent on the narrowness of the possible front of the fighting line, (2) that unanimous evidence of historians points to the fact that the

THE ACCOUNT OF SALAMIS IN HERODOTUS. 233

event showed it to be a sound argument. But, if the fleets were ranged on either side of the strait in the actual battle, and not across it, it is somewhat difficult to see why so much success should be attributed to the design.

Day preceding the battle.

Ship despatched to Aegina to fetch the Aeacidae (H. viii. 64).

Persian fleet at Phaleron (H. viii. 66—T. 12).

Persian fleet, so says Herodotus, puts out towards Salamis and quietly forms line of battle. It is too late to fight that day (H. viii. 70),

This movement, then, presumably took place in the late afternoon.

Here I think Herodotus’ mistake begins. He represents this movement as made before Xerxes received Themistocles’ message, i.e. not causally con- nected with it at all. But both Aeschylus and Diodorus almost certainly describe the same movement as having been made after Themistocles’ message had been received and in consequence of it.

Cf. Aesch. Pers, 374-83 describing the embarkation of the Persians with a view to (368-9) ranging the ships in three lines so as to guard the exits and the ‘roaring friths’ and (370-8) sending ships by a circuit round the Island of Ajax to cut off the Greek retreat.

Cf. also Diod. xi. 17.

In other words neither Aeschylus nor Diodorus has any mention of a movement of the Persian fleet from Phaleron until after Themistocles’ message was received.

What Herodotus describes and mistimes is really the movement which the Persians made at night to block the eastern strait.

After the receipt of Themistocles’ message the Persian fleet did on the night before the battle put out, and these measures were taken :—

(1) Psyttaleia was occupied (H. viii. 76, Aesch. Pers. 449, and apparently Plut. Avist. 8)

because Xerxes expected evidently that it would be an important position to hold during the battle (v. H. and A. P. ad loc. cit.).

Professor Goodwin thinks, as indeed appears probable, that it must have been near the centre of the Persian lines of battle as at first arranged.

(2) The strait was blocked on the East (P. Them. 12, A. P. 368-373,

(3) The strait was also blocked on the West} H. viii. 76, Diod. xi. 17.

With regard to (2), the consideration suggested by the occupation of Psyttaleia and the fact that the measure was carried out unknown to the Greeks, might suggest that the Persian line was extended from behind Kyno- sura, South of Psyttaleia to the Attic coast near Piraeus harbour.

As regards (3), Diodorus says expressly that the Egyptian contingent was sent to block the strait between Salamis and the Megarid (xi. 17).

This is a curious triangular concord at this point in the history.

Diodorus says that the Egyptian contingent was sent.

Plutarch speaks of 200 ships sent off by night.

Herodotus tells us that the Egyptian contingent numbered 200.

The assertion of Diodorus, if it stood alone, might not be held to be very

234 THE ACCOUNT OF SALAMIS IN HERODOTUS.

convincing, but Aeschylus practically says the same thing, viz., that Xerxes sent ships by a circuit round the Island of Ajax to cut off the Greek retreat.

Professor Goodwin would apparently argue that Herodotus’ description of the movement really implies, if rightly interpreted, the same thing. Now what Herodotus does say is this (vili. 76), ἐπειδὴ ἐγίνοντο μέσαι νύκτες, ἀνῆγον μὲν τὸ am’ ἑσπέρης κέρας κυκλούμενοι πρὸς THY Σαλαμῖνα, ἀνῆγον δὲ οἱ ἀμφὶ τὴν Κέον καὶ τὴν Κυνόσουραν τεταγμένοι, κατεῖχόν τε μέχρι Μουνυχίης πάντα τὸν πορθμὸν τῇσι νηυσί. The earlier part of the passage evidently refers to the blocking of the strait on the west. On the general question of meaning, I do not think that this can mean anything else but that the west wing, as he calls it, moved through the strait, not round Salamis, to block the west end of the strait at the point where it enters the Bay of Eleusis, z.c. between Aegaleos and Salamis island. This, indeed, is the natural interpretation, which those who follow what I have called the ‘old’ scheme of the battle put upon the words.

But I venture to think very strongly that Herodotus is describing in this passage, (though he, of course, was not aware of the fact,) the movements in the actual battle. We have seen that he has ascribed to the previous after- noon a movement which Aeschylus and Diodorus say took place at night, and he is now ascribing to the night a movement which was really made in the morrow-morning’s battle. He has antedated his information, and hence he can tell us practically nothing of the general movements in the fight, for he had already used up all the information he had with regard to the move- ments of the two fleets.

T shall have, of course, to recur to this point, when we come to discuss the actual manueevring in the battle.

The movements in the night were made quietly, in order that the enemy might not get knowledge of them (H. viii. 76).

They took up the whole night (H. and A. P.).

This tells in favour of the movement round Salamis island to the west strait between the island and the Megarid.

The Greeks were unaware of the movements until Aristides came from Aegina to inform them of them (H. viii. 79, Plut. 7.12, A. 8). Aristides further told them that he had had great difficulty in getting through the blockading squadron (H. viii. 81). It would be mere guesswork to surmise how he got through.

Day of the battle.

When the day dawned the position of the fleets seems to have been :—

The Greek fleet close to Salamis town, probably in bay north of it; the Persian fleet in a line across the strait almost due east and west, with Psyttaleia near the centre.

These positions are to be conjectured from the following facts :—

(a) The Greek position, from Diodorus’ statement that, after embarking, the Greeks sailed out and occupied the strait between Salamis and the Herakleion (D. xi. 18). Plutarch tells us that the Herakleion was where the

THE ACCOUNT OF SALAMIS IN HERODOTUS. 235

Island of Salamis is separated from the main land by a narrow passage. It must therefore have been on Aegaleos opposite Salamis town.

(Ὁ) The Persian position is defined by very strong evidence :

(1) Respecting the occupation of Psyttaleia already quoted ;

(2) (And most markedly) by the statements of Aeschylus and Diodorus.

A. P, 415 says that the Persian ships fell foul of one another when they came into the narrows.

D. xi. 18 says that the Persians in their advance at first retained their order, having plenty of sea room, but when they came to the strait, they were compelled to withdraw some ships from the line, and fel! into much confusion.

Their position at dawn must then have been in the broad part of the strait, just before the narrows begin. It will be seen that there is a most marked diminution in breadth so soon as Kynosura is rounded. I say above ‘just before the narrows begin,’ because the evidence of Aeschylus and Hero- dotus shows that the two fleets came in sight of one another very soon after they began to move (cf. A P. 400, H. viii. 84).

It was after dawn when the Greeks embarked (H. viii. 83 ad init.).

When we come to the question of the positions of the various con- tingents in the two fleets Aeschylus unfortunately does not aid us.

Herodotus and Diodorus are in agreement as to:

Medised Greeks on Persian L. wing (H. viii. 85, D. xi. 17) ;

Phoenicians on Persian R. wing (H. viii. 85, D. xi. 17).

Diodorus further tells us :

Cyprians with Phoenicians on R. wing (D. xi. 19).

Cilicians, Pamphylians, Lycians next them (D. xi. 19).

Curiously enough the two authorities differ with regard to the few details they give of the Greek array.

Both agree that Athenians were on Greek L. wing (H. viii. 85, D. xi. 18).

But the Lacedaemonians according to Herodotus (viii. 85) were on the R. wing, but according to Diodorus (xi. 18) on the left with Athenians.

Aeginetans and Megareans according to Diodorus were on the R. wing (xi. 18).

It is difficult to say whether the evidence on the point where the authorities are in conflict, viz., as to the composition of the Greek right, inclines to the side of Herodotus or Diodorus. I shall give reasons for believing that the account of Herodotus is more probably the right one.

The Movements of the Fleets in the battle.

The Persian fleect.—The information which we do get as to the advance of the Persian fleet and its manceuvring is wholly in accord with what seems to me to be almost certainly defined as its position at dawn.

Aeschylus (P. 368-9) tells us that it took up its original position in three lines.

Describing its appearance as it advanced into the narrow part of the strait, he speaks of it coming on in a ῥεῦμα, which can only refer to some formation in column, or something resembling a column.

236 THE ACCOUNT OF SALAMIS IN HERODOTUS.

Diodorus supplies the connecting link between the two formations when he says that the Persians when they came into the strait were compelled to withdraw some ships from the line (ΧΙ. 18).

If the chart of the strait at this point be examined it will be seen that after passing Psyttaleia and Kynosura it not only narrows, but turns west- ward at right angles. The Persian fleet had consequently to accomplish a most difficult manceuvre of a double kind, viz. :

(a) To reduce their front.

(Ὁ) To execute a wheeling movement to the left, of which their extreme left wing would form the pivot.

What seems to have taken place is this; their right, and possibly their centre having reduced their front passed through the strait east of Psyttaleia in some sort of column formation; then wheeled to the left to turn the corner of the strait, while their left wing marked time, as it were, in the strait west of Psyttaleia. The latter would be hidden from the Greek fleet by the some- what lofty rocky promontory of Kynosura. But the right wing passing east of Psyttaleia would almost immediately come into sight, and would present to the Greeks that appearance of a ῥεῦμα, which the eye-witness Aeschylus describes.

The Greek fleet.

Meanwhile the Greek fleet had moved. It formed line in the first instance in the strait between Salamis and the Herakleion, so Diodorus tells us (xi. 18). From that position it advanced along the strait, and soon after starting, came in sight of the Persian (A. P. 400). Needless to say that this detail, given by Aeschylus, is singularly in disaccord with the ‘old’ scheme.

We now come to the difficult question as to the position of the fleets when contact took place.

Aeschylus gives a most important detail bearing on this point, when he says (A. P.401) that the Greek right wing led the advance and the remainder came behind. If we consider what this implies, and further take into con- sideration another fact mentioned by Aeschylus (A. P. 411) and supported by Herodotus (viii. 84, ad init.) to whit, that a Greek (1.6. Athenian) ship began the battle by attacking a Phoenician, we see that the Persian right, 1.6. that part of their fleet which had wheeled on the outside had got in advance of the rest of their line, 1.6. that the two fleets were in a kind of échelon formation when contact took place, not in line direct across the strait, which would imply positions running north and south, but slantwise in positions running from north-west to south-east nearly. In order to make clear what I mean, I append three small sketch maps of the straits, showing (1) original position, (2) advance, (3) contact, of the two fleets.

It is, I believe, this position of the fleets at the point of contact which Herodotus describes in vili. 76, and viii. 85.

First, with respect to viii. 76.

Herodotus had already ascribed the Persian movement from Phaleron to the position at Psyttaleia to the afternoon of the day preceding the battle, orior to the receipt of Themistocles’ message by Xerxes.

THE ACCOUNT OF SALAMIS IN HERODOTUS. 237

This, we have seen, was a mistake. That movement was made, as Aeschylus and Diodorus show clearly, subsequent to the receipt of the message, and in the night.

What is the first immediate consequence of his mistake? He had separated with respect to time the Persians’ movement to the position at Psyttaleia from their movement to block the western end of the straits, whereas these two movements were synchronous,

Now let us consider what his position as an historian would have been after making this mistake.

He would be quite aware of the notorious fact that a movement was made in the night to block this western issue, and was made in consequence of Themistocles’ message.

1 ORIGINAL POSITION

mt AEGALEU®

Y

-HERACLEUM

59 Ζ

ι ὌΠ" ὕ

He would naturally suppose that the second movement indicated in his sources of information was this movement, whereas it was really the movement in the actual battle itself.

The description of the movement in viii. 76 is very closely applicable to what we have seen from other evidence must have been the movement in the battle. He Bays: ἀνῆγον μὲν τὸ ἀπ’ ἑσπέρης κέρας κυκλούμενοι πρὸς τὴν Σαλαμῖνα, a ἀνῆγον δὲ οἱ ἀμφὶ τὴν Κέον τε καὶ τὴν Κυνόσουραν τεταγμένοι, κατεῖχόν τε μέχρι Μουνυχίης πάντα τὸν πορθμὸν τῇσι νηυσί.

In the first place we find out from chap. 85 what he meant by the west wing, viz., κατὰ μὲν δὴ ᾿Αθηναίους ἐτετάχατο Φοίνικες (οὗτοι γὰρ εἶχον τὸ πρὸς ᾿Ελευσῖνός τε καὶ ἑσπέρης κέρας)" κατὰ δὲ Λακεδαιμονίους Ἴωνες" οὗτοι δ᾽ εἶχον τὸ πρὸς τὴν ἠῶ τε καὶ τὸν Πειραιέα.

It sems to me to be fairly clear that he supposed the wheeling movement

H.S.—VOL, XVII. 8

238 THE ACCOUNT OF SALAMIS IN HERODOTUS.

of the right wing (the Phoenicians) which brought that wing as we have seen in front of the rest of the line, which movement was evidently accurately described in his source of information (cf. especially κυκλούμενοι πρὸς τὴν Σαλαμῖνα), to be a movement through the strait to block up the narrows north of Salamis town, near the Bay of Eleusis. The remaining words, ἀνῆγον δὲ οἱ ἀμφὶ τὴν Kéov te καὶ τὴν Κυνόσουραν τεταγμένοι, are taken also, no doubt, from his source of information and refer to the movement of the left Persian wing in the battle.

Of course Herodotus misinterprets the meaning in consequence of the mistake he has made, and adds the words which follow κατεῖχόν τε, etc. He has to bring this latter body of ships in touch with those which he has in imagination moved right up the strait.

2 ADVANCE

-HERACLEUM

SMILES

The τὸ ἀπ᾽ ἑσπέρης κέρας of ch. 76 is no doubt of the same origin as the τὸ πρὸς ᾿Ελευσῖνός te καὶ ἑσπέρης κέρας of ch. 85, and the latter is used in describing the position in actual battle. Strict geography might take excep- tion to the description of this wing as the west wing, though when the battle began it did apparently lie north-west or thereabouts of the other wing, but the distinction between the two wings in ch. 85 as πρὸς ’EXevoivos and πρὸς τὸν Πειραιέα is, as will be seen, correct and good.

Before I close I should like to say one word as to the battle itself, and the question of the conflicting evidence as to the position of the Aeginetans and Lacedaemonians in the fighting line. ‘The Aeginetans come in for the highest praise in Herodotus’ account. That they would not have got the praise from him unless they deserved it in some notorious way, we may be sure. The philo-Athenian historian was not likely to commend them without

THE ACCOUNT OF SALAMIS IN HERODOTUS. 239

cause. But Herodotus also mentions a detail about them in the battle which he is extremely unlikely to have invented, namely, that they fell on the Phoenician ships which fled before the Athenians (vil. 91). I think we may fairly suspect from Herodotus’ language that the Aeginetans were credited with dealing the decisive blow in the battle, though the Athenians had apparently the hardest fight (H. viii. 93). But this detail about the Aeginetans falling on the Phoenician ships makes it highly probable, if not certain, that they were close to the Athenians in the line, probably in the centre or left-centre, ic. not on the right, as Diodorus says, and we may, I think, conjecture that the decisive action of the battle was the breaking of the Persian line at the centre by the Aeginetans, after which the latter were able to fall on the flank of the Phoenicians, the

3.CONTACT

-HERACLEUM

Wily ὡς ~ Wi

Ss)

CYNOSURA

Persian right wing, whom the Greeks regarded, as may be seen from various hints in the accounts of the battle, as the really formidable element in the Persian fleet.

It is possible that the actual fighting may have gone in accordance with the original plan of attack of the Greek commanders, with such modifications as the confusion into which the Persian fleet fell in the course of executing their difficult mancuvre of necessity introduced into it. Of that confusion the Greeks apparently took full advantage but still the crushing of the Persian centre, probably consisting of the poorest material in their fleet, by what was looked upon, if Diodorus be right, as, after the Athenians, the most capable portion of the Greek fleet,

to be followed by an attack on the flank of the powerful Persian right 5.2

240 THE ACCOUNT OF SALAMIS IN HERODOTUS.

wing, may have been in outline the original design of Eurybiades and his colleagues.

It is not necessary, perhaps, for me to acknowledge further the extent to which I am indebted to Professor Goodwin’s article, since it will be fully appreciated by those who are interested in this side of Greek history, to whom certainly that article will be well known.

G. B. GRUNDY.

THE TEXT OF THE HOMERIC HYMNS: IV. 241

THE TEXT OF THE HOMERIC HYMNS. Parr IV, APOLLO.

RECENT literature on this Hymn is almost limited to the notes of R. Peppmiller, Philologus, 1884, p. 196 8ῳ., 1894, p. 2538 8ᾳ., and H. Pomtow, Neue Jahrb. f. Phil., 1886, p. 176, and the articles of A. Kirchhoff, Sitzungsberichte der preuss, Akad, xlii. 1893, and A. W. Verrall in this Journal vol. xiv. pp. 1 sgg. (1894). For Apollo we have the account by Roscher in his Lexicon ; Mr. Farnell (Cults of Greek States) has not yet treated him.

Want of illustration and of positive information upon the topics with which the Hymn deals, is the chief stumbling block to its interpretation. We are practically entirely ignorant, so far as other sources are concerned, at these places : the geographical names Εὐρεσίαι 32, Αὐτοκάνη 35, Aicayén 40, Λέκτος 217, ᾿Αρήνη ’Apyudén Αἶἷπυ 422, 423; the nature of Eilithyia’s necklace 103; the recitations at the Delian festival 156 sg., Apollo’s ‘brides’ 208 sg., the observances connected with chariots at Onchestus 230 8ῳ., the epithets of Apollo πύθιος 373, δέλφειος 496; the part taken by Cretans in the Delphic worship 393 sg. It is the more disappointing that the newly found Delphic inscriptions, so far as they have been published, contribute nothing to the elucidation of these points. The fragments of Hymns to Apollo, whatever perturbation they may have caused in the theory of Greek music, are singularly barren as literary documents.

This condition of ignorance has brought the usual result, that the Higher Criticism has marked the document for its own. Even the usually judicious Ruhnken divided the Hymn into two. I venture to think that consideration tends to show that even where the full import of the context is unrealisable, the grammatical sense presented by the tradition is clear, and that therefore the text may vindicate its soundness. When the darkness that surrounds the subject-matter of documents is deep, a prudent editor will, pending the arrival of better lights, at least guard the wording and the order of the texts for which he is responsible.

In three places in the poem the text has literally disintegrated, at 59, 152, 211; the first two of these singular corruptions have been fairly satisfactorily healed. Several lacunas, but of small extent, appear necessary.

18. ὑπ᾽ ᾿Ινωποῖο ῥεέθροις. Reiz and A. Matthiae substituted ἐπ᾽, the necessity of which with Ilgen I doubt. As Ilgen remarks, the Inopus is con-

242 THE TEXT OF THE HOMERIC HYMNS: IV.

ceived as a mountain torrent (vv. 17, 26), and Leto might be said to be ‘under’ its waters as it fell steeply. Or ὑπὸ may have the more general sense of ‘near, about,’ as in Apollonius ii. 794 ὄφρ᾽ ἐβάλοντο | οὖρα βαθυρ- pelovtos ὑφ᾽ εἱαμεναῖς Ὕπίοιο. Φ 87 ἐπὶ Σατνιόεντι ; ἐπὶ Strabo, ὑπὸ

all MSS.

20. νόμος βεβλήαται ὠδῆς. Barnes removed the singular, which was kept by Maittaire (Miscellanea Graecorum aliquot Seriptorum Carmina, 1722, p. 166): cf. Aratus 817 καὶ μᾶλλον μελανεῦσα, καὶ εἰ ῥηγνύατο μᾶλλον, where Maass quotes y 438 θεὰ κεχαροίατ᾽ ἰδοῦσα, as several MSS. have it for κεχάροιτο ἰδοῦσα; A 660 one MS. has βεβλήαται for βέβληται μὲν ὁ Τυδείδης. To keep the singular νόμος here would imply a strong view upon the unfamiliarity of the author with the epic dialect, but the principle of the preservation of linguistic anomalies presented by MSS. is one to which I incline. Cf. κατενήνοθεν with plural, Dem. 279, é in the plural Aphr. 267. Whether νόμος should be kept, or altered with Barnes into νομός, may be doubted. Βάλλειν νόμον is in any case an unusual phrase; βάλλειν must be taken, I suppose, in the sense of ‘lay, found, and in this sense may suit better with νόμος ‘custom’ or ‘strain’ than with νομὸς ‘range’ or ‘course.’ Also some weight perhaps should be given to the unvarying accentuation of the MSS. Hes. Theog. 66 μέλποντας πάντων Te νόμους, one MS. has νομούς. The conjectures πεπλήχαται (Matthiae), μεμέληται ἀοιδῆς (Hermann), νόμοι μεμβλήατ᾽ ἀοιδῆς (Nitzsch) do not assist.

26. πρὸς κύνθος ὄρος. On the united authority of the MSS. and of Steph. Byz. (sv. παρ᾽ ᾿Αντιμάχῳ ἐν πρώτῃ Θηβαΐδος. ὁ οἰκήτωρ κύνθιος" Kai θηλυκῶς καὶ οὐδετέρως) who can hardly refer to any passage but ours, I retain the neuter, notwithstanding the gen. Κύνθου v. 141. Barnesis the last

editor, D’Arnaud, quoted by Ilgen, the last critic, who has not departed from the MSS. |

29 sg. With Hermann and Baumeister it must be felt that the connection of the enumeration of places, vv. 30-44, is uncertain. If, as is usually the case, we print a comma at the end of 29, the places are introduced as those over which Apollo rules; but when we get to the end of the list we find they are regions over which Leto wandered. Unless we are to suppose that Apollo’s dominion coincided with the spgts through which his mother when big with him wandered, either a sign of interrogation, as Gemoll, or a full stop, as in the Oxford text, must be put after 29; the slight abruptness finds many parallels in the Hymns.

32. αἰγαί τ᾽ εἰρεσίαι te. Πειρεσίαι Ruhnken, It is admitted that no connection is known between Piresiae and Apollo, and as we have Iresiae standing in the texts of Livy xxxii. 13 it seems safe to leave Ecpeodau here. It is true that the Livian Iresiae and Piresiae must have been in the same neighbourhood, and Leake (Northern Greece, iv. 493) wished to simplify the matter by abolishing Iresiae. But is it even certain that our Iresiae is the

THE TEXT OF THE HOMERIC HYMNS: IV. 243

same as the Livian? For the name cf. the deme Εἰρασίδαι. Two other unknown names preserved in this catalogue are Aleayén and Αὐτοκάνη. It is a pity that Strabo did not extend his studies on B to this document.

35. αὐτοκάνης ὄρος αἰπύ. Αὐτοκάνη is not found; it does not follow however that we need the conjectures ἀντικάνη, ἀκροκάνη, αἰγοκάνη, which are equally non-existent. Κάνη or Κάναι is the name of a considerable mass of mountain opposite the south point of Lesbos, mentioned often by Strabo in his account of Asia Minor and described p. 615. The name applied to a town also, and (according to Stephanus s. v.) to a lake ; the district in general was called ἡ Καναία. Hence (at Ilgen’s suggestion) I take it that αὐτοκάνη may mean the centre of the geographical name κάνη, ‘Heart of Κάνη, i.e. the original peak of which Strabo says αὐτὸ xa’ αὑτὸ ἱκανῶς συνέσταλται, προσνεύει δὲ ἐπὶ τὸ Αἰγαῖον πέλαγος, although no parallel use of αὐτο- seems to be preserved. Αἰσαγέη v. 40 is still unidentified.

46. εἴ τίς cou γαιέων υἱεῖ θέλοι οἰκία θέσθαι. Oi which is generally read, is now found to be the emendation of H, and is made probable by ἑκηβόλον in 45. At the same time Apollo is addressed immediately before this parenthesis begins, v. 25, as σε, and afterwards v. 120, and therefore σοι may not be impossible here.

53. ἄλλος δ᾽ οὔτις σεῖό ποθ᾽ ἅψεται οὐδέ ce λίσσει. Mr. T. L. Agar (Classical Review, Nov. 1896,) has removed the scales from our eyes, and with the help of the unknown writer of S, seen οὐδέ σε λήσει in the end of the line. Thus Ernesti’s τίσει, Kirchhoff’s ἐσελάσσει and my ἐσδύσει retire into their proper limbo. Τίσει had no graphical possibility, and it is singular that it should have occupied the field for a century; Kirchhoff’s contribution is curiously inappropriate to an island like Ithaca οὐχ ἱππήλατος (or Zacyn- thus, of which Simonides fr. 15 ἱπποτροφία yap ob Ζακύνθῳ): my own effort rested on the graphical support given by p 276 δύσεο δὲ μνηστῆρας, λίσσεο ‘J marg.,” E 811 δέδυκεν, λέλυκεν ‘L. For the effects produced by the simplest case of itacism cf. Hes. Opp. 2, δεῦτε δὴ ἐννέπετε, δεῦτε δί᾽ ἐννέπετε.

79. ἀλλ᾽ εἴ μοι τλαίης γε θεὰ μέγαν ὅρκον ὀμόσσαι ἐνθάδε μιν πρῶτον τεύξειν περικαλλέα νηόν ἔμμεναι ἀνθρώπων χρηστήριον, αὐτὰρ ἔπειτα πάντας ἐπ’ ἀνθρώπους ἐπειὴ πολυώνυμος ἔσται.

How Gemoll can say ‘der Sinn iiasst nichts zu wiinschen tibrig’ passes comprehension. Leto was not to include in her oath (nor does she actually 84 sq.) that Apollo should proceed to other men after building a temple at Delos, she was to engage that he should build such a temple at Delos; after which says Delos with a sigh, let him continue his favours, ἐπειὴ πολυώ- νυμος ἔσται. No possible compression can get this into the passage: supply rather, with Hermann, such a verse as τευξάσθω νηούς τε καὶ ἄλσεα δεν- δρήεντα, which fell out from its identity with 76. Cf. 35-40, 371-4, 505-8, where similar endings have had this effect at four lines distance. The phrase

244 THE TEXT OF THE HOMERIC HYMNS: IV.

is repeated, vv. 148, 221, 245, so that one more instance need not give offence. Pomtow’s objections (NV. Jahrb. f. Phil., 1887, p. 176, sg.) to Her- mann’s notion of the contents of the lacuna seem unfounded, and his sugges- tion that 81 is not genuine gratuitous; the same epithet applies to Pepp- miiller’s bracketing of 81, 82 (J. ὁ. p. 198).

103. μέγαν ὅρμον | χρυσείοισι λίνοισιν ἐερμένον, ἐννεάπηχυν. Whether any work of art resembling this necklace ever existed in rerum natura at any period of Greek art is for archaeologists to settle: the commentator need have no difficulty in translating the words as they stand: ‘a great necklace, nine cubits long, set with golden threads.. The Greek will bear the inter- pretation either of gold wire, or of tassels of thread or string gilded or strung with gold thread; the latter seems the more likely, if we consider some of the objects to which χρύσεος is applied in Homer: thus @ 42 horses’ manes, ® 44, N 26 a whip, E 727 reins, T 382, X 383 plumes of Achilles’ helmet. In all these cases material cannot be implied, but decoration. We get closer to the context in the Scutwm Herculis 224, ἀμφὶ δὲ μιν κίβισις θέε θαῦμα ἰδέσθαι | apyupén: θύσανοι δὲ κατῃωρεῦντο φαεινοὶ | χρύσειοι, of the golden tufts or tassels round Perseus’ bag. Further to anyone who objected that no such objects are known from excavations (if indeed tassels and such like can survive) I would answer that this necklace and most of the other objects I have quoted are the work of Gods, and may therefore possess unusual refine- ments of art. When we find in some Mycenae a necklace nine yards long, we may expect to find one ‘set with golden threads, a wonder to behold.’ The alterations besides being uncalled for, are all more or less improbable. Barnes’ χρύσεον ἠλέκτροισιν ἐερμένον is graphically impossible (and Gemoll’s palaeographical observations merely illusory) ; λέθοισιν (Matthiae and Pepp- miiller) is commonplace and can never have been corrupted into the rarer word λίνοισιν ; γλήνεσσιν like most of Bergk’s conjectures is brilliant but scatterbrained.

᾿Ἑερμένον (Barnes) for ἐεργμένον is a very proper correction on the analogy of Εἰ 89 σ 296.

Matthiae, in his Animadversions and edition, Franke, and Burckhardt in a dissertation quoted by Gemoll, keep the reading λίνοισιν.

116. τὴν τότε δὴ τόκος εἷλε should be restored from Ilgen’s most needless alteration δὴ τότε τὴν. The amount of emphasis conveyed by the position of τὴν 18 quite in place.

133. ὡς εἰπών ἐβίβασκεν ἀπὸ χθονὸς εὐρυοδείης. ᾿Επὶ Matthiae, which of course gives an easy sense. I am inclined to think however that ἀπὸ may without violence be given a pregnant sense, ‘he began to walk [getting up] from the ground,’ where up to this time he had been lying. E13 τὼ μὲν ἀφ᾽ ἵπποιιν, ὁ δ᾽ ἀπὸ χθονὸς ὥρνυτο πεζός is somewhat parallel, in so far as it shows how ἀπὸ may be used out of its strictly literal sense. So Hermes as soon as he was born, οὐκέτι δηρὸν ἔκειτο---ἀλλ᾽ Sy’ ἀναΐξας κιτλ. (Herm, 21, 22).

THE TEXT OF THE HOMERIC HYMNS: IV. 245

"142. ἄλλοτε δ᾽ αὖ νήσους τε καὶ ἀνέρας ἠχάσκαζες. Ilgen, Peppmiiller and Tyrrell seem right in denying that ἠλασκάζειν can take a simple accusative ; av therefore must be altered to dv; cf. B 198 ὃν δ᾽ ad δήμον, ὃν δ᾽ ἂν Eust. Here the ν fell out before νήσους, and v was added to make metre. It is unnecessary to alter ἀνέρας, as has been proposed: νήσους τε καὶ ἀνέρας is a Hendiadys for the ‘inhabited islands, in contrast to Delos, For a similar omission of ἂν cf. Dem. 7.

152. of τότ᾽ ἐπάντια σεῖο τ᾽ idoves ἀθρόοι elev etc. codd. Martin’s brilliant ὃς for οἱ is made necessary by ἴδοιτο and τέρψαιτο of 153. Οἱ no doubt came in after ἐπαντιάσει᾽ had decomposed, in order to give an apparent subject to εἶεν, the only verb then left. 356 ὃς τῇγ᾽ ἀντιάσειε. The corrup- tion has the marks of being very early. Conversely Herod. i. 124 ἀντῆσαι for ἀντία σεῦ.

160. The apodosis starts here, as Gemoll rightly says. The πάντων ἀνθρώπων φωναί are of course the various dialects, which in strongly decentralized countries assume to their speakers the dignity of languages. The case of Gorgo and Praxinoa is in point. In modern Italy recitations in different dialects may now and then be heard. Κρεμβαλιαστύς or βαμβα- λιαστύς is, as Gemoll sensibly decides, the accompaniment. The /forestieri at this great pilgrim centre hear their own speech and their own music. Pepp- miiller’s alteration of αὐτὴ ἑκάστη in 163 misses the point sadly. Matthiae in his Animadversions appears to realise the scene, but in his edition, with the inexplicable violence to which the subjective critic is chronically liable, cuts out all three lines, the most graphic and racy in the Hymn. A study of the arrangements at Rome or Einsiedeln would convince commentators that there is nothing ‘inept’ in making the pious feel at home.

166 sq. ἐμεῖο δὲ καὶ μετόπισθε fs yee / 4 > / > , μνήσασθ᾽ ὁππότε κέν τις ἐπιχθονίων ἀνθρώπων ἐνθάδ᾽ ἀνείρηται ξεῖνος ταλαπείριος ἐλθών' Φ “ / δ᾽ Ν > \ “ὃ > ὃ la] ἣν ὦ κοῦραι, τίς δ᾽ ὕμμιν ἀνὴρ ἥδιστος ἀοιδῶν κ.τ(λ.

I quote here, since I do not find it in any commentary, the remarkable fragment of Hesiod. No. 227 :—

ἐν Δήλῳ τότε πρῶτον ἐγὼ καὶ Ὅμηρος ἀοιδοί μέλπομεν, ἐν νεαροῖς ὕμνοις ῥάψαντες ἀοιδήν, Φοῖβον ᾿Απόλλωνα, χρυσάορον, ὃν τέκε Λητώ.

The coincidence of subject and place is so marked that one can hardly imagine that the fragment and the Hymn are without connection with one another.

171. ὑμεῖς δ᾽ εὖ μάλα πᾶσαι ὑποκρίνασθε ἀφήμως. ἀφήμως Thucydidis codd. antiquiores, εὐφήμως deteriores, ἀφ᾽ ἡμέων Mx: ἡμῶν Aristides ii. p. 589: ὑμέων, ὑμῶν p. I should like to withdraw my note on this line, vol. xv. p. 310. I think now that ἀφήμως, the reading of the older

246 THE TEXT OF THE HOMERIC HYMNS: IV.

MSS. of Thucydides is literally correct, and that we do not even require the rough breathing of Bergk’s ἁφήμως, Griech. Literaturgeschichte i. p, 750 π. ‘einstimmig (ὑποκρίνασθαι ἀφήμως oder besser d¢7juws).’ Compound words consisting of a primitive+ in the sense of the primitive are not unfrequent, see Kiihner-Blass ὃ 339 δ, e.g. ἄπεδος ‘flat’ from πέδον, Herod. i. 110, ἄβρομοι αὐίαχοι ‘noisy’ from βρόμος, ἰαχή N 41, and especially I 404 οὐδ᾽ ὅσα λάινος οὐδὸς ἀφήτορος ἐντὸς ἐέργει, Where the usual derivation was from ἀφίημι, but Aristarchus glossed the word by ὁμοφήτορος, and this derivation is approved of by Prellwitz. There is, I think, no need to refer to ἅμα either in origin or in sense, and if ἀφήτωρ means ‘the speaker, ἀφήμους will mean ‘clearly’ or ‘loudly, not ‘unanimously,’ as the scholiasts on Thucydides render, acknowledging the word but misinterpreting it, ἡσύχα, ἀθρόως. I read therefore with Bergk ὑποκρίνασθαι ἀφήμως, recommending myself to the mercy of the etymologists. ;

Assuming ἀφήμως, the reading of the oldest MSS., to be the original, the corruptions are easily accounted for: on the one hand, ἀφήμως retained as a single word fell into εὐφήμως by the most usual process of graphical corruption ; on the other, ἀφημως, the preposition separating, naturally gave rise to the conjectures ἀφ᾽ ἡμέων or ἡμῶν. In the editions, ἀφ᾽ ἡμέων starting as the x reading from Demetrius Chalcondyles, lasted down to Ruhnken, and was translated by Barnes ‘responderitis a nobis. Ruhnken took from the younger MSS. of Thucydides the reading εὐφήμως, palpably the worst of any. a may explain ev, but not ev a. Normann, in his edition of two speeches of Aristides, Upsala, 1687, and after him Bergk, defended ἀφήμως.

178. Keep the present ἀριστεύουσιν ; ‘whose songs have the greatest fame after,’ 6. after he has sung them once. He has fame within his lifetime, his songs are more demanded than those of others (e.g. the Hesiod of the fragment). Μετόπισθε 166 is used of time during the poet's life, and πωλεῖται, τέρπεσθε, οἰκεῖ are all present. ᾿Αριστεύσουσιν (Barnes) would invest the Delian maidens with prophesy in addition to their other accomplishments.

The criticisms of Ruhnken, Ilgen, and Matthiae upon the excellent word ἀριστεύουσιν are typical of that age (‘Deinde quale istud est, ἀριστεύουσιν ἀοιδαί. Tua te lingua prodit,o bone. Digna haec sunt Nonni aetate, non Homeri.’). Hermann vindicated the word. The lines which, even as late as Bergk, have been thought unworthy of the poet, are surely original and most characteristic of the professional bard.

185. ἄμβροτα εἵματ᾽ ἔχων τεθυωδέα. There would be no objection to θυώδεα, on which Pierson’s εὐωδέα is no improvement (in its favour may be brought ε 264 θυώδεα, edwdea Plut. de vitando an, al. 831 ἢ), Ap. Rhod. iv. 1155 éavods εὐώδεας), but that it involves the awkwardness of τε fourth in the sentence. Barnes’ usually accepted τεθυωμένα may therefore stand,

M and we must suppose that the participle, written as usual τεθυω with an

THE TEXT OF THE HOMERIC HYMNS: IV. 247 indeterminate scrawl to indicate omission, was at an early period misdeciphered Δ τεθυω, i.e. τε θυωδέα.

204-6. Peppmiiller’s (Philologus, 1894, p. 256) discussion of these lines, which lands him in either the alteration of θυμὸν into υἱὸν (!) or the trans- position of 206 before 205, is a striking instance of the results of overfamiliarity with a document in a foreign language.

207 sq. This passage seems as far off as ever from salvation. The only opinion I can express is that as all the lines with the exception of 211 make . a bare sense as they stand the text should be left untampered with. The various suggestions that have been made can be refuted one by one, even where they do not mutually destroy each other. I will merely notice the frivolity of Gemoll’s ἀναμνήσω or ἐπιμνήσω for ἐνὶ μνηστῇσιν; no one will believe that the omission of γένος (211) in yis a proof that it was interpolated into the other MSS. Schneidewin’s ἢ ὡς φόρβαντα for ἡ ἅμα φόρβαντι is now given up, and my own assertion (vol. xv. p. 276) that τρίοπος 218 is genitive is as uncertain. The passage waits, and may do so to eternity, for an interpreter,

218. λέκτον +’ ἠμαθόεντα. Baumeister’s Λάκμον and the earlier conjecture Aedxoy are unconvincing and therefore to be rejected. There may have been a Λέκτος in Europe as there was in Asia, and the name lost, ep. Aicayén and the other names p. 2. , The only geographical ‘corrections that seem indispensable are ’Evijjvas in this line and Ἕλος τ᾽ ἔφαλον v. 410, both due to Matthiae.

227. οὐδ dpa πω τότε γ᾽ ἦσαν ἀταρπιτοὶ οὐδὲ κέλευθοι Θήβης ἂμ πεδίον πυρηφόρον ἀλλ᾽ ἔχεν ὕλην.

TAxn for ὕλην is Barnes’ best conjecture. The accusative comes from the tendency of scribes to be influenced by the nearest apparent construction ; similar cases are N 104 οὐδ᾽ ἔπι χάρμη, where the suggestion of the preposi- tion has been irresistible to ‘H’ Ven.,,, ,, M,, which give χαρμήν and to L, M, Vat.,, Ven.A BC, which give χάρμῃ; $ 177 τρὶς δὲ μεθῆκε Bin; many MSS. βίη and Bekker needlessly Bins. A 174 σέο δ᾽ ὀστέα ricer ἄρουρα ; ἄρουραν B. M. Pap. 136.

230 sq. The custom at Onchestus. My rendering of this passage is as follows: ‘there the new-tamed horse breathes again, tired though he be with dragging a fair car, and the driver good though he be leaps to ground from the chariot and walks the road; meanwhile the horses rattle empty cars and have lost their lords. Now if the chariot be broke in the planted grove, they groom their horses, but the chariot they lean up [against a wall or the temple] and leave there, for so is it the custom from the beginning; they make their prayer to the king, but the chariot is the god’s portion to keep,’ I think that this is intelligible in itself, and it involves only Cobet’s alteration of dynow into ἀγῇσιν. The current interpretation of the passage down to

248 THE TEXT OF THE HOMERIC HYMNS: IV.

Baumeister followed an account given to A. Matthiae by Bottiger. Gemoll exhibited disbelief in this account, and Peppmiiller in an interesting note has reviewed the whole situation (Philologus, 1894, pp. 257-260). He rightly remarks that the interpretations of Béttiger and of Preller are incorrect in several vital points, and that the passages quoted from Pausanias do not refer to the local custom in question. They are however none the less extremely interesting and pertinent, as illustrating the terrifying effect upon horses attributed to Poseidon in particular places. So in the hippodrome at Olympia (p. 504) there was on one side, in a sort of cutting, κατὰ τὴν διέξοδον τὴν διὰ τοῦ χώματος, ‘ the fear of horses’ 6 ταράξιππος.ς At this point τοὺς ἵππους φόβος τε αὐτίκα ἰσχυρὸς ἀπ᾽ οὐδεμιᾶς προφάσεως φανερᾶς καὶ ἀπὸ τοῦ φόβου λαμβάνει ταραχή; the chariots as a rule are broken, and the drivers hurt.: There were other ταράξιπποι in Greece, at the Isthmus and at Nemea; and a certain suspicion attached to the hippodrome of Apollo at Delphi (p. 893). Pausanias believes the divinity at the bottom of these various manifestations to be Ποσειδῶν Ἵππιος : a celebrated case of his action is that of Hippolytus.

In our passage Peppmiiller objects to the slight alteration ἀγῇσιν, although it has the undeniable analogy of Pausanias’ ta τε δὴ ἅρματα καταγνύουσιν ὡς ἐπίπαν, and reads app’ ἀγάγωσιν. This is open to more than one objection; the translation must be ‘if they bring the chariot into the grove’; but ἐν ἄλσει δενδρήεντε cannot be used to express motion after dryeew—seeing which Peppmiiller would connect ἐν ἄλσει Sevdpyjevte with the next line; this however is forbidden by μέν, which plainly marks the beginning of the apodosis. Secondly, the sense of ἀγάγωσι is very fiat. The young horse is left to himself, and the question is how he will behave ; will he get safe past the temple, or will the influence of ταράξιππος be too strong and will he bolt and smash the chariot among the sacred trees? ‘Bring the chariot to the grove’ could only have a meaning if we suppose the horse liable to turn tail. Also the horse and his driver were already ἐν ἄλσει; the road doubtless ran past the temple, and the driver will have got down where the precinct began. There is therefore no question of the horse ‘finding his way to the goal,’ and becoming ἄφετος. Far from that it is implied that his master in any case kept him.

I conceive the statement not to refer to any special festival or ἀγών, but to have been the ordinary rule of the road in these parts. The God of Horses was offended at wheeled traffic that passed his home; but he gave travellers so much grace that their cattle were allowed a chance, without guidance. If the horse withstood his influence, well; if he bolted and wrecked the chariot, the traveller compounded by leaving the broken carriage—of which it is to be presumed the priests undertook the repair and eventual sale at second-hand. This very interesting use died out with the decay of Onchestus, of which in Pausanias’ time (p. 76) there were left the ruins of the town, the temple and the grove: Strabo (p. 411) saw the temple, but thought the poets had invented the grove. Lastly no particular stress is to be laid upon νεοδμής, as if only young horses underwent the ordeal. Rather it was only in the case of

THE TEXT OF THE HOMERIC HYMNS: IV. 249

a veoduns πῶλος that his owner felt the anxiety; old hacks’ nerves were beyond the reach even of an Earthshaker, Leake, Northern (reece, 11. p. 213 describes the site of Onchestus, on a low ridge.

’ i ᾽ 260. ἠμὲν ὅσοι Ἰ]ελοπόννησον πίειραν ἔχουσιν, »Ο» “« > ’ \ > / \ / nd ὅσοι Εὐρώπην τε καὶ ἀμφιρύτους κατὰ νήσους.

That the name Εὐρώπη, like ᾿Ασέα and ᾿Ελλάς, extended its original connotation, is suggested by the ancient authorities (Steph. Byz. and the Etym. Magnum, who point to Macedonia) and by modern geographers (e.g. Bunbury, History of Ancient Geography, i. p. 89). To gut the document, and substitute the impossible ἤπειρον written by Reiz on the margin of his edition, is unworthy of a responsible editor.

299. κτιστοῖσιν λάεσσι. It seems impossible to apply κτέζειν to the materials out of which the temple is made—‘fabricatis lapidibus’ as Barnes translates. Ernesti’s ἕεστοῖσιν is too far from the letters of κτιστοῖσι, and the other epic epithet ῥυτοῖσι is farther still. Perhaps τυκτοῖσι ‘ wrought,’ comparing ὃ 627, p 169 206, ν 306 ν.]., τυκτὰν μάρμαρον Theocr. xxii. 210. The stages of the corruption are TYKTOICIN, TIKTOICIN by itacism, TIICTOICIN(K = IC), (IK)TICTOICIN to make a word.

331. ὡς εἰποῦσ᾽ ἀπόνοσφι θεῶν κίε χωομένη περ.

Barnes’ κῆρ for mep has been accepted from his time till Gemoll’s, but the non-adversative force of περ, though rare, can hardly be denied in these places :-—

a 315. μή p’ ἔτι viv κατέρυκε λιλαιόμενόν περ ὁδοῖο' p 12. ἐμὲ δ᾽ οὔπως ἐστιν ἅπαντας ἀνθρώπους ἀνέχεσθαι ἔχοντά περ ἄλγεα θυμῷ" ib. 47. μῆτερ ἐμὴ μή μοι γόον ὄρνυθι μηδέ μοι ἧτορ ἐν στήθεσσιν ὄρινε φυγόντί περ αἰπὺν ὄλεθρον" and it may well be absent from Γ 200 οὗτος δ᾽ αὖ Λαερτιάδης, πολύμητις ᾿Οδυσσεύς ὃς τράφη ἐν δήμῳ ᾿Ιθάκης κραναῆς περ ἐούσης. The amount of emphasis in περ in these instances very fairly suits our line: ‘ she left the god, angry as she was.’

340. ὡὼς dpa φωνήσασ᾽ ἵμασε χθόνα χειρὶ παχείῃ.

Mr. Platt in a recent number of the Journal of Philology prefers ἔλασε, on the precedent of 333, to ἵμασε. Certainly graphically the words are not far off as in minuscule μ᾿ produces, not unfrequently, AA: N 372 ἐλλάσιν οὐκ ἐθέλοντα six or seven MSS. give ἔμασιν or ἵμασιν for ἐλλάσιν. However ἵμασε is forcible, of Hera’s rage, and is supported by I 568 πολλὰ δὲ καὶ γαῖαν πολυφόρβην χερσὶν ἀλοία, an exact parallel: and of Zeus scourging the earth

250 THE TEXT OF THE HOMERIC HYMNS: IV.

or his enemies B 782 ὅτε τ᾽ ἀμφὶ τυφωέι γαῖαν ἱμάσσῃ, Hes. Theog. 857 πληγῆσιν ἱμάσσας.

361. λεῖπε δὲ θυμὸν φοινὸν ἀποπνείουσ᾽"

The incredibly bad substitutions for this fine phrase repay study. Gemoll, who collects them, justly decides that the text is sound. The unusual λεῖπε δὲ θυμὸν is defended by the passage Pind. Pyth. 111, 180 first brought by Matthiac, and by the usual tendency to regard these human physical phenomena (death, etc.) from two alternate points of view, as the man becomes object or subject.

380. προρέειν καλλίρροον ὕδωρ. Φ 366 προρέειν and προχέειν are variants, and we have προχέειν alone ® 219 and here 241, but the cognate accusative after προρέειν seems made out, and is certainly the more difficult construction,

382. 7 καὶ ἐπὶ ῥίον ὦὧσεν ἄναξ ἑκάεργος ᾿Απόλλων meTpains προχυτῇσιν, ἀπέκρυψεν δὲ ῥέεθρα.

This fall of rocks has buried more than one critic; Ruhnken altered 3838 to metpains προχοῇσιν, the latter word not a very violent change (Ion of Chios fr. ii. 8 προχύταισιν ἐν ἀργυρέοις, as quoted by Ath. 463 B προχοαῖσιν) but hopeless as to sense; a Dutchman may be excused unfamiliarity with mountain phenomena, but Gemoll is no better, who thinks that the change of ῥίον into ῥόον makes all straight.

The dative, cause of all this mischief, is not governed by ἐπὶ---ὦὧσεν, but is of circumstance: ‘he pushed a rock over, with a shower of stones,’ pro- fusis lapidibus. Another ignored dat. of circumstance is at Hes. Scuwt. 288 οἵ γε μὲν ἤμων | αἰχμῇς ὀξείῃσι κορυνιόωντα πέτηλα : commentators, ancient and modern have had doubts about reaping grain with spears, and Paley brings in ἅρπῃς to do duty: the ears, however, ‘bristle with sharp spears, 1.6. their stalks, as Burns has it.

Apollo’s Bergsturz followed the usual laws of such things ; first the heavy crag detached itself, then a shower of stones and earth followed, and effectually filled the river. Travellers (see Bursian, Geog. von Griechenland, 1. p. 284) have identified the spring and the mountain behind it, wooded below but ending in sharp rocks, but they do not say if any appearance suggests a catastrophe—another local legend lost except for this Hymn. Other descrip- tions of falling stones and the damage done by them may be read in Scwt. 374 sq. and 437 80.

408. κραιπνὸς δὲ Νότος κατόπισθεν ἔγειρε | νῆα Oonv. Ruhnken’s érrevye is usually accepted for ἔγειρε, but passages like the following suggest that the text may stand: Herod. vii. 49 ἐγειρομένου χειμῶνος, Ap. Rhod. 1. 1159 ἐγρομένοιο σάλου ζαχρηέσιν αὔραις, iii. 295 of flame, Anth. Pal. vi. 21 πρασιὴν διψεῦσαν ἐγείρειν. Quintus ix. 271 of a wave, 6 τ᾽ ἐξ ἀνέμοιο

THE TEXT OF THE HOMERIC HYMNS: IV. 25)

διεγρομένου φορέηται. ‘he ship had been running under canvas; now the wind strengthened and ‘ woke’ it into a quicker course.

417. εἰς oldu’ ἅλιον πολυΐχθυον ἀμφὶς dpovoet.

Pierson’s αὖθις is not as violent a conjecture as might appear at first sight (αμφις, agus, αὐφις, αυθις) but I keep ἀμφὶς in the sense of ‘apart, away’ or more shortly ‘out.’ Examples with a genitive are given in the Lex. Hom., p. 108b; it is but a step to the absolute use, for which the nearest parallels in this sense are w 218 ἠέ κεν ἀγνοιῇσι πολὺν χρόνον ἀμφὶς ἐόντα, y 486 0184 σεῖον ζυγὸν ἀμφὶς ἔχοντες, Ap. Rhod. iii. 1069 μνώεο---οὔνομα Μηδείης" ὡς δ᾽ adr’ ἐγὼ ἀμφὶς ἐόντος | μνήσομαι, oracle ap. Herod. i. 85 τὸ δέ σοι πολὺ λώιον ἀμφὶς | ἔμμεναι.

422. In handling this geographical passage, Gemoll, with much judg- ment, decides that in default of further knowledge the text is to be left unaltered. The passage partly coincides with B 591 sg. and 0 294 sg. Strabo 348 sq. pointedly ignores the Hymn, though he quotes v. 425 (but with καλλιρέεθρον and p. 447 πετρήεσσαν instead of καὶ παρὰ Δύμην) as from the Odyssey. In view of the abundant extra lines that papyrus is adding to the Homeric poems, it is perfectly probable that Strabo found this verse in his copies.

488. νῆα δ᾽ ἔπειτα θοὴν ἐπὶ (ἐπ᾿ M) ἠπείρου ἐρύσασθε.

Mr. Agar’s θοὴν av’ ἐπ᾽ ἠπείρου to save the hiatus is neat and corre-

sponding to the wording of 506.

491. Ilgen’s δ᾽ after πῦρ is unnecessary if we make a comma at θαλάσσης and take the two participles with 490; the conclusion then comes with natural weight at εὔχεσθαι δὴ ἔπειτα.

521. Pierson altered ἔμελλεν and τετιμένος into the plural—most need- lessly, for while both temple (479, 483) and priests (485) are to enjoy honour, the MSS. may be allowed to turn the scale here in favour of the former,

529. οὔτε τρυγηφόρος ἥδε γ᾽ ἐπήρατος οὔτ᾽ εὐλείμων is certainly harsh, but the construction (‘this land is not desirable as corn-producing nor as fair-pastured ’) of adjectives qualifying adjectives in amply covered by ν 246 αἰγίβοτος δ᾽ ἀγαθὴ καὶ BovBoros, where no other translation is possible but ‘it is good as goat-feeding and as ox-feeding. The conjectures, most of them incredible, are collected by Gemoll: Peppmiiller (/.c. p. 275) in an evil hour added aid γ᾽ for ἥδε γ᾽!

538. νηὸν δὲ προφύλαχθε, δέδεχθε δὲ φῦλ᾽ ἀνθρώπων, ἐνθάδ᾽ ἀγειρομένων καὶ ἐμὴν ἰθύν τε μάλιστα. ἠέ τι τηύσιον ἔπος ἔσσεται, ἠέ τι ἔργον. Various attempts have been made to complete the construction of 539. Ἰθύν seems too good a word to be given up: it is used tropically in Homer, Z 79, ὃ 434 πᾶσαν én’ ἰθύν and π᾿ 304 σύ τ᾽ ἐγώ τε γυναικῶν γνώομεν ἰθύν,

252 THE TEXT OF THE HOMERIC HYMNS: IV.

the ‘inclination, bent’ of the women. Here it is peculiarly appropriate to the will or guidance of the God, the straight path made plain through the oracles of Loxias. The expressions ἐθείῃσι δίκῃσι, δίκην ἰθύντατα εἴποι, δίκῃ δ᾽ ἴθυνε θέμιστας are analogous. The word should therefore be kept, and here I find myself in agreement with Peppmiiller. I cannot, however, accept his parenthesis (δέδεχθε δὲ φῦλα ἀνθρώπων, ἐνθάδ᾽ ἀγειρομένωνλ) : rather a lacuna must be made, to contain a verb to govern ἐθύν, a transition to the threat of 540, the construction of which as it stands is abrupt, and a singular to antecede σὺ---φύλαξαι of 544, I can think of nothing better than δείκνυσθε θνητοῖσι, σὺ δὲ φρεσὶ δέξο θέμιστα. Homoeoteleuton of some sort naturally is wanted.

HERMES,

Critical work on the Hymn to Hermes during the last ten years consists for the most part of the labours of one man, Arthur Ludwich. Articles by him are to be found in the Rheinisches Museum for 1888, ’89, and ’90, and the Neue Jahrbiicher fiir Philologie, 1886, ’87,’88, and ’89, and their results are collected in the extremely useful edition of the hymn, Regimontii, 1890. Homer owes more to Professor Ludwich than perhaps to anyone else of his generation, and it is well that this debt should be put on record at a moment when he has lately been the victim of a gratuitous impertinence, not, we may be glad to think, on the part of an Englishman.}

I have also to refer to notes by R. Peppmiiller, Newe Jahrb., 1887, pp. 201, 805; Herwerden, Rhein. Mus., 1888, p. 73 sg. The myth is well treated by Gemoll, and in Roscher’s exhaustive article in his Lewicon.

The Hymn is admittedly the most difficult of the collection: and this not so much on account of its subject, for the story was often treated in literature, e.g. by Alcaeus, and accounts corroborative in the main, though divergent in detail, remain in Apollodorus, Antoninus Liberalis, Ovid, and Pausanias, as from its language and style. The view that the hymn is late is generally abandoned: we have rather a specimen of early, half simple, half ironic, epos: the style, though admirable narrative on the whole, is in places apparently intentionally riddling and dark; absence of cognate literature for comparison has produced unusual corruption; the continuity of sense is broken in several places, and a large number of voces nihili remain to baffle the reader. The attem,‘s of the learned upon them have been more than usually unsuccessful. They belong, or seem to belong to the desperate category of difficulty, where either there is no corruption and it is our knowledge that is at fault, or the corruption is but a step removed from the tradition—a step which is beyond our skill to make. I hope closer study of these documents may make it plain that violent conjectures do not win acceptance, and that the right method is, either that of new interpretation of

1 I referto P. C. Molhuysen, DetribusHomeri 1896. Mr. Mulvany, Classical Review, June Odysseae codicibus antiquissimis. Lugd. Bat. 1897, has overrated this performance.

THE TEXT OF THE HOMERIC HYMNS: IV. 253

the existing word-forms, or of corrections that approach the type of Mr. Agar’s palmary οὐδέ σε λήσει for οὐδέ σε λίσσει, Apoll. 53.

Lacunas seem necessary at 91, 409, 415, 526, 568, and these expressions are either corrupted or still uncertain: ἀναπηλήσας 41, αὐτοτροπήσας 86, τὸ σὸν αὐτοῦ 93, περῆν 133, ἤ σε λαβόντα μεταξὺ 159, τιτύσκεαι 163, βουλεύων 167, the whole line 188, ἄγρης: elveréov τε 242, ὀφρύσι ῥιπτάζειν 279, 282 sq., εὐμυλίη 325, ὅδ᾽ ἐκτός 346, κραίνων 427 and elsewhere, μέμηλας 437, θυμὸν ἐπαίνει 457, ἡγεμονεύσω 461, σήματ᾽ ἐπεὶ 509.

With the higher criticism of the Hymn I have not to do, There is however, one peculiarity of the story which must strike every attentive reader, the variations in the different accounts of Hermes’ journey with the oxen from Pieria to Pylos and Apollo’s search after him, The difficulty is real, but I entirely agree with Franke, Gemoll (p. 187 and note on 211), and Ludwich ‘ Angebliche Widerspruche im hom. Hermeshymnus’ Neue Jahrb., 1887, p. 321 sg. that the inconsistency is original and native to the poem, It is satisfactory to find literary criticism at length becoming historical and taking account of conditions and standards other than those of its own time. Tyrrell, 1.0. p. 42 sg. has fallen into a misapprehension with regard to Hermes’ descent from Pieria which it is unnecessary to examine in detail. There was, of course, only one journey.

The integrity of the document apart, the geographical outlook of the writer is curious; he is very vague as to continental Greece, and evidently thought Pieria was connected with Onchestus by a sandy road along the sea. This vagueness contrasts with the accuracy of the author of the Hymn to Apollo, who gets Apollo’s journey from Euboea to Delphi marked out with great correctness. It would be an easy guess that the writer of the Hymn to Hermes was a Peloponnesian; the reference to the skin surviving outside the cave at Pylos (v. 125) implies connection with the Alpheus country. Bergk (Griech. Lit., i. p. 766 n.), upon the same evidence thinks the author was an Ionian; so differently do things present themselves to different

people.

6. ἄντρον ἔσω ναίουσα παλίσκιον. With Ludwich I restore this, the MS. reading. "ἄντρου or ἄντρῳ is needless; xviii. 6 ἄντρῳ ναιετάουσα παλισκίῳ has no binding force, and ἔσω is absolute and parenthetic; ‘inhabiting the cave, within’; cf. 49 ap εἴσω κίε δῶμα, H 13 δῦναι δόμον ἀίδος εἴσω, @ 549 κνίσην δ᾽ ἐκ πεδίου ἄνεμοι φέρον οὐρανὸν εἴσω, and other passages; Ψ 23 νέεσθαι | αὗτις ἔσω μέγαρον, where the variant μεγάρων arises from the same misapprehension, Theocr. Hpigr., 3. 5, ἄντρον ἔσω ° στείχοντες. The use being parenthetic can accommodate itself as well to rest as to motion: Ilgen brings some exx. of the former sense.

15. πυληδόκον. Certainly not ‘ porter, as Ebeling, Lez. Hom., and the older commentaries, since Hermes never appears in so sedentary a function ; but =‘thief’ as Matthiae suggests and Baumeister decides; ef. ὁδοιδόκος and (in a different sense) πολεμαδόκος.

H.S.—VOL. XVII. T

254 THE TEXT OF THE HOMERIC HYMNS: IV.

32. πόθεν τόδε καλὸν ἄθυρμα | αἰόλον ὄστρακον ἐσσὶ χέλυς ὄρεσι ζώουσα. Tyrrell’s brilliant and humorous ἕσσο must command universal acceptance. When τόδε καλὸν ἄθυρμα, αἰόλον ὄστρακον become accusatives, the difficulties of construction and punctuation, which had endlessly embarrassed the commentators, vanish.

41. ἔνθ᾽ ἀναπηλήσας γλυφάνῳ πολιοῖο σιδήρου αἰῶν᾽ ἐξετόρησεν ὀρεσκῴοιο χελώνης

In »

Αἰῶν᾽ literally ‘marrow’ is by a natural semi-comic metaphor ‘ flesh,’ which is to the tortoise’s shell what marrow is to the spine: in the sense of ‘life’ αἰῶν᾽ could not in this sort of poetry be joined with a concrete word like ἐξετόρησεν. If then v. 42 expresses the process of clearing the flesh out of the shell, we should expect v. 41 to contain the act of killing ; ἀναπηλήσας however has resisted all the interpretation and conjectures of the learned. The latter, many of them evidently impossible, may be seen ap. Ludwich: Barnes’ ἀναπηδήσας is not bad, if the action be unnecessarily violent: Hermann’s ἀναπιλήσας is the best and perhaps may satisfy. ᾿Αναπιλεῖν must mean to squeeze, and denotes I suppose choking, a possible method no doubt of executing the job, though now-a-days we hear more of cutting off the heads of tortoises. Πιλεῖν occurs in epos Ap. Rhod. iv. 678 πιληθεῖσα.

44, ἀνέρος ὅντε θαμιναὶ ἐπιστρωφῶσι μέριμναι. Ruhnken kept @apivai, quoting μεσημβρινός, ὀπωρινός, ὀρθινός as instances of ambiguous quantity, Blass-Kiihner, Awsfiihrl. Gramm. § '75, 9, Lobeck Pathol. p. 200 sg. Choero- boscus in Cramer An. Oz. ii. p. 180 quoted by Lobeck J. ὁ. p. 188 recognises a form θαμεινός, and I can bring the derivative of ὑδατ- which in Hipp. 4ér. c. 15, 19 is written ὑδατειναί, and has ¢ necessarily long in Matro 79. This evidence would incline me, rather than accept Barnes’ θαμειαί (the loss of which I cannot account for), to leave θαμῖναί.

48. πειρήνας διὰ νῶτα διὰ ῥινοῖο χελώνης. Whether we can have πειραίνω in the sense οὗ πείρω is a question I would not decide; in any case Matthiae’s τετρήνας may be justified by Herod. ii. 11 συντετραίνοντας v. 1. συμπεραίνον- ras. Of the words that follow, διὰ puvoto are unanimously considered corrupt: κραταιρίνοιο, λιθορίνοιο, ταλαρίνοιο are proposed, but pace Mr. Sikes (Classical Review, 1894, April) and Mr. Tyrrell, they do not convince. Why should these elegant adjectives have broken up into διὰ ῥινοῖοῖ To my mind the second διὰ has driven out another preposition that originally occupied the . place of the first; this phenomenon,—where two prepositions occur in the same line and one expels the other—may be seen K 54 ῥίμφα θέων παρὰ νῆας" ἐγὼ δ᾽ ἐπὶ Νέστορα δῖον: for mapa ‘ACHS’ read ἐπὶ; 141 1/60 οὕτω κατὰ νῆας ἀνὰ στρατὸν οἷοι ἀλᾶσθε, κατὰ νῆα κατὰ στρατὸν ‘L’; 298 ἂμ φόνον, ἂν νέκυας διά 7 ἔντεα καὶ μέλαν αἷμα, ἀνά τ’ ἔντεα Eust. Κατὰ and διὰ are exchanged simply Ν 388, σ 341, and for the sequence κατὰ---διὰ οἴ, ἡ 40 ἐρχόμενον κατὰ ἄστυ διὰ σφέας, Ap. Rho.d, iv. 1002, κατὰ στόμα καὶ διὰ πέτρας. Here, of the two, διὰ with ῥινοῖο is clearly the more appropriate

THE TEXT OF THE HOMERIC HYMNS: TV. 255

(Hes. Opp. 515 καί τε διὰ pivod βοὸς ἔρχεται): I would therefore write meipnvas κατὰ νῶτα διὰ ῥινοῖο χελώνης ‘at, or on, the back, through the

shell.’

58, dv πάρος ὡρίξεσκον ἑταιρείῃ φιλότητι. “Ov πάρος is untranslat- able, nor do I see how to explain its corruption from ὡς πώρος the correction of I’, nor Clarke’s οὗ πάρος, the sense of which also is poor. A bolder critic than I might think of ὁππόσ᾽ dp’.

79. σάνδαλα δ᾽ αὐτίκ' ἔριψεν ἐπὶ ψαμάθοις ἁλίῃσιν ἄφραστ᾽ ἠδ᾽ ἀνόητα διέπλεκε θαύματα ἔργα συμμίσγων κ.τ.λ.

Vv. 79 and 80 have evidently to be brought into grammatical connection, and this is most neatly done by Dr. Postgate’s puplv. I had thought of making ὄφρ᾽ dot’ (ic. ἀϊστά) out of ἄφραστ᾽, but this word is evidently sound and not to be disturbed. "Epayev which Gemoll and others liked, never helped.

83 sg. ᾿Αβλαβέως must mean securely, so as not to come undone and trip (βλάπτειν) him. Metaphorically Theognis 1153 εἴη μοι πλουτεῦντι κακῶν ἀπάτερθε μεριμνέων | Swe aBraBéws, μηδὲν ἔχοντι κακόν, ‘ without a check.’ ᾿Αλεείνων 85 may stand if it can mean avoiding (the toil of) way- faring, 1.6. helping him to walk through the sand, but the variants on 361 lend probability to Windisch’s ἀλεγύνων, ‘preparing. Old τ᾽ ἐπειγόμενος 86 agrees with this, ‘being, as he was, in haste’ utpote qui festinaret. Tyrrell’s αὐτοπορήσας for αὐτοτροπήσας or αὐτοπρεπὴς ὡς though not certain is better than the monsters collected in Ludwich’s note, (vol. xv. p. 270).

88. The other accounts of the myth do not mention Onchestus: Antoninus Liberalis gives as the scene the rocks called βάττου σκοπιαί on Mt. Maenalus in Arcadia. Bergk’s notion that Onchestus was chosen because it was half-way between Pieria and Pylos is too ‘modern:’ it would be more to the point to notice that Onchestus wason a rising ground. Really, we have as in the hymn to Apollo another lost local legend, and it is curious that in both hymns the story attaches itself to the same village.

90. ὦ γέρον, ὅς Te φυτὰ σκάπτεις ἐπικαμπύλος ὦμους, ἢ πολυοινήσεις εὖτ᾽ ἂν τάδε πάντα φέρῃσι. καί τε ἰδὼν μὴ ἰδὼν εἶναι καὶ κωφὸς ἀκούσας, καὶ σιγᾶν, ὅτε μή τι καταβλάπτῃ τὸ σὸν αὐτοῦ.

A most enigmatical passage, perhaps intentionally so. The absence of construction in 92 makes the lacuna between 91 and 92, started by Groddeck, indispensable ; and the absence of this line or lines in its turn makes the meaning of 93 doubtful. The purport seems to be twofold: (1) an imper- tinence : ‘you will have plenty to drink when these vines bear.’ (2) A hint to be blind, deaf, and dumb, as to Hermes and the oxen. Gemoll is perfectly

right in seeing no threat nor entreaty in Hermes’ language; the whole is 7 2

256 THE TEXT OF THE HOMERIC HYMNS: IV.

ironic. The lacuna might be supplied by a line to this effect, 7 (or ὡς) μέλλεις μάλα παῦρα νοῆσαι ἐνὶ φρεσὶ σῆσι, (which I offer as a mere stop gap): homoeoteleuton is thus set up, and a construction supplied for 92.

What are we to make of 93? The context will run: ‘you will some day be full of wine, and are like seeing not to see, and hearing to be deaf, and to hold your peace except when Καταβλάπτῃ may be either active or passive; τὸ σὸν αὐτοῦ may be either nom. or acc. It has often been taken to mean ‘your own interests,’ and there is no difficulty in the combination of possessive pronoun and genitive, cf. Z 446 ἐμοὶ αὐτοῦ, 490 τὰ σ᾽ αὐτῆς ἔργα, and the neuter easily stands for ‘interest, concern’: cf. @ 211 ἕο δ᾽ αὐτοῦ πάντα κολούει, Θ 430 τὰ ἃ φρονέων Dion. Chalcus fr. i. 5 τὸ σὸν εὖ θέμενος Eur. Her. Fur. 507 τὸ δ᾽ αὑτοῦ σπουδάσας. Still the phrase is harsh, as Gemoll says, and also no question of the old man’s good or ill enters into the Homeric story; in the later account he was punished, but in the hymn all parties go scot free.

After several years reflection, the passage seems to me to turn entirely upon the vineyard and the taunt Hermes gets out of it: ‘when these vines bear, won’t you be full of wine! you won’t see what you see, you won't hear what you hear: you'll hold your tongue except Except when the wine has some similar effect on his speech, 1.6. except when his tongue is loosened and he blabs. How is this to be got out of καταβλάπτῃ τὸ σὸν αὐτοῦ! I offer τόσον ad τοῦ for consideration: lit. ‘except when you are hindered as much in that too,’ sc. τοῦ σιγᾶν. Exx. of βλάπτειν c. gen. are supplied by the Lexx. Cf. generally Aesch. P. V. 196 δίδαξον ἡμᾶς, εἴ τι μὴ βλάπτῃ λόγῳ, and for the language of 92, 7b. 463 of πρῶτα μὲν βλέποντες ἔβλεπον μάτην, | κλύοντες οὐκ ἤκουον.

94. φὰς συνέσευε for φασὶν ἔσευε is the simple and convincing correction of Demetrius Chalcondyles, and needs no tinkering. Now that he was on hard ground, he drove his herd head-forward and ‘together,’ not straggling as before.

103. ἀδμῆτες δ᾽ ἵκανον ἐς αὔλιον ὑψιμέλαθρον.

᾿Ακμῆτες Ilgen, but we have no reason to suppose that cows driven first backwards through wet sand, and then forwards across hills and ravines and plains all one night would be ‘unwearied.’ In a weak moment I conjectured ἄκμηνοι, since Hermes promptly feeds them (105), but I must not fall into the habits I denounce. ᾿Αδμῆτες seems to correspond to ἄξζυγες in Ant. Lib. 23, 3 εἶτα δ᾽ ἀπελαύνει πόρτιας δώδεκα Kal ἑκατὸν Bods ἄζυγας καὶ ταῦρον. ‘All unyoked they came’: the epithet gives an idea of the value of the theft. Certainly the adjective in this place is rather harsh, but cf. ἄφθιυτοι ἠγερέθοντο 326.

109. δάφνης ἀγλαὸν ὄξον ἑλὼν ἐπέλεψε σιδήρῳ ἄρμενον ἐν παλάμῃ, ἄμπνυτο δὲ θερμὸς ἀύϊτμή.

Cf. vol. xv. p. 285, 6. On again considering the passage, I think a lacuna between 109 and 110 absolutely necessary. Besides that it is hardly

THE TEXT OF THE HOMERIC HYMNS: IV. 257

conceivable that in an aetiological account of the origin of fire the essential act of friction should be omitted, the word ἄρμενον can only apply to the ‘recipient’: it is not necessary in order to prune a branch to hold it ‘firmly fixed’ in one’s palm: such an action on the contrary is peculiarly appropriate to the στορεύς. ’Emédewe over which difficulties have been made, is to trim, prune, point: the Lexx. recognise the force of é:—‘to a point, cut down,’ in ἐπικόπτειν, ἐπιτέμνειν. ᾿Απέλεψε as Herwerden and possibly others prefer, would mean ‘cut off’ the tree; but this is already given in ἑλών.

116. τόφρα δ᾽ ὑποβρυχίας ἕλικας Bods εἷλκε Ovpate.

Ὑποβρυχίας is still uncertain; but as the Lexx. give two verbs, ὑποβρυχάομαι and ὑποβρύχω meaning ‘to roar or bellow a little, there seems no reason to deny the existence of an adjective in the same sense, Or, having regard to the humorous style of the hymn, it might be thought that the cows in the dark cavern were called ‘drowned, i.e. ‘hidden away.’ In any case an alteration like ἐριβρύχους is not to be thought of.

124, ῥινοὺς δ᾽ ἐξετάνυσσε καταστυφέλῳ ἐνὶ πέτρῃ ὡς ἔτι νῦν τὰ μέτασσα πολυχρόνιοι πεφύασι δηρὸν δὴ μετὰ ταῦτα καὶ ἄκριτον.

The only cognate form to μέτασσα is the feminine, « 221 χωρὶς μὲν πρόγοναι χωρὶς δὲ μέτασσαι. I see no reason why μέτασσα may not be the neut. pl. used adverbially—‘in the intervening time’; a record of such a use is preserved in Cramer, An. Oz., i. p. 280, quoted by Lobeck, Pathol., p. 148, ὥσπερ παρὰ τὴν ἐπὶ γίνεται ἔπισσα---οὕτω καὶ παρὰ τὴν μετὰ μέτασσα [not μέτασσαι]. Baumeister inserted μέταξε, but the fact that μέταξε has given place to μεταξύ (a gloss) Hes. Opp. 394 is no argument for its corrup- tion into μέτασσα, a word that must have seemed and did seem nonsense to the scribes. “Axptov about which Gemoll doubts, naturally means ‘boundless, endless’ and here is adverbial: so 577 ἄκριτον ἠπεροπεύει, Pan xix. 26 καταμίσγεται ἄκριτα, and very similarly @ 505 τοὶ δ᾽ ἄκριτα πόλλ᾽ ἀγόρευον. The expression denotes simple belief: ‘a long, an endless time after these things.’

As to the facts, the view first expressed by J. P. D’Orville (Jowrnal of Philology, xxv. p. 254) and then by O. Miiller (Hyperbor-Rom. Studien, p. 310, quoted by Baumeister) seems nearly certain, that the writer of the Hymn saw what professed to be these skins, preserved or shewn by priests; the commentators speak of caves, at Pylos or in Arcadia, where the natural conformation of the rock in some way resembled skins. This miracle I must confess seems harder of belief than the conservation of the actual hides: but relics in general were abundant in the ancient world; D’Orville quotes Ovid Met. viii. 29, and I have noticed the skin of Marsyas Herod. vii. 26, the Alban sow preserved in brine Varror. r. ii. 4. 18, and Eur. Her. Fur., 416 τὰ κλεινὰ δ᾽ Ελλὰς ἔλαβε βαρβάρου κόρας | Λάφυρα καὶ σώξεται Μυκήναις, and many more instances no doubt can be produced.

258 THE TEXT OF THE HOMERIC HYMNS: IV.

132. ἀλλ᾽ οὐδ᾽ ὥς οἱ ἐπείθετο θυμὸς ἀγήνωρ καί τε μάλ᾽ ἱμείροντι περῆν ἱερῆς κατὰ δειρῆς.

To me as to Gemoll περῆν is incomprehensible; certainly περάω cannot mean ‘send down,’ and περαίνω seems out of the question: possibly ἱμείροντί περ εἶν᾽ ἱερῆς κατὰ δειρῆς. Καθίέημι is quite in place, Ὡ 642 λαυκανίης καθέηκα, cf. for the expression T 209 πρὶν δ᾽ οὔ πως ἂν ἐμοί γε φίλον κατὰ λαιμὸν ἰείη | οὐ πόσις οὐδὲ βρῶσις. I am aware that καί τε μάλ᾽ ἱμείροντί περ εἶν᾽ ἱερῆς involves diaeresis after the third foot and elision of az ; if the former objection be fatal perhaps παρεῖν᾽ might stand, ‘admit,’ though it is almost burlesque; cf. below 152, and Ge. xxx. 8 περ ἐστι πάρεστι Pp. But I prefer my first suggestion. Ludwich’s πιεῖν is not very suitable to an ὁσίη κρεάων.

136. ἐπὶ δὲ ξύλα κάγκαν᾽ ἀείρας; Ilgen needlessly altered ἀείρας into ἀγείρας. Hermes we may suppose threw the hands and feet upon the embers of his old fire in the κατούδαιος βόθρος ; he then piled more wood upon the top, and consumed everything in the heart of the fire. v. 113 ἐπέθηκεν, and for the general use of ἀεέρω I 214 Σ 615 a 141 ¢ 120. The writers neither of this nor of the other hymns object to the assonance ἀνάειρε ἀείρας.

147. Αὔρῃ ὀπωρινῇ ἐναλίγκιος, ἠύτ᾽ ὀμίχλη, ‘like a draught in autumn’ is a comparison that comes home to anyone living in a cave or a chalet, but there is surely no justification for a mythologist (like Roscher) to see in it a proof of Hermes’ original function as God of Wind; cf. £ 20 of a dream ἡ δ᾽ ἀνέμου ὡς πνοιὴ ἐπέσσυτο δέμνια κούρης : ὃ 802 a spirit makes an entry very like Hermes’—és θάλαμον δ᾽ εἰσῆλθε παρὰ κληῖδος ἵμαντα. Quintus iv. 111 αὔρῃ ὑπηώῃ ἐναλίγκιος.

152. λαῖφος ἀθύρων ‘playing with the clothes’ may seem a singular phrase, but it is undoubtedly defended by the passive ἀθυρομένη ‘being played’ ν. 485, and other accusatives from Pindar and the Anthology may be seen in the Lexx.

159. ἢ σὲ λαβόντα (φέροντα M) μεταξὺ «.7.r. In the Oxford text Matthiae’s λαθόντα was adopted as a stop-gap, but it belongs to the class of unmotived corruptions, and the right remedy for the line seems quite uncertain. Neither Ludwich’s Xdovra μάλ᾽ ὀξὺ nor the lacuna proposed vol. xv. p. 287 are convincing; and φέροντα also must be accounted for in any conjecture that is to hold water.

160. Ruhnken’s τάλαν is only ingenious; πάλιν ‘get you back the way you came’ is perfectly in point.

163. τί με ταῦτα τιτύσκεαι. In meaning τιτύσκεαι might well stand : to aim is a metaphor easily transferable to words; ταῦτα also as.a cognate acc. is possible, but the accusative we is a stumbling block. Before accepting Pierson’s δεδίσκεαι one would wish to see instances of the exchange of δ and τ,

THE TEXT OF THE HOMERIC HYMNS: IV. 259

167. βουλεύων ἐμὲ καὶ σὲ. βουκολέων (Ludwich) or βουκολέειν (Gemoll) is exceedingly ingenious and graphically not too far off.

187. ἔνθα γέροντα κνώδαλον εὗρε νέμοντα παρὲξ ὁδοῦ ἕρκος ἀλωῆς.

Κνώδαλον is usually held to be corrupt, but the conjectures (τρόχμαλον, νώδαλον "κώδαλον, etc.) are unsatisfactory, and the passage falls into such hopeless confusion without this object to νέμοντα, that perhaps we may allow the well-known lines Aesch. !Prom. 462 to save the word. Prometheus says :—

κἄζευξα πρῶτος ἐν ζυγοῖσι κνώδαλα ζεύγλαισι δουλεύοντα σώμασίν θ᾽ ὅπως θνητοῖς μεγίστων διάδοχοι μοχθημάτων γένοινθ᾽, ὑφ᾽ ἅρμα τ’ ἤγαγον φιληνίους ἵππους.

Horses then being excluded, κνώδαλα in Prometheus’ mouth must imply oxen and mules, and given the humorous style of the Hymn, one may without much violence take κνώδαλον, literally ‘ beast,’ to mean here ‘ox.’ "Epos ἀλωῆς is metaphorical, not literal—‘ prop of the vineyard ’—of the ox who ploughs between the vines, and draws the grapes and the fodder. Translate ‘there he found an old man grazing his ox, the stay of his vineyard, on the edge of the road.’ Gemoll’s grammar and agriculture are equally strange.

226. αἰνὰ μὲν ἔνθεν ὁδοῖο τὰ δ᾽ aivorep’ ἔνθεν ὁδοῖο.

With Franke and Ludwich, Newe Jahrb., 1887, p. 327, n. 15, I quite agree that Hermes’ footprints only are intended. The cows Apollo has noticed, v. 220, and recognised them: the other spoor baffles him, and he expresses a naif astonishment ‘wonderful here, and more wonderful there’; Hermes ‘waddled,’ ἐπιστροφάδην δ᾽ ἐβάδιξεν 210, partly because the cows straggled, partly because his peculiar foot-coverings made him flounder. Cf. 357.

231. Gemoll suggests that the ‘delightful smell’ came from Maia’s fire, as that in ε from Calypso’s. This, however, is too ‘literary’; the ancients had noses for natural smells, and the fragrance of the Alp is intended here. Martial, iii. 65, 4, gramina quod redolent quae modo carpsit ovis.

239. The transposition ἀνεείλε᾽ for ἀλέεινεν is simple and probable; ἀνεείλε᾽, a8 Dr. Postgate pointed out te me, is more correct than ἀνεείλει (Lohsee) ; it also represents more exactly the letters of ἀλέεινεν, and this is of importance when a metathesis is in question,

242. ἄγρης" elveréov τε χέλυν ὑπὸ μασχάλῃ εἶχεν.

Martin’s ἐγρήσσων for aypns εἰν is probable; P 660 for ἐγρήσσοντες Ly, M,, Vat.,, have dy-, A551 there seems no variant, v53 ἀγρήσσοντα ‘P.

260 THE TEXT OF THE HOMERIC HYMNS: IV.

’Ereds also isa word that has a tendency to dissolve, cf. the variants T 255. Read therefore with Martin and Hermann

ἐγρήσσων ἐτεὸν δὲ χέλυν K.T.X.

which is a little simpler than ἐτεόν γε" χέλυν δ᾽. Hipponax 89 ἑρμῆ μάκαρ καθ᾽ ὕπνον οἶδας ἐγρήσσειν is in point and seems to have escaped the com- mentators.. In the line before, it is Barnes, not Hermann, that deserves the credit of the excellent conjecture $7, half-confirmed afterwards by y’s 6%. Ludwich’s objection, WV. J., 1887, p. 325, n. 10, that in a hundred similar cases ὡς, not $7, is used would have weight if the MSS. 6) and δή did not point strongly to the rarer word, another instance of which may now be drawn from the new fragments of Callimachus’ Hecale (see Ellis, Journal of Philology, xxiv. p. 158): in the same fragments φηλητής (spelled φιλητής) occurs.

258. ὑπὸ yain ἐρρήσεις ὀλίγοισι μετ᾽ ἀνδράσιν ἡγεμονεύων.

Hermes’ subjects have been called by different critics λυγροῖσιν, ἀλυτροῖ- σιν, δολίοισιν, ὀλοοῖσιν, φθιμένοισιν (!) “ ψυδνοῖσιν vel potius ψυδροῖσιν (olim φήλοισιν vel λιροῖσιν)᾽ (!!). I quote Ludwich’s note, omitting the names of the authors of these pearls of learning.

Mr. Tyrrell, and Boissonade quoted by Franke, defend ὀλέγοισιν, which has much point. Hermes will go to Tartarus and.be king among men of his own size, 1.6. other bad babies. It is to be presumed that ancient ideas of the

next world kept a place for children among the ‘ matres atque viri, heroes and girls,

272. βουσὶ per’ ἀγραύλοισι. It is certainly a small step from -σὶ per’ to -σὶν ἐπ᾽, yet I think that the MS. reading gives as good a sense as Schneidewin’s conjecture. To have passed through the door with oxen is even a greater feat for an infant than to have gone out after them, and it is

the former that Apollo accuses Hermes of when he examines Maia’s house- hold stores.

279. ὀφρύσι ῥιπτάζεσκεν. There is no quotation to support ῥιπτάζειν in any connection with the eyes, but it occurs absolutely, in the middle or active in Hippocrates of tossing in bed, of patients unable to sleep (e.g. Acut. ii. 18), and a substantive ῥιπτασμός exists in the same sense. Therefore, since the MSS. give ὀφρύσι, it seems as well to preserve the intransitive use here also. Hermann preferred the acc. ὀφρῦς ; the alterations of the excellent word ῥιπτάζεσκεν do not need enumerating.

280. ἅλιον τὸν μῦθον ἀκούων. On ὡς which M ὦ; give instead of τὸν and p gives together with τὸν, see vol. xv. p. 304. Tyrrell defends the tradi- tion, but ἅλιον predicative is certainly hard, ‘for naught.’ The occurrence of ὡς in so many MSS. suggests that it may after all not be a gloss, and that emendation should take account of it. ᾿Ακούων is sound, the attempts upon

it are unsuccessful: Stadtmiiller, who decidedly has not la main heureuse, thought of ὑλακτῶν.

THE TEXT OF THE HOMERIC HYMNS: IV. 261

282. ὦ πέπον, ἠπεροπευτά, δολοφραδές, ἣ σε μάλ᾽ οἴω πολλάκις ἀντιτοροῦντα δόμους εὖ ναιετάοντας ἔννυχον οὐχ ἕνα μοῦνον ἐπ᾽ οὔδεϊ φῶτα καθίσσαι.

It is to be wished that a better parallel than Theoer. i. 51 ἀκράτιστον ἐπὶ ξηροῖσι καθίζειν (especially as in modern texts the conjecture ἀκρατισμὸν or ἀκρατισμὼ is generally printed) could be brought to ἐπ᾽ οὔδεϊ καθίσσαι. To ‘seat a man upon the floor’ is an odd expression for to rob; besides that oeas the words run may be either subject or object to the verb. A somewhat similar odd phrase is ἐπ᾽ ἀκινήτοισι καθίζειν Hes. Opp. 750. The Paroemiographi do not help.

315. ὁ μὲν νημερτέα φωνήν οὐκ ἀδίκως ἐπὶ βουσὶν ἐλάξυτο κύδιμον “Ἑρμῆν.

The alterations of φωνήν into φωνῇ, φωνεῖν, φωνῶν are unacceptable ; they belong to the class of petty, unmotived, and therefore unconvincing changes. If φωνῶν or φωνεῖν had originally stood, there is no probability of it being altered by any reader into -ήν. After Hermann’s brilliant correction of φωνῆς into φωρῆς 136, based upon the variation φωνήν, φωρήν 385, the same emendation is obvious here (and Windisch already had attempted g@pa). The sense will,then be ‘he was attempting to convict Hermes of clear theft in the matter of oxen,’ a good Attic construction (ἑλεῖν twa τι) which Matthiae seems to have contemplated here. If it be thought too forensic my only other suggestion is to return to φωνήν with a lacuna con- taining ἑείς or an equivalent word. Οὐκ ἀδέκως, if prosy is sound, ‘ not without justice,’ as opposed to Hermes’ ‘arts and crafty words,’

325. εὐμυλίη (εὐμιλίη M) δ᾽ ἔχ᾽ Ὄλυμπον ἀγάννιφον. The older con- jectures endeavoured to produce a word somewhat resembling εὐμυλίέη ; so Heyne’s αἱμυλίη accepted by Ilgen but rejected by Matthiae, Hermann’s εὐμελίη or ἐμμελίη, Franke’s εὐελέη. The later attempts desert the letters of the MSS. and may well be left unquoted. I will have the courage to inter- pret. Hes. Opp. 529 of animals in a storm, καὶ τότε δὴ κεραοὶ καὶ νήκεροι ὑληκοῖται | λυγρὸν μυλιόωντες ἀνὰ δρία βησσήεντα | φεύγουσιν. Μυλιεόωντες was a rare word and variously interpreted, and Crates indeed read μαλκιόωντες. However, the second of the explanations of Proclus seems correct, τὰ χείλη κινοῦντες ὑπὸ τῆς ψυχρότητος, chattering from cold ; cf. μοιμύλλω and μύλλω which Prellwitz connects with the simple form μύω. When we consider the close connection between the vocabulary of the four greater Hymns and Hesiod (brought out by Fietkau, De carminum Hesiodeorwm atque hymnorum quatuor magnorum vocabulis non homericis, Reg., 1866), perhaps εὐμυλέη may express the action exactly opposite to λυγρῶς μυλιοῦν, ‘a pleasant buzz or hum.’ The sense (much the same as D’Orville’s στωμυλέη, Journ. Phil. xxv. p. 255) would suit the easy style of this Hymn, ‘a pleasant hum possessed Olympus’; the gods were exchanging morning salutations, μετὰ χρυσόθρονον 7d. Right or wrong, I think this attempt at interpretation better than inventing another

262 THE TEXT OF THE HOMERIC HYMNS: IV.

word of the measure οὖ -- and with much timidity I submit it to the etymologists.

326. ΓΑφθιτοι is difficult as a predicate, but finds a parallel in ἀδμῆτες δ᾽ ἵκανον v. 103. Groddeck’s ἀθρόοι certainly is quite inadmissible ; the later conjectures ὥθετοι, ayy θεοὶ, and Tyrrell’s ἄφθονοι do not, I am afraid, help matters. Probably, as Gemoll suggests, ἀθάνατοι is used as a complete substantive, qualified by ἄφθιετοι : ‘the Immortals gathered, deathless.’

a \ \ \ > > \ fal 344. τῇσιν μὲν yap βουσὶν ἐς ἀσφοδελὸν λειμῶνα > ‘ / 2’ , ? / le ἀντία βήματ᾽ ἔχουσα κόνις ἀνέφαινε μέλαινα.

I see no difficulty in the dative βουσὶν: ‘in the case of the oxen the sand, which held them, showed the foot-prints facing the meadow. ᾿Αντίος with ἐς is unexampled ; p 333 τὸν κατέθηκε φέρων πρὸς τηλεμάχοιο τράπεζαν] ἀντίον is obviously different. One may, I suppose, translate ‘facing towards. “Eyouca must mean ‘took and kept. The sentence is as the commentators say, awkward; but the antithesis to αὐτὸς δ᾽ οὗτος was the cause of its contortion.

346. αὐτὸς δ᾽ οὗτος ὅδ᾽ ἐκτὸς ἀμήχανος.

Ὅδ᾽ ἐκτὸς is completely dark. Barnes thought ἐκτός might mean ‘supra modum’ and Ruhnken gave effect to this interpretation by writing éEoy’. The modern conjectures ὀπηδός, ὄλεθρος, ὁδαῖος, etc. are evident stop- gaps. Formally Hermann’s ἄϊκτος is still the best, as it makes a good parallel to ἀμήχανος, but the sense is poor. ‘O δεκτός and ὅδ᾽ ἑκτός do not promise, Can ὁ ἐκτός mean the ‘outsider, stranger, one who is not yet admitted into Olympus, according to a sense given by the Lexx. not earlier than Plato? Αὐτὸς δ᾽ οὗτος is certainly sound; Apollo’s rage rises in stages: ‘the cows have their footsteps the wrong way about, but this, this very ἐκτός here ν᾽

849. ὡς εἴ τις ἀραιῇσι δρυσὶ βαίνοι. Δρυσὶ is simple instrumental dative, like ποσσὶν and χερσὶν in 346 and 347, ‘as if one walked with trees, instead of feet. Solon xi. 5 ἀλώπεκος ἴχνεσι βαίνει, M 207 πέτετο πνοιῇς ἀνέμοιο. The conjectures σὺν (Gemoll) ἐν (Herwerden) are needless.

357. διαπυρπαλάμησεν and 361 ὠμόργαξε are two of Ilgen’s best contributions, the former a joining together of διὰ πῦρ παλάμησεν given by M (as I should have noticed in Part II.), the latter for the vox nihili ὠμάρ- tate. Ludwich, indeed, alone of editors keeps the latter, but his χέρσῳ explain who can,

409 sg. It is impossible not to feel that a certain quantity of matter has perished in this context; verbal alteration does not suffice to restore (1) the construction and (2) the sense. Under the former head the fem. plur. ταὶ 410 cannot possibly follow δεσμὰ ἄγνου, even κατὰ σύνεσιν : ἐγκρύψαι 416 wants an object, and no word can be supplied out of the preceding lines. (2) The motive of ‘twisting the chains’ 409 is entirely unexplained, and also ‘chains’ are not made of agnus castus. Something intervened between 409

THE TEXT OF THE HOMERIC HYMNS: IV. 263

and 410; the chains were perhaps changed into shoots of willow. Again in 415 Hermes ‘flashes fire, but in the same sentence 416 he is ‘soothing’ Apollo. In short the scene is not stated; we have six lines left from a much longer passage. We may think we see what the action may have been, but no literature could have left it originally in such obscurity. I therefore leave the language as it stands (and every word seems sound), and suppose with Baum. two lacunae of unknown size after 409 and 415.

I conjecture, from the surviving fragments, the course of events to have been this. Apollo, irritated at seeing the skins, and inferring that two cows were lost from his herd, began to ‘twist strong chains’ in his hands (for Hermes, naturally, not for the unoffending cows), They by magic, ἑρμέω βουλῇσι κλεψίφρονος, fell off Hermes, or fell to the ground before they could be put on him, and turned into shoots of willow, took root in the ground, and in a moment (αἶψα) grew up and made a bower or pergola over all the cows, at which Apollo, with reason, θαύμασεν ἀθρήσας. Hermes’ next action, to ‘look askance at the ground, his eyes glancing fire’ is certainly inexplicable : his desire ‘to hide’ can only refer either to the skins or to the fat and flesh which was stored inside the cave. Lastly, one or the other gap must have contained a mention of the lyre, which is referred to without definite intro- duction in 417. Πῦρ in 415 is rightly restored by Lohsee and Ludwich for Martin’s πύκν᾽ : cf. Hes. Theog., 827 (quoted by Clarke) and Quintus, viii. 28. ‘Pela te καὶ πάσησιν 412 is well defended by Gemoll with Hes, Theog., 87.

426. γηρύετ᾽ ἀμβολάδην, ἐρατὴ δέ οἱ ἕσπετο φωνή, κραίνων ἀθανάτους τε θεοὺς καὶ γαῖαν ἐρεμνήν.

Hesych. κραίνειν' τιμᾶν may perhaps establish the meaning ‘celebrate,’ ‘tell’ which is definitely maintained by Maurophrydes (Kuhn’s Zeitschr., vii. 346 84., quoted in the Lex. Hom.) here, 531 and 559. Let etymologists pro- nounce. The conjectures (e.g. κλείων) are unconvincing and Stadtmiiller’s οὐρανόν (adopted by Ludwich) one of the worst that has disfigured a text. Ἐρατὴ---φωνή is generally recognised to be parenthetical; parentheses are frequent, ¢.9. :

A 429, οἱ δ᾽ ἄλλοι ἀκὴν ἴσαν, οὐδέ κε φαίης / \ “ Μ ya? 40 iP) / τόσσον λαὸν ἕπεσθαι ἔχοντ᾽ ἐν στήθεσιν αὐδήν, σιγῇ δειδιότες σημάντορας.

Herm. 175.

436. Μηχανιῶτα is analogous to σπαργανιῶτα 310, χαριδῶτα Herm. Xviil. 12, εἰραφιῶτα Dion. i. 2,17, 20 Πὰν σκοπιῆτα, Anth. Pal., vi. 34, 5, 109, 7, λοφιῆτα 79, 1, ὑλειῶτα 106, 1. I see no objection to πονεύμενε ‘labourer, industrious.’ Apollo as throughout is ironical, and congratulates Hermes on the variety of his accomplishments: ‘butcher, trickster, workman, minstrel.’

Δαιτὸς ἑταίρῃ (Ludwich)is ingenious, but the God may be said to be present where his invention is used or at what he inspires ; πονεύμενε δαιτὸς Eraipy

264 THE TEXT OF THE HOMERIC HYMNS: IV.

in combination is intolerably prosaic. Cf. Hipponax i. ἔβωσε Mains παῖδα Κυλλήνης πάλμυν | ‘Epp κυνάγχα, Μῃονιστὶ Κανδαῦλα | φωρῶν ἑταῖρε.

437. πεντήκοντα βοῶν ἀντάξια ταῦτα μέμηλας. I can do nothing with μέμηλας ; there is no ex. of μέλομαι c. ace. nor in fact of the 2 perf. except in the third person. Μεμηλώς (Ludwich) does not ease the construction, μέμηλε (Eberhard) lacks sense, and this verb is really not wanted at all; μεμελέτηκας or μέμολπας is the sense; Lohsee’s μέλη σά is too desperately ingenious. Gemoll has added to the enormity of Stadtmiiller’s ἀνέφηνας by putting it in his text. Are there, or are there not, principles of emendation ?

447, τίς τέχνη, τίς μοῦσα aunyavéwy μελεδώνων ;

The ‘muse of hopeless cares’ passed muster till Schneidewin, who made the prosy and inaccurate alteration ἀμήχανε σῶν μελεδώνων. Hermes’ cares were material, and confined to admission among the Olympians. For the construction I may quote Franke who shows often excellent judgment: ‘est genitivus ut dicitur objecti: cantus contra sollicitudines et curas. ᾿Αμηχανέων is not from the unheard of ἀμηχανής, but is gen. fem. from ἀμήχανος; I may refer to a list of compound adjectives of three terminations, vol. xv. p. 261. The word itself is greatly in point, cf. 434 ἔρος ἀμήχανος, Theocr. xiv. 52 apnyavéovtos ἔρωτος : μελεδῶνας occurs Apoll. 532. Herwerden’s and Gemoll’s conjectures, which would destroy the general predication of Hermes’ art, will not bear repeating. The sentiment is that of Hes. Theog. 55 (the Muses) λησμοσύνην τε κακῶν ἄμπαυμά τε μερμηράων., and the well-known lines, Cypria fr. 10 οἶνόν τοι Μενέλαε θεοὶ ποίησαν ἄριστον | θνητοῖς ἀνθρώ- ποισιν ἀποσκεδάσαι μελεδώνας.

453. ἀλλ᾽ οὔπω τί μοι ὧδε μετὰ φρεσὶν ἄλλο μέλησεν οἷα νέων θαλίης ἐνδέξια ἔργα πέλονται.

Θεῶν (Herwerden, Gemoll) is excessively weak for νέων ; the comparison is the same as in 55 ἠύτε κοῦροι | ἡβηταὶ θαλίῃσι παραιβόλα κερτομέουσιν. Nor is τοῖα (Ludwich) necessary: construe οὔπω μοι ἄλλο TL ὧδε μέλησεν ἐκείνων ola θαλίῃς νέων, ἐνδέξια ἔργα, πέλονται (as Matthiae).

450, 7. νῦν δ᾽ ἐπεὶ οὖν ὀλίγος περ ἐὼν κλυτὰ μήδεα οἶδας ἷζε πέπον καὶ θυμὸν ἐπαίνεε πρεσβυτέροισι.

That M alone preserves these two lines is nothing against. their genuine- ness, as indeed has been recognised since Ruhnken’s time: but M’s character for uncorrected corruption would admit mistakes in tradition and allow of bolder remedies. That some corruption has happened is obvious.

To take the words in order. {fe is usually accepted, though as Gemoll notices, there is no motive for Apollo ordering Hermes to sit down and as a matter of fact he does not do so. The first word of a verse is peculiarly exposed to corruption, and for instances of loss or addition of initial, cf. Z 185 δύμεναι, ἔμμεναι ‘Li, 208 Ἴσανδρον, Πείσανδρον Strabo, Hes. Theog. 970, ᾿Ιασίῳ, ᾿Ασέω and ᾿Ασσίω, Theognis, 477 δείξω, ἥξω and ἥκω, Hippocr. Vet.

THE TEXT OF THE HOMERIC HYMNS: IV. 265

Med. 22 πίη, ein, in. » Schneidewin tried εἶκε, but this evades the problem ; three years ago in the Academy, Sept., 1894, I proposed δῖξε πέπον καὶ θυμὸν ἐγεῖραι πρεσβυτέροισιν, 1.6. ‘think twice before’; ἐγεῖραι is a fair uncial permutation for ἐπαίνει, but the construction of 8/f@ is always with #, and καὶ also is difficult. The accepted method is Ruhnken’s μῦθον for θυμόν, but (1) metatheses should be avoided except under the clearest proof; the only place in the Hymns where a metathesis seems necessary is v. 256, Ilgen’s λαβών for βαλών. (2) μῦθον ἐπαινεῖν, approve or agree to a speech, is an odd phrase for general humility; and again, how does sitting come in 7 (3) the dative in such a construction is unintelligible. Other attempts, to be seen in Ludwich’s note, hardly need mention. After much reflection I incline to think the sentence springs out of Hermes’ musical accomplishments; ‘since at your young age you are so clever, use your gifts for the general good’; ἦξε will then be of the bard, at the table in Olympia. ‘Sit, and the spirit in your elders, The missing word must be an equivalent of ‘comfort,’ and what but faivey? And this I now see was Schneidewin’s view. He read ἴαινε, but the synizesis ἐπέαινε does not seem impossible (Monro, Hom. Grammar, ὃ 378 and for the elision of -c cf. wep’ ἐἰγνύος Herm. 152, περεβάλοντο Aesch. Ag. 1144, περεσκήνωσεν Ewmen. 637 and schol.), and the rarity of the word together with the metrical license will have given ἐπαίνει.

460. τόδε κρανάϊον ἀκόντιον. That some adjective from x«pdvoy or xpaveca is intended I do not doubt; the usual form is xpaveivoy, which Ilgen restored. Cf. δορυκράνου λόγχης ἰσχὺς Aesch. Persae, 151. For omission or insertion of v cf. the forms of ἐλατιονίδη Apoll. 210; for the quantity Kpaveivoy One may perhaps compare θαμιναὶ v. 44.

460.

4 \ b] , ἢ μὲν ἐγώ oe \ ’ ; / ΔΝ e 4 κυδρὸν ἐν ἀθανάτοισι καὶ ὄλβιον ἡγεμονεύσω ΄ ϑι 2 \ aA eee / > > / δώσω T ἀγλαὰ δῶρα Kal és τέλος οὐκ ἀπατήσω.

For ἡγεμονεύσω which is enigmatical Mr. Tyrrell suggests ἡγεμόν᾽ εἵσω, Mr. Agar ἡγεμόν᾽ ἕσσω. This close coincidence cannot but have weight. No advantage results from transposing ἡγεμονεύσω and οὐκ ἀπατήσω, as Ludwich after Waardenburg prints.

471, καὶ τιμὰς σέ γέ φασι δαήμεναι ἐκ Διὸς ὀμφῆς μαντείας θ᾽ ἑκάεργε Διὸς παρὰ θέσφατα πάντα.

This is the punctuation and reading of the MSS., which at Gemoll’s suggestion (in his note, for in his text he goes with the majority) I have restored. Usually, following Matthiae a colon is put after τιμάς, ye is changed into δὲ, and θ᾽ after μαντείας is suppressed. The documentary reading, however, gives τιμαί and μάντειαι as two gifts of Zeus to Apollo, and this corresponds to the division 531 sg. The accent on παρὰ is best retracted.

473. τῶν [y, καὶ Map] viv αὐτὸς ἔγωγε παῖδ᾽ ἀφνειὸν δεδάηκα.

A line unmetrical and most mysterious. The older critics omitted γε to help the metre and took ἀφνειὸν often of money ; others tried to turn it into

266 THE TEXT OF THE HOMERIC HYMNS: IV.

something resembling αἰφνίδιος. Hermann’s πανομῴφαῖον (too good), for a long time won acceptance. I think most scholars will now admit that τῶν--- ἀφνειόν, ‘rich in which’ is sound; but how to deal with até’? I still admire the amazing ingenuity of Tyrrell’s wed’ ἀφνειῶν, but after much consideration, I think the simplest expedient is to write ἐγώ oe παῖ ἀφνειόν. The ὃ was inserted to avoid the apparent hiatus. This is a phenomenon of wide occurrence and that shews itself under very different forms: one or two cases are collected vol. xv. p. 275, here I may add E 4 δαῖε οἱ, date δ᾽ of T 194 δῶρα ἐμῆς, δῶρα δ᾽ ἐμῆς. Lats might be thought precocious in Hermes’ mouth, but he calls Apollo κοῦρε 490, and he may have thought like Aristophanes παῖδα yap κἂν ἢ γέρων | καλεῖν δίκαιον ὅστις ἂν πληγὰς λάβῃ ( Wasps, 1297).

479. Gemoll’s ἐπιστάμενος for ἐπισταμένως is very neat, and gives a good sense. ᾿Επισταμένην (Barnes) is very bad, and ought not to have been adopted so generally. The accusative would never have corrupted into the adverb. I agree with Gemoll that the passage 478-480 is sound: Ludwich’s transposition of εὐμόλπει and εὔκηλος does not assist.

485. peta συνηθείησιν ἀθυρομένη μαλακῇσιν. An affected way of describing a musical instrument, but the sense is plain: ‘easily played by gentle practice,’ the harp will respond to the executant who takes the trouble to ‘learn its ways,’ to ‘painful labour’ ἐργασίη δυήπαθος it refuses itself, Franke is right with his interpretation conswetudines molles pro consuetudine molliter tangendi fides. ᾿Αθυρομένη is nothing but passive.

497. Ἑρμῇ δ᾽ ἐγγυάλιξεν ἔχων μάστιγα φαεινήν. χων naturally offends, but Matthiae’s generally accepted ἔχειν belongs to the category of unmotived corruptions, like φωνῶν for φωνήν v. 315. If ἔχειν had been original, who consciously or unconsciously would have changed it to ἔχων 7 Martin’s ἑκών is better, but I venture to write ἑλών, permutation between which and ἔχων is graphical, and frequent in Homeric MSS. Δ. E 136 H 197 (ἑκών) A 488 Ψ 219 Ω 735 a 95 ε 387.

509. I can make nothing of σήματ᾽ ἐπεὶ. The sense is so complete without it that no clue is given to its possible meaning. If it were joined with ὡς ἔτι καὶ νῦν a verb would be wanted, but ὡς ἔτι καὶ νῦν (v. 125 ὡς ἔτι νῦν) ‘as still now’ requires φιλεῖ and seems a simple expression like δηρὸν δὴ μετὰ ταῦτα καὶ ἄκριτον v.126. If σήματ᾽ belongs to 509 and means ‘as a token’ it must at least be dative, and its position makes such a sense very doubtful.

526 sq. The transition to direct oration is quite intolerable, ἐκ requires a verb, τέλειον is senseless with σύμβολον, lastly the pronoun ce is required. So many conditions can only be fulfilled by a lacuna, which might have con- tained such a line as αἰετὸν ἧκε πατήρ' ὁ δ᾽ ἐπώμοσεν ἢ σε μάλ᾽ οἷον. Zeus to approve of the compact let fly the τελειότατος πετεηνῶν, at sight of which Apollo made oath,

THE TEXT OF THE HOMERIC HYMNS: IV. 267

558, Schneidewin’s correction ἄλλοτε ἄλλῃ for ἄλλοτ᾽ ἐπ᾿ ἄλλῃ is justified by Hes. Opp., 713, where for ἄλλοτε ἄλλον various MSS. give ἄλλοτε τ᾽ ἄλλον, ἄλλοτ᾽ ἐς ἄλλον. Desire to avoid hiatus produced alike ἐπ᾽, τ᾽ and és. Some other examples are given in Rzach’s note

ad loe.

568. The construction here is absolutely broken; one or two lines are wanted to pave the way to the orat. obliqua and provide a principal verb for ἀνάσσειν. They may, as Gemoll says, have contained a reference to Zeus and have run thus:

ὡς ἔφατ᾽ - οὐρανόθεν δὲ πατὴρ Ζεὺς αὐτὸς ἔπεσσιν θῆκε τέλος + πᾶσιν δ᾽ ὁ μὲν οἰωνοῖσι κέλευσε.

T. W. ALLEN.

268 INSCRIPTIONS FROM MYSITA.

INSCRIPTIONS FROM MYSIA.

Most of the following inscriptions were copied on a tour through the country of the Rhyndacus and Macestus in the autumn of 1894. A few were added on a hasty visit to Balukiser and Balia in September 1896. I have ar- ranged them geographically in a zigzag line, from Brusa westwards to Cyzicus, then south-eastwards to Tavshanli and the plain of Simav, and then westwards again by Balat, Kebsud, Balukiser, and Balia to the Aesepus. For the precise position of the several places, and for topographical details, I may refer to a paper by Mr. H. M. Anthony and myself published in the Journal of the Royal Geographical Society, February and March 1897. Prof. W. C. Ἐς, Anderson, who accompanied me on both journeys, has kindly allowed me to compare his readings of the inscriptions with mine, and in one instance I have adopted his copy in preference to my own. I have not repeated any inscriptions which I have found to be already published, unless I believed that I could make material additions or corrections. Many of the following inscriptions will appear to be of little or no importance, but in Asia Minor much light may be thrown on the ethnology, religion, and civilization of a district by mere nomenclature and spelling. The work of Prof. W. M. Ramsay has shown how much may be learnt from the epitaphs and dedications of semi-barbarous peasants. Several of these inscriptions, however, are of some interest, for example No. 27 for political history, Nos. 13 and 67 for historical geography, and Nos. 48 and 64 for the religion of the country.

1. Tachtali, a village about three hours west of Brusa: in the Church of Hagios Theodoros. Ornate marble stele with gable and spirally fluted columns, 2 feet high, 1 foot 10 inches broad. Letters about 1 inch.

AT A®HTYXH ᾿Αγαθῇ τύχῃ.

ΘΡΕΠΤΗΡᾺ Θρεπτῆρα μουσῶν καὶ λόγων κοσμήτορα MOYCUIN Kopvodtov οὕτω Φέρμος ἀντημείψατο. KAIAOPUIN

5 KOCMHTOPA The metrical form of the inscription is

meant as a compliment to Cornutus, from

KOPNOYTON whom Firmus learnt the art of writing OYTU verses. ΦΙΡΜΟΓ

ANTHMEIYATO

INSCRIPTIONS FROM MYSTA. 269

2. Tachtali: under a stair outside the Church of Hagios Theodoros, Block of coarse marble broken at both ends. Letters about 24 inches,

¥XANCOITTA εἸὐχάν σοι π[αῖδας καὶ 1 AIAWNTTAIA π]͵αίδων παΐῖδ[ας

3. Tachtali: in the floor of the bath. Marble slab, about 24 feet long, broken below. Letters about 2 inches.

+ TIC TOC CEPA ἜἪ Πιστὸς θερά- πων εν θα δε πων Θ(εο 0] ἐνθάδε KA TAK 4 WANMICO κατάκ[ιται] ᾿Ιωάννης

ETTIC KoTHGQHOEIC ᾿Επισκόπης. Μηθεὶς του τὸν

4. Akcheler, near the eastern shore of the lake of Apollonia: in a lane to the west of the village. Rough boulder. Letters about 24 inches, rudely inscribed.

tn GW) P AIADEPON TECTONK OTA

Apparently a boundary stone. The character of the writing resembles that of another boundary stone, built into the wall of the mosque. See Le Bas No. 1095.

5. Apollonia (ad Rhyndacum): outside the wall, not far from the bridge. Fragment of marble column. Letters about 2 inches.

ΚΑΙΛΙΟΣ Καΐλιος LTEKOYNAL Σεκοῦνδος ΟΡΗΤΩΡ ὁ ῥήτωρ.

6. Apollonia: in a wall just inside the gate. Marble block, broken to right. Letters 1} inch.

ΜΑΙ witAAM! Μαϊ[γ]νῶλλα[ν φιλό- ΞΟΦΟΝΜΑΓΝ -σοφον Μάγν[ου b/AOLOPVOYO’ φιλοσόφου bu[ya- TEPAMANIC -τέρα, Μηνίο[ν φιλο- -- τοντν ὅ -σόφ]ου γυναῖκα.

H.S.— VOL. XVII. U

270 INSCRIPTIONS FROM MYSIA.

7. Apollonia: (a) in a wall near the gate. Large rough block. Letters about 3 inches, rudely inscribed.

TTP W TOC

(b) Forming a step to a shop on the other side of the street. Similar block, similarly inscribed. TEXEIAI

8. Apollonia: in a stair leading down to the shore. Round marble pillar, broken below. Letters about ? inch.

AT A@HITYXHI ᾿Αγαθῆι τύχηι.

TAICANOAICTHC ταῖς avoais τῆς ΘΕΟΥ θεοῦ €IMIANOCOAKT//// ‘E[p]ucavos ὁ ᾿Ακτ[ίου 5 ZHCACATIEAUKEN 5 ζήσας ἀπέδωκεν

€YXAPICTHPION εὐχαριστήριον TAWTAKAITONBW Ta ὦτα Kal τὸν Bo- MONETTIIEPEIAC -μὸν ἐπὶ ἱερείας TIPOC/////////THC [τῆς δεῖνος]

dvoais apparently for ἀνωγαῖς the y being softened to y, as often in modern Greek.

9. Apollonia: ina house. Marble stele, broken above and at both sides, Beneath the inscription is a panel (74 x 7 inches) containing a relief of a lady, seated to right, holding a child in her lap, while a maid facing her presents a basket. Letters } inch.

A ATTON ANAPOY AZKAH XAIPE

10. Apollonia: in a house. Rough marble fragment, broken above and at both sides. Above the inscription are carved the talons of an eagle in high relief.

ONTAGHHGOE « gan τῆς Θεί..... εὖ-

ΧΗΝ -χῆν.

11. Apollonia. Copy communicated by a dealer in Brusa. Stone described as a square base, with mouldings above and below, broken to left.

ΑΓΑΘΗΙ TYXHI ᾿Αγαθῆι τύχηι.

AYTOKPATOPI Αὐτοκράτορι ΣΑΡΙΑΔΡΙΑΝΟΙ Καί]σαρι ᾿Αδριαν[ὦ }ν ΜΠΙΩΙΣΩΤΗΡ ᾿Ολυ]μπέωι σωτῆρϊἰ:

ΙΚΚΑΑΙΚΤΙΣΤΗ καὶ κτίστηἰι.

INSCRIPTIONS FROM MYSIA. 271

The titles σωτὴρ καὶ κτίστης were often given to Hadrian by grateful communities. Compare 6... Perrot, La Galatic, No. 59, and Le Bas, No. 1721a.

The great inscription built into the walls of Apollonia (Hamilton, No. 304, Le Bas, No. 1068), is a testimony to the munificent patronage of Hadrian,

12, Ulubad (Lopadium): in the café. Fragment of marble block, broken on all sides. Letters about 1 inch.

AKAIIEPHZAMENUNKAIN, NTATQANIZBAL TANEYTE AIIEPOMNHMONIIEZEANTAAI TEL TATAKAILE TPATUII: ΤΊ λ καὶ ἱερ[ευ]σάμενον καὶ ν[εωποι- -noalyta τῶν Σ[ε]βαστῶν εὖ τε [καὶ καλ- -ῶς κ]αὶ ἱερομνημον[ἤσαντα δὶΪς ev- -πρε]πέστατα καὶ στρατηγήσαντα

Doubtless from Cyzicus, like other stones at Ulubad. See Cichorius, Sitzungsb. der Berl. Akad. 1889, p. 367.

13. Mikhalich: built into the corner of the Ulujami mosque, rather high up. Published by Cichorius, Ath, Mitth. xiv. p. 248. My copy presents such curious differences that I give both versions side by side.

Dr, Cichorius. My copy. ETINOTONAY (a) €TANOTONAV NAIM ον Ἄπολις νὰ μιον ἅπολις έ 6 (Ὁ) On the other face of the same stone. PA TA © δ - ΠΩ Δ

Dr. Cichorius’ suggestion that MTTOAIC stands for Μιλητόπολις is extremely probable, but if my reading is correct, and it is confirmed by Dr. Cichorius’ own reading in (Ὁ), we have to restore Με(ιλητό)πολις, cf. Μειλάτης, Aristides, Or. xxv. p. 499 (Dindorf).

can scarcely mean anything but ὅρος. The initials below I would interpret as Πο(ιμανηνῶν) and Με(ιλατῶν). Stones travel far over the plains of Mysia, as Dr. Cichorius knows. He “has not identified Miletopolis with Mikhalich simply on the evidence of this inscription, but other people have done so. The identification is probable υ

272 INSCRIPTIONS FROM MYSTA.

enough on other grounds (although I cannot convince myself that Melde near Kirmasti is an altogether improbable site for the Milatian town), but the fact that a boundary stone, which once delimited the very wide territories of Poemanenum and Miletopolis, is found built into a mosque at Mikhalich, is obviously insufficient proof of it. Poemanenum seems to have lain far to the west. Probably Miletopolis and Poemanenum enlarged their borders at the expense of the decayed Cyzicus. Perhaps the first two lines of the inscription may be read as ἐπὶ νότον Δυνδέμου (1.6. Δινδύμου) Μειλητόπολις. But until the readings have been verified interpretations are premature.

14. Mikhalich: built into the wall of a mosque, high up. Fragment of marble block. Letters about 1 inch. Published, less completely, by Cichorius, Ath. Mlitth. xiv. p. 248.

=ZTPATO +) SP OTpATO sa INOPANOYMHTPOANPOYASKAI Μηνοφάνου Μητροδώρου ᾿Ασκλ[ηπιαδου \IPE XAIPE ΧΙ! χ]αῖρε. χαῖρε. χαῖρε. ΡΥΚΗΝΝΕΚΎΕΣΣΙΝΚΕ νεκύεσσιν ‘TATPIAQAEXETEIS θυ]γατρὶ δωδε[ κ]ετεῖ QNHIENEZTEOPHN ἤιεν ἐς τέφρην QNNEKYONTIPOYK τ]ῶν νεκύων ΛΔΡΘΕΝΟΣΕΥΓΕΝΕ π]αρθένος εὐγενέϊτης ΟΦΡΑΜΕΤΑΥΘΙ ὄφρα μεταῦθι[ς ~OMENON

The verses are too fragmentary for restoration.

15. Mikhalich: in the pavement in front of the door of the Garibche mosque. Fragment of stone block, broken to left. Large letters.

A KAI ZINOT eres rete ἐὰν δέ τις ΤΟΛΜΗ τολμήσ[ῃ ἕτερον θεῖναι δώ- Σ ΕΙΤΩΦΙ -σει τῷ φίσ[κῳ δηνάρια---.

16. Mikhalich: in the same place as the preceding. Similar stone.

Similar letters. pounder εἰ N A-FEHETEPONI

KI AYTIOXOZEZTAIT XIAZEKAHMATI

INSCRIPTIONS FROM MYSIA.

bo ~I ws

ie AXb)s €repov .... . ὕποχος ἔσται τ[ῷ τῆς τυμβωρυ- -χίας ἐ[γ]κλήματι

17. Yenije Keui, between Mikhalich and Panderma: in the pavement in front of a house. Fragment of a marble sepulchral relief of the Funeral Feast type. Letters (below the relief) about 3 inch, almost effaced.

MENEKPA ΠΕΡΜΟ Μενεκρά[τε]ι ᾿ Ερμο[τίμου

18. Omar Keui, on the verge of the plateau above Panderma: in the cemetery above the village. Milestone of coarse marble, much weathered.

DONN DDNN CONSTANT LVALENTINANVS VICTORI ETFLVALENS VIC TORES SIMPEIAVGG

ETFLCONS M Η

CONSTANTIOET CONSTANTIN B ETNOBILL OBIN COeCTAN

MH

ΜΠ

There seems to have been a third inscription on the same stone, but only faint traces of it remain. We made out an isolated CAKES, but nothing more, by the light we had.

The inscriptions are probably to be restored as :—

(1) D(ominis) n(ostris) (3) D(ominis) n(ostris) ΕἸ. Constant[ino Fl. Valentin[iJanus victori [ac et Fl. Valens [triumfatori] s[e]mpe[r] Aug. = victores s[e]mpe[r] Augg. et ΕἸ. Cons[tantino et M. ἡ.

ΕἸ.] Constantio et

F].] Constanti n[o]b[b. ac (2) et nobil. [Caes.]

flJo[re ]n[tiss. Caes(aribus) Co[n]stan[tio Mi. ἡ. VIII.

I suppose that the fourth line of (1) has been worked into (3) with the substitution of. victores for triwmfatori, and take (2) to be a later addition to (1) with reference to Constantius Gallus.

274 INSCRIPTIONS FROM MYSIA.

The caput viae must be Cyzicus, which is quite ten miles from Omar Keui, so the stone has been carried. The stone probably belongs to the road from Cyzicus to Pergamum via the Macestus valley and Hadrianutherae. Other milestones probably to be referred to this same road have been found at Debleki, some miles to the south of Omar Keui, and at Chamandra, on the left bank of the Macestus, three hours above Mikhalich, 6.1.1. iii. 7179-80 and 463-5.

19. Aksakal, about fifteen miles from Panderma on the Susurlu road : in the fountain. Marble block, 1 foot 10 inches x 2 feet 2 inches. The inscription is over a much damaged relief, in which there is a serpent coiled about a tree. Letters inch. There was more both before and after these lines, but illegible.

KAIZYIEQTTAPOAEITAXAPOIZOTITOYTO TOKOINONEITTAZEMOIXAIPEINENEKEN EYZEBIH>

a 4 5 nr ΄ ee lal \ , καὶ ov [γ]ε, ὦ παροδεῖτα, χάροις, ὅτι τοῦτο TO κοινόν εἶπας ἐμοὶ χαίρειν, ἕνεκεν evoeBins.

20. Panderma: outside the Church of the Trinity. Marble relief, bust of a lady, with inscription below.

YTTOMNHMA “Ὑπόμνημα. “TITIMIA- AYPHAIA-W Σε]πτιμίᾳ Αὐρηλίᾳ ’Q- ΞΛΙΜΗ: ΜΑΡΙΚΟΓ -ΟΥ̓ΑΗΙ -φ]ελίμῃ Μάρκος Οὐαϊλέρι-

ΤΡΟΦΙΜΟΓΤΗΓΛΥΚΎΤΑ -ος] Τρόφιμος τῇ γλυκυτά[τῃ YNBIWZHCACHETHME 5 σ]υνβίῳ ζησάσῃ ἔτη μ' ε΄ IEIAC PIN μυ]είας [χάϊ]ριν.

21. Panderma: in a wall outside the Church of the Trinity. Small marble sepulchral relief of the funeral feast type, broken to left.

NEQNTPODIMHEQE

OYTOYFAYKQNOEIr YNH XAIPE ἜΝ ΡΝ ΔΝ. . . « Τροφίμη Σωσ- -ιβίου θυγάτηρ. . . 2Jov τοῦ Γλύκωνος γυνὴ χαῖρε.

I give Mr. Anderson’s copy, which seems to be more complete than my own.

22. Panderma: in a wall outside the Church of the Trinity. Small marble sepulchral relief, broken above, with two panels: (a) upper panel,

INSCRIPTIONS FROM MYSIA. 275

funeral feast ; (6) lower panel, man riding, followed by a boy who holds the horse’s tail. Letters 1 inch.

<!tVIOKAITIATEPIQN XAIPE

ὁ δεῖνα Διονυ]σίο[υ] ὁ καὶ Πατερίων

χαῖρε.

23. Hammamli Keui, above Cyzicus: in the steps at the door of the mosque. Marble base. Letters 1} inch,

EYDHMOY Εὐφήμου ΤΟΥΜΕΙΔΙΟΥ τοῦ Μειδίου.

24, Edinjik: ἴῃ ἃ yard, Round marble cippus, Letters 11 to 1 inch.

YTOMNHMA Ὑπόμνημα. NYNPEPQTOLOKAINIKANOP Νυνφερωτὸς ὁ καὶ Νεικάνωρ ΝΜΙΙΚΟΠΟΛΕΙ HENEIKHZAZAPENL Νεικοπολείτης νεική[σ]ας "Apews

ΝΕΙΚΑΣΕΩ ΔΕΑΠΟΚΕΙΜΑΙ͂ΤΡΟ νείκας ve’ ὧδε ἀποκεῖμαι. Τρο[φίμη EYNBIOLEKTANIAIQNMEIALXAP ὅ σύνβιος ἐκ τῶν ἰδίων μνείας χάρ[ιν. ΤΙΣΔΑΝΑΔΙΚΗΣΗ ΤΟΝΒΩΜΟΝ τίς δ᾽ ἂν ἀδικήσῃ τὸν βωμὸν ANE ELEIE TANTTOAINXP δώσει εἰς THY πόλιν χρ[ήματα.

XAIPE ETTAPOAEITAI χαίρετε παροδεῖται.

25. Edinjik: an inscription published (΄.1.6(. 11. 3678; by Hamilton, Researches, vol. ii. No. 306; and best by Lolling, Ath. Mitt. ix. p.19. The marble block on which it is written has been bored through, and now forms the mouthpiece of a well.

Lolling’s interpretation runs :—

πολ]͵]λάκις [τε τειμηθεὶς πρ]οβλ[ηθέντων πολλ]ῶν ἄλλαις τε τει[μα]1}ς] κ[ αὶ μ]υσταρχίαις πολλαῖς.

My copy, which shows rather more than Lolling’s, confirms his interpre- tation in all but two points. (1) I doubt whether there is room for the te in the first line, and suspect that it may have crept in from the second. (2) I read the end of the first line as Y . OBA, which is inconsistent with Lolling’s restoration. Considerations of space are also against it. Hamilton read the last letter as A. I would suggest [τειμηθεὶς] ὑ[π]|ὸ AlacliAé]wv x.7.r. The inscription belongs to Cyzicus, and the βασιλεῖς at Cyzicus seem to be sacred officials of the μύσται. See for example C.Z.G. ii. 8663, with Boeckh’s notes.

276 INSCRIPTIONS FROM MYSIA.

26. Chepne, near the north-west corner of lake Manias: near the well. Small marble cone with a disc-like top, standing about two feet out of the ground. Letters 1} inch, very irregular.

+ WPOOECIA YAPEWNEI! -AIWSTITAZIC OEWZOSOV A PEAT IC ὡρο[θ]εσία is plain, and the second word may be Ὕδρέων, but I can make nothing of the rest. f

27. Eski Manias: in the south wall of the castle, rather high up. Large marble block, upside down. Published, but from incomplete copies, by Dorigny, Rev. Arch. xxxiv. p. 106, and by Mordtmann, Ath. Mitth. xv.

p. 156. OIENTHIASIAIAHMOIKA////AEOW//

KAIOIKATANAPAKEKPIMENOIENTHITIP//// POMAIOYEOIAIAIKAITQNAAAQNOIEIPE// MENOIMETEXEINTONZEQTHPIQN KAI 5 MOYKIEIQN δ)ιᾶΈετιμησαν APOETPATONAOPKAAINNOZANAPAALAOONTE NOMENONKAIAIENENKANTATIIZTEIKAIAPETH KAIA////AIOZYNHKAIEYSEBEIAIKAITIEPITOXON ZYNDEPONTOSTHNTTAEISTI//NEISENHNED ME ‘10 NONETTOYAHNKAITIOAAAKAIMELr AAATIEPITIO// HEANTATOIZKOINOIZ TOYZ YNEAPIOYTIPATMA ZINTQANITPOZSAOZANKAIMNHMHNAIQONIOI . ANHIKONTOQNAPETH//////////////IKAIEYNOIAS THZEEZEAYTOY Οἱ ἐν τῆι ᾿Ασίαν δῆμοι κα[ὶ τ]ὰ ἔθν[η καὶ οἱ Kat’ ἄνδρα κεκριμένοι ἐν τῆι πρ[ὸς “Ῥωμαίους φιλίαι καὶ τῶν ἄλλων οἱ εἰρη- -μένοι μετέχειν τῶν Σωτηρίων καὶ Μουκιείων ἐτίμησαν “Ἡ]ρόστρατον Δορκαλίωνος ἄνδρα ἀγα[θ]ὸν γε- -νόμενον καὶ διενένκαντα πίστει καὶ ἀρετῆ[ι καὶ δ[ικ]αιοσύνη[ι] καὶ εὐσεβείαι καὶ περὶ ταῦ κο[ι]ν[οῦ συνφέροντος τὴν πλείστ[η]ν εἰσενηνεγμέ- 10 -νον σπουδὴν καὶ πολλὰ καὶ μεγάλα περίπο[ι- τήσαντα τοῖς κοινοῖς τοῦ συνεδρίου πράγμα- -σιν τῶν πρὸς δόξαν καὶ μνήμην αἰώνιοϊν ἀνηκόντων ἀρετῆΪς ἕνεκεν] καὶ εὐνοίας τῆς εἰ ἐς] ἑαυτούϊς.

or

INSCRIPTIONS FROM MYSIA. 277

The first five lines, which form the heading, are rather shorter than the rest. There is not space for Dr. Mordtmann’s addition καὶ ai πόλεις to the first line. In line 6 Dorigny and A. D. Mordtmann were no doubt right in reading the first letter as ἢ, In line 8 I am inclined to trust my copy for the combination of Ὑ and K, for which there is a reason in the lack of space. Owing to the position of the stone or the light upon it there is some difficulty in detecting the letter |, a fact which I noticed at the time of copying and find confirmed by a comparison of the three versions,

The Μουκέίεια are the festival in honour of the famous Q. Mucius Scaevola, governor of Asia about 98 B.c. Cp. Cicero, im Verr. Act 11. 11. 21 51, Mithridates in Asia..... Mucia non sustulit.

It is, I think, probable that Herostratus was the agent sent by Brutus to Macedonia (Plutarch, Brut. 24, εἰς Μακεδονίαν ἔπεμψεν ᾿ΗἩρόστρατον οἰκειού- μενος τοὺς ἐπὶ τῶν ἐκεῖ στρατοπέδων), and the inscription may date from the spring of 42 B.c. when Brutus was in Asia organizing the province in his interest and equipping a fleet at Cyzicus, (2b. 28).

I have elsewhere given reasons for rejecting the idea that Poemanenum was at Eski Manias, and for believing that the inscribed stones there have been brought from Cyzicus.

The mention in an inscription which is evidently of pre-Augustan date of a συνέδριον representing the communities and associations in Asia is of some interest as bearing on the origins of the κοινὸν τῆς ᾿Ασίας. In the somewhat vague and loose expressions περὶ τοῦ κοινοῦ συνφέροντος τὴν πλείστην εἰσενηνεγμένον σπουδὴν and πολλὰ καὶ μεγάλα περιποιήσαντα τοῖς κοινοῖς τοῦ συνεδρίου πράγμασιν, there is nothing to invalidate the view so lucidly set forth by Dr. Brandis in the new edition of Pauly’s Real-encyclopddie, pp. 1556-7, that, before Augustus, meetings of representatives of the whole pro- vince were merely occasional, to transact a particular piece of business which happened to turn up and seemed to require common action. Yet to my mind the set formulae of the heading (which recur in other inscriptions), the πολλὰ καὶ μεγάλα ascribed to Herostratus, the comprehensive τοῖς κοινοῖς πράγμασιν, and the familiar reference to τὸ συνέδριον, suggest that the sessions of the assembly were no longer extraordinary, but habitual, if not regular, and its business was already enlarged beyond special occasions to current affairs of general interest.

On other points it is enough to refer to the comments of M. Dorigny and Dr. Mordtmann, and to the parallels quoted by Dr. Brandis.

28. Kassaba Kirmasti. An inscription published in Le Bas and Waddington No. 1764 a, by Perrot, La Galatie No. 63, and by Cichorius, Ath. Mitth. XIV. p. 251.

Λεωνίδης κὲ ΓΑμίμ]ιον οἱ γοναιεῖς ᾿Αλεξάνδρῳ

τέκνῳ καταθυμίῳ μνήμης χάριν: (leaf)

ὃς ἂν τὴν στήλην ἀφανίσει ἢ τὰ ἐν αὐτῇ γε-

-γλυμ[μ]ένα ἢ μετάρῃ αὐτὴν, αὐτὸν ἐξώλη καὶ γέ- ᾿ -νος αὐτοῦ.

278 INSCRIPTIONS FROM MYSTIA

The Γ in γοναιεῖς is still legible although the stone is damaged. Cichorius is perhaps right in explaining this strange form as a mistake of the stonecutter’s, who wrote ΑἹ and then corrected it by adding ΕἸ, or possibly (I think) doubled the N into ΝΑΙ. Cichorius has done good service in getting rid of the place-name ‘Oeonaea,’ into which Waddington was betrayed by Vernouillet’s copy, and has corrected some other errors. But he has introduced some fresh ones of his own. He has substituted @ for Q throughout. In line 2 he has changed μνήμης into μνείας. In line 3 he has omitted the final re, And he has tacked line 5 on to line 4.

The grammatical constructions of the imprecation are not elegant. With the second clause we have to understand γενέσθαι.

The sketch of the stone with its two reliefs given by Perrot is fairly good, I have compared it with a photograph and find that the only impor- tant omission is a wreath in the right hand of the seated figure facing to right in the upper panel.

29. Kestelek: in the yard of the old konak. Thick marble slab, forming the lowest step to a wooden stair. Letters 3 inch. Above the inscription is a panel 164 inches long, 10? broad, containing a relief. A male deity stands en face wrapped about the waist in a himation. He holds a spear in his left hand, a patera in his right. | Opposed to him is an eagle on an altar, in front of which a servant sacrifices an ox, while another behind it bears a tray of offerings. Behind the eagle stands a male figure in an attitude of adoration. Only the ends of the lines of the inscription are legible.

KANQKAIFO

ENIAZIAAI PPOYZTPATH ΑΣΤΗΣΕΙΣ

90. Kestelek: in the yard of a house. Marble stele broken below. The inscription is between two reliefs; the upper is a funeral feast of man and wife carved in a square panel surmounted by a gable, in which is an altar, and flanked by fluted Ionic columns; the lower, under a round arch, is broken away, but traces of two heads remain. On the top of the arch is a small male figure carrying a big goblet, at each side is a small female figure. These figures interrupt the inscription, which is irregularly engraved over an earlier epitaph imperfectly erased. The upper panel measures 1 foot 5 inches high, and 1 foot 7 inches broad including the columns. Letters 1 inch.

AAAAOYTATHP OTAKIAIOZ

Elder AZKAA A.NHP Younger Girl. Boy. AAAAS Girl. AEKMOS CNAIZ

The names seem to refer to the figures. Adda is the wife at the foot of the couch, Oraxddvos ἀνὴρ Λάλας her husband reclining, θυγάτηρ "Ασκλα

INSCRIPTIONS FROM MYSTA. 279

seems to be the left hand figure over the arch, Γναὶς the right hand figure, and Aéxpos the boy with the goblet.

31. Narlinar: beside a fountain in the village. Fragment of marble block with moulding above, broken on both sides and below. Letters 1 inch.

YTATPIAI Τῇ ἑαυτο]ῦ πατρίδι GPEREDRO! ‘orth... Jos ᾿ΕΓχ]έφρο[νος a ee νοὶ Ἰίτης , ΓΑΙΒΑΙΟΜ ΥΕΣΟΥ IKAIENO ΓΕΜΑΙΡΙῚ ΙσΣστιαδα α THMEP (EMT \PRIKI AUK.

an

32. Beyje: in the pavement of the main street. Marble slab, broken above and to right. Letters about 1 inch.

Mas ae

TWEA*1 τῷ; ἑαυτ[ῆς συνβίῳ 1

MAKPOBIWO¥ MaxpoBip....

THINXAPINEXOMEN τὴν χάριν ἐχομέν[η 1

ὙΠΕΡΠΟΛΛΙΙΓΤΕΙΜ ὑπὲρ πολλῆς τειμ[ῆς, MNEMECXAPIN μνήμης χάριν.

33. Tavshanli: in the Armenian cemetery. Marble sepulchral stele of the door type. The door is between fluted pilasters surmounted by an arch and gable. In the tympanum of the arch is an eagle, in the gable above, a floral pattern. The door has six panels. In the left uppermost panel is a tablet, in the right a wreath and a keyhole; the middle panels are filled with a geometrical design, a circle in a lozenge; the bottom panels represent gratings. The inscription runs round the face of the arch. Letters } inch.

AAMOZENOSTIAIAEYTHZEEAYTQIZQN Aapokevos παιδευτὴς ἑαυτῶι ζῶν.

The writing tablet doubtless refers*to Damoxenus’ profession, — The prevalence of the door type of tombstones on the upper Rhyndacus

280 INSCRIPTIONS FROM MYSIA.

and its rarity to the west of this region indicate a difference in beliefs about the dead, and this difference of belief probably indicates a difference in population. The dividing line is marked by the rock Tomb at Delikli Tash about two hours west of Tavshanli, the westernmost of the great Phrygian tombs, And the Delikli Tash tomb is itself a testimony to the antiquity of the type in the locality. The plains of the upper Rhyndacus belong geographically to the Phrygian country to the east. They are cut off from the Mysian country to the west by the rugged district enclosed between the Rhyndacus and the Macestus.

The great number of tombstones of this type at Tavshanli and Moimul naturally leads one to suspect the neighbourhood of some important ancient city, but I am not convinced that the stones may not have been brought from Aezani, where the same type prevails.

34. Tavshanli: in the Armenian cemetery. Pillar of coarse marble. In the top, which is cut square, is an arched niche with fluted pilasters, containing a relief of a male figure standing en face, with long hair, and a staff in his left hand. The first inscription is poorly cut below the niche. Beneath it is an ornate capital of late type, and on the shaft below is the second inscription. Letters inch and 1 inch respectively. (a) is broken to right, (Ὁ) to left.

(a) YTEPTHS OYKY CW THPIACAOYKIO AIOYITAPAAAA XPHCT ATEAl = [APOZHNWEY.

“Ὑπὲρ τῆϊς τ]οῦ κυ[ρίου 1 σωτηρίας Λουκίου [. ..

-δίου Ἰ]αρδαλᾶ χρηστο[ῦ ἀ[π]έϊδωκεν] 'Ιαροζήνω εὐχ[ὴν

KAITOY Avi CANECTHCE καὶ τοῦ! ..... Ἰς ἀνέστησεν. (0) HNOTENOYC ... Z]nvoyévous ENOYC ... ylevous

35. Tavshanli: built into a fountain in the bazaar. Marble tombstone of the door type, with the inscription on one of the panels. Letters about 1 inch.

EB=PMHAPTEM “E(rous) B’E'p’ μη(νὸς) ᾿Αρτεμ- EICIOYZK.ACKAHTTIA -εἰσίου On’. ᾿Ασκληπιά- AHCKAITIMOOEOLCKAI -δης καὶ Τιμόθεος Kai AEKMIOCAIODANTOLKA Δέκ[εμ]ος Διόφαντος κα- ΙΕΥΝΟΙΓΤΑΤΙΩΜΗΤΡ -ἰ Εὔνοις Τατίῳ Μητρ- ΟΔΩΡΟΥΜΝΗΓ. -οδώρου μνή[ μη]ς

ΕΝΕΚΕΝ ἕνεκεν.

INSCRIPTIONS FROM MYSTA. 381

The stone-cutter has shifted the circumflexes over the numerals, one letter to the left, transposed ἐμ in Δέκεμος, and omitted μὴ in μνήμης.

Assuming that the Sullan era is used, and that the mouth Artemisios corresponds to February (see Ramsay, Cities and Bishoprics of Phrygia, vol. i. p. 204), the date is Feb. 27th, Α.}. 78.

36. Tavshanli. Marble stele of the door type, now converted into a fountain. Inscription round the arch. Letters 1 inch. TIPEIMOCKAITATEIGNMHTPO IIpetuos καὶ Ἱ᾿Γατείων Mnrpo- III L/MNHMHC>////P1N -δώρωι] μνήμης [χά]ριν. 97. Tavshanli: built into a fountain in the bazaar. Marble architrave or lintel, broken at both ends. Letters about 2 inches. IKAICFENETHP .. ALYIEIE//ECK - καὶ yeveThple κ]αὶ υἱεῖ.

Probably Christian. _

38. Moimul. Stele of the door type, with double gable. Published by Perrot, Za Galatie, etc., No. 68.

Left arch. MIOPHCTATIWFYNAIKICWIA Right archh MNHMHCXAPINKAIEIAIOYW

MéOpns Τατίω γυναικὶ Σωίᾳ μνήμης χάριν καὶ εἰδίο ὑῴ.

Perrot has TATNO for TATIW, COIA for CWIA, and |W for YW, 39. Emed: in the east cemetery above the town. Square limestone

block, a little chipped at both sides near the top. Letters about 1 inch Beneath the inscription is a wreath.

OYOCKMOYEMH ΜΊόψος κὲ Mois Mn- ΖΖΏΩἧΩῪὝἯΝΙΚΗ -κιο]νίκῃ PAYKYTATH γλυκυτάτγ.

40. Emed: in the east cemetery. Square limestone base. Letters rather over 1 inch.

¥TIEPLTOMA ὑπὲρ στομά- XO¥E¥XHN -you εὐχήν.

282 INSCRIPTIONS FROM MYSIA.

Beneath the inscription is a much damaged relief of a female figure en face holding a long torch in each hand, a familiar type of Hecate or Artemis. It appears from the following inscription that vows were paid at Emed to a deity whose name begins with A. Combining the two pieces of evidence we may infer that the dedications are to Artemis, Doubtless the names of the deity and the dedicator appeared elsewhere on this monument.

The inscription evidently records a medicinal cure. Now at the lower end of the town of Emed there is a copious hot spring of very high tempera- ture, slightly impregnated with sulphur. It is natural to suppose that the cure was effected by the medicinal properties of this spring, which must therefore have been sacred to Artemis. An exact parallel is to be found in the hot springs on the Aesepus, to which the orator Aristides resorts in a similar case, These springs were sacred to Artemis Thermaea, and Mr. W. M. Ramsay has very plausibly conjectured that the place at which they were situated is to be identified with the Artemea of Hierocles. See Aristides; 11.1.. iv., Dind. vol. i. p. 508, and Ramsay, Hist. Geogr. p. 154.

41. Emed: in the east cemetery. Square block of coarse marble, broken at the upper corners.

MNO ANH//////// ᾿Αριστ]οφάνηϊΪς κὲ An- /I//ATPLOCA//////// -μ]ήτριος ᾿Α[ρτέμιδι €YXHN εὐχήν.

Compare the preceding inscription.

42. Emed: in the east cemetery. Square block of coarse marble. Beneath is a mutilated something in relief. Letters 1 inch.

AAMACKTASIA Aapas κὲ Ταθια- ΝΗ(ςΓςΥΕΙΩῶΩΤΡΟΦΙ -νὴ«ς; νεΐἱῷῴῷ Τροφί- M@QMNHMHCXAP -μῳ μνήμης χάρ-

ΙΝ τιν.

The ς added to Ταθιανὴ seems to be a mistake of the stonecutter.

43. Emed: in the east cemetery. Limestone stele of the door type, inscribed round the arch. Letters 1} inch and # inch.

AWCIMOLCKTPODIMOCM////////A //////|/[AKEAQHMNHMHCX APIN

Δώσιμος κὲ Τρόφιμος M... a ἀδελφῇ μνήμης χάριν,

INSCRIPTIONS FROM MYSIA. 283 44. Kmed: in the east cemetery. Square block of coarse marble Letters 1 inch. TPODIMACEDELIQ

MNHMHCXAPIN Τροφιμᾶς ᾿φεσίῳ μνήμης χάριν.

45, Emed: in the east cemetery. Square block of coarse marble. Letters 1 inch,

ATOAAW ᾿Απολλω- NI AIG, -νίδ[ ns MENE Meve MAXW -μάχῳ ΠΑΤΡΙΆ ὅ πατρὶ μ- NHMHC -νήμης ΧΑΡΙΝ χάριν.

40. Emed: in the east cemetery. Square block. Letters 11 inch.

TEKNATIATPI Τέκνα πατρὶ KAIOEGTIAAHN καὶ θεῷ τιμήν.

47. Near a hot spring with baths not far from Yenije keui, south-west of Emed. Limestone stele of the door type, broken above. The inscription runs round the arch, and in the tympanum are carved a basket, a mirror, and acomb. Letters 1 inch, rudely cut.

Left of the arch. Light of the arch. PIOUS ee a en. gy ΜῊΝ SORT PUL Paadscos Mai. δὲ aseunlt- ᾿ἰοϑαύνονπι ρα ψή ans gape.

48, Assarlar: built into the side of a fountain. Letters 1 inch.

\¥NIANIK¥NTIAN - ΚἸΊυντιανὴ Κυντιαν[οῦ ¥TIEPTEKEWCANEHE ὑπὲρ Téxews ἀνέθηκε ΓΑΕΚΑΤΑΚΑΙΜΑΝΙΤΟΝ τ] “Exata καὶ Mavi τὸν /IEATWHAIANOL v]iéa τῷ LLacavos.

This inscription is an important addition to our scanty knowledge of the religious beliefs of North Phrygia. M. Paul Perdrizet in a recent article on Men (Bull. corr. hell. xx., 1896, pp. 55-106), can only quote one little dedica- tion to Men (Mnvi εὐχήν) from North Phrygia. He gives, however, an excellent photographic reproduction of the well known relief in the Imperial

284 INSCRIPTIONS FROM MYSIA.

Museum at Constantinople, which is almost certainly to be referred to this same district. It bears the inscription”“Aayuov τὸν ἑαυτῆς σύνβιον Taevov κατεειέρωσεν Σωτήρῃ ᾿κάτῃ" «.7.r.,and on it Men is represented standing side by side with the triple Hecate. This relief and inscription together form a useful commentary on our text. A still closer parallel for the con- secration of the child to Hecate is furnished by an inscription of Cotiainm (C.1.G, 3827 4, Le Bas and Waddington 805) Σωτεέρῃ[ς] “Exarn [Τρόφι]μος Ke "Addn Δημοσθένη τὸν ἑαυτῶν υἱὸν τει[μη]θέντα ὑπὸ Σωτείρης ‘Exarns κατειέρωσαν. Σώτειρα is at once the Goddess of birth and of death. She appears also as Hecate triformis on coins of Apamea (Head, Hist. Num. p. 558).

Men is often found associated with a Goddess, especially with Artemis Anaitis. They are one of the numerous pairs, Goddess and God, so common in Asia Minor. Their names vary, even in the same place, but their persona- lities remain fundamentally the same. It is likely enough that the Hecate of our inscription is one with the Goddess at Emed on the other side of the valley, to whom we saw reason for giving the name of Artemis, and perhaps Men is not to be distinguished from Apollo whose festival in the grove is mentioned in another inscription, of public and official character, at Assarlar (6.1.6. 3847 b, Le Bas and Waddington 1011). It may be noted that the coin-types of Tiberiopolis usually refer to the worship of Artemis (Head, Hist. Num. p. 570).

The relation between the God and the Goddess is variously, but in- differently, conceived, now as male and female, for the words husband and wife, father and daughter, cannot be appropriately applied, now as mother and son. How was it conceived at Assarlar? Ramsay has very plausibly placed Tiberiopolis at or near Assarlar (Hist. Geogr. pp. 146-7), and has brought into connection with this identification the inscription, now at Egriguz a little down the river, in which mention is made of the ὁμοβώμιοι θεοὶ Σεβαστοί. The imperial mother and son, Livia and Tiberius, took the place, Ramsay argues, of the divine mother and son. (See 6... 3847 m, Le Bas and Waddington 1021, and compare several inscriptions of Aezani.) This conjec- ture is confirmed by our inscription. Hecate is probably regarded as the mother of Men, for express mention is made of his father Paean.

The words τῴ Ilasavos come in rather awkwardly at the end. They seem to be an afterthought, added perhaps by a different hand, for the straight-barred A is here changed for A, We may compare the way in which Bevvet Σοηνῶν is added at the end of an inscription of Altyntash (CLG. 3857 1, cp. Ramsay, Hist. Geogr., pp. 144-5). The form of appellation Μὴν ὁ Παιᾶνος is, to say the least of it, rarely applied to deities. Is it intended to distinguish this Men from others? or to justify his invocation in childbirth as the son of the Deliverer? I can find no other reference to the parentage of Men, although Manes, whom Ramsay (rightly, I believe) regards as the original Men, was the father of Acmon and Doias, the eponyms of Acmonia and Doiantos Pedion (Ramsay, Cities and Bishoprics of Phrygia, vol. 11. pp. 625-6, with the references there given), Paean is no doubt, as Mr. Ramsay

INSCRIPTIONS FROM MYSIA. 285

first suggested to me, a Hellenizing name for the native Sabazios. He was doubtless also known as Asclepios, who appears on the coins. Asclepios Soter and Hecate or Artemis Soteira would be the God and the Goddess, Men the Son, who is not ultimately distinct from his father. The nomenclature of the district is mostly derived from Asclepios, Meter, Artemis, Men, and Apollo,

On the whole subject of the religious conceptions prevalent in Asia Minor see especially Ramsay’s Cities and Bishoprics of Phrygia. I have followed Ramsay with confidence rather than Roscher or Perdrizet, whose traditional methods seem to me wholly inapplicable to Anatolian mythology, whereas what small experience I possess entirely supports Ramsay’s principles of interpretation.

49. Assarlar. Gable of a stele of the door type. The inscription runs round the arch as usual. It is broken at both ends. In the field of the gable is carved a pair of dolphins.

CHETVAAIOCCY//////// \PQMNHM X//// \PIN

«σῆς Τύλλιος συ[ντρ]όφῳ μνήμ[ης χάριν.

50. Assarlar: near the mosque. Square limestone stele, with vine branches carved in the side panels. Published in Le Bas and Waddington, vol. i. No. 1012.

The number of the date as published is CNTF, but Waddington notes that Le Bas’s manuscript has ΝΕ. My copy has CNE.

There are two more lines after ᾿Αχιυλλεῖ :-—

ἈΠ AAW μ[νή]μης II/IITIN χάρι]ν.

51. Near the north-east shore of the lake of Simav: in a fountain by the wayside. Limestone stele of the door type. In the panels, besides a mirror and other objects, there is a little stele carved, which bears inscription (Ὁ). Letters 1} inch.

(a) YMTTIANOCAAEZANAPWAAE | KCABEINIANHENATPIMNEIACXAPIN

᾿Ολ]υμπιανὸς ᾿Αλεξάνδρῳ ἀδελφῷ κὲ Σαβεινιανῇ [θυγ]ατρὶ μνείας χάριν. H.S.—VOL. XVII. x

286 INSCRIPTIONS FROM MYSTA.

(b) KMAPKI xe Μαρκι- ANWA -av@ ἀ- AEA -δελ- ΦΟΥΥ -φοῦ υ- IWMN -ἰῷ μν- ΕΙΑΝ -είαν.

52. Near the north-east shore of the lake οἵ Simav: in a fountain by the wayside. Limestone stele of the door type, with two doors under two wreaths, and in the tympanum a basket. Letters about 1 inch.

AZKAHTTEIAAHCE YNEKEIMNH MHEXAPIN

᾿Ασκληπειάδης γυνεκεὶ μνή- -μης χάριν.

53. Saujilar: in the north wall of the old mosque to the east of the village. Marble stele with gabled top and pilasters, between which is a garland. Letters about ? inch.

ETOYEPMRMTIANHMOYAIOIZYNHOEIS PIAOIETIMHE ANAIOLENH Hi//-YPOY τους ρ΄ μ΄], w[n](vos) Πανήμου 8'¢, οἱ συνήθεις φίλοι ἐτίμησαν Διογένη οὐνύφου.

There are traces of a small H over the Μ of μηνός.

Assuming the use of the Sullan era the date will be a.p. 58.

54, Sdujilar: in a hedge on the south side of the village. Limestone stele, square, with acroteria. Letters about 14 inches.

AYPIEPOKAHC ST: Αὐρ. Ἱεροκλῆς γ' BACCIANOC ΤῺ Βασσιανὸς τῷ CAYKY TATQATTAT PI yAUKUTUT@ πατρὶ AY PIEPOKAEI€ Ta AP Αὐρ. “Ἱεροκλεῖ [8] τῷ ap- ΧΙΑΤΡΩΓΥΝΚΑΙΤΗ 5 -χιάτρῳ σὺν καὶ τῇ

ΑΔΕΛΦΗΑΥΡΒΑΓΓΗ ἀδελφῇ Αὐρ. Βάσσῃ MEIALE XAPIN μνείας «“ς;» χάριν.

INSCRIPTIONS FROM MYSIA. 2387

In line 4 there can be no doubt that β΄ ought to be read. Aur. Hierocles III. must be the son of Aur. Hierocles II. On the use of Aurelius as praenomen see Ramsay, Cities and Bishoprics of Phrygia, vol. ii. pp. 388-9. By ἀρχίατρος we probably have in this connection to understand the public officer of health for the district. See Pauly-Wissowa, Real-Hncycl.

The inscriptions at Saujilar are probably to be referred to the neigh- bouring Ancyra rather than to Synaos, which is farther off.

55, Saujilar: in the wall of the cemetery. Limestone sfe/e. Letters about 1 inch, rudely inscribed.

ETOCTNZM

HNOCAPTE Willie TOVYEKT | -μεισί]ου [ 1κ΄,..

4AAIAYPHCYNECIC .. . Adpn. Σύνεσις

APKAPTEPIWEAIACIN A[v]p. Καρτερίῳ ε[ἐ]δί[ῳ] σ[υ]ν-

Biw ANH -βίῳ μνή-

MHC +A ht -μης χάρι-

Ν -Ψ.

"Ε το[υ]ς τίν΄ μ- -ηνὸς ᾿Αρτε-

Assuming the Sullan era the date would be the spring of Α.0, 273.

56. Siujilar: in the wall of the cemetery. Marble stele with pediment. Letters 1} inches.

AAEZANAPAAAE ᾿Αλεξάνδρα 'AXe- =ANAPSITIATPI -ξάνδρῳ πατρὶ ΜΕΙΑΝ μνείαν.

57. Siiujilar: in the wall of the cemetery. Square limestone stele, broken below and on both sides, Letters 1} inches.

CA ἼΣ Ε K Md A 9A | Ta τέκνα Ai- ©) Co (dl | TTA ii fo -θωνι sage Mel | A ("ol CY μνίαν.

288 INSCRIPTIONS FROM MYSIA.

58. Baddelu, west of the lake of Simav: in the wall of a house. Small marble stele with gabled top. Letters 1 inch.

ΔΗΜΟΣΘΕΝΗΣ Δημοσθένης ZYNMHTPAAAEA σὺν Μητρᾷ ἀδελ- ΦΩΑΜΜΙΆΜΗΤΡΙ -φῷ ᾿Αμμίᾳ μητρί.

ὅθ. Baddelu: in the wall of a house. Marble stele with gabled top. Below the inscription are carved a basket, a mirror, a vine, and a pruning: hook. Letters 1inch. The first line is on the lintel above the panel.

[///AZIANEAWAE Mlatia νέα ὧδε KAAAKEIMAIEYN καδάκειμαι [σ]ύ[μ- MOIPOCTEKNW -μοιρος τέκνῳ. ZWCLIMOCFAYKYTA Ζώσιμος γλυκυτά-

9 ΤΗΓΥΝΑΙΚΙΚΑΙΤΕΚΝ ὅ -τῃ γυναικὶ καὶ τέκνῳ MANIACXAPIN μνίας χάριν.

60. Assar keui, south-west of the lake οἵ Simav: built into the corner of a house. Marble block with moulding to right. Possibly incomplete above. Letters about 1 inch, but irregular.

~-AHMENGE ᾿ΕἸλήμενοϊς 1 ΕΝΑΘΑΝΑΤ ἐν ἀθανάτ- ΟΙΓΚΑΤΑΚΕ τοις κατάκε- ITAITOY TO -ttat. Τούτῳ AQPONEAOK 5 δῶρον ἔδωκ- ESEOCBO -ε θεὸς βο- H®EINTHE -ηθεῖν τῇ ἐ- AYTOYTTA -αυτοῦ Tra- ΤΡΙΔΙ -τρίδι.

An interesting Christian inscription. One would like to know in what ways this ‘gift of helpfulness’ was exercised: probably Deo dante dedit.

61. Assar keui: in a wall. The top of a gabled stele. Letters 14 inch.

AHMOZOQENHEKAEOMAXC//// Δημοσθένης Κλεομάχζου.

62. Yemishli: in the wall of the roadside cemetery. Marble stele with gable. Letters about ξ inch, much worn,

INSCRIPTIONS FROM MYSIA. 289

MAN TELACATO//ATIATPI AMANELACXAPIN

. tea [Ἰ]σαγόρᾳ πατρὶ μνείας χάριν.

63. Yemishli: in the wall of the roadside cemetery. Gabled marble stele on which are carved a comb, a wreath, and a mirror.

EPMHEryY “Ἕρμης γυ[ναικὶ Ν] 111]

64. Tash keui, about midway between Simav and Balat: in a field half a mile below the village to the south-west. Square marble base with moulding above and below. The inscription is framed in a~panel of incised lines. Letters } inch.

AIIMANAHHKN Aut Πανδήμῳ EYXHNAAMAL εὐχὴν Δάμας ΑΠΠΑΓΥΝΑΚΜ ᾿Αππᾶ σὺν ’Ap- MIQTHI¥NALIL -μίῳ τῇ γυναι- KIEKTQNIAIQN 5 -κὶ ἐκ τῶν ἰδίων ΔΝΕΓΤΉΗΓΕΝ ἀνέστησεν.

The title Πάνδημος is given to Zeus elsewhere, 6.9. at Synnada (Head, Hist. Num. p. 569). It doubtless expresses in each case the religious aspect of the political unity of the community. Now Tash keui obviously lies in. the district of Abrettene (Strabo, 574, 576). We may therefore identify this Zeus Pandemos with the Zeus Abrettenos to whom the robber chieftain Cleon was priest in the time of Strabo. It is possible that Tash keui was a main centre of his worship, for there are traces of a considerable sanctuary in the field in which the stone lies, down in the river valley below the village.

65. Tash keui: in the same field. Marble base, 2 feet square. Letters 14 inch. OA¥MTTIOAQ RQI AAEZ ANARO¥ HRQI

Ὀλυμπιοδώ- -ρωι ᾿Αλεξάνδρου ἥρωι.

290 INSCRIPTIONS FROM MYSIA.

66. Russiler: in the street, near the west end of the village. Square marble stele, broken to right. Letters 14 inch.

EATIICEICTI// Ἔλπις Eiot[p- |ANS2CYNBI//// -ιανῷ auvBil CAYK////F A//// γλυκ[υ͵τάϊτῳ MNHMHCX//// μνήμης χίά-

ΡΙΝ -ριν.

67. Balat: at the Church of S. Demetrius. Square limestone base, broken below. Letters 14 inch. The first liné is on the moulding above.

ATAQHI THXHI ᾿Αγαθῆι τύχηι. HBOYAEFKAIOAIH Ἢ βουλὴ καὶ ὁ δῆ- MOZAAPIANEQKE -μος ᾿Αδριανέων ἐ- ΤΕΙΜΙΣΕΝΚΑΣΑΤΟΡ -τείμησεν K[A]. Σατορ- NEINANOYTATEPA -νεῖναν θυγατέρα ΤΙΚΛΑΥΛΟΥΠΕΣΣΙΑΝΟΥ Τρ χα δῆ δεσσιανοῦ,

CYNAIKAFIOYA:AIAIA γυναῖκα I". Tour. Δίλια- NOY KAEANAPO -νοῦ Κλεάνδρου

8.4 6 8

The stone may have been brought from Hadriani (Beyje), although the distance is considerable. Both the forms, AAPIANEQN and AAPIANQN, appear on the coins (Head, Hist. Num. p. 455). On the other hand there is some difficulty in believing that a heavy block (and that not a marble one) was transported about 40 miles over bad roads and mountainous country. Ancient remains exist at Balat (notably the great sarcophagus figured in Le Bas), but the name to be assigned to the ancient town is uncertain. Ramsay suggests Neo Caesarea (Hist. Geogr. p. 181). Did the territory of Hadriani once extend to Balat, and was the bishopric of Neo Caesarea carved out of it? Or is it possible that there was after all an Adriania, the Adraneia of certain of the Notitiae, distinct from Hadriani? This question is discussed by Ramsay, Hist. Geogr. pp. 160-1, and the possibility gives a certain interest to our inscription.

68. Balat: in the court of the priest’s house, opposite to the Church of S. Demetrius. Square limestone base. Letters 1} inch.

MAYPHAIQKPITS M. Αὐρηλίῳ Κρίτω- ΝΙ ΤΘΑΓΑΘΩΚΑΙ -νι τῷ ἀγαθῷ καὶ ΑΣΥΝΚΡΙΤΩΘΡΕ ἀσυνκρίτῳ θρέ- YANTIOIATTEAEY -ψαντι οἱ ἀπελεύ- ΘΕΡΟΙΑΥΡΗΛΙΟΙΦΙ -θεροι Αὐρήλιοι Φι- ΛΗΜΩΝΚΑΙΠΡΑΞΖΙ τλήμων καὶ IIpaki-

ΑΣΖΩΝΤΙ ας ζῶντ

INSCRIPTIONS FROM MYSIA. 291

69. Kebsul; outside the iosque, in the street. Large marble base Letters 1} inch.

MAI ZHONAMEIAIKIO CKTANE MOIPHAPFIKAY “QNTIA@IHZAYAMEN NOAAAMOANENOEL AOYXOULIOILIAY REX ONAITEINTIAIAE EPO NEAAOPOICEIAEOE EICFNONAITON Mo BIONQTTAPOAFI TAM MENMOIEXMAAOZO OCOYNOMAMEIAIACE - = AEQEOYCANAYCAK MAQANATOICIM EIMI ΟΓΓΟΥΓΓΑΡΦΙΛΕΟΥΓ NEOIONCKOYEINATIAI ree

> ᾽ Ν > Ν / /

Soe e PPS μ᾽ αἰξηὸν ἀμειλικ[ τ]ὸς κτάνε Μοίρη

Μ a ͵ὔ Φ / /

ἄρ[τ]. κλυ[τ]ῶν Παφίης ἁψάμεν[ο]ν θαλάμων' υτέμθεσ οἹῖδ' ovy ootoroe!s 1.1.0 2) Sip παρα ΠΣ Ἰς

fod AL οἱ Ὁ. λιπεῖν παῖδ᾽ ἕτερον μελάθροις. Εἰ δὲ θέζλ]εις γνῶναι τὸν ἐμὸ[ν] βιον, ὦ παροδεῖτα, ἦν μέν μοι τέχνη λαοξόος, οὔνομα Μειδίας, 3 \ \ De f Ass , ΄ ἐς] δὲ θεοὺς ἀνέλυσα κ[α]ὶ ἀθανάτοισι μέτειμι, ὅσσους γὰρ φιλέουσι νέοι [θ]νήσκου[ σιν ἅπα[νἾτες.

ce

Hamilton’s copy, from which this inscription has been published (Researches, vol. ii. No. 325, and Le Bas and Waddington, vol. 111. No. 1771 B.), is imperfect, but it was the stonecutter, not Hamilton (as Waddington naturally assumed), who omitted a line. All that survives of this lost line is AYFPEXON. The beginning of the first line may have been on another stone placed on the top of this one.

The inscription on the side of the same stone is fairly well rendered by Hamilton, and Le Bas and Waddington. My copy has the following variants. Inline 2, APTIMEIC < ; in line 4, add Ε at the beginning, and y in ἑαυτοῖϊς ; in line 5, read χάρειν for χάριν ; in line 8, add Ε at the beginning; in line 10, read IC for E.C, és for εἰς.

The last line of our inscription is a familiar Greek sentiment, perhaps best known in the form Ὃν yap θεοὶ φιλοῦσιν ἀποθνήσκει νέος.

292

INSCRIPTIONS FROM MYSIA.

70. Kebsud: built into the west wall of the mosque, rather high up.

Whitewashed stone. The right edge covered by the roof. Letters about 2 inches. €YAr Εὐάγ[ης CWTHP Σωτήρ[ι- +OCENVI -yos Ev[n- ΘΙ cyt -θίῳ συ[ν- τροφω -τρόφῳ πρῶτο πρωτοΐ κ- WM MT -ωμ[ή]τίῃ CYNTOIC σὺν τοῖς [ἐ- AIOICAY -δίοις αὐ[τ- WTEKN -@ τέκν[οι- CHMNHA -ς μνή[μη- CEAFTA -ς χάρι[ν GrPAYA ἔγραψα.

Hamilton’s copy, from which this inscription has been published in his Researches, vol, ii. No. 327, and in Le Bas and Waddington, 111. No. 1772,

appears to have been defective in several points, but supplies many of the letters missing at the ends of the lines.

71. Balukiser: in the abutment of a bridge on the Susurlu road about

five miles out of Balukiser, not far from the village of Eshibaji. block, 1 foot 8 inches long, broken to right.

waterworn.

TIKAA

ATH

Blue marble Letters 14 inch, well cut but

a *e, ὧν 7 ἃ

MAT KAAVAIO//MENANAPOV KAAVAIOLCAOMITIOL//////= ANSE YIQTHNEZZAPANEK ΙΔΙΩΝΣΠΟΙΗΓΣΝ

Κλαυδίο[υ] Μενάνδρου [υἱῷ Κλαύδιος Δομέτιος. ... ἀνεψίῳ τὴν ἐ[ξ]έδραν ἐκ [τῶν ἰδίων ἐποίησεν.

= is used for Ε throughout. += for = in ἐξέδραν is a mistake of the stone-cutter’s. This inscription and the next, which also refers to an architec-

tural work, are probably derived from the site of Hadrianutherae, which ought to lie somewhere in the great plain of Balukiser.

72. Balukiser: built into a mill close by the same bridge. architrave block, 7 feet long, 1 foot 2 inches broad. Letters 1 inch.

TAIOZKAAOYEIZIOZLAIOYYIOZOPATIAZEIARNEKTRNIAIRNANEOHKENTOETIZT)

Marble

4 Γάιος Καλουείσιος, Τ᾿αίου υἱὸς, “Opatias, Εἵλων ἐκ τῶν ἰδίων ἀνέθηκεν τὸ ἐπιστύλιον.

INSCRIPTIONS FROM MYSIA. 293

73. Balia Maden: formerly in a Turkish fountain, recently in the house of the late director of the mines. Small marble relief of a male figure, half face to left, with an altar in front of him. The inscription is below the relief. Letters 4 inch, much worn.

FAAAIQNOCOEY Γαλλίωνος Θεύ- AA//\OCAIIYPAM//////// -δα[μ]ος Aud Κραμ[ψηνῷ

Ζεὺς Κραμψηνὸς appears in another inscription of Balia, which has been published by Kontoleon, Ath. Mitth. xiv. p. 90, by Anastasiades in the “Αρμονία of Smyrna, quoted Bull. corr. hell. xviii. p. 541, but most correctly by Fabricius, Sitzwngsber. der Berlin. Akad., 1894, p. 903. I copied the inscription also, and can confirm the reading KPAMYHNQ, wkich Fabricius rejected for KPAMYHNQ. My copy is supported by Prof. Anderson’s, and as we had the stone specially cleaned, we are not likely to have made a mistake. It may be added here that I read the first letter of the second line as E not Σ, and made out the M of Δέκμου, and the A= of Ποπλικίας.

74, Yenije, on a western affluent of the Aesepus, north of Assar kale : in a garden. Marble slab, with gable top and reliefs of sacrificial scenes above the inscription, 4 feet 2 inches high, 1 foot 5 inches broad. Letters 1 inch. Copied in failing light.

HI/EKAIAIOZATIAIOSIEPEY//// MLL WEYXANSTH///// {T/T

ἐνὸς Καίλιος ᾿Ατέλιος ἱερεὺϊς .... ee ΤῸ, ῳ εὐχ[ὴν] ἀν[έ]στησεν .. ..

J. ARTHUR R. Munro.

294 CEANEUS AND THE CENTAURS: A VASE AT HARROW.

CAENEUS AND THE CENTAURS: A VASE AT HARROW.

[Prare ΜΠ |

THE vase that is here published, by the kind permission of the authorities of the Harrow School Museum, is the gem of the collection of antiquities presented to that Museum by Sir Gardner Wilkinson; it is described by Mr. Cecil Torr as No. 50 in his catalogue. It had been repainted and restored in such a way as to suggest that it had been through the hands of an Italian dealer ; and this conjecture as to its provenance is confirmed by the fact that a tracing of the design exists in the apparat of the German Institute at Rome; the vase comes from Vitorchiano and had been seen in the possession of Depoletti: the tracing was communicated by Gerhard. Dr. Wernicke describes the vase from this tracing in the Archacologische Zeitung, 1885, p. 262; but it is clear that the tracing was not accurate enough to give him any adequate notion of the beauty and character of the drawing; though he notices the extraordinary foreshortening of the Centaur on the right, he suggests that the design is a variant derived from a vase signed by Polygnotus at Brussels, a suggestion that could not have been made by any one who had seen the vase or a good drawing of it; the style, as we shall see, points unmistakably to an earlier and finer stage in the history of vase- painting. All the most important vases of the Harrow Museum have recently been cleaned by the skilful hands of Mr. Sharp, of the British Museum ; the scientific value of the collection has thus been enhanced, and our vase, in particular, has improved greatly in appearance. Only a few insigni- ficant details have disappeared with the restorer’s work, while the thorough tests to which the vase has been submitted enable us to be confident that all that is now left is due to the original painter—an assurance the more necessary in view of the remarkable character of some of the drawing. The design is faithfully reproduced in the drawing by Mr. Anderson, from which Pl. VJ. has been made.

The vase is of the shape commonly known as a kelebe, or crater with columnar handles (vaso a colonnette); its height is 194 inches, its diameter (including handles) 19 inches. The main lines of the figures are shown by outlines of the same black varnish as is used for the field ; in addition to these

a TA NUL

πηι U κυ

ΕἾ be ie - ‘ | dae | ~ ἢ Billing ἱ 7, ᾿ ν-- 25 ἈΔΔι " <a = ᾿ ‘Ne e a LER eae , gm weg Tite Bs y “4 ᾿ ἐκ SS WY wy j “ γι. ἵν ftv i Vf. ma WW Sy i / τ DW 4 7 '

CAENEUS AND THE CENTAURS: A VASE AT HARROW, 295

there are lighter inner markings in light brown,' and the same light brown pigment is used, as is shown in the plate, to render the hair of both Caeneus and the Centaurs, and the tawny fur on the panther skins which they wear. There are also purple retouches, to render the leaves of the branches carried by the Centaurs.

The main design is contained in panels; that on the obverse of the vase, which is by far the finest in execution, represents the contest of Caeneus and the Centaurs; the panel is bordered on each side by a row of palmettes, alternating to right and to left, in black figured technique on a red strip left for the purpose; the panel on the reverse contains merely a careless and conventional design of satyrs and a maenad ; this panel is framed on each side by an ivy wreath. Above the panels is a row of framed bars; below them a purple band running round the vase ; rays ascend from the base. Outside the rim is an ivy wreath, and on its horizontal top is a row of lotus with

ΠΝ"

Bee 777 soba Madd be ᾿ ΨΨ ΝΥΝ

interlacing stalks and petals, with a large palmette and volutes over each handle. On the bottom is an incised inscription (graffito), A4|A; and there are four drilled holes in the bottom and four more opposite them inside the bottom rim; these look as if they were to hold rivets, but the vase shows no sign of having been broken.

The design on the reverse need not detain us very long. It represents a draped maenad, who holds in her hand an object pointed at both ends —perhaps a branch—seized by two nude satyrs, one of whom holds a thyrsus ; they are baldheaded, and have horses’ or asses’ ears; they are infibulated. The drawing is of the ordinary and careless style often found even in the best period, and calls for no further attention.

1 These lighter markings have to a great the destruction of the surface when the vase extent disappeared, owing chiefly, no doubt, to —_ was repainted.

296 CAENEUS AND THE CENTAURS: A VASE AT HARROW.

The drawing of the principal scene, on the other hand, is of quite extra- ordinary boldness and vigour. In order to appreciate this, one has only to compare it with the rendering of the same scene on other vases, for instance that in the British Museum (Catalogue, vol. iii. E. 473).

A little to the right of the centre is Caeneus, still upright, but buried to the waist in the earth, σχίσας ὀρθῷ ποδὶ γᾶν, as Pindar has it. His body is seen from the front, and his head is seen in profile turned to his right; he Jeans to his left, so as to gain room for his sword arm, with which he stabs the Centaur above him in the human abdomen. The blow has not gone home, for the blade runs almost parallel to the Centaur’s body, only the point being imbedded in his flesh; the stroke is evidently borne back by the onward rush of the monster. On his left arm Caeneus holds his shield, seen about three-quarter face; it rests on the ground, but does not sink into it, as in some other examples of the scene; its device, in black- figured technique, is a running Centaur to the left, with a branch raised in both hands over his head—a design full of go and spirit. Caeneus wears a Corinthian helmet, and a breast-plate with a star as ornament on the shoulder-piece, and pteryges hanging from its waist, which look as if made of pleated linen. The Centaur wounded by Caeneus advances towards him from the left, and also slightly forward, so that his body is slightly fore- shortened; he supports on his shoulder with both hands a mass of rock with which he is about to overwhelm the hero; he wears a panther skin, knotted _ round his neck by its fore-paws, and hanging down his back, the tail, with a twisted knot in it, being visible beneath his equine body. He places his fore-legs on Caeneus’ shoulders, as if to force him yet farther into the ground. He is bald, and his mask-like face, with shaggy hair, shapeless eyes, and snub nose, is in marked contrast to the fine features of his two comrades; like them, he has horses’ ears! No blood is now visible from his wound; what was visible before cleaning was due merely to the hand of the restorer, and certainly did not exist in the original design. The Centaur behind him, on the left of the design, is seen in profile; the end of his body is cut off by the border of the panel; he wears a panther skin in the same way as the middle Centaur. His chest is turned to front the spectator, as in most drawings and reliefs of Centaurs from the finest period, and with both hands he holds over his head a pine tree, which he is about to dash down trunk foremost on to Caeneus. But for his equine ear and shaggy head, the type of his face has nothing bestial about it, but has dignity combined with its fierceness; it is the face of a baldheaded man of middle age. The head is turned slightly beyond the profile, so that the outline of the further brow stands out against the background,—a peculiarity noticeable in the case of the other Centaurs also.

The third Centaur, on the right, is the most remarkable figure on this vase, and among the most remarkable figures in all Greek vase-painting. The remarkable foreshortening of his equine body would alone suffice to

1 The top of his ear has been lost in a small fracture of the surface.

CAENEUS AND THE CENTAURS: A VASE AT HARROW. 297

distinguish the design, though it can be paralleled elsewhere!; yet even if it is open to criticism in perspective, the very difficult position is attempted with boldness and skill. But the type and expression of his face are not so easy to match; his bent brow, aquiline nose, and masses of overhanging hair, together with the way his head is turned back over his shoulder, combine to enhance the brutal fierceness of his expression; and the bold outline of his shoulder-blades emphasises the violence with which he dashes down his pine- branch on to the hero.

When we come to consider the style and the period of this design, we can have no hesitation as to the. position we must assign to it. Such work can only be found in the later productions of the cycle of Euphronius, and especially among those assigned by Dr. Hartwig to Onesimus.2 The mag- nificent Centaur Cylix, Hartwig, Pl. lix, lx, has many points of resemblance with our vase, especially in the vigour of the drawing and the boldness of the foreshortening, 6.9. in the fallen Centaur on the inside. For the fore- shortening on our vase we shall, however, see nearer analogies—some of them at least within the same cycle of Euphronius; we must first notice other points of style which confirm our attribution of this vase to an artist closely connected with Euphronius. The drawing of the eye is a safe indication of period; on our vase it is neither full face nor profile, but drawn in that compromise which is characteristic of the age and school; the inner ends are left open, and even slightly diverge, so as to give the effect of eyelashes, while the pupil, indicated by a dot and a circle round it, is placed so far towards the inner corner as to give a very near approach to a true profile drawing. The only exception is in the case of the eye of the middle Centaur, which has an unnatural and almost fishy appearance, adding to the repulsiveness of his mask-like face. I do not know of any exact parallel to this, but the intention of the artist is obvious. The mouths have not the outlines of the lips inserted, but are drawn in a freer manner. The variety in the treatment of the hair also is just what we should expect at this period; sometimes it is in black masses, sometimes drawn with delicate detail in individual tresses, the effect being enhanced by an addition of brown pigment; a treatment of which the value had been learnt by the vase- painters of this school from their practice in painting the beautiful vases with white ground that are among their finest productions. The contrast between the crisp and curly locks of Caeneus and the lank thin tresses of his monstrous adversaries is admirably worked out; and even the Centaurs are differentiated among themselves by variety of treatment, from the masses of unkempt hair of the Centaur on the right to the thin and straggling locks of the one in the middle. The types of face are differentiated with even more subtlety ; there is hardly more contrast between the delicate and conventional Greek profile of Caeneus and the strongly marked features of the Centaurs,

1 See below. course no opinion is expressed as to the correct- 2 The name is a convenient one for the iden- 688 of his restored name, tification of this set of vases; in using it, of

298 CAENEUS AND THE CENTAURS: A VASE AT HARROW.

than there is between the different types of these Centaurs themselves. The one to the left is hardly inhuman, only of heavy and somewhat truculent type ; the middle one has the conventional snub-nosed satyric mask, while the one on the right is characterised with a brutal vigour hardly to be surpassed or even matched among the extant products of Greek art. But the tendency to the choice of quaint and individual types, of which this is so striking an example, may easily be paralleled among the works of Euphronius and his colleagues. The foreshortening of this Ceutaur’s body, which we have already noticed as the most remarkable piece of drawing, finds its nearest parallel in an early work of Euphronius (Hartwig, Pl. X.). In his text. p. 108, n. 1, Dr. Hartwig mentions other examples of similar foreshortening, either in horses or Centaurs; to these may be added a Centaur on a vase from Rome (Annali, 1860, Pl. A), which belongs to a decidedly later date than the Harrow vase, and a very similar foreshortening of a dog! (Gerhard, Avwserl. Vasenb. Pl. CCLXVII.), which is on a vase of style decidedly earlier, and is perhaps the earliest example of an experiment in drawing that finds its best known if least pleasing repetition in the famous horse of the Issus mosaic.* Most of these horses are even to the raising of the tail in exactly the same position ; and they seem to be a series of attempts to adopt and improve on a bold invention in drawing ; but we cannot say to whom this invention is to be assigned, unless it be, perhaps, to Euphronius himself in his younger days. It is interesting to note that the main error of drawing in this fore- shortened figure consists in a tendency to draw the two hind legs diverging, as if seen from the side. Thus it corresponds exactly to the conventional per- spective of early art, by which an object seen from the front often has its two sides represented as extending away from the front on either hand.*

The type of the group of Caeneus and the Centaurs, known to us both on vases and reliefs, has been derived by Loeschcke* fromm the conventional group of the man between two horses which is familiar from the earliest days of Greek art, and can be traced back to still earlier sources. In the Harrow vase, which may perhaps claim to be the finest of all the repetitions of the subject, it concerns us most to note the deviations by which the painter has improved the scene; by the addition of the third Centaur, and the fore- shortening of the one behind Caeneus, he has escaped entirely from that conventional and over-symmetrical grouping which we still see even in the western frieze of the Theseum. Whether the vase-painter originated these changes in the design it is hard to say; but the vigour and originality of his drawing make us inclined to assign to him the excellence of the composition as well. It was of course a tempting hypothesis to associate this fine design with the paintings of Lapiths and Centaurs with which the artist Micon decorated the Theseum ; and such a suggestion was confirmed by the presence of the same scene on the sculptured frieze of the temple which we now call

''This comparison was suggested to me by 3 See Murray, J.7/S. ii. 318 and PL. xv,

Mr. J. ©. Hoppin. 4 Bonner Studien, p. 252. 2 Baumeister, Tal. xxi.

CAENEUS AND THE CENTAURS: A VASE AT HARROW. 299

by that name; but even if we overrule the objections that have been brought against the identification of the building, we cannot of course, with our present knowledge of the chronology of Greek vase-painting, admit the possibility of any such connexion; for the Harrow vase must be earlier by nearly a genera- tion than the paintings of Micon in the Theseum. Nevertheless it may show us the vigour and the variety of the types on which the painter could draw for his subject.

1.

The legend of Caeneus is one of the most imteresting in Greek mythology; while it is difficult to explain in some details, it contains elements which connect it unmistakably with those primitive and popular rites that underlie so many mythological stories. But the legend has been so thoroughly recast in the workshop of poetical fiction that its original character has been obscured. Let us first take the tale as it has been pre- served to us in literature and note the features that are either inexplicable in themselves or inconsistent with other parts of the story or with artistic repre- sentations; for it is from these intractable details, still cropping up through the smooth and even narrative with which the poetical mythologist has endeavoured to overlay them, that we can infer the true form of the myth.

Caeneus figures from Homer down among the leaders of the Lapiths in their great battle with the Centaurs at the wedding feast of Pirithous.?2 He rashly pursued them in their flight; they turned on him, and, finding him invulnerable to their weapons, overwhelmed him by piling pine-trees and rocks over him. He was crushed by the weight, but emerged from the heap in the form of a tawny bird. Such is an outline of the story as given by Ovid. There are other features about the tale that only complicate it, while some do not harmonise with this version. One is, that Caeneus was at first a woman, who was beloved by Poseidon, and that the god granted her wish that she should become a man and invulnerable; another, that, in the shades below, he was changed into a woman. It is curious that Ovid, who frequently refers to the first change, seems to know nothing of the second; while Virgil says only ‘vir quondam, nunc femina, Caenis.’ It looks as if the essential thing in the tradition from which the tales of the Latin poets are ultimately derived was merely the change of sex, but the relation of this change to the story was doubtful.

Then there is the tale of his going straight through the earth to the realms below, when he was buried by the Centaurs—a tale associated with the interesting words of Pindar, σχίσας ὀρθῷ ποδὶ γᾶν. We have seen how

1 Throughout this mythological discussion I * The most important passages are: Homer, am indebted to valuable hints given me by Mr. 7], i. 264 and Scholia ; Hesiod, Asp. Her, 179 ; J. G. Frazer. At the same time I cannot hold Pindar, p, 168; Apoll. Rhod. i. 57; Verg. Aen. him responsible for the application I have made νὶ. 448 ; Ovid. Met. 12, 489; Hyginus, p. 14; of them, though I am glad to be able to quote Orph. Argonaut. 168.

is general approval of iny conclusions.

300 CAENEUS AND THE CENTAURS: A VASE AT HARROW.

this expression fits in exactly with the type of the scene as usually depicted on Greek vases and reliefs; while that type is by no means a natural way of rendering the fight as it is recorded in literature. It must however be added that the words of Pindar, though they coincide so remarkably with the scene on the vase, do not offer any satisfactory explanation of it. They rather seem to point to a common origin, from which both the literary and the artistic tradition were derived, but which neither the literary nor the artistic tradition understands. Another fact that may help us in tracing the origin of the tradition, though it has no organic connexion with the story in its accepted form, is that Caeneus’ father is called Elatos, and that he himself is called Elateius and Phyllaeus.

It will help us in an attempt to trace the origin of the tale of the burying of Caeneus, if we arrange the points we have to consider in a tabular form, and then discuss them in turn.

(1) The tale is associated, apparently from the earliest times, with the battle of the Centaurs and Lapiths.

(2) Caeneus is associated with the pine-tree by his parentage (Ἔλατος), and is buried in or by means of pine trees (ἐλάταις) and stones.

(3) He is buried upright, or goes upright into the ground; he is always represented in art as standing upright, and buried to the waist.

(4) He undergoes a change of sex.

(5) He undergoes resurrection in the form of a bird, or else goes down alive among the dead.

The love of Poseidon and the invulnerability of Caeneus may be passed over for the present; they may well be invented to explain the later form of the legend; the intervention of Poseidon is readily suggested by his appear- ance as a giver of magical gifts in the early form of the tale of Peleus and Pelion.

(1) The myth of the Centaurs and Lapiths has always been a puzzle to mythologists, but Mannhardt’s explanation ! is as simple as it is convincing, Its novelty, as he himself points out, lies not so much in the actual pheno- menon with which it associates the myth, as in the aspect under which the phenomena are viewed, his great advantage over his predecessors resulting from his substitution of the comparative and inductive method for imagin- ation and theory. Others had suggested that the Centaurs were imperson- ations of natural phenomena, such as storms or torrents. Mannhardt regards them ‘as spirits of the forest or the mountain, to whose action these phenomena were assigned.’ Abstract generalisation and personification are highly improb- able in the period to which the origin of the myth must be assigned; while a belief in wild creatures of the woods is universally prevalent in Europe among peoples still in a primitive stage of thought. Such a belief is found, for example, among the Russian peasants, who believe that ‘the devastation wrought by hurricanes is the result of a battle between the spirits of the woods, battles in which the combatants hurl tree-trunks of a century’s growth

1 Ant. Wald- und Feldkulte,

CAENEUS AND THE CENTAURS: A VASE AT HARROW. 301

and rocks of four thousand pounds’ weight at one another, over a distance of a hundred versts.’' The stones and pine trees that are always the weapons of the Centaurs both in literature and in art here find their exact counterpart.

The ethical tendency of later Greek mythology has transformed and obscured the story of the combat of the Centaurs and Lapiths. The fact of a combat may have suggested a difference in character between the com- batants ; the tendency to take sides in relating a fight is irresistible; and as a result the Lapiths who fight against the wild and bestial Centaurs first won credit for their prowess in meeting such a foe

κάρτιστοι μὲν ἔσαν Kal καρτίστοις ἐμάχοντο,

and then came to be adopted in a way as the champions of humanity and civilisation, until in the age of the Persian wars the battle of Lapiths and Centaurs came to be a favourite type of the great struggle between Hellene and Barbarian. But we have only to examine the myth to see that it must have travelled very far from its original significance. The Lapiths are no Greek people, but are closely akin to the Centaurs—a kinship that is all the better attested because its exact manner is variously related. The devastation of storms is wrought by the contest of the wood-spirits, not against human antagonists, but against others of their own kind. It is made out with great probability by Mannhardt that Centaurs and Lapiths are in their origin but two different forms of the same wild men of the woods; only in the one case the anthropomorphic tendency has had more scope than in the other. Or it would perhaps be more correct to say that the wild men of the woods were originally thought of merely as rugged and hairy monsters; in the case of the Lapiths they. have come to lose everything inhuman except their super- human strength ; while in the case of the Centaurs their bestial characteris- tics have assumed a very peculiar form. The Centaur with which we are familiar in Greek art is by no means identical with the shaggy brutes of Homer and Hesiod, which, as Mannhardt points out, have nothing whatever distinctively equine about them. The appropriateness of the form of a horse, or of association with a horse, to spirits that ride the storm, is both obvious in itself and attested by innumerable instances from folk-lore, but the peculiar form taken by this association in the earliest Greek Centaurs, which are merely men with a horse’s body and hind quarters growing out of their back, is probably due either to some accidental combination or to some too literal interpretation of a metaphor used by an early poet; it really has no more to do with the origin of the Centaur than has the late and more artistic combination of man and horse that we see in the sculpture and painting of the fifth century.

We may then adopt Mannhardt’s explanation of the Centaurs and Lapiths, and regard them but as two different developments of the same original conception—of the wood-spirits whose combats left their traces

1 Op. cit. p. 96. ne VOL. XVI. ¥

302 CAENEUS AND THE CENTAURS: A VASE AT HARROW.

behind them in pine-trunks and rocks hurled by storms about the slopes of Pelion.

(2) A close association with various trees is naturally enough to be expected of wood-spirits, whether such relationship be so definitely realised as to cause them to be regarded as the children of tree spirits (Dryads, &c.) or not. In some cases the fact is definitely stated ; thus Pholos is the son of Melea (the Dryad of the ash) ; Dryalus who is called Πευκείδης and Elatus are names that speak for themselves. So Caeneus also is called Elateius (the pine-tree man) and Phyllaeus (the leaf-man). By later mythologists his father is called Elatus ; it seems likely that the epithet Elateius, which could just as well come from ’EXarn (pine) may have existed before the name coined to explain it. The use of pine-trees to overwhelm the Lapith hero demands of necessity no further explanation, since the pine-tree is the recognised weapon of the Centaurs; but, in view of other indications, it is worth while to note that the pine-tree had a peculiar sanctity in Greece, especially in cases which seem to point to a ritual of human sacrifice. Thus Attis wounded himself and died under a pine-tree; Pentheus was set up in a pine-tree, stoned, dragged down, and torn to pieces by the Theban maenads!; and the robber Sinis, the pine-bender, slew his victims by fastening them to two pine-trees and was himself slain in the same manner by Theseus. It is possible then that the appearance of the pine-trees in this case may have some significance beyond their ordinary use as weapons by the Centaurs.

The stones thrown at Caeneus are even more significant. We have just noticed how this feature occurs also in the tale of Pentheus; at Troezen a festival called the stone-throwing (λεθοβόλια) was held in honour of Damia and Auxesia,? and the legend went that these two maidens had been stoned todeath. Mr. Frazer writes: “It is practically certain that Damia and Auxesia were spirits of vegetation and growth. Their images are said to have been made of the sacred olive wood of Athens in order to restore to the land of Epidaurus the fertility which it had temporarily lost, and the making of the images had the desired effect.2 Their names, too, point in the same direction. Now battles more or less serious, conducted in the fields with stones as weapons seem to have been regarded as a means of promoting fertility in many parts of the world. Why they should have been so regarded is more than I can say at present, but the fact seems to be undoubted. For European examples, see Mannhardt, Bawmkultus, pp. 548-552. In my note on Paus. II. 32. 2, I quote more examples, of which I will mention one or two. Among the Khonds of Orissa, who sacrificed human victims and buried their flesh in the fields to fertilise them, a wild battle was fought with stones and mud just before the flesh was buried in the ground (S. Macpherson, Memorials of Service in India, p. 129). In Tonga an essential ceremony to procure a good crop of yams was a battle between the islanders, one half of the island against the other half.

1 See Bather, J.H.S. xiv. 251. batim from a letter of Mr. Frazer. 2 Paus. II. 32 2. A good deal of this 3 Herod. v. 82-87. evidence about stone-throwing is quoted ver-

CAENEUS AND THE CENTAURS: A VASE AT HARROW. 303

The fight was obstinate and lasted for hours (see Maurice, 7'onga Islands, 2, p. 207). In Gilgit an elaborate sham-fight marked the time for pruning the vines and the first budding of the apricot-trees (Biddulph, 7'ribes of the Hindoo Koosh, p. 102). These battles of stones, associated in myth with victims who were slain by stoning, seem in every case to be regarded as con- ducive to the fertility of fields or trees. In the case of Caeneus, who is buried with stones in such a battle, the association with a pine-tree is already otherwise attested. It is an obvious inference that his stoning and burial is regarded as conducive to the growth and fertility of the tree with which he 15 associated.

(3) We now come to the most curious part of the whole myth, a part that perhaps may show its real origin. Both in the literary evidence and in the artistic representations we have noticed a fact that is inconsistent with the rest of the story. When describing how Caeneus is slain by the Centaurs, smitten with green pine-branches, Pindar adds that he cleft the earth with unbended knee ; and this is just how he is represented in art, buried to the waist but upright. Now this is not the position in which anyone would be buried ‘who was overwhelmed by the mass.of unwieldy missiles hurled at him in the confusion of combat. It is hardly too much to say that such a manner of burial implies a deliberate and intentional act, and that its interpolation in the battle-scene is more or less accidental,! while the invulnerability of Caeneus is a mere invention to explain it. Now it is not easy to say why either Centaurs or Lapiths should bury one of their own number in this strange way; but there is another explanation which suggests itself. The Centaurs or Lapiths, as we have seen, are wood-spirits, whose life is closely bound up with the pine forests in which they live; and it is a very common thing for divinities or superhuman beings to have tales told about them which are merely derived from the ritual practised by men in relation to the function or phenomena with which such divinities are associated. One has only to recall the way in which the wanderings of Demeter are related in imitation of the wanderings of the mystae at Eleusis, or the tale of Lycaon’s slaying his son to feast the gods is coined in imitation of the cannibal sacrifice οἵ Mt. Lycaeus. If we apply a similar solution to this problem, we should naturally look for the rite from which the myth is derived among those customs that are associated with tree or wood-spirits or divinities on which the growth of vegetation is dependent. We have already noticed examples in which the pine-tree, doubtless as containing such a spirit, was associated with human sacrifice in Greece; and the analogy of popular customs throughout Europe leads us to see in such sacrifices, real or symbolical, a mystic connexion between the life of the man and the life of the tree.” That Caeneus was a man in this condition is implied both by his epithets and by the company in which we find him. That he should be stoned to death is

1 Compare however the practice of the Khonds, coincidence due to a similar contamination of quoted above, in which the burial of the victim _ two distinct rites. is associated with a battle. But this may be a 2 See Frazer, Golden Bough, passim. Y 2

304 CAENEUS AND THE CENTAURS: A VASE AT HARROW.

in accordance with a common custom in such cases; we need only remember, once more, the tale of Pentheus. And in Russia, for example, the burial of Jarilo! (the spring) is associated with practices like the setting up of a tree which contains the successor of the victim. Perhaps however this burial to the waist may have a more exact significance ; it is a practice best known in witchcraft like that of Horace’s Canidia,? or fanaticism like that of the Suffering Ivan at Kief or the Russian devotees who even yet follow his example. But it may go back to ἃ notion that by planting the man who represented the tree-spirit as if he were himself a tree, the growth of the tree would be assured.* I only give this conjecture, as it was suggested to me by Mr. Frazer, with all possible reserve. The chief justification for it lies in the fact that it exactly meets the required conditions, and explains what is other- wise inexplicable in the traditional form of the myth.

(4) Caeneus’ change of sex is significant, since a change of sex, or a disguise of sex, is an extremely common feature in popular customs that are connected with the tree-spirit. I need only quote again the case of Pentheus, who was disguised as a woman before he was set up in the pine-tree whence he was dragged to his death. There is no need here for us to seek an explanation of a fact which has hitherto baffled mythologists ; but the existence of this peculiarity in Caeneus is a striking confirmation of the view that he too is to be regarded as representing the tree-spirit.

(5) Resurrection, real or simulated, is another very common feature in the rites so often quoted. Resurrection in the form of a bird is not indeed known to me in any clear example, though the tale of the Phoenix suggests itself, and the practice of liberating an eagle from an emperor’s funeral pyre is well known. Perhaps this fact may make us suspect the form of the resurrection, which is recorded only by Ovid, but it is hardly likely to have been an entire invention, and the resurrection in some form must have existed in the early myth. This is confirmed by the fact that according to Pindar and Apollonius Rhodius Caeneus seems to have gone down alive among the dead,

ζωόν τ᾽ ἐν φθιμένοισι μολεῖν ὕπο κεύθεα γαίης

Orph. Argon. l.c.

Finally, we may find yet another independent proof that we are right in regarding the tale of Caeneus as a survival from the primitive rites connected with the tree-spirit that are familiar wherever the may-pole is set

1 Mannhardt, W.F.K. p. 265.

2 Epod. v. 32.

3 Burial in these cases was up to the arm-pits or to the shoulders. The motive is recorded to be in one case to produce pining, in the other the mortification of the flesh ; but in both cases the practice is probably earlier than its explana- tion.

4 A curious analogy is offered by the crop of

watriors who come up when Jason sows the dragon’s teeth, Ap. Rhod. iii. 1374, &. Many of them are slain while still buried to the waist, like Caeneus ; and the first comparison, which may well be traditional, is to ‘ pine-trees or oaks, that are hurled down by the blasts of the storm.’ If this is only a coincidence, it is a very curious one,

CAENEUS AND THE CENTAURS: A VASE AT HARROW. 305

up. For one more fact is recorded about him which has no rational connexion with the rest of the tale, but which is easily explained on this supposition. In the Scholia to the Iliad! it is said that he πήξας ἀκόντιον ἐν τῷ μεσαιτάτῳ τῆς ἀγορᾶς θεὸν τοῦτο προσέταξεν ἀριθμεῖν. What can this mean but that he set up ἃ may-pole on the village green, thereby proving, if further proof be needed, the true nature of the tales that were told about him ? ERNEST GARDNER.

1 Schol. A on A 264,

306 VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM.

VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM. [PLates VII, VIII.]

THE Terra-cotta reliefs which form the subject of the present article have been for some years one of the minor attractions of the Acropolis Museum, and I am indebted for permission to publish them to the kindness of the Ephor General M. Cavvadias, and of the Ephors, MM. Stais and Castriotis, The latter as superintendent of the Museum most obligingly put at my disposal all the information in his possession and afforded me every facility for photographing the tablets. No single tablet in the collection is perfect and of the 73 fragments! which comprise it, the great majority came to light during the systematic excavations of the Acropolis which were carried on from 1885-1890; they were all found at some depth below the surface and as far as could be learnt, to the north, the east and the south-east of the Parthenon. The rest have been in the Museum since 1863 ; these were all found on or near the surface and it is possible that others were carried away by visitors and are now hidden in private collections. The publication of the Museum fragments may lead to their identification and the completion of some of the tablets whose design cannot at present be determined.

As already stated no complete tablet has been preserved, but a sufficient variety of fragments exists to show that they were of uniform size, clay and technique, about 22 centimetres long by 16 wide and 1 thick, bordered top and bottom by a rim projecting } a centimetre beyond the background, from which the relief rises to a height not exceeding 14 centimetres. They are pierced by three holes, usually one in each of the upper corners and another in the centre of the lower edge. The clay is very hard and fine, of a pale red colour which on fracture shows brighter red streaks ; the firing is well done and very few of the fragments have any trace of the warping common in the reliefs from Epizephyrian Locri. The tablets are all covered with a layer of white lime-wash as a basis for the colour used, pale blue for the background, bright red, green (?) black, brown and yellow for the different portions of the relief. The border, the outside edge and, in one case, the back were painted crimson red, so that the general effect must have been brilliant in the extreme ; the colouring, however, though vivid, was much more carefully done than is usual in terra cotta work, and we do not find that the green

1 Reg. Nos, 1318-1391.

VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM. 307

border of an aegis is continued on to a red robe or that the red of the chariot wheels has trespassed on to a blue background.

In style the designs are archaic, but it is an archaism due more to conservatism than to want of skill. Those who have studied the cases of terra-cotta figures in the Acropolis Museum or Dr Winter’s account of them * will remember that they are distinguished by the same fineness of clay, perfection of firing and precision of colouring. Another technical peculiarity which they share with the tablets is the use of lines of pale grey colour to sketch in that portion of the design which is not rendered in relief. The statuettes and the tablets must therefore come from one and the same locality, and the reasons which caused Dr. Winter to describe the statuettes as of local, ὁ... Athenian, origin derive fresh force from the evidence supplied by the tablets. In subject as in technique the latter are extremely local. All the designs, save three, one of which is too fragmentary for interpretation, refer to various conceptions of Athena as Ergané, Polias, Archegetis or Promachos, nor is this to be wondered at. It is impossible to separate Athena from her chosen city; the one idea includes the other. In other states she was worshipped as one of the divinities; in Athens she was the divinity, her reputation and cultus increased in proportion as the reputation and power of the city grew, her nature underwent the same transformation as did the nature of the state; originally she was a goddess of agriculture, the goddess to whom the earliest corn sowing and the fields? were dedicated, but as the leader of an ambitious race, fighting its way to power, she became a goddess of war. The complete transformation of a divinity to suit the character of a particular set of worshippers is not uncommon in Greek mythology, witness the warlike Aphrodite of Sparta, but the development of Athena proceeds on logical lines. The central point of the conception is mind or rather intelligence, the practical mind which turns matter to the best account and subdues brute force, therefore not only every art but every craft however humble was under her protection; as Athena Ergané she invented the flute, she helped to build the Argo, she wove wondrous garments for herself and Hera, she taught the Rhodian artists te people the island with statues which seemed to live and breathe; as Archegetis or Polias, by her wise counsel she protected the fortunes of the state and if necessary armed herself in its defence. The conception of Athena as a goddess of wisdom does not come within our province ; it is a later one which arose when Athens, no longer paramount in the political counsels of Greece, sought and obtained a wider and less disputed sway in the intellectual world.

We have therefore two distinct sides to her character, the peaceful and the warlike, and the latter assumed undue prominence because the existence of a Greek state depended on its fighting power—therefore tle popular idea of Athena Polias was of a combative goddess, though originally her function was both peace and war. Eustathius*® describes thellian palladion thus:

1 Arch. Anzeiger, 1893, pp. 140-148. 3 Eustath. on Z 91, p. 627, 2 Suidas, s.v. Procharisteria.

308 VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM.

διοπετὲς μὲν εἶναι, στέμμα τε ἔχειν καὶ ἠλακάτην, ἐν δὲ TH κεφαλῇ πῖλον (πόλον) καὶ δόρυ ἐν τῇ δεξιᾷ. Apollodorus gives much the same description of it, τῇ μὲν δεξιᾷ δόρυ διῃρμένον ἔχον, τῇ δὲ ἑτέρᾳ ἠλακάτην καὶ ἄτρακτον. It is under this form that it appears on late silver coins of Novum Ilion? Again the Athena Polias of Erythrae,? an archaic wooden statue ascribed to Endoios, had a distaff in either hand; in the liad, though Athena mainly appears as an active partizan of the Greeks and therefore in her warlike character, attention is also drawn to her skill in womanly arts* which are again insisted on in the Homeric Hymn ‘H δέ τε παρθενικὰς ἁπαλόχροας ἐν μεγάροισιν | ἀγλαὰ ἔργ᾽ ἐδίδαξεν, ἐπὶ φρεσὶ θεῖσα ἑκάστῃ. Later on the two conceptions were somewhat sharply sundered; the arts of peace were allotted to Athena Ergané who took a lower position and was worshipped only by craftsmen, for Plutarch ® in contrasting the worshippers of Athena Ergané who work with their hands, with those of Athena Polias who use their brains quotes from Sophocles : 7 πᾶς ὁ χειρῶναξ λεὼς οἱ τὴν Διὸς γοργῶπιν ᾿Εργάνην...... PA, cts satel d 19s eh: προστρέπεσθε...

At the period to which our reliefs belong (the end of the sixth century and beginning of the fifth) the difference had not been emphasized and in this indistinctness of thought it seems to me that we should seek the solution of the problem as to whether Athena Ergané had a special temple on the Acropolis or not. I do not think she had. Pausanias states® that the Athenians boasted of having been the first to worship her, and inscriptions to her have been found on the Acropolis, to the north, south and west of the Par- thenon, among them a grave inscription ® wherein the relatives of one Euanthé state that they have dedicated ‘a painted tablet in the precinct of Pallas the laborious’ (εἰκόνα μὲν γραπτάν...θήκαμεν ἐργοπόνου Iladdaébos ἐν τεμένει), but there is no reason why this description should not be applied to Athena Polias. Most of the inscriptions to Athena Ergané are set up by the relations of women whose special interest in the Polias was obviously the peplos woven for her, and to whom she might well be ἐργοπόνος, as the robe was begun at the feast of Athena Ergané under the supervision of her priestess and of the Errephorae ; therefore offerings to Ergané might well be placed in the Polias temple and the latter goddess be described as ἐργοπόνος.

The reliefs fall naturally into two main divisions: (44) those which- represent Athena, and those (B) which, so far as we can tell, represented some other personage. The latter division whieh only includes 3 fragments out of a total of 73, is figured under numbers 8, 9, 10.

1 Apoll. iii. 12, 3. 5 iii. in Venerem, lines 14-15. 2 Brit. Mus. Cat. Coins of Troas, Plate XI. 6 Praec. de ret. ger. 5.

3-7. 7 Soph. Frag. Dindorf 724. 3 Paus. vii. 5, 9. 8 Paus. i. 24, 3.

v. 735; xiv. 178. 9 C.I.A. ii. 1830.

VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM. 309

Division A, Representations of Athena, contains four clearly marked types :-—

(1) Athena Ergané, a seated figure spinning (Fig. 1 and Plate VII. 1).

(2) Athena Polias, seated, unarmed in gala array (Fig. 2 and Plate VII. 2).

(3) Athena Polias, seated armed but not combative (Fig. 3).

(4) Athena Promachos. (Figs. 4, 5, 6, 7 and Plate VIII, 1, 2).

SS

INS St

4%,

Of the two aspects under which the reliefs represent Athena, armed and unarmed, seated and standing, the unarmed seated type is the more attractive and gives rise to several interesting problems. One type of it is shown in Fig. 1 and Plate VIJ.1.1 A young girl dressed in the ordinary house costume, an

1 Two examples. the old wall’ in the forecourt of the museum. (1) Reg. No. 1827 ; length 0°19 cent. x 0°16. All the drawings in the text have been skil- Traces of black on chair. See Plate. fully put together by Mr. F. Anderson from my (2) 1880; length 0°22 x 0°16. Draperyincised. photographs. The left hand in Fig. 1 is re- Traces of red on footstool, kerchief, and rim, stored from a vase-painting. and of blue on background. Found 1886 ‘near

310 VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM.

Tonic chiton of fine material, her hair concealed by a handkerchief, seated, in a somewhat inelegant attitude on a long bench, her feet resting on a foot-stool. Two examples exist, the less perfect of which is represented on Plate VII. as it supplies the key to the action; the peculiar turn of the right hand and the spindle seen below it show that the girl is spinning; the left hand is missing but the raised forearm is in the position required for holding a distaff (cf. Steph. C.R. 1863. Plate II. 17). Given the locality of the find, and the special circumstances which connect Athena Ergané with spinning and especially with the manufacture of the peplos, there seems little doubt that the tablet represents her in the likeness of one of the ἐργαστῖναι, a young girl spinning in the women’s chamber and therefore represented in indoor costume and an easy attitude which form a piquant contrast to the prim position and elaborate costume shown in the second type (Fig. 2). The strong resemblance to the ‘Peitho’ of the Parthenon frieze is of course evident, but the attribute of the spindle seems to allow no room for doubt as to the person depicted. The dedication may have been made either by a young girl or perhaps it was the gift of her relations who, like Euanthé’s friends, offered an εἰκόνα γραπτάν in memory of her, not of course a portrait as these tablets were made by the dozen. .

An interesting point in connexion with this figure is the possible light it throws on the subject of the ‘catagusa’! statue. It is now generally admitted that κατάγειν means ‘to draw out the thread,’ and Forster (Phil. Supp. Bd. IV. pp. 720, 21) has already made the suggestion that the ‘catagusa’ might prove to be a representation of Athena Ergané; therefore in our relief we may have an indication of the main lines of the subject. The ‘ catagusa’ was of course a statue in the round, but its novelty is as likely to have consisted in the adaptation of the ‘motif’ of a relief as in an original conception.

Fig. 2 gives another representation of Athena? in a dignified somewhat hieratic pose, corresponding to her elaborate gala costume. In her right hand she holds a bowl, the left is tightly clasped over some object which was not indicated in relief and has therefore disappeared. The treatment of the face (Plate VII. 2) shows less of archaism than the preceding and there is a dignity and nobility about the figure which, to compare the infinitely small with the infinitely great, recalls the Parthenon frieze and is shared by one other head in this collection (Fig. 3). There is no direct evidence to prove that this figure is an Athena at all, still less an Athena Polias, but the very absence of any distinctive attribute is in favour of the attribution, and the obvious connexion between Figs. 1 and 2 helps to strengthen it. An Athena Polias seated in much the same attitude and costume and holding a dish in her outstretched right hand is shown on an Etrurian hydria where the goddess is

1 Pliny, Nat. Hist, xxxiv. ‘Praxiteles fecit Reg. No. 1838 (PI. ix. 2). Upper part of relief item catagusam.’ For κατάγειν in this sense see 0'15x0'16. Stephané red, background blue. Plat. Soph. p. 226°, Pollux, vii. 29, &c. See Hair and chiton incised. Traces of burning. also Forster, loc. cit. p. 719. Reg. No. 1387; 0°14x0°15. Slight traces

* Two examples : of blue on background and of red on chair.

ATHENIAN VOTIVE RELIEFS

J.H.S VOL. XVII. (1897), PL. Vil.

wach

ἥ

we.

ATHENIAN VOTIVE RELIEFS.

J. H. S. VOL.

XVil. (1897), PL. Vill.

a

VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM. 311

identified by her helmet, spear and snake.! Further, an inscription * found to the west of the Parthenon recounts how the Boulé was approached by the fathers of the maidens who span the wool for the peplos of Athena Polias, with the request that as they had fulfilled all their duties creditably and accompanied the procession, they might now offer a silver phialé, value 100 drachmas, to the goddess as a mark of piety to her and goodwill to the Demos. This inscription is assigned by MM. Kohler and Foucart to 98-97 B.c.,

Z\tr rrr ({ ζζ WHE" γ

but the custom of the dedication of a silver dish by the ἐργαστῖναι can hardly have been instituted for the first time at that date, and under like circumstances this tablet would be an especially appropriate offering.

Only three small fragments ° exist of a tablet which shows the goddess

1 Gerh. A.V. iv. 242, 1. This vase has dis- 3 Reg. No. 1321; 0°10x0°9. Background

appeared and has apparently not been seen since __ blue, hair red. the publication by Gerhard. It is not in Berlin. No. 1318; 0°4x0°2. 3 0.1.4... 477. Completed 8.0. Η. vol. xiii. No. 1855; 0°6 long x 0°2.

pp. 170, 1, No. 6.

312 VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM.

accoutred with helmet, aegis, and gorgoneion (Fig. 3), but her elaborately- dressed hair and fine Ionic tunic prove that she is the victorious Polias enjoying the blessings of peace for which she fought. The scale on which the helmet and head are given show that the figure was a seated one,and we may restore the design by placing a spear in one hand and an owl either in the field or in the left hand (cf. Gerh. Trinkschalen wu. Geftisse, Pl. XIII., 1.)

The most popular representation of the goddess was as Athena Promachos in her war chariot (Figs. 4,5,6, Plate VIII.,1,2). The design is common on black figured vases, but its peculiarity here lies in the absence of the horses, which are left to the imagination, though the goddess’ right hand is outstretched to hold the reins, and rests on the front rail. Two variations of the design exist; in one, the goddess with a spear in her right hand is mounting the chariot,! in the other, she has already mounted and holds a shield or a

Fie. 3:

spear on her left arm.” The face, probably for religious reasons, is more emphatically archaic than in any of the reliefs already discussed, with a thick nose, prominent eyes and chin. One fragment (Fig. 5) shows heavy features and coarsely incised locks of hair, and further differs from the others in having no owl in the field.* The owl was adopted as Athena’s crest probably because

1 Fifteen large fragments, of which the most 0°'18x0.8.

important are: Reg. No. 1333. Plate viii., Reg. No. 1322; zigzag pattern in grey lines Fig. 1. 0°22x0°16. on the aegis. Background blue, chariot and robe red, aegis 2 Two specimens : black border. Reg. No. 1334. Plate viii. 2. 0°15 x 0°16. Casque of helmet black, crest red, lips red. Reg. 1340. Fig. 4. 0°10x0°9. Owl yellow-brown, details in black. 3 Reg. No. 1336; 0°12 x 0°14. Reg. No. 1835 ; 0°22 x 16. Colouring asin*. Relief 14 cent. high.

Reg. No. 1341, showing left side of tablet;

VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM. 313

it abounded on the Acropolis, for otherwise it was a bird of evil omen,' but under her protection it flourished so much that γλαῦκα εἰς ᾿Αθήνας was the Greek equivalent of ‘ taking coals to Newcastle. * Attention may be drawn to the care with which the bird is represented; the feathers are first incised in the clay, and then carefully picked out in black; in fact, one of the most striking points about these little reliefs is the care expended on details, the most perfect specimen being shown in Fig 6, where the scales of the aegis are outlined in black and coloured alternately red and blue, the inside of the aegis green, the robe red and the chariot rail black.

i; a

wz

Fic. 4.

The main interest of the Promachos design centres in Athena’s accoutre- ments, helmet, aegis, and shield. The helmet is of the high-crested Attic type, the shield also Attic, the aegis cloak-like in form and edged with a double wavy line which does duty for a snake border, (the elaborate scale aegis [Fig. 6] has a nearly straight edge), the gorgoneion does not appear on it at all, though the seated Athena (Fig. 3) has one. The aegis in this

1 Ael, de Nat. Anim. x. 37. * Schol, ad Arist. Hqwites, 1102.

314 VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM.

form is that worn in all archaic representations of her, and much more closely resembles the Homeric conception of it than the scale gorget of later

Fic. 5.

art. Curiously enough, the tradition as to the aegis is extremely confused and uncertain. Homer states that she received it from her father Zeus, ac-

cording to Euripides! it was the skin of Gorgo, and according to a still later tradition ? it was the skin of the Titan, Pallas, whom she slew. Probably,®

1 Jon 1000, and a gem in the Brit. Mus.: 3 Fora discussion of this question see Reichel, Murray, Handbook of Arch. Pl. XII. 9. Homerische Waffen, pp. 65-72. * Apoll. i. 6, 2.

VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM. 315

it and the lion skin of Heracles were the sole survivals of a time when the only thing available for protective armour was a skin, worn as a cloak in time of peace, and brought round over the left arm in battle ἐν προβολῆ An unsuccessful attempt to represent the tufts of hair on the skin may be the basis of the scales, for the fleece on an early Rhodian pinax? is rendered by a series of triangular lines which are not unlike rough scales, but the main

Frc. 7.

reason for them arises from the combination of the aegis and the gorgoneion., This took place when the general acceptance of the Argive version of the Medusa story (according to which Perseus was inspired by Athena to slay the Gorgon), led to the Gorgon’s head becoming as essential an attribute of the goddess as was the aegis. At a very early stage we find that the fringes of

Fic. 8.

the aegis had developed into writhing serpents, either mechanically, or to increase its terrifying power; but when the gorgoneion was transferred to it from the shield where Athena first placed it,’ this power centred in it and the Medusa legend with its snakes dominated the conception. The archaic

1 Gerh. A.V. ii. 127. 8 Apoll. ii. 4, 3, 7. 2 B.M. First Vase Room, Case A, No. A 750.

316 VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM.

Tonian (2.¢e., Attic) Gorgon type is that shown in Fig. 7, broad, full, and fleshy, with little or no trace of snakes, the lines of the mouth are horizontal but the eye teeth do not show and few of the steps have been taken by which a simple mask grew into the snake-wreathed horror of later times.

It will thus be seen that the design is consistently archaic in all its details: where the gorgoneion appears at all it is on the shield, and that only in 2 fragments out of 40, while the shield on Plate VIII. has none at all. It therefore belongs to the archaic Athena type reconstructed by Studniczka, and assigned by him to the 6th century.

I have no suggestion to make for the restoration or interpretation of the fragments shown in Fig. 8.8

The draped male figure, standing by a bench‘ of which a portion is shown in Fig. 9 can be completed by the help of Fig. 94, a tablet seen by

Stackelberg and published in his Graeber der Hellenen (Plate LVI. 4). Both he and Overbeck (Kunstmythologie III. p. 68) interpret the figure as Apollo, and there is a striking resemblance in the treatment of the hair to the colossal Apollo head of the West Pediment at Olympia, but as the same treatment is shown by a head of Hades in two terra-cotta reliefs from Locri, now in the British Museum, the evidence in favour of this identification is not conclusive. The hind, in Fig. 9A, is an unusual attribute for Apollo, but I cannot vouch for this detail as I have not succeeded in tracing the tablet.

1 Two examples : 3 No. 1391.

Reg. No. 1367 ; 0:10 x 0°7. Shield black, rim Feet of draped figure moving to left ; 0°11 x 0°7. red. Found 1886. 4 No. 1389.

Reg. No. 1372; 0°8x0°7. Gorgon’s eyes out- Draped figure standing against a bench; lined in black. 0°10x 0°11. Pink in folds of drapery.

2 Studniczka, Ath. Mitth. xi. 185.

VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM. 317

The Heracles relief! (Fig. 10) has already been published and discussed by Dr. Reisch, and is only included here to complete the series, I differ,

Vie 9a.

Fie. 10.

however, from him in regarding it as an offering to Heracles, and think that like the other tablets, it was offered to Athena. Ample evidence of the

1 Reg. No. 1323; 0°12 x 0°9. τῆς ᾿Ακροπόλεως. Reisch, Ath. Mitth. 1887. Hair incised: Found in 1886. παρὰ τείχου

ΗΝ, -ΨῸ , XVIT, Z

318 VOTIVE RELIEFS IN THE ACROPOLIS MUSEUM.

custom of dedicating to one divinity the image of another is afforded by the temple favissae.

In describing these tablets I have said little about their artistic charm ; though photography does not reveal it, they have all the graceful precision of line and somewhat prim beauty, which finds its highest expression in the sculp- tures of the Treasury of the Athenians at Delphi, and I, therefore, assign them on stylistic grounds, to the end of the 6th and the beginning of the 5th century, a conclusion which is strengthened by Dr. Reisch’s attribution of the Heracles-cum-lion schema to the end of the 6th century, and by the consist- ently archaic treatment of Athena’s panoply in Fig. 4. They certaiuly afford interesting evidence of the high artistic level of the age, for the potter was a craftsman who followed, but did not guide, the public taste, and they also serve to illustrate the varied aspects under which the Athenians regarded her to whom they prayed

Παλλὰς Tpitoyéver’, ἄνασσ᾽ ᾿Αθηνᾶ, v / , Ν / ὄρθου τήνδε πόλιν TE Kal πολίτας ἄτερ ἀλγέων καὶ στάσεων καὶ θανάτων ἀώρων σύ τε καὶ πατήρ. C. A. Hutton.

ON THE TUMULUS OF CHOBAN TEPEH IN THE TROAD, 319

ON THE TUMULUS OF CHOBAN TEPEH IN THE TROAD.

I DERIVE the materials of the present paper from some memoranda which I find amongst my archaeological notes and which relate to certain explora- tions to which I was not a party, made so long ago as 1887. I have thought that the particulars then obtained may be deemed sufficiently interesting to deserve a record in the history of Trojan archaeological discovery.

The subject is one of the four small tumuli dotted about and near the hill of Balh-Dagh, the crest of which according to the now exploded theory of Le Chevalier (1785) was supposed to represent the Pergamos of Troy. Ih a memoir contributed to the Jowrnal of the Archacological Institute of 1864, I proved that the site in question was no other than that of the ancient city of Gergis. In the same paper I gave an account of the results of the excavation of one of the group of three tumuli on Balli-Dagh, the so- named Tomb of Priam. The other two, namely Le Chevalier’s Tomb of Hector, and an unnamed hillock, were excavated respectively by Sir John Lubbock (about 1878) and Dr. Schliemann (1882) without result. The present relates to the fourth mound on the road between the villages of Bournarbashi and Arablar (as shown in the published maps), which goes by the name of Choban Tepeh (Shepherd’s hillock) and the Tomb of Paris, according to Rancklin (1799). This tumulus was secretly excavated at night by some workmen under the direction of a Turkish village priest, in the usual hope of finding treasure, and the enclosed tomb was rifled on the 6/7th March, 1887. Some valuable and interesting objects were in effect found, with others of archaeological interest as affording still further proof of the non-identity of Balli-Dagh with ancient Troy. The Turkish authorities, having got wind of the matter, imprisoned the priest, and took possession of the objects found, which they forwarded to the Imperial Museum at Constan- tinople. I had the opportunity of seeing the articles at the Government House before they were sent on.

They consisted of the following :—

A solid golden chaplet with thin oak leaves and small acorns on long vibrating stems.

Three golden fillets, with embossed pattern.

A number of fine strips of gold.

Fragments of sprigs of myrtle, with stems in lead and bronze leaves and berries gilt,

A bronze speculum—plain,

| Ζ 2

320 ON THE TUMULUS OF CHOBAN TEPEH IN THE TROAD.

A small bronze patera. An alabastron of alabaster. Some iron nails.

The above were found in the tomb resting on the solid rock near the centre of the tumulus. I lost no time in visiting the place and the results of my examination are best shown in the accompanying diagrams of the tumulus and tomb. An open trench intersected the mound to the surface level of the tomb, of which one of the covering slabs had been broken in effecting entrance. The position of the tomb in the tumulus is more towards the east which is the longitudinal direction.

The dimensions of the tumulus are :—

Diameter at the base, about 130 feet, Height above the level of the slope, about 20 feet ;

Section of TUMULUS of CHOBAN TEPEH

Section of TOMB

Frank Calvert

of the tomb :— ft. in. Interior length of chamber . αἴ .1|8: Breadth ¥ ᾿ ἦν τολδας att % & Height Ἢ δ writ wads doid&- Thickness of masonry ..... 10.

I have only to add that the masonry of site iets is of well dressed stones fitted closely together, without cement, and the roof formed of five slabs. The material is from an ancient quarry on the banks of the Scamander at the foot of the hills. It is not of the same formation as the rock on which the tumulus stands, but is a peculiar one, composed of the débris from the heights on the side of the rivers Scamander and Thymbrius, which have been indurated by the coulées of trachyte flowing over them in the valleys of these rivers. The stone is of good quality and easily dressed.

FRANK CALVERT, DARDANELLES, 25TH November, 1897.

PL. Xl.

\‘1897)

J. H.S. VOL. Xvi.

ἜΝΞΗΙν LY

QV3SH LIVYLHYOd τ

Bees, tt

nee aw a“ Ῥ ὌΝ Je ν᾿: δῦ ς at "6 iY secnaelee eatin” πὰ τ' “4 ᾿ ἐν re a SF wall ἔμ ΤῊ gat ΜΝ es «| 85" ὃν» a Te a - he 4 aL ᾿ ae , be be i> P a ur Ya Gee f μ " ΚΝ + oe "» “ ’ δ, Ἢ > 7 q \ ‘e . ! é ay ae, 7 Py y ὃ Yi, . uw /*

4h. it LA ἣ ms

A THRACIAN PORTRAIT. 321

A THRACIAN PORTRAIT:! [PLATE ΧΙ]

WHEN I first saw this head,? I was at once struck by its marked individuality : if any portrait could be recognized from a coin, it seemed to be this, for features so personal the poorest engraver could scarcely conceal. My hopes were realized, as a comparison of the accompanying photographs with the coin*® reproduced beneath will I hope prove.

In both we see the same treatment of the hair in front, the same fashion of wearing it behind: the long upper lip, the nose with its curiously distended nostrils—the marble preserving just enough to make the agreement certain—the long ears, the deep lines on the cheek, the shape of jaw and forehead, the prominent Adam’s apple; these too are common to both. In one point only is a slight difference noticeable: the eyes of the bust are rather small, those of the coin decidedly large, but this is precisely the feature which an artist in little would naturally exaggerate. And any doubts, which I at first had, were finally dispelled by the existence of two inscriptions at Athens, completely bearing out the numismatic evidence.

The coin bears the legends—

(Obv.) ΒΑΣΙΛΕῪΣ KOTYS (Rev.) BASIAEQE ΡΑΙΣΙΚΟΥΠΟΡΙΔΟΣ or PAIZKOYTIOPENS.

11 am indebted to Dr. Imhoof-Blumer for 2 Cavvadias, Catalogue, 581; of Pentelic his kindness in sending me a cast of the coin marble; found in A‘hens in 1887, here reproduced, to Mr. Warwick Wroth for a 3 Imhoof-Blumer, Portrdtképfe etc. Taf. similar courtesy, and to Mr. Charles Clark for 2, 17. having photographed the Athenian head.

322 A THRACIAN PORTRAIT.

But, apart from this, the almost Bacchic wildness of the locks above the forehead and the square face—curiously reminding me of the Franconian type as drawn by Holbein, Strigel, Diirer, and others—would point unmis- takably to a barbaric origin: no one could for a moment think our subject was either Greek or Roman. The lines of forehead, cheek, and mouth lend an expression of nervous determination to the character, but, though strong and decided, he would not scruple also to commit acts of treachery when they served his interests, a vigorous but shifty man, if we may so interpret a sinister look about the eyes and the thin lips. A wreath represents perhaps royal pretensions, and the tightly drawn flesh, the crowsfeet round the eyes and the fulness under the chin point to a possibly early maturity. Whomever this head portrays, it is a real contribution to ethnography, for in a free unstereotyped fashion it gives us clearly all the features which historians attribute to the Thracian character, and combines them in a physical setting which no guess-work could have recovered.!' Too often, as in the crude provincial work from Adam-Klissi, our ethnographical documents are of an inferior order: this work however has real artistic merit, like in kind to the Pergamene ‘Galatians, and like them it aids us somewhat. in unravelling a very obscure history—a history so obscure indeed, that despite the coin it is difficult to find out who is the person here portrayed. The Athenian inscriptions above referred to, are as follows :—*

(1) BAZIAEA PASKOYTIOPIN KOTYOE APETHS ENEKEN THE ΕΙΣ EATON. ANTIFNQTOS EMOIHEEN

O ΔΗΜΟΣ BAZIAEA KOYTYN ΒΑΣΙΛΕΟΣ PAIZKOYTIOPIAOS YON APETHS ENEKEN KAI EYNOIAS THE ΕΙΣ AYTON ANTIFNQTOS ΕΠΟΙΗΣΕΝ.

(2)

Unfortunately both coin and inscriptions have been the subject of much controversy, but one or two facts may be laid down which will lessen the ground of dispute. Almost all numismatists agree that the coin belongs to the Augustan period : as to the relation between its two legends, there is less unanimity, but the most reasonable view seems to be that of von Sallet 3—

1 It is interesting to contrast it with another Thracian head, the Capitoline portrait of the Emperor Maximin, the face of aman who like

2 C.I.A. iii. 1, 552 ; ib. 555. Loewy, 1.4.B., 314, 315. This writer has criticised his pre- decessors exhaustively, and so I have tried not

Kotys had come into not unfriendly contact with a high civilization, yet had by no means lost his barbarism. The differences between the two are as instructive as the points of resem- blance,

to repeat arguments of his, to which I have nothing to add.

8 Beschreibung der Antiken Miinzen (Berlin), i, 334, 3365.

A THRACIAN PORTRAIT. 323

Rhaskouporis (rev.) is the son of Kotys (obv.) and possibly his associate in the royalty.

Secondly, although Mommsen does otherwise, few will hesitate to connect the two inscriptions: in spite of slight differences in spelling, the artist’s name is sufficient to justify this. And the character of the letters and the form of words used again point to the Augustan period.

In order, therefore, to discover the subject of our head, we must find a king named Kotys, of the Augustan age, both son and father of kings named Rhaskouporis.

Several reconstructions of Thracian history have been offered, but as I am unable to accept any of them completely, it will be necessary once more to enter this labyrinth, following, however, only those paths which have any bearing upon our quest.

During the half century previous to the death of Caesar we have evidence of a powerful Odrysian dynasty. In B.c.70,aking named Sadala (1) was reigning :” in 48 another Sadala (2) or Sadales was sent by his father Kotys (1) then king of Thrace to assist Pompeius. This man was pardoned at Pharsala, and, after succeeding his father about the same time, died child- less in 42, leaving his dominions to Rome. Thereupon Brutus overran this part of Thrace and was vigorously supported in his campaign by a Sapaian dynast, named Rhaskouporis‘ (1).

Further, Appian® tells the story of a certain Polemokratia, the widow of a murdered Thracian kinglet, who fled with her children to Brutus and was placed by him at Kyzikos. And there is extant an inscription from a monument erected by a certain Kotys to his parents Sadales and Polemokratia.®

Mommsen’ has combined these three facts: impugning Dion’s truthful- uess, he argues that Sadales (2) did not die childless, that he was the husband of Appian’s Polemokratia, and that the Kotys of the Bizye inscription is their child, and identical also with a king of that name whom we know from other sources to have reigned later. This later Kotys (2) is known to have had a son Rhaskouporis (2), and two brothers Rhoimetalkes (1) and Rhaskouporis (3) all afterwards kings of Thrace. Mommsen’s pedigree therefore connects this dynasty with the previous Odrysian line; but at the cost of denying Dion’s account, and of assuming the existence of an unknown dynasty to whom the Athenian inscriptions may be assigned. Further objections to this theory will be pointed out in the version of Thracian history which I suggest, a version which at least brings all authorities literary and epigraphi- cal into complete harmony.

1 For the form ἑατόν, cf. Homolle, B.C.H., 3 Dion. Cassius, xli., 51, 63. viii. p. 188, and contrast with Loewy, 1. 6.8. 4 ib. xlvii., 25. 316. 5 Appian, iv., 75.

2 Cic. Verr. ii. 1, 24. The figures in 6 Rangabé, <Antiquités Helléniques, ii, No. brackets after the king’s names are inserted for 1236 ; from Bizye. the convenience of the present paper, and do not 7 Mommsen, Ephem. Epigraph, ii. (1875), correspond with any others. p. 253. seq.

324 A THRACIAN PORTRAIT.

King Kotys of the Bizye inscription, I identify with the father of Sadala (2): the repetition of the name Polemokratia is not at all unlikely, and this date suits the character of the letters.1 With the death of his son, the Odrysian dynasty I believe came to an end, as Dion says. The second Polemokratia is described simply as the widow of a Thracian kinglet, whose name the writer does not know: possibly it was Koson,? but this ignorance and the title βασιλίσκος would both be strange, if the murdered man were really an Odrysian dynast as well-known as Sadales (2).

After the death of Sadala (2) two Sapaian princes rose into prominence, the brothers Rhaskouporis (1) and Rhaskos.2 In 48 the former sent only 200 horsemen to assist Pompeius, in 42 he helped Cassius with 3000 and his brother brought the same complement to Caesar: after the “ Liberators’ ” defeat, Rhaskouporis was pardoned on his brother's intercession. The forces which these two princes could summon, show that they had extended their power, presumably over the land once ruled by the Odrysians, and this extension must have been the result of the previous campaign of Brutus, the protector, according to Mommsen, of the Odrysian heir.

This Rhaskouporis (1) I believe to be the father of Kotys (2), therefore also of Rhoimetalkes (1) and Rhaskouporis (2), for the following reasons. First, Strabo, a contemporary authority, describes Kotys (3) the son of Rhoimetalkes (1) as a Sapaian: therefore it is natural to identify him with a Sapaian, not an Odrysian house.

Secondly, this dynasty was very unpopular with the Odrysian tribe.®

Thirdly, so far as I know, the name Rhaskouporis occurs nowhere among the Odrysian kings: in the later dynasty it is frequent while the name Sadala is unknown,

And lastly, this version discovers for us the object of our quest as set by coin and inscriptions—a king Kotys of the Augustan age, at once son and father of kings named Rhaskouporis. Of this king we know further, that he died before 17 B.c. leaving his kingdom to his son who was then a . Ininor.®

When we turn to ask what actions earned for this king a statue from the people of Athens, we must rest content with the vague words of the inscrip- tion. Between Thrace and Athens there were numerous bonds of connexion. Ovid’ addressing a later and more interesting Kotys, nephew of our subject, refers to his descent from Eumolpos, a legendary tie which reversed the historic sequence of events. What civilization had reached Thrace seems to have come mainly from Athens or from Athenian colonies,’ and two later Thracian kings held civic dignities in Athens.®

Kotys and Rhaskouporis were perhaps Phil-Hellene princes, like

1 Cf. Loewy, loc. cit. his theory.

2 Brit. Mus. Catalogue of Coins, Thrace, 5 Tacitus, Ann. iii. 88.

p. 208. § Dion. liv. 20, 34. 3 Cf. reff. above. 1 iin ,ex.) Ponto, 11, 1X. 11.52; 1,9. 4 Strabo, xii. 3, 29. p. 556. Mommsen 8 Dumont, Mélanges, pp. 201,202.

wishes to change this text in accordance with 9 C.I.A. iii. 114, 1077, 1284.

A THRACIAN PORTRAIT. 325

Ariobarzanes II. and III. of Kappadokia,! who restored somewhat of the damage inflicted upon Athens in the Mithridatic and later wars, doing what the kings of Egypt and Pergamon had previously done more splendidly. Or perhaps as Antonius was very popular in Athens and the Thracian kings at first sided with him, there may have been a connexion in this way, but further speculation cannot in the present state of our knowledge lead to any profit- able result, for no historian has chronicled a single act of king Kotys. His coins, which usually represent a more youthful type, are not uncommon, so we may assume for him some material importance, and our portrait indeed represents a character worthy to act a small part in the world struggles of the Romans, though one not likely to lessen the difficulties of his masters. The coins issued by his successor Rhoimetalkes are those of a Roman vassal. Kotys and his son therefore seem to have failed at last to maintain their independence against greater hostile tribes, and his dynasty to have won restoration only by accepting the suzerainty of Rome.

The artist Antignotos has signed a third portrait-basis at Athens :? unfor- tunately only the first name MAPKON has been preserved. Pliny? says that he made ‘luctatores perixyomenum tyrannicidasque supra dictos,’ but at present none of these have been identified: it is possible that the ‘tyranni- cides’ were not as is usually supposed Harmodios and Aristogeiton, but Brutus and Cassius, whose statues were placed by the Athenians near the old ones. From a study of the Kotys head, we can readily believe that Antignotos was one of the first sculptors of the day and would probably receive such a commission.

Under the influence of Augustus, a new spirit pervaded portrait-sculpture at this time, a return to a more abstract and typical art, a tendency to sub- ordinate details to an ideal likeness. If Antignotos was at all touched by this current, this work shows that he knew how on occasion to find escape, for though vivid and full of vigorous thought, it is executed throughout with equal care and individuality. In its truthful modelling of details it recalls several Hellenistic works, following at some considerable interval the ‘Antiochos Soter’ and the ‘Barberini Faun. The slight marking on the eyebrows I am inclined to attribute to a later hand: they are neither like the raised eyebrows on the works above mentioned, nor like the plain treatment of ordinary Augustan heads. The slightly opened lips again might be more easily paralleled on heads before than after Augustus. The names of the other works of Antignotos—the ‘Wrestlers’ pointing to Rhodes, the Perixyomenus yet further to Lysippos— corroborate the position which these technical details suggest. Comparing this portrait with the finer and warmer flesh-treatment of the Rhodians or the Pergamenes, or again with the liquid softness of Antonine art, we are conscious of a certain dryness, but it would be rash to attribute this to Roman

1 Vitruvius, v. ix. i., 6.71.6. i. 357 shows 2 Loewy, 316. that they were honoured in the same way. Cf. 8 H.N. xxxiv. 86. also Hertzberg, Geschichte. i. p. 436. 4 Dion. xlvii. 20.

326 A THRACIAN PORTRAIT.

influence, and we shall be safer in seeing in Antignotos an artist trained in the normal traditions of Hellenistic sculpture. J. ΝΥ. Crowroor.

P.S.—Since the above was written, the head in question has been published by Arndt (Nos. 348, 344). He adds a note. “Kopf cines unbe- kannten alten Griechen........ Im Haar ein Kranz von Oel (7) bliittern ; danach ein Priester? Ein herrliches griechisches Original, ausserordentlich fein in Arbeit und Ausdruck, wohl aus der spiiteren Diadochenzeit.” The period to which he assigns it I may regard perhaps as a corroboration of the

stylistic influence traced above, H Beh als

J. Η. 5. Χν!!. (1897), PL. ΙΧ.

PRIMITIVE CRETAN PRISM-SEALS OF STEATITE.

» " eo

as Pot > ι ᾿Ν

mn ives!) wee μι

9

᾿ ἢ δὴ » * ᾿ μόνης... Ν ι 3 4 = — 4: Ὁ

ΣΝ

τ

ἀντ ai A VERA AS POMC AR. Δ

@9ITAST® 30 2)A22-wecina HATARS.

ἐμάς πο, ἜΧΩΝ

vy q ae ΩΝ

ἜΣ

= ᾿

" ial yMegthia ἐ ᾿ "ΙΝ Ll + Wane vos ἀν» Orig ost, a we

Oa

¥

J. H. 8. XVII. (1897), PL. X.

PRIMITIVE CRETAN PRISM-SEALS OF STEATITE.

Π 7 | ; ἢ J ; τὰ . ν»

el wast) tee κε Be. ᾳ paps, ,

— ue ee, ᾿ ᾿ ν» ᾿ | + -

ΓΔ * Π δ ͵ Pe a iors, ek ie Act a Vets ω a a 8 Ὗ ‘> 5 Cpe Reon Vie ἔν, νεῖν Ὕ ma," Ι 7 ws i 4 7 bal »- γ ᾿ 3 ae VEN τ " ἫΝ a 7" é Ῥ.- “ats 1 ty we 7 4 ae Ν᾽ μὰ ὧν 4 " Ὶ ee oe Ah V4 sh . ‘ ἊΝ »» Ἶ ἰ τ (Ms ,, . “- " , “«" ἠ ay 7 fs ᾿ οὐ a ie . δΣ.- ἊΨ» fn ῃ Φῷ 4 oe e = νά ‘ a ’ oun ᾽ =] i ese τὴ ἃ, ‘ a): » ᾿ ners δ « ve 5 oe ¥ a rT *) 4 [2 ἈΠ ΑΚ τὰ + .

ἢ" Ἂν: ; cic : ‘ Ἢ ἣ 219% MATRA SVMS τς

Ὁ ZJALe-'

ITF) TAITe

FURTHER DISCOVERIES OF CRETAN AND AEGEAN SCRIPT. 327

FURTHER DISCOVERIES OF CRETAN AND AEGEAN SCRIPT: WITH LIBYAN AND PROTO-EGYPTIAN COMPARISONS.

[Puates IX, X.].

Part I.—FurtTHER DISCOVERIES OF CRETAN AND AEGEAN SCRIPT. § 1. Introductory.

In a former communication ! attention was called to an indigenous system of writing in Crete, the earlier stages of which go back, not only far beyond the date of the first introduction of the Phoenician alphabet among the Greeks, but to a period considerably anterior to the most ancient monumental record of the Semitic letters.

From the evidence of ancient Cretan seals it was possible to demonstrate the existence of a form of pictographic writing from its simplest beginnings to a more conventional and abbreviated stage. Side by side with this a variety of data supplied by seals, vases, and inscribed stones, showed the further existence of a linear system of writing, connected with the other and presenting some striking comparisons on the one hand with certain characters found by Professor Petrie in Egypt and by Mr. Bliss at Lachish ; on the other hand with the syllabic script of Cyprus and some Anatolian regions. It was further pointed out that in some instances Cretan linear characters displayed a remarkable correspondence with Phoenician and early Greek letter forms.

It was, moreover, possible to show from the evidence of finds like that of Hagios Onuphrios and from the imitation of certain characteristic ornamental | motives, that the more purely pictorial class of the Cretan seals went back at least as far as the period of the Twelfth Dynasty in Egypt, and to the first half of the third Millennium before our era. I have since been able to accumu- late further proofs of a very early contact between Crete and Egypt, going back to a considerably earlier period than that of the Twelfth Dynasty.

Although, however, various decorative motives in this primitive class of Cretan seals were due to Egyptian influence, it nevertheless appeared that

1 Primitive Pictographs and a Prac-Phoenician cated to the Hellenic Society in November Script from Crete, &c. J.H.S. vol. xiv. 1894 p. 1896. The second part containing the proto. 270 seqqg. ; and, London, Quaritch, 1895. The Egyptian and Libyan parallels has been added first part of the present paper was communi- _ since that date,

328 FURTHER DISCOVERIES

the representations as a whole were of indigenous character,—the later conventionalised pictographs showing perhaps a greater affinity to the ‘Hittite’ characters of Anatolia and Northern Syria than to the Egyptian.

Two further visits to Crete in the springs of 1895 and 18961 have enabled me to add to the material previously collected, and my most recent investigations in the island have resulted in the discovery of one monument of capital importance. I was also able to ascertain the existence of a geo- logical phenomenon which goes far to explain how it was that this island became at such a very early date a centre of the glyptic art, and was thus able to produce the engraved designs on seals which eventually gave rise, by a gradual evolution, to a conventional system of writing. This was the existence, throughout a considerable tract of south-eastern Crete, of rich beds of steatite or soapstone, a soft and, in some of its phases, attractive material, of which all the earlier engraved stones and seals found in the island are composed. Following up a clue given me by Dr. Hazzidakis, the President of the Syllogos of Candia, I found plentiful beds of steatite of a translucent greenish hue, in the valley of the Sarakina stream, about half- an-hour below the site of the ancient Malla; and [ subsequently obtained information of the existence of equally prolific deposits on the coast at the Kakon Oros, a little east of Arvi, and in the range that separates Kastelliana from Sudzuro, in the territory, that is, of the ancient Priansos.

In dealing with the new materials bearing on the Cretan script it will be convenient to begin with the earliest class of seal-stones, presenting designs and characters of a linear kind; to pass thence to the seals on which, though still early in execution, designs are seen of a more definitely pictographic style, and from these to their direct offshoot, the Eteocretan seal-stones with a more conven- tionalized pictographic writing. New examples of the fully developed linear system of writing on seals and other objects will next be passed in review, including the most important object of this class as yet brought to light, namely, a steatite Libation Table presenting part of an inscription.

Attention will finally be called to a prism-seal from Karnak, revealing a connexion between Crete and the Nile Valley at an extremely early period, and to the far-reaching results of this early intercourse on the prehistoric arts of the Aegean world.

§ 2. Primitive Prism-Seal with Linear Characters and Figures.

The remarkable seal-stone seen in Fig. 1 was first observed and described by the Italian archaeologist, Dr. A. Taramelli, who found it in the possession of an inhabitant of the village of Kalokhorio in the Pedeada province. It has now been acquired for the Museum of the Syllogos at Candia.

From its superior size, its somewhat irregular shape, and the rude character of the designs, it claims a very early place in the series of Cretan

+ A short account of my journey in 1896 appeared in the Academy, June 13, 20, July 4, and 18 of that year.

OF CRETAN AND AEGEAN SCRIPT. 329

bead-seals, and from its exceptional character I have preferred to place it in a separate class. It is of yellowish-brown steatite, and, like the other seals,

ΧΑ - OO! alma we:

Fic. 1.—RupE THREE-SIDED SEAL OF STEATITE FROM KALOKHORIO. [7]. (a, section: ὃ, ὁ, d, sides. )

perforated through its axis. Its irregular three-sided section places it in the same category as a perforated triangular steatite with rude linear engraving

330 FURTHER DISCOVERIES

from central Crete, already described in my preceding paper (‘ Pictographs,’ &c., Fig. 17). These examples point to the conclusion that the trilateral bead-seals originate from more or less natural triangular splinters of steatite which, with the surface somewhat smoothed and engraved in the simplest linear fashion, were adapted for wearing by being bored through their axis.

The character of the designs in the present instance bears a great resem- blance to those of certain engraved objects from the Hagios Onupliios deposit at Phaestos,! the early elements of which as is shown by the Egyptian evidence go back at least to the first half of the Third Millennium, p.c. The rude male figure with outstretched hands on the first side of the Kalokhorio seal recalls a figure on a terra cotta cylinder from that deposit.2 On the other hand, the animal—for so it must be interpreted—in the lower part of the ficld on the second facet presents a distinct parallelism with that on the Phaestos whorl.2 Some of the signs or characters also show a certain resemblance to those on tle whorl.

The conclusion to which both the Phaestos whorl and the Kalokhorio seal point is that the linear characters of the Cretan and Aegean scripts go back to a very early period and may be rather derived from the primitive school of engraving in which the objects are indicated by mere lines,—like the first drawings of a child ona slate—than from the more developed pictographic style. The conventionalised script derived from this more advanced style must therefore in the main be regarded as parallel with the linear characters rather than as their immediate source.*

It must still be observed that in some cases both systems—the linear and the more pictographic—show a close approximation and certain common elements. Purely pictographic and linear characters are, as has been already pointed out,’ occasionally found upon the same stone. On an early steatite seal of the four-sided class (‘ Pictographs, &c., Fig. 36), we see a rude figure of a man on one side and on the other three well- pronounced linear characters. This seal, both from its style and material, belongs to an earlier date than what I have called the conventionalised picto- graphic class, and illustrates the fact that linear signs had already been evolved from linear drawings in this primitive period. The same conclusion may be deduced from other examples (Cf. ‘ Pictographs,’ &c., Figs. 29, 30).

1 See Cretan Pictographs and Prae-Phoenician Script with an account of a Sepulchral deposit of Hagios Onuphrios near Phestos. London: B. Quaritch ; New York: G. P. Putnams, 1895, p. 105 segq.

2 Op. cit. p. 107, Fig. 81b.

3 Op. cit. p. 15, (284) Fig. 11a.

4 On the relation of the Cretan Pictographs to the linear characters, more will be found below, p. 358 seqq.

5 ‘Pictographs’ &c. p. 32[J.H.S. xiv. p. 301).

* I have called attention to this point in my ‘ Pictographs’ &e. p. 95 [J.H.S. xiv p. 364]: ‘ In instituting the comparisons (on Table 17),

the pictographic signs have been taken from the somewhat advanced types represented on the Mycenaean seal-stones of Eastern Crete, but inasmuch as the linear forms...go back to a very early date it would not be literally true to say that they are derived from pictographs in the stage represented by, these Eteocretan seals. The actual prototypes of the linear forms would probably have been pictographs of a ruder ‘ graffito’ and almost linear type them- selves, such as we find on some of the most archaic Cretan stones and on the whorls of the earliest settlements of Hissarlik.’

OF CRETAN AND AEGEAN SCRIPT. 331

Yet the community existing between this purely linear and the later pictographic class is well illustrated by the appearance on the very seals above referred to of linear forms of the gate symbol which is also one of the most frequent on the more pictorial class, It is probable that both systems reacted on one another,

§ 3. Karly Pictographie Prism-Seals,

The fresh examples of this class figured on Plates [X., X. are all of steatite three-sided and perforated through their axis.

Nos. 1—5! refer apparently to an owner of flocks and herds. Asin other examples he is seen either standing (PI. IX. Nos. 1, 2) with round-bottomed pots suspended from a pole before him, or either seated or standing and holding a high-spouted vessel (Pl. 1X. Nos. 3, 4) in one case with another before him. No, 5 is unfortunately somewhat fractured, but the object held by the seated figure is more like a drinking-horn.

In three instances (Nos. 1, 3 and 4), the human figure, which must be taken to represent the owner of the seal, is followed on another face of the seal by a goat. In No. 2 its second face contains an imperfect delineation of three human figures? The third face shows a greater variety of symbols—on No. 1 a spider, on No. 3 an uncertain animal, probably a dog, on Nos, 2 and 4 a star or sun with revolving rays, on No. 5 perhaps a four- petalled flower.

The vase-holding seated figure on No, θα shows a general resemblance to those of the above group. It is however to be observed that the vase in front of him stands upside down. On the next face of the seal are further seen four round-bottomed pots, two as if slung on either side of a central pole, and all together contained in a quatrefoil compartment. The possibility suggests itself that we have here the signet of a potter, and that the vessels hung up in the enclosed space illustrate some primitive method of baking pottery. The third face of this stone represents a scene of the chase in which a hound springs from the side at the hind leg of a running deer. This is an interesting anticipation of a scheme that occurs on lentoid gems of the Mycenaean period.‘

On No. 75 we see a standing male figure, the head of aram or mouftlon ° and four globules or pellets. The same number of pellets is found on other seals and agrees with the duodecimal numeration which seems to have been in vogue in the island at a very early date.’

1 No. 1. from Gonias, Pedeada; No. 2. flying in the same way at the hind-leg of a wild Lasethi; No. 3. Koprana, Lasethi; No. 4. _ bull.

near Gortyna; No. 5. Spelia near Lamndn, 5 From Milato.

Siteia. δ On a Mycenaean gem recently found at 2 Of. Cretan Pictographs, p. 75, Fig. 69(J.W.S. Kastri near Turloti in E. Crete there is what

xiv. p. 344). appears to be a representation of a moufflon. 3 From Mallia, Pedeada, in the Museum of This animal is no longer found in the island,

he Syllogos at Candia. 7 Cf. Cretan Pictographs, &c., pp. 73, 74;

4 Upon one in my collection a dog is seen (J. H.S. xiv. pp. 342, 343).

332 FURTHER DISCOVERIES

No. 8} shows two male figures in reversed positions, followed by a goat and two fish. On a Cretan gem in the Berlin Museum, two men are followed by three fish, a group which also occurs on No. 16 below.

No. 9 from eastern Crete exhibits on the first facet three serrated bars, recalling the later spray and tree symbols, (‘ Pictographs’ No. 58, 59). This is followed on successive faces by a horned animal,—deer or goat,—and a hipppocampus, apparently the hippocampus guttulatus or brevirostris of the Mediterranean, which in a modified form seems to have supplied many sea monsters to later Greek art.2 Two hippocampi are also seen on the transitional Cretan stone, Fig. 46 below, now in the Copenhagen Museum. In Crete this marine animal was specially chosen as a symbol by the inhabitants of Itanos at the easternmost corner of the island, where two confronted hippocampi form the principal types on the reverse of its Fifth Century coins.

On No. 10, from Mallia near Chersonésos, the serrated bar appears between two heads of what seem to be short-horned goats. These symbols are followed by three goats’ heads of the same kind, but two hornless. On the third side appears another version of the floral design. Variations of the same figure will be seen on No. 13 headed by the §-shaped double animal already familiar on these early seals. (See ‘ Pictographs’ Figs. 62, 63).

On Nos. 11,‘ 12, and 14, the two latter found at Mallia between Cherso- nésos and the site of the Cretan Miletos, we see rude delineations of pigs, in the second instance a group of three. The pig is also found on a three-sided seal of the later class (‘ Pictographs, &c. Fig. 24. ¢). The other three animals on No. 12 must be regarded as uncertain. The long-legged, long- necked birds repeated on No. 14, as well as on a Cretan seal in the Copenhagen Museum® recall an example on another early seal stone, (‘ Pictographs’ ἄς. Fig. 64 a.). In the case of the latter stone the suggestion has been made that the bird may represent an ostrich, once more attesting the early commercial relations between the Aegean island and the African Coast. It is remarkable that both Nos. 11 and 12, and 14, show the same succession of pigs and long-legged birds. We have here another instance on the early seal stones of the grouping together of symbols in a parallel sequence, which shows, if any proof were still needed, that these figures were not chosen at haphazard, and that the collective group on the different sides of the stone has a connected and cumulative meaning.

The third design on No. 14 seems to be a spider very naturalistically

Keller, Tier- und Pflanzenbilder auf Miinzen und Gemmen, p. 73 and Taf. viii, 39; xii. 34,

1 From Milato. 2 Cretan Pictographs, &ec., p. 70, Fig. 59

{J.H.S. xiv. p. 339].

3 Salinas, Ripostiglio di monete antiche di argento (Rome 1888 p. 7), regards one of these as the prototype of the so-called sea serpent (pistrix) seen on so many Sicilian and Magna- Grecian coins. Cf. Imhoof- Blumer und

35 ; xiii. 18.

4 Bought at Athens by Mr. J. L. Myres and presented by him to the Ashmolean Museum. The seal is clearly of Cretan fabric.

5 T have to thank Dr. Blinkenberg and the Director for an impression,

OF CRETAN AND AEGEAN SCRIPT. 333

rendered in profile! The spider has been already seen as viewed from above, on No. 1 and it recurs on No. 15 so that it seems to have been a favourite Cretan symbol. We shall find it again on two stones representing the more conventionalised stage of the ‘ pictographic’ script.?

The frequency of the spider on these seals is specially remarkable when itis remembered that this insect is conspicuous by its absence on the engraved stones, and coin-types of the classical period of Greece, though other insects such as the ant, the bee, or the cicada are common enough. In Greek mythology the spider appears in the legend of Arachné as the representative of Lydian textile art, and with that old Anatolian race this insect evidently typified the spinning industry. The undoubted affinities between the earlier indigenous elements of Crete and those of Western Asia Minor makes the prominence of the spider in its primitive pictographs the more suggestive, and we may infer that here, too, the insect as a symbol indicates the possession of looms. In this connexion it is worth while recalling the fact that the three seals representing spiders, of which the provenience is exactly known, come from that part of the island in which the Cretan Miletos, now the village of Milato, the reputed mother-city of the better-known Carian and Ionian homonym, was, from Homeric times onwards the chief civic centre.‘ The localization of the myth of Arachné at the once Lydian and Maeonian Kolophén, and the occurrence of the spider signets in the mother-country of the not distant Milesians are, perhaps, not altogether accidental coincidences.

It will be shown in a succeeding section® that the spider,—probably with the same significance,—recurs on a primitive class of Egyptian cylinders and on Libyan seal-stones. On the stone No. 15 the spider is coupled with a floral emblem resembling that on No. 5, and the solar or stellar disk with revolving rays. The former association recalls the fact that, on one of the conventionalised pictographic seals referred to,® the spider and a similar quasi-floral design succeed one another at the end of one line and the beginning of another.

The two birds on No. 16, from Mokhos, Pedeada,’ are shorter-legged and apparently of a different kind from those described above. They somewhat recall the bird on aseal previously described (‘ Pictographs,’ &c., Fig. 65a) from central Crete, in which I ventured to trace a resemblance to a cock. The group of three fishes also recurs on another early seal (‘Pictographs, &c., Fig. 59c). The design on the third face of No. 16 is a four-handled vase, a type which is also seen on No. 110.

On another three-sided seal, not figured in the Plates, from Kavuse in

1 It is probable that the two objectsonaCretan _—classischen Alterthums. gemin the Berlin Museum (‘ Pictographs’ &c. 4 The seals Nos. 14 and 15, were from Fig. 59b.), described by me loc. cit. as ‘polyp- Mallia the site of an ancient settlement on the

like,’ are alsc intended to represent spiders. north-east coast, a little to the west of Milato. 2 See below, p.335, Fig. 5b and p. 336, No. 1 came from Goniais in the hill-country Fig. 60. above.

3 No single representation of a spider occurs 5 See below p. 364, Fig. 29 and p. 368, Fig. 32. n Imhoof-Blumer und Keller, Tier- und 8 See below, p. 336 Fig. θα, ὃ. Pflanzenbilder auf Miinzen und Gemmen des 7 In the Museum of the Syllogos at Candia.

H,S.— VOL, XVII, AA

334 FURTHER DISCOVERIES

eastern Crete,! the disk with revolving rays, seen on Nos. 2, 4 and 15, occurs in a variant form. It is accompanied on the other faces of the stone by the figure of a rude animal and a goat. The rayed disk is also found in juxtaposition with a goat’s head on a seal of the later class (‘ Pictographs,’ &c., Fig. 5980). In connexion with the goat it has already appeared in No. 4 above.

Fic. 3.—YELLOW STEATITE, ELUNDA (Otous). [83].

§ 4. Later Seals with Conventionalised Pictographs. Class A. Transitional (in soft stone).

The three following seals, Figs. 2,3 and 4, which stand in a very close relation to one another, though in many respects fitting on to the preceding class, show an elongated form more characteristic of the later series with the advanced pictographic script, and of the time when hard stone such as cornelian or chalcedony had begun to supersede steatite for such purposes. This is particularly the case with Fig. 3 where the group of three . high-spouted vases corresponds with that on a prism-shaped seal already published (‘ Pictographs,’ &c., Fig. 21). This seal, though also of steatite and of primitive execution, is there, nevertheless, classed with the later pictographic group, owing to the appearance on the second face of the stone of two of the most characteristic signs belonging to that series. Both it and the present seal may, in fact, be regarded as transitional in type.

Fig. 2, found at Mallia, between Chersonesos and the site of the Cretan Miletos (Milato), and Fig. 3, found at Eluda, or Elunda, the site of the ancient Olous, show on the first face in the order here given, a ship, and in the second respectively, a single vase and a group of three vases. The third place is filled

1 Seen by me there in 1893,

OF CRETAN AND AEGEAN SCRIPT. 335

on Fig. 2 by an animal, on Fig. 3 by a group of comb- and rake-like objects, the arrangement of which, however, is of decorative origin.

The ship on Fig. 2 is remarkable from the fact that the mast is only connected by ropes with the forepart of the ship. The other vessel (Fig. 3) has, as usual in all early Cretan seals and gems representing ships, ropes attached to the mast on either side. The discovery of this seal on the site of Olous is interesting, as conveying a hint of the very early maritime enterprise of that port,—now the land-locked lagoon of Spinalunga,—whose sheltered waters must have afforded every facility for primitive navigation.

Fig. 4.—STEATITE, CRETE. (COPENHAGEN MusEum.) [#].

With these two maritime signets may be grouped Fig. 4, a Cretan specimen in the Copenhagen collection,! also of steatite, and belonging to the same transitional class. It exhibits on its first face an instrument, perhaps an arbelon for cutting leather, which is of :frequent occurrence in the later series. It is here placed between two trumpet-like scrolls, also found on some

later seals.2 There follow on the two other faces a pair of hippocampi and two S-like scrolls,

Fic. 5.—STEATITE PrisM-SEAL, FROM IMPRESSIONS OBTAINED AT CANDIA. [f].

The dull white steatite seal, Fig. 5, taken from an impression obtained at Candia, is a typical example of the earliest of the more advanced pictographic class. The soft stone of its material, and the style of its engraving, place it,

however, very near the transitional seals, Figs. 2, 3 and 4, with which it is here grouped.

Peres a es lg, BE eee See

1 Impressions of this seal and that figured on 2 See below, p. 342, Fig. 11 and p. 843 Plate X. No. 18 were due to the kindnessof Dr. Fig. 13. Chr. Blinkenberg. AA2

336 FURTHER DISCOVERIES

The collocation of the gate and bent leg symbols occurs on three other seals (‘ Pictographs, Figs. 220, 25a, and 34)b,—the gate somewhat variant in the latter case), We have here, therefore, another proof that the grouping of these pictographic characters was not arbitrary but that they were combined according to a definite system in order to give expression to ideas.

This repeated combination of the bent leg and gate makes it prob- able that the 1shaped sign coupled with the gate in the group of linear characters on the early seal (‘ Pictographs, Fig. 36d) is the linearised equivalent of the bent leg. The spider of the following facet has been already noted on Nos. 1, 14 and 15 of the early series as probably connected with the spinning industry. The animal is perhaps a wolf; witness the appearance of the wolf's head among the more abbreviated pictographs,! of which a fresh example is given below.? It is possible, however, that in all cases we have todo with a dog of wolf-like breed. At the present day the dogs in parts of the Balkan peninsula are hardly distinguishable from wolves in their external aspect.

Fic. 6.—Four-sIDED SEAL FROM SITEIA, [3].

Class B. Seals with fully-developed conventionalised pictographs. (Hard stone).

Fig. 6, a perforated quadrangular stone from Siteia, is a pictographic seal of considerable interest. The seated figure with which the first line (as given above) begins, recalls the same subject on so many of the seals of the earlier class, and establishes a new link of connexion with them. The spider at the beginning of line ὃ and the quatrefoil of line ὁ also recur on the earlier series, The second sign of line ὃ may be regarded as a variant of the ‘arbelon’

1 See ‘Pictographs,’ ete, p. 41 [υἱ. 7. δ. xiv. p. 310, No. 43]. * See below, p. 3438,

OF CRETAN 'AND AEGEAN SCRIPT. 337

symbol (‘ Pictographs,’ No. 16), while the third is certainly the adze (No. 22). The other types are already well ascertained, namely, the gate,! double axe,? mallet,’ spiral,* and mountain-symbol.®

Fig. 7 represents a white steatite seal found in Crete and now preserved in the Central Museum at Athens, It resembles in every way the stone seals of this class. An imperfect figure of this seal was given as far back as 1872 by Dumont in his Inscriptions Céramiques de la Grece.® Dumont—who erroneously described its material as ivory—compared it to the gladiatorial tesserae, and explained the ship as an allusion to the naval sham-fights of the amphitheatre. The fourth face of the seal he allowed to be enigmatic, but in the reduplicated symbols of line 3,—similar to that above identified with the plough,—he saw pairs of wrestlers, and in the goat’s head an aplustre. The comparative materials now collected will at least have served to set at rest some of these speculations. They illustrate the difficulty, which all archaeologists must experience, in interpreting isolated objects of an unprecedented type.

Fic. 7,— WHITE STEATITE SEAL, CRETE. (CENTRAL Museum, ATHENS.) [#].

The two broader sides of this seal seem to stand by themselves; to judge by the bird and ship, they are somewhat more pictorial in character than the others. Both in these and the others we notice that the position in which the same sign is placed is capable of variation. The spray or tree- symbol occurs with its central stem running upwards in ὦ and downwards in ὁ. The plough-like symbol in ὁ faces two ways. The instrument at the

1 ¢ Pictographs,’ No. 25. 6 Pp. 415, 416. No attempt was made to

2 Ib. No. 10. represent face a. Iam indebted to Professor 3 7b. No. 18. Halbherr for this reference and to Dr. Stais and 4 Ib. Nos. 69, 70. Μ. Gilliéron for a cast of the object in question.

5 Ib. No. 66.

338 FURTHER DISCOVERIES

right end of ὁ and d is seen in reversed positions. Most of the signs are placed as if to be looked at horizontally, but the two animals’ heads—the second character from the right end of face c, and the fourth from the left end of line d—and the mountain sign in ¢, are placed as if the column was to be read vertically. Their relative positions seem to indicate that line ὁ is to be read from left to right, and d from right to left— another instance of the bouwstrophedon arrangement already noted in other cases.

The ship on face ὃ recalls a somewhat similar example on a four sided chalcedony seal from Crete (now in the Berlin Museum), figured in ‘Picto- graphs,’ &c., No. 34. Here as there it may betoken that the seal belonged to a merchant who traded over-sea. The characteristic ends, like an open beak, and the double rudder recur on lentoid gems of Mycenaean date from Crete.

Several of the signs represented on the present seal are already familiar on the Cretan seal-stones. Thus we find the arrow-head and shaft (‘ Pictographs,’ No. 186.), the ‘arbelon’ (7b. 16), the mallet (2b. 18), an instrument of a similar kind, but with a triangular handle, the ‘plough’ already seen on the seal Fig. 6 above, and another indeterminate object (4 from the left end of line d), also paralleled by the second symbol on the same seal. The ship, already noticed (‘ Pictographs,’ No. 32) the tree-symbol (7b. No. 58), the crosses, plain and knobbed, the goats’ head (2b. No. 35), the ‘deer-horns’ (1b. No. 38) the figure like an imperfect caduceus (1b. No. 71), all recur here. In other cases we have variants of known forms, thus the zigzag (3 of 1. 2) seems to be the same as ‘ Pictographs, No. 75 with a terminal flourish, the cross-legged bird may be regarded as an alternative form of No. 47, and the animal’s head (No. 4 of 1. d) may be identified with the ass’s head of the above-cited seal (Fig. 62).

Six of the signs here represented, however, occur apparently for the first time. These will be considered separately on pp. 339, 340 below.

A noteworthy feature of this seal is the frequent repetition of the symbols in the same line. Thus in line 6 we have the tree-symbol occurring twice separately and in a group of four. In line ὁ the plough-sign appears six times, divided into three groups of two each by that which seems to signify mountains or, more generally, land. In line d the arrow occurs three times.

The same characteristic has already to a certain extent been exemplified by other Cretan seals. Thus the vase symbol occurs three times in succes- sion on the stone engraved in ‘ Pictographs,’ Fig. 21 and again in Figure 36 above. The cross pommée begins and ends another line of a seal (tb. Fig. 34d) and the S symbol is twice repeated in the same way (id Fig. 2la and 23c).

In ancient Egyptian the plural was sometimes formed by repeating a hieroglyph of either the ideographic or alphabetic class, three times, and reduplications of signs are also frequent. Such repetitions are, however, especially characteristic of the Hittite inscriptions. In the first line, for

OF CRETAN ‘AND AEGEAN SCRIPT. 339

instance, of the Hamath Inscriptions, Nos. 1 and 2,1 we find one sign repeated four times in two successive groups of two, and another forming a group of three.

It is worth remarking that the beginning of line d of the present seal, which from the direction of the goat’s head may be taken to read from right to left, contains the same symbols as those of the last line of the quadrangular stone seal from Siteia (Fig. 6 above), though the arrangement is somewhat different and the goat’s head is here substituted for the Ssymbol. In both groups as will be seen from the comparative figures below Fig. 8, we find the oblong instrument with triangular handle, the plough (in the present case twice repeated), and the ‘ mountain,’ or ‘land’ sign. It is highly improbable that this parallel grouping is accidental.

Bo

ΓΤῚ

ANNSi

Fig. 8.—CoMPARATIVE Groups ΟΕ SYMBOLS FROM Fias. 6 AND 7.

The following is a detailed list of the conventionalised symbols that occur on the two last-mentioned seals (Figs. 6, 7) for the first time, including the spider already seen on Fig. 50. For convenience of reference the numbers follow on to the list of pictographs? in my former work,

83.

a δ 84. Apparently an ass’s head. Compare the charac- teristic Hittite sign 4 (Wright, Zmpire of the Hittites, Plates gga VIII, IX., X., ΧΙ, and XIX. 5, Jerabis). 455 85. The Spider. Cf. p. 338 above, and for Libyan and 85. Proto-Egyptian parallels, pp. 364, 368 below.

85d is from Fig. 50 above.

1 Wright, Emp. of the Hittites, Platei, H. 1 and H. II. 2 Pp. [802] 33-[315] 46.

340

86.

87.

88.

89.

90.

91.

‘palace’ or ‘altar. We character in the Cretan

FURTHER DISCOVERIES

Compare the floral designs on the earlier series of prism-seals.

I have placed this sign in the above position as it — seems to represent the primitive form of plough, in

which the pole and the share-beam were in one piece, the handle only being attached.

(No. 4 of Fig. 7a). Seems to represent a woman’s breasts. Compare the Egyptian sign ye Mna.=a

nurse, &c.

-- (No. 8 of Fig. 7a). Apparently ἃ gourd.

may possibly be compared.

(No. 2 of Fig. 7d). A kind of crook. The Egyptian | 5

N

A

Ὃ

(No. 5 of Fig. 74). The sign is here placed with the pro- jections uppermost since some other symbols in this line— notably the familiar instrument No. 11, and that at the end —are in a reverse position to that in which they are usually found. Placed as above, this character is identical with the Egyptian =ha, ah, the meaning of which is a

have here therefore a clear example of a

series directly borrowed from the Egyptian.

(No. 6, 10 and 18 of Fig. 7d). It resembles the stem-less Mycenaean type of arrow-head, here shown without the shaft.

(No. 7 of Fig. 7d). Possibly a mirror; the more oval

figure at the end of line a is apparently only a variant of this.

Fig. 9a and ὦ, a white agate with translucent veins, from Gortyna, belongs to a class already signalised in my former communication ! as seal- stones with a single engraved face and with their upper part convoluted.

* Pictographs’ &., p. 19 [288], Fig. 21 and p. 29 [298], Fig. 38.

OF CRETAN AND AEGEAN SORIPT. 341

The lower part of its field is occupied by a conventionalised lion’s head full face, with a kind of fleur-de-lys, more probably an abbreviated palm-tree ἢ rising above it. Above this—which may, perhaps, be regarded as a badge of a more personal character—are two symbols, a kind of extended N, and what appears to be a species of polyp.

ψ

I

σι:

Fic. 9.—CoNVOLUTED AGATE SEAL-STONE FROM GORTYNA. [3].

This conjunction, again, is of great interest, since the same two symbols occur in juxtaposition and attached to one another by a kind of network on the four-sided seal stone (‘ Pictographs,’ &c., Fig. 34d). This network, or cross-

a b

Fic. 10.—CoMPARATIVE GrRouUPS OF SYMBOLS.

hatching, is frequently found as an adjunct of Cretan symbols. It does not seem to have an independent value, being, sometimes, a merely ornamental fill-up, covering the whole background of the seal as in Fig. 11, to be described below.2 Occasionally however it seems to mark off one symbol from others

Compare op. cit. p. 43 [312] No. 5and the suggests the palmettes seen behind conven- Hittite fleur-de-lys symbol from MHamath tionalised lions on one of the shields from the (Wright, Empire of the Hittites, Pl. iv., Idaean Cave, Halbherr and Orsi Antichita dell’

os ii. 2and 8). Inthe presentcase however Antro di Zeus, Atlas Tay. ii. its conjunction with the lion’s head 2 P. 342 ; of. too, ‘ Pictographs,’ &o, Fig. 33a.

342 FURTHER DISCOVERIES

of a group,! or, as in the above instance, to bring two into a connexion separate from the rest.

The ‘ polyp,’ sign seems to have played an important part in the Cretan series. We shall find it in connexion with linear characters in the important monument to be described below.?

§ 5. Stgnet-shaped Stones with Conventionalised Pictographs and Other Figures.

In the course of recent explorations in the Eparchies of Siteia, Girapetra and Mirabello, I came across examples of a wholly new class of Cretan seal- stones, in shape very much like modern seals, cut out of jasper and cornelian ὅ (Figs. 11, 12, 13, 16,17). Though of smaller size, their essentially modern form shows a certain parallelism with some Hittite and Syrian types, amongst which, as in the case of the silver seal from Bor, metal forms also occur.

These Anatolian types are inferior, both in form and material, to the Cretan. They are thicker and heavier, and instead of the jasper and cornelian, are formed of haematite,—which gives them a very metallic appearance,—and of light-coloured steatites. Examples have been found both at Tyre and at Sidon,‘ and one of a similar form was obtained at Palaeokastro on the Laconian coast.°

The signet) shape of these Cretan stones and the analogy that they present with Hittite seals, is of special value as showing that the symbols engraved on them had a direct personal significance.

The designs themselves are both pictorial and ornamental and of the conventionalised pictographic type.

Fig. 11.—GREEN JASPER SIGNET, STO DAso. [3].

The seals, Figs. 11, 12, must be included in the true pictographic class. Fig. lla and ὁ, of green jasper, was found in a prehistoric phrourion called

1 Cf. Op. cit. Fig. 840, where the central sign ὅς. No. 88, p. 10 and Taf. 8. See below. is thus marked off, 4 These are in my own collection. That 2. See p. 359, Fig. 27. from Tyre shows a ‘hatted’ Sphinx boldly cut

® Since this was written, a similar example, also from Crete, has been published by Dr. Furtwangler in his description of the engraved stones in the Berlin Autiquarium, Beschreibung

in a style somewhat recalling the coarser Melian work.

5 In the Ashmolean Museum, Oxford. See ‘ Pictographs,’ &., p. 74. [J.H.S. xiv. p. 848.]

OF CRETAN AND AEGEAN SCRIPT. 348

’Std Daso in the upland glen of ᾿βύὰ Limnia, visited by me on the way from Xero to the site of Ampelos, The symbols consist of the ‘ polyp, the goat’s head, a coil very like the figure 6, and the mallet, of which all recur on other seal-groups, the 6-shaped scroll answering to the trumpet on Fig. 4 above.

Fic. 12.—Rep CorngeLian SicNer From KEDRIE NEAR GIRAPETRA (HigrapyTna). [{].

Fig. 124 and ὃ was found at. Kedrie, above Girapetra, an ancient site, abounding in Mycenaean remains, probably answering to the ancient Larisa, which stood in the neighbourhood of Hierapytna. The seal is of red cornelian and of exquisite work, though the top is slightly broken. The quatrefoil and moulding is most delicately wrought, and the intaglio on.the face of the seal is finely engraved. The subject is specially interesting as representing the wolf’s head with protruding tongue, a symbol equally characteristic of the Cretan and the Hittite series, standing alone within an ornamental border. It follows that this symbol could represent some object or idea by itself, without copulation with any other sign.

Fic. 18.—GREEN JASPER SIGNET FROM SITE OF PRAEsOS. [4].

On a green jasper signet of the same class from the site of Praesos (Fig. 13) three coils resembling the third symbol on Fig. 11, are symmetrically grouped, and the design must be regarded as of decorative origin, whether or not the three coils had afterwards acquired a more definite meaning. The design, in fact, goes back to a triple scroll which already occurs on a very early class of Cretan button-seals of dark steatite, on others of which distinct imitations of Twelfth Dynasty scarab decoration are visible.’ An example of this class with a triple coil from central Crete is given in Fig. 14, A still further link in the chain is supplied by Fig. 15, a seal of brown steatite, which has been placed among the Assyrian specimens in the Louvre, but the Cretan

1 The seals ¢., 9. and ἢ. Fig. 49 in ‘ Pictographs’ &c., p. 58 [J.H.S. xiv. p, 827] belong to this class.

344 FURTHER DISCOVERIES

origin of which may be regarded as highly probable. It shows the same kind of trefoil scroll as that of the button-seal, but in a more developed form. It has at the same time acquired a stem and taken the characteristic signet

ἧι -.-:..- (3 ὃ

Fic. 14.—STEATITE BUTTON-SEAL: CENTRAL CRETE. [1].

shape. It would thus appear probable that the very ancient sub-conical type ot Cretan bead-seals,—perhaps under the influence of Anatolian example,— gradually developed into the signet proper. The specimen represented in

a b Fic, 15.—STEATITE SicnNeT, Louvre. [4].

Fig. 15 shows that this evolution was‘already effected in the prae-Mycenaean period. The three scrolls of the Mycenaean signet, given in Fig. 13, will thus be seen to preserve arecord of its remote ancestry on Cretan soil.

Fig. 16.—YELLOW CORNELIAN SIGNET FROM Kuapra. [3].

A good example of the pictorial style is afforded by a yellow cornelian signet from Khadra, in Siteia (Fig. 16). It represents two wild goats browsing on a rocky peak, and is of great importance as supplying from its artistic style a chronological equation for the pictographic seals of the same class.

OF CRETAN AND AEGEAN SCRIPT. 345

The purely naturalistic treatment of the design distinguishes it alike from the rude representations of the geometrical class and the conventionalism of the orientalising school of engraving, illustrated by the ‘Melian’ gems. The intaglio on the other hand lacks the boldness of the earlier Mycenaean art and must. be placed somewhat late amongst objects of that category. The interlacing scroll work round the centre of the stem of the seal shows a certain approximation to the guilloche ornament frequent on Cretan pithoi from the eighth century onwards, though the more oval form here seen still bears a closer affinity to some Egyptian scarab-borders of the Eighteenth Dynasty.! On a jasper seal of the same type from Goulds (Fig. 17) is seen a lion of conventional pose, the head of which, except for the absence of the fleur-de-lys, bears a great family likeness to the lion’s head on the convoluted seal, Fig. 9, described above,— an interesting indication of the synchronism of these two types of seal.

Fic, 17.—JaspER ΒΊΟΝΕΤ, Gounls. [8].

A similar figure of a lion also occurs on a triangular prism-seal of the elongated class in the British Museum. Like the others it was found in Crete, and the designs on all three faces are in the same, curiously mannered style.2 It may be regarded as one of the latest representatives of its class, which is thus seen also to have overlapped the ‘signet-shaped stones’ with which we are dealing.

A broken crystal signet, with a lion of a conventional type, allied to the above, was also observed by me m the village of Mallia, and a certain approximation to the later class known as ‘Melian’ is unmistak- able in these types. This is further borne out by a Cretan signet stone of the same kind in the Berlin Museum.’ It is of yellow jasper and bears two dolphins with spiny backs, the general character of which betrays distinct affinities with certain Melian types. The pellets surrounded by dots, which occupy the central space between the dolphins on this seal, are also suggestive of a somewhat late date. A similar dotted rosette is seen in the field of an archaic scaraboid in the British Museum.‘

1 #.g. Scarab of the Princess Nefrura (6. 1500 B.c.), Petrie, History of Egypt during the Seventeenth and Eighteenth Dynasties, p. 78 Fig. 39.

2 B.M. Catalogue of Gems, No. 99. It is thus described, p. 45: ‘Triangular prism with ounded ends (a) Lion to 1. Chiefly executed by means of circles and semi-circles (Ὁ) Goat lying down to /: tree(?) and circles in field. (c)

Deer with large horns lying down tol; circle in field. Green Jasper. Crete.’

3 Furtwangler, Beschreibung der geschnittenen Steine im Antiquarium, No. 88, p. 10 and Taf. 3.

4 B.M. Catalogue of Gems, Pl. B. 113 Ῥ. 47. ‘Deer standing to/., looking back and suckling young ; branch in field and pattern of drilled holes above, Hematite. Egypt,’

346 FURTHER DISCOVERIES

There are, it will be seen, strong indications that the Cretan class of signet-shaped intaglios continued in use to the later Mycenaean period in the island, and coincides in the main with the whole duration of what may be called ‘the Early Hard-Stone Period’ of Aegean glyptic art. That period, which answers to the period of Mycenaean art in its widest extent, was characterised by the use of harder materials, such as cornelian, jasper, crystal, and amethyst for engraved seals and gems. The more primitive artists of the preceding age had confined themselves to the soft steatite, and in the time of diminished technical skill, characterised by the Melian class of ‘island stones’, which succeeded the close of the Mycenaean period, the engravers relapsed into the use of the same soft material. The ability to work harder stones may have survived somewhat longer in Crete,’ but the general tendency of the evidence precludes us from bringing down even the latest examples of this Cretan class of signets beyond the eighth or ninth century before our era. The earlier and bolder types go back considerably before that date.

§ 6. Seals and other Objects with Linear Signs.

The lentoid bead of dark steatite, Fig. 18, was found on the site of Knésos. It seems to be an early representative of its class, otherwise so frequent among Mycenaean gems, The engraving here is of a linear kind, and is very different from the bold cutting usual on gems of that period, and the dark steatite of which it is composed, though not unknown among the Mycenaean intaglios of Crete, is more generally associated with primitive work.

Fic. 18.—LentTorp BEAD oF Dark STEATITE, KNosos. [3].

In the centre is a kind of dart or arrow symbol with a lozenge-shaped butt, and on either side of this, two branches or sprays. These vegetable motives with a star between recur on another dark steatite lentoid gem of the same character from central Crete,2 on the other side of which are two more

1 An octagonal signet of simple conical form fact that in Crete too during the succeeding and of green steatite from Crete in the Berlin period there was a return to the softer material. Museum, (Beschreibung &c. No. 81, p. 9 and 2 In my collection.

Taf.. 2.) representing a Sphinx, exemplifies the

OF CRETAN AND AEGEAN SCRIPT. 347

sprays of a different shape. A double spray also occurs on an early lentoid gem of light green steatite from Amorgos.' All three gems agree in the early character of the design and material, though the latter is more deeply cut. They belong in fact to a well-marked though hitherto unrecognised class of prae-Mycenaean or proto-Mycenaean lentoid gems in soft stone.

The early character of the present gem lends a special interest to the two linear signs which appear outside the spray on either side of its margin, and which are almost identical with the Cypriote signs A =/o and [6 =e.

Fia. 194a.—BLack STEATITE Wnhort, Knosos. [?].

Fia. 192.

The black steatite whorl, Fig. 19a, was also found on the site of Knésos. The monogrammatic characters (Fig. 19d) on its upper circumference have in some respect such a comparatively modern aspect that they might be thought to be a recent addition. A minute examination with a strong lens reveals the fact, however, that the edges of the incisions are slightly worn and that in fact they belong to the same date as the whorl itself,—probably the latest prae-Mycenaean period. The characters themselves, moreover, find some close analogies among certain primitive signs found in Crete and else- where. This may be seen from the comparative forms given on Fig. 20. Of these a is from the vase handle found at Mycenae,? 6 from the early Cretan pot found at Prodromos Botzano, ὁ is a proto-Egyptian sign from Naqada,‘ d, a form of S on the Minaeo-Sabaean inscriptions of Southern Arabia which go back to about 1500 B.c.®

The monogrammatic sign reproduced in Fig, 21—long-stemmed like the central character of Fig. 18—was engraved on a sherd of pottery, picked

1 In my collection. 3 Pictographs, etc., p. 10 [279], Fig. 56.

3 Tsountas Μυκήναι, p. 214, Figs. 3; Tsountas 4 Petrie, Nagada, Pl. LIV., No. 262. and Manatt, The Mycenaean Age, p. 269, Figs. 1, 6 Fritz Hommel, Sid-Arabische Chresto- 138, 139, mathie.

348 FURTHER DISCOVERIES

up on the height of Keraton, one of the loftiest among the ‘Cyclopean’ strong- holds of Crete.1_ The sherd on which this graffito occurs is of a reddish and some-

Fic. 20.—CoMPARATIVE EXAMPLES OF MONOGRAMMATIC SIGNs.

what micaceous clay of the same character as that of hundreds of plain fragments of vases, which from their association here with portions of painted Mycenaean

Fic. 21.—LINEAR CHARACTER ON SHERD FROM KERATON.

larnakes seem to represent a local fabric of that period. There was no trace of any later Hellenic occupation of this primitive stronghold, which from

Fic. 22.—YELLow STEATITE SEAL, KALAMAFEA. [3].

a height of about 2000 feet, commands a large part of the southern coast of the island, from Girapetra to the spurs of Ida.

1 See Academy, July 18, 1896 (p. 54

OF CRETAN AND AEGEAN SCRIPT. 349

The seal Fig. 22, was found in a primitive akropolis above the village of Kalamafka in south eastern Crete, where I obtained it in April, 1896. It is even more purely natural in its formation than the rude three-sided stone already mentioned in Fig. 1. It is simply an almost unworked finger-end of steatite which seems to have been thought handy for sealing purposes, and the end of which has been cut flat and engraved with three characters one over the other. Of these the top one, a plain oval and the lowest, perhaps intended to represent a pair of curving horns, are new to the Cretan series. The central sign somewhat resembles the ‘polyp’ symbol, not infrequent in the pictographic series (see above p. 343), but the two upper tails are here more elongated. In its more usual form, it occurs with linear characters on the libation-table from the Diktaean Cave to be described below.! The horned symbol which

occupies the lowest place on this signet somewhat resembles the Egyptian hieroglyph ap.

§ 7. Inscribed Vase from Cerrgo.

The island of Cerigo,the ancient Kythera, may be regarded as a stepping- stone between Crete and the Peloponnesian mainland. Professor Sayce informs me that a prism seal with a variety of conventionalised pictographic

Fic. 24,—CHARACTERS ENGRAVED ON VASE FROM CERIGO.

symbols has been recently found in the island, apparently in company with a lentoid gem of the ordinary Mycenaean type. I have not however been able to obtain an impression of the seal.

Thanks to the kindness of Dr. Stais and of its proprietor, M. Spiridion Stais, Deputy for that island, 1 am able to give a representation of a small marble vase (Fig. 23), also found in Cerigo, containing three characters (Fig.

1 See p. 352. H.S.—VOL. XVII. ΒΒ

350 FURTHER DISCOVERIES

24) the two latter of which at any rate are of the linear class. The material employed and the simple form of the vase, seem to show that it belongs to the latest prae-Mycenaean or ‘ Amorgian’ period.!

Of the signs represented, the second greatly resembles the Egyptian ®

Sep = times (Vices). The third sign has been regarded in the first part of this work? as a form of the eye-symbol. It may also be compared with the

hieroglyph ©) signifying the solar disk.

The first sign, is of a more remarkable character, and has the appearance of a truncated obelisk standing on a base, with another slab resting on its summit. Obelisk-like figures are found among Egyptian hieroglyphics* but in this case the slab above and below suggests an altar, or perhaps, an aniconic image, analogous to that representing the Paphian goddess.

§ 8. Inscribed Libation Table from the Diktaean Cave.

Hitherto, with the exception of some more or less isolated signs on the gypsum blocks of the prehistoric building at Knésos, the evidence of the early Cretan Script has been confined to seal stones and graffiti on vases. I am now able to call attention to a monument of a different class, bearing what appears to be a prehistoric dedication in well-cut characters belonging to the linear type of the Cretan writing.

The scene of this discovery was the great cave on the steep of Mount Lasethi, above the village of Psychro, which must certainly be identified with the Diktaion Antron of the Lyttian traditions. It lies, in fact, only four and a half hours distant from the site of Lyttos, with which it was connected over a low mountain pass by what appears to have been a very ancient road- line. This cave, according to the Lyttian legend preserved by Hesiod,‘ was the birthplace of Zeus, and the votive relics discovered in extraordinary abundance attest, in fact, the existence there of a cult identical with that of the Cave of Zeus on Mount Ida. .

The ancient remains of the Psychro Cave were, for the first time, explored in 1886, by the Italian archaeologist, Professor Halbherr, in company with Dr. Hazzidakis, President of the Syllogos of Candia. In their work on the

1 The inscription is given in Tsountas and 3 The ideograph haz is a rounded stele on a Manatt, The Mycenaean Age, p. 279, but the base; ἐχῆ, an obelisk also on a base. first sign is there imperfectly rendered, the upper 4 Theogonia v. 477 seqq. (Rhea has taken slab above the truncated obelisk, of which dis- counsel with her parents Ouranos and Gaia) :

tinct traces are visible, having been omitted. πέμψαν δ᾽ és Λύκτον, Κρήτης és πίονα δῆμον, The vase is there described as being of a familiar ὁππότ᾽ ἄρ᾽ ὁπλότατον παίδων ἤμελλε τεκέσθαι, ‘*Tsland” form, from which 1 infer that Dr. Ζῆνα μέγαν: τὸν μὲν of ἐδέξατο Γαῖα πελώρη Tsountas also refers it to the earlier Aegean Κρήτῃ ἐν εὐρείῃ τραφέμεν ἀτιταλλέμεναί τε. period. Fig. 24 was executed, with the aid ot ἔνθα μιν ἵκτο φέρουσα θοὴν διὰ νύκτα μέλαιναν photography, by Mr. F. Anderson from the cast πρώτην és Δίκτην' κρύψεν δέ ἑ χερσὶ λαβοῦσα supplied me by Dr. Stais. ἄντρῳ ἐν ἠλιβάτῳ, ζαθέης ὑπὸ κεύθεσι γαίης 2 ‘Pictographs’ &c., p. 34 [J.H.S. xiv: Αἰγαίῳ ἐν ipet, πεπυκασμένῳ ὑλήεντι.

p. 303).

OF CRETAN AND AEGEAN SCRIPT. 351

Idaean Cave, Halbherr and Orsi describe the results of some excavations near the mouth of the Cave and also various relics discovered there by the peasants in the course of ‘ tumultuary’ diggings.! In 1894, during my travels in that part of Crete, although unable at that time to visit the spot, 1 procured from the peasants many additional objects in the shape of bronze arms, votive and otherwise, and small figures of men and animals. In the following year I was able to visit the Cave in company with Mr, J. L. Myres, and to secure further materials illustrating the character of its deposits. At the time of our visit in 1895, it being the holiday-time of the Greek Easter, a large part of the male inhabitants of the village were engaged in grubbing in the interstices of the boulders. The huge masses of fallen rock with which almost the whole of the vast entrance hall of the Cave is strewn, preclude anything like systematic excavation on a large scale within the Cave except at an enormous expense. Here and there, however, a few square metres of less encumbered soil enabled us, at least, to gauge the character of the deposits.

Among the excavators was a youth who, shortly before my return to the Cave in April, 1896, and in anticipation of it, dug down to its rock floor in a comparatively unencumbered part at the lowest level of the vast entrance hall. On my arrival he showed me several clay bulls and figures of the usual Mycenaean class obtained through his dig, together with several plain terra-cotta cups to be presently referred to, As a matter of comparatively minor importance, he informed me that he and a friend, who had helped him in the excavation, had found at the bottom of the hole a ‘broken stone, with writing. It may readily be imagined that I lost no time in securing the stone and also in ascertaining on the spot the exact circumstances of its position. The stone proved to be a dark steatite fragment of a low table exhibiting cup-shaped hollows with raised rims, similar to those of the stone libation tables of ancient Egypt. The form of the table had been oblong with four short legs and a central stem. It had originally possessed three cups, the central one somewhat larger than the other two, but the part of the stone containing the left-most of these was broken away. Its most remarkable feature, however, was part of an inscription clearly cut along the upper margin of the table in the prae-Phoenician script of Crete.

A view of the remaining portion, as well as of the whole table restored in outline, will be seen in Figs. 25a, ὃ and c.

I at once made arrangements to continue the excavation at the spot where the inscribed object was found, partly to ascertain if the remaining fragment of the stone table was discoverable, partly to gain an accurate idea of the deposit from beneath which the part now brought to light had been extracted.

The inscribed block lay at the point indicated in the sketch plan (Fig. 26) of the great entrance hall or ‘Atrium’ of the cave near its inner wall and on the rock floor, at this point about two metres below the existing surface of the ground. I dug out a space of about sixteen square metres all

1 Antichita dell antro di Zeus Ideo, p. 216 seqq. BB 2

(‘our]yno ur pasoyser 5908} todd Q) ΤΠ “HAVO NVHVLIHIG FHL WOUA TIAV], NOILVAT] GaaIMosN[—‘vez “1

<=

(‘aUT]}NO UT par04seyy) [Ὁ] “HAVO NvavixIq woud aTavy, NOILVGIT 10 MAGIA LNOU™I—"9GZ “org

ΡΥ.

va

Be A

Fic. 25c.—UNDER-SIDE OF LIBATION TABLE FROM DIKTAEAN CAVE.

(Restored in outline. )

FURTHER DISCOVERIES OF CRETAN AND AEGEAN SCRIPT. 355

round down to the rock which in most places lay somewhat over two metres below the surface. No trace of the remaining portion of the stone was to be found, but about 1} metres down, we found a continuous layer containing what appeared to be a sacrificial deposit of bones, horns and ceramic objects, imbedded in ashes and charcoal. The bones were of deer, oxen, and goats, the horn of an agrimi or wild goat found in this stratum being about a foot and a half in length. Amongst the ceramic relics was a small clay figure of an ox of a rude character, common among the late Mycenaean remains of Crete, and of which a small deposit had been found in the same ash-layer nearer the inner wal] of the cave. There was also the head of

ie + auntie Ἦ >. GY 5 w OZ. «ἀρ,

ΝῊ !

ye “yh ", “ aD Gif!

ss

HV R orl ας g ἤ SS Ae ea ie Ὕ , HL <= ren i NE aE ra Mi,

AN ha flay ba A "" Qi my 1 D WIDTH OF MOUTH OF CAVE ABOUT 23 METRES wil sy

SKETCH PLAN OF ATRIUM OF CAVE OF PSYCHRO [DIKTAAN CAvE]

Fic. 26.

another ox of superior fabric, and fragments of two rough clay pipkins with flat bottoms, and handles sticking out like projecting fingers of clay. But the most characteristic vessels were small cups of plain reddish clay of a type found in the votive deposit in the Idaean Cave and in Cretan beehive tombs of the Mycenaean period. Of these I extracted over a score and almost all without a break, some arranged in ‘ nests’ inside one another, a circumstance which sufficiently demonstrates that the stratum in which they lay had remained undisturbed since the time of their deposit. Two bronze oxen of rude fabric also occurred in the same layer.

The fact that the remaining part of the steatite libation-table was found

356 FURTHER DISCOVERIES

beneath this well-defined sacrificial stratum is of considerable chronological importance. This becomes the more evident when we come to survey the votive deposits of the Psychro Cave taken as a whole. Remains of the historic period are curiously rare. I was able to observe a plain proto- Corinthian aryballos and one or two fragments of glazed black Hellenic ware in a superficial layer, and, in 1895, was shown a terra-cotta griffin’s head apparently from a tripod bowl, and a small trunk-like block of white marble with a tail of a snake coiled round it, belonging to a later cult, whether of Asklepios or of some local hero it is difficult to say. I further obtained a very remarkable bronze openwork figure of a huntsman carrying a wild-goat, analogous, though in a superior style, to that referred to by Milchhoefer,’ and now in the Louvre, representing two huntsmen with a similar animal. A few specimens of Cretan geometrical ware contemporaneous with that of the ‘Dipylon’ period in Greece also occurred and a fibula? with coils in its bow and a small square catch-plate, showing similar affinities.

But the great bulk of the relics found in the Diktaean Cave go back to the prehistoric period—and a large proportion of these may be described as ‘Late Mycenaean.’ A characteristic sword handle of that period found here (also common to Southern Italy) greatly resembled types represented in the bronze hoards discovered in the later houses of Mycenae. Certain double- axes, knives, adzes, and dagger-blades from the Psychro deposit bear the same affinities, while the bronze knives with slightly curving blades? also occur in Mycenaean tholos tombs of the island and as imported objects in the later Italian Terremare. The coarser bronze figurines of men and animals approach those of the earliest deposits of Olympia, and of the Italo-Hallstatt Province, while other specimens obtained by me show a purer Mycenaean spirit. Amongst these may be enumerated flounced female figures and a small statuette of a man wearing the Mycenaean loin-cloth and showing a method of knotting the hair and two long locks behind very similar to that seen in the case of the men on the Vapheio gold cups. |

Nor were there wanting retics ofa still earlier period. Among these may be mentioned a fragment of a large dark brown vase with a goat rudely moulded in high relief, recalling some Cypriote ceramic products of the Copper and early Bronze Age. A small bronze dagger of quasi- triangular form, and short swords of very simple fabric seemed also to be distinctly prae-Mycenaean in character. A broken basin of dark steatite,* somewhat -heavily made, also probably belongs to the same early period. The fact that the fragment of the inscribed table lay below a well defined and apparently undisturbed sacrificial layer of Mycenaean date makes it possible, so far as the actual conditions of its discovery are concerned, that it too may belong to the earlier cultural stratum represented in the Psychro relics. The

1 Anfange der Kunst p. 169 Fig. 65. obtained by myself are in the Ashmolean

2 Halbherr and Orsi, Antro di Zeus, &c., 1 Museum. A similar knife was procured by me Τὰν. xiii. 6. from a tholos tomb at Kamares on the

8 Compare for example Halbherr and Orsi, southern slopes of Mt. Ida. op. cit. I Tay. xiii. 8. Other specimens 4 ‘Pictographs’ &c. p. 122, Fig. 121.

OF CRETAN: AND AEGEAN SCRIPT. 357

character of the libation-table itself is at least not inconsistent with such a possibility. To judge by the abundance of small terracotta cups, it looks as if the later votaries of the cave preferred to set their libations in these more portable receptacles.

The use, indeed, of steatite vessels, certainly survived in Crete into Mycenaean times, but, as I have elsewhere shown, the most flourishing period of the fabric of such objects in the island goes back to the time when ceramic arts had been less perfected. A series of examples demonstrates the fact that at a very early period Egyptian vases in hard stone were imitated by Cretan artificers in their native steatite, and the absolute correspondence with some Twelfth Dynasty models, including the imitation of the returning spiral orna- ment then rife,1 shows that many of these Cretan stone vessels go back to the first half of the Third Millennium before our era. A special centre of the discovery of these early steatite vases is Arvi, a very ancient sanctuary of Zeus on the southern coast, and here examples were found in a tomb con- taining a clay suspension vase of early Aegean bucchero belonging to the period of the cist-graves of Amorgos or even of the Second City of Troy. Curiously enough, my recent journey led to the acquisition on this site, together with other vases of the same material, of an example of a steatite libation-table affording a very close parallel to that of the Diktaean Cave. It was of greyish hue, somewhat smaller than the other, and with only a single cup-shaped hollow, as before, with a raised rim. It had four short feet, but no central stem. This relic, at the time of my visit, had recently come to light at a knoll called Tartari, a little below the monastery which still keeps up the sanctity of the spot.

The early associations of other stone vases from this site and its neighbourhood, in which the imitation of Egyptian old empire models is clearly discernible, make the general correspondence of the libation-tables from Arvi and Psychto with Egyptian prototypes of early date the more suggestive. The characteristic features of the whole, the small portable table with cup- shaped hollows having their rim raised above the flat surface of the table, are here faithfully reproduced. It further appears that stone libation-tables of this kind were specially in vogue during the Twelfth Dynasty, and it is to that period that their imitation in Crete must remount. Professor Sayce informs me that a Twelfth Dynasty libation-table, which in form is simply an enlarged repetition of that from the Diktaean Cave, was discovered last year at Lisht by Messrs. Gautier and Jéquier and is now in the Gizeh Museum.

In the case of the libation table from Arvi the small proove which uate the peper surface of the slab near the margin is sigs continued

type in Pare stone, and with a similar cover, now in the collection of Dr. Julius Naue at Munich. It is also found on Egyptian cylinders

1 The MAT. ite ornament in Twelfth Dynasty Egypt was not confined to scarabs. A dark bucchero vase found in Egypt, of a type

characteristic of that and the succeeding Thir- teenth Dynasty is surrounded by a decoration of this kind inlaid with white gypsum. The returning-spiral ornament recurs on a Cretan steatite vase, resembling a Twelfth Dynasty

and is imitated on primitive Hgean examples from Amorgos. The imita.ion of similar orna- ment on similar objects is a strong proof of the common origin of both.

358 FURTHER DISCOVERIES

round three sides. The fact that it is omitted on the fourth side seems to show that it was here set back against some other object. A groove in a similar position is observable on the remaining portion of the Diktaean Table ; and in the restoration indicated in Fig. 25a it, too, has, on the analogy of the Arvi example, been omitted on the back side. It is probable that the Diktaean Libation Table was also set back against a flat surface, perhaps in this case the wall of the cave itself, close to which it was found.

The threefold receptacle of the Diktaean Table suggests some interesting analogies with a ritual usage which goes back to the earliest religious stratum of Greece. In the case of such primitive worship as that of the Shades of the Departed, and again in that of the Nymphs, a triple libation was frequently offered. According to the old Arcadian rite (specially sig- nificant in a Cretan connexion,) recorded in the Odyssey, the offering to the Dead before the Falls of Styx was of this kind?:

Πρῶτα μελικρήτῳ, μετέπειτα δὲ ἡδέϊ οἴνῳ

Τὸ τρίτον αὖθ᾽ ὕδατι. The heroic and chthonic character of the primitive Zeus-worship of Crete makes it probable that a similar usage may here also have obtained, and in the very cave where according to the legend the infant Zeus has been fed by the Nymphs with ‘mingled milk and honey,’? the offering of the μελίκρητα would have been specially appropriate. We are, indeed expressly told that the ritual performed in honour of the Cretan Zeus set forth the miraculous preservation of the infant and his nourishment by Amalthea and Melissa.’

§ 9. The Inscription.

It is time however to turn to the inscription itself. If the position of the punctuations can be taken as a guide, the characters run from left to right. It is possible, however, that, as in the later Libyan alphabets,‘ these full-stop-like marks had themselves the value of letters.

The first character, though imperfectly preserved, is obviously the same as No, 3, and presents an elongated variety of what I have called the four- barred gate symbol. This occurs both on the linear and the pictographic ser‘es. As connected with a linear group it appears on an early white steatite seal-stone from Praesos, ‘ Pictographs, etc., Fig. 36d, p. 28 [296]. On the pictographic series it is twice coupled with the bent leg, and, as already noted above, it is linked on the linear group referred to with a 7-like sign which may well be the linearised equivalent of the leg symbol. The ‘door’ or ‘ gate’ symbol has already been compared with the Boeotian E with four parallel bars,° which points in turn to an older form of the Semitic Cheth with four bars instead of three. (See Table 1.)

1 Od. x. 519, 520. 5 *Pictographs’ &c. p. 92 [J.H.S. xiv. 2 Cf. Diod. v. 70. p. 261] and cf. Table III. No. 3 p, 95 [ 5. δ 3 Lactantius, De falsa Religione, 21, 22. xiv. p. 365].

+ See below p. 386.

OF CRETAN: AND AEGEAN SCRIPT. 359

The second sign is altogether new. The Ὑ contained in it reminds us of Pictograph No. 54 and allied linear forms, but its combination with the arch suggests a comparison with the Egyptian hieroglyph repre- senting a vault of a roof supported by a column,!—especially the first example given in my comparative Table (I.) in which the capital of the column has a threefold division.

Among the meanings given for the Egyptian sign are hall, (seh), assembly, or festival (kb); to meditate or consider (wa-wa), science, wisdom, and incense (sent).

{

teed

)

i

ΓΦ, Υ Lets

Fic. 27.—INSCRIPTION ΟΝ DIKTAEAN Ἐπ} TABLE.

The fourth and fifth signs are identical with that referred to above as the ‘polyp’ symbol, common among the Cretan pictographic signs and also apparently seen in a slightly variant form on the early linear seal, Fig. 22, above. (See Table I.) The reduplication of this character finds a parallel on a seal of the conventionalised pictographic class, Pictographs, etc, p. 30 (ὦ... xiv. p. 299), Fig. 39, where it appears at the end of one line and the beginning of another.

The sixth sign must certainly be completed as 2, This form occurs

in the conventionalised pictographic series (Pictographs, No. 69, and cf. 70,

78), being found once more on line 4 of the four-sided seal described

above (Fig. 6d). I have already compared it with the Egyptian hieroglyph

a coil of thread, signifying ‘to reel.’ In the Cypriote syllabary ? = pe. (See Table 1.)

The seventh character is too imperfect to admit of probable restoration.

The remaining portion of the eighth letter is curiously grouped with the last of the series. It looks like the upper part of the Semitic Resh. The ninth character, which lies apparently on its back above the preceding, has a greater affinity with a Beth.

In the present state of our enquiry it would not be safe to go beyond general comparisons. Nor shall I, at least, attempt an interpretation which

1 No. 1 is taken from Birch’s edition of No. 2 is from De Rougé Chrestomathie Egyp- Bunsen Egypt's Place in Universal History tienne PI. xiii. 36 p. 105. Vol. 1 p. 541 No. 578 and cf. p. 542 No. 582.

360 FURTHER DISCOVERIES

could hardly fail to be premature. The great value of the present group of characters consists in the fact that it is impossible to doubt that we have

SIGNS FROM DICTAAN

1 AND3.2. 4 ANDS AND 6 OF LIBATION TABLE CAVE WITH PARALLEL FORMS.

LIBATION TABLE

NEAR ON SEAL

etre

CRETAN PICTOGRAPAS

TOGRAPHS ON SEALS

EGYPTIAN CRETAN PIC- |] CRETAN LI-

HIEROGLYPHS FESTIVAL ETCy

SIGNS FOR HALL

LIBYAN

SYLLABARY | LETTERS

x wn

CYPRIOTE ]

SEMITIC LETTERS

Fu ae yz ω ΕΞ 22 #2

ω Oe 5 a fe Kx we Ww Pee a :

TABLE I,

here to do with a formal inscription. For.the first time we have to deal with a series of signs of alphabetical form,—though probably in part at least of

OF CRETAN AND AEGEAN SCRIPT. 361

syllabic force,—not merely of a personal nature like those engraved on seals or scratched on vases, but, as far as can be gathered from their association, in the strictest sense of the word monumental. These letters, clearly cut and accompanied even by what has the appearance of a regular punc- tuation, on a stone Table of Offerings, brought to light in the earliest stratum of a sanctuary of remote antiquity, must in all ‘probability be regarded as part of a formal dedication.

The correspondence observable with known symbols of the early Cretan script, both linear and pictographic, shows that the present inscription belongs to the same series as those of the seals and vases. But the inserip- tion itself must on the whole be classed with the more linear group, and the balance of evidence shows that the linear type of writing in Crete belongs in the main to what may be called the ‘Early Soft-Stone Period’ of seal- engraving, which preceded the Mycenaean Age when harder materials like cornelian and chalcedony were successfully attacked.

In the three instances that I was able to cite in my former work in which characters of this linear class appear on three- or four-sided bead seals, they are all of the ‘soft-stone’ class, and all display ornaments or figures similar to those of the most archaic type of pictographic seals, in the decorative designs of which Twelfth Dynasty models are clearly traceable. The three-sided stone given there in Fig. 36 with rude linear signs and figures, certainly stands near the beginning of its class, and the inscribed whorl found in the Hagios Onuphrios deposit, engraved in the same primitive manner, also belongs, as the associated relics show, to the same early period. Of the examples cited in the course of the present paper, the rude steatite seal from Kalamafka has every appearance of primitive workmanship, and the inscribed vase from Cerigo has already been referred to the prae-Mycenaean period of Aegean culture.

On the other hand, the comparisons instituted between certain characters on the libation table, with some of those of the conventionalised pictographic class, such as the ‘polyp’ sign, the four-barred gate, and the @ seem to bring down its date to a period approaching that of this later class which has proved Mycenaean affinities. This might take us to the beginning of the Second Millennium B.c., a date which would be still reconcileable with the fact that the Table itself is apparently based on a somewhat earlier Egyptian model. The converging lines of chronological induction at our disposal make it on the whole unsafe to attribute this monument to a later time.

If, as would thus appear probable, this monument goes back to about 2000 B.c. the antiquity of the pra-Phcenician system of writing in Crete receives a new and remarkable illustration. Brief and incomplete as it is, the Psychro inscription stands alone among the written records of our Continent, It is not only separated longo intervallo from the most ancient examples of Greek writing, but it distances by at least a thousand years the earliest specimens of the Semitic alphabet as seen on the Baal Lebanon bowls and the Moabite stone.

362 FURTHER DISCOVERIES

ParT I].—Proro-Ea@yptiaAN AND LIBYAN COMPARISONS.

δ΄. Early Prism-seal of Steatite Jrom Karnak.

In connexion with the Early Cretan remains described in the pre- ceding sections I am able to cite a remarkable piece of evidence pointing to the existence in the Nile Valley or its borderlands at a very early Pharaonic date, if not of actual settlers from Crete, at any rate of a ‘ Libyan’ population closely allied to the primitive Cretans in the most distinctive property of early culture. This is a triangular bead-seal of black steatite (Fig. 28) obtained some years since, with other small relics from Karnak, by the late Mr. Greville Chester, and presented by him to the Ashmolean Museum. Though somewhat larger in size, it reproduces the characteristic form of the triangular Cretan bead-seals as well as the material of their earliest class.

Fie, 28.—PRISM-SEAL OF Buack STEATITE FROM KARNAK. [8]. a, section: b, ς, d sides.

Among the signs which appear on the three sides, the bee or hornet, the scorpion with upturned tail (twice repeated), and the beetle, are common to the Egyptian hieroglyphic series, the others are variant or foreign forms.

As so often on the early Cretan seals we have at the head of what may

OF CRETAN AND AEGEAN SCRIPY. 363

be taken to be the initial column, a rude human tigure—in this case standing and apparently holding up a crocodile by its tail. In the next column is a seated animal with uplifted paw. Beneath this is a figure in a violent attitude, as if in the act of running. ‘The body and limbs are those of ἃ man, but the head, which is turned back, is that of an animal. It has

two long slightly curving horns as of some kind of goat ov ox. On the third face of the seal appears another monstrous form—the linked forequarters of a pair of ibexes.

The form of the Cretan prism-seals had suggested to me from the

first a certain parallelism with the oriental cylinders. I went, indeed, so far as to observe that they might in some sort be described as ‘ threc-sided eylinders. That these trilateral seals are in Crete itself the direct descendants of the rude perforated splinters of steatite which characterise the most primitive stage of the glyptic art in the island is a natural supposition.1 But the elongated type with large central perforation? shows such an approximation to the cylinder that some influence from that type of signet might reasonably be suspected. The rude irregular form of the original’ bored splinter has been as it were crystallised into a geometrical shape in conformity with the early Egyptian and Oriental cylinder-seal. In Crete itself, however, there seems to have been no sufficient opportunity for such influence. Rude and distant imitations of the early cylinder type have indeed been found at Hissarlik and in Amorgos, but not a single specimen of the primitive cylinder has as yet been discovered in Crete.

The occurrence, however, of the prism-type of bead-seal in Egypt suggests that the more primitive ‘ wedge-seal ’ may have been modified by the cylinder type on Egyptian soil itself by a population having both an Agean and a Nilotic range. For we now know that the earliest form of signet among the dynastic Egyptians themselves was not the scarab but the cylinder. The re- markable royal tombs, explored by M. Amélineau at Abydos,and by M.de Morgan at Naqada, though they contained not asingle scarab, produced a series of clay cones used as stoppers of vases, exhibiting impressions from cylinders. The crowning discovery of Dr. Borchardt, who has identified the royal tomb excavated by M. de Morgan at Naqada with that of the first Egyptian monarch, shows that the signet of Menes himself was a cylinder.

Some of the cylinders of this earliest dynastic period have been actually preserved to us. On one of white stone in the Ashmolean Museum Professor Sayce has recently deciphered the name of Atota, a grandson of Menes, while on another of green steatite, found in an early tomb excavated by Mr. Quibell at El Kab, he has recognised the name of King Khaires of the Second Dynasty. Some of these early cylinders are of copper,’ and it is perhaps owing to the influence of this type that from about the Fourth to the Sixth

1 See above, p. 330. class,

2 A good example of this Cretan type with 3 A specimen of this class, also from a tomb, abnormally large perforation is seen in Picto- excavated by Mr. Quibell at El Kab, and now graphs, &c., Fig. 36, p. 28 [297]. This seal; in the Ashmolean Museum, apparently bears the with linear characters, belongs to a very carly name of Men-Kau-ra of the Fourth Dynasty.

364 FURTHER DISCOVERIES

Dynasty the stone cylinders show an abnormally large perforation. In the time of the Twelfth Dynasty an ivory tube was inserted into this perforation, and shortly after that period the cylinder type of signet in Egypt was finally

Fic. 31.—Ciay CYLINDER IN ΟἼΖΕΗ MusEuM. AFTER DE Morcan.

superseded by the scarab, a form apparently unknown under the earliest

dynasties. The Egyptian cylinders above referred to are the true fabric of the

OF CRETAN AND AEGEAN SCRIPT. 365

Pharaonic race and bear regular hieroglyphic inscriptions. But, side by side with these, from the earliest dynastic period, there existed another class of cylinders exhibiting signs of a more pictorial character, which, though contain- ing Egyptian elements, are by no means of the orthodox Egyptian class. These cylinders seem for the most part to be formed of black steatite similar to that of the Karnak prism, and the figures with which they are engraved show most striking correspondences with those on this Cretan type of seal.

The closeness of this affinity will be sufficiently illustrated by the examples of this class of cylinder given in Figs. 29,1 30,2 31.3 The standing figure apparently holding up a crocodile by the tail on the first face of the prism receives a double illustration in Fig. 31. The running figure with a man’s body and the head of a horned animal, Fig. 280, repeats the movement of the running human figure on Fig. 29. If in the former case we have a kind of Minotaur, we find here the figure of a man with a hare’s head. The scorpion with upturned tail reappears in Fig. 31, the beetle on Fig. 29, the bee on all three cylinders. The two-headed goat finds a close analogy in the linked forequarters of the oxen on Fig. 30, and further parallels in the double lion and lion-ibex of a cylinder impression from the tomb of Menes to be referred to below.

It is obvious that both prism and cylinder represent the same traditions and are of the same contemporary fabric. A closely-allied work is also to be seen in the impression of a cylinder on a clay cone found by M. de Morgan in the tomb of Menes at Naqada.t The special group with which we are con- cerned seems, however, to belong to a somewhat later date, though containing very early traditions. From its exceptionally large perforation and certain peculiarities of technique, Professor Petrie inclines to refer the cylinder in his possession (Fig. 29), to the time of Pepi 11., of the Sixth Dynasty, for whose reign he assigns the approximate dates, 3443—3348 B.c. The close resemblance observable between this and the Karnak seal both in the material, which is the same black steatite, the size of the perforation, and the character and style of the figures, shows that it must, approximately at least, be regarded as contemporary with the Petrie cylinder. It thus appears that—if we accept the chronology of Professor Petrie—the Karnak prism-seal was executed about 3400 B.c. and that the prototypes of the primitive Cretan class must go back to that remote epoch.

We can have no hesitation in dealing with the above cylinders and the Karnak prism as a homogeneous group, and the interesting question arises,—T'o what racial element does it belong? It represents, as we have seen, together with certain types common to the historic Egyptians, other forms of extraneous origin.

1 This cylinder is in Professor Petrie’s collec- 8 Clay cylinder in the Gizeh Museum, from tion, to whose kind permission the present re- De Morgan’s Recherches swr les Origines de production is due. It is of black steatite, with U#gypte (ii.) Fig. 857, p. 257. an exceptionally large perforation. 4 Origines de V Egypte (ii.) Fig. 560, p. 169.

2 From Lajard, Culte de Mithra, Pl. xiii. 8,

H.S.—VOL XVI. Gc

366 FURTHER DISCOVERIES

Among these is a curious bow-legged figure which occurs on all the examples of cylinders given above. On Fig. 29 we see it with both hands raised, on Fig. 30 leading a cynocephalus, on Fig. 31 it is four times repeated.

This figure is of great interest. The characteristic form of the lower limbs shows that we have in fact to deal with the same grotesque personage who so often makes his appearance in a secondary position on Babylonian cylinders.! Allied, and perhaps derivative, figures may be seen in the pygmy or ‘embryonic’ form of Ptah-Socharis-Osiris and its offshoot the Phoenician Pataecus,? but there can be no question that the type scen on these early cylinders is the direct reflection of that which appears at a very early date upon those of Chaldaea. The horned man of the Karnak prism may itself be due to a composite and distant reminiscence of Gilgames and Eabani.

The true cylinder types of dynastic Egypt, as we now know them from the First Dynasty onwards, show, from the first, purely normal representations of contemporary hieroglyphic forms. They reproduce a system of signs already fully evolved by at least as early a date as the time of Menes. Though the cylinder form itself is oriental, and though some few hieroglyphs may go back to the same common origin as the Chaldaean, there is no sign of direct borrowing of Asiatic types. On the group, however, with which we are immediately concerned, we are here confronted with a figure taken direct from the cylinders of Babylonia. In the naked male figure, indeed, between two crocodiles on Fig. 31, we find the actual adaptation of a familiar Baby- lonian scheme—the hero between two bulls or lions, sometimes held in a reverse position.

We see thus upon the present series evidence of borrowing both from Asiatic and dynastic Egyptian sources—the latter naturally preponderating, — while at the same time both classes of borrowed elements are reproduced with a certain barbaric fantasy, and combined with other features which are neither Pharaonic nor Chaldaean. .

To what Nilotic population, then, are these hybrid works to be ascribed ? The answer to this inquiry will probably be found in the evidence supplied from other quarters of the partial survival in the Nile Valley of the earlier

* It appears in diminutive dimensions in the we are dealing, but the legs in this case are not

inter-spaces between the principal figures on a series of Babylonian cylinders. At times it is associated with the small image of a nude female divinity, apparently Sala, a form of Istar. (Menant, Collection De Clereg, Pl. xxiii. 231, Pl. xxvii. 277; Lajard, Culte de Mithra, PI. xxxix. 5, Pl. xl. 9). For Sala, see Nikolsky, Rev. Arch. 1891, ii. p. 41, who cites a cylinder on which this name accompanies the nude female type. In this case Sala-Istar is coupled with a nude male divinity, also of diminutive size, and identified by the inscription with Ram4nu, the Syrian Rimmon. The arms of this male figure, crossed on the breast, resemble those of the pre- valent Chaldean version of the type with which

bow-legged. The fact however that the bow- legged type is repeatedly associated with the nude Goddess, and like it occasionally appears on a kind of base of the same form, makes it probable that the diminutive and grotesque male figure was regarded as a satellite of the small female figure. This male figure occurs on cylinders of extremely archaic type (cf. es- pecially Lajard, Culte de Mithra, Pl. xl. 9). Above it is not unfrequently seen the combined symbol of Sin and Samas, and sometimes a crescent or a star. (Menant, Coll. De Clercq, Pl. xiv. 123). ? A parallel but variant type is seen in Bes,

OF CRETAN AND AEGEAN SCRIPT. 367

indigenous stock which was in possession of the land at the time of the first coming of the Pharaonic conquerors from the South-East. This earlier population, of which something more will be said in the succeeding pages, has been with great probability identified by Professor Wiedemann and others with the ancient Libyan race. In a large number of the tombs excavated at Naqada and elsewhere, we must, in view of the most recent discoveries, recognise the remains of this pre-historic race in Egypt.

To a certain extent the whole later civilisation of historic Egypt was influenced by this pre-existing indigenous element, the assimilation of which was only the work of centuries. In the case of the present group of cylinders, the traces of the traditional ‘Libyan’ art seem to be especially strong. The form of the human figures may be compared with the rude sketches on some of the prehistoric pots of Naqada.!' The scorpions with upturned tails, the crocodiles, the ibexes, the long-legged birds, probably ostriches,? are all plentifully represented. The semi-processional arrange- ment of the animals on the cylinder impression from the tomb of Menes recalls the zones of animals on the prehistoric ivory handle from Sohaj,3 and the gold-plated knife in the Gizeh Museum.* A further parallel is afforded by the two-headed animals of the cylinders and prism-seal. These combinations recall one of the special features of this proto-Egyptian art, as illustrated by the slate palettes and ivory combs, which repeatedly take the shape of double birds ® and, at times, of ibexes.? On a proto-Egyptian slate tablet? a double bull is seen in association with bow-men whose hair, divided into separate plaits and adorned with two plumes, recalls later, Libyan, fashions.

These comparisons seem to show that the class of cylinders with which we are at present concerned, and with them the Karnak cylinder, were the work of some more or less independent Libyan elements which still survived in the Nile Valley, or had perhaps partially re-intruded themselves there, as late as the Sixth Dynasty. The traces of Asiatic influence, such as the Pataecus-like figure on these signets, point to a race who had intimate relations with the traditional enemies of the Pharaohs on the Syrian side.

If this conclusion is correct, and we have here the handiwork of a Libyan population, the prism seal of Karnak, though as yet an isolated phenomenon, may prove to be of capital importance in its relation to the early bead-seals of Crete, and in a more general sense to the origin of the primitive Cretan and Aegean culture. For, as will be seen from the com- parisons given below, the points of resemblance are by no means confined to the three-sided form of the seal itself, Taking this and the allied

1 Compare the vases, Petrie, Nagada Pl. % Petrie, Nagada, Pl. Ixxviii. In the Pitt xxxv. 7 and De Morgan, Origines de’ Egypte i. _ Rivers collection. Pl. iii. 48. 4 De Morgan, Origines, &c., i. p. 115, Fig.

2 Cf. the cylinder impression from the tomb 136, and ii. Pl. v. of Menes, De Morgan, Origines de I’ Egypte, ii. 5 Cf. Petrie, op. cit. Pl. xlix. Fig. 62 seqq. Fig. 560, p. 169, where a similar lonyg-legged 6 Jb. Pl. xlvii. Fig. 11. bird occurs, 7 In the British Museum No. 27090,

coazZ

368 FURTHER DISCOVERIES

cylinders as a homogeneous group, a whole series of interesting and con- clusive parallels can be established between the figures that they contain and certain distinctive types of the primitive Cretan cycle. This correspond- ence, moreover, extends to some of the characteristic designs on gems of the Mycenaean period.

The bird, probably an ostrich, that occurs among the animals on the impression of a cylinder of the indigenous class found in the tomb of Menes seems to be the forerunner of the long-legged race seen on the Cretan bead- seals. In the one series we find the ibex, in the other the wild goat. The scorpion with upturned tail, which is seen already on the prehistoric pottery of Egypt, and is so characteristic a feature on the Karnak prism and the allied group of cylinders, is also one of the most frequently represented objects on the early three-sided steatite seals from Crete. The spider on the Petrie cylinder is of special interest in view of the fact, that while on the one hand it is unknown amongst Egyptian and Chaldean representations, it is now seen to take a prominent place among the Cretan pictographs.

Fic. 32.—ANoIENT LipyAN BEAD-SEAL OF STEATITE FROM NEAR CONSTANTINE (CIRTA).

But the spider itself as a symbol on ancient signets also appears in an undoubtedly Libyan connexion. In the collection of antiquities formed by Captain Farge, Director of the Bureau Arabe at Constantine, the ancient Cirta, once the capital of the Numidian kings, I observed a bead-seal of brown steatite found near that place, a sketch of which, made with the owner’s kind permission, is reproduced in Fig. 32. It represents a spider of the Tarantula class, not unlike that on the early cylinder, and the bead-seal on which this object appears seems, both from its form and material, to belong to a relatively early date. This hemispherical type of bead-seal belongs in Greece to the Geometrical Period and to the ninth and tenth centuries B.c.1

On Table II. will be seen a further series of comparisons between certain characteristic schemes and subjects of the ‘Egypto-Libyan’ group—if such

1 Examples of this form have been found at quariwm (Berlin), No. 70. Similar types of Olympia. Cf. Furtwingler, Olympia, p..188, bead-seal have been found on the site of the and Beschreibung der geschnittenen Steineim Anti- Heraion at Argos and in Anatolia,

369

OF CRETAN. AND AEGEAN SCRIPT.

—_—_—— es ES LL -

fw @) 3.139} WOUS W39 ΟΙΟΙΝ3Ὶ NO ΘΝ ΒΟ3 NV3VN "το: ἼΑΝΝ ONY ΝΥ 138 Alyy]

HLIM G3¥¥dWOD S¥3QNITAD Se tases “OaI TY ONG, werwd WWNYYH NO

>@08>'> S3IdALIVWINY ONY NYWNH wvisgnin TD 31e8vi

SoSQON™M 8 avaag ge = a σεν GlOLNaT NVoVNaDAWe 118 WLiNBAS σν3θ ΟΙοιν31 ΞΝνϑη ιν SModouny

8 GVa8 CIOLN3] NW3WN3DAWANO ΟΥ38Β S101N34 NY YNID AW ΝΟ ΝΥ ΣΥΝ ΣΘΑΙ. a Sivas wsidd ΝΥ1399 ΑἸῊν3 34λ. νν.3Ἐ 420 ) | ) Ὕν39 WSlud NO 2YOHM φολφϑνη

᾿Αιἰϑυνλα,οῖ vovowN % NOISS 3ugw υ JONINAD

WIGNITAD WWiagan3_

naz/2}w3am Ἕ WSiud FA

a ἔς

Views Ne 984 AYN Y YO BIONNAD, NYILdADZ, . AD, NVILdAD3, WSI8d WYNUYy : ; ‘ ,

370 FURTHER DISCOVERIES

a term be allowable—and others on the primitive seals of Crete and on some Mycenaean gems. It will be noticed that the rude square-shouldered human figure that occupies the first column of the Karnak prism-seal shows a great family likeness to the primitive figures which stand in the same position on the analogous class of Cretan signets. The figures with bowed arms, seen on the cylinder, Fig. 31, correspond with another Cretan type. The running figure on the Petrie cylinder, the scheme of which is repeated by the horned man on the Karnak prism, bears a striking resemblance to that on the steatite bead-seal of Cretan type published in my former work on the Pictographs.!

The contorted schemes illustrated by the hare-headed human figure on Fig 29, and still more by the tumbler on Fig. 30, also find analogies among the Cretan designs. A certain common element may be detected between the first-mentioned and the attitude of the rude horned man on the Phaestos whorl. The tumbler—which except that on the cylinder he is seen naked, corresponds with an Egyptian ideograph ?—presents a design admirably fitted for the circular field of the lentoid class of gems. If we may be allowed to assume—in the absence of direct evidence—that the tradition of this type was perpetuated through the intervening period by ‘ Egypto-Libyan ’ art, we should be able to trace to this source a scheme applied by the Mycenaean engravers of Crete to representations of the Minotaur. The close conformity of arrangement will be seen by a glance at the annexed diagram.

But the most interesting of all the parallels supplied by the Karnak prism is the first appearance of the ‘Minotaur’ itself. The type of the horned human figure, though in a different pose, is seen on one of the earliest examples of Cretan engraving, the whorl, namely, from the prehistoric deposit of Hagios Onuphrios, near the site of Phaestos. In Mycenaean Crete the type is frequent, but in this case it is of different composition. We no longer see a human body and bull’s head, but the whole forequarters from the waist up are here bovine. It is in fact one of a parallel series of Cretan representations of this period, in which the lower part of a man is coupled with the upper part of a wild goat, or a lion, or a flounced female figure terminates above in a flying eagle* In Greek art, as is well known, the monster reappears with human arms and body and only the head of a bull.

So far as the horned man is concerned it looks as if through all these corporeal variations we had still to do with essentially the same fabulous form, and the later Cretan version as seen on the coins of Knésos may thus

1 Fig. 53.

2 xtb=to tumble.

8 On the Cretan lentoid gem in the British Museum (Cat. No. 76, Pl. A.) published by Milchhofer (Anfange der Kunst, p. 78, Fig. 50) the foreparts of a goat and bull are both at- tached to the lower part of a man. Separate figures of a goat-man and a lion-man are in my own collection. Milchhofer himself (loc. cit.) expresses his opinion that the Minotaur type

did not originate from a mere arbitrary con- junction of this kind, due to the caprice of an engraver, or, as in the case of the Chimaera, to a misunderstanding of gem perspective by later copyists. The antiquity of the horned man type in Crete seems also to show that the Mycenaean engravers in this case simply gave a new expression to an already existing idea. 4. On some unpublished gems from Crete.

OF CRETAN AND AEGEAN SCRIPT. 371 be regarded as a reversion to the type which precedes the Mycenaean. Whether the whole myth is of iconographic origin or with what oriental elements it possibly connects itself, need not be discussed here. But if the connexion between the primitive type of the Karnak cylinder and the insular examples holds good, we may here have caught a glimpse of the Minotaur on his way to Crete as early as the fourth Millennium before our era.

Another interesting point of agreement between the early prism and cylinders from Egypt and the Mycenaean cycle is supplied by the two-headed animals, The conformity with the two-headed terminations of the prehistoric slate palettes of Naqada and the double bull on the proto-Egyptian tablet has already been noted. A certain analogy to these monstrous forms on the early cylinders of the non-Pharaonic inhabitants of the Nile Valley, is also supplied by the composite animal forms of the still earlier Chaldaean cylinders, due to the coalescing of two crossed animals. As a rule, however, these Chaldaean forms differ from the ‘ Egypto-Libyan’ class, since in their case the upper part of a single body is attached to two hind-quarters.!

The examples before us, on the contrary, show two fore-quarters united— in one case of two lions, in another of a lion and ibex, in a third of two goats or ibexes, while in the lower part of the more complicated design on Fig. 30 the forequarters of two bulls are seen to coalesce.

On the more primitive class of Cretan seals we find certain designs, such as that seen on Pl. X. No. 18, with two foreparts of animals in reversed position. A still closer analogy is shown by some Mycenaean types, of which two examples are given in Table 11. The first is a double goat on a lentoid bead of dark steatite found on the Akropolis at Athens,? the other is a double ox on a similar bead of green serpentine from Sybrita in Crete® curiously recalling that on the slate tablet referred to above. The existence of such animal ligatures‘ may have assisted the evolution of the Chimaera from the misinterpreted perspective of the Mycenaean gem type, representing a goat behind the back of a lion.® It is at any rate a remark- able coincidence that the coupling of the lion and the ibex should already occur on a primitive cylinder of the age of Menes. :

The reappearance—per saltwmn—upon Mycenaean intaglios of the figures of conjoined animals that characterise these early cylinders and prisms, taken in connexion with the other correspondences already indicated, is a fact of great suggestiveness, We may well suspect that the missing links in the

1 Compare the cylinder figured in Menant, Glyptique Orientale, vol. i. pp. 60, 61, Figs. 26, 27, 28. In Fig. 29 however the upper part of two coalesced goats terminates in two heads and forequarters. This type might form the starting- point for the ‘ Egypto-Libyan’ form.

2 Now in the Central Museum at Athens.

3 In my collection.

4 Double-headed animals as pendants or

ornaments are also frequent in the Late Bronze and Early Iron Age, perhaps spread through Mycenaean influences. (See especially 8. Reinach, Sculpture en Europe, p. 118 δεᾳᾳ.).

6 See Milchhofer, Anfdnge der Kunst, pp. 82, 88 and Fig. 53. The type is placed for com- parison on Table II. Another lentoid: bead of steatite with the same type was observed by me at Xero in Eastern Crete.

372 FURTHER DISCOVERIES

pedigree will eventually reveal themselves whenever the early archaeological strata of the Libyan borderlands of Egypt come to be investigated. The lentoid type of bead itself comes from the Egyptian side. It was fashionable in the days of the Twelfth Dynasty, and from the occurrence of this form among the relics from the royal tombs of Abydos it would appear to have been already in use in the days of the first Pharaohs.

§ 2. Crete the Meeting-point of Thraco-Phrygian and Libyan Elements.

It will be seen that the new evidence supplied by the Karnak prism and the inscribed Libation Table from the Diktaean Cave leads us in the same direction. In both cases we find the clearest indication of a very early connexion between Crete and the Nile Valley. In my account of the Hagios Onuphrios deposit, evidence has already been given that a form of Cretan stone vase shows a close correspondence with an Egyptian type belonging to the Fourth Dynasty. The Karnak prism indicates that as early as the Sixth Dynasty, and at a date which cannot be brought down much later than 3,500 B.c., the typical Cretan form of the seal was in use by a probably Libyan population in the Nile Valley. The Libation Table, on the other hand, brings home to us for the first time the fact that by the time of the Twelfth Dynasty the Cretans were so far affected by Egyptian in- fluence as not only to have received—as we know from other evidence—a series of decorative motives from that source, but to have adopted, apparently for similar use, an article of Egyptian cult.

The imitation of the characteristic scarab decoration of that period on the Cretan seal-stones of the prae-Mycenaean class has, I venture to think, been proved to demonstration! Such imitation, moreover, proves even more than the sporadic discovery of the Twelfth Dynasty scarabs themselves in the island, for primitive peoples are not antiquarian revivalists, and content themselves with copying the contemporary fashions of their more civilised neighbours.

But the occurrence of the inscribed Libation Table of Twelfth Dynasty type in the prehistoric stratum of a Cretan votive cave must be considered to indicate something more than the borrowing of external forms. The adoption, in this case, for indigenous purposes of cult, of the early Egyptian form of libation table shows truly an intimacy of religious contact for which the other evidences of Egyptian influence, striking as they are, could hardly prepare us. The phenomenon opens up whole vistas of new possibilities as to the primitive relations of Crete with the Nile Valley, and the con- formity here brought to light is of such a kind as almost to necessitate the invocation of Libyan intermediaries. Such a degree of influence, not on the externals only of articles of use or ornament, but on a fundamental object of primitive cult, can hardly be due to mere mercantile relations. It points surely to the presence among the inhabitants of Crete of an element which

1 To the comparative examples in Pictographs, &c., p. 58 [827], Fig. 49, I can now add others equally cogent.

OF CRETAN AND AEGEAN SCRIPT. 373

had experienced a prolonged land contact with Egypt—to an element astride the Libyan sea, with one foot on the Aigean island and the other on the African shore.

That the Libyans had largely imbibed the religious teachings of Egypt appears from a variety of indications. This is even shown by their personal names, in the composition of which are found the Egyptian appellations of the Sun- and Moon-Gods—Ra and Ah—and possibly of Isis and Hathor.! The opposite process is seen in the Egyptian adoption of the Libyan divinities Neit and Set.

The strong Egypto-Libyan ingredients in the primitive Cretan culture— pointing,.as they do, to the possibility of early settlement from that side— may some day indeed supply the clue to more than one characteristic feature in the insular religion and mythology. In my earlier communication stress has already been laid on the impossibility of explaining the deep-seated com- munity between some of these and Semitic types by the comparatively late Phoenician contact. Such parallel appearances, for instance, as Minos and Moses—both divine legislators, receiving the law ‘mouth to mouth’ in repeated visits to the God of the Mountain,—point to very early derivation from a common source.

But the Libyan element was itself well qualified to supply certain links of connexion with the Semitic world as well as the Egyptian. Evidences of a religious contribution from this side, quite apart from that derived from the Phoenician settlements on the North African Coast, are indeed supplied by some of the indigenous Libyan inscriptions, which contain the names of three Nabataean divinities.2 The early ‘Egypto-Libyan’ cylinders referred to in the preceding section show not only the influence of the oriental form of signet, but are accompanied by the Pataecus-like figure which on the Babylonian cylinders appears as the satellite of the nude figure of Sala- Istar, and apparently as a representative of the typically Syrian God, Ramanu or Rimmon. The direct relation in which these, ex hypothesi, Libyan signets stand to the Cretan prisms has been sufficiently illustrated; and we may therefore trace in them the further links of a chain of primitive intercourse with the Semitic world.

But over and above these archaeological evidences, it must be borne in mind that the Libyan dialects themselves, as illustrated by their surviving members in Algeria, Morocco, and Sahara, stand in a close relation to the Semitic family.2 In their grammar, and, to a certain extent, in their vocabulary,—and notably in such rudimentary elements as the numerals and

1 Halévy, Etudes berbéres, p. 122.

2 Halévy, loc. cit.

* For the modern Libyan languages see es- pecially Venture de Paradis, Dictionnaire ber- bére and the Dictionnaire frangais-berbére (gener- ally known as Brosselard from the name of one of its chief collaborateurs) Paris, 1844. A. Hanoteau, Essai de Grammaire Kabyle (Paris, 1858) and Essai de Grammaire Tamachek (Paris,

1860). ᾿ Barth, Travels in North and Central Africa. v. 565 segg. H. Stanhope Freeman, Grammatical Sketch of the Temahug or Towarek Language. (Ghat Dialect), (London, 1862), Prof. F. W. Newman, Notes on the Libyan Lan- guages, (R. Asiat. Soc., July, 1880) Libyan Vocabulary, (London, 1882) and Kabail Vocab- ulary (London, 1887).

374 FURTHER DISCOVERIES

personal pronouns,—they show a decided affinity not only with Coptic and Amharic, but with Hebrew, Aramaic, and Assyrian. They are in fact sub- Semitic.?

It must yet be borne in mind that from another point of view the early archaeological remains of Crete bring it like the other Mgean islands into close relation with the western coastlands of Asia Minor, the mainland of Greece, and even the Danubian basin. The marble ‘idols’ of the Cretan deposits, like that of Phaestos, belong to precisely the same class as that of the other igean islands and of the first and second cities of Troy. The primi- tive clay hanging vessels and the most rudimentary incised figures on Cretan whorls and seal-stones also find their nearest analogy in the earliest strata of Hissarlik. So, too—to turn to the remains of Mycenaean date—the megaron of the Cretan Goulas preserves with only slight modifications the ground- plan of the far earlier prototypes that occur in the second city of Ilios.? All this is quite in keeping with the well-marked group of early traditions and pre-Hellenic place-names implying the existence of a strong Phrygian element in the primitive population. Sufficient evidence of this connexion is supplied by names like Pergamon and Ida, and by many characteristic features of the Cretan religion in which the Mother Goddess Rhea, the Idaean Daktyls, the Kuretes and Korybantes with their orgiastic dances, all reappear.

While, therefore, we must admit the great infusion of Egypto-Libyan elements in primitive Cretan culture, we must at the same time never lose sight of that other side of its early traditions and remains which implies the presence here of members of the great Thraco-Phrygian race. If, as seems to be a natural conclusion from the intensive character of the Old Empire influence in the island, there were Libyan settlements here at a remote date, these may have either been gradually merged in an earlier population of European stock, or may have continued to coexist with it, just as at a later date Eteokretes, Dorians, Achaians, and Pelasgians lived side by side.

The evidence which makes Crete the meeting-place of Thraco-Phrygian and Egypto-Libyan elements fits in with a parallel series of indications supplied by ancient tradition and corroborated by Egyptian records. In these we see the ‘ Trojans’ and their neighbours engaged at a very early date on the African side. The ease with which the Libyan princes during the Nineteenth Dynasty allied themselves with the maritime races of the fEgean shores reveals a very intimate connexion between the two, and the abiding tradition of the Maxyes,—perhaps the most civilised of the Libyan tribes,—that they were of Trojan origin, may point to some still recognised blood-relationship. So too the Trojan band of Antenor appear among the

1 The expression is Prof. F. W. Newman’s. Others, like Renan, have preferred to apply the term ‘ Hamitic’—a distinction, perhaps, with-

. out a difference.

2 See my article, Goulds, the City of Zeus

(Annual of the British School at Athens, 1895-

1896), p. 188. 3 Cf. Hoeck, Kreta, i. 109 segg., 148, 208 seqg., &c., Milchhoefer, Anfinge der Kunst 129. 4 Herodotus, iv. 191: φασὶ δὲ οὗτοι εἶναι τῶν ἐκ Τροίης ἀνδρῶν. é

OF CRETAN: AND AEGEAN SCRIPT. 375

pre-Hellenic settlers in Cyrene, reminding us of legends which made Teucer land in Crete and Aineas found the Cretan Pergamon.? Elymos, the son of Anchises, the Eponymos of the Elymian inhabitants of Western Sicily,’ bears a name almost identical with that of a Libyan prince;* his people themselves had touched on the Libyan coast δ before continuing their course to Lilybaeon® and Eryx. Aineas, on the same Sicilian journey, tarries at Carthage, and his intercourse with Dido is the more important that we have here the female form of the most characteristic and at the same time the most ancient of Libyan personal names.’

Closely connected with these phenomena are some striking correspond- ences between the tribal and geographical names in the North African coastlands with those of the Greek and Thraco-Illyrian peninsula. The Numidian Μυκῆνοι are very suggestive. The legend of the Argonauts brought Jason from Iolkos to the Triton’s lake. The Maxyes of that very region claimed, as we have already seen, a ‘ Trojan’ descent, and the names of other bordering tribes point clearly to European kinship. If there were Thessalian Magnétes in Crete, there were Dolopes beyond the land of the Lotus-Eaters, whose neighbours again, the Eropoei, have a Boeotian and Illyrian ring. The foundation sagas of Cyrene, in all their variant forms, show that the Greek settlers recognised a very close pre-existing connexion between Crete and the opposite Libyan coast. How far-reaching was the scope of early Cretan enterprise may be gathered from the fact that the traditions of ‘Minoan’ settlement extend from Gaza to the Sicilian Herakleia.

The first colonists from Thera have to find a pilot from the Eteocretan district, a purple-shell fisher, namely, of Itanos,? whose legendary name, Korobios, has been compared with Korybas.!° The first Battos was held to

1 Pindar, Pyth. v. 82-88, and Schol. Pind. Pyth. v. 108. Cf. Studniczka, Kyrene, pp. 129, 130. There was an ᾿Αντηνοριδῶν λόφος between Cyrene and the sea.

3 Virg. den. iii. 133, cf. Serv. adloc. Virgil makes the Trojans come from Crete.

3 It is worth recalling that in Sicily as in Crete the archaeological evidence also points to an early ‘Trojan’ influence. A clay ‘idol,’ certain remarkable bone ornaments and several forms of clay vessels found by Prof. Orsi in ‘aeneolithic’ rock-tombs of the province of Syracuse are identical with those from the early strata of Hissarlik. (Orsi, La necropoli Sicula di Castelluccio, Bull. di Paletn. 1892, &c. p. 1, seqg. Cf. Patroni, Anthropologie, 1897, pp. 134, 139, 140.)

* Diodorus, xx. 17 records of Agathoclés, Ἔλυμαν τὸν βασιλέα τῶν Λιβύων εἰς συμμαχίαν προσελάβετο. See my note in Freeman’s Sicily, vol. iv. p. 419.

δ Thucydides vi. 2.

6 The spring Lilybaion, from which the town was named, seems to contain the Libyan word

for water=lily, according to Hésychios (s.v.). The word for ‘ water’ in use among the existing Libyan dialects—Kabyle, Shilha, and Tuareg— is however aman.

7 The name is not only frequent in the Libyan sepulchral inscriptions, but under the form Didi appears as that of the father of the Libyan Prince Marmaiou, who headed the great attack of Euro- pean and West Asiatic confederates on the Egypt of Menephtah. Another Didi, perhaps the son of Marmaiou (Maspero, Hist. Anc. des peuples d’ Orient. p. 266) fought against Ramses III.

8 Compare Oropus and the Illyrian Aeropus. The points of comparison between the early tribal and geographical names of North Africa and Italy are, perhaps, still more numerous. Compare for instance the Ausenses and Ausones, the river-name <Ausere (perhaps Wed Neffetia) and the Auser, Uthina, and Vedinum (Udine), Salassii, and Salassit. The Libyan connexion with Spain is still more conspicuous.

® Herod. iv. 150 seqq.

§tudniczka, Kyrene, p. 129;

376 FURTHER DISCOVERIES

be a grandson of a King of Axos,! and the nymph Kyréné herself was carried by Apollo to Crete* before she reached the land that was to bear her name. These more or less mythical traditions were not simply called into being to account for the fact that a third of the settlers in Cyrene were of Cretan stock. They imply that an earlier connexion than that established under Theraean leadership existed between Crete and Barka, and the fact that the pilot was chosen from that easternmost Cretan region, which, as we know from the Praesos inscription, retained its non-hellenic speech to the sixth century before our era, has perhaps ἃ special significance.

The race affinities subsisting between the early Cretan population and the ‘Trojan’ settlers in Barka and elsewhere may well have had a reflex action on the island. The Agean settlers on the North African coast may have become partly fused with the Libyan indigenes. The story of Dido and Mneas is, indeed, the poetic record of such a blending of the Thraco-Phrygian and native elements, just as at a later date the Greek colonists of Cyrene blended with their ‘ yellow-haired’ Libyan neighbours.‘ The fortune of war may from time to time have obliged some of these already half-acclimatised Aigean settlers to return to their older seats on the northern shores of the Libyan sea. In the same way the traditions of Danaos and Mgyptos—though these rather concern Rhodes and Argos— seem to point to a similar return wave of a European (or Anatolian) population from the Delta.

The ebb and flow of these early tides of Hgean enterprise and migration may have contributed towards the diffusion of Egypto-Libyan elements in primitive Crete. But there is every reason to infer an impulse of a more direct kind from the Nile Valley and its borderlands. In the forgathering of the Thraco-Phrygian and Libyan races there is no reason to suppose that the passive réle was always on the Libyan side. On the contrary, in the earliest historic records of this connexion between the mainland borderers of Egypt and the Agean peoples, it is Libyan princes, with distinctively Libyan names, who take the lead in the Confederacy. The enterprise of Marmaiou, the son of Didi, against the Egypt of Menephtah seems to have had its counterpart in the West. The companion and charioteer of Héraklés, whose Hellenized appellation, Iolaos,> covers the name of a Libyan divinity, finds his

1 Herod. iv. 154.

2 Libyka of Agroitas, Fr. ἢ. Gr, iv. 294. See Studniczka, op. cit.. p. 127.

3 Herod. iv. 161.

as Ἰόλαος in the Greek transcription. Halévy however supposes that the connexion of Ialaou and Iolaos the son of Herakles is a mere coin- cidence. ‘Les grecs, ayant entendu prononcer

* An interesting reference to the social inter- course between the Theraean colonists and the earlier Libyan inhabitants of Cyrene is found in Kallimachos, Hymn 2, 86: ὠρχήσαντο μετὰ ξανθῇσι Λιβύσσαις.

5 Halévy, Etudes berbéres, p. 157, where the form Jalaow of a bilingual inscription appears

en Libye le nom Jala, Jalaou, ont été naturelle- ment portés & y voir une de leurs divinités qui avait un nom semblable.’ But this does not explain the specially Libyan connexion of Iolaos even in Greek legend. The double ‘coincidence’ is rather too improbable.

OF CRETAN AND AEGEAN SCRIPT. 377

special sphere of action in Sardinia. Perhaps we may even detect a still further stage of Libyan colonisation in the name of Massalia.!

These echoes of more distant enterprise make it the more probable that Crete, where the records of a primitive intercourse with the Nile Valley are.so unmistakable, should have been betimes the goal of Libyan settlement. There were doubtless successive waves of migration in this direction, the impulse to which may have occasionally been the triumph of Egyptian arms over the Libyan tribes bordering on the Delta. In particular, the special relation in which the early Cretan remains have been shown to stand to the typical products of the Twelfth Dynasty period may be not unconnected with the Libyan triumphs of Amenemhat I. An abiding tradition of a historic ᾿ς episode of this kind, as well as of the fusing of Libyan and Cretan elements, may indeed be traced in a legend preserved by Diodoros. Ammon, expelled from Libya, settles in Crete, and marries Krété, the daughter of one of the Kuretes.? That a certain community of type between Cretans and Libyans was really recognised, appears from the ethnic classification of Polemd6n, the physiognomist, who divides the Libyans—he is not here speaking of Cyrenaean Greeks—into two classes: Negroes (Αἰθίοπες) and Cretans.3

ὃ 8. Proto-Egyptian or Egypto-Iibyan Comparisons.

The Egypto-Libyan connexions of prehistoric Crete invest any attempt to trace affinities with its early script on that side with a certain degree of ὦ priort probability. But this is heightened by the fact that the signs them- selves are found in the case of the Libation Table, of the stone vases, and of the prism seals engraved on objects the prototypes of which seem to be on the one hand of Old Empire Egyptian, on the other of ‘ Libyan’ origin. So far as the pictographic class of Cretan signs is concerned, although its general independence is clear, the influence of certain Egyptian hieroglyphic forms is unmistakable, and examples of this have been already referred to in my previous paper.‘ Fresh parallels of the kind may perhaps be detected in the two-horned symbol of Fig. 20 and still more clearly in the fringed or ‘door’ symbol on the seal Fig. 7d answering to the Egyptian sign for ‘ palace’ or altar.5 The second character on the Libation Table has also been cited as a probable example of this indebtedness.

1 Compare the Massyli and Massaesyli of the province of Carthage. Mas in the modern Berber dialects still means ‘son’ or ‘ descen- dant’ (Tissot, Afrique Romaine, i. p. 446); hence the frequency of this element in Libyan tribal and personal names.

2 Diod. Hist. iii. ὁ. 71.

8 Of μὲν Λίβυες Αἰθίοψιν ὅμοιοι, of δ᾽ εἰσὶ Κρῆτες. Polemon, Physiognom. lib. i. (in Scriptt. Physiognomici Veteres, ed. J. G. F. Franzius, Altenburg, 1780, p. 184). Polemon

who was personally acquainted with Cyrenaean Greeks, could not have embraced them under the —to a Greek—barbarous designation of ‘Libyans.’ Had he done so moreover, he would in this passage have committed the further absurdity of confounding the blonde, European-like Libyan element with Negroes !

* E.g. the adze (Pictographs, &c., No. 22), the saw (ἐδ. No. 23), the spouted vase (ib. No, 28), and the coil (2b. No. 69).

5 See above, p. 340,

378 FURTHER DISCOVERIES

But if we have here a derivative form of an Egyptian sign which in its primary sense of ‘ hall’ was connected with festivals and gatherings, we have just such an element as the Libyan borderers of the Egyptians may have borrowed but which the indigenous Cretans would hardly have sought so far afield. In other words the imitation of such a sign is on all fours with the imitation of Twelfth Dynasty decorative designs and of the form of the Libation Table itself which, as already observed, would most naturally have effected itself among a population actually bordering the Nile Valley.

The signs found on the ‘Libyan’ pottery of Naqada afford an interesting parallel to this phenomenon. There too, side by side with exclusively native symbols, others of which it may at least be said that they are common to the Egyptian hieroglyphic series occasionally appear.1 Among these may be noted a kind of vase (an), the water sign (mw), the signs for the king of Upper and Lower Egypt.

The field for comparison on the Egyptian side has been greatly enlarged by recent discoveries. In Table I. of my first work on the Cretan ‘ Picto- graphs’ the parallelism between the Aegean linear signs and these found by Mr. Petrie at Kahun and Gurob has been already set forth.2 But these signs, occurring on sherds and other objects for the most part of Twelfth Dynasty date, can now no longer be explained as the work of Aegean foreigners in Egypt. The further discoveries at Naqada and Abydos show that they fit on to an indigenous class which makes its appearance in the Nile Valley before the time of the first Egyptian Dynasties.

In the early stratum of the sanctuary of Min at Koptos, Professor Petrie had already discovered colossal statues of the God engraved with primitive hieroglyphs together with figures of animals all of which from their archaic style he ascribed to the prehistoric age of Egypt. Animal forms, lions and hawks of precisely identical types occurred in the cemeteries excavated by him and Mr. Quibell at Naqada and Ballas on the opposite bank of the Nile which revealed the existence of a hitherto unknown form of early culture in Egypt. The weight of local evidence however at that time inclined Mr. Petrie to assign the Naqada relics to a period posterior to the Fourth Dynasty though in any case earlier than the Twelfth.

Yet the difficulties in the way of such an explanation were from the first unsurmountable. How, for instance, explain the fact that among the contents of nearly 3,000 graves, no single scarab nor any familiar ornament of Egyptian fabric was discovered? If the remains unearthed at Naqada were to be simply ascribed to a sporadic settlement made by successful invaders of Pharaonic Egypt during the period between the Fourth and the Twelfth Dynasty, how was it possible that they should contain no scrap of the vast stores of plunder accumulated by such ‘ Spoilers of the Egyptians’ ?

1 See below, pp. 383, 384 Libyan Goddess, Neit, was probably the Libyan ® Petrie, Nagada, Pl. 111. Nos. 55, 74, 75, 76. crown generally, since its value bat corresponds The sign for the crown of Lower Egypt is in with the Libyan word for king, preserved, as relief. Mr. Petrie points out (op. cit. p. 64) that | Herodotos records, in the Greek Battos. this, which was the characteristic crown of the 3. Pictographs, &c., p. 80 [349].

OF CRETAN AND AEGEAN SCRIPT. 379

But these settlements are not sporadic. The evidence of their wide diffusion in the Nile Valley is rapidly accumulating and a wholly new light has been thrown on their date and ethnic relations by the discovery by M. Amélineau at Abydos of tombs belonging to the first two Dynasties and at Nagqada itself by M. de Morgan of the royal tomb, now identified with that of Menes, the founder of the Egyptian dynasties. The culture first brought to light at Ballas and Naqada now stands before us in its true relation to that associated with the earliest monuments of Pharaonic Egypt.' In part, no doubt it overlaps these earliest dynastic relics, but in the main it belongs to the true prehistoric age and to the indigenous stock which held the Nile Valley before the conquering race of ‘copper-smiths’ poured into it from the Red Sea littoral. It is at most survivals of the older form of culture such as we see it at Naqada that we find associated with the remains of the First and Second Dynasties.*

As to the character of the autochthonous race of Egypt, there seems to be good reason for accepting the view that they are to be identified with the people of the Oases—the Tahennu or Tamahu, a race of Libyan stock who in early times extended as far as the Nubian borders of Egypt. Members of this white-skinned race—so European in its affinities—still formed a distinct part of the Egyptian population as late as the fourth Dynasty though reduced to the position of helots* We may provisionally apply the term ‘ Egypto-Libyan’ or ‘ proto-Egyptian’ to this early indigenous population of the Nile Valley.

It is probable that the influences brought to bear in an intensive form by the Pharaonic conquest, were already beginning to operate on the primitive population of the Nile Valley long before the time of Menes. The native race had no doubt attained great proficiency in the fabric of stone vases at a period when there is no trace of actual contact with the dynastic Egyptians.

1k. Amélineau, Les Nowvelles Fouilles @ Abydos (1895-1896), Angers, 1896, 1.68 Now- velles Fowilles d’ Abydos (1896-1897), Paris, 1897: De Morgan, Recherches sur les Origines de U Egypte (Paris, 1896), p. 76 seqgg. The Abydos and Naqada finds and the views ex- pressed by M. de Morgan, M. Maspero, and Mr. Petrie on the civilisation to which they belong, are discussed by M. Salomon Reinach, Le préhistorique en Egypte d’apres de récentes publications, Anthropologie, 1897, p. 327 seqq. Thanks to the kindness of M. Amélineau, I have been able to inspect the greater part of the objects obtained by him during his two campaigns and thus to express an independent opinion on the bearing of these discoveries. The appearance of the second volume of M. de Morgan’s Origines with the account of the royal tomb, and the essays contained in it by Professor Wiedemann and Μ, Jéquier have

greatly strengthened the argument.

2 The contents of an intact tomb excavated by Mr. Quibell at E] Kab and presented to the Ashmolean Museum by the ‘ Egyptian Research Fund’ are specially important in this connexion. The tomb itself was dated by a cylinder bearing the name of King Khaires of the Second Dynasty, and in it, side by side with relics of the ‘Pha- raonic’ class, were painted vases representing a late development of ‘Naqada’ types.

3 See Wiedemann, Question de l'origine du peuple Egyptien in De Morgan, Origines de U’Egypte, ii. p. 219 seqy. It will be seen that, though divergent on some other points, Pro- fessors Petrie, Maspero, and Wiedemann are agreed in attributing the culture of Naqada to a people of Libyan stock.

4 78. p. 221. Cf. De Morgan, Origines, &c., i. p. 197 they are depicted with flat heads and red beard.

380 FURTHER DISCOVERIES

It has however been pointed out by Dr. Schweinfurth,! that the porphyritic and crystalline materials of which a large proportion of these vases consist must have been derived from the eastern parts of Nubia or even further afield.

But in the main this prehistoric culture of Egypt, like the race itself, has a Mediterranean range.2 It even shows some distinct points of sympathy with primitive Aegean culture. The obsidian knives* take us to Santorin and recall the very ancient relations between Thera and the Libyan coast. The early use of stone vases is equally characteristic of both areas. The primitive ‘idols’ of the Proto-Egyptians in some respects point to a similar relation- ship. The tattooed female figure from Naqada‘ which illustrates a practice foreign to historic Egypt, but universal among the Libyans down to their Berber descendants of to-day, finds its counterpart on the Aegean side. A primitive marble image of a squatting female found near Sparta, has its bare arms engraved with square and zigzag decorations,° recalling the tattoo-marks seen on the arms of Thracian women on Greek fifth-century vases. In certain bronze needles with steatite handles found in the prae-Mycenaean tombs of Amorgos, where various colouring materials also occur, Dr. Blinkenberg has recognised the actual tattooing instruments of the early Aegean population.” A rude Egypto-Libyan clay figure from the pre-historic cemetery of Gebel-el-Tarif’ though differing from the primitive marble ‘idols’ of the Aegean islands in its bent knees® and arms held close to the side, yet shows a remarkable resemblance to them in its general shape, while in its recurved flat topped head it reproduces one of their most characteristic features (see Fig. 33).

The steatopygous female figures of clay and limestone from the Proto- Egyptian graves, while also betraying a close analogy with certain types of prehistoric Greece and Thrace, are almost the exact counterparts of the primi-

1 De Vorigine des Egyptiens et sur quelques-uns de leurs usages remontants ἃ lage de la pierre (Extrait du Bulletin de la Société Khédiviale de Géographie, iv. Série No. 12 (1897), p. 16 seqgq. Dr. Schweinfurth points out that the materials of many of these vases point to the crystalline region east of the Upper Nile, and shows that something of this early industry still survives among the Bishareen and Ababdehs. The crystal bowls from Abydos are the most remark- able of all.

2 See especially Petrie, Nagada, p. 62 seqq.

3 Actual vases of obsidian were found in the tomb of Menes (De Morgan, Origines, &c., 11. p. 180, Figs. 625-627). Obsidian, however, is also found in Armenia (op. cit. p. 174).

4 Petrie, Nagada.

5 Dr. Wolters (Mitth. d. Arch. Inst. in Athen. 1892, p. 52 segg.), who considered that the en- graving simply indicated painting. But Dr. Blinkenberg, Praemykeniske Oldsager, p. 42 seqq. (Antiquités prémycéniennes, p, 46 seqq.),

has demonstrated the much greater probability that we have here to do with tattoo-marks. In the red streaks on the forehead and beneath the eyes of a large head from Amorgos (Wolters, l.c. p. 46) I have ventured to see the bloody nail-marksof amourner. Dr. Blinkenberg, how- ever (Joc. cit.), regards these also as tattoo-marks

6 Schéne, Museo Bocchi, No. 167, Pl. 3, 2; Blinkenberg, op. cit. p. 48.

7 Op. cit. p. 44. One of these instruments, found with marble ‘idols’ in a tomb at Amorgos, is in the Ashmolean Museum.

8 De Morgan, Origines, &c.. vol. i. p. 151, Fig. 373; reproduced vol. ii. p. 54, Fig. 111. I am indebted to M. de Morgan’s work for the representation given in Fig. 33.

® A slight bending of the knee is however visible in a marble figure from Phaestos ; (see my Sepulchral Deposit of Hagios Onuphrios near Phaestos, in Cretan Pictographs, &c. (Quaritch} 1895), p. 126, Fig. 129.

OF CRETAN AND AEGEAN SCRIPT. 381

tive Maltese figures! in the same materials, which reproduce moreover the curious side-squatting attitude This connexion with Malta fits in well, as Mr. Petrie has noticed, with the Libyan hypothesis, and another interesting parallel may be adduced which points in the same direction.

The exquisite flint implements of the ‘ proto-Egyptian’ tombs display in their most characteristic refinement, the chipping of a surface previously ground, an analogy with Spanish and even Irish Neolithic fabrics. This point of agreement with the extreme West may of course be an accidental coincidence, but the reappearance of the highly characteristic flint rings of Naqada and Abydos in a more westerly African region has an unmistakable

Fic. 33.—ProTo-EGypriaAN CLAY FIGURE, Fic. 34.—MARBLE ‘IpoL,’ AMORGOs.

GEBEL-EL-TARIF. AFTER DE MORGAN.

significance. Flint rings of the same type recur in a series of Neolithic stations extending trom the province of Constantine by the Wed Rir (Oued Rir) towards the central Sahara? Once more we are led in a Libyan direc-

1 See op. cit. p. 129; Petrie, Nagada, pp. him while making the artesian wells in the Wed 13, 34. Rir. These, together with exquisitely worked 31 ascertained this fact during a journey, in flint arrow-heads and other implements, were the spring of 1897, to the Constantine borders found embedded in layers of broken ostrich-eggs. of Sahara. My thanks are specially due to The flint rings are not mentioned in M. Jus

Captain Farge, of the Bureau Arabe at Constan- tine, and to the engineer, M. Jus, at Batna, who had found flint rings, such as those described above, in the Neolithic settlements explored by

H.S.— VOL. XVII.

earlier report on these discoveries, Stations pré- historiques de V Owed Rir. (Rev. d’ Ethnographie, 1887). The stations extend beyond Wargla.

DD

382 FURTHER DISCOVERTES

tion. The non-Egyptian practice of burying the dead in a contracted posture also recalls that of the Nasamones as described by Herodotus and reappears in the Dolmens of Tunisia, Algeria and Morocco.! The actual dismemberment of the dead body, practised by the Proto-Egyptians, has been compared with Diodoros’ account of the Balearic islanders who pounded together the limbs and body of the deceased so as to fit them into their funeral jars.?

It is probable that when the prehistoric remains of Barka come to be unearthed, the same habit of using stone vessels, which is so characteristic of the Proto-Egyptians on the one side and of the Cretans and Aigean islanders on the other, will be found to receive a wider Libyan illustration. On the further shores of the Syrtes the use of large stone vessels of very primitive aspect is still a native characteristic. During a recent journey through southern Tunisia, I had myself occasion to observe a primitive form of stone water-vessel identical with a type of at least Mycenaean antiquity of which I had previously seen several examples above or near cisterns in the great prehistoric city of Goulas in Crete.? It is needless, after what has been written in the previous section and in the account of the Phaestos Deposit, to dwell on the striking parallels presented by the primitive stone vessels of Crete and the Agean islands with those of the early inhabitants of the Nile valley.

The same Mediterranean range of affinities is perceptible in the charac- teristic red-faced pottery of the Proto-Egyptians, which as being most pro- lific in the early linear signs has a special bearing on the present subject. Similar forms of red-faced pottery have been observed by Captain Lyons in the Western Oasis of Dakhlah.* Elsewhere it finds its nearest parallels in the Copper Age cemeteries of Cyprus and the early strata of Hissarlik and of the prehistoric sites of the Greek islands and mainland. In the Libyan region proper it fits on to the red-faced pottery of the Tunisian and Algerian Dolmens and, like the steatopygous figures, reappears at Malta.®

These affinities of the early red-faced pottery of Naqada have an impor- tant bearing on the origin of the linear signs which appear engraved upon them, and tend to show that they belong to the non-Pharaonic, indigenous element of the Nile Valley. The culture with which they are connected has, as we have seen, a Mediterranean, especially a North African, range. In contrast to this, the most primitive hieroglyphs, such as those on the colossal

1 Cf. Bertholon, Exploration Anthropologique οἵ these—the conical cavity and ear-like ledge- de la Khoumirie, p. 66; Carton, Lw Nécropole handles are identical with those of the Cretan

de Bulla Regia Bull. Arch. 1890), ete. vessels (see Goulds, the City of Zeus, in the 5 Diod. v. 18. Cf. Wiedemann (in De Annual iof the British School at Athens, 1895- Morgan, Origines, &c., ii. p. 221). 1896, pp. 189, 190. For the general form of

3 Below the akropolis site of Takrouna was a the Libyan vessels, compare Fig. 11. The ears broken limestone vessel of this kind not far are seen better in Fig. 12. from the village well. In the garden of the 4 Cf, Petrie, Nagada, p. 63. neighbouring village of Dar-el-Bey was placed ° An interesting triple vase of similar ware another of the same kind, and probably from and primitive fabric is preserved in the Museum: the same locality. The characteristic features at Valetta.

OF CRETAN AND AEGEAN SCRIPT 383

Mins trom Koptos, show forms of marine shells and saw-fish, derived Professor Petrie has pointed out, from the Red Sea.!

The linear characters found on the Proto-Egyptian pottery at Naqada recur to a considerable extent on pottery found in tombs of the earliest dynastic period at Abydos, side by side with true hieroglyphic forms. At Abydos there is perceptible a certain reaction of linear indigenous signs on the more elaborate and pictorial characters of the Pharaonic people. Thus in several cases the linear forms here are simply Egyptian hieroglyphs very rudely scrawled.

In the examples—taken from rude vessels of the Abydos tombs—given in Fig. 35, there can be no doubt as to the hieroglyphic derivation of some of the characters such as the beetle (kheper) and the ka sign. There is therefore a presumption that the other signs grouped with them may be also linear sketches of true Egyptian forms, though it might puzzle an

as

SIGNS ON CLAY VESSELS, FROM ROYAL TOMBS OF ABYDOS. EXCAVATED BY M.AMELINEAU.

Fie. 35.

Egyptologist to identify certain of these. It is however noteworthy that while the hieroglyphic inscriptions in these early tombs occur on objects of higher artistic execution, and of exotic materials, such as the vases of crystalline and porphyritic rocks, the simpler signs are found on rude clay vessels made for humbler domestic use.

This reduction of the more elaborate hieroglyphic forms to simple linear signs, which at Abydos is quite unmistakable, finds a certain amount of analogy on the still earlier indigenous vessels of Naqada,? and suggests some curious questions. We now know that by the time of Menes the highly developed hieroglyphic script of the dynastic Egyptians had taken firm root in the country. But a large proportion of the hieroglyphic signs—the lotus-

1 History of Egypt from the Earliest Times to rude linear ka-sign, Petrie, Nagada, Pl. lv, the Fifteenth Dynasty, pp. 13, 14. No. 319. 2 A clear instance of this may be seen in the

DD2

384 FURTHER DISCOVERIES

sprays and river-craft, the water-birds, fishes, crocodiles, and other character- istic animals, already by the time of the First Dynasty become convention- alised types,—are of indigenous Nilotic origin. It follows then that many of the elements of hieroglyphic writing had been growing upon the banks of the Nile long before the time of the first historic dynasty. If the race that brought these pictorial elements to maturity is to be regarded as distinct from the old inhabitants of the land, whose remains have now been recog-

T A.B. EE. ΠῚ PROTO-EGYPTIAN OR EGYPTO-LIBYAN CRETAN AND AEGEAN SIGN-GROUPS SIGN-GROUPS’

nised at Naqada and elsewhere, it must at least have been brought into very early contact with them. Hence there is a possibility that the beginning of hieroglyphic script reacted on the linear native signs at a much more remote date than that of Menes. And the hieroglyphic figures themselves—how far may they not simply represent the coming to life of still earlier linear types? The same inquiry meets us in Crete.!

1 See below, pp. 394, 395.

OF CRETAN' AND AEGEAN SCRIPT. 385

There can be no doubt that the linear signs at Abydos on the whole belong to the same indigenous family as those from the pre-historic graves of Nagada. The correspondence, in form and arrangement, must be regarded as conclusive. In both cases these primitive signs may have been used for a limited purpose—often perhaps to indicate the contents of the vessels—but it will also be seen that some of the most characteristic recur among the ensigns found on the early painted pottery.

Both at Naqada and Abydos, characters of more pictographic aspect— and in some cases identical with Egyptian hieroglyphic forms—are at times coupled with the linear signs. Several of the groups of linear signs are found accompanied by one of this more pictographic class—an interesting point of correspondence with early Cretan and ‘Aegean’ usage. ‘This is illustrated by some examples! on Table III., where specimens of proto- Egyptian and Aegean sign-groups are placed side by side for comparison.

Thus in the first group the frequent y-like character of the Naqada series is headed by a vase like that which forms the Egyptian hieroglyphic an, standing for a tributary. In the second group two geometrical signs are preceded by a figure resembling the bowl or basket=neb, a lord. A third group which appears on two pots from Abydos, shows a more or less pictorial fish accompanied by a linear square. The fish is of frequent occurrence among the Cretan pictographs.

Taken as a whole, the two series of proto-Egyptian or Egypto-Libyan and Aegean sign-groups as seen in Table III. present a remarkable parallelism. It is true that there is no combination of signs which absolutely corresponds in the two classes. But the general arrangement is strikingly similar, and of thirty-two proto-Egyptian characters represented nearly a third show a close resemblance to forms that occur in the parallel series of Cretan and Aegean sign-groups.

A comparison of the individual signs of the two classes will be found in the first and second columns of Table IV. which are to a certain extent supplementary to those given in Table I. of my former work.? Due allow- ance being made for the selective process requisite for such a tabulation, it must nevertheless be allowed that the amount of parallelism in the two groups is very considerable.

§ 4. Comparisons with the Libyan and Tuareg Alphabets.

It is time to recall the existence on the Libyan side, at a considerably later date than the remains with which we have been dealing, of a very remarkable indigenous script which enables us to supplement these comparisons.

1 In representing the Naqada signs I have Kahun and Gurob, and were then described as eliminated tentative scratches due to want of ‘Aegean Signs found in Egypt.’ In view of skill in the engraver, and adhered to the the new evidence, especially that of Naqada, essential outlines. this description must be definitely abandoned.

2 The Egyptian forms there given were taken In Table iv. they are called ‘ Egypto-Libyan from the pottery discovered by Mr, Petrieat or proto-Egyptian’ signs,

FURTHER DISCOVERIES

386

Tie Ἐ7

" w%Zoam

1oxr

zo 2 = ~

ο ~ :

ΕΞ] ὃ Φ & >| = | =5 IDX] ΟΙΘΘ᾽ e~ |—

c

oO

az

LIBYAN IN

Gi Ue ΟἹ:

Sak

021 SL | IO: air] τ 8 }Ξ EFC HEE) 5000 τ ὃς = z = : a Ρ» π Ε- Ξ ly = — ΞΕ Ἐπ | ΤΙΝ HE] Pe Bical se. ΤΠ O@ Pe) | ἘΦ Ξ ὦ < £833 SK IEIO JOOP HTS |S) s| ὁ Co 22

| __8 ajw“a] sl 8} | AN νὰ 3 223 r πρ ἢ a] ΞΙ ἃ Ἢ : = ep = εἰ = ΗΕ he) πῇ ΒΞ | aS =| πῇ ἘΣ ΞΕ AL εν] ΓΕ] σΣ £335 Ἵ ¥ * 1 Fy} = 2 = x “ Ἑ ὼ ἐπε 3423) Sia edt NIK: ΞῈ -Ξ By εἰ Σ LSB τ γῇ S| ge ΕΠ: τ" >) FB) + De

CRETAN AND AEGEAN

rebel bal alee co "ἢ ἘΠΙ ΗΒΡΕΠΕΒΗΞΞΠΒΗΝΒΙΞ

OF CRETAN AND AEGEAN SCRIPT 387

The later Libyans possessed an independent system of writing which had taken such a strong hold on their national life that it maintained itself intact side by side with the intrusive systems of the Carthaginian, the Roman, and even of the Arab conquerors, and survives to this day essentially unchanged in the alphabet known as the ‘Tifinagh, of the Tuaregs, a Berber race of Sahara. The first known example of this Libyan script, the bilingual inscription of Thugga! containing a Phoenician as well as a native text, was discovered as early as 1631; the archaeological exploration of Roman Africa, for which the world is so deeply indebted to French enterprise, has now accumulated a whole series of Libyan inscriptions and among them some of the bilingual class, Carthaginian, Greek, and Latin. The diffusion of these Libyan records is very wide, extending at least sporadically as far as Sinai in one direction, and the Canary Islands in the other. The great mass of them, however, have naturally been found within the limits of the best explored region of Roman Africa, in eastern Algeria, otherwise known as the Province of Constantine, and now, since the French occupation, in many parts of the Regency of Tunis. The form of script thus revealed is quite distinct from the Phoenician and Latin forms with which it is at times associated, and would therefore seem to have been an independent Libyan possession before the days when the North African population was seriously influenced by the Carthaginian or the Roman types of culture. The earliest approximately dateable inscription of this class appears to be that of Thugga, which has been referred to the third or fourth century B.c., but the probability that much earlier examples will ultimately be discovered is heightened by the fact that the more easterly region where on general grounds its source is most likely to lie, is at present practically sealed to observation. That this script also held its own side by side with that of the Greek colonists in the region of Barka, is shown by an inscription in a cave at Derna, a district bordering on Cyrene. We have thus evidence of the existence on the part of the North African coast nearest to Crete of an ancient and independent Libyan script, which had struck such deep roots in African soil that the whole prestige of European and Asiatic conquest, of higher civilizations, and even of Mahometan fanaticism, has failed entirely to eradicate it.

The survival among the modern Tuaregs of this old Libyan form of writing is a phenomenon of great retrospective utility. The ‘Tifinagh’ or

1 Cf. Gesenius, Monwmenta Phoeniciae, Tab. xlviii. De Saulcy, Observations sur V Alphabet Tifinag, Journ. As. 1849, p. 247 seqg. Judas,

_ Etudes Phéniciens, Pl. xxxi. The inscription fixed into the fagade of a Mausoleum of a Libyan Prince, remained in situ at Dougga till 1842, when the British Consul-General at Tunis, Th. Read, ruined the whole wall of the monu- ment in order to obtain possession of it. At his death it was sold, and it is now in the British Museum. See P. Gauckler/’ Archéologie dela Tunisie, p. 18 seqq.

2 Faidherbe, Inscriptions Numidiques, Paris,

1870. Halévy, Etudes berbéres, Jowrnal Asiatique, 1874 (Ser. vii. Τὶ iii. p. 73-203 ; T. iv. p. 369- 416); Letourneaux, Du déchiffrement des In- scriptions Libyco-Berbéres (Fourth Interna- tional Congress of Orientalists, Florence, 1878), vol. i. p. 57 segg. ; and the papers in the Recueil des Notices et Mémoires de la Société Archéolo- gique de la Province de Constantine, notably those by Dr. A. Judas (T. xiii. p. 69 segg. T. xiv. p. 298 seqq.) Dr. V. Rebout (T. xvii. p. 55segq., Pl. I.—XII. ; T. xix. p. 211 segg., Pl. VI.—XIV.).

8 Some of these may now be seen in the Museum of the Bardo,

388 FURTHER DISCOVERIES

Tuareg alphabet which was first noted by the English traveller Oudney? in the Oasis of Ghat, has preserved almost intact? the great majority of the old letter-forms together with their values. This evidence added to that already supplied by the bilingual inscriptions, enables us to ascertain with certainty the sounds of most of the original Libyan letters. At the same time the long survival* of this ancient script among the tribes of Sahara, affords additional proof of the deep root taken by it in North African soil.

The correspondence between the old Libyan script and the Tifinagh is not confined to the form and value of the letters. It also extends to the arrangement, which on the Libyan inscriptions is almost indifferently in vertical or horizontal columns. Oudney in his account of the Tuareg writing remarked a similar peculiarity.°

The great simplicity of the Libyan script and the extent to which it is made up of mere reduplications of straight lines arranged alternately upright or lengthwise has suggested the theory that it is of purely artificial origin.® That an artificial element may exist in it is always possible, yet in other respects affinities can be traced with a much wider and geographically connected group of alphabets and the balance of probability greatly inclines towards the simplification of more complicated forms as against the spontaneous invention of a perfected system.

One or two isolated comparisons may be found between Libyan and Phoenician letters.’ A somewhat greater community may be traced between certain Libyan characters and those of the Minaeo-Sabaean alphabet of South Arabia, which have partly survived in the Himyaritic and Ethiopian. This Sabaean script is now known to be of great antiquity and its existence is alone

1 See Travels and Discoveries in Northern Africa in 1822, 1823, and 1824 by Major

stone in places they frequent, the Tuarick char- acters are hewn out. It matters nothing

Denham, F.R.S., Captain Clapperton and the late Dr. Oudney, London, John Murray, 1826, Vol. I. pp. xlvi., xlvii., Ixxxvii., lxxxviii.

* For the Tuareg or Berber Script and comparisons with the ancient Libyan see especially De Saulcy, Observations sur l’ Alpha- bet Tifinag, Journ, Asiat. 1849, p. 247 seqgg. ; A. Hanoteau, Essai de Grammaire de la Langue Tamachek, p. 3 seqgg. Letourneaux, JVth Congress of Orientalists, Vol. I. p. 57 seqq. Judas, De 0 Ecriture Libyco-Berbere, Rev. Arch. N.S. vi. 1862. Tissot, Province Romaine @ Afrique, I. p. 517 seqq.

8 An exception is found in ©, ΕἸ, equivalent to B in the old script, but now representing S.

4 The Tifinagh script is known to be still in use in the Ahaggar range of Sahara. It seems to have been also current within recent times in Morocco. M. Tissot was informed that MSS. of the Koran in the Berber alphabet existed in the Rif Mountains. Tissot, Province Romaine @ Afrique 1. 527.

5 Op. cit. I. p.

Ixxxvii, ‘On almost every

whether the letters are written from right to left, or vice versa, or written horizontally.’ As this last position is meant to be different from the others it is obvious that ‘horizontally,’ is here a slip for ‘vertically’ or in upright columns.

8 Meltzer, Geschichte der Karthager, I. p. 438 n. 26, in view of the ‘strong geometrical con- structive character’ of the Libyan alphabet suggests that it was a creation of Massinissa, in furtherance of his national Numidian policy. But it appears to go back at any rate consider- ably before his date.

7 As for instance the forms of the A, I,§ and Τὶ

8 Dr. Isaac Taylor, 7'he Alphabet (18838), Vol. I.. p. 153 observes that ‘in many respects the Libyan agrees curiously with the South Semitic Alphabets.’

Dr. Judas De Beriture Libyco-Berbére. Revue Archéologique, N.S. VI. (1862) p. 167, compares Himyaritic and Ethiopian forms,

OF CRETAN: AND AEGEAN SCRIPT. 389

sufficient to place the problem of the origin of the North Semitic or Phoe- nician group in a wholly new light. These somewhat distant affinities may be found to have a value of their own whenever the mutual relationships of the earliest scripts of south-western Asia and the east Mediterranean basin come to be satisfactorily elucidated.

But it is on Libyan ground itself that still closer materials for comparison may now be found. Wide as is the gulf of time that separates the earliest monuments of the class now under consideration from the

‘inscribed vases of Naqada and Abydos their general character corre-

sponds in a remarkable degree with that of these earlier Egypto-Libyan or Proto-Egyptian signs. In the later Libyan and Berber, a process of selection and differentiation has reduced their number and adapted simple linear characters of this primitive type to the needs of a regular alphabet. But the third column Table IV. in which the later Libyan, including a few Tuareg forms, are set beside the signs of Naqada and Abydos shows sufficient degree of correspondence with the earlier Egypto-Libyan forms to warrant the supposition that they may have been derived from a very ancient source. The fact moreover that throughout the course of over two thousand years, the Berber letters have remained practically unchanged, removes the improbability of their having retained their shape for a much longer period.

These linear forms indeed consist of simple geometrical figures which, unlike the more complicated pictorial class, were little susceptible of modi- fication. A cross, a circle or crescent, a line and its multiples, a square or two or three sides of it, two parallel lines crossed or joined, a zigzag, a triangle with crossed ends are distinguishing marks of such simplicity that they have little or nothing superfluous to throw off. It is however these simple linear forms that we already find on Egypto-Libyan vases at a date as early certainly, as the first appearance of Egyptian hieroglyphs.

The slight variation of form among the Libyan signs and letters at very remote periods suggests the further possibility of instituting a fruitful comparison between the later group of these and the Cretan and Mgean char- acters. In Table IV. the latter are compared with the Libyan script as seen on the gravestones of the native race in Carthaginian and Roman Africa, one or two variant examples of letters from the Tifinagh of the modern Tuaregs being also introduced. In the case of the slightly more elaborate forms the possibility of a certain degree of simplification must not be excluded, and, for this reason, conjectural comparisons like Nos. 8, 9, and 10, have been tentatively inserted in the Table. It must also be borne in mind that in the case of the Libyan forms a difference in the position of the letter counts for little. Apart from the fact that the Libyan characters are arranged indifferently in vertical and horizontal columns it also appears that the individual letters are habitually placed upright in the script of one African district and recumbent in another, so that they may be turned either way about for purposes of comparison. This variability of arrangement, which is, as has been already noticed, an Egypto-Libyan tradition, is shared, it will also be recalled, by the Cretan script. Upon the seal-stones the characters

390 FURTHER DISCOVERIES

are there found in upright columns! as well as in a horizontal order, and in some cases the lines apparently follow one another in boustrophedon fashion, alternately from right to left and left to right.

Oudney and his fellow-traveller, when their attention was first directed to the Tuareg letters, were at once struck with their European aspect. ‘We imagined,’ he writes, ‘that we could trace some resemblance to the letters of Europe, and conjectured that they had been hewn out by some European traveller at no very distant period.* In the same way Mr. Petrie first described the ‘ Egypto-Libyan ’ signs at Kahun and Gurob as ‘ Aegean,’ and M. Amélineau writes of ‘Greek inscriptions’ on the rude pottery from the Royal Tombs of Abydos. The early script of Crete has produced a similar im- pression. On first inspecting the characters on the Diktaean Table I found it hard myself not to believe that I had before me some archaic form of classical Greek writing, and the signs on the Phaestos whorl were considered by more than one archaeologist who had seen them to be Byzantine !

The comparisons above instituted between the early Cretan and Aegean characters and those of Kahun, Naqada, and Abydos on the one hand, and of the Libyan alphabet and the modern Tifinagh on the other, show a very real amount of correspondence. Not only do certain simple linear signs of the same class appear in use at a very remote date among the primitive inhabi- tants of the Nile Valley, but there, as in the Aegean area, they occasionally appear grouped in a way which indicates their application as a form of writing. At a later date we find a selected series of similar signs used throughout a vast West-African region with alphabetic values. The Naqada and Abydos characters moreover show another striking point of parallelism with those of prehistoric Crete. There too groups are found in which the linear signs are headed or supplemented by others of a more pictorial class resembling Egyptian hieroglyphics.

These correspondences become the more significant when taken in connexion with the other indications cited above of a very early and direct inter-relation between Crete, the Nile Valley, and the opposite Libyan coasts. The conclusion to which they seem to point is that the Cretan and Aegean linear script must in a certain sense be regarded as a branch of a very ancient stock having a wide North-African extension.

1 E.g. Pictographs &c., Figs. 21b, 24b, 25b, cases the arrangement is still more irregular, 30b, 806, and Fig. 32a, Ὁ, c, d, and inthe recalling that of Hittite inscriptions. present series Figs. 5a, 9b, and 22. Jn other 2 Op. eit. Vol. I. p. xlvi.

OF CRETAN' AND AEGEAN SCRIPT. 391

Part III.—Conc.upiIna ORSERVATIONS.

The vast antiquity which the discoveries at Naqada and Abydos now enable us to attribute to the use of linear signs among the primitive population of the Nile Valley makes it no longer admissible to assume that they were introduced there from the Aegean side. When already at an earlier date than that of the first Egyptian dynasty, we find ‘alphabetic’ signs already grouped in such a way as to suggest a definite system of writing, we have some warrant for inferring that the proto-Egyptians were ahead of the Aegean peoples in the evolution of their linear script. On the other hand the very ancient relations which have been shown to have subsisted between Crete and the Egypto-Libyan world would lead us to expect that the early script of the island like its stone vases and various ornamental motives may have been influenced, and partly derived, from that quarter.

That the Cretan linear forms were wholly of exotic origin it is impossible to believe. Simple as these signs are, and early as they appear, we are entitled by all analogy to suppose that the linear characters are themselves only the worn survivals of a primitive system of picture-writing, in which, like the first drawings of a child on a slate, various objects are indicated by a series of lines. And that this rudest form of pictography was practised on European soil there is abundant evidence. A good instance has already been pointed out in the rude horned animal or ‘ Minotaur’ which appears in linear strokes on one side of the Phaestos whorl, while on the other is seen the head alone. The clay whorls from the early strata of Hissarlik, the contents of which, as already noticed, afford some very close parallels to the primitive Cretan remains, supply a series of similar examples. A linear figure of a quadruped, for instance, is reduced by successive stages of degradation to one horizontal, and four or even three dependent lines! Identical examples are to be seen on the whorls and pottery of Broos in Transylvania and elsewhere in the Danubian regions, and very close parallels to the Trojan linear figures may be found as far afield as the sculptured rocks of Andalusia.2 In the ‘ Maraviglie’ and the still better examples, more recently discovered

1 Compare the figures on the whorls repre- sented in Schliemann’s Jlios Nos. 1867, 1879 1886, 1908 and 1912. The ornamental charac- ter of the zones on the Hissarlik whorls and the constantly occurring repetitions of what are really only variants of the same figure all round the whorl make it difficult to recognise in those of the primitive class any definite ‘inscriptions.’ Nevertheless the analogy which Professors Gomperz, Haug and Sayce have pointed out between certain Trojan signs and those of the Cypriote and Anatolian syllabaries can hardly

be gainsaid.

* Examples of these inscribed figures on the ‘Piedra Escrité’ near Fuencaliente are given by Don Manuel Géngora y Martinez, Antigiledades prehistoricas de Andalucia, pp. 65—67. The same reduction of the quadruped to 4 lines is perceptible. The Andalusian signs afford a very close comparison with those of the ‘ Writ- ten Stones’ (‘Hadjra Mektouba'), described by M. Flamand, in the south of the Oran Pro- vince of Algeria, Anthropologie, 1897, p. 285 seqq.

392 FURTHER DISCOVERIES

at Fontanalba in the Maritime Alps,! as well as in the linear figures on pre- historic stone monuments such as those of Brittany, Ireland and Scandinavia, we find analogous designs. It is in fact evident, without going back to the still earlier and very remarkable signs painted on the pebbles of the Mas d’Azil grotto,” that there exist throughout a wide European area the records of a primitive usage of linear picture-writing which already in prehistoric times showed a tendency to Simplify itself into abbreviated linear signs.

‘Nondum flumineas Memphis contexere byblos Noverat: et saxis tantum volucresque feraeque Sculptaque servabant magicas animalia linguas.”

But these general considerations are quite compatible with the view that the early linear script of Crete and the Aegean coastlands stands in a specially close relation to that of the Egypto-Libyan group. The existence of a primitive European stock of rude pictographs and their simplified derivatives need not be called in question. But there are many indications that in Crete at any rate the beginnings of writing like the beginnings of many other arts were influenced from the Nile Valley or its borderlands. In the case of the more pictorial class of Cretan characters this influence can be proved to demonstration.

It is on the face of it difficult to explain the appearance in a small and isolated area like Crete of a system of writing so fully developed as to present linear forms that have practically remained unchanged to modern times. Comparisons have already been instituted in my former communication between many of these and the characters of the Cypriote and Anatolian syllabaries and even with the letters of the Semitic alphabet. But to whatever extent the converse may be true, it is impossible to derive the older forms seen in Crete and some other parts of the Aegean world from the systems which first show themselves on the Syrian and Canaanite coastlands at an apparently later date. In saying this, however, it is not meant to exclude the probability that a branch of the same great family of primitive linear signs which have left their traces throughout such a wide North African region may have spread over Canaan at a very early date. The Lachish signs, so closely related to those of Kahun as well as to certain Aegean forms, seem to be an indication of this. On the Asiatic side, however, these primitive linear characters, if they existed there at a date as early as that of some of the Cretan signs, were overlaid and obscured by the spread of the cuneiform system which, as we know from the Tell-el-Amarna tablets, was the current form of writing throughout Syria and Palestine in the fifteenth century B.c. It is not till five centuries later that a more perfectly equipped form of linear writing, the Phoenician alphabet, was able as it were to shake off the

? See Mr. C. Bicknell’s communication to the go back to the early Bronze Age (see my obser- Society of Antiquaries, Dec. 9, 1897; Athen- vations Athenaeum, loc. cit.) aeum Dec. 18. These figures as is shown by 2 See M. D’Acy’s account of these discoveries, the appearance of the halberd with three rivets 3 Lucan, Pharsalia, III. 220.

OF CRETAN AND AEGEAN SCRIPT. 393

Assyrian yoke. It was the superior development of this, aided by the commercial enterprise of its possessors, that enabled it to oust, in part perhaps to assimilate, the more primitive and imperfect forms of writing existing on the Aegean shores.

The general results of the fresh materials that my recent journeys have enabled me to add to those already published may be summed up in a few words. The evidence that in early times, and long before our first records of the Phoenician alphabet, the art of writing was known to the Cretans receives striking corroboration. The view is also confirmed that we have to deal with two distinct yet inter-related systems, one pictographic in its character the other more purely linear. The generally indigenous character of the pictographic system emerges the more clearly from the occurrence of fresh examples illustrating the evolution of the conventionalised symbols from purely pictorial prototypes which occur on the more primitive class of seals. Thus we find the seated figure of a man, the disk with revolving rays, the spider, and a floral design common to the earlier and the later seals.

New evidence is also forthcoming of similar collocations of the later picto- graphs on different stones, such as the \\-shaped symbol and the ‘ polyp,’ the bent leg and gate,—collocations, which, like others already signalised, are specially valuable as showing that we have not to deal with the random insertion of chance figures but with a methodical graphic system. The discovery of a new class of pictographic seals of a form which could not have been used as an ornament, but is on the contrary that of a typical signet and closely akin to inscribed Hittite types, is also a valuable indication of the purpose- fulness of these groups of symbols.

The most recent discoveries fully corroborate the view, already expressed by me, that the later pictographic seals of the conventionalised class are mainly confined to eastern Crete, though a few like the convoluted bead-seal from Gortyna belong to the central area. The suggestion is thus confirmed that this quasi-hieroglyphic class which comes down to the borders of the historic period was the special property of the Eteocretan stock. Elsewhere in the Aegean area, as to a certain extent in Crete itself, the linear char- acters still continued in use, and they seem to have had a closer relation to the dominant elements of the Mycenaean world on the Peloponnesian side. In a more general sense, however, the name ‘Mycenaean’ must be equally applied to the peculiarly Cretan group of pictographic signs.

The linear system on the other hand, though it also overlaps the other, goes back to a very remote period. It seems to have reacted on the picto- graphic class, and to have been partly incorporated in it, but in this case, unlike the other, the proofs of evolution on Cretan soil from pictorial originals are not always so clear. The rude linear figures of men and animals on the very earliest class of seals partially indicate indeed an indigenous source: and in the Phaestos whorl we see the head of what is a rude linear animal on one side, becoming on the other a detached symbol. But the impression derived from the new materials supplied by the Psychro Libation Table. is that this linear script had at a very early date attained a maturity

394 FURTHER DISCOV ERTIES

of form and a systematic application which secius to imply a long antecedent evolution, and is best explained by the influence of an older civilisation such as that of the Nile Valley. The Libyan clement may, as suggested in the preceding sections, ultimately supply the link of connexion and explain how a more advanced system was brought to bear on the ruder family of Cretan and Aegean linear signs.

The evidence that has here been put together is in part indeed of such a nature as to place the very early relations between Crete and Egypt beyond the range of controversy. We have not only to deal with borrowings of Twelfth Dynasty decorative designs, of types of stone vases peculiar to the Old Empire, and even in the case of the libation tables of articles of cult. The Karnak prism seal shows the most typical form of Cretan seal in use among a probably Libyan population in the Nile Valley as early as the Sixth Dynasty, while the allied group of cylinders brings a whole series of Cretan and Aegean types into connexion with the same primitive element. Finally, the linear signs themselves, and a whole series of early fabrics tend to show that a close relation existed between the indigenous population of the Nile Valley and those of Crete and the Aegean Islands at a period so remote that it goes back beyond the earliest historic dynasty.

That the linear or quasi-alphabetic signs, whether of primitive Egypt or of the Aegean area, were in the main ultimately derived from the rudely scratched line pictures belonging to the infancy of art can hardly be doubted. This consideration helps to explain the intimate relation in which Cretan linear signs stand to the later and more pictorial characters. For certain purposes fuller and more literal representation was still adhered to in the linear series, and a pictograph pure and simple appears at the head of linear signs in which the prototypes are no longer so easily recognisable. It has been noted that both in the Naqada and Abydos groups the same com- bination of the two kinds of character is found as on the early Cretan prism-seals.

But this partial survival of the practice of pictorial representation in place of linear ‘shorthand’ was as nothing to the wholesale revival of the pictographic style which took place in Crete during the Mycenaean period. This revival corresponds in the island with a renewed period of intensive Egyptian influence under the Eighteenth and Nineteenth Dynasties, so clearly marked in the borrowing of decorative and other designs. It is not unreasonable therefore to believe that it was this Egyptian influence which here, as in the neighbouring Hittite regions, promoted a reaction orn a more pictorial style of script.

The linear figures assume a more realistic aspect in keeping with an age in which the engraver’s art and the artistic sense were more highly developed. On older stones like the Phaestos whorl? or the Arvi pendant? we see a mere. outline representation of a horned head like the Phoenician aleph. The symbol now takes a fuller form and clothes itself as it were with flesh and

1 Pictographs, &c. p. 15 [284] Fig. 11b. 2 Jb. p. 17 [286] Fig. 16.

OF CRETAN AND AEGEAN SCRIPT. 395

bluod. A mere circle completes itself asa human eye. The upright and cross lines that seem to have stood for a tree take again a more vegetable shape. In this way we may obtain from the more advanced representations of certain objects a retrospective light on the meaning of an original linear form. At the same time a whole series of new symbols, a few of them direct borrowings from Egyptian or Hittite sources, is introduced, of which no prototypes can be found in the earlier linear series. The repertory becomes larger; more complicated, but also more expressive.

From the alphabetic point of view indeed this process must be regarded as in the main reactionary, though thoroughly characteristic of the in- fluences predominant in dynastic Egypt. However imperfectly applied as yet to the purposes of a formal script, the old linear forms,—such as we see them both in the primitive Aegean strata and in prehistoric Egypt,—were those that ultimately triumphed in the Phoenician letters. The primitive engraver who had made an ox’s head with an angle and cross strokes or a tree with an upright and three horizontal lines was nearer the beginnings of alphabetic writing than the artistically trained Egyptian whose picture-sign informs us of the genus and species."

ARTHUR J. EVANS.

1 On this, as on the former occasion, 1 materials and institute comparisons. To those have refrained from any attempt to interpret who care to embark on more ingenious specula- either the linear or the pictographic script. It tions Dr. Kluge’s work Die Schrift der Mykenier has seemed to me that in the present stage of may afford food for reflection but hardly for the enquiry the main object should be to collect | encourageinent.

396 A SUMMER IN PHRYGIA: I,

A SUMMER IN PHRYGIA: I.

[PLatTe XII.]

DurinG the summer of 1897 I had the opportunity of making extensive exploration in Phrygia, and the following paper gives, as a first instalment, an account of the more important results of the season’s work there. I have given a map (Plate Χ11) based on the Ottoman Railway Survey to illustrate the watercourses of the Laodiceian district, but I regret that a map to show the new sites has had to be deferred. At the outset I must make acknow- ledgment of the valuable help I have received from Prof. W. M. Ramsay, who has kindly sent me some criticisms and suggestions. For the numerous references to his volumes on Phrygia no apology is necessary. Every student of its history must use his brilliant pages as the basis of his study; and the work of the explorer in the districts which they cover must naturally be to. endeavour to amplify the information, and to confirm or correct the views, which he finds there.

Few parts of Phrygia have been so frequently traversed as the Lycos valley with the adjacent Carian and Lydian frontiers: yet anyone who studies this district will be astonished at the number of unsolved problems which it presents. To begin with the Carian borderland and go round the valley of the Lycos, the first problem that confronts us is the site of

KIDRAMOS,

a city without annals, but important enough to possess a coinage of its own, at least from the time of Augustus to that of Julia Maesa. A review of the evidence for its situation will be found in Prof. Ramsay’s Cit. & Bish. of Phrygia, i. p. 184.1 His conclusion is that ‘it is to be looked for . . . between Antioch on the west,and Attouda or Karoura on the east, perhaps somewhere opposite Ortakche, on a spur of the hills that fringe the valley.’ In accordance with this view, which seemed very probable, I searched the district carefully. We were exploring the right bank of the Maeander to see if any site could be found there for DALDIS, and after an unsuccessful search? we intended to cross

1 This work will be referred to as CB. The place Daldis in the Hermos valley near Hiero- other abbreviations need no explanation. caesareia (CB. i. p. 179). See also Buresch,

2 Although exploration can hardly ever claim Reisebericht in Ber. der Kgl. Sachs. Ges. ἃ. to be final, this will be regarded asanindication Wiss, (Leipsic), 1894, p. 91: and cp. M. Imhoof- in favour of the alternative view which would Blumer in Rev. Suisse, 1897, p. 211 f.

J.H.S. νοι. XVII. (1897) @

n=

MAP OF THE

DISTRICT “OR ΟΕ Based on the Railway Survey.

\ ᾿ OR Ὁ ST ἡ» ALS ald) > OF “¢ a et eRe — Tches SS δας πος πόδε “Ὁ : -S=0F Boe 0 cot OL, SH SBRENGSSOP 28-10 6 gi ἃ ὩΙῸΝ iPuk= υω a th oe, α J SS == fy? Kizil Kak Net et “9: SSS τς hal ΣΩ͂Ν ΣΞ fey ae to” Zz “ass at 2a 40d i o% 46 τὰ ὦ Kodya Ba ἤ τ: a, Dia » Ve . ἢ os) ὑψ' Se — Ρ ‘~ (Gp » erekeut Va Ag Εν ΤῈ 433 Εν ἵ iY Sivri Kalek Laci Se. “ Hm MN See i,5 Pgh alekeu uajalt > 95 2 Feat, ἡ Pr “fl GRudjal eS Ae Bride yee? fis ne SAY chiftlik 3336 q i HA Gerekeus i πο. Ἂν .} xe : co OSS ie Ἔν Γ᾿ ΨΥ, 2%. : eu ' " ie πε Re "tile Ve ra ae ewes ἢ »Ἥ sot, ad gk N ne ΤῈ PCR 7 “ἢ ae, + Menteshe i τὰ Κ᾿ ee Ὁ αι ΟὟ ΣΝ ΐ sent eae * Mills Jem Ἂν» τ. ἃ ὋΣ τῷ , bce SEE of | IM: ; Py ae τἀ ἀν... 1600 Bed of Ry, δ ᾿ ' ape >> nya 3 Ϊ ψο εκ Ἀγ τ αὐ ραν, / Of Like lt | ES 5 TANG See ae es\&." « FS. ΩΣ 3 PS Jit Wer es hw ye SE ἘΠ ~ Ἐπ # ἱ eke A In htt γι ΟΝ TRUE Ke 5 , we ἜΣ ἐκ 2640 Μὸν Pas, o NAW APSR Tohukur a. ~ ee τ cA 1:3: ον σα ΤΡ Δ “πο ~ δέρεη, ἐδ Α oe, Nena 3700 i ἃ ean = . . @ Springs ΟῚ ἃ ᾿ eho Sat,5 Soriova He) ‘Shae Pe Shes Pe πιο Μ Gulub ‘ 1S Ἂ 8 ting Aq a . hy Ὁ ᾿ ps 5% ἢ ey, ῳ τὶ k ᾿ δα ἃ 9 Εν Aes Wey ὸ © -«- , Yurei/ 310. Ὑ fal * Yataghan . Merengue ὃ = Kuyujak es ¢ ο Ο τὶ Ρ: ; ο Scale of ο 1 2 3 4€ng. Miles ; (bet ttt eed THEMISONION ase ] Heights wn feet above sea level. Kara-eyuk Bazar σ᾽

τ ᾿ 5 CORNER αν ομμ νοι ἜΣ ΒΞ ete τος: pie ee een Pon ' ἸΉΤ 0 ἩΧ et [HDIQOAL FO WI rele ν"

evry you \eA sds we Weel, Ne

te -

1 κ Ae he hus ya oh

bah

᾿ ae bhe vein had! ὙΠ fees * ; Mn at the ides inble |

᾿ mAs Mealy avin heelerredt ies i i eye

i a Nieto y Tater eu + tah Gp copes

Δ, ΨΥ ΔΝ

ἜΗΝ ν᾽ ich eet ap τ

ΐ workvol the ΧΡΊ ΚΣ Ὁ] he CameTi CRs ὑοῦ - hha potar Το asiggllh tly en

wis hy tie πηάξᾳειτοιι ελμὰ ἐφ᾽ ν 1 Hew ὃ τ. ny yee bx γς ἐμὰ ΚΝ ΤῊ

"ὦ δ) wash iio Bayle oT Cm! nia? εὖ ἜΝ “ἢ ΡΝ tiews:

idee ay Ratxwt walt 3 Aah ἡ} Figg ἢ ‘a the Αι πο ip

vohaah ἡ ᾧ ) Τὸ apie λὶῷ ue i 5 horde anhd ᾿

ver boy νῷ i #4, Ligh ine ᾿ δ Stat Welw 7 Ν ἡ we [ὁ Lhe sito

ae ng wes he : eyo "" ᾽ ou

» make δου ᾿ς δι ge Laut? rom thy Oo F we \eaet tiv wee MO ἢ cies an, Ν᾿ ᾿ 7 | d ete on yh mi ἐτι: \¢ ela ᾧ γον, Α͂ ett τῷ

| cya ABA WOES ἐν, RSE PO οι

) ’ Al beds ma ¢ pon, at sch Kare ifn ὉΠ the '

een he © née, ga 5 εν bile thst [raion Ata alle ®

nth eile VP Dia μ᾿ ve probable, 1 searched

i ἢ [Ἢ } ‘@ wi Ἂν νὴ . hole Lye VMucun ter te

| » fara 4 LBs, vs ὅν. an anaes) δ

0 φιι ᾿ ᾿

' ᾿ “orgies aft Ν ἈΝ γον σαν σὸς

ial id rhA auth itl bo balhyret we ὦ CH σὰ pally Dalle shox eblrwr'ert'gata tors wh pays ὧν, omentee (ΟὟ. LB ATER

σ' λϑνοος ἰν κα νϑυνογν < ῳ herGe peewee ὦ in Der dae ;

«te be Pee), Chlp WIS ἣν fy "εῦνν Spi oon ie. ty), FSPa pe Bhs ood ep ἯΙ

te ντωι! af ἡμενου οδροῦνο glewsge ἡ νην μῳχῖς υυἱϑμάνν ia Re Sotn, 2007, ΚΙ’ ἘΠ ©

᾿ Meet AT eee ew 6.0. et ®S ‘epee duane Α heel) eee κὸν ἢ ἶ a Apes ae eae —~ sd ere

"Tey } . ὃ mG ὦ... 7 . De ἘΝ

A SUMMER IN PHRYGIA: I. 397

the river beside Ortakche and explore the spurs of Tchibuk Dagh, which here come down almost to the water’s edge. Finding the wooden bridge broken, and unable to ford the river, we were forced to go down the right bank nearly as far as Antiocheia to find a bridge. Here there is a wide and fertile plain narrowing a little at the village Yamalak and then widening again, as one goes eastward, up to the slopes of Tchibuk Dagh nearly as far as Ortakche on the opposite side. Crossing this plain we reached the ¢ch iftlik Budjak keui which lies high up the hill side, about 500 ft. above the Maeander valley} and three miles or so from the river, hidden from the view of the traveller in the plain. The village was seen to be full of ancient stones, and the desired site was found a short distance to the south on level ground running out from the hillside and looking down to the valley, about 100 ft. lower than the village. The remains on the site are all late: we noted what seemed to be the line of fortifications on west and south-west, two ends of a low arched way built of small stones with cement rising a little above the level of the ground (late Roman or Byzantine work), the lower part of a sarcophagus hewn out of the natural rock, foundations on the hill-side, and nearer the village the foundations of a Byzantine chapel, etc. A short distance below the village there has recently been excavated a rectangular chamber built of fine marble blocks and roofed over with flat stones cramped together (as we were informed) with lead. It was probably a tomb.

Only one inscription was found. It is built into the wall of the mosque.

ἢ ΖΡ ΟΔΉΜΟΣΞΕΤΊΛΗΞΣῳ

ΑΙΣ ΤΑΙΞΣΜΕΙΓῚΣ T AISKAIKFY

ΔΙΣ ΤΑΙ ΤΙ ΜΑΙ ATIOAANNI YY

ΘΗΝΑΙΌΡΟΥΤΙ Ἀττι αν ANAPAAY

ΘΟΝΜΆΑΙΦΙΛΟΙΠΑ ΤΡΙΝΓ ENOMEY

KAIAGAT ENOY STIAZSHAPETHW NHNOX OTA

ἡ βουλ]ὴ καὶ ὁ δῆμος ἐτέίμησ[αν maa jas ταῖς μεγίσταις καὶ κ[αλ- λ]ίσταις τιμαῖς ᾿Απολλώνι[ον ᾿Α- θηναγόρου Παπίαν ἄνδρα ἀγα-

5 θὸν καὶ φιλόπατριν γενόμενον καὶ διὰ γένους πάσῃ ἀρετῇϊ διε- νηνοχότα.

It is perfectly safe to assign the name Kidramos to this site. Prof. Ramsay’s inference, as will be seen, hit the mark remarkably well. He has brought out the fact (CB. i. p. 166) that the city belongs to a small numis- matic group, of which Attouda, situated on the other side of the hill, is another member. In this connection it is worth remarking that of a few coins which I saw there, those which were not Byzantine were coins of Attouda.

: Heights astininted by aneroid (except in the Map, PI. ΧΙ]... i H.S.—-VOL. XVII. E E

398 A SUMMER IN PHRYGIA: I.

KAROURA.

Continuing along the southern side of the Maeander, we crossed Tchibuk Dagh en route for the famous hot springs. On the sides of the hill there are traces of later ancient life (e.g. several tombs opened by the villagers in search of gold) and remains of an old paved path leading up to a site which was perhaps a Byzantine refuge, 1,000 feet above Budjak keui. On the level summit of the ridge, where there is a Yuruk yaila 1,800 ft. above the valley, we left the path to Attouda (Assar, reckoned four hours from Kidramos), here a good and easy road, and descended towards the springs, north-east of Tekke keui. This spot is one mass of springs which have formed an enormous marsh, emitting evil sulphurous vapours ; towards the western end there are two conspicuous sources, whence boiling water bubbles up and sends off steam: and between the marsh and the old ruined bridge over the Maeander, near the river bed (ὑπὲρ τοῦ χείλους, Strabo p. 578), we saw several dried up springs. ΑἹ] around the soil is white with the lime deposit. There can be no reasonable doubt that KAROURA was situated here (CB. i. pp. 2, 170, Strabo 578, 580).1 In an old overgrown Turkish cemetery near the village Kab-agatch there is an extraordinary number of column drums (some with dowel holes), varying in diameter, though several are of equal sizes. These heavy stones would not be carried far: they must have come from beside the hot springs and perhaps belonged to a temple of the god at Karoura. No other evidence bearing on the question was discovered.

ATTOUDA.

The village Assar occupies part of the site of Attouda. Approaching the village from Gumuldjak (between Karoura and Khas keui) and keeping along the banks of the Assar Tchai, we came after one hour to the foot of a steep hill, up which there winds in fine curves an old paved road,? which may be old Turkish, but is more probably the remains of an ancient road from the valley of the Maeander and Lycos. This plateau is divided from the hill on which the village stands by the deep cafion of the Assar stream. The village itself occupies a fine site (fully 1,600 ft. above the Lycos valley), surrounded by deep ravines on all sides except the south-west, where the ground slopes gently down to a wide depression which merges again into the hills beyond. It is full of ancient remains of all kinds, built into walls or lying about serving no purpose, and it stands upon what was clearly the acropolis of the ancient city. On the top of this acropolis, above the roofs of the houses which climb up the slope, the natives have recently excavated a large square cistern, arching slightly towards the top, of the ordinary Byzantine type.

1 Prof. G. Radet, in the map attached to Hn Serai keui, where no remains exist. On the Phrygie (1895), gives up his former identifi- recent growth of Serai keui CB. p. 164, 168. cation of Karoura and Kydrara, and now places 2 Traces of pavement were also seen quite Karoura here, while still leaving Kydrara at near Gumuldjak.

A SUMMER IN PHRYGIA: I. 399

Some new inscriptions ! were recovered ; there are probably many more to be found, but the villagers were very suspicious and obstinate.

The only inscription which fixes the site is given in 0.1.4. 3950 from an imperfect copy of Sherard, in which Franz detected the name ’Arrovdéwy. It deserves to be repeated.

2. Lying before the mosque enclosure :

ATAOH TYXH ᾿Αγαθῇ Τύχῃ.

Μ ΑῪΡ ΑὙΔΙ © NE TT! M. Avp. Λύδιον ἐπί- TPOTIONLCEBAL TWN τροπον Σεβαστῶν

HROY AHKAIOAHMOL ἡ βουλὴ καὶ ὁ δῆμος SLT O YAEWNTON ᾿ΑἸττουδέων τὸν

EYEPTE ΤῊΝ αὑτῶ]ν εὐεργέτην.

In the last line there seems to be no room for [ἑαυτῶ]ν. The inscription probably dates ca. 162-180 a.p. (cp. CB. ii. No. 235), in which case the emperors would be M. Aurelius and L. Verus or Aurelius and Commodus.

M. Aur. Lydius proc. Augg. was probably a freedman of M. Aurelius. His duties would be to superintend the /iscus (CB. i. p. 71).

3. Ina garden, very small letters engraved on the narrow end of a rectangular block :

Jp E OVE TIMINS AD E PMOFENH XAPIY

ETTAIN QUITTPOEAPIAI ΚΡΥΣΩΙΣ TEGAN EIKONIFPATITHEIKONIXAAKH vec \

PPETHZEENEKENKAIE YNOIAS ve YHETIPOS TOYEINEOYS

Οἱ] Νέοι ἐτέμ[η]σαν ‘Eppoyévn χάριτι, ἐπαίνωι, προεδρίαι, χρυσῶι στεφάϊΪνωι, εἰκόνι γραπτῇ, εἰκόνι χαλκῇ, ἀ]ρετῆς ἕνεκεν καὶ εὐνοίας τῆς πρὸς τοὺς Νέους. Προεδρία granted by the Neoi can refer, of course, only to their own meetings.

4. Ibid., a small tablet: 3 OAHMOC Ὁ δῆμος

ΤΕΙΜΗΓΕΝ ἐτείμησεν τὸ PION KAAY Τιβέριον Krav- AION BHPYANIA διον Βηρυλλια- NON-A ANAPA 5 vov, ἄνδρα K AAONKAI ATA καλὸν Kab aya- OONAPETHCE θὸν ἀρετῆς ἕ-

ΚΕΝ (ve)xev.

1 Inser. of Attouda, C./.@. 8949 ff. ; Le Bas- pp. 238-9; CB. i. p. 181-38. Wadd. 743; B.C.H. 1887, pp. 348 f. and 1890, EE 2

400 A SUMMER IN PHRYGIA: I.

5. Built into the wall of a house, upside down :

YTATITIATPIAI PAS TOSTY XIKOY 2 T1 SAIABIOYKTA YNTTAN ΤΙΤΩΠΕ “<MOEKTANIAI ~S\ENZYN

SS

TH γλυκ])υτάτῃ πατρίδι Αὐρ. "Adlpactos Τυχικοῦ ἀγωνοθ]έτης διὰ βίου κὲ τα- μίας, σ]ὺν παντὶ τῷ πε-

5 ριόντι κόσ]μῳ ἐκ τῶν ἐἰδί- ων -ἀνέθηκ]εν συν- ὀλως].

Adrastos, son of Tychikos, may be the person mentioned in CB. u. No. 71, restoring [’A]épdo[tov Τυχικ]οῦ for [Λοκρί]ου. The name Adrastos is known to be common at Aphrodisias, and occurs at Trapezopolis: it was no doubt very common at Attouda which worshipped Μήτηρ" Αδραστος.

6. In the wall of a house:

ἡ βο[ υ]λὴ ἐτείμησαϊ[ν (sic) ᾿Απολλώνι[ον ᾿Ασ 1- κάντου.

7. In the cemetery wall; the first part was engraved on a separate stone:

[ἡ γερουσία (or ἡ βουλὴ) ἐτείμησεν τὸν δεῖνα) ταῖς μεγίσταις καὶ καλλίσταις τειμαῖς, ΝΜ \ ἈΝ 9 ’ ἄνδρα καλὸν καὶ ἀγαθὸν γενόμενον περὶ τὴν πο-

ὅ λειτήαν ἐπί τε ταῖς τῶ[ν προγόνων αὐτοῦ [εὐ- δοξίαις ἐπί τε tails χρεί-

ἃ Ν Ἁ Ν \ ais ἃς κατὰ τὸΪ καλὸν Kal / > μεγαλό[φρον ἐν παν- 10 τὶ καιρῷ τ[ῇ γερουσίᾳ (or βουλῇ) ὑπέσ-

χετο ane TOS ocr φῦ Sth τσ, les

Before leaving Attouda, we should note the fact that to the south of the modern village, separated from it by a ravine, there is a high conical hill whose summit is crowned by a Tiirbe (the tomb of a local saint, as we may say). This T%irbe,as Prof. Ramsay has pointed out in other cases, preserves

the sanctity attaching to the old town, the seat of the worship of Men Karou.!

Δ At Apameia there is a Tiirbe on the hill above the town, to which the natives go up to pray when they want rain.

A SUMMER IN PHRYGIA: 1. 401

TRAPEZOPOLIS.

Another important problem in the topography of the Lycos valley is the exact situation of Trapezopolis, which is assigned by the evidence to this district. The arguments are stated, with the author’s usual acumen, in CB. i. p. 171 f. He infers that ‘Trapezopolis lay north of Mt. Salbakos near the frontier between Caria and Phrygia and west of Laodiceia in such a position that, when the Phrygian frontier was moved a little further west, Trapezopolis came to be in Phrygia, not in Caria’ (p. 171): that it was conterminous with Attouda’: and consequently that it was probably near Kadi keui and ‘corresponding to it, though perhaps not on the actual site’ (pp. 165, 172),— a judicious proviso which is merely the application of a general rule proved by all experience. Exploration has confirmed each of these inferences, and justified the order of Hierocles. The ruins of the city were discovered on a plateau north east of Kadi keui and about an hour and a quarter south-east of the railway station at Serai keui,’ just where the higher spurs of Baba Dagh (Salbakos) merge into the curiously-shaped alluvial hills of the Lycos valley.

The east side of the plateau forms one bank of the cafion called Gebe Deressi,t through which flows the streain Dondjali Tchai (the small river nearest Sara keui on the east), and the west side is bounded by a similar ravine; on the south the slope is short and easy but steeper again on the irregularly shaped north side. The eastern and western sides, composed as they are of horizontal strata of clay, dipping perpendicularly down, have fallen away to some extent, carrying stones and foundations with them: at the river side, 400 or 500 feet below, we saw numerous large blocks of marble, and on the top one could see the foundations stopping abruptly at the edge.

The site, which still retains the latter half of the name in the form ‘ Bolo,’ conspicuously justifies by its shape the title ‘Table-City, ® especially when one gets a view of it from the side of Baba Dagh (Mt. Salbakos). A search over the plateau, which is now turned into cornfields, revealed numerous remains of all kinds.6 The foundations of the aqueduct, which brought a supply of water down from Mt. Salbakos, can be easily traced for a consider- able distance, and lying about we saw some of the stone pipes, which are of exactly the same form as those that are found in such quantities, largely

1 An interpretation of a coin (described p. name which occurs elsewhere, and seems to be

166), which will be seen to be justified.

2 The rule is stated ¢g. p. 168 πη. 1. M. Radet places Trapezopolis at Kadi keui, but without any proviso. At Kadi keui we did not see the slightest trace of an old settlement. In this district between Tchibuk Dagh and Laodiceia he crowds together a host of towns, several of them on sites where there is no vestige of ancient remains. method.

3 The name is so given by the Railway Com- pany, but it is always pronounced Sara keui, a

This is wrong’

the correct form here.

4 This was given me as the right form of the word : I was corrected when I used the form Djebe.

5 The suggestion that the name was probably significant is thrown out in CB. i. p. 172.

8 The nearest village is Seine keui at the foot of the Dere by the river side. It contains no remains: it would be very difficult to transport heavy blocks down the steep side of the caiion, but even those blocks which have fallen down have not been carried away.

402 A SUMMER IN PHRYGIA: I.

in situ, at Laodiceia. It is possible that a semi-circular recess in the northern side contained a theatre: there are a considerable number of blocks there, and clear traces of building. On the east slope, standing out from the hill side, just below the level of the plateau (but mostly beyond the actual site, ze. towards north), we found an enormous number of low archways built of flat stones cemented together, and closed by a wall at the back, the sides and ends being often pierced with window-shaped openings. These perhaps served as foundations for buildings. There are two similar ones on the plateau at the western side, almost the same as we saw at Kidramos.

Two inscriptions only were found on the site, but a search through fields of full-grown barley is rather difficult, and others may be discovered. Neither of them gives the city name, but the identification is certain.

8. Engraved on a large block (length 97 cm., breadth 773, thickness 57); most of the inscription was underground :

“Ὁ ΔῊ Sroeatcit ὦ ΝΔΕΔΟΜΕΝΟΝΤΙῚ METI M E. ΛΗ T-NYTIOTOYMETI > YM TV OKPAT OPO MAIS APOS TiAl 5 | WiiOYAAPIANOY ZEIWAS ΤΟΥ ΑΙ Ke Mi YK TTIONAAMANKAT YAAE INO MWONASTAP XI-IN K TIS THNKAIE LEP LE a tal labile Gis ii G2 Ao.

“LINANAS TASIN TIOIH2AM:NI> 10 WiWf\EMOS1A FIX ΠΟΛΕΩΣΔΙΕ TIME ALTCY TITOYPAAOY | OYMAZIMOYAY ΣΙΟΥ TTPQTAPXONOZ ΤΙΣ TOAEQSTOA

ἡ βουλὴ κα]ὶ ὁ δῆμος [ἐτείμη- σαν τ]ὸν δεδομένον τῇ [πόλ- ει] ἐπιμελητὴν ὑπὸ τοῦ μεγίστου Αὐτοκράτορος K[a]icapos II. Αἰ[λ- ὅ ἤου ᾿Αδριανοῦ Σε[βα]στοῦ Μάρκον

ΟἸὔλπιον Δάμαν Κατυλλεῖνον, tlov ᾿Ασιάρχην, κτέστην καὶ εἰ ὑ- ε]ργέτην τῆς πόλεως" τὴν ἀνάστασιν ποιησαμένης

10 δημοσίᾳ τῆς πόλεως bu’ ἐπιμελητοῦ Τίτου Φλαουίου Μαξίμου Λυσίου πρωτάρχοντος τῆς πόλεως τὸ δεύ-

τερον.

A SUMMER IN PHRYGIA: 1. 403

A comparison of this inscription with the coinage of Trapezopolis furnishes the final and conclusive demonstration that Bolo is the site of that city. Mr. Head publishes a coin of the city (Catalogue of Caria, p. 177), struck in the imperial period but without the head of an emperor: it reads on obverse BOYAH TPATIEZOTIOAITQN, and on reverse AIA T DA MA AYCIOY. The Lysias of the inscription and the Lysias of the coin must be one and the same person. Lysias was raised to the civitas under the Flavian dynasty, while M. Ulpius Damas Catullinus probably received it from Trajan. Catullinus is appointed by the emperor ἐπιμελητὴς (curator) of the city. The question arises, what were his duties? It seems clear that ἐπιμελητὴς is merely an alternative for the common term λογιστής, the whole expression being a rendering of curator rei publicae Trapezopolitarum, an imperial official sent to look after the finances of the city. We have, then, at Trapezopolis a further example of the incompetency of the municipal governments. On this whole question see CB. ii. pp. 369 f., 376, etc. As the inscription belongs to the first half of the second century, Catullinus was probably not a citizen of Trapezopolis.? The title κτίστης implies, as usual, merely that he had obtained advantages for the city.

Lysias is first Archon for the second time. It is possible that the chief board of magistrates in Trapezopolis bore the name archons and not strategoi. Now on the coins of Trapezopolis archons are several times mentioned ; but no coins are as yet published that mention strategoi. ,

9.

ὁ δῆμος ἐτείμησεν Γάϊον "Αττί[εον

OAHMO® E TEIMH ZENIAIONAT TILT OY YIONKAAPON E TIAPX ONEYEPTETHN

ΚΑΙΣΩ TH PAKAI TAT Ps NATH 2 ΠΌΛΕΙΣ

Τίτου υἱὸν Κλᾶρον

ἔπαρχον, εὐεργέτην καὶ σωτῆρα καὶ πάτρ[ω- να τῆς πόλεως.

C. Attius T. 7. Clarus praef. belongs to a Roman family, and was doubt- less a Roman officer who had had the opportunity of rendering some service

1 Cod. Iust. i. 54, 3: Curator rei publicae qui Graeco vocabulo logisla nuncupatur, ᾿Ἐπιμελητὴς (τῆς wéAews) is the natural rendering of cwrator (ret publicae), and the adoption of λογιστὴς was probably due to the fact that ἐπιμελητὴς had almost become specialised in the sense of curator operis (as, for example, in 1. 10 of this inscr. )

2 He may have been a Trallian, brother of

Claudianus Damas (his Latin name being pro- bably M. Ulpius Damas Claudianus) who left a large bequest to Tralleis to found games in the reign of Antoninus Pius: see Pappaconstan- tinos Τράλλεις, Nos. 80 and 81. The two brothers in that case were probably sons of a Trallian named Damas. Antoninus Pius sent a native of Aizanoi as curator to Aphrodisias (C.1.@. 8884, 2741).

404 A SUMMER IN PHRYGIA: I.

to Trapezopolis. He was perhaps praefectus fabrwm in attendance on a high Roman official and possessing influence with him, or supposed by the Trapezopolitans to possess it.

THE RIVERS OF LAODICEIA.

The identification of the site of Trapezopolis leads us on to the vexed question of the Laodiceian rivers. The discovery of the city renders untenable Prof. Ramsay’s earlier assignation of the river names (CB. 1. p- 35 f.) and proves that he was right in recurring to the evidence of Pliny (‘ Corrections, p. 785f.). I shall try to show that his later suggestion is the correct view. The first essential in any scientific discussion of this question is evidently to know the course of the various streams or to have a correct map to show it; the opinions of a critic cannot be of much value if he is in ignorance about the rivers on which the whole controversy turns. Kiepert’s large-scale map of Westliches Klein Asien is absolutely untrustworthy here: it shows several rivers which do not exist, and makes the important river Tchukur Su (called also Geuk Bunar Su) flow in an impossible direction. M. Radet’s map goes even further astray: he retains Kiepert’s mistake about Geuk Bunar and commits the additional error of making Bash Bunar Tchai (which he names Geuk Bunar) flow direct into the Lycos—which suits a theory, but not facts! Prof. Ramsay’s map gives Geuk Bunar Su rightly, and though it does not show Bash Bunar Tchai, he has a correct knowledge of its course (p. 785, n. 1).

We must first indicate the course of the rivers which concern us here. (1) Gumush Tchai, which passes Laodiceia on the west side, is known by everyone, and it is therefore unnecessary to describe it. (2) Bash Bunar Tchai,” which flows by the ruins on the east side, has its source in a number of copious springs at Denizli. The water, however, is diverted for irrigation purposes, and very little of it is carried away by the stream. It is therefore a mere insignificant brook, with no claim to be called a river. In ancient times it was undoubtedly the same. We must suppose that there was always a settlement of some kind in the fine, well-watered plain of Denizli:? as Prof. Ramsay says of Eumeneia, ‘such a fine situation must have been occupied from time immemorial; the bountiful fountains would attract the peasantry of a primitive time’ (CB. ii. p. 354). After the foundation of Laodiceia, it was doubtless one of the villages in its territory. The Bash Bunar Tchai, then, was always the little brook it is today: and what water it brought down was almost certainly used up in the city. Be it noted that

1 He himself says ‘La carte hydrographique (Pl. XII.) the stream ought to be represented de ce district reste ἃ faire. Toutes les cartes by a much lighter line. existantes fourmillent d’erreurs,’ (Rev. des Univ. 3M. Radet justly remarks (Rev. Univ. Midi du Midi, 1896, p. 22, n. 2). p. 22) that ‘le site de Dénizly, l’un des plus * It is sometimes called Bashli Tchai, but frais, des plus enchanteurs qui soient dans la Bash Bunar Tchai is clearly right, ‘the stream _péninsule, n’a certainement jamais (i.e. before that flows from the Head Source,” In the map 514 B,c.) été inoccupé.’

A SUMMER IN PHRYGIA: T. 405

it does not fall into the Lycos, but slinks away to join the Geuk Bunar Su: at its junction it is hardly visible,a few shrubs being sufficient to conceal it. (3) Geuk Bunar Su, which has generally been identified with the R. Kadmos (probably incorrectly, see infra), is a fine, rapid river, quite as important and conspicuous as the Lycos. It rises in the hills on the south of Tchukur Ova, flows in a deep gorge between Mt. Kadmos (Khonas Dagh) and Mt. Salbakos (Baba Dagh), past Geuk Bunar [Kara Gidl], the copious fountains which form a duden (κατάβοθρον) on the left bank of the river, and passing Ak Khan (the old Seljuk building on its left bank) falls into the Lycos above its junction with Gumush Tchai. In its upper reaches it is called Tchukur Su, in its middle course Geuk Bunar Su, and after passing through Baghirsak Dere, Emir Sultan Tchai. (See Pl. XII.)

Now the Laodiceian rivers were the Lycos, Asopos and Kapros. There is a general agreement that the Asopos is to be identified with Gumush Tchai. Which is the Kapros? The Bash Bunar [Bashli] Tchai, say some critics, amongst whom is M. Radet (whose map represents the stream as flowing direct into the Lycos).1_ They point to the evidence of Pliny (v. 105, ce. 29), Inposita est (Laodiceia) Lyco flumini, latera adluentibus Asopo et Capro, and as the Bash Bunar Tchai flows by the line of fortifications, they regard the identification as obvious. But there is other evidence to deal with. Strabo (p. 578), speaking of Laodiceia,? says ἐνταῦθα δὲ καὶ ὁ Κάπρος καὶ ὁ Λύκος συμβάλλει τῷ Μαιάνδρῳ ποταμῷ, ποταμὸς εὐμεγέθης, ad’ οὗ καὶ ἡ πρὸς τῷ Λύκῳ Λαοδίκεια λέγεται. The correct meaning of this sentence has been suggested by Dr. Partsch (Berl. Phil. Woch., 1896, p. 466) in the words ‘ Auch fiir Strabo bilden Kapros und Lykos zusammen nur einen Fuss,’ and explained by Prof. Ramsay (p. 785-6) ‘Kapros and Lycos, two streams, join the Maeander in a common channel, forming together a large stream, which is called Lycos and on which Laodiceia is situated ’—a meaning which would be very well expressed by the term Λυκόκαπρος, which occurs frequently in the account of the miracle performed by St. Michael at Khonas.? This evidently forms an insuperable objection to the identification of Kapros with Bash Bunar Tchai. M. Radet ‘readily regards the expression καὶ ὁ Κάπρος καὶ ὁ Λύκος as the equivalent pure and simple of the term Λυκόκαπρος᾽ but his reason is that the Kapros and Lycos ‘mingle their waters before Laodiceia.’ The reason is untrue, and the objection therefore remains in its full force. (2) Moreover, how can the advocates of this view explain the coin (described CB. p. 35) representing, in the usual way, the chief rivers of the city, ΚΑΠΡΟΣ and AYKO=? Why is it that the Kapros is always named alongside of the Lycos as the other chief river of Laodiceia (e.g. by this coin, Strabo, Cinnamus and the term Avxoxampos)? The tiny Bash Bunar Tchai is the most insignificant of the streams, and it is inconceivable that it should have been selected for such special prominence (instead of the Asopos, for

1 Revue des Univ. du Midi, 1896, pp. 20-22 ; 3 Quoted by M. Radet, J.c.: see Bonnet, Narr. map in En Phrygie. de miraculoa Michaelo archangelo Chonis patrato

2 Strabo’s authority is especially high here, (Paris, 1890). It belongs to the eighth or ninth where he was probably an eye-witness. century ( Church in R, E, ο. xix,),

406 A SUMMER IN PHRYGIA: I.

example). (3) On this view, what stream can be found for the ELEINOs ? The χῶρος ᾿Ελεινοκαπρία was probably so named because its territory lay between the Eleinos and Kapros (or because these rivers flowed through it’). Cp. CB. 1. p. 86. Where can it be placed ?

I believe that Prof. Ramsay’s provisional suggestion (‘ Corrections’ pp. 785-6), which assigns the name to the Geuk Bunar Su, is the correct view. It explains every one of the difficulties just stated. Pliny’s expression (/atera adluentibus) is by no means inconsistent with this view. If we take it in the most literal sense, it may be interpreted in the way Prof. Ramsay suggests, viz. that the suburbs of the city extended to the Geuk Bunar Tchai or near it (p. 785 n.). There is nothing improbable in this. On the contrary, there is every probability that Laodiceia extended well out beyond the walls towards the east. Remains can be traced nearly up to the Geuk Bunar water: perhaps these are only relics of the tombs lining the great road to the east, but it is not impossible that they represent buildings.” Laodiceia was originally a small town and its fortifications enclose only a small space ; but it grew great and rich under the Roman rule (Strabo p. 578), and must have extended far beyond the walls. Excavations may yet reveal that the Geuk Bunar Su actually washed the edge of Laodiceia. At the same time it must be pointed out that there is no necessity to take latera adluentibus in an absolutely literal sense. It is a vivid expression used quite commonly in modern writers to mean merely that a river is close to a city: and this is what Cinnamus actually says, ἔστι δέ τις ἄγχιστα Λύκου καὶ Κάπρου τῶν Φρυγίων ποταμῶν κειμένη πόλις ὄνομα Λαοδίκη (1. 2, p. 5 ed. Bonn, quoted by M. Radet).

On this view it is easy to understand Strabo, the term Λυκόκαπρος, and the selection of the Kapros alongside of the Lycos as the two chief rivers of Laodiceia. Eleinokapria may be placed between Geuk Bunar Su and the stream which flows past Dere keui. The stone mentioning the Eleinokaprian canton, which now stands near the Kaive at Budjali, has probably not been carried. In Turkey, stones are ‘carried only when they are wanted for some purpose; but this stone stands by itself on open ground and may quite probably be in its original position. Prof. Ramsay, when encumbered by his original error about the Kapros, was forced to suppose that the stone had been carried: but he now welcomes the view that the stone is in its original position. In that case the stream at Dere keui is probably the Eleinos,3 and Budjali Kaive may very well be the exact representative of the ancient meeting-place of the χῶρος ’EXevvoxarpitov.* .

But is not Geuk Bunar the R. Kapmos? It has been so identified, but

1 As Waddington explains it (No. 1693a): wne des dewx rivieres qui traversaient le territoire de Laodicée 8᾽ appelait le Caprus, et le nom du village en est dérivé.

2 This is reported to me by W.M.R.

3 The stream called Kale Tchai is a mere torrent-bed, which never flows except during

heavy rains.

4 Waddington (on 1693a) identifies Eleino- kapria with Budjali, 1.6. the village (tchi/tlik), on the left bank of the Dere keui stream, in the corner between it and the Lycos. The Kaive stands close to the Railway, a very short dis- tance E. of the Station,

A SUMMER IN PHRYGIA: I. 407

without evidence. The Kadmos is probably the river that comes down from Khonas, joining the Lycos at Colossae, after irrigating the gardens and vine- yards below the village. Hamilton calls it ‘a copious stream,’ ‘a considerable stream’: he names it Bunar Bashi Su and says its source is a copious spring at the foot of the mountain near Khonas (i. pp. 509, 513). One branch comes down from the village, but the-water is mostly used up there: this part at least is called Dere Tchai. Lower down we found it flowing with a copious stream, and on enquiry it was explained that the water came from a bwnar. I understood that the spring was in, or close to, the channel of the river and unfortunately did not return to examine it. But Hamilton’s account is probably correct.

The Kadmos was identified with the Geuk Bunar, on the ground that ‘Strabo describes a duden in the former, and there is a duden in the latter’ (CB. p. 785; see A. H. Smith in J.H.S. 1887, p. 224-5). But Strabo’s words refer to the Lycos, not to the Kadmos!: ὑπέρκειται δὲ τῆς πόλεως ὄρος Κάδμος [Khonas Dagh], ἐξ οὗ καὶ ὁ Λύκος ῥεῖ καὶ ἄλλος ὁμώνυμος τῷ ὄρει.

3 w ξς 3 “(ὦ 3 a be

τὸ πλέον δ᾽ οὗτος ὑπὸ γῆς ῥυεὶς εἶτ᾽ ἀνακύψας συνέπεσεν εἰς ταὐτὸ τοῖς ἄλλοις ποταμοῖς, K.T.A. Οὗτος refers to the Lycos, which Mr. Smith admits to be a possible interpretation: the rule that ἐκεῖνος... οὗτος =ille...hic is not absolute even in the best writers. Moreover, Strabo’s description would not apply to the Geuk Bunar: the river does not disappear, the duden is a separate phenomenon on the left bank.?_ This is shown by the accompanying section. The water flows from the pond N in a shallow channel to form another pond M, which has no visible outlet.*

1 This was pointed out to me by Prof. Ramsay himself after I had been making a fruitless search for another duden, being unable to accept the identification of Kapros with Bash Bunar Tchai, I had not a copy of Strabo with me.

2 Hence it is not quite accurate to say that ‘the disappearance actually takes place at Kara Gél as well as on the Lycus’ (J.H.S. l.c.). I did not hear the water from the duden ‘ flowing

from the side of the deep gorge and falling down to the bed of the river’; I thought the noise was merely the roar of the river, but the fact may be as stated, for the water must issue somewhere.

3 For this section I am indebted to the kind- ness of Mr. 8S. Watkins of the Ottoman Railway.

408 A SUMMER IN PHRYGIA: 1.

Prof. Ramsay takes our view of the duden in CZ. p. 210, placing it on the course of the Lycos, not of the Kadmos; and it is apparently a slip that leads him on pp. 36 and 786 to accept the other opinion that there was a duden on the Kadmos.

I give here a few inscriptions of Laodiceia.!

10. Karak-Ova. November, 1896.

HAT ΠῚ

ΛΝ BO¥

Marble base, now used as ἃ dibek tash: copied in

ἡ ᾿Ατ[τ]αλὶς [φυ- λὴ τῶν βου-

wy MB TON λευτῶν' ΠΡΟΝΌΟΜΣΑΜΕΝΟΥΤΗΣ προνο[η]σαμένου τῆς ANAL re τ τοαρνι ἀναστάσεως Λογγεί- NOY vou τρίς, ἐκ να nee Hild ΣΦῪ τοῦ μῶν a τῆς φυ- 6, λῆ!ς.

We have here the name of a new tribe. Ina Seleucid city the name Attalis represents a foundation subsequent to the extension of Pergamenian influence in 190 Bc. In CB. p. 34 ἢ Prof. Ramsay inferred from the occurrence of Thracian and Epirote names, such as Seitalkas and Molossos, at Laodiceia that ‘a settlement of Thracian mercenaries had been made in the Lycos valley to counterbalance the colonists of Laodiceia, who were attached to the Seleucid kings’; but he wrongly supposed that the Per- gamenian settlement was made only at Tripolis, and that some of these settlers migrated afterwards to Laodiceia. It is now clear that the Attalidae did not restrict themselves to the planting of new cities over against the old Seleucid colonies (such as Tripolis, Dionysopolis, Eumeneia &c., CB. pp. 193, 199 f., 258, &c.), but actually introduced into the Seleucid foundations bodies of new citizens likely to be faithful to themselves.

The inscription shows that the constitution of the Boule was exactly of the ordinary Greek type. It is earlier than CB. No. 7 (= Ath. Mitth., 1891, p. 146), where the Boule is still organized on the Greek system. πρυτάνεις and ἐξετασταὶ are mentioned in the early inscription published in Ath. Mitth. 1895, p. 207 f., and also in Inser. in Brit. Mus. iii. No. 421 where, as Mr. Hicks points out, the Prytaneis change from time to time and enjoy σέτησις ἐν

πρυτανείῳ in the regular way; while the judicial system is also of the

1“In® CB. No, δ ΞΟ 8949), “the fragments of which I copied hurriedly, read Ait Μεγίστωι Σωτῆρι, and [Δομιτιαν] oo | (with Prof. Ramsay), which exactly fills the erasure: the next μεγίστῳ has no iota adscript. The inscription is engraved on architrave blocks below the triglyphs and above it there was a Latin inser. of which one fragment remains, DEDI CANT ESFEX (carved on the metopes) 1.6,

dedicante Sex[to...... pro consule]. The stones, which have suffered somewhat since they were copied, now lie at the south-east extremity of the ruins beside the Bash Bunar Tchai. It is possible, therefore, that they belonged to the Syrian Gateway, but they cannot have been part of the Ephesian gate, as Prof. Ramsay supposes (from the inaccurate accounts given about them).

A SUMMER IN PHRYGIA: 1. 409

ordinary Greek type. Increase of evidence tends to show that the constitution of Graeco-Phrygian cities like Laodiceia was closely after the usual Greek model.

11. A metrical epitaph, which I was told about early in the summer, but could get no chance of copying tll late autumn, gives the only example we have of the use of Doric at Laodiceia. We should expect to find a mixture of Doric in cities where Macedonian colonists were settled: but it is unsafe to draw any such conclusion from an epigram of this kind. When this paper was almost finished, the third part of Ath. Mitth. (for 1897) appeared, and I find myself forestalled in the publication of this epigram (see p. 358, No. 8) and some other inscriptions, e.g. Nos. 1, 4 and 6 (which I copied in the autumn of 1896). I therefore omit these from this paper, giving only some necessary corrections. The ecpigram is rightly restored, except the last two lines. The epigraphic text of the last line is

OYAAXIAAEYEAEQS TF!!'7919////ALOETIAOS,

and I restore both

"Addo μὲν ᾿Επέγονος μνᾶμα ζωιοῖς διασώζει, οὐδ᾽ ᾿Αχιλλεὺς δ᾽ ἔφυγεν μοῖραν ἁ]χλὶ Θέτιδος.

The type used in Ath, Mitth. gives no idea of the lettering, which is not good.

In No. 4 1. 2 I read SQSIHEAYTALI, tc. Τάρσος, ζῇ, ἑαυτῶι &e., (where M. Weber reads an unintelligible X1M); inl. 4 KAQAIQI as required, and in the last line IQSIN.

12. No. 6 adds something to our knowledge of the Laodiceian festivals but we cannot be quite sure of the restorations. M. Weber has not noticed a fragment lying beside the block and fitting to line 8 (which indecd increases the difficulty of restoring line 9). The fragment reads

8 | TH //// TiK1//// 10 | MAYP////

There is only a slight space between | and AZAN, hardly room for more than one letter. I estimated that after line 6 there was a possible space for about 8 letters.

We may attempt a provisional restoration thus :—

yl Deep tier se -rov Παπε[ιριανόν, πατέρα καὶ θεῖον κα[ὶ ἀδελφὸν συγ-

1 Inl. 6(end)Iread ΟΣ TC///, 1. 8 NIE, appeared to be nothing inscribed below the

1. 10 PODS, 1. 11 PEN//MATI///;. there middle of this last word.

410 A SUMMER IN PHRYGIA I.

κλητικοῦ' Kai [ἀρχιερέα 5 καὶ ἀγωνοθέτην [τῆς τρίτης ? πενταετηρίδος τ[ῶν μεγά- lov ἀγώνων Δίων κ[αὶ ᾿Ισολυμ- (or τῶν Ὀλυμ-) πίων ἱερῶν εἰσελα[στικῶν π͵]ᾶσαν τὴν οἰκουμένην 10 Μ. ‘Aip? θ]ρέμμα ἈΥΜΙ͂Νν ἀνέθ. ηκεν 1]

L. 6. The Δεῖα Σεβαστὰ οἰκουμενικὰ is already known: we now see that it was a four-yearly festival. Another Laodiceian festival was the ᾿Αντιόχεια founded by Antiochus II. (261-246 B.c.) and celebrated every year (Znscr. Brit. Mus. iii. 421).2 Lines 7-9 seem to give an additional one called [Ολύμ]πια (or Ἴσολ-) ἱερὰ εἰσελαστικὰ οἰκουμενικὰ: in 1. 9 we want some- thing like the εἰσελ. εἰς τὴν οἰκουμένην of C.I.G. 2932. Games called Ὀλύμπια Ἡράκλεια ᾿Αδράστηα are found at Attouda. On εἰσελαστικὰ see Hicks, Hphesos No. 607 and C.L.G. iii. 5804, where Franz says such games were called vulgo ἀγῶνες ἱεροί πενταετηρικοί. On οἰκουμενικὰ

Hicks l.c. No. 505.

13. I add an improved copy of an interesting fragment imperfectly given by Dr. Judeich in Ath. Mitth., 1890, p. 258, leaving it for the present without any attempt at restoration. It clearly refers to a vain contention περὶ πρωτίων between those ‘ brought up in the new faith’ and those of the old (ἡ ματ[α]ία φιλονικεί[α7), and gives an admonition (the nature of which is not clear) to put an end to it. The inser. is repeated in CB. no. 410.

arm ME O@Y ME MMII (ACTINECEICINKAINH AO ZHTPE Wil! \T OY TOYE COL XP HNEPITT PW TTIW N Tae GAIKATAAYET WEAN THNAM@ICBHTH VON TOL HMAGAMIRDINON ΠΕ AMIN μὰ} Wire WON, EE NV SY HEC AY MH δ. αι i! IN Oy ol Oy, GICAIn POKATAPXETUDLAN δι IMJOPMMMENOYLLCEMNOTEPOYLTAPEAY TY IINOYCINTOYL MPOLAZIAN TIMHCEIMHZ W/E [PAINOIN TO vac

ARE aR We Sipe 6G 2 AN oh Bs

11 had restored μεγάλ]ων, and Ath. Mitth. 1895, p. 207, should be restored [ἐν τοῖς συν- shows an A where my copy has a vacant space. τελουμένοις κατ᾽ ἐνιαυτὸν ἀγῶσιν ἐν [τοῖς ᾿Αντιο * L. 16 of inscription published Ath. Mitth. χείοις. .1, by comparison with Br. Af. 421.

A SUMMER IN PHRYGIA: I. 411

HIERAPOLIS AND ITS VILLAGES.

14. The following important psephisma was found near Tchindere keui, a village high up on the plateau above Hierapolis, overhanging the Maeander eafion. It was said by one of the natives (who gave it as a tradition) to have come from a site SSW. from Geuzlar (Thiounta), where we saw some remains. The tradition is probably trustworthy.!. The slab is unfortunately broken at the top and two sides: it is much worn, and could not be read except by placing the stone athwart the sunlight. The restorations are intended to indicate what appears to be the general sense.

AOY TOY ὃ ΝΕΩΤΈΡΟΥΜΙΕΙΟΣ AEKAT

( IEPATIONE | TO.N THBOYAH ETT! ΤΩΝΑΡΧΑΡΈΣΙΩΙ,

HWE NINE NET ANENTAIZKOMAIS ΠΟΙΕΙΣΘΛΙΕΠΊΔΙΡ

5 (ONAYTOIZTIAP EXE INIMONON 2 YAAKAIAXY PAKAIMON

ΛΈΝΧΘΕΝΤΑΤΙ ἘΠΟΙΗΚΕΝΑΙ ΠΡΟΣΤΕΙΜΟΥΟΝΟΜΆΤΙΕΗΙΣΑΙ ENAIAY ΤΟΝΟΣΑΑΝΕΛΈΝΧΘΗΕΙΛΗ ΤΦΩ 2 ΠΑΡΑΤΙΝΟΣΑΤΙ WE ΛΈΝΧΘΕΝΤΑΣ TIAPAPYARKA> MHAAMBANEINTAD ΠΆΡΑΤ 10) RETR PSM apy \9.\KON TAD. Σ TEPANOY NTIAPA\ @APFYPI ONHTIZ AANKOMHROYAHOH 2 TEPANN ZAI ΠΑΡᾺ “TAMH@ENT EINE ZOAIEIAEM TONY TIENANT IQS ITOH? ἸΟΛΛΩΝΟΣΑΜΑΘΗΜΑΤΆΟΝΤΟΣ TOY TOY TOYYH9I MAT ETTEXONTOD

~~

(

λοι ον ᾿Επὶ στρατηγοῦ Θεοφ 1]έλου τοῦ β΄ νεωτέρου, μ[η]νὸς δεκάτου... ἔδοξε τῶν Π]εραπολειτῶν τῇ βουλῇ ἐπὶ τῶν ἀρχα[ι]ρεσιῶν[ τοὺς παρα- φύλακας κατὰ vou ἴ]ον ἀπ᾽ ἑαυτῶν ἐν ταῖς κώμαις ποιεῖσθαι ἐπιδημίαν ὡς μὴ 5 προσῆκ]ον αὐτοῖς παρέχειν ἢ μόνον ξύλα καὶ ἄχυρα καὶ μον[ήν, ἄλλο δὲ μηδὲν μηδενὶ ἄλλῳ ᾧ ἄν ποτε τρόπῳ' ἐὰν δέ τις παρὰ ταῦτα ποιήσῃ ἤ ἐϊπι- χειρήσῃ πλέον λαβεῖν, τὸν ἐϊλενχθέντα πεποιηκέναι προστείμου ὀνόματι εἰσ[άγειν... καὶ κατατιθ]έναι αὐτὸν ὅσα ἂν ἐλενχθῇ εἰληπφὼς παρά τινος ἀτίμως, πρὸς δὲ τού- 1 Of the value of such statements, which are zwverldssig, a striking proof will be given under often dismissed (c.g. by Von Diest) as stets wn- MEROS,

412 A SUMMER IN PHRYGIA: I.

τοις τοὺς] ἐλενχθέντας παραφύλακας μὴ λαμβάνειν τὰς παρὰ Th[s κώμης τιμάς: ἐὰν δὲ

10 Eup Bay TAS. nominee fe ἢ κωμάρχ[ας] ἄκοντας στεφανοῦν παραφ[ύλακα, ἀποδοῦναι αὐτὸν τὸ] ἀργύριον᾽ ἥ τις δ᾽ ἂν κώμη βουληθῇ στεφανῶσαι παρα[φύλακα, δεῖν \ A ae ' ee ERK , Ν e , ΄ πρὸς ταῦ]τα μηθὲν 1 τείνεσθαι" εἰ δὲ μή, τὸν ὑπεναντίως ποιήσαντα ἕξειν

τὰ ᾿Απ]όλλωνος ἀναθήματα, ὄντος τούτου τοῦ ψηφίσματος κυρίου καὶ] ἐπέχοντος.

Notes.—L. 2. There seems to be a gap between AOY and TOY, though the copy exaggerates it slightly : νεωτέρου is equivalent to τοῦ A’, ‘ Theophilos, son of Theophilos, i.e. Theophilos the younger” L. 4-5. ‘The para- phylakes must live at their own expense in the villages, since the villagers are not required to supply them with anything but merely wood etc.’ L. 10. The letters before HK QM are far from certain: the impression seems to show something like what is given.

For ἐπὶ τῶν ἀρχαιρεσιῶν in 3 cp. ἐν ταῖς ἀρχαιρεσίαις (CLG. 2693 c and d); and expressions like ἀρχαιρεσιακῆς ἐκκλησίας (J..S. 1895, p. 118). The title ἡγεμόνες which occurs at Melokome (CB, no. 64 = J.H.S. 1887, p. 399) is apparently not equivalent to κωμάρχαι.

This inscription (which is one of the copies of the decree set up publicly in all the κῶμαι) throws some light on the relation of Hierapolis to its subject villages, a point which is discussed in CB. pp. 123-5. It is there argued from the failure of any allusions to self-government among the inscriptions of Mossyna and Thiounta,’ the two ancient κῶμαι near Geuzlar on the plateau behind Hierapolis, and from the fact that Hierapolis and Dionysopolis were probably conterminous, that these two villages, and doubt- less others (see p. 141), were subject to Hierapolis. The author proceeds to indicate the probable relations of the metropolis to the subject κῶμαι and remarks (p. 125) that a παραφύλαξ mentioned at Thiounta (inscr. 31) was most likely an official whose authority emanated from Hierapolis, rather than a mere Thiountene officer. Our inscription shows that Hierapolis appointed a set of παραφύλακες for its villages.2 These officials were undoubtedly ‘heads of police’ charged with the maintenance of order in the territory of the ruling city. The police were styled παραφυλακῖται or φυλακῖται (the two names being probably equivalent), and they were perhaps a Pergamenian institution (CB. p. 258, Frankel, Inschr. Perg. 249). About the constitution of these police forces our knowledge is scanty: but they were employed in

1 Cp. especially No. 29. each xwun. These παραφύλακες are also men- 2 The expression in CB. p. 125, might suggest _ tioned in the mutilated decree found at Develer that a single paraphylax was appointed for the and published by Hogarth in J.H.S, 1887, whole Hierapolitan territory: that, however, p. 892 (no. 21). was not the case: there was probably one for

A SUMMER IN PHRYGTIA: I. 413

hunting down and keeping in custody brigands, Christians, and other disturbers of the peace.

Officials of the subject villages must be carefully distinguished from officials of the ruling city or metropolis (here Hierapolis): this distinction is always clearly brought out in the Egyptian documents, which are our best authority for the relation of a metropolis to its κῶμαι. It is natural that the charge of order in the territory as a whole should be vested in officials appointed by the metropolis: and it is satisfactory to find documentary confirmation of Prof. Ramsay’s conjecture to that effect, a conjecture founded mainly on the consideration of natural suitability.

The Paraphylakes were in a position of power, and could make illegal requisitions upon the villagers or extort honours from them against their will (Il. 5, 10). This decree enacts pains and penalties with a view to the prevention of such abuses: the Paraphylakes are required to live at their own expense, and the articles they are authorized to demand from the villagers are strictly defined.

On the whole subject, see Prof. O. Hirschfeld in Berlin. Akad. Sitzungsber. 1891, pp. 845—877, 1892, pp. 815—824, 1893, 421—441; Frankel, Jnschr. Perg. No. 249: CB. p. 68, 258 ff., 307 f. (where bodies of παραφυλακῖται are stationed in villages).

CHRYSORHOAS

In speaking of the Lycos valley, Strabo refers to τὸ πολύτρητον τῆς χώρας καὶ τὸ εὔσειστον: An interesting phenomenon, not mentioned by any traveller, is related by Prof. Ramsay about the stream Chrysorhoas, the most important of the rock-forming cascades which flow over the cliffs at Hierapolis. He says (p. 86, . 2),‘My friend Mr. Walker told me that its waters, after tumbling over the cliffs, flow for a short distance south through the plain until they reach a hole in the ground into which they disappear, etc. An investigation of this point showed that the statement was quite true, but a few years ago the phenomenon disappeared owing (as the natives also said) to the gradual choking up of the underground passage by incrusta- tion. The hole where it vanished can be seen and easily identified from the deposit formed at the sides. I was assured, however, that after its under- ground course it reappeared down in the plain near the village Kutchuk Shamli where it formed a marsh, at least in winter, when the water was not used to irrigate the fields. Now the stream flows above ground and is carried down in the same direction.

1 Earthquakes still occur in the valley: a rather violent one took place during one of my visits to Laodiceia.

H.S.—VOL, XVII. ae

414 A SUMMER IN PHRYGIA: I

ANAVA—SANAOS.

“Avava, the town passed by Xerxes on his way from Kelainai to Colossai beside the salt lake of the same name (Herod. vii. 30), has been identified by Prof. Ramsay with the later Sanaos1 (through the stricter form Sanavos) and placed at the village Sari-Kavak on the edge of the northern hills overlooking Adji Tuz Gidl, ‘ Bitter Salt Lake ’ (Amer. Jour. Arch., 1888, p. 275, CB. p. 230, 218). M. Radet, disagreeing with this view, separates the two names, and leaving Anava at Sarikavak places Sanaos at Tchardak, on the western end of Lake Anava. The following inscription proves the correctness of the former view. It is engraved on a sarcophagus cut out of the rock in a necropolis between the middle and western quarters (mahalla) of the village: the stone is much weather-worn, and the first part is quite illegible.

15.

TY NBIOY MT OKA ΜΡ ΠΡ ΗΠ 7} ᾿ΗΤΠΉ Η Ή:;ΓΤΟΛΜΉ TAS ATIOA@ FEIMOIEPSTAT | TAMI@ATIK AZ XBOKT@ TANAHN@NAHM QI 2 ΤΕΙΜΑΣ TOY ZE 3: Mili CWAUN ATIKAZ’B$ Z @NTEZAEE TEAZZANEAYTOIZ TON @NIONOIKOM νας

-“ A A \ [ot δεῖνες κατεσκεύασαν TO μνημεῖον ἑαυτοῖς καὶ

ἑτέρῳ δὲ obder! ἐξέσται τεθῆναι. χωρὶς τῆς] - συνβίου al τοί) 0) καὶ [τῶν τέκνων (3) ἐπεὶ ὁ] τολμήσας ἀποδώσει [τ]ῷ ἱερωτάτῳ ταμείῳ Atixas% ,Bd' κὲ τῷ Σαναηνῶν δήμῳ ἐ ἐς τειμὰς τοῦ Σεβ. ᾿Ατικὰς ,(βφ΄. ζῶντες δὲ ἐτέλέσαν ἑαυτοῖς τὸν ἐώνιον οἶκοῖν.

In this and the following inscription the fine is to be paid in Attic drachmae? as at Apameia (CB. ii. No. 321, quoting also Thyatira); this suggests a connection with Apameia, and is a further indication that Sanaos was subject to that city (see CB. p. 280, ii. p. 428, etc.).

On αἰώνιος οἶκος, see a paper in the forthcoming Annual of the British School at Athens for 1897.

16. Ona similar sarcophagus near the former.

τὸ ee τ id Oe GS Per (OT NEAY M&K AI ΤΗΓΥΝΒΙ

᾿ς DIQIEINOYAENI AEE ZEMTAIETEPONTINA

ponies, IETIEIOTOIONOR ITOAPHC AC ATTOAME EITTPOCT EIMOYEILTOIEP@TATON TAME! ΠΤ ἘΞ ACTTENTARICAIAIAC <3

a eee eee

1 Σαναός Strabo p. 576, Xdvaos Hierocles, Zavls reading due to the notches in the stone for ‘BS, Ptolem. v. 2, 26, Zuvads or Xwads Notitiae, as in next line, but probably it is used to denote ? ᾿Ατικὰς )ξ seems curious : it is possibly a mis- δραχμάς.

A SUMMER IN PHRYGIA: T. 415

[ὁ δεῖνα κατεσκεύασε τὸ 7 Clee ζῶν éav[td] καὶ τῇ συνβίϊῳ αὐτοῦ οὐδενὶ δὲ ἐξέσται ἕ ἕτερόν τίνα τὐδεθοΊαϊ, ὁ ἐπεὶ ὁ dase DG τι vin wifes ἀποδ[ώἼ]σει προστείμου els τὸ ἱερώ- τατον ταμεῖ- ον ᾿ΑἸτικὰς πεντακισχιλίας.

Some other inscriptions of Sanaos will be given in Part II.

BRIA.

The city Bria was placed by Prof. Radet and Prof. Ramsay, independently of each other, at the modern village of Burgas.1. M. Radet judges from the order of Hierocles, the importance of the modern village, and from its name which ‘semble n’étre qu’une adaptation Turque de l’ancien.’ Prof. Ramsay goes further and shows in an admirable commentary on inscription 218 (see also p. 577) the etymological affinities of the word Bria and how the name Burgas arose from the form Berga: but he is careful again to add the qualifi- cation that though the name remains, the exact site may not be at the modern village; for no remains have been found there, and it is the exception, not the rule, to find modern villages exactly on ancient sites. Towards the end of the summer I passed through this district and found the old site about a mile and three-quarters north-west of Burgas on the left of the road to Tatar keui. It is situated in the open plain in the midst of acorn-producing (palamit) trees and is hardly visible, and certainly not noticeable, from the road. Such a defenceless situation was of the Pergamenian type (Hist. Geogr. Ῥ. 86), which looked to commercial rather than to military considerations. With this accords the Thracian name Bria, for it is known that Thracian colonists were often settled in Pergamenian foundations. We may therefore safely infer that the city was founded after 190 8.6.

There is very little to be seen now on the site. The most conspicuous part of the ruins is what we may best describe by saying that it looks like an extensive square-shaped entrenchment, banked right round, the general surface being raised above the ground level to the height of several feet. About two yards or so from the outer edge a low narrow ridge runs round, evidently concealing the foundations of a wall, the blocks of which appear here and there in situ. This then was the fortified part of the city: and the natives have appropriately given it the name hendek, ice. ‘dyke’ or ‘trench.’ The buildings, however, extended over a large extent of ground especially towards the south-west. Here several big rectangular blocks may still be seen on the surface and the villagers of Tatar keui have recently laid bare some foundations formed of fine blocks with some cemented work: at this spot were found the stones bearing the inscriptions given

1 Radet, En Phrygie, p. 112; Ramsay, CB. i. p. 248-4. -FF2

416 A SUMMER IN PHRYGIA: I. .

below. At the inner base of the narrow ridge (where the fortifications seem to have been) they dug up a large Byzantine column with a late inscription on it together with some other inscriptions which they broke into fragments to put into the foundations of their new mosque.}

The question arises, what has become of all the surface stones? Burgas seems to possess none, though some are almost certainly concealed in the walls of the mosques,? and Tatar keui is built of mud. They must have been carried to greater distances (perhaps to Sivasli and neighbouring villages).

The following are the only inscriptions which we can certainly assign to Bria.

17 In the unfinished mosque at Tatar keui:

AILKAIZEBA Διὶ καὶ Σεβασ-

TALKAIZ API τῷ Καίσαρι EY =ENOZAZKAH Εὔξενος ᾿Ασκλη- HbA ΔΟΥΧΘΙΕ ΕΣ: πιάδου ὁ ἱερεύς.

Euxenos was priest in the Imperial cultus, the worship of the Emperor being associated, in the usual way, with that of the native deity Zeus.

18. iid, In two fragments: in the epigraphic text they are placed together

‘O βωμὸς]καὶ ἡ Kat’ av- τοῦ σορὸ]ς M. Αὐρ. © Διογέν ?]ous ᾿Αθη-

νοδώρου ? Bou Ξ]ανοῦ,3 εὐ- δοξοτάτου β]ουλευτοῦ φυλῆς β' 1 π]άσας ἀρχὰς καὶ λειτουρ]γίας τῇ πα- τρίδι τελέσ]αντος ἐκ προ-

γόνων, στε]φανηφόρου,

1 It seems probable however that No. 19 declarations of the natives, who were kindly came from this spot. disposed) three previous expeditions found 2 Continuous heavy rain prevented a proper _ nothing. examination of the village, but (apart from the 3 Perhaps a name like ’A@nvodwpiavod.

A SUMMER IN PHRYGIA;: I. 417

In 1, 1 the expression κατ᾽ αὐτοῦ is unusual! but we may compare an inscription of Laodiceia (Ath. Mitth., 1895, p. 209) where we have πλάτοι δύο, εἷς ἔνγαιος καὶ κατ᾽ αὐτοῦ ὑπέργειος. In 1. 8, the expression ἐκ προγόνων (like ἱερεὺς διὰ γένους, etc.) means merely that members of his family had often undertaken these offices: the ἀρχαὶ were by this time as much burdens (munera) as the λειτουργίαι. On στεφανηφόρος see CB. i. p. 56 ἢ

The date is shown by the name M. Aur. to be the latter half of the second, or the first half of the third, century of our era.

19, Jbid. A late Byzantine inscription on a rectangular block: the spelling is atrocious, but it shows the modern pronunciation.

tELOHOGNAH( YKOAOMH C46 TONG OM K€AH AOTON KE ¥ FOXOPAdH N TOA ΚΙ, ANAAYCE TON AOPOMOY HN T Milil. THNEKAHCHANH( TH NHMEP A Nii HNQEXHTHNAH KHNM E TQT fill,

APOMON Ἔ +éyo ᾿Ηοάνης (Ἰωάννης) ὑκοδομήσας τὸν (ν)αόν Ke Δηδοτον DATA TOUA LYSTHIATHA ἀνάλυσε τὸν λόγο(ν) μου Ava (1.¢. ἵνα) τ τὴν ἐκλησήαν ἧς τὴν ἡμέραν. . .. ἥνα ἔχῃ τὴν δήκην μετὰ τ[ὸν ἐπιχθόνιον δρόμον

Owing to limitations of space and of time, I must conclude this paper by giving the more important of the results attained in the north-west of

Phrygia.

THE IMPERIAL ESTATE OF TEMBRION.

20. At the village Yapuldjan, close to Altyn Tash and the site of Soa, I copied the following inscription, which gives important evidence, both topo- graphical and historical, regarding the large Imperial Estate on the Tembrogios (Tembris).

2 An exact parallel occurs at Hierapolis (J. H.S. 1885, p. 346, No 75).

418 A SUMMER IN PHRYGIA: 1.

ATAQH

1M?“ CARL MH: ὌΠ LULL OU SR il (IMU (Cre (WAU PEADIDYMUMNXY IU C ENERUAY PROCOM/ULE US PEBSRECTAFIDEEORUM QUAEY QUIDINIURIOfECERATURAD (O({ICITUDINEMSUAMREUO CA AIT ΧΑ

win Sis

---->

AYTOKPATOPIK-CAPIM lOYAWN OIAITITINEYCEBEIEYTYX EICEBK 5 |PIAITITINETIIPANECTATWKCAPIAEHLI CTIA PAAYPHAIOYELKAEY NOYTWNNAPATOYHNWNITAPOIKWNKITEWPLWNTWNY MET EPA Sheu ea eee NKATAPYTIANTOTTINNAIATOY CTPATIHTOY* TIANTWNENToICMAKAPINTATOICYMWN KAIPOILEYE EBEL TIOTATOIT WNITWTTOTE BACIAEWNHPEMON KAI FAAHNONTON BION AIA T# 10 |NFPIACKAIACEIC MWNTTE AY MENKNMONO] EME | LAAAGT PIATINE 1% KAIPWNIACXONTECTINAE FNIK ET Egy ENT POLATOMENE XE HEWCENTOY ΤΟΙ ψ' *XWRONYMETEP ONZCMENIEPNTA WM MOLOAOK AHOLOIKATA$ EYPONTZCKTEINOMENOITHY METEPAC C EIOME@AAETIAPATOA ACTON KTIAPATI PALCOME @AYTIEKEIN WNG 15 -| CIONOPAE WME EIOIPAPTY i XANONTELKMITETTAPACT PATE ! BOMENAAAGTPIATNYMTEPHNMAKAPINTATWN KAI PANN Y TOATITT ANWNKAIMATTAPAAIMTTANONTELTACAEW$ OPoYLy MINTAIK AYNACTAITWNITPOYXONTINK = ATHNTOA|N MET EPo! ETI El CEB<OMENOIK KATAAIMTITANONTELTACAE: 20 |EPTHNMACA$ICTANTELK TOY CAPOT PAL BROAN AOMENAAYTOICTIAPATPACCOYCINKCYNAINEIO* OY TOYAAIKEIC@AIA] ACEIOMENOYE TTE PI W NATIA CEBACTEM PE OOCOMOTETHNETIAPXON AIE I TT EY NOLKOTIHCTIEPIT OY THNEKEINd ΘΗ LOYH 97) ENTETADMENA QUAE(IR MOCONPLEXIESTIY QUIDAAITOP ERAMNEDMATIUIS QUERELLYy ETT EIAHOYNOYAE NO} EAOMMINERTAYTH ΓΤ 14 BHKENAEMMACKATATHNALPOIKIANTAMIO$E 1/7 TTENMSAINOIZTENT INNKCY NTTATOY NT ANIMAL I 30 “YTIOT AN KCAPIANNOYTATY XONTA AWMEIEL KECOAIKTAX HW PIAEPHMOYCAIKWANS CK OY TIAPAWIM ML. ONKKTOIKOYN ἢ AY NAMENAYIMIIT AY T VE

OMAR MY YEO ἡ

ς᾽ “Ὁ

᾿Αγαθῇ Τύχῃ Imp. Caes. M. [Iul. PJhilfippus Pius Felix Aug.] et [M. Iul. Philippu]s nfo}bi{ljissimus Caes. M. Au[r....... Ap- peae ? Didymum M generum pro consule . . , perspecta fide eorum quae [seribit Eglectus .. . quia iniuriose geratur, ad sollicitudinem suam revoca(n)t.

on

2

10

15

20

25

A SUMMER IN PHRYGIA: T. 419

Αὐτοκράτορι Κέσαρι M. ᾿Ιουλίῳ Φιλίππῳ Εὐσεβεῖ Εὐτυχεῖ Σεβ(αστῷ) κ[ὲ Μ. ᾿Ιουλέῳ

Φιλίππῳ ἐπιφανεστάτῳ Κέσαρι δέησις παρὰ Αὐρηλίου ᾿Ἐγκλέκ- [Tov .... ta- n lal ’ fal , \ al A ξ΄ / / νοῦ τῶν Apayounvay παροίκων κὲ γεωργῶν τῶν ὑμετέρων, [δημοσίᾳ δαπ- ἄνῃ δήμου κοινο(ῦ) Μο(ξ)εανῶν Σοηνῶν τῶν κατὰ Φρυγίαν τόπων, διὰ ΠΟ Fa, ΣΝ ,, ΄ > - 4 ig a “ ᾽ / στρατιώτου. Ἰ]άντων ἐν τοῖς μακαριωτάτοις ὑμῶν καιροῖς, εὐσεβέσ[τατοι κὲ ἀλυ- ’ lal / / »Μ \ \ ‘ , πότατοι τῶν πώποτε βασιλέων, ἤρεμον καὶ γαληνὸν τὸν βίον διαγ[ο- μένων, πο- , \ “Ὁ / / € “ , / a J vnpias κὲ διασεισμῶν πε[π]αυμένων, μόνοι ἡμεῖς ἀλλότρια τ[ῶ]ν ε[ὑ- τυχεστάτων καιρῶν πάσχοντες τήνδε τὴν ἱκετείαν ὑμεῖν προσάγομεν, ἐχέϊγγυοι ἐνθυμ-

ἤσεως ἐν τούτοις: χωρίον ὑμέτερόν [ἐΐσμεν ἱερώτατον κέ, ὅταν ἢ σεισ-

μὸς ὁλόκληρος, οἱ καταφεύγοντες κὲ γεινόμενοι τῆς ὑμετέρας [προστα- σίας" δια-

σειόμεθα δὲ παρὰ τὸ ἄλογον κὲ παραπρασσόμεθα ὑπ᾽ ἐκείνων ο[ὺς μὴ ἀδικεῖν τὸν πλη-

σίον ὀφίλει' μεσόγειοι γὰρ τυνχάνοντες κὲ μ[ή]τε παρὰ yan) OL μηδενί, πάσ-

χομεν ἀλλότρια τῶν ὑμετέρων μακαριωτάτων καιρῶν [ἐπεὶ οἱ ἐνοι- κοῦντες '

τὸ ᾿Αππιανῶν κλίμα παραλιμπάνοντες τὰς λεωφόρους ὁδοὺς γίνονται στρα-

τ]ιῶται κὲ δυνάσται τῶν προὐχόντων κ[ατ]ὰ τὴν πόλιν [ἡμῶν, γείτονες δὲ ἡ-

μέτεροι ἐπεισε[ρ]χόμενοι KE καταλιμπάνοντες τὰς λε[ωφόρους.. κὲ τῶν

ἔργων ἡμᾶς ἀφίσταντες κὲ τοὺς ἀροτῆρας βοᾶν [πειρώμενοιϊ τὰ μὴ ὀφει-

λόμενα αὐτοῖς παραπράσσουσιν κὲ συνβαίνει οὕτως ἡμᾶς ἐκ τοῦ τοι-

ovtov ἀδικεῖσθαι δι(α)σειομένους περὶ ὧν ἀπ’ ἀρχῆς πρόσταξιν ἐποιήσω, :

Σεβαστὲ Μέγεθος, ὁπότε τὴν ἔπαρχον διεῖπε[ς ἐξουσίαν ΤΟΝ

vos κὲ ὅπως περὶ τούτων ἐκεινήθη σοῦ ἡ θεία κέλευσις ἐν ταῖς Sénrots

ἐντεταγμένη: Quae li[b]e[r]o (or li[b]e[ll]o ?) conplexi esti[s . . .

-quid, agit operam ne d{iuJti(n)is querell[is ...... ᾿Ἐπειδὴ οὖν οὐδὲν ὄφελος ἡμεῖν ἐκ ταύτης τῆς ταραχῆς γείνεται, συνβέ- βηκεν δὲ ἡμᾶς κατὰ τὴν ἀγροικίαν τὰ μὴ ὀφει[λόμενα παραπράσσεσ- θαι, ἐ-

1 Better (Τ)οτεανῶν, see infra.

420 A SUMMER IN PHRYGIA: I.

πενβαινόϊν]των τινῶν Ke συνπατούντων ἡμᾶς [παρὰ τὸ δίκαιον (or ἄλογον), ἐπειδὴ 30 07 ὑπὸ τῶν Κεσαριανῶν οὐ τὰ τυχόντα δι[ασ]είεσ[θαι δεῖ οὐδὲ πάσ- χεσθαι Wego: ἐμ -κεσθαι κὲ τὰ χώρια ἐρημοῦσί(θ)αι KE... ye he feast! ς κὲ ov παρὰ τ[ῶν ἔν]δον κατοικούντων...

Notes.—The centre of the stone (as I estimated it) is shown by the dotted line at top and foot. This makes a possible space for eleven or twelve letters after 1.1. L. 2, the second letter may be F; before GENERUM the stone seemed to show two letters (possibly L1), not U (but perhaps merely bad engraving for U). L. 5, mapa seems to denote that Enclektos drew up the document for the commune: perhaps he was the headman (magister, προάγων, κωμάρχης in J.H.S., 1887 p. 498). If so, διὰ would denote the person who presented the appeal. L.7, Μοτεανῶν may possibly be a variant form, but the T seemed different from the other letters T, and it is perhaps a miscut (like ὀφ'λει in 15, AIC. in 22, and probably others). 1.10 /in., read probably TWN asin 16. Τ, 15,can otpard[pyns] mean ‘a governor with a force at his dis- posal’ (like στρατηγὸς ὕπατος for procos.)? L. 18 init., we want a word for ‘marauders’: perhaps στρατιῶται may bear such a meaning (‘they have become foragers’: cp. στρατιωτικῶς ‘ brutally’). L. 23, διεῖπες from διέπω ‘directed the proconsulship of the province’ or simply ἔπαρχον διεῖπες ‘ were arranging the affairs of the prov. L. 24, émws=‘when. L 26, the stone has TIUIS, which is clearly another miscut for TINIS or TIUS. The inscrip- tion was carelessly engraved.

Later consideration makes it seem more probable that in AAOTEANWN (1. 7) it is the A, not the T, that is miscut. The lapidary cut KOINOMA for KOINOYT. We thus get the form Τοτεανῶν, 1.6. the people of Tottoia,! the ancient name of the village Besh Karish Eyuk, about five miles S. of Altyn Tash (see J.H.S. 1887 p. 513). This is exactly what is wanted. The suggestion is due to Prof. Ramsay.

The date is 244-246 A.D.: in the latter year the younger Philip assumed the title Augustus.

The sense of the Latin heading is not clear. Perhaps the reply was sent through Didymus M—gener, the proconsul, who would forward it to the procurator. LI. 2-3, ‘having examined the truth of what Enclektos writes . . . because (Appia ?) conducts itself (geratur=se gerat) wrongfully,” they ? take the matter under their care.’

In the Greek part, though the restorations are often uncertain, the general sense is fairly clear. The coloni on the estate (χωρίον) appeal to the Emperor as their lord to put a stop to the violent conduct (twmultwosum vel iniuriosum adversus colonos Caesaris, Dig. i. 19) of the inhabitants of the Appian district who have ceased to confine themselves to the high roads and have

? The name occurs with one T or with two indifferently (cp. Hist. Geog. p. 240 &c.). * Iniuriose is the technical term (Dig. i. 19),

A SUMMER IN PHRYGIA; I. 421

taken to marauding on the estate, making themselves masters of Soa, inter- rupting agricultural work and blackmailing the coloni, in defiance of the Emperor's edict issued at the time of the appointment of the provincial governor and inserted in the archives of the town (7), ete. After the appeal there was added, no doubt, the imperial reply. The whole correspondence was set up publicly at the expense of the coloni under the superintendence of the ‘headman’ (as in the African analogies C.J.LZ. viii. 10570, Comptes Rendus de Vv Académie des Inscr. 1897, p. 146 ff.).

Important topographical questions are decided by this inscription. The estate referred to is that called in later times Tembrion or Tembre, because the river Tembrogios or Tembris (Porsuk Su) flowed through it. The exist- ence of an estate was detected here and the name Tembrion assigned to it in Hist. Geogr. p. 177-8 (see also CB. ii. p. 615). This view has now received complete confirmation. Our inscription makes it clear that the people of Soa (near Altyn Tash) and part at least of the Moxeanoi! were coloni on the domain. From this we should infer that the estate extended south to the borders of the territory of Alia (Islam keui)—a conclusion already reached by Prof. Ramsay (CB. p. 615, No. 527) from an inscription found at Gumulu, a village between Alia and Siokharax (Otourak), by MM. Legrand and Chamonard (B.C_H. 1893 p. 272). The northern limit is given by a boundary stone (6.1.1. Suppl. 7004), standing on a low ridge running out from the western hills and narrowing the plain opposite the villages Haidarlar and Nuh-éren: this stone probably marked off the estate from the territories of Apia and Aizanoi (cp. Hist. Geogr. p. 178), which perhaps met here. The estate there- fore included the whole valley of the upper Porsuk Su. Now, just as in other cases (6.9. Augustopolis), there must have been a bishopric for this vast stretch of country, and we are therefore compelled to agree with the view expressed in Hist. Geogr. p. 146, which assigns to this district the name Εὐδοκίας given by Hierocles between Appia and Aizanoi. Soa was perhaps the ecclesiastical centre. The name ’Apayounvoi is new.

The historical importance of the inscription lies in the fact that it supplies a fresh piece of evidence as to the status of the coloni on an Imperial estate in the third century. Previous to the fourth century, the colont whom we meet in literature and in law are free tenants, occupying holdings under a lease (= conductores); in the fourth century, the status of the colonus as defined by public law is altogether altered: he is still free, but his tenure is permanent and hereditary, he is ‘bound to the soil.’ This change has been traced by Prof. Pelham, in his clear and incisive style, to the influence of the regulations prevailing in the Imperial domains since Hadrian’s time: these

1 This agrees with the situation assigned to the Moxeanoi in CB. ii. p. 631 f.—a situation indeed already confirmed by epigraphic evidence (No. 615). I should now, however, prefer to say ‘the people of Soa and of Tottoia’; but the suggestion was received too late to be incor- porated in the text. The change does not,

H.S.—VOL. XVII.

however, affect what is said about the boundaries of the estate.

21 heard that Gumulu and Hassan keui are villages near the Devrent (on the eastern side) between Otourak and Islam keui: but I have not seen them. ι

GG

422 A SUMMER IN PHRYGIA: I.

were recognised to be advantageous and were merely stereotyped by the legislation of the fourth century.1. We know that the Imperial estates were reorganised by Hadrian?: and the status of the coloni under this reorganisa- tion (at least in Africa and on the Milyadic estates) was essentially that prescribed by the law of the fourth century (Pelham /.c. pp. 11-17%). Further, Prof. Pelham has shown that, although ‘we know nothing of the regulations by which the Caesars finally bound their own coloni to the soil, economic and political reasons (as well as the attractions offered to the colon?) all operated in this direction. This view is confirmed by our inscription, which gives similar evidence for the estate of Tembrion in 244-246 a.p. The coloni are clearly bound to the soil: they describe themselves as Caesar’s husbandmen, some of them at least having been planted on the estate by their Imperial lords (πάροικοι κὲ γεωργοὶ οἱ ὑμέτεροι, ἀροτῆρες, ἄγροικοι, Keoapsavoé), to whom they ‘flee for refuge, placing themselves under his protection’ when there is a general upheaval of law and order (1. 18). Compare the expressions used by the dwellers on the Saltus Burunitanus in Africa sixty years earlier (180-183 A.D.), homines rustici tenues manum nostrarum operis victum tolerantes, or rustici tut vernulae et alumni saltum tuorum (CLL. viii., 10570, col. ii., 20 and 28).

No mention is made of the conductores: that is natural, for the complaint is against blackmailing by outsiders. But what has become of the procurator whose duty it was to protect the ‘men of Caesar’ (1. 30)? He may have been mentioned, for the inscription is incomplete; but perhaps the force of παραφυλακῖται at his command was insufficient to cope with the marauders.

MEROS.

In Hist. Geogr. p. 144 (quoting J.H.S. 1887, p. 498, No. Ixvi.) Meros® was placed at Kumbet, where there are considerable remains, especially on the acropolis.© The evidence consisted only of the order in Hierocles, and the fact that it was the boundary between the Opsikian and Anatolic Themes. The identification was generally accepted by critics, including Prof. Kiepert and M. Radet. During an expedition to the country of the Praipenisseis in the beginning of September, I passed the village of Elmaly in the hilly country north-north-east from Altyn Tash (at or near which was the site of Soa), and copied there the following inscription.

1 The Imperial Domains and the Colonate, Rendus del’ Acad. des Inscr. 1897 p. 146 ff.

London, 1890. 3 The organisation of Milyadic Estates is also 3 Proved for Africa by inscriptions, especially described CB. i. p. 281 ff.

C.I.L. viii. 10570 (discussed by Mommsen, 4 πάροικοι, sojourners, resident foreigners, as

Hermes xv. 1880, p. 385 ff.) and for the Mily- in C.1.G. 1625, 45; 1631 ; 2906, &c.

adic or Killanian Estates in Asia Minor by 5 Mijpos in Hierocles and the Notitt., Mnpds

Ramsay (CB. i. p. 284). The lex Hadriana in Not. Basilii and Not. Leonis (ed. Gelzer) and undoubtedly applied to all the other estates. Const. Porphyr. de Thematibus i. pp. 14 and 25. Prof. Pelham points out (p. 18) that the idea 6 Κι. the Lion Tomb and Palace described of the new system originated with Vespasian by Prof, Ramsay in his Study of Phryg. Art and Trajan, and this is confirmed by the (J.H.S. 1889, p. 176 ff.).

African inscription recently published in Comptes

A SUMMER IN PHRYGIA: I. 423

ATAOFRITY XE KOPNHAIANCA AQNEINAN CEBACTHN FHMEIPFINGN ΚΑ O 1K IA

᾿Αγαθῇ Τύχη. Κορνηλίαν Σα-

λωνεῖναν

Σεβαστὴν ἡ Μειρηνῶν

κατοικία.

The inscription was evidently carried, along with a few other stones (including a richly ornamented sarcophagus now used as a fountain trough), and after copying it, I asked the circle of onlookers whence it came. The answer was: ‘It has been here a long time but we have heard from our fathers that it was brought from Malatia, while this other stone [an inscribed Byzantine column] came from Kara Agatch Oren.’ At the moment I did not recognise Meros as the town named in our inscription and as I had already heard that there were ruins at this place Malatia, I was eager to know what surprise was in store there. When we reached the spot, it was soon seen to be an ancient site. It lies between Doghan Arslan and Gerriz, half an hour from the latter, and in recent years a colony from Gerriz has built a village beside the old town. The ruins, which run out ftom the base of an oval-shaped hill, the acropolis no doubt of the old city, are mostly characterless ; but we were told that formerly there were many marbles there, most of which have been carried off by natives of the district to Kutaya (Kotiaion) 1\—twenty-five, they said, were taken away by mosque-builders from that city six or seven years ago—while the German Railway (which passes through the narrow plain) had destroyed great numbers ‘ written and — unwritten’: we ourselves saw the proof of their vandal depredations in the heaps of marble chips lying beside the foundations of a large building. In default of evidence, I determined to assign the name ἡ Μειρηνῶν κατοικία to this site: but, fortunately, our search resulted in the discovery of the fol- lowing inscription, which puts the identification with Meros? beyond doubt and proves the trustworthiness of the statement of the villagers of Elmaly.*

22. Ona rectangular block standing amongst the ruins: inscription much worn, but decipherable with certainty in favourable sunlight.

1 At Kara Agatch Oren (sPoRE), north-east of Altyn Tash, we were likewise told that many stones had been taken thence to Kutaya.

2 The form Miros occurs in 536 A.D. (Labbe, viii. p. 974).

3 I do not mean that all Turkish statements

re true, but the traveller can distinguish. If

e.g. one is visiting a frequented district, and were to ask whether a well-known inscription, in the possession of a villager, had been copied before, your friend (in expectation of bakshish) would of course answer ‘No.’ But when there is no motive for deception, there is less reason to disbelieve.

424 A SUMMER IN PHRYGIA: 1.

— eee

!

in en ] ATAOH TY KH ᾿Αγαθῇ Τύχῃ.

d OTHE ho 4Oi Ὁ NI Dr. “Orr:pov τὸν AIlACKMHTEMON Υ, διασημ(ότατον) ἡγεμόνα HMEIPENGONTTIOA ly ἡ Μειρηνῶν πόλις TON EYEPFET HN τὸν εὐεργέτην KAICGOT FIPATFICETTAP καὶ σωτῆρ[ α] τῆς ἐπαρ- OP oe χίου.

ALIN KV WMV. & NTT E PI

ΕἸ. Optimus is called perfectissimus pracses provinciae, and the inscription therefore probably dates after the reorganisation of Diocletian (it might, however, be shortly before Dioclet., cp. Mommsen, Staatsrecht, 11. p. 230,.n. 2). ἡ ἐπάρχειος occurs B.C.H. vii. p. 17, No. 3; so C.I.G. 6627 where ἐπαρχείου is wrongly taken as neuter.

Meros here calls itself πόλιες, and in the former inscription (dating shortly after the middle of III. Cent. after Christ) κατοικία. What sense does the latter term bear here? We cannot think of a military colony settled by the Greek kings. That is, no doubt, the most common meaning of κατοικία: but the term is also used to denote a settlement of the citizens of any given city living in an outlying part ofits territory and managing their own internal affairs.1_ _In the Imperial times it comes to mean merely a village (κώμη). This is probably the sense it bears here. Meros was most likely a village of the Praipenisseis, which was raised to the rank of a bishopric (before the time of Hierocles, ca. 530 A.D.) in accordance with the usual Byzantine policy. It may possibly have been a κατοικία subject to Kotiaion (or even Prymnessos)*: but this is less prob- able. Even in the tenth century it is called ἃ κωμόπολις by Constant. Porphyr.

The situation now assigned to Meros is about thirteen miles nearly due west of Kumbet. The question remains, what was the ancient name of this village? A village Pontanos (or —a) is proved for this neighbourhood in Hist. Geogr. p. 435, but it seems too unimportant to represent Kumbet. Unfortunately my visit there preceded the discovery of Meros, and as I accepted the generally received identification, and was at the time more specially interested in the Phrygian monuments, I did not make a careful search in the village. Two inscriptions of Kumbet relating to Epinikos, a native of the town who rose to high office in the Imperial service, and is known from literary sources, have just been published, by Prof. Mommsen from Prof. Ramsay’s copies in Hermes, 1897, p. 660 ff. Another inscription is published by Prof. Ramsay in J.H.S. 1887 p. 498.

J. G. C. ANDERSON.

1 See, forexample, CB. p.583and Nos. 498,499. 3 Not of Nakoleia, whose territory could ? As M.Radet says, Rev. Univ. Midi, 1896, p.6. hardly extend to the west side of the mountains.

RULES

OF THE

Society for the Promotion of Hellenic Studies.

1. THE objects of this Society shall be as follows :—

I. To advance the study of Greek language, literature, and art, and to illustrate the history of the Greek race in the ancient, Byzantine, and Neo-Hellenic periods, by the publication of memoirs and unedited documents or monuments in a Journal to be issued periodically.

II. To collect drawings, facsimiles, transcripts, plans, and photographs ot Greek inscriptions, MSS., works of art, ancient sites and remains, and with this view to invite travellers to communicate to the Society notes or sketches of archzological and topographical interest.

III. To organise means by which members of the Society may have increased facilities for visiting ancient sites and pursuing archeological researches in countries which, at any time, have been the sites of Hellenic civilization.

2. The Society shall consist of a President, Vice-Presidents, a Council, a Treasurer, one or more Secretaries, and Ordinary Members. All officers of the Society shall be chosen from among its Members, and shall be ex officio members of the Council.

3. The President shall preside at all General, Ordinary, or Special Meetings of the Society, and of the Council or of any Committee at which he is present. In case of the absence of the President, one of the Vice-Presidents shall preside in his stead, and in the absence of the Vice-Presidents the Treasurer. In the absence of the Treasurer

the Council or Committee shall appoint one of their Members to preside. b

x

4. The funds and other property of the Society shall be administered and applied by the Council in such manner as they shall consider most conducive to the objects of the Society: in the Council shall also be vested the control of all publications issued by the Society, and the general management of all its affairs and concerns. The number of the Council shall not exceed fifty.

5. The Treasurer shall receive, on account of the Society, all subscriptions, donations, or other moneys accruing to the funds thereof, and shall make all payments ordered by the Council. All cheques shall be signed by the Treasurer and countersigned by the Secretary.

6. In the absence of the Treasurer the Council may direct that cheques may be signed by two members of Council and countersigned by the Secretary.

7. The Council shall meet as often as they may deem necessary for the despatch of business.

8. Due notice of every such Meeting shall be sent to each Member of the Council, by a summons signed by the Secretary.

g. Three Members of the Council, provided not more than one of the three present be a permanent officer of the Society, shall be a quorum.

το. All questions before the Council shall be determined by a majority of votes. The Chairman to have a casting vote.

11. The Council shall prepare an Annual Report, to be submitted to the Annual Meeting of the Society.

12. The Secretary shall give notice in writing to each Member of the Council of the ordinary days of meeting of the Council, and shall have authority to summon a Special and Extraordinary Meeting of the Council on a requisition signed by at least four Members of the Council.

13. Two Auditors, not being Members of the Council, shall be elected by the Society in each year.

14. A General Meeting of the Society shall be held in London in June of each year, when the Reports of the Council and of the Auditors shall be read, the Council, Officers, and Auditors for the ensuing year elected, and any other business recommended by the Council discussed

χὶ

and determined. Meetings of the Society for the reading of papers may be held at such times as the Council may fix, due notice being given to Members.

15. The President, Vice-Presidents, Treasurer, Secretaries, and Council shall be elected by the Members of the Society at the Annual

Meeting.

16. The President and Vice-Presidents shall be appointed for one year, after which they shall be eligible for re-election at the Annual Meeting.

17. One-third of the Council shall retire every year, but the Members so retiring shall be eligible for re-election at the Annual Meeting.

18, The Treasurer and Secretaries shall hold their offices during the pleasure of the Council.

19. The elections of the Officers, Council, and Auditors, at the Annual Meeting, shall be by a majority of the votes of those present. The Chairman of the Meeting shall have a casting vote. The mode in which the vote shall be taken shall be determined by the President and Council.

20. Every Member of the Society shall be summoned to the Annual Meeting by notice issued at least one month before it is held.

21. All motions made at the Annual Meeting shall be in writing and shall be signed by the mover and seconder. No motion shall be submitted, unless notice of it has been given to the Secretary at least three weeks before the Annual Meeting.

22. Upon any vacancy in the Presidency, occurring between the Annual Elections, one of the Vice-Presidents shall be elected by the Council to officiate as President until the next Annual Meeting.

23. All vacancies among the other Officers of the Society occurring between the same dates shall in like manner be provisionally filled up by the Council until the next Annual Meeting.

24. The names of all candidates wishing to become Members of the Society shall be submitted to a Meeting of the Council, and at their next Meeting the Council shall proceed to the election of candidates so proposed: no such election to be valid unless the candidate receives

the votes of the majority qf those present. 62

XI

25. The Annual Subscription of Members shall be one guinea, payable and due on the Ist of January each year ; this annual subscription may be compounded for by a payment of £15 15s., entitling compounders to be Members of the Society for life, without further payment. All Members

elected on or after January 1, 1894, shall pay on election an entrance fec of one guinea.

26. The payment of the Annual Subscription, or of the Life Composition, entitles each Member to receive a copy of the ordinary publications of the Society.

27. When any Member of the Society shall be six months in arrear of his Annual Subscription, the Secretary or Treasurer shall remind him of the arrears due, and in case of non-payment thereot within six months after date of such notice, such defaulting Member shall cease to be a Member of the Society, unless the Council make an order to the contrary.

28. Members intending to leave the Society must send a formal notice of resignation to the Secretary on or before January I ; otherwise they will be held liable for the subscription for the current year.

29. If at any time there may appear cause for the expulsion of a Member of the Society, a Special Meeting of the Council shall be held to consider the case, and if at such Meeting at least two-thirds of the Members present shall concur in a resolution for the expulsion of such Member of the Society, the President shall submit the same for con- firmation at a General Meeting of the Society specially summoned for this purpose, and if the decision of the Council be confirmed by a majority at the General Meeting, notice shall be given to that effect to

the Member in question, who shall thereupon cease to be a Member of the Society.

30. The Council shall have power to nominate British or Foreign

Honorary Members. The number of British Honorary Members shall not exceed ten.

31. Ladies shall be eligible as Ordinary Members of the Society, and

when elected shall be entitled to the same privileges as other Ordinary Members.

32. No change shall be made in the Rules of the Society unless at least a fortnight before the Annual Meeting specific notice be given to every Member of the Society of the changes proposed.

xiii

RULES FOR THE USE OF THE LIBRARY

AT 22, ALBEMARLE STREET.

I, THAT the Library be administered by the Library Committee, which shall be composed of not less than four members, two of whom shall form a quorum.

II. That the custody and arrangement of the Library be in the hands of the Librarian and Assistant-Librarian, subject to the control of the Committee, and in accordance with Regulations drawn up by the said Committee and approved by the Council.

III. That all books, periodicals, plans, photographs, &c., be received by the Librarian, Assistant Librarian or Secretary and reported to the Council at their next meeting.

IV. That every book or periodical sent to the Society be at once stamped with the Society’s name.

V. That all the Society’s books be entered in a Catalogue to be kept by the Librarian, and that in this Catalogue such books, &c., as are not to be lent out be specified.

VI. That, except on Christmas Day, Good Friday, and on Bank Holidays, the Library be accessible to Members on all week days from eleven A.M. to six P.M. (Saturdays, 11 A.M. to 2 P.M.), when either the Assistant-Librarian, or in her absence some responsible person, shall be in attendance. Until further notice, however, the Library shall be closed for the vacation from July 20 to August 31 (inclusive).

VII. That the Society’s books (with exceptions hereinafter to be specified) be lent to Members under the following conditions :—

(1) That the number of volumes lent at any one time to each Member shall not exceed three.

(2) That the time during which such book or books may be kept shall not exceed one month.

(3) That no books be sent beyond the limits of the United Kingdom. VIII. That the manner in which books are lent shall be as follows :—

(1) That all requests for the loan of books be addressed to the Librarian.

(2) That the Librarian shall record all such requests, and lend out the books in the order of application.

(3) That in each case the name of the book and of the borrower be inscribed, with the date, in a special register to be kept by the Librarian.

(4) Should a book not be returned within the period specified, the Librarian may reclaim it.

“xiv

(5) All expenses of carriage to and fro shall be borne by the borrower.

(6) All books are due for return to the Library before the summer vacation.

IX. That no book falling under the following categories be lent out under any circumstances :— (1) Unbound books, (2) Detached plates, plans, photographs, and the like. (3) Books considered too valuable for transmission. (4) New books within one month of their coming into the Library. X. That new books may be borrowed for one week only, if they have been more than one month and less than three months in the Library.

XI. That in the case of a book being kept beyond the stated time the borrower be liable to a fine of one shilling for each week after application has been made by the Librarian for its return, and if a book is lost the borrower be bound to replace it.

The Library Committee. PROF. PERCY GARDNER. Miss JANE HARRISON, LL.D. ΜΕ. WALTER LEAF, Litt.D. MR. GEORGE MACMILLAN (fon. Sec.). Mr. ERNEST MYERS. REv. W. G. RUTHERFORD, LL.D. Mrs. 5: ARTHUR STRONG, LL.D. Mr. ARTHUR HAMILTON SMITH. (Hon. Librarian). Sir E. MAUNDE THOMPSON, K.C.B., D.C.L. Mr. TALFOURD ELY.

Assistant Librarian, MISS FANNY JOHNSON, to whom, at 22, Albemarle Street, applications for books may be addressed.

SESSION 1897—1898.

General Meetings will be held in the Rooms of the Royal Asiatic Society, 22, Albemarle Street, London, W., for the reading of Papers and for Discussion, at 5 P.M. on the following days :—

1897. Thursday, November 4th. 18098. Thursday, February 17th. Thursday, April 2151. Thursday, June 23rd (Annual). The Council will meet at 4.30 p.m. on each of the above days.

THE SOCIETY FOR.THE PROMOTION OF HELLENIC STUDIES OFFICERS AND COUNCIL FOR 1897—1808,

President. PROFESSOR ΚΕ. C. JEBB, Litt.D., D.C.L., LL.D., M.P.

Vice-Presidents.

PROF. 5. H. BUTCHER, Lirtt.D., LL.D.

PROF. INGRAM BYWATER.

REV. PROF. LEWIS CAMPBELL.

MR. SIDNEY COLVIN.

PROF. PERCY GARDNER, Lirtt.D.

MR. D. B. MONRO, Provost of Oriel College, Oxford. MR. A. S. MURRAY, LL.D.

PROF. H. F. PELHAM, President of Trinity College, Oxford.

MR. F.C. PENROSE, F.R.S. PROF. W. M. RAMSAY, D.C.L. MR, J. E. SANDYS, Lirtt.D. REV. PROF. A. H. SAYCE, LL.D.

SIR E. MAUNDE THOMPSON, K.C.B., D.C.L. REV. H. F. TOZER. PROF. ΚΟΥ. TYRRELL, Litt.D., D.C.L.

Council.

PROF. W. Ο. F. ANDERSON. REV. A. G. BATHER.

MR. H. G. DAKYNS.

MR. LOUIS DYER.

MR. TALFOURD ELY.

MR. ARTHUR J. EVANS.

LADY EVANS.

PROF. ERNEST A. GARDNER. MR. B. P. GRENFELL.

MISS JANE HARRISON, LL.D. ‘MR. J. W. HEADLAM.

MR. G, F. HILL.

MR. D. G. HOGARTH. SIRHENRYH. HOWORTH, K.C.I.E., M.P., F.R.S. MR. H.STUART JONES.

MR. F. G. KENYON.

MR. WALTER LEAF, Lirtt.D.

MR. WILLIAM LORING.

REV. PROF. JOSEPH MAYOR. MR. J. A. R. MUNRO.

PROF. G. G. A. MURRAY. MR. ERNEST MYERS.

MR. J. L. MYRES.

MR. R. A. NEIL.

MISS EMILY PENROSE.

MR. G. H. RENDALL, Lirtt.D. PROF. WILLIAM RIDGEWAY. MR. R. W. SCHULTZ.

MR. CECIL SMITH, LL.D. MR.A. HAMILTON SMITH. MRS, 5. ARTHUR STRONG, LL.D. MR. H. B. WALTERS.

REV. W. WAYTE.

Hon. Treasurer. MR. DOUGLAS W. FRESHFIELD.

Hon. Secretary. MR. GEORGE A. MACMILLAN, ST. MARTIN'S STREET, W.C.

Assistant Secretary. MR. W. RISELEY.

Hon. Librarian. MR. ARTHUR H. SMITH.

Assistant Librarian. MISS FANNY JOHNSON.

Acting Editorial Committee.

PROF, ERNEST GARDNER. |

MR. WALTER LEAF. |

MR. A. H. SMITH.

Consultative Editorial Committee. PROFESSOR JEBB | PROFESSOR BYWATER | SIRE. MAUNDE THOMPSON | MR. SIDNEYCOLVIN and Mr. D. G. HOGARTH (ex officio), as Director of the British School at Athens.

Auditors for 1897-98.

MR. ARTHUR J. BUTLER.

MR. STEPHEN SPRING-RICE, C.B.

Bankers. MESSRS. ROBARTS, LUBBOCK & CO., 15, LOMBARD STREET.

CAMBRIDGE BRANCH

OF

THE SOCIETY, FOR*#P?HE, PROMOTION OF HELEERNIG:ST UDIES:

OFFICERS AND COMMITTEE FOR 1897-1898.

Chairman. PROV, ΓΕΒ birt. 5 ΟΣ lls... ΜΕΡ:

Vice-Chairman. Mr. J. E. SANDys, LitT.D.

Committee.

Mr. J. G. FRAZER. | Mr. E. ἘΣ SIKEs.

Pror. ERNEST A. GARDNER. Mr. ARTHUR TILLEY.

Mr. Henry JAckson, LitT.D. | Mr. A. W. VERRALL, {τὺ} Ὁ). Mr. M. R. James, LitT.D. | Mr. C. WALDSTEIN, LiTT.D.

Pror. W. RIDGEWAY.

Hon. Secretarp.

Mr. ARTHUR BARNARD COOK, TRINITY COLLEGE.

XVll

HONORARY MEMBERS.

HIS MAJESTY THE KING OF THE HELLENES, 4 M. le Secretaire du Koi des Hellenes, Athens, Greece.

Sir Alfred Biliotti, K.C.B., 4.8... Consul for Crete.

Prof. D. Comparetti, /stituto di Studii Superior, Fiorence.

M. Alexander Contostavlos, Athens

Mr. George Dennis c/o Lloyds Bank, Limited, 16, S¢ James's Street, S.W.

Prof. Wilhelm Dorpfeld, Ph.D., Director of the German Archeological Institute, Athens.

Monsieur P. Foucart, 13, Rue de Tournon, Paris,

Prof. Furtwangler,

Monsieur J. Gennadius, 21, Hyde Park Place, W.

His Excellency Hamdy Bey, Keeper of the Museum of Antiquities, Constantinople.

Prof. W. Helbig, Villa Lante, Rome.

Monsieur Homolle, Director of the French School, Athens.

Monsieur P. Kavvadias, Ephor-General of Antiquities, Athens, Greece.

Prof. A. Kirchhoff, The University, Berlin.

Prof. U. Kohler, Zhe University, Berlin.

Prof. 5. A. Kumanudes, 7he University, Athens.

Mr. Charles Merlin, 19, Ladbroke Grove, VW’.

Prof. A. Michaelis, Uuzversity, Strassburg.

Prof. E. Petersen, /mstituto Archeologico Germanico, Monte Tarpeo, Rome.

LIST OF MEMBERS.

* Original Members. Ὑ Life Members. The other Members have been elected by the Council since the Inaugural Meeting.

Abbott, Evelyn, Balliol College, Oxford. *Abercromby, Lord, 14, Grosvenor Street, W. + Abercrombie, Dr. John, 23, Upper Wimpole Street, W. Abram, Edward, 1, Middle Temple Lane, E.C. *Acland, Sir H. W., Bart., K.C.B., M.D., F.R.S., Broad Street, Oxford. Adam, James, Emmanuel Colleye, Cambridge. Adams, Miss Mary G., 43, Campden Hill Square, Kensington, W. Agnew, Philip L., 11, Devonshire Terrace, Hyde Park, W. Ainger, A.C., Eton College, Windsor. Ainger, Rev. Canon, Master's House, The Temple, E.C. +Ainslie, R. St. John, Zhe School, Sedbergh. Alford, Rev. B. H., St. Luke's Vicarage, Nutford Place, W. Alford, ΚΕ. G., Queen Anne’s Mansions, S.W. Allbutt, Professor T. Clifford, M.D., F.R.S., Chaucer Road, Cambridge. Amherst, Lord, Didlington Hall, Brandon, Suffolk. Anderson, J. C. G., 166, Union Street, Aberdeen. Anderson, J. R., Latrbeck, Keswick. Anderson, Prof. W. C. F.(Council), Férth College, Sheffield. Anderton, Basil, Public Library, Newcastle-on- Tyne. *Antrobus, Rev. Frederick, 71 Oratory, S.W. Apostolides, S., 24, Montpelier Road, Brighton. Archer-Hind, R. D., Trinity College, Cambridge. tArkwright, W., ddbury House, Newbury. Awdry, Herbert, Wellington College, Berks. Bagley, Mrs. John, Washington Avenue, Detroit, Michigan. Bailey, J. C., 118, Ashley Gardens, S.W. Baker, F. B., The College, Great Malvern. Baker, H. T., Mew College, Oxford. Baker, Rev. William, D.D., Merchant Taylors’ School, E.C. *Balfour, Right Hon. A. J., M.P., 4, Carlton Gardens, S.W. *Balfour, Right Hon. G. W., M.P., 24, Addison Road, W.

xvill

Ball, Sidney, St. John’s College, Oxford. Barclay, James W., M.P., 5, Clarendon Place, Hyde Park Gardens, W. +Barlow, Miss Annie E. F., Greenthorne, Edgworth, Bolton. Barlow, Mrs., 10, Wimpole Street, W. Barnewall, Sir Reginald A., Bart., 23, Cliveden Place, Eaton Square, S.W. Barnsley, Sidney H., Pembury, near Cirencester. Barran, J. N., Weetwood, Leeds. Bather, Rev. Arthur George (Council), 8, Kingsgate Street, Winchester. Bayfield, Rev. M. A., Eastbourne College, Eastbourne. Beare, Prof.-J. Isaac, 9, Trinity College, Dublin. + Beaumont, Somerset, Shere, near Guildford. Beebee, M. J. L., New Travellers Club, 97, Piccadilly, W. ft Benn, Alfred W.,70, Via Cavour, Florence. Bennett, 5. A., Audley House, Richmond, Surrey. Benson, E. F., King’s College, Cambridge. Bent, Mrs. Theodore, 13, Great Cumberland Place, W. Bevan, E. R., 14, Beaumont Street, Oxford. Bickford-Smith, R. A. H., 45, North Bailey, Darlington. ft Bikelas, Demetrius, 50, Rue de Varenne, Paris. Blomfield, Sir A. W., A.R.A., 6, Montagu Place, Montagu Square, W.C. Blomfield, Mrs. Massie, Port House, Alexandria, Egypt. Blore, Rev. Dr., St. Stephen’s, Canterbury. Blumenfeld, Ralph Drew, 64, Cheyne Court, Chelsea, S. W. Bodington, Prof. N., Principal of the Yorkshire College, Leeds. Bond, Sir Edward, K.C.B., LL.D., 64, Princes Square, Bayswater, W. Bond, Edward, M.P., Elm Bank, Hampstead, N.W. Bosanquet, Rev. F. C. T., The Hermitage, Uplyme, Devon. Bosanquet, R. Carr, Rock Hall, Alnwick, Northumberland. Bosdari, Count Allessandro di, 20, Grosvenor Square, W. Bougatsos, Christos Ch., Alexandria, Egyft. Bousfield, William, 20, Hyde Park Gate, S.W. Boyd, Rev. Henry, D.D., Principal of Hertford College, Oxford. Boys, Rev. H. A., North Cadbury Rectory, Bath. Bramley, Rev. H. R., The Precentory, Lincoln. Bramwell, Miss, 73, Chester Square, S.W. Branteghem, A. van, 28, Rue des Buisson, Bruxelles. Brinton, Hubert, Eton College, Windsor. Broadbent, H., Eton College, Windsor. *Brodie, E. H., H.M.I.S., Grasendale, Malvern. Brooke, Rev. A. E., King’s College, Cambridge. Brooke, Rev. Stopford A., 1 Manchester Square, W. Brooks, E. W., 28, Great Ormond Street, W.C. Brooksbank, Mrs., Leigh Place, Godstone. Brown, Horace T., F.R.S., 52, Mevern Square, South Kensington, S.W. Brown, Prof. G. Baldwin, The University, Edinburgh. *Browning, Oscar, King’s College, Cambridge. *Bryce, The Right Hon. James, D.C.L., M.P., 54, Portland Place, W. Buller, Lady Audrey, 29, Bruton Street, W. Bulwer, Sir Henry, K.C.B., 11, South Street, Park Lane, W. Burnet, Prof. J., 1, Alexandra Place, St. Andrews, N.B. Burton, Sir F. W., 43, Argyll Road, Kensington, W. Bury, Prof. J. B., Tvinzty College, Dublin. Burge, Hubert M. University College, Oxford. Burgh, W. de, University Extension College, Reading. Burrows, Prof. Ronald, University College, Cardiff. Butcher, Prof. 5. H., Litt.D., LL.D. (V.P.), The University, Edinburgh. f Bute, The Marquis of, K.T., St. John’s Lodge, Regent's Park, N.W. Butler, Arthur J., Wood End, Weybridge. *Butler, The Rev. H. M., D.D., Master of Trinity College, Cambridge. Buxton, F. W., 42, Grosvenor Gardens, S.W. Buxton, Mrs. Alfred W., 32, Great Cumberland Place, W. Bywater, Prof. Ingram (V.P.), 93, Onslow Square, S.W.

xix

+Bywater, Mrs., 93, Onslow Square, S.W. Calvert, Rev. Thomas, 121, Hopton Road, Streatham, S.W.

tCalvocoressi, L. M., Messrs. Ralli Bros., Mellor’s Buildings, Exchange Street East, Liverpool.

Cameron, Dr. James, Registrar of the University, Capetown. Campbell, Rev. Prof. Lewis (V.P.), 35, Kensington Court Mansions, W. Campbell, Mrs. Lewis, 35, Kensington Court Mansions, W. Capes, Rev. W. W., Bramshott, Liphook, Hants, Carapdnos, Constantin, Député, Athens. Carey, Miss, 13, Colosseum Terrace, Regent’s Park, N.W. *Carlisle, A. D., Hatleybury College, Hertford. Carlisle, Miss Helen, Houndhill, Marchington, Stafford. +Carr, Rev. A., St. Sebastian’s Vicarage, Wokingham. + Carmichael, Sir T. D. Gibson, Castlecraig, Dolphinton, N.B. Carter, Prof. Frank, McGill University, Montreal. Carthew, Miss, 15a Palace Gardens, W. Cartwright, T. B. Case, Miss Janet, 5 Windmill Hill, Hampstead, S,W. Cates, Arthur, 12, York Terrace, Regent’s Park, N.W. Cave, Lawrence T., 13, Lowndes Square, S.W. Chambers, C. Gore, Hertford House, De Parry’s Avenue, Bedford. Chambers, Charles D., The Steps, Bromsgrove, Worcestershire. Chance, Frederick, 51, Prince’s Gate, S.W. Chavasse, A. S., Kempsey, Worcestershire. + Chawner, G., King’s College, Cambridge. +Chawner, W., Master of Emmanuel College, Cambridge. Cheetham, J. C. M., Zyford Park, Bourton-on-the- Water, R.S.O., Gloucestershire. Cheetham, J. Frederick, Eastwood, Staleybridge. *Christie, R. C., Ribsden, Bagshot, Surrey. Christian, J. Henry, 18, Devonshire Place, Portland Place, W. Christian, Rev. G., Redgate, Uppingham. Churchill, E. L., Eton College, Windsor. Clark, Charles R.R., c/o EZ. P. Warren, Esg., 18, Cowley Street, Westminster, S.W. +Clark-Maxwell, Rev. W. Gilchrist, Clundury Vicarage, Aston-on-Clem, Salop. Clarke, Joseph Thacher, 3, College Road, Harrow, N.W. +Clauson, A. C., 12, Park Place Villas, Paddington, W. Clarke, Somers, 22, Whitehall Court, SW Clay, C. F., 38, Great Ormond Street, W.C. Clerke, Miss Agnes, 68, Redcliffe Square, S.W. Cobbold, Felix T., Zhe Lodge, Felixstowe, Suffolk. *Cobham, C. Delaval, H.2.M. Commissioner, Larnaca, Cyprus. Colby, Rev. Dr., 12, Hillsborough Terrace, Ilfracombe. Cole, A. C., 64, Portland Place, W. Colfox, William, Westmead, Bridport. Colvin, Sidney (V.P.), British Museum, W.C. Collins, Miss F. H., 3, Bramham Gardens, South Kensington, S.W. Collins, J. Churton, 51, Worfolk Square, W. Colvill, Miss Helen H., Overdale, Shortlands, Kent. Compton, Rev. W. C., The College, Dover. Connul, B. M., Zhe Yorkshire College, Leeds. *Constantinides, Prof. M., Coundouriotes Street, Munychia, Petraeus, Athens. Conybeare, F. C., 13, Norham Gardens, Oxford. Conway, Sir W. M., The Red House, 21, Hornton Street, W. Cook, Arthur Barnard, Trinity College, Cambridge. Cookson, C., Magdalen College, Oxford. Cookson, Sir C. A., C.B., H.B.M. Consul, Alexandria. Cordery, J. G., C.S.1., 63, Goldington Road, Bedford. Corbet, His Honour Eustace K., Matrve Court of Appeal, Cairo, Corgialegno, M., 21, Pembridge Gardens, W. Courtney, W. L., 53, Belsize Park, N.W. Courtenay, Miss, 34, Brompton Square, S.W,

XX

Cowper, The Right Hon. Earl, K.G., Panshanger, Hertford. Craik, George Lillie, 2, West Halkin Street, S.W. Crawley, C., 3, Regent Street, S.W. Crewdson, Wilson, 7he Barons, Reigate. Crooke, W., Westleigh, Arterberry Road, Wimbledon, S.W. tCrossman, C. S., The College, Winchester. Crowfoot, J. W.,, Bzshop’s Hostel, Lincoln. Cruikshank, Rev. A. H., The College, Winchester. Cust, H. J. C., Ellesmere, Salop. Cust, Lionel, 9, Bryanston Square, W. Cust, Miss Anna Maria, 63, E/m Park Gardens, Fulham Roaa, S.W. Cust, Miss Beatrice, 13, Eccleston Square, S.W. Dabis, Miss, Holloway College, Egham, Surrey. Dakyns, H. G. (Council), Higher Coombe, ypewenie Surrey. Danson, J. T., F.S.A., Grasmere, R.S.O. David, W., 8, Hyde Pane Terrace, W. David, Hew. W. H., Kelly College, Tavistock. Davidson, H. O. D., Harrow, N.W. t Davies, G. A., Trinity College, Cambridge. Davies, Rev. Gerald S., Charterhouse, Godalming. Deibel, Dr., care of Messrs. Asher, Beriin. Delamarre, Jules, 4, /passe Royer-Collard, Paris. , De Saumarez, Lord, Shrubland Park, Coddenham, Suffolk. Dickson Miss Isabel A., Dumnichen House, Forfar. Dill, Prof. S., Montpelier, Malone Road, Belfast. Dobson, Miss, 77, Harcourt Terrace, Redcliffe Square, S.W. Donaldson, James, LL.D., Principal of The University, St. Andrews, Donaldson, Rev. S. A., Eton College, Windsor. Dragoumis, M. Etienne, Athens, Greece. Draper, W. H., 52, Doughty Street, W.C. Drisler, Prof. Henry, 48, West 46th Street, New York City, U.S.A. Drummond, Allan, 7, Eanismore Gardens, S. VW. Duchataux, M. V., 12, Rue de l’Echauderie, ἃ Reims. Duckworth, H. T. F., Merton College, Oxford. Duhn, Prof. von, University, Heidelberg. Duke, Roger, 8, Neville Terrace, Onslow Gardens, S.W. fDunham, Miss, 37, Hast Thirty-Sixth Street, New York. Dunlap, Miss Mabel Gordon, 425, 721 Street, Portland, Oregon, U.S.A. Durham, The Right Rev. The Lord Bishop of, Auckland Castle, Bishop Auckland. Dyer, Louis (Council), Sunbury Lodge, Banbury Road, Oxford. Earl, Mrs. A. G., Ferox Hall, Tonbridge. Earp, F. R., King’s College, Cambridge. Edwards, G. M., Sidney Sussex College, Cambridge. Egerton, Mrs. Hugh, 11, 7216 Street, Chelsea, S.W. tEgerton, Sir Edwin H., K.C.B., H.B.M. Minister, British Legation, Athens, Greece. Egerton, Miss M., Whitwich Hali, York. Eld, Rev. F. J., Polstead Rectory, Colchester. TEllis, Prof. Robinson, 7vinity College, Oxford. Elwell, Levi H., Amherst College, Amherst, Mass. Ely, Talfourd (Council), 73, Parliament Hill Road, Hampstead, N.W. Emens, Edgar A., Syracuse University, New York. Erichsen, Miss Nelly, Woodlands, Elmbourne Road, Upper Tooting, S.W. Eumorfopoulo, A.,1, Kensington Park Gardens, W. Evans, A. J. (Council), Ashmolean Museum, Oxford. Evans, Sir John, K.C.B., D.C.L., F.R.S., Mash Mills, Hemel Hembstead. t Evans, Lady (Council), Mash Mills, Hemel Hempstead. Eve, H. W., 37, Gordon Square, W.C. Ewart, Miss Mary A., 68, Albert Hall Mansions, S.W. Fanshawe, Reginald, 37, Pembroke Road, Clifton. Farnell, L. R., Exeter College, Oxford. Farrar, Rev. Canon A. S., Durham.

XX

Farrow, Frederic R., 2, Mew Court, Carey Street, W.C.

Farside, William, 7horpe Hall, Robin Hood's Bay, Yorkshire.

*Fearon, Rev. W. A., D.D., The College, Winchester.

Fenning, W. D., Hazleybury College, Hertford.

Field, Rev. T., Radley College, Abingdon.

Firminger, W. K., Merton College, Oxford Fisher, H. A. L., New College, Oxford.

Fisher, Miss Edith S., 21, Chapel Street, Belgrave Square, S.W.

{ Fitzmaurice, Lady Edmond, 2, Green Street, Grosvenor Square, W. Fitz-Patrick, Di T., 30, Sussex Gardens, Hyde Park, W. Flather, J. H., 52 Bateman Street, Cambridge.

Flower, Wickham, O/d Swan House, Chelsea Embankment, S.W.

tForbes, ΝΥ. H., Balliol College, Oxford.

Ford, His Excellency the Right Hon. Sir Francis Clare, G.C.B.,G.C.M.G., H.B.M. Ambassador, British Embassy, Rome.

Forster, Miss Frances, 46, E/m Park Road, S.W.

Fowler, Harold N., Ph.D., Western Reserve University, Cleveland, Ohio, U.S.A.

*Fowler, Rev. Professor, President of Corpus Christi College, Oxford. Fowler, W. Warde, Lincoln College, Oxford.

Franklin, T. M., St. Hilary, Cowbridge, S. Wales.

Frazer, J. G., Trinity College, Cambridge.

Freeman, C. E., Parkhouse, Southborough, Tunbridge Wells.

Freeman, W. George, Lee Street, Plumstead.

*Freshfield, Douglas W. (Hon. Treasurer), 1, Azv/ie Gardens,Campden Hill, W. tFreshfield, Edwin, LL.D., 5, Bank Bailes: Ee.

Freston, Henry W., PAPER. Prestwich, Lancashire.

Bry. ἢ.» ΕΑ ΗΝ Leigh Wood, Clifton.

Fry, Right Hon. Sir Edward, Fazland House, Failand, near Bristol.

Fullerton, W. Morton, Rue Vignon,Paris.

+Furley, J. S., 10, College Street, Winchester.

Furneaux, L. R., Rossall School, Fleetwood. Furneaux, Rev. W. M., Refton Hall, Burton-on-Trent.

tGardner, Prof. Ernest A. (Council), Radnor Cottage, Sandgate.

*+Gardner, Prof. Percy, Litt.D. (V.P.), 12, Canterbury Road, Oxford. Gardner, Miss Alice, 7216 Old Hall, Newnham College, Cambridge. Gardner, Samuel, Oakhurst, Harrow-on-the-Hill.

Gardner, W. Amory, Groton, Massachusetts, U.S.A.

Garnett, Mrs. Terrell, 3, Queen Anne's Gate, S.W.

Geddes, Sir W. D., Principal of the University, Aberdeen.

Geikie, Sir Archibald, F.R.S., 10, Chester Terrace, Regents Park, N.W. Gibbs, F. W., Q.C., C.B., 38, Cornwall Gardens, South Kensington, S.W. Gibson, Mrs. Margaret D., Castle-brae, Chesterton Road Cambridge.

Giles, P., Emmanuel College, Cambridge.

Gilkes, A. H., The College, Dulwich, S.E.

Gilliat, Rev. E., Harrow, N.W.

Glazebrook, Rev. M. G., Clifton College, Bristol.

Godden, Miss Gertrude M., Ridgfeld, Wimbledon.

Gonino, Miss G., 90, Warwick Street, Warwick Square, S.W.

Goodhart, A. M., Eton College, Windsor.

Goodrich, Prof. F. S., Albion College, Albion, Michigan, U.S.A.

Goodwin, Prof. W. W., D.C.L., Harvard College, Cambridge, Mass., U.S.A, Gow, James, Litt.D., High School, Nottingham. Gower, Lord Ronald, 27, 7rebovir Road, Earl’s Court, S.W. Granger, F. S., Untversity College, Nottingham. Graves, A. S., Felsted School, Essex.

Gray, Rev. H. B., Bradfield College, Berks. Green, Mrs. J. R., 14, Kensington Square, W. Grenfell, B. P. (Council), Queen’s College, Oxford. Greenwell, Rev. Canon, F.R.S., Durham.

Griffith, G., Harrow, N.W.

Griffith,Miss Mary E., 4, Bramham Gardens, S.W.

Grundy, George Beardoe, 27, St. Maryaret’s Road, Oxford.

XXll

Gurney, Miss Amelia, 69, Eanismore Gardens, S.W. Hadow, W. H., Worcester College, Oxford. Haigh, A. E., 2, Crick Road, Oxford. Hales, Rev. C. T., Aysgarth School, Newton-le-Willows, R.S.O. Yorks. Hall-Dare, Francis, 10, Bury Street, St. James’s, S.W. Hall, Rev. F.H., Oriel College, Oxford. Hall, Miss S. E., 37, York Street Chambers, Bryanston Square W. Hall, Harry Reginald, 13, Chalcot Gardens, N.W. Hall, Rev. F. J., Morthaw Place, Potter’s Rar, Herts. Hall, F, W., Westminster School, Dean’s Yard, Westminster, S.W. Hallam, G. H., Zhe Park, Harrow, N.W. t Hammond, B. E., Trinity College, Cambridge. Hardie, Prof. W. Ross, Zhe University, Edinburgh. Hardinge, Miss. Hardwich, J. M., 55, South Street, Durham. Harrison, Miss F. Bayford, Suffolk House, Weybridge. Lh σας Miss J. E., LL.D. (Council), 37, Barkston Gardens, Earl's Court,

Harrower, Prof. John, The University, Aberdeen. Hartshorne, B. F., 41, E/m Park Gardens, Chelsea, S.W. Haslam, S., The School, Uppingham. Haussoullier, Β., 8, Rue Sainte-Cécile, Paris.

+ Haverfield, F. J., Christ Church, Oxford. Hawes, Miss E. P., 89, Oxford Terrace, W.

tHay, C. A., 127, Harley Street, W. Hay, A. T., The College, Brighton.

tHaynes, Miss Lucy, 7, Thornton Hill, Wimbledon. Hayter, Angelo G. K., 74, Adelaide Road, N.W. Headlam, Rev. A. C., Welwyn Vicarage, Herts. Headlam, C. E. S., Trinity Hall, Cambridge. Headlam, J. W. (Council), 6, E/don Road, Kensington, W. Headlam, W. G., King’s College, Cambridge. Heard, Rev. W. A., Fettes College, Edinburgh.

tHeathcote, W. E., Round Coppice, Ivor Heath, Uxbridge. Heberden, C. B., Principal of Brasenose College, Oxford. Hedgcock, Mrs. Harrison, 21, Caversham Road, N.W. Hereford, The Lord Bishop of, The Palace, Hereford. Herschell, The Right Hon. Lord, 46, Grosvenor Gardens, S.W. Heyer, G., The College, Weymouth. Hicks, Rev. E. L., 21, Leaf Sguare, Pendleton, Manchester. Higgins, Alfred, 16, King Street, Portman Square, W. Hill, George F. (Council), British Museum, W.C.

Hill, Arthur, British Vice-Consul, Athens, Greece. Hobhouse, Rev. Walter, Zhe School House, Durham. Hodgson, F. C., Education Department, Whitehall, S.W.

Tt Hodgson, J. Stewart, 1, Audley Square, W. Hogarth, David G. (Council), British Archeological School, Athens. Holiday, Henry, Oak Tree House, Branch Hill, Hampstead, N.W. Holland, Miss Emily, 27, Homefield Road, Wimbledon. Hopgood, Harold B., 17, Whitehall Place, S.W. Hopkinson, J. H., University College, Oxford. Hoppin, J. C., c/o 7. 5. Morgan & Co., 22, Old Broad Street, E.C. Housley, Samuel J., Gynsdal, Waterloo Road, Epsom. Hornby, Rev. J. J., D.D., Provost of Eton College, Windsor.

t+ Hort, Arthur F., Adoyne, Harrow. Howorth, Sir,Henry H., K.C.1.E., M.P. (Council), 30, Collingham Place, S.W. Huddart, Rev. G. A. W., Kirkiington Rectory, Bedale, Yorks Hiigel, Baron Friedrich von, 4, Holford Road, Hampstead, N.W. Hughes, Rev. W. Hawker, Jesus College, Oxford. Hughes, Miss C., 22, Albemarle Street, W. Hulse, Miss Caroline Μ,

XX1l

Humphreys, Robert Henry, Royal Societies Club, 63, St. James Street, S. W. Hunt, A. S., Queen’s College, Oxford. Hutton, Miss Ὁ. A., 18, Cheyne Court, Chelsea, S. W. Huyshe, Wentworth, ‘ Dazly Graphic’ Office, Milford Lane, Strand, W.C. Image, Selwyn, 6, Southampton Street, Bloomsbury, W.C. lonides, Alex. A., 1, Holland Park, |W. Ionides, Luke A., 47, Marloes Road, Kensington, W. Jackson, Henry, Litt.D., Trinity College, Cambridge. Jackson, Miss Rose, Longdene, Haslemere. Jackson, Rev. Blomfield, 29. Mecklenburgh Square, W.C. Jackson, Rev. W. W., Rector of Exeter College, Oxford. James, A. C., Eton College, Windsor. *James, The Rev. H.A., D.D., School House, Rugby. James, Lionel, St. Peter's College, Radley, Abingdon. James, M. R., Litt.D., Aznug’s College, Cambridge. James, Rev. S. R., The College, Malvern. Jannaris, A. N., Ph.D., The University, St. Andrews, N.B. Jeans, Rev. G. E., Shorwell, Newport, Isle of Wight.

*Jebb, Prof. R. C., D.C.L., LL.D., Litt.D., M.P. (President), Springheld, Newnham, Cambridge.

Jenkin, Miss M. L., Carfax, King Charles’ Road, Surbiton. Jenkinson, F.J.H., Zrinzty College, Cambridge. Jenner, Miss Lucy A., 39, Adadtson Road, Kensington, W. Jevons, F. B., The Castle, Durham. Jex-Blake, Miss, Girton College, Cambridge. Jobling, G. C., 5, Park Villas, Cheltenham. Jones, H. Stuart (Council), 7rznzty College, Oxford. Karo, George, 19, Ptazza del Carmine, Firenza, Italy. Keep, R. P., Ph.D., Free Academy, Norwich, Conn., U.S.A. Keene, Prof. Charles H., 11, Dyke Parade, Cork. Kelly, Charles Arthur, 30, Cheyne Walk, Chelsea, S.W. Keltie, J. S., Glendevon House, Compayne Gardens, Hampstead, N.W. Kennedy, Rev. John, Grammar School, Aldenham, Elstree, Herts. Kenyon, F. G. (Council), Brztish Museum, W.C. Ker, Prof. W. P., 95, Gower Street, W.C. Kerr, Prof. Alexander, M/adison, Wisconsin, U.S.A. Keser, Dr. J., 11, Harley Street, Cavendish Square, W. Kieffer, Prof. John B., 232, Lancaster Avenue, Lancaster, Pa., U.S.A. King, J. E., Grammar School, Manchester. King, Rev. J. R., St. Peter’s Vicarage, Oxford. King, Mrs. Wilson, 19, Highfield Road, Edgbaston, Birmingham. Kirwan, Miss Evelyn, 1, Richmond Gardens, Bournemouth. Knowles, James, Queen Anne’s Lodge, St. James Park, S.W. Krohn, H. A., 103, Cannon Street, E.C. Lambros, Spiridion, Athens, Lang, Andrew, LL.D., 1, Marloes Rd., Kensington, W. *Lang, Sir R. Hamilton, K.C.M.G., The Ottoman Bank, Constantinople. Lathbury, Miss, 19, Lingfield Road, Wimbledon, S.W. Lautour, Miss de, 85, Harcourt Terrace, Redcliffe Square, S.W. Lawford, Frederick le Breton, 65, Fitzjohns Avenue, Hampstead, N.W. Lawrence, Sir Edwin, Bart., M.P., 13, Carlton House Terrace, S.W. Leaf, Mrs. C. J., Beechwood, Tunbridge Wells. Leaf, Herbert, Zhe Green, Marlborough. ft Leaf, Walter, Litt. D., (Council), 6, Sussex Place, Recon: Park, N.W. Legge, Miss, 3, Ked/e Riad: Oxford. Lecky, Mrs., 38, Onslow Gardens, S.W. Leeper, Alexander, Warden of Trinity College, Melbourne. Leichtenstein, Moritz, 46, Aurzol Road, West Kensington, W. Leigh, Rev. A. Austen, Provost of Kimg’s College, Cambridge. Leigh, W. Austen, 2, Vorfolk Crescent, Hyde Park, W. Lethbridge, Sir Roper, 36, Victoria Street, S.W. Lewis, Harry, 5, Argyll Road, Kensington, W.

XX1V

t+ Lewis, Mrs. 5. S., Castle-brae, Chesterton Road, Cambridge.

t+ Lewis, Prof. T. Hayter, 12, Kensington Gardens Square,W, Lewis-Poole, D., Royal Societies Club, 63, St, James? Street, S.W.

*Leycester, Mrs. Rafe, 6, Cheyne Walk, S.W. Liddell, Very Rev. H. G., D.D., The Wood House, Ascot, Berks. Lindley, Miss Julia, 10, A¢dbrook Terrace, Shooter's Hill Ra, S.E. Lindley, William, 10, Kzdbrook Terrace, Shooter’s Hill Και, S.E. Lingen, The Right Hon. Lord, K.C.B.,13, Wetherby Gardens, S.W. Lingen, Lady, 13, Wetherby Gardens, S.W.

Lister, Hon. Reginald, British Embassy, Constantinople.

Litchfield, R. B., 31, Kensington Square, W.

Lloyd, Miss A.M., Caythorpe Hall, Grantham.

Lloyd-Roberts, H., 1, Pump Court, Temple, E.C.

London, The Right Rev. the Lord Bishop of, The Palace’ Fulham, S.W.

tLock, Rev. W., Keble College, Oxford.

Lockyer, Sir Norman, K.C.B., F.R.S., 16, Pen-y-Wern Road, South Kensington, S.W. Long, Prof. Albert Limerick, Robert College, Constantinople.

ft Loring, William (Council), 2, Hare Court, Temple, E.C.

*Lubbock, The Right Hon. Sir John, Bart., M.P., High Elms, Hayes, Kent. Luce, Rev. E., 9, Royal Crescent, Brighton.

Ludlow, T. W., Cottage Lawn, Yonkers, New York. Lupton, J. M., The College, Marlborough.

Lupton, Miss M., 7 Zarl’s Terrace, Kensington, W. Luxmoore, H. E., Eton College, Windsor.

Lyttelton, Hon. and Rev. E., Hazleybury College, Hertfori.

Lythgoe, A. M., 15, Warland Buildings, Harvard College, Cambridge, Mass., U.S.A. Macan, R. W., University College, Oxford.

McAnally, H. W. W., War Office, Pall Mall, S.W.

McDaniel, J. H., Hobart College, Geneva, N.Y.

Macdonald, Miss Louisa, Women’s College, Sydney University, Sydney, N.S.W. Macdonell, W. R., LL.D., Coleat, Bycullah Park, Enfield.

MacEwen, Rev. Alex. Robertson, 25, Woodside Place, Glasgow. Macgillivray, J. Pittendrigh, Ravelstow Elms, Murrayfield, Edinburgh. McKechney, Mrs. W. C., 3, Berkeley Place, Wimbledon, S.W. Mackennal, Miss E. M., Beechwood, Bowdon, Cheshire.

Mackenzie, Duncan, British School, Athens.

MacLehose, James J., 61, St. Vincent Street, Glasgow.

Macmillan, Mrs. Alexander, Bramshott Chase, Shottermill, Surrey.

*Macmillan, George A. (Hon. Sec.), 29, Bedford St., Covent Garden, W.C. Macmillan, Mrs. George A., 19, Earls’ Terrace, Kensington, W.

+Macnaghten, Hugh, Eton College, Windsor.

Macnaghten, The Right Hon. Lord, 3, New Square, Lincoln’s Inn, W.C. McPherson, Miss Florence, Bank House, Maghull, Liverpool.

tMagrath, Rev. J. R., Provost of Queen's College, Oxford.

*Mahaffy, Rev. Prof. J. P., D.D., D.C.L., Trinity College, Dublin. . Maidstone, The Viscountess, Hurstmonceaux Place, Hailsham, Sussex. Manning, Percy, New College, Oxford.

Mano, Constantin, Balliol College, Oxford. Manos, Grégoire, Greek Legation, Vienna. Marchant, E. C., St. Paul’s School, West Kensington, W.

+ Marindin, G. E., Broomfield, Frensham, Farnham,

+ Marquand, Prof. Allan, Princeton College, New Fersey. Marshall, R., 31, Zhe Waldrons, Croydon.

Marshall, T., Highfield, Chapel Allerton, Leeds. Martin, Charles B., The College, Oberlin, Ohio, U.S.A.

+ Martin, R, B., M.P., το, Hzl/ Street, W.

+ Martyn, Edward, 7i/lyra Castle, Ardrahan, County Galway. Mason, H.C. F., Hatleybury College, Hertford.

Matheson, P. E., Mew College, Oxford. Mavrogordato, Pandeli, South Sea louse, Threadneedle St., E.C. Maynard, H. L., Junior School, Westward Ho, N. Devon.

XXV

Mayor, Rev. Prof. Joseph B. (Council), Queensgate House, Kingston Hill, Surrey. Mayor, R. G., Education Department, Whitehall, S.W. Merry, Rev. W. W., Rector of Lincoln College, Oxford. Milliet, P., 95, Boulevard St. Michel, Paris. Millington, Miss M. V., 27, Morland Square, W. Milne, J. Grafton, Mansfield House, Canning Town, E. Milner, His Excellency Sir Alfred, G.C.B., Government House, Capetown, 5, Africa Minet, Miss Julia, 18, Sussex Sguare, Hyde Park, W. Minns, E. H., Pembroke College, Cambridge. Mitchell, C. W., 195, Queen’s Gate, S.W. tMocatta, F. D., 9, Connaught Place, Edgware Road, W. Moline, Miss J. R., 172, Church Street, Stoke Newington, N. Monk, C. J., M.P., 5, Buckingham Gate, S.W. Monson, His Excellency the Right Hon. Sir E. J., G.C.B,, G.C.M.G., H.}.M, Ambassador, Parts. Monro, D.B. (V.P.), Provost of Oriel College, Oxford. Morgan, Miss Sarah,c/o Miss Colville, Overdale, Shortlands, Kent. Morice, Rev. F. D., 10 Azl/morton Road, Rugby. *Morley, The Right Hon. the Earl of, 31, Princes Gardens, S.W Morris, J. E., Zhe Grammar School, Bedford. Morrison, F. A. C., Jesus College, Cambridge. t Morshead, E. Ὁ. A., The College, Winchester. Moss, Rev. H. W., The School, Shrewsbury. Moule, C. W., Corpus Christi College, Cambridge. Moulton, Rev. W. F., D.D., The Leys, Cambridge. Mount, Rev. C. B., 14, Nerham Road, Oxford. tT Mount, J. T., Eton College, Windsor. Mudie, Mrs., Budleigh, Maresfield Gardens, Hampstead, ΜΟΙ. Munro, J. A. R. (Council), Lzzcoln College, Oxford. Murray, A. S., LL.D. (V.P.), British Museum, W.C. Murray, Prof. G. G. A. (Council), 5, The University, Glasgow. *+Mycrs, Ernest(Council), Brackenside, Chislehurst. tMyres, J. Linton (Council), Christ Church, Oxford. Naef, Conrad J., The Admiralty, S.W. Neil, R. A. (Council), Pemdroke College, Cambridge. Newman, W.L., Pittville Lawn, Cheltenham. Nicholson, Sir Charles,Bart., The Grange, Totteridge, Herts. O’Connor, Arthur, M.P., 5, Essex Court, Temple, E.C. Ohnefalsch-Richter, Dr. Max, 14, Wharfedale Street, Earl’s-Court, S.W. Ommanney, Admiral Sir Erasmus K., 29, Connaught Square, W. Ormiston, Miss F. M., Girls’ High School, Leeds. t+ Oxford, The Right Rev. the Lord Bishop of, Cuddesdon Palace, Wheatley, Oxon. Page, T. E., Charterhouse, Godalming. Pallis, Alexander, Zatoz, Sefton Park, Liverpool. Parker, Francis W., Cook County Normal School, Englewood, /ll., U.S.A. tParry, Rev. O. H., 5, Salem Hill, Sunderland. Parry, Rev. R. St. J., 7vintty College, Cambridge. Paton, James Morton, Wesleyan University, Middletown, Conn., U.S.A. Paton, W. R., British Pos Office, Smyrna. Pears, Edwin, 2, Rue de la Bangue, Constantinople. Peckover, Miss Alexandrina, Bank House, Wisbech. Peel, Hon. S. C., The Lodge, Sandy, Beds. Peers, C. R., Harrow Weald Vicarage, Stanmore, Middlesex. Peile, John, Litt.D., Master of Christ’s College, Cambridge. Pelham, Hon. Mrs. Arthur, Woorcroft, Monmouth. Pelham, Professor H. F. (V.P.), President of Trinity College, Oxford. Pember, E. H., Q.C., Vicars Hill, near Lymington, Hants. *Penrose, F. C., F.R.S.,(V.P.), Colebyfield, Wimbledon, S.W. Penrose, Miss Emily (Council), Royal Holloway College, Egham. *t Percival, F. W., 2, Southwick Place, Hyde Park Square, W. Perry, Prof. E. D., Columbia University, New York City, U.S.A. Philips, Mrs. Herbert, Sutton Oaks, Macclesfield.

Xxvl

Pickard, Miss Esther M., Overdale, Settle, Yorkshire. Pirie, Miss A. A., Sunnybank House, Old Aberdeen. Pirie, Miss Emily, Cozntesswell House, Aberdeenshire. f Platt, Prof. Arthur, 23, Powys Sguare, WW’. Pollock, Sir Frederick, Bart., 48, Great Cumberland Place, W. +Pond, Prof. C. A. M., University College, Auckland, New Zealand. Port, Dr. H., 48, /zusbury Square, E.C. Porter, Miss Sarah, Farmington, Connecticut, U.S.A. +Postgate, Prof. J. P., 77znzty College, Cambridge. Powell, Sir F. S., Bart., M.P., 1, Cambridge Square, Hyde Park, W, Powell, John U., S¢. Fohn’s College, Oxford. Poynter, Sir Edward J., P.R.A., 28, Albert Gate, S.W. Pretor, A., St. Catherine's Ci ollege, Cambridge. Prickard, A. O., New College, Oxford, Proctor, R. G. C., British Museum, W.C, Prothero, Prof. G. W., The University, Edinburgh. ¢ Pryor, Francis R., Woodfeld, Hatfield, Herts. Psychari, A., 25, Boulevard des Capucines, Paris. Radcliffe, W. W., Fonthill, East Grinstead, Sussex. Radford, Dr. W. T., S¢dmouth. tRaleigh, Miss Katherine A., Zerrick House, Tring. *Ralli, Pandeli, 17, Belgrave Square, S.W. + Ralli, Mrs. Stephen A., 32, Park Lane, VW’, f Ramsay, Prof. W. M. D.C.L. (V.P.), The University, Aberdeen. Rawlins, F. H., Eton College, Windsor. : Rawnsley, W. F., Parkhill, Lyndhurst, Hants. Reece, Miss Dora, 26, Bullingham Mansion, Pitt Street, Kensington, W. Reid, J. S., Litt.D., Cazus College, Cambridge. fReinach, Salomon, 31, Awe de Berlin, Paris. Rendall, Rev. F., 82, Phzlbeach Gardens, S.W. +Rendall, G. H., Litt.D. (Council), Charterhouse, Godalming. Renieri, M. Mario, A¢hens. Richards, Rev. G. C., British School, Athens. Richards, F., Azvugswood School, Bath. Richards, H., Wadham College, Oxford. Richmond, Sir W. B., K.C.B., R.A., Bevor Lodge, West End, Hammersmith, W. Ridgeway, Prof. W. (Council), Fen Ditton, Cambridge. Ridley, Sir Edward, 48, Lennox Gardens, S.W. Rigg, Herbert A., 12, Stanhope Place, Hyde Park, W. Robb, Mrs., 46, Rutland Gate, S.W. Robins, Miss Julia, 95, Mount Vernon Street, Boston, Mass., U.S.A. Roberts, Rev. E.S., Catus College, Cambridge. Roberts, Professor W. Rhys, Unversity College of North Wales, Bangor. Roberts, Herbert F., 836, Mew York Life Buildings, Kansas City Mi. U.S.A. Robertson, Charles, Redfern, Colinton Road, Edinburgh. Robertson, Rev. Archibald, A7zuzg’s College, Strand, W.C. Robinson, T. P. G., Ashfield, Rothsay Place, Bedford. Rochester, The Right Reyv.the Lord Bishop of, Bishop’s House, Kennington ParkRoaa,S.E. Rogers, Major-General, 14, St. Margaret’s Road, St. Leonards-on-Sea. Rome, W., Oxford Lodge, Wimbledon Common, S.W. 1 Rosebery, The Right Hon. the Earl of, K.G., 38, Berkeley Square, W. Rotton, J. F., 3, Boltons, West Brompton, S.W. Roundell, C. S., 16, Curzon Street, W. Rous, Lieut.-Colonel, Worstead House, Norwich. ftRouse, W. H. D., 4, Bilton Road, Rugby, Rubie, Rev. Alfred E., Zhe Royal Naval School, Eltham, S.E. Runtz, Ernest, 22, Moorgate Street, E.C. Rushbrooke, W. G., 13, Cathcart Hizil, Highgate, N. Rutherford, Rev. W. Gunion, LL.D. 19, Dean’s Vard, Westminster, S.W. Rylands, W. H., 6, Holland Park Terrace, W.

TRyle, Rev. Prof. H. E. D.D., Prestdent of Queens’ Col ‘ } “olle , Samuel, Mrs. Sylvester, 80, Onslow Gardens, S.W. ge, Cambridge.

XXvil

Samuelson, Sir B., Bart., 56, Princes Gate, S. Kensington, S.W. Sandbach, Miss R., 16, Draycott Place, Cadogan Gardens, S.W. Sandwith, T. B., C.B., 29, Bramham Gardens, Earl’s Court, SAW. tSandys, J. E., Litt.D.(V.P.), St. John’s College, Cambridge. tSandys, Mrs., AZerton House, Cambridge. tSavage-Armstrong, Prof. G. F’., Queen’s College, Cork. t*Sayce, Rev. Prof. A. H., LL.D. (V.P.), 23, Chepstow Villas, Bayswater, W. tScaramanga, A. P., 18, Barkston Gardens, S. Kensington, δ. ΗΖ, Schilizzi, John S., 6, Cromwe// Houses, S. Kensington, SW. Schultz, R. Weir (Council), 6, .Wandeville Place, W. Schuster, Ernest, 12, Harrington Gardens, S. W. Scot-Skirving, E., Zhe College, Cheltenham. Scouloudi, Stephanos, A¢hens, Greece. Scull, Miss Sara A., 1100, MZ. Street, N.W., Washington, D.C. Seaman, Owen, Zower House, West Hill, Putney, S.W. Seebohm, Hugh, 714 Hermitage, Hitchin. t Selwyn, Rev. E. C., School House, Uppingham. ft Sendall, Sir Walter J., K.C.M.G., Colonial Office, S.W. Seymour, Prof. Thomas D., Vale College, Newhaven, U.S.A. Shadwell, C. L., Ortel College, Oxford. Sharkey, J. A., Chrtst’s College, Cambridge. Sharpe, Miss, Harold House, Lansdowne Road, W. Shewan, Alexander, c/o AZessrs. IV’. Watson and Co., 27, Leadenhall Street, E.C. Shuckburgh, E. S., Granchester, Cambridge. Sidgwick, Arthur, Corpus Christi College, Oxford. Sidgwick, Prof. Henry, Litt.D., 7yiutty College, Cambridge. Sikes, Edward Ernest, .52, John's College, Cambridge. Simpson, H. B., 3, South Street, Park Lane, W. Sinclair, Captain H. Μ. R.E., Junior United Service Club, S.W. *Skrine, H. D., Claverton Manor, Bath. *Skrine, Rev. J. H., 7véntty College, Glenalmond, Perthshire. Smedley, J. F., 15, Loudon Road, Tonbridge. Smith, A. Hamilton (Council), 121, Bedford Court Mansions, Bedford Avenue, W.C. Smith, Cecil, LL.D.(Council), British Museum, W.C. Smith, Eustace S.,62, Banner Road, Victoria Park, E. Smith, H. Babington, C.S.1., Stokeleigh, Queen’s Road, Weybridge. tSmith, Prof. Goldwin, The Grange, Toronto, Canada. Smith, Prof. T. Roger, 7, Gordon Street, Gordon Square, W.C. Smith, R. Elsey, 7, Gordon Street, Gordon Square, W.C. Smith, Reginald J., Q.C., 11, Hyde Park Street, W. Smith, W. G., 5.2. John’s College, Oxford. Smith, F. E. J., 2, Tanfield Court, Inner Temple, E.C. Smith, J. A., Balliol College, Oxford. TSnow, T. C., St. Fohn’s College, Oxford. tSomerset, Arthur, Castle Goring, Worthing. Sonnenschein, Prof. E. A., Greenfield Cottage, Harborne, Birmingham. tSouthwell, The Right Rev. the Bishop of, Thurgarton Priory, Southwell. Spiers, Phené, Carlton Chambers, 12, Regent Street, W. Spooner, Rev. W. A., New College, Oxford. Spring-Rice, S. E., C.B., 1, Bryanston Place, Bryanston Square, W. Stannus, Hugh, 61, Larkhall Rise, Clapham, S.W. Stanton, Charles H., Field Place, Stroud, Gloucestershire. Statham, H. Heathcote, 40, Gower Street, W.C. Stawell, Miss Melian, Newnham College, Cambridge. Steele, Dr., 33, Via S. Gallo, Florence. Sterrett, J. R. Sitlington, Amhurst College, Amhurst, Mass., U.S.A. Stevenson, Miss E. C., 13, Randolph Crescent, Edinéurgh. Stewart, Prof. J. A., Christ Church, Oxford. Stillman, W. J., 50, Via S. Basilio, Rome. Stillwell, James, 1, Victoria Park, Dover. Stogdon, J., Harrow, N.W.

XXVIll

Stokes, Miss Margaret, Carriy Brene, Lowth, co. Dublin. Stone, E. W., Eton College, Windsor. Stone, Rev. E. D., Hillingdon, Uxbridge. Strachan-Davidson, J. L., Balliol College, Oxford. Stretton, Gilbert W., The College, Dulwich, S.E. Strong, Prof. S. Arthur. Strong, Mrs. S. Arthur, LL.D. (Council). Sturgis, Julian R.. Elvington, Eyethorne, Dover. Sturgis, Russell, 307, East 17th Street, New York. Sullivan, John, Reform Club, Pall Mall, S.W. Surr, Watson, 57, O/d Broad Street, E.C. Swanwick, Miss Anna, 23, Cumberland Terrace, N. W. Tait, C. J., Cathedral Yard, Exeter. + Tait, C. W. A., Clifton College, Bristol. Tancock, Rev. C. C., Leck, Kirkby Lonsdale. Tarbell, Prof. F. B., University of Chicago, Chicago, Ill., U.S.A. Tarring, C. J., Chief Justice of Grenada, West Indies. Tarn, W. W., 13,-Vorfolk Square, Hyde Park, W. Tatton, R. G., Passmore Edwards Settlement, Tavistock Place, W.C, | Taylor, Rev. Charles, D.D., Master of St. John’s College, Cambridge. Thomas, Rev. T. LI., Jesus Colleye, Oxford. *Thompson, Sir E. M., K.C.B., D.C.L. (V.P.), British Museum, W.C. Thompson, E. S., Christ's College, Cambridge. Thompson, F. E., The Cottage, Preshute, Marlborough. Thompson, Henry F. H., 35, Wimpole Street, W. Thompson, J. Eyre, 8, Stone Buildings, Lincoln's Inn, W.C. Thursfield, J. R., Fryth, Great Berkhamstead. Tilley, Arthur, King’s College, Cambridge. Tottenham, H. R., St. John’s College, Cambridge. *+Tozer, Rev. H. F. (V.P.), 18, Norham Gardens, Oxford. tTruell, H. P., F.R.C.S., Cloumannon, Ashford, Co. Wicklow. Tubbs, Prof. H. A., University College, Auckland, New Zealand. *+Tuckett, F. F., Frenchay, near Bristol. Tuckerman, Mrs. Mary F., 12, Jacopo da Diacceto, Florence. Tudeer, Dr. Emil, Helsingfors, Finland. {Turnbull, Mrs. Peveril, Sandy-Brook Hall, Ashbourne. Tylor, Prof, E. B., D.C.L., F.R.S., The Museum House, Oxford. Tyrrell, Prof. R. Y., Litt.D., D.C.L. (V.P.), Trinity College, Dublin. *Tyrwhitt, Rev. R. St. J., Ketilby, Oxford. Underhill, G. E., Magdalen College, Oxford. Unwin, T. Fisher, 10, Hereford Square, S.W. Upcott, L. E., The College, Marlborough. *Valetta, J. N., 16, Durham Terrace, Westbourne Park, W. tValieri, Octavius, 2, Kensington Park Gardens, W Vanderbyl, Mrs. Philip, 51, Porchester Terrace, W. Vardy, Rev. A. R., King Edwards School, Birmingham. t Vaughan, E. L., Eton College, Windsor. Venning, Miss Rosamond, 8, Balcombe Street, Dorset Square, N.W, Verrall, A. W., Litt.D., Trinity College, Cambridge. Verrall, Mrs. A. W., Selwyn Gardens, Cambridge. Vickers, Rev. W. V., Knowle Grange, Sidmouth. Vincent, Sir Edgar, K.C.M.G., Esher Place, Surrey. tVlasto, T. A., Bouevaine, Sefton Park, Liverpool. + Wackernagel, Prof. Jacob, Steinenberg, 5, Bale. t Wagner, Henry, 13, Half Moon Street, W. t Waldstein, Prof. Charles, Ph.D., Litt.D., King’s College, Cambridge. Walford, Mrs. Neville, 1, Ashburn Place, S.W.

Walker, Rev. F. A., D.D., Dun Mallard, Shootup Hill, B Walters, Henry Besuchara : 2 Hill, Brondesbury, N.W,

p (Council), British Museum, W.C.

Walters, W. C. Flamstead, 3, Selby Road, Anerley, S.E.

t Wantage, Lord, K.C.B., V.C., 2, Carlton Gardens, S.W. Ward, John, F.S.A., Lenoxvale, Belfast. :

ΧΧΙΧ

Ward, T. H., 25, Grosvenor Place, S. W. Warr, Prof. G. C., 16, Earl's Terrace, Kensington, W. Warre, Rev. Edmond, D.D., Eton College, Windsor. Warren, Col. G. E. Falkland C.M.G., 57, Cornwall Road, Westbourne Park, W. Warren, T. H., President of Magdalen College, Oxford. Warren, E. P., Lewes House, Lewes, Sussex. Waterhouse, Edwin, Fe/demore, near Dorking. Waterhouse, Miss M. E., 59, Edge Lane, Liverpool. Way, Rev. J. P., Zhe Hall, Rossall, Fleetwood. Wayte, Rev. W.(Council), 6, Onslow Square, S.W. ft Weber, F. P., M.D., 19, Harley Street, W. Weber, Herman, M.D., 10, Grosvenor Street, W. Wedd, N., Xing’s College, Cambridge. Weekes, G. A., Stdney Sussex College, Cambridge. Weir, Miss Edith, 4, Frognal, Hampstead, N.W. t Welldon, Rev. J. E. C., The School, Harrow, N.W. Weld-Blundell, Herbert, 7, Par& Place, St. James's, S.W. Wells, J., Wadham College, Oxford. Westlake, Prof. J., LL.D., Zhe River House, Chelsea Embankment, S.W, Westcott, Rev. F. B., School House, Sherborne. Wheeler, James R., Ph.D., Columbia College, New York City, U.S.A. t White, A. Cromwell, 3, Harcourt Buildings, Temple. White, Mrs. Andrew, c/o Prof. White, Cornell University, Ithaca, U.S.A. White, Prof.J. W., Cambridge, Massachusetts, U.S.A. White, J. N., Rockland, Waterford. White, John Forbes, LL.D., Craigstay, Dundee. Whitehead, R. R., Box 144, Santa Barbara, California, U.S.A. Whitehouse, F. Cope, 8, Cleveland Row, St. Fames’, S.W. Wickham, The Very Rev. E. C., The Deanery, Lincoln. Wilkins, Rev. George, 36, 7rinzty College, Dublin. Wilkins, Prof. A. S., LL.D., Litt.D., The Owens College, Manchester. Williamson, J. W., Lzmasol, Cyprus. Willson, S. B. Wynn, St. John’s College, Cambridge. Wilson, Donald, Wavertree, Beverley Road, Hull. Wilson, H. C. B., New College, Oxford. Wilson, Harry, 447, Oxford Street, W. Windley, Rev. H. C., St. Colomba’s, Sunderland. Winkworth, Mrs., Holly Lodge, Campden Hill, W. Wiseman, Rev. Henry John, C/z/ton College, Bristol. Wither, H. S. Wood, Rev. W.S., Ufford Rectory, Stamford. Woodhouse, W. J., Sedbergh, Yorkshire. + Woods, Rev. H. G., President of Trinity College, Oxford. +Wren, Walter, 7, Pow7s Sguare, W. Wright, Sir R. S., 14, St. James's Place, S.W. tWright, W. Aldis, Vice-Master, Trinity College, Cambridge. Wright, Prof., John Henry, Cambridge, Mass., U.S.A. + Wyndham, Rev. Francis M., St. Mary of the Angels, Westmoreland Road, Bayswater,W. +Wyse, W., Trinity College, Cambridge. Yates, Rev. S. A. Thompson, 43, Phzliimore Gardens, W. Yorke, V. W., Forthampton Court, Tewkesbury. *Young, Rev. Canon E. M., Rothbury, Northumberland. tT Yule, Miss Amy F., Zarradale House, Ross-shire, Scotland.

XXX

LIST OF SUBSCRIBERS TO THE JOURNAL OF HELLENIC SU TTES:

The University Library, Aderdeen. The University College of Wales, Aderystwith. The Amherst College Library, Amherst, Mass. The Andover Theological Seminary, Andover, Mass., U.S.A. The Peabody Institute, Baltimore, U.S.A. The Johns Hopkins Library, Baltemore. The Enoch Pratt Library, Baltimore. The Royal Museum Library, Berlin. The Royal Library, Berlin. The Central Free Library, Ratcliffe Place, Lirmingham (J. D. Mullins, Esq.) The Mason Science College, Lirmingham. The Public Library, Boston, U.S.A. The Boston Atheneum, Boston, U.S.A. The University Library, Breslau. The University College of South Wales, (γώ. The Library of Clifton College, C//ton, Bristo/. The University Library, California. The Harvard College Library, Cambridge, Mass., U.S.A. ft The University Library, Caméiridge. The Library of Trinity College, Cambridge. The Library of St. John’s College, Cambriage. The Library of King’s College, Caméridge. The Fitzwilliam Archaeological Museum, Cambridge. The Girton College Library, Cambridge. The University of Chicago Press, Chzcago, Illinois. The Lewis Institute, Chicago, [/linots. The University Library, Christiania, Norway. The Library of Canterbury College, Christchurch, N.Z. The Public Library, Crzcznnatz, U.S.A. The Adelbert College, Cleveland, Ohio, U.S.A. The University of Colorado, U.S.A. The University Library of State of Missouri, Ce/umbia, Missouri, U.S.A. The Public Library, De¢rozt. The Royal Museum of Casts, Dresden. The Library of Trinity College, Dudlin. The National Library of Ireland, Dudlin. The King’s Inns Library, Dudiin. The Royal Irish Academy, Dudlin. The University College, Dundee. The Durham Cathedral Library, Durham. tThe Advocates’ Library, Edinburgh. The University Library, Erlangen. The University Library, Frecburg im Baden, Germany. The Philologische Seminar, Gzessen. The University Library, Glasgow. The Ducal Library, Gotha (Dr. W. Pertsch). The University Library, Gottingen. The Royal University Library, Gretfswala. The Dartmouth College Library, Hanover, U.S.A. The University Library, Heédelberg (Dr. Zangmeister). The School Library, Harrow, NV. W. The Cornell University Library, /thaca, A. Y. The University Library, Jena. The Royal and University Library, Kénigsberg. The University of Kansas, Lawrence, U.S.A. The Leeds Library, Commercial Street, Leeds. The Public Library, Leeds. The Bibliothéque Universitaire, 3, Rue des Fleurs, Lille, Nora. The Free Library, Liverpool. The University College, Zzverfool. +The British Museum, W.C. The Department of Greek and Roman Antiquities, Brztish Museum, W.C. The Library of University College, London. The Athenaeum Club, Pal/ Mall, London, S.U. The Burlington Fine Arts Club, Savile Row, London, W. The Library of St. Paul’s School, Kensington, W. The London Library. St. James’s Sguare, London, S.W. The Reform Club, Pad/ Mall, London, S.W. The Royal Institution, Albemarle Street, W.

XXXi

The Library, Westminster School. S.W. The Oxford and Cambridge Club, Pall Mall, c/o Messrs. Harrison & Sons, 59, Pall Mall. The Foreign Architectural Book Society (T. H. Watson, Esq.),9, Nottingham Place, W, The Sion College Library, Victoria Embankment, EC The College Library, Dulwich, S.E. The City Library, Lowel/, Mass., U.S.A. The Bibliothéque Universitaire, Pa/azs Saint Pierre, Lyons, The Library of the University of Wisconsin, Madison, U.S.A. (E. Εν Riley, Esq.). The Whitworth Institute, J/anchester. The Chetham’s Library, Hunts Rank, Manchester. The Grammar School, Manchester. The Royal University Library, A/ardurg. The Public Library, A/e/éourne, Victoria (c/o Messrs. Melville, Mullen ἃ Co.). The Library of the University of Milan, A/7/an. The McGill University Library, Montreal (C. H. Gould, Esq.) The K6nigliche Paulinische Bibliothek, /umster, 1.11. The Royal Library, A/unich. The Archeological Seminary, Munich. The University of Nebraska, Vebraska, U.S.A. The Forbes Library, Morthampton, U.S.A. The University Library, A/zinster. The Free Public Library, Vewark, New Jersey. The Newberry Library, Vewderry, U.S.A. The Library of Yale College, Vewaven. The Free Public Library, Jersey Czty, New Jersey, U.S.A. The Astor Library, Mew York. The New York State Library, A/bany, New York. The Library of Columbia University, Wew York. The Hamilton College Library, C/inton, New York. The Metropolitan Museum of Art, Vew York. The Library of the College of the City of New York, New Yor. The Sachs Collegiate Institute, Vew York. Tt The Bodleian Library, Oxford. The Library of All Souls College, Oxford. The Library of Worcester College, Oxford. The Library of Balliol] College, Oxford. The Library of Christchurch, Ox/ord. The Library of Exeter College, Oxford. The Library of St. John’s College, Oxford. The Library of New College, Oxford The Library of Oriel College, Oxford. The Library of Queen’s College, Oxford. The Library of Trinity College, O2/ord. The Library of University College, Oxford. The Union Society, Oxford. The University Galleries, Oxford. The Bibliothéque de l’Institut de France, δ 725. The Bibliothéque de I’ Université de France, Parts. The Bibliothéque des Musées Nationaux, Parvs. The Bibliothéque Nationale de Paris, Pars. The Ecole Normale Supérieur, Paris. The Library Company, PAzladelphia. é' The Library of the University of Pennsylvania, Philadelphia, Pa., U.S.A. The Vassar Library, Poughkeepsie, N.Y. The Archaeological Seminary, Zhe University, Prague (Dr. Wilhelm Klein). The Bibliothéque de l’Université, Rennes. The Library of Brown University, Providence, Rhode Island, U.S.A. The American School of Classical Studies, Xume, /taly. The Rossall Library, Rossall, Fleetwood (the Rev. W. H. E. Worship). The School Reading Room, Rugby, care of Mr. A. J. Lawrence. The St Louis Mercantile Library, Sz. Louis, U.S.A. " The Royal Library, Stockho/m (Messrs. Samson & Wallin). The Archaeological Museum, 7he University, Strassburg (per Prof. Michaelis). The Imperial University and National Library, Strassburg. The Free Library, Sydney, New South Wales. The University Library, Syracuse, New York. The University Library, Zovonto. The Library of the University of Illinois, Urdana, [llinoi?s. The Boys’ Library, Eton College, Wznasor. The Library of Eton College, Windsor. The Public Library, Winterthur, (Dr. Imhoof. Blumer). The Free Library, Worcester, Mass., U.S.A. The Williams Collegé Library, W7//iamstown, Mass., U.S.A.

t Libraries claiming copies under the Copyright Act.

ΧΧΧΙΙ

LIST OF JOURNALS, &c., RECEIVED IN EXCHANGE FOR THE JOURNAL OF HELLENIC STUDIES.

American Journal of Archeology (Professor J. M. Paton, Middleton, Connecticut, U.S.A.)

Bulletin de Correspondance Hellénique (published by the French School at A¢hens).

Bullettino della Commissione Archeologica Comunale di Roma (Prof. Gatti, Museo Capitolino, Rome).

Ephemeris Archaiologike, Athens.

Siam sete of German Imperial Archaeological Institute, Corneliusstrasse No. 2 II.

erlin.

Journal of the Royal Institute of British Architects, 9, Conduzt Street, W.

Mélanges d’Histoire et d’Archéologie, published by the French School at Rome.

Mittheilungen of the German Imperial Archaeological Institute at Athens.

Mittheilungen of the German Imperial Archaeological Institute at Rome.

Mittheilungen and Abhandlungen of the Archaeolog. Epigraphisches Seminar of the University of Vienna.

Mnemosyne (care of Mr. E. J. Brill), Leiden, Holland.

Numismatic Chronicle, 22, A/éemarle Street.

Philologus. Zeitschrift fiir das klassische Altertum (cave of Dietrich’sche Verlags- Buchhandlung, Gottingen).

Praktika of the Athenian Archaeological Society.

Proceedings of the Hellenic Philological Syllogos, Constantinople.

Publications of the Imperial Archaeological Commission, S¢. Petersburg.

Revue Archéologique, Parzs (per M. Georges Perrot, 45, rue d’ Ulm).

Revue des Etudes Grecques, Publication Trimestrielle de l’Association pour |’En- couragement des Etudes Grecques en France, Parts.

Transactions of the American School, Athens.

Transactions of the Cambridge Philological Society and Journal of Philology,

SESSION 1896-97.

THE First General Meeting was held at 22, Albemarle Street, on November 2nd, 1896, Professor L. Campbell, V.P., in the chair.

Mr. A. J. Evans read a paper ‘On Further Discoveries of the Early Cretan Script.’ The additional materials described had been collected by the author during his recent researches in the island, and fully corroborated the evidence first brought forward by him two years since of the existence in Crete in prehistoric times of two interrelated systems of writing, one picto- graphic, the other linear. A fresh series of early seals was described, showing the evolution of purely pictorial types into a conventionalized pictographic script of Mycenaean date, having points of resemblance with the Hittite. Very primitive examples of seals with linear characters were also illustrated, and it was pointed out that this linear class in Crete, which presented some curious resemblances to Greek and Phoenician letters, went back, on the whole, to a distinctly earlier period than the conventionalized pictographic class, and might be largely described as pre-Mycenaean. Hitherto the evidence had mainly rested on seals and graffiti on vases. Mr. Evans was now able to describe the discovery in the Cave of Psychro— the ‘Diktaion Antron’ of Zeus—beneath a votive and sacrificial stratum of Mycenaean date, of part of a libation table of steatite, imitated from a twelfth dynasty Egyptian model, bearing the remaining half of what seemed to be a dedication in Cretan linear characters. The inscription consisted of nine letters with two punctuations, and was of the highest importance as showing that this pre-Phoenician script was applied to monumental as well as personal objects. The Egyptian affinities of the libation table itself fitted in with other signs of intimate connexion between Crete and the Egypt of the twelfth dynasty supplied by the decorative designs of sealstone and steatite vases. Here, however, in the imitation of an object of cult they had proof of a community so deep-lying that it could hardly have been due to mere commercial intercourse. It pointed to continuous land-contact in the population so influenced, and the probability became great that this and other vestiges of the influences of the old empire of Egypt in Crete were due to Libyan settlements in the island. If so, the beginnings of the Cretan linear script, which also seemed to show Egyptian influences, might be ultimately found in Tripoli. A remarkable parallelism was, in fact, shown between the Cretan signs and the early Libyan alphabets. Converging lines of evidence showed that the inscribed libation table from the Dictaean

4

XXXiV

cave could not be brought down later than about 2000 B.C. (/.H.S. vol. xvii. p. 327). Mr. Hogarth, Sir J. Evans, Prof. Ridgeway, and others took part in the discussion which followed.

The Second General Meeting was held at 22, Albemarle Street, on February 15th, 1897, Mr. Talfourd Ely, in the chair.

Prof. P. Gardner read two papers: (1) ‘Ona Stone Tripod at Oxford.’ The tripod was given to All Souls’ College by A. Lefroy in 1771. It was found at Corinth. It consists of a basis intended for the support of a large basin, probably meant to hold lustral water. There is a central column, around which stand back to back three draped female figures, each on a recumbent lion, and holding in one hand the tail of the lion. Froma comparison with a very similar tripod of which fragments were found at Olympia, it appears that this was a fixed type for vessels of the class. The date of the Oxford tripod was fixed by Prof. Gardner, from considerations of style, as the earlier half of the fifth century (/.4.S. vol. xvi. p. 275). A discussion followed, in which Prof. Waldstein, Prof. E. Gardner, and Miss Harrison took part. (2) ‘On the Mantinean Basis. This basis, bearing reliefs by a pupil of Praxiteles, was submitted by Prof. Gardner to a close examination. He maintained: (4) That the phrase in which Pausanias describes the basis should be read Μοῦσα καὶ Μαρσύας αὐλῶν, and must be regarded as referring only to one slab of the reliefs, which represents the conflict of Apollo and Marsyas. (ὁ) That the three slabs which we possess were the whole of the relief. We need not suppose a slab to have been lost, and it is quite possible that six Muses rather than nine were repre- sented. The group of Apollo and Marsyas would be in the midst, three Muses on each side as spectators, the whole occupying the front of the pedestals. (c) That the figures of Apollo, Leto, and Artemis which stood on the pedestal were not arranged as a group, but stood side by side, as they appear in the Praxitelean group copied on a late coin of Megara (7.1.5. vol. xvi. p. 280). In the discussion which followed, Prof. Waldstein argued that the proposed arrangement of the slabs was too asymmetrical for Greek art, and dwelt upon the difficulty of departing from the number of nine Muses, which was supported both by monumental and literary evidence. The practice of vase painters in varying the number was to be explained by artistic convenience, without regard to mythological considera- tions. Prof. Waldstein preferred to adhere to the arrangement of the slabs which he had himself publicly advocated, and which assumed that they had originally been four in number. Prof. E. Gardner, though pointing out some difficulties in detail, was on the whole inclined to accept the rearrange- ment proposed by Prof. Percy Gardner.

The Third General Meeting was held at 22, Albemarle Street, on April” 12th, “1897, Mr. FOC; Penrose, ΡΟ in’ tne ehait:

Miss Harrison read a paper on the Danaides. She contended that the origin of the Danaid myth had been misunderstood, especially as regards

XXXV

the supposed punishment of the water-carrying in leaky vessels. This was really no punishment at all, but simply the carrying on in Hades of their upper-world function as well-nymphs ; the pztios the Danaides had to fill was bored only at the bottom, as shown on ancient monuments, and it was a well cistern. The labour of well-filling was endless from the beginning, because Argos was πολυδίψιον. The idea of water-carrying in Hades as a penalty for the ἀμύητοι was also, she contended, not a mere later moralizing addition, but inherent in the primitive Danaid myth, and the leaky vessels pointed to an ‘ordeal by the sieve, such as was undergone, according to Pliny, by the vestal virgin Tuccia (‘ Nat. Hist., xxviii. 2, 3). The forty-nine guilty Danaides would fail in the ordeal and be proved as ἀμύητοι in the rites of Demeter Thesmophoros. Referring to Prof. Ridgeway’s recent paper in the He/lentc Journal on the Pelasgian origin of the ‘ objects called Mycenaean, Miss Harrison expressed her view that though the Olympian gods would be found, on analysis, to be part Hellenic, part Pelasgian, the remaining denizens of Hades would prove, like the Danaides, to be of Pelasgian origin.

Prof. Ernest Gardner read some notes on a vase in the museum at Chicago, which seemed to him to represent the myth of Athamas.

The paper announced by Prof. Gardner, on a vase in the museum at Harrow, was postponed to a subsequent meeting.

The Annual Meeting was held at 22, Albemarle Street, on July 5th, 1897, Prof. Jebb, M.P., President, in the chair.

The Hon. Secretary read the following report on behalf of the Council:—

The Session just ended presents no very striking features, but the work of the Society has been quietly and effectively carried on.

Two Parts of the Journal have appeared as usual, and the contents speak for themselves. There has, however, been a change of Editorship to which reference should here be made. Professor Percy Gardner, who has borne the chief burden of Editorship ever since the Journal was started, has found it necessary to retire from its active management, though he will remain a member of the Consultative Committee. It would be difficult to exaggerate the value of the services rendered to the Society by Professor Gardner during the seventeen years of his close connection with the Journal. Until five years ago, when the present Editorial Committee was constituted, the main duties of Editorship, corre- spondence, choice of papers, reading of proofs, were performed by Professor Gardner single handed. And even since Mr. Leaf and Mr. Arthur Smith were associated with him in the work, a full share of labour and responsibility has naturally fallen to the colleague whom long experience and wide knowledge have so eminently qualified to form a sound judgment on the matter in hand. Professor Gardner will carry with him in his retirement from these arduous duties the warm thanks of all members of the Society, and not least of those on the Council who best know how devoted his service has been,

ad 2

XXXVi

Before leaving the ‘subject of the Journal, it may be added that an Index is in preparation to Volumes IX.—XVI. inclusive, and to the Supplementary Papers.

The Society has suffered some serious losses by death during the past year, and especially during the last few months. Among those who have passed away may be mentioned Archbishop Benson ; Sir Wollaston Franks, who recently madea valuable gift to the Library ; Dr. Hubert Holden, who had for years been an active member of the Council and latterly a Vice-President, and had also filled the office of Hon. Librarian ; Mr. John B. Martin, who had most efficiently filled the office of Treasurer since 1888 ; and quite recently Mr. J. Theodore Bent, who had been a member of Council for many years, and whose death at a comparatively early age leaves a serious blank in the ranks of archaeological explorers.

The Council are glad to be able to announce that Mr. Douglas Fresh- field kindly consented to act as Treasurer after Mr. Martin’s death, and he is to-day formally nominated to the office. Mr. Stephen Spring Rice has consented to take Mr. Freshfield’s place as one of the Auditors.

It will be remembered that last year a change was made in the arrange- ments for the Library, Miss Johnson being appointed Assistant Librarian. The new arrangements have worked very well, and the Library is now in better order than ever before. Two months ago, however, Miss Johnson represented to the Council that she found it impossible, for the modest salary which the Society is in a position to pay, to give her whole time to the work. The question was very carefully considered by the Library Committee, and they recommended that in order, if possible, to retain Miss Johnson’s services, her hours of attendance should be reduced. These recommendations were accepted by the Council, and Miss Johnson consented to retain her post on the understanding that she is to attend from 2.30 to 5, on every day but Saturday. This arrangement is under- stood to be provisional, but the Council hope that it will on the whole be found convenient to members. With so small a Library, the Society can hardly hope to retain the exclusive services of an efficient librarian. Due provision being made for the custody of the books, the attendance of a librarian for a stated period on five days in the week (the usual holidays excepted) seems likely to serve all practical purposes.

It may be of interest to record that about seventy members have in the course of the year made use of the Library, either on the spot, or by borrowing books or lantern slides. During the lecture season the slides were in constant demand, so that this privilege of membership is evidently appreciated. A new Catalogue is about to be issued, including numerous recent additions [see p. liii.]. If members have slides to present, it would be an advantage if they could offer them before this Catalogue is printed. Additions would be most welcome in the department of views of architec- tural details (other than the Parthenon) and of sculpture. Of donations made to the Library in response to the appeal issued last year special mention is due to that received from the late Sir Wollaston Franks, who

XXXVii

presented twenty volumes of the Numismatic Chronicle, Falkener’s Ephesus and the Temple of Diana, and Winckelmann’s Monumenti Antichi Inediti. The Society has also acquired by purchase or exchange the following among other valuable works—the facsimile recently made in Florence of the Laurentian codex of Aeschylus, the two volumes of Collignon’s Histoire de la Sculpture grecque, the official record of the German Excavations at Olympia, and 7he Mycenaean Age by Tsountas and Manatt.

In the course of the Session, the annual grant of £100 to the British School at Athens has been renewed for a further period of three years. Although the School is on a more satisfactory financial basis than during the first nine years of its existence, it can still not afford to dispense with this grant in aid from the Hellenic Society, and the Council feel that there is no object to which the funds of the Society could more properly be devoted. Not only is the School the one institution which gives facilities to British students for original research on Greek soil, but the Society receives an adequate return for its subscription in the valuable articles contributed by members of the School to the Journal of Hellenic Studies. Members will be glad to learn that in spite of untoward circumstances in Greece the School has had a satisfactory season. The number of students has been considerably above the average, and it has been found possible to continue both in Athens and in Melos the excavations which were begun last year and of which some account has already appeared in the Journal. The results of the further work upon the site of Kynosarges, and in Melos, will as usual be reported to the Annual Meeting of Subscribers in July, but some preliminary information will be communicated to the Society to-day by the Director of the School.

The only other grants made by the Council during the past Session have been the sum of £50 to Mr. W. R. Paton in aid of some proposed explorations in Asia Minor, and of £30 to Mr. W. J. Woodhouse, for additional illustrations in a forthcoming work on Aetolia. Unfortunately the unsettled state of things in the East has prevented Mr. Paton from doing much at present in fulfilment of his object, but it is hoped that there may be some results to record in next year’s Report. The grant to Mr. Woodhouse is a somewhat new departure, such help from the Society’s funds having hitherto been given rather towards the collection of new material than to its publication when collected, except where such publication has been undertaken by the Society itself. Mr. Woodhouse’s researches in Aetolia were carried out while he was a student of the British School at Athens. The volume in which the results are recorded is to be published by the Delegates of the Clarendon Press, but they did not see their way to provide all the illustrations the author thought necessary. The Council were then approached with the suggestion that they should make a grant for additional illustrations, and after careful inquiry by a Committee the case was held to come within the scope of the Society’s work, and the grant was authorised.

Three General Meetings have as usual been held during the Session

XXXVili

at which papers have been read by Mr. Arthur Evans, Professor Percy Gardner, Miss Harrison, and Professor Ernest Gardner. The meetings have been well attended, and have as a rule been followed by good discussions, It having been pointed out that the fact of these meetings being held on Monday prevented schoolmasters from attending them, the Council have decided to revert to Thursday, the day on which meetings were held in the earlier years of the Society’s existence. It is believed that this day will be more convenient to a majority of members.

The Treasurer’s Accounts show the financial position of the Society to be satisfactory. Ordinary receipts during the year were £815 against £915 during the financial year 1895-96. The receipts from Subscriptions, including arrears, amount to 4623, against 4655. Life Compositions amount to #15, against £63, and receipts from Libraries and for the purchase of back volumes £127, against £116. The receipts for loan of Lantern Slides amount to 45, against 47, but other items of ordinary income show no change.

The ordinary expenditure for the year amounts to £617, against £621. Payments for Rent 480, Insurance £15, Salaries £52, and. Stationery, &c. #29, are practically the same as in the preceding year, but the cost of purchases for the Library shows £94 against £39. The cost of the Journal, Vol. XVI., has amounted to £345, against £394. The grants, as mentioned above, to the British School at Athens, to Mr. Paton and to Mr. Woodhouse, amounted to £180. The balance carried forward at the close of the year under review amounted to £360, against 4340 at the end of the previous financial year.

Since the entrance fee was imposed in January, 1894, about 490 have been received from this source, a very substantial addition to the Society’s income. :

Twenty-six new members have been elected during the year, while twenty-one have been lost by death or resignation. This shows a net increase of five, and brings the total number of members to 778.

Six new Libraries have joined the list of Subscribers, which now amounts to 133; or with the five Public Libraries to 138.

On the whole the Council feel that the Society, if not making any definite advance, is at least holding its own, and continues to do useful work. As stated earlier in the Report the losses by death of prominent members of Council have been during the last few months unusually severe, but other good men have been found to take their places, and the Council have no fear that the work of the Society in the future will be less efficient than in the past. The responsibility of management necessarily rests with the Council, and there is happily no sign that this body has lost the confidence of the members at large. But members should bear in mind that without their support at meetings and otherwise the Council would lack the needful stimulus and encouragement to further effort, and that in particular it rests mainly with the members at large to see that a due supply of candidates is forthcoming to fill up

ΧΧΧΙΧ

the inevitable gaps in the ranks, so that the Society, if it cannot extend its operations, may at least maintain them in undiminished efficiency.

The adoption of the Report was moved by the Chairman, who alluded to the recent discovery of papyri in Egypt, including some MSS. of Bacchylides. Prof. Jebb also expressed the sympathy of the Society with the present unhappy condition of Greece. The adoption of the Report was seconded by Sir John Evans, and carried unanimously.

The former President and Vice-Presidents were re-elected, the name of Prof. W. M. Ramsay being added to the latter. Prof. W. C. F. Anderson, the Rev. A. G. Bather, Mr. B. P. Grenfell, and Principal G. H. Rendall were elected to vacancies on the Council.

Mr. Cecil Smith, Director of the British School at Athens, gave a very interesting account of recent archaeological work in Greece, and especially of the excavations carried on by the British School on the site of Kynosarges in Athens, and at Phyllakopi in the island of Melos, where extensive remains had been found of an important pre-historic city.

The proceedings closed with the usual vote of thanks to the Auditors

and to the Chairman.

PROCEEDINGS OF THE CAMBRIDGE BRANCH OF ΤΕ ΤΠ ΠΡ.

SESSION 1806--7.

On Saturday, December 5th, 1806, ἃ meeting was held at the Chairman’s house.

Prof. Ridgeway read a paper on ‘ The Trident of Poseidon,’ in which he controverted the view put forward by Mr. H. B. Walters (/.H.S. vol. xiii. p. 13 ff.) that it was a development from a lotus-headed sceptre. He main- tained that Poseidon was a fisherman’s god, and was therefore equipped as a fisherman with the ordinary fishspear, or glaive. Other sea deities such as Triton and perhaps Palaemon were similarly furnished ; a semipiscine deity on the coins of Itanus in Crete is armed with a trident with which he is going to strike something beneath; Taras the eponymous of the Tarentines, a population living largely by fishing, is seen on coins striking at a fish with his trident. Aeschylus describes the trident as the ‘ fish- smiting engine’ of Poseidon. Τριοδοντία, ‘ fishing with a trident, is one of the recognized forms of seafishing in Plato’s Sophist and is also mentioned by Pollux. Both τρίαινα and τριόδους are used of the Trident of Poseidon, though τρίαινα is the word most commonly employed for it.

The fishspear with two or more prongs is one of the most universally distributed implements. In New Guinea and Polynesia they are made of as many as six pieces of barbed wood tied together. It is the common eelspear with five prongs of the Fens; the eelspear of Ireland is a trident ; such a spear is used for taking flat fish off the east coast of Scotland; a two-pronged spear is used for capturing lobsters on the coast of Devon. The trident is commonly employed at this very hour in the Mediterranean, as it was in Pliny’s time, who tells us in two passages that it was used for catching tunnies. Old Dictys, the kind fisherman of Seriphos, who found Danae and Perseus, was armed with his trident and net. When a fishing population went to war, they used their fishing gear for weapons, as rustics used their scythes, pitchforks, and bills. So Pittacus of Mitylene, when he challenged to single encounter the Athenian captain, entangled the latter in his fishing net and despatched him with his trident. The Roman gladiator called ‘retiarius’ was only a fisherman armed with a net and trident, for he said to his opponent, ‘non te peto, piscem peto; quid me fugis, o Galle?’ His opponent wore a fish in the front of his helmet. When then we find Poseidon using his trident as a lance even on horseback (see the coins of

ΧΙ

Potidaea) he is only doing what was the regular practice among the maritime populations of the Mediterranean.

(The subject will be treated at full length in Prof. Ridgeway’s forthcoming Early Age of Greece.)

Dr. Postgate exhibited two terra-cotta figures representing actors from the comic stage.

On Saturday, February 27th, 1897, a mecting was held at the Vice- Chairman’s house,

Mr. Adam read a paper ‘On some archaeological difficulties in Plato’s Republic’ The passages discussed were (1) iii. 398A ἀποπέμποιέν Te... στέψαντες, (2) iii. 399C (the παναρμόνιον), (3) iv. 439E παρὰ τῷ δημίῳ, (4) iv: 427C ἐπὶ τοῦ ὀμφαλοῦ καθήμενος. On the first of these passages he endeavoured to show that the current interpretation is right, as against the explanations offered by Ast and Mr. G. B. Hussey in the Proceedings of the Am. Philological Ass. Vol. 22 (1891) p. xliii. In discussing iii. 399C, Mr. Adam reviewed the evidence for the view that παναρμόνιον denotes a musical instrument, and argued that the word was always used to signify a certain form of musical composition, a sort of Panharmonic mode, in fact. On iv. 439E the reading of the manuscripts was defended against the conjectures of Valckenaer and Hemsterhuis. Leontius probably entered the city by the Μελιτίδες πύλαι, which were within a stone’s throw of the Barathrum. The executioner was standing by the dead bodies which he was about to throw into the pit. See Milchhofer Schrzftquellen etc. pp. i.-ii. Mr. Adam complained of Herwerden’s rashness in bracketing ἐν μέσῳ in ily. 427C, and expressed a doubt whether Herwerden knew what the ὀμφαλός really was. Plato is thinking of representations of Apollo in Greek art, for he is constantly depicted as seated on the ὀμφαλός. See Imhoof-Blumer and Professor Percy Gardner in ἢ H.S. viii. p. 18 and Middleton 2d. ix. 308, with Eur. Jon 5-6. Mr. Adam, however, professed himself unable to explain why Apollo should be seated on the ὀμφαλός when he prophesies. If the priestess sat there when delivering her oracles, it would be natural enough to identify her with the god, but we know that she sat on the tripod. Is it possible to suppose that she did occasionally occupy the ὀμφαλός, or that the tripod was fixed on the ὀμφαλός on some occasions? Pindar’s χρυσέων Διὸς αἰητῶν πάρεδρος (Pyth. 4, 4) might be adduced in support ‘of either view, for the eagles flanked the ὀμφαλός. No stress was laid on this conjecture.

Mr. A. B. Cook exhibited an impression from an early gold signet-ring, now in the Brit. Mus., representing a man with a wolf's head and tail stabbing a lion; the intaglio comes from the recent excavations in Cyprus and possibly illustrates a primitive wolf-cult.

On Tuesday, May 18th, 1897, a meeting was held at the Secretary’s

house. Dr. Postgate read a paper on ‘Cerberus and other polycephalous

xlii

monsters, in which he argued that the serious discrepancy in the number of heads attributed to Cerberus was to be explained by supposing that, when the dog was described as hundred-headed, etc., the reference was to the snake-heads which encircled his single, double or triple neck; while his heads proper were never conccived of as exceeding three in number. The same reference was to be seen in the hundred arms of the Giants, the multitudinous heads of Hydra, the fifty heads of Scylla, etc. The legend about the imitation of the νόμος πολυκέφαλος by Athene (Pindar), with many other references and expressions in Greek and Latin writers, derives new light from the observation that many-headedness imports snaky character or personality.

ΟΡ ‘L6g1 φηη yzbz ‘AOIN-ONINdS NAHdALS ‘aaanspeal “UuOFT ‘(TAIAHSAYA “M SVTONOG NTILAGI NHOL MAHLUVS ‘e170 γι puy pure ‘yoo βιϑ 84 pur sJaYyoNOA ay} YI 11 pareduoo ‘juNodoe 51|} pourMEXa aAvYy 9A

- --ο---- ---Ο--αἈ--- a I! ost 1% EL 6x GSEH ΚΡ τὰ τ L6gi ‘Kew 1515 ‘srayueg ye souryeg ΟἹ, -πἰκπππ = & αἱ onney fe) fe) ¢ <n eRe ene scans, Teer ee eee eee ee eeeeee seyoD “e ζ δὲ ἕξ oss αὺ ον ϑυώμου πονοῦν ος sisyurg ye aourleg “cc ae Ὁ ‘bsq Sqstaqry "ΔΑ “ΠΟΠΒΌΟΩ Areiqr] “ee Se, |) nn MEE Sy slg tae junosdy jeuimof jo souvjeg ‘‘ ΘΕᾺ Treseeeeeeeeceoses ἈΒΘΡΠΘ ΠΑΘΊΒΥΣΙ jouvoy ** o Oo ogr — ᾿ ΟΙ 61 τ΄ © Ὁ oor “"""" suayyy 18 JooyNS ysnug 91, ¢ 9 z Seeececscees L6g1 “ kew O 0 OG wre ttsensenessstecesctenceees uojieg Ἢ "ΔΑ ς 9 ς OS ae SORT ty *AO N—"jU99 jod € O O OF crrteresssseseseseseeees Qsnoypoo AA “f “AK wieysutyj0N jo uorjeiodi05 ςς ςς --ϑΊΖάθαο “cc g ς 61 seeebeoee L6g1 2 judy NE SSS SME oy, iene τς yurg sad ‘uorsstuu0s) “Ὁ Ste 61 Ὁ μα 6} ΠΘΟΩΙ Ἴ90---90ης 090 €-v 6z Sunuug Aipung pur ‘ase}sog ‘Arauorje}g “ἢ τοῦ $€ sayeqy πος MON UO ΞΡΌΘΡΙΔΙΩ “ 6 £ 6 ainjwin J pue syoog—junosoy Areiqry ‘* 9 ΟἹ QgzI ie? 7 28 4 9 O10 “ soUINJOA YOvg puvsivauy “Ὁ “ 0 0 Of 3 Ὁ L6g1 ‘key γ515 0 0 gzi “" L6gI—9g6gr ‘suonduosqns Aresqry ** 0} 1vah auo ‘Aivjamag εν. “Ὁ is “S| «CF 8. 989 ; re) oI ze ah weet eee eee L6g1 “(ἘΠῚ ys1f re) v v eee eee eee ee eeeee slvolly ςς ςς ΟἹ reat 900 ‘ueeiqry “yssy ‘sarees “" Onamwsn *"*"2" “" suonduosqns 911 big < © 0 SE —_—s_—_——CCiCids ttt t RSH ERE EEE eteleweten etter eens Qouvinsuy « ΟΙ6 919 ** L6g1—96g1 ‘suonduosqns ΘΙΘΑΌΙΘΙΝΙ ςς οο og ΠΤ 96g1 ὍΘΕ 0} teak auo uo kg II ΟῚ οες τ ρον OHO “(ἘΠῚ 515 ye aoureg oL pay ps F Zier Sept? “LNNODDV HSVO L 9 ovbF ; L 9 οὗν fe) ¢ ς ΠΡ ΞΞΞΦΞΞΞΦΦΦΦΦΠῆΠΠ ῸῸ uorlj}e[sueL “e ce L v τεῦ RR eee ee eee eee we eee eneee ee re jodeg “ce “ce oO gI τε ΟΞ Saucy pue Surmviq ce ‘ec I II 06 tee Saye[g “ce ce II 6 1lz (ase bv 6 SPE Berea meee e sees areesneee yunos0y 1580 0} ‘g0ueleq ee ae -18ὴ — Surjuug 11 pue Ἵ ‘sig ἿΛΧ TOA Aq | € Li ἢ ** g6g1 ὍΣ aunf 0} ‘S6g1 ‘1 ΔΠπ| Ἴξαχπο [10 saeg ΟἹ, ps

*L6g1 ‘AWIX ISIE OQNIGNA NVAA AHL ΟἹ LNNODOV «SAIGALS OINATTIAH AO TVNUNOL AHL,,

A comparison with the receipts and expenditure of the last ten years _

is furnished by the following tables :—

ANALYSIS OF ANNUAL ΚΕΓΕΙΡΤΟ FOR

Subscriptions) yes τούτο greene τι ὁ] A PRORESH 2.3. προ ΕΣ Cxes < Shas Life Compositions ............... Libraries and Back Vols. Dividends: -3#hso.d.cechrwceses Special Receipts ...... ......0000.

Laurentian MS.

Mr. D. G. Hogarth (Alex- andria Grant Refunded)

Loan of Lantern Slides ... ἘΠ ΟΠ... τε ce εν,

Royalty on Sales of Photo- PTRDNS oo τὰν ΟΕ sce

Donations—J. Vansittart, Esq.

E. H. Egerton, Esq.2....5 Library, Mrs. Cohen ...... W. Arkwright, Esq.

Balance from preceding year ...

THE YEARS ENDING :—

188s. | 1 895. ἄν ων: £ ee L 5 539 | 545 585 | 554! "564. 671 | 678 | 41 | 32 89. : 256i | tae] ‘oaahe ma 79| 47 79 | 126] 95 79 50 119 122 118 | 233 161 186 122 50} oy38 B5C ris? απ 891 ates lees 53 31 37

4 4 4 2 5 5 | 2 2 I | 100 | 5 ! | ; |

861 gto 898 976 «878 1,034 910 489 | 255 150 F255.) 29% | 259) 20am

1,350 | 1,165 1,049 |1,231 1, 117 | 1,293

3: τὴ 31) Mey 3 May, 31 May, 3: May, 31 May, (3 May, 31 M: ay. 31 sone : May, Ϊ 1891. 92. 1803 : |

& & 645 | 617 9 4. 63 15 117 126 43. | 43} 30 ΒᾺΝ 7 5: a I | 915 816 | 169 | 340 |

1,124 E 084 | 1,156

ANALYSIS OF ANNUAL EXPENDITURE FOR THE YEARS ENDING :—

31: Mays 31 May, 31: τ 31 Μαν,18 31 may, 31: May, 31 May,| 31 May, 80 sor 31 Bay, | 1889. jo. | 1891. | 1892. 37 1894. 1895. 897. | sai PN ; ἐς Ξε εν 3 & & & 4 & Ae x Rents dc tice nae otacteeneanncstees | 15 30 | 30 50 ΣΙ 80 80 80 INSUMANCE: ........Ὁλύν ον ΟἿΣ εἶ τον τος 3 5 | Il II II ri | ag 15 15 Galanviesi., 5683... τὸ eves ensceans canoes 46 39 39 | 44 49 49 49 47 52 Library τ Ξ νι τς, ΟΞ ὡς 41 15 | 16 8 41 75 : 96 39 94 Stationery, Printing,and Postage 54 61 62 41 71 49 49 46 29 Cost of Journal (less sales)...... 583 |} *873 440 |. 610 | 532] 475 | 441 | 394] 346 Grants 2 0 wike.c.0v.c'60 0.900.000 00 dlep θυ»... 9 350 100 150 Ϊ 125 100 185 225 100 180 IMYASEMENtS) cee τορι τ tae eee 46 | 100 158 | EgyptExplorationFund—t1, 100 copies of Mr. Hogarth’s Report 23 LoapgRepaid ........2.4%.2..>.. Photo Enlargements, Albums, Lantern Slides, &c. 18 4 .| SundFies: νὰ... ἕο δι, cys 3 : I ,095 1,123 : 794. 992 | 858 1,079) 955 | 744 796 Balance sca ceceasccr ease tress ore 255 42 255 239 | 259 214 169 | 340] 360 1,350 | 1,165 1,049 | 1,231 [1,117 | 1,293 | 1,124 | 1,084 | 1,156

* Includes cost of reprinting of Vols. IV. and V. (= £437) less the amount received from sales. 1 The grant of £100 to the School at Athens has been paid since the accounts were made up: see Cash Account.

LIST OF BOOKS AND PAMPHLETS

ADDED TO THE

LIBRARY OF THE SOCIETY FOR THE PROMOTION OF HELLENIC STUDIES

1897.

Anderson (W. C. F.). Journey from Mount Athos to the Hebrus. 8vo. Sheffield. 1897. (Presented.)

Antiquités du Bosphore Cimmérien. See Reinach (S.)

Apostolides (B.). Essai sur |’Hellénisme Egyptien, et ses Rapports avec |’Hellénisme classique et 1|’Hellénisme moderne. Vol. I. 1. 8vo. Paris. 1898. (Presented.)

Aristotle. Politics. Revised text, with introduction, analysis, and commentary. By F. Susemihl and R. Ὁ. Hicks. Books I-V. 8vo. London. 1894. (Presented.)

Aristotle. Constitution of Athens. Ed. F. G. Kenyon. See British Museum.

Arrian. Anabasis. Ed. Καὶ. Abicht. 8vo. Leipzig. 1871. (Presented.)

Bacchylides. Poems. Ed. F.G. Kenyon. 8vo. 1897.

Beloch (J.). Griechische Geschichte. Vol. II. 8vo. Strasburg. 1897.

Beulé (C. E.). L’Acropole d’Athénes. 8vo. Paris. 1862. (Pre- sented.)

Boston Museum of Fine Arts. Catalogue of Casts. Pt. III. Greek and Roman Sculpture, By E. Robinson. Revised Edition. 8vo. Boston. 1896. (Presented.)

British Museum PuBLicaTIONs.

The following works have been presented by the Trustees of the British Museum. (The Library also contains several Museum publications, which have been already reported.)

Department of Coins and Medals.

Catalogue of Greek Coins in the British Museum. 8vo, Italy. By R.S. Poole. 1873.

xlvi

Tauric Chersonese...Thrace, &c. By B. V. Head and P. Gardner. 1877. Seleucid Kings of Syria. By P. Gardner. 1878. Macedonia. By B. V. Head. 1879. Thessaly to Aetolia. By P. Gardner. 1883. Ptolemaic Kings of Egypt. By R.S. Poole. 1883. Central Greece. By B. V. Head. 1884. Crete and the Aegean Islands. By W. Wroth. 1886. Attica, Megaris, Aegina. By B. V. Head. 1888. Lycia, Pisidia, and Pamphylia. By G. F. Hill. 1897. Caria, Cos, Rhodes, &c. By B. V. Head. 1897. Catalogue of Indian Coins, Greek and Scythic Kings of Bactria and India. By P. Gardner and R.S8. Poole. 8vo. 1886. Guide to the Principal Coins of the Ancients from circ. B.c. 700 to a.D. 1, with 70 plates. By B. V. Head. 8vo. 1889. Guide to the Principal Coins of the Ancients from cire. B.c. 700 to a.D. 1. (Pamphlet.) Fourth Edition. 8vo. 1895. Roman Medallions. By H. A. Grueber and R. 8. Poole. 4to. 1874.

Department of Greek and Roman Antiquities.

Catalogue of Engraved Gems. By A. H. Smith. 8vo. 1888. Catalogue of Greek and Etruscan Vases. 2 Vols. 8vo, 1851-70. Collection of Ancient Greek Inscriptions. I. Attica. By E. L. Hicks. 1874. II. Greece, etc. By C. T. Newton. 1883. ΠῚ. Priene, Iasos, and Ephesus. By Εἰ. L. Hicks. 1890. IV. 1. Knidos, Halikarnassos, and Branchidae. By G. Hirschfeld. 1893. Description of the Collection of Ancient Marbles. Parts I.-XI. 4to. 1812-1861. Description of the Collection of Ancient Terracottas. By Taylor Combe. 4to. 1810. Guide to the Mausoleum Room. 8vo. 1886. White Athenian Vases. By A. S. Murray and A. H. Smith. Folio. 1896.

Department of Manuscripts.

Aristotle. Constitution of Athens. Ed. F. G. Kenyon. ὅνο. 1892.

Catalogue of Ancient Manuscripts. Part I., Greek. Part II., Latin. Folio. 1881-4.

Catalogue of Greek Papyri with Texts. Ed. F. G. Kenyon. 4to. 1893.

Classical Texts from the Papyri, including the newly-discovered Poems of Herodas. Ed. F.G. Kenyon. 4to. 1891.

Description of the Greek Papyri. Part I. 4to. London, 1839,

Greek Papyri. Facsimiles. Folio. London, 1893,

xl vii

Department of Printed Books. Excerpts from the General Catalogue of Printed Books—

Aeschylus, 1883. Horatius. 1885. Aesop. 1883. Ptolemaeus. 1895. Aristotle. 1885. Virgil. 1882. Homer. 1890.

Broéndsted (P. O.). Reisen u. Untersuchungen in Griechenland. 4to. Paris. 1826.

Burlington Fine Arts Club. Catalogue of Objects of Greek Ceramic Art exhibited in 1888. 4to. London. 1888.

Cesnola (L. P. di). Cyprus. Its Ancient Cities, Tombs and Temples. 8vo. London. 1877.

Chipiez (C.). See Perrot (G.).

Collignon (M.). Histoire de la Sculpture Grecque. 2 Vols. 4to. Paris. 1892-1897.

Conze (A.). Melische Thongefiisse. Folio. Leipsic. 1862.

Curtius (E.) and J. A. Kaupert. Karten von Attika, IX.1. 4to Berlin. 1897.

Curtius (E.) and F. Adler (edd.). Olympia. See Olympia.

Curtius (E.). Historische u. Philologische Aufsitze zu seinem 70" Geburtstage gewidmet. 8vo. Berlin. 1884. Presented.) Daremberg (Ch.) and K. Saglio. . Dictionnaire des Antiquités Grecques et Romaines d’aprés les Textes et les Monuments, 23 and 24

(—Jmage). 4to, Paris. 1897.

Dilettanti Society. Unedited Antiquities of Attica, comprising the Architectural Remains of Eleusis, Rhamnus, Sunium and Thoricus. Second Edition. Folio. London, 1833.

Dilettanti Society. Specimens of Antient Sculpture. 2 Vols. Folio. London. 1809-1835.

Doerpfeld (W.) and E. Reisch. Das Griechische Theater. Beitrige zur Geschichte des Dionysos-Theaters in Athen u. anderer Griechischer Theater. 4to. Athens and Leipsic. 1896.

Englefield Vases, drawn and engraved by H. Moses. 4to. London. 1848.

Fleming and Tibbins. English and French, and French and English Dictionary. 2 vols. 4to. Paris. 1845-1846.

Froehner (W.). La Collection Tyszkiewicz. Plates XXXIII.-XL. Folio, Munich. 1897.

Furtwaengler (A.). Intermezzi. Kunstgeschichtliche Studien. 4to. Leipsic. 1896.

Gaertringen (Hiller v.). Die Archaische Kultur der Insel Thera. 8vo. Berlin. 1897. (Presented.)

Gardner (E. A.). A Handbook of Greek Sculpture. 2 Parts. London. 1896-1897.

Gardner (P.). Sculptured Tombs of Hellas. 8vo. London. 1896.

Gori (A. F.). Museum Florentinum. See Reinach (S.), Pierres

gravées,

xl viii

Harrison (J. ἘΝ) and D. 5. MacColl. Greek Vase Paintings. Folio. London. 1894.

Helbig (W.). Das Homerische Epos aus den Denkmilern erliutert. 8vo. Leipsic. 1884. (Presented.)

Hermann (K. F.). Lehrbuch ἃ. griechischen Antiquititen. Vol. I. Sixth Edition. Vol. II. Sixth Edition. 8vo. Heidelberg. 1889 and 1892.

Herodotus. Ed. K. Abicht. 2 vols. Third Edition. 8vo. Leipsic. 1874-1883. (Presented.)

Hill (G. F.). Sources for Greek History between the Persian and Peloponnesian Wars. 8vo. Oxford. 1897. (Presented.) Homer. Opera et Reliquiae. Ed. D. B. Monro. 12mo. Oxford. 1896.

(Presented.)

Hope (T.). Costume of the Ancients. 2 vols. 8vo. London. 1841. (Presented.)

Hultsch (F.). Griechische ἃ. Rémische Metrologie. 8vo. Berlin. 1862. (Presented.)

Index in Tragicos Graecos. 2 vols. 4to. Cambridge. 1880, (Pre- sented.)

Inwood (H. W.). Erechtheion at Athens. Folio. London. 1831.

Jahn (0.). Die Ficoronische Cista. 4to. Leipsic. 1852.

Jahn (0.). Uber Darstellungen d. Handwerks ἃ. Handelsverkehrs auf antiken Wandgemilden. 8vo. Leipsic. 1868.

Jannaris (A. N.). Historical Greek Grammar. 8vo. London. 1897.

Kavvadias (P.). Athens, National Museum. Τλύπτα τοῦ ’EO. ’Apy. Μουσείου κατάλογος περιγράφικος. 8vo. Athens. 1890-1892. King (C. W.). Antique Gems, their origin, uses and value as interpreters of ancient history and ... . art. Second Edition.

Royal 8vo. London. 1866.

King (C. W.). Handbook of Engraved Gems. 8vo. London. 1866.

King (C. W.). Natural History of Precious Stones and Gems, and of the Precious Metals. 8vo. London. 1865.

Klein (W.). Praxiteles. ὅνο. Leipsic. 1898.

La Chau et Le Blond. Pierres gravées du... Due d’Orléans. See Reinach (8.).

La Ville de Mirmont (H. de). Le Navire Argo et la Science Nautique d’Apollonios de Rhodes. 8vo. Paris. 1895. (Presented.) Lévesque de Gravelle. Recueil de Pierres gravées. See Reinach (S.). Liddell (H. G.) and R. Scott. Greek-English Lexicon. 4to. Oxford.

1897.

Louvre Museum. Catalogue des Vases antiques. I. Les Origines by E. Pottier. 8vo. Paris. 1896.

Louvre Museum. Vases antiques du Louvre. Salles A—-E. by E. Pottier. 4to. Paris. 1897.

Louvre Museum. Catalogue sommaire des Marbres antiques by E. Michon. 8vo. Paris. 1895.

Mariette (P. I.). Traité des Pierres gravées. See Reinach (8,).

Marlborough Gems. See Reinach (8.).

xlix

Millin (A. L.). Peintures des Vases antiques. See Reinach (S.).

Millin (A. L.). Pierres gravées inédites. See Reinach (8.).

Millingen (J.). Peintures antiques de Vases Grees tirées de diverses Collections. Folio. Rome. 1813.

Millingen (J.), Peintures antiques de Vases Grees., See Reinach (8.).

Mueller (1. v.). Handbuch des klasaischen Altertumswissenschaft. Vols. I1f.3; V.2; VI. Atlas. ὅνο, ete. Munich. 1897.

Murray (A.8.). Handbook of Greek Archeology. 8vo. London. 1892.

Nicole (I.). Le Laboureur de Ménandre. 8vo. Geneva. 1898.

Olympia. Die Ergebnisse der von dem deutschen Reich veranstalteten Ausgrabung. Edd. E. Curtius and F. Adler.

Textb. I. Topographie und Geschichte. 4to. Berlin. 1897.

Textb. 11. Tafelb. I., 11. Die Bau-denkmaeler. 4to and folio. 1892-6.

Textb. III. Tafelb. 111. Bildwerke in Stein und Thon. 4to and folio. 1894-7.

Textb. IV. Tafelb. ΓΝ. Die Bronzen. 4to and folio. 1890.

Textb. V. Die Inschriften. 4to. 1896.

Portfolio of Maps and Plans. Large folio. 1897.

Overbeck (J.). Die Antiken Schriftquellen zur Geschichte d. bil- denden Kiinste bei den Griechen. 8vo. Leipsic. 1868.

Oxford. Ashmolean Museum. Catalogue of Greek Vases, by P. Gardner. 4to. Oxford. 1893.

Pausanias. Description of Greece. Translated, with a Commentary, by J.G. Frazer. 6 Vols. 8vo. 1898.

Penrose (F. C.). On the Results of an Examination of the Orienta- tions of a Number of Greek Temples. Supplement, (Pamphlet). 4to. London. 1897. (Presented.)

Perrot (G.) and R. de Lasteyrie. Fondation Piot. Monuments et Mémoires publiés par l’Académie des Inscriptions et Belles- lettres. Vol. IV. 1. 4to. Paris. 1897.

Perrot (G.) and Ὁ, Chipiez. Histoire de l’Art dans ]’Antiquité. Vol. VII. fase. 346-8.

Perry (W.C.). Greek and Roman Sculpture. 8vo. London. 1882.

Plutarch. Moralia. Ed. G. N. Bernardakis. Vol. VII. 8vo. Leipsic. 1896.

Ramsay (W. M.). Cities and Bishoprics of Phrygia. Vol. I. Pt. 2. West and Central Phrygia. 8vo. Oxford. 1897. (Presented.)

Reinach (S,). Bibliothéque des Monuments figurés Grecs et Romains. II. Peintures de Vases antiques recueillies par Millin (1808) et

Millingen (1813). 4to. Paris. 1891. T1I. Antiquités du Bosphore Cimmérien (1854). 4to. Paris.

1892. IV. Pierres gravées des Collections Marlborough, ete. 4to. Paris. 1895.

Reinach (S.). Répertoire de la Statuaire, I. Clarac de Poche. &8vo. Paris. 1897.

Robert (K.). Marathonschlacht in der Poikile u. weiteres iiber Polygnot. 4to. Halle-a-S. 1895. (Presented.)

l

Roscher (W. H.). Ausfiihrliches Lexicon d. Griechischen u. Roémischen Mythologie, Nos. 34-36 (—Myton). 8vo. Leipsic. 1897.

Schildt (A.). Giebelgruppen v. Aegina. 8vo. Leipsic. 1895.

Schliemann (H.). Mycenae. 8vo. London. 1878.

Schliemann (H.). Tiryns. 8vo. London. 1886.

Sellers (E.) and K. Jex-Blake. The Elder Pliny’s Chapters on the History of Art. 8vo. London. 1896.

Stosch (P. von). Gemmae antiquae celatae. See Reinach (S.).

Torr (C.). Interpretation of Greek Music. (Pamphlet.) Oxford. 1896. (Presented.)

Treu (G.). Hermes mit dem Dionysosknaben. Folio. Berlin, 1878.

Tsountas (C. T.) and J. Irving Manatt. Mycenaean Age. ὅνο. London. 1897.

Tyszkiewicz Collection. See Froehner.

Vienna, Austrian Museum. Sammlung der antiken Vasen u. Terra- cotten. Katalog τι. historische Einleitung von K. Masner. 4to. Vienna. 1892.

Vienna. Choix des Pierres graveés. By Eckhel. See Reinach (S.).

Woodhouse (W. J.). AXtolia, its Geography, Topography and Anti- quities. Royal 8vo. Oxford. 1897. (Presented.)

Wordsworth (C.). Greece, Pictorial, Descriptive and Historical. New Edition. Revised, with Illustrations and a History of the Characteristics of Greek Art, by G. Scharf. 4to. London. 1859. (Presented.)

Xenophon. Translated by H. G. Dakyns. Vols. II., III. 8vo. London. 1892 and 1897. (Presented.)

Yarborough Collection. Catalogue of Antiquities in the collection of the Earl of Yarborough at Brocklesby Park. By A. H. Smith. 8vo. London. 1897. (Presented.)

A LIST OF THE PERIODICAL PUBLICATIONS IN THE SOCIETY’S LIBRARY, Dec. 31, 1897.

American Journal of Archaeology. I—XI. 3. (1896.)

American Journal of Philology. XIV.—XVIII. 1. (1897.)

Annali dell’ Instituto Archeologico. LII.—LVII. (1885.) End.

Annuaire de |’ Association des Etudes Grecques. XV.—XXI. (1887.)

End.

Annuaire de la Société Francaise de Numismatique. 1896. 9, 10.

Annual of the British School at Athens. I.—II. 1895—1896.

Antike Denkmaeler des Archaeologischen Instituts. I.—II. 2.

Archaeological Institute of America. Reports I.—X VII. (1880—96.) Papers of Institute; American Series. I.—V.; Classical Series. I.; Papers of American School at Athens. I.—V.

Archaeologische Zeitung. XXXVJIJT.—XLIII, (1885.) End.

li

Athenaion. I.—X. (1881.) End.

Berliner Philologische Wochenschrift. XI., XVII. (1897.

Berliner Studien. I.—XI. (1890.)

Bulletin de Correspondance Hellénique. I.—XXI. 8, (1897.)

Bullettino della Commissione Archeologica Comunale. XXYV. 2, 3. (1897.)

Bullettino dell’ Instituto di Corrispondenza Archeologica. 1880—1885, End.

Bursian’s Jahresbericht iiber die Fortschritte d. classischen Alter- tumswissenschaft. I.—XXV. (1897.)

Byzantinische Zeitschrift. I.—VI. (1897.)

Cambridge Philological Society. Transactions I.—III. (1893); Proceedings I.—XLII. (1896.)

Classical Review. I.—XI. (1897.)

Commission Impériale Archéologique. Compte Rendu. 1878—9 and 1882—8; Atlas 1878—1888; Russian continuation viz.: ‘ Materials,” Nos. 4—20 (1890—96) and “ Reports ’’ for 1889—1894 (1892—1896). For General Index, 1859—1881, see Reinach’s Bibl. des Monuments, III., p. 145.

Deltion of the Historical and Ethnographical Society of Greece. I.—V. 18. (1896.)

Egypt Exploration Fund, Reports. 1895, 1896.

Ephemeris Archaiologike. Third Series. 1884—1897. 2.

Goteborgs Hogskolas Arsskrift. 1895, 1896.

Hellenikos Philologikos Syllogos (of Constantinople). IV.—xXVI. (1871—1885.) XX.—XXV. (1891—1895.)

Hermes. XXVII.—XXXII. (1897.)

Institute (Royal) of British Architects. Proceedings, N.S. I1.—IX. (V. Imperf.) (1886—1893). Transactions, 1880—1892. Journal, 3rd Series. I.—V.1. (1898.)

Jahrbuch d. Kais. Deutsch. Arch. Inst. I.—XII. 3. (1897.)

Journal of Classical and Sacred Philology. 1854—1857.

Journal of Hellenic Studies. I.—XVII.1. (1897.) (Two copies.)

Journal of Philology. J.—XXV. (1897.)

Journal of the Royal Asiatic Society. XII.—XIV. (1883.)

Mélanges d’Archéologie et d'Histoire. I—XVII. (1897.)

Mittheilungen d. Kais. Deutsch. Arch. Instituts. Athenische Abth. I.—XXII. 3. (1897.)

- Mittheilungen d. Kais. Deutsch. Arch. Instituts. Rémische Abth, I.—XII, 2. . (1897.)

Mittheilungen (Arch.-Epigr.) aus Oesterreich-Ungarn. XVII.—XIX, (1897.)

Mnemosyne. I.—XXV. (1897.)

Monumenti Inediti dell’ Instituto Archeologico. XI. pl. 13—XII. (1885.) End.

Monuments Grecs. I.—IIJ. 22, (1895.)

Neue Philologische Rundschau. XII.—XVII. (1897.)

Numismatic Chronicle. Ist Series. 1836 and 1848—54.° New Series, Vols. I.—XX. Third Series. I.—XVII. (1897.)

δ 2

ΠῚ

Parnassos (Philologikos Syllogos). Vols, I.—V., VI.—X. (Iimperf.), and XI., XII. (1888). Epeteris I. 1897.

Philistor. I.—IV. (1863.)

Philologus, Neue Folge. 47—56. (1897.)

Praktika of the Athenian Archaeological Society. 1873—187.

Revue Archéologique. 2nd Series. I.—XXXIII., XLI., XLIIT.— XLIV. (XXXIV., XLII. imperf.). 3rd Series. I.—XXXI. (1897.\ (III. imperf.).

Revue de Bibliographie Analytique. 1840, 1841.

Revue des Etudes Grecques. I.—X. (1897.)

Revue de Philologie. XX., XXI. (1897:)

Rheinisches Museum fiir Philologie. XULVII.—-LII. (1897.

Wochenschrift fiir Klassische Philologie. XI.—-XIV. (1897.)

LOAN COLLECTION OF LANTERN SLIDES.

THE collection, here catalogued, of lantern slides illustrative of the subjects coming within the purview of the Society, has been formed, by the kind- ness of members and others, and by purchase, for the purpose of loans to members under the following conditions. Members of the Teachers’ Guild are also admitted to the same privileges in return for a corresponding concession (see below p. Ixxviii.). The control of the collection is vested in the Library Committee.

bo

REGULATIONS FOR USE OF SLIDES.

The slides shall be lent only to members of the Society or members of the Teachers’ Guild desiring to use them for the purposes of demonstration.

Those members who have presented slides to the Society shall have a right to the free loan of two slides for every slide thus presented.

For the loan of slides beyond this number, and for loans to members who have not presented slides, a charge of 3d. for each slide shall be made.

All applications must be made to the Assistant Librarian, Hellenic Society, at 22 Albemarle Street. If desired, slides will be packed and forwarded to any address within the United Kingdom at the risk and cost of the borrowers.

The sum of half-a-crown must be paid for any slide broken while at the risk of the borrowers.

The slides may be kept for a period not exceeding fourteen days. If for exceptional reasons it is required to keep them fora longer period, special application must’ be made to the Library Committee. Slides required at a particular date may be booked for not more than three months in advance, on payment of the fee of 3d. per slide for the loan (except in the case of those who have presented slides as already provided).

If the Slides are returned within three days the charge will be reduced from 3d. per slide to 2d.

31 December, 1897.

liv

CATALOGUE OF SLIDES.

THE Magic Lantern slides in the Society’s collection are catalogued in the fullowing order, the letters prefixed being those which distinguish the various serles :—

TOPOGRAPHY.

Athens.

Attica.

Northern Greece. Peloponnese. Islands, etc. Cyprus.

Ao Ob Pp

Each of the above sections is sub-divided as follows :— (a) Maps and Plans, (b) General Views, (c) Architectural Views and Details, (4) Byzantine Buildings &c.

P. THe PARTHENON.

S. ScuLprure, including Reliefs, Terracottas, etc.

Ὁ

. Archaic period. Reliefs and Statues. b. Fine and Later periods. Reliefs. c. Fine and Later periods. Statues and works in the round.

ν. VAsEs:

a. Prehistoric, Mycenaean and other early wares.

b. Black-figured Vases, arranged according to subjects.

c. Red-figured Vases, and other later wares, arranged according to subjects.

Ia. INSCRIPTIONS, M. MISCELLANEOUS SUBJECTS.

a. Mycenaean and early periods. Ὁ. Later periods.

lv

The slides in the topographical classes are mainly from negatives taken by members of the Hellenic Society. A few have been taken, by permission from the photographs of the German Archeological Institute.

Those in classes P and § are for the most part taken from the originals, but in some cases from engravings, etc. In the case of sculpture, slides marked with * have been taken by photographic methods from the originals ; if marked { they have been derived from casts. If not thus distinguished they have been taken from drawings and engravings.

In class V, most. of the slides are derived from published illustrations. Where there is ἃ choice of publications, reference is made by preference to that which was used for making the slide, except when it is difficult of access.

The following is a list of the principal contractions employed :— 'OA.M. Mittheilungen des Arch. Inst. Athenische Abtheilung.

A.Z. Archiologische Zeitung. B.C.H. Bulletin de Correspondance Hellénique. B.D), Baumeister, Denkmdler.

B.M. British Museum.

Conze. Conze, Die Attischen Grabrelie/s.

Gardner. E. A. Gardner, 4 Handbook of Greck Sculpture. G.A.V. Gerhard, Auserlesene Vasenbilder.

Eb. Overbeck, Gallerie Heroischer Bildwerke.

J.H.S. Journal of Hellenic Studies.

Jahrbuch. Jahrbuch des K. Deutschen Arch. Instituts.

MI. onumenti inediti dell’ Inst. Arch.

Mich. Michaelis, Der Parthenon.

Myce. Schliemann, Mycene.

P. Prisse d’Avennes, Hist. de l Art Egyptien, 1863. R.& C. Rayet and Collignon, Hist. de la Céramique grecque. Schuch. Schuchhardt, Schliemann’s Excavations (Eng. Tr.). W.V. Wiener Vorlegeblatter.

&25> Members ordering slides are requested to be careful to quote the class letters (Aa, etc.) as well as the numbers.

TOPOGRAPHY. Athens—Maps and Plans.

Aa 1. Plan of Athens. 2. Plan of Acropolis. aA ᾿ (Harrison and Verrall, p. 343). . Sections __,, (Jahn and Michaelis). . Plan of Propylaea. » 4, Dionysiac Theatre.

o> Or "αὶ OO

Abl. 2.

τ Sod ee

8.

9. 10. at. 19. 12a. 13. 14, 15. 16.

bo

lvi Atheus—Generul Views.

Athens and the Piraeus. Athens from Pnyx. Panorama No. 1. ἣν Ἶ + », 2(Odeum of Herodes).

» τῷ “ » 9 (Acropolis). »» + i , 4 (Lycabettus). » » > » ὃ (Theseum).

3 _ + ΠΟ Acropolis from E.

» and Olympieum from S.E.

» toEK. and SE.

» and Theseum from N.W.

» from S.W.

᾿ » ».W., with Frankish Tower.

. » road Κα. of Zappeion.

is » foot of Areopagus.

: » Lycabettus.

τ » Church of Bombardier.

» and Theseum from Railway.

. Acropolis Restored.

Propylaea and Cave of Pan from N.

. Theseum from Prison of Socrates.

» and Modern Town.

. Areopagus from Gate of Acropolis.

a and Grotto of Eumenides.

”

. Olympieum from Acropolis. > Payx, . Street of Tombs.

ΕΣ)

. Observatory.

. Callirrhoe.

. Colonus and the Cephisus.

.- Demeter Euchloos.

. Mount Hymettus from road to Observatory.

5 ς » American School. » Pentelicus.

. Salamis and Psyttalea over Piraeus. . Tatoi from Acropolis.

Athens— Architecture, etc.

(See also the Parthenon.)

. Acropolis, W. front, entrance.

τε W. front, from the approach.

»

”

᾽)

. Nike Apte

”

lvil

. Acropolis, Bastion of Odysseus. S. side (Turkish period), A.J/, ii. pl. 2.

Walls, with Old Columns.

. Propylaea, W. front.

N.W. Hall (Pinacotheca).

S.E. Hall (unfinished). N.E. Hall (unfinished). from top of Parthenon. Pyrrhus Inscription. ros from N.E.

. Ge Karpophoros Inscription,

. Theseum.

. Asclepieum. from E.

᾽ν)

})

. Erechtheum from N.W.

and Parthenon from N.W,

and Old Temple from top of Parthenon.

and part of Old Temple from §, N. Porch.

N. Door.

Portico of Caryatides.

”

3)

Excavation on Ν. side. (Stuart and Revett). Ornaments.

2) . Theatre of Dionysus.

2)

᾽)

front view.

from Acropolis.

from 8.W.

from N.E.

with the Two Temples. Auditorium from E.

Stage from E.

Stage of Phaedrus from EK, Remains of Stages.

Stage with Old Orchestra, Altar.

Priest's Chair.

. Monument of Thrasyllus (present state).

(Stuart and Revett).

Boundary Stone.

lviii

Ac49, Asclepieum Retaining Wall of Theatre. 50. - Gate of Well. 51. " Interior of Well-House (from sketches). 52. Monument of Lysicrates. 53. Tower of the Winds. 54. Basilica of Hadrian, W. end. 55. ε δ᾽ ὅθ. Pnyx, Bema. 57. Corinthian Capital. 58. Altar of Dionysus (in Limnae?) A.M. xxi., pl. 9, fig. 1 Πρ πὰ α View. 60. Ancient Greek Wine-Press.

»”»

Athens—Byzantine.

Ad 1. Small Metropolis, S. side. ag hen a from S.E.

3. Asomaton Monastery from British School. 4. 5. Theodore.

Ba. Attica—Maps and Plans—none. Attica—General Views.

Bb1. Piraeus Panorama 1.

2. 22) ” 2. 3. Pore » 3. 4. Aegina from Old Phalerum, —

5. Bay of Phalerum.

6. Eetionea.

7. Zea, Piraeus and Salamis.

8. Straits of Salamis.

9. Eleusis and Salamis.

10. Marathon, from N.E., with Pentelicus.

11. oe the Mound: Lg. = from Vrana. 13. ” from the Mound. 14. - from the S. road.

15. Phyle, Fortress Walls. 16. ,, View over Attica to S.

1%. §,,° Power:

18. ,, Entrance.

19. Icaria, Dionyso, the Cave. 20. ,, Rapendosa Cave.

21. ,, Cliff, View towards Marathon.

lix Attica—Architecture, cle.

Be 1. Eleusis, Sekos from N.W. angle looking E. a § Sekos from N.W. looking 5.10,

3 » » » >) is

4, a " Be:

Ὡς ᾿ " Substructure.

6 » * View to S.E.

i, , Precincts of Pluto from N.

8. δ ᾿ ΟΝ ὭΣ

9. » Details of Appius Pulcher’s Gate and Capital. 10. Icaria, Ruined Church. ae... Ἢ » pulled down.

12. ,, Acroterion from Byzantine Church. 13. Sunium from N.E.

Attica—Byzantine. Bd 1. Church of Omorphi.

Northern Greece—Maps and Plans.

Ca. Map of Greece.

Northern Greece—General Views.

Cb 1. Delphi.

2. τ 3. Delphi from Cirrhean Plain. 4, Cirrhean Plain from Delphi. 5. Delphi. Phaedriadae showing Temple. 6. From Delphi looking E. 7. Plain of Boeotia from Kokla (Plataea). 8. Plataea from N. 9. Scironian Cliffs.

10. Euripus from N.

Northern Greece—Architecture.

Cc 1. Portico of Athenians.

Northern Greece—Byzantine.

Cd 1. Meteora, Monastery of Barlaam. 2: " ἊΝ » Metamorphosis. 3. St. Luke, Stiris and Parnassus. 4. Megara, the Easter Dance. 5

Ix

Peloponnese—Maps and Plans.

Dal. Plan of Mycenae (Schuchhardt), 2. ,,_ ,, Tiryns (Schliemann),

3. Megaron of Tiryns (Schuchhardt).

4+. Homeric House, Plan (P. Gardner),

5. Epidaurus, Plan of Hieron.

6. τ: » » Lheatre.

Peloponnese—General Views. Db 1. Corinth.

2. a Canal.

3. Acro-Corinthus, Old Fortifications.

4, View from.

5. Nauplia, General View.

6. " View of, from Tiryns.

Ἑ i Harbour and Island.

8. Mycene, General View.

9. Epidaurus, Distant View of Theatre. 10. Olympia, before Excavation, from a Print. 11. ᾿ View with Cladeus.

12. a Panorama 1.

13. a - 2.

14, 3 5 3.

15. Megalopolis, Mound from N.W.

16. Ἢ Theatre from W. wing. τὰ - Looking across Scena. 18. * Excavations.

19. sf At Work.

20. ᾿ Wheeling away Earth. 21. 5 A Barrow Load.

22. Ἢ Our Street.

23. ‘a Priests.

24. τ Peasant Women.

25 Market Place.

26. Greek Ploughs.

97. ᾿ Washerwomen,

28. Holiday Dress of Workmen, 29. Asea ἢ rancovrysi), Site of Acropolis. | ae » Distant View.

a: . ‘ Walls of Acropolis. 32. ‘ Acropolis.

33. Hysiac (Achladocampo).

34. ,» Bit of Wall.

35. Mantinea, river Ophidi, near.

Ixi

Db36. Sparta Taygetus, S. View.

37. A N. View. 38. Gytheam from Steamer. 39. Langada Pass, View from Summit. 40. oe ἣ ᾿ 5 δ. Mt. Rindomuo. 41. Patras. 42 » another View. 43. Vostitza, View in. 44. a Currant Factory. 45. = View. 46. δ: Xs Peloponnese—Architecture, ete. De 1. Corinth. 2. Mycene, from Treasury of Atreus. 3. ss Wall. 4. ᾿ Lion Gate. 6." bs ¥ another View. 6, ‘3 Postern Gate and N. Wall. 7. Pa Gallery leading down to Well in N. Wall. 8. r Stone Circle. 9. τ Ms Ε 10. : Walls of Palace. bl. 5 Palace Staircase. 12. 3 Treasury of Atreus. 13. i Dromos of Treasury of Atreus. 14. » Interior of if ἢ 15. 4 Restoration of Capital ἔνα, Treasury of Atreus. (Puchstein, Das Ionische Cap., tig. 42.) 16. ds Profiles of Capitals (Middleton, J.7/.S., vii., p. 163). ἘΣ Mrs. Schliemann’s Treasury. 18. Tixyne from W. 19. , Great Portal on E. side. 20. » N. Tower of E. Gate. 21. » Ramp to E. Entrance. 22. ,, ‘Sally Port and Staircase in W. Wall. 23. » N. Wall and Postern. 24. » Ν᾿. Gallery. 25. , E. Gallery. 26. Restoration of Hall. 27. Hpidaurna Theatre. 28. ss 4 from E. 29. . ᾿ κι a 90, ἡ Fs 3 Stage. 51 i » Orchestra and W, Parodus, 32, 3 Capital from Tholos,

Ixii

Dc33. Epidaurus Cyclopean Bridge near. 34. Olympia, Temple of Hera.

35. Ἢ 7 ”

36. - ns , and Cronos Hill. 37. > Ne » from Gymnasium. 38. ἐ " Zeus from Heraeum. 39. ἣν es » Restored.

40. Bassae, Temple from S.E.

41, z - “ΟΝ.

42, 3 τὸ N. end.

43. ᾿ - Κ᾽ end:

44, ὡς » from N.W.

48. Ὲ ᾿ » YH

46. " 6 Interior.

47. Υ another View.

48. Lycosura, Temple of Despoena. 49. Ithome, Walls of Epaminondas.

Peloponnese—Byzantine,

Dd 1. Mistra, Church of “ Zoddochos Pege.” 2. Ithome, Catholicon Monastery.

Islands, ete—Maps and Plans.

Ea 1. Map of Troas (Schliemann Troja). 2. Plan of Hissarlik (E. Mayer, after Dorpfeld).

Islands, ete.—General Views.

Eb 1. Delos, Mt. Cynthus from Lake. 2. Delos, Lake of Leto. 3. Ithaca from Cephalonia. 4. Hissarlik from Plain.

Islands, etc.—Architecture.

Ec 1. Delos, Ruins of Temple of Apollo.

. Delos, Cynthian Cave Temple from Roman House.

. Delos, J Ὰ M near View.

Η Precinct of Isis.

. Aegina, Temple of Athena from below.

᾿ be >. » hear View.

. Paestum, Temple of Posidon.

» » ‘5 from 8.W.

᾿ " 5 » ».E., Basilica behind,

DOTA oP ὦ pH

10. 11. 12.

Fa 1.

xiii

Pergamum, Reconstruction of Acropolis. » " Great Altar. Corfu, Restoration of Capital of Xenfares (Puchstein, Das Ionische Cap., fig. 29.)

Islands ete. Byzantine.

Cyprus—Map. Map of Cyprus.

Cyprus— Views.

. Village of Suskiu and Valley N.E. of Kuklia (Old Paphos). . Valley W. of Kuklia.

. The Village Mosque, Kuklia.

. Threshing Floor, Kuklia.

. Village of Kulklia.

. Pay-Day, Kuklia.

. Gorge near Aschelia.

. Summit of Mt. Troddos, and Summer Encampment. . View from above Village of Amargetti.

. A Street in Nicosia.

. Episcopi from W. .

. Parapedhia and Cyprus Company’s Wine Factory.

. Curium, Acropolis and Site of Excavations, 1895,

_ Workmen, 1893. a Staff and Workmen, 1895.

Cyprus—Architecture.

. Monoliths by the Sea, Old Paphos. . Old Paphos, S.W. Angle Blocks; S. Wing or Tomb of Cinyras,

Temple of Aphrodite. ¥ Part of S. Wing. Κὶ » S. Wing from 8. Porch. 7 Digging in 5. Wing. s Central Court ; Breakfast. sf Clearing S. Porch. ἐς S. Porch, W. End. - S. Porch, S.E. Angle. ᾿ S. Porch from §.E. Angle. f N. Wall, W. End. Cesnola’s N.W. Angle Block. δὲ S. Chamber from E. Inscribed Pedestal. E. Entrance from N. From House, E. Part.

μι Θ Ὁ οὦὐ -1ὺ ι ιΚ WN μὰ

tpt

Ixiv Cyprius—Byzantine and Gothic. . Nikosia, St. Sophia.

τ = Interior. é Desecrated Church.

. Leondari Vouno, Crusaders’ Fortress from 8. . Famagusta, Cathedral.

another View. Front.

E. End.

S. Side. Chantry Door.

SE: Wicholae ΘΙ Catia Limasol.

. Colossi, Castle of Knights Templars.

. Bellapais, the Cloister.

. Aschelia, Carved Wooden Church Screen.

τ Baldachin. - Pulpit. > Rood, ete.

THE PARTHENON. The Building.

. Parthenon from N.E.

. Parthenon from N.W.

. Parthenon from §.E.

. Parthenon. Interior, looking West. . Parthenon. Plan.

. Parthenon. Sectional view of Εἰ. end restored (Niemann). . Diagram, shewing positions of sculptures. . Substructure, E. end of S. side.

. Steps on N. side, shewing curvature.

. Capital of column (B.M.).

. Unfinished drums.

Athene Parthenos.

. The Varvakeion copy.* Side view. » ™* Front view (Gardner, fig. 52). The ΕΠ copy + (Gardner, fig. 53). . Head of Parthenos on Athenian coin.

The Metopes.

. Metope.* Centaur and Lapith, Mich. 111. 2, B.M. 305. noe” ᾿ ἢ , Mich. in. 3, B.M. 806. ” μὰ 7) 7) 2} Mich. ill. 4, B.M. 307,

Ixv

P19. Metope.* Centaur and Lapith Mich. iii, 7, B.M. 810. 20. me ‘ is » Mich. iii, 26, B.M. 315. 21. ΤΥ Ἢ ᾿ » Mich. iii, 27, Β.Μ. 316, 22. mu - is » Mich. iii. 28, B.M, 317. 28: ta of” sa J i, Mich. iv. 32, B.M. 321.

The East Pediment.

24, East Pediment. Carrey’s drawing.

a A b South end (View in Elgin Room),

an Ἢ (Carrey’ s drawing).

W's, if North oad (View in Elgin Room).

ae <i » » (Carrey’s drawing).

2; 4, ν᾽, 1% yf aod, avi pl, 8.

a i Ἰ Horses of Helios, Theseus.*

ak Ὁ ᾿ Three Fates.*

32. ‘5 Selene + and horse.*

33. Birth of Athene, on Madrid Puteal. (Schneider, Geburt dev Athena pl. 1).

. The West Pediment. 34. West Pediment. (Carrey’s drawing).

35. ,» North end (View in Elgin Room). 36. sa, 2 North end and centre (Carrey’s drawing, facsimile) τς, ᾿ South end (Carrey’s drawing, facsimile). pe. | 2) J Llissos.* _ a Ὁ Cecrops and daughter.t 40. , . Central Fragments (A. J. xvi. pl. 8). The Frieze

41. Diagram shewing order of Panathenaic procession. 42. East frieze. Hermes to Ares.* Mich. xiv. 24-27, ae ε Zeus, Hera and Iris.* Mich. xiv. 28-31.

4. ε Head of Iris.t

45. ᾿ς, 7 Central group,* Mich. xiv. 32-35.

46. , 7 Athene and Hephaestos.* Mich. xiv. 36, 37.

ἈΠ. — 53 Poseidon, Dionysos, Demeter.t Mich. xiv. 38-40.

48, rf Aphrodite, Eros, Elders.* + Mich. xiv. 41-46. 49. Αἱ pe Maidens.+ Mich. xiv. 49-56.

50. North frieze. Cattle.f Mich. xii. 3-6.

aa 7 Sheep.t Mich. xii. 8-12.

bz. ls * Pitcher Carriers.+ Mich. xii. 13, 16-19. ae a Chariot group.*+ Mich. xii. 45-47. | - Chariot group.t Mich. xii. 54-58.

55. ys } Horsemen.* Mich. xiii. 110-114.

Ixvi

P56. North frieze. Horsemen.* Mich. xiii. 115-118.

57.

< ᾽ Youths and horses,* Mich. xiii. 180-134.

58. West frieze. Horsemen.* Mich. ix. 2, 3.

59. 60. 61. 62. 63.

- "μὰ oo tb =

DO ἢ -τ' δὴ Ob ὡϑ de

es 2 Horseman. Youth.f Mich. ix. 11, 12.

iF + Horse and man.t Mich. ix. 15.

wn Horse. Youths.t Mich. ix. 22-24.

Γ Ἂ * Photographed in situ.

Γ ᾿ = ie sabes (Inferior to last.)

SCULPTURE.— Archaic.

. Phrygian Lion Tomb. J.H.S8. ix. p. 368. . Spartan Stelé. Helbig, Hom. Epos,’ p. 217. . Selinus. Metope* from first Temple. Perseus and Medusa. . Head of Triton.* Acropolis. . Typhon.* Pedimental figure. Acropolis. *

’ ”

. Statue of Chares. Branchidae. B.M. 14. . Hera of Samos,f Niké of Archermos,} and dedication of Nikandra.t+ . Figure dedicated by Nikandra.*

*

2)

. Archaic female figure* from the Acropolis.

. Niké of Archermos.*

. Bronze Head.* Acropolis. E. A. Gardner, fig. 43. 14.

Figure carrying calf.* Acropolis.

. Head of ‘Antenor’ figure.t Head of Harmodius,t Jahvo. ii. pl 10. . Archaic female statue* (with fruit). Acropolis. Rhomaides, pl. 9.

ἊΣ: τ᾿ ss (Acropolis). Gardner, fig. 28.

iy » ™ Acropolis (two views). Rhomaides, pls. 7. 8. » » »» ‘ = 19 (back view).

»» 5, , ™* Acropolis.

»» ᾿ » ™* Acropolis.

»» fs » ™ (upper half). Acropolis.

. Archaic female heads.* Acropolis. . Archaic female figure on pedestal, with inscr. of Antenor. Acropolis,

Jahrb. ii. p. 141.

3. Head of archaic female figure* (two views). Acropolis. Gardner,

fig. 31. 27. Head of Ephebos.* Acropolis. 28, Spartan Relief.t A.M, i. pl. 22. 29. 55 " A.M. ii. pl. 20. 30. " Me J.ELS, v. p. 123. 31. +

2. Harpy Tomb, West Side.* B.M. 9+. 1.

Apollo of Tenea.* . Bronze Chorus.* Olympia, Bronzen, pl. 16.

. Bronze statuette* from Ligourio. Gardner, fig. 39,

. Aristogiton.* Naples. 4

. Harmodios and Aristogiton.* Naples. Gardner, fig. 35.

. Dying Warrior.* Εἰ. Pediment, Aegina. Gardner, fig. 42.

. Figure bending forward.* E. Pediment, Aegina. Gardner, fig. 41. . Central group.* W. Pediment, Aegina. Gardner, fig. 40.

. The Naxian Colossus.*

Ixvii

. Harpy Tomb, North Side.* B.M. 94. 2,

» South Side.* BM. 94. 4,

; Πρ οϊαυ Relief.

. Stelé of Aristion,* Stelé of Alxenor.*

. Steld of Aristion.*

. Stelé of Alxenor.*

. Warrior’s Stelé, etc.: from Ikaria.*

. Selinus. Metope* from second Temple. Europa on Bull. . Selinus. Metope from later Temple. Zeus and Hera.

. Apollo Ptoos.*

» "(= 42).

Youthful male figure from Sanctuary of Apollo Ptoos.*

” ” ” ”» ” » (= 44).

SCULPTURE.—Reliefs of Fine and Later periods. [See also the section on

Sb 1.

the Parthenon].

Temple of Zeus, Olympia. Metope.* Heracles and Bull. Gardner, fig. 48. a ᾿ ” Metope.* Heracles and Atlas. Gardner, fig. 49.

ss ᾿ Mi Athene* from Metope of Heracles and Augean Stable. " # feos Head of Athene* from Metope of Heracles and Lion. ᾿ Μὴ ἐν Metope.* Heracles and Geryon. τ tas” Gs » ™ Heracles and Augean Stable. »» ἪΡ » ™ Heracles and Eurystheus. +9 ἜΤΙ ,» ™ Heracles and Mares of Diomede. » ™ Heracles and Cerberus.

. Temple at Phigaleia. Metopes* and frieze*. B.M. . Temple of Niké Apteros. Frieze.* ΒΜ.

Balustrade. Victories with Bull.* Victory loosing sandal.+

» ” ”

. Attic Grave Relief * of tes, mer fig. 93.

* Conze i. pl. 78.

* of Pamphile and Demetria. Conze i. pl. 110.

᾿ » “of Mynnion. Conze ii. pl. 176.

* Man and Woman.

* Hegilla and Philagros. Conze i. pl. 105.

* Girl with doll, bird, and dog. Conze pl. 157,

f2

Ixvili

21. Attic Marble Lekythos* of Aristonike. Conze ii. no, 456. . Attic Grave Relief * of Menekrateia and Meneas. Conze i. pl. 50.

ε ἢ , ™of Ariphrades. Conze pl. 139. } , » ἢ of Phaidimos. Conze ii. pl. 187. 3 ‘ i Sof Nake. " Ἧ » *“ofSelino. Conze i. pl. 76. =. sy » ™*of Mys and Meles. Brueckner, Griech. Grabrel. p. 12, L. τῇ τ » *of Aristomache. Conze ii. pl. 15+. * of Corallion. Conze i. pl. 98.

: me Marble Lekythos * of Nikostrate. Conze i. pl. 90.

. Attic Grave Relief * from the Tlissos. Conze ii, pl. 211, no. 1055.

2. Mourning Figure from Tomb. Sabouwroff Coll., pl. 15.

. Monument of Knights Slain at Corinth. Curtius, Atlas von Athen, p. 3. . Attic Grave Relief * of Aristion (a youth) 4.2. 1871, p. 23, no. 50.

. Fragment of Relief * at Delphi. Nude Athlete (?).

. Votive Relief to Asclepios.* Annali 1873 pl. MN.

» ” ” ” Ἐπ ΟΣ ΠΩ, pl. 9. » » " ᾿ PAM, We pl. Le. ” ” 2) ” μὴ A.M. li. pl. 14.

* ”» ” 2) ” τ A.M. li, pl. 16. ” » ” ” Ξ (=58.) ” » » “5 A.M. ii. pl. 15.

ἘΠῚ} : Votive Relief to Pan ond Nymphs.* Sabowrof Coll., pl. 28.

»» stan” Ὁ is dedicated by Archandros. A.M. v. pl. 7.

. Eleusis Relief.* Demeter, Persephone, Triptolemos. Gardner, fig. 71.

tenia πι

; Attic Relief. Girl dancing.* Heydemann, Verhiillte Tdnzerin, p. 9.

no. 8 2.

. Attic Relief. Girl dancing.* Rev. Arch. N.S. 1867, pl. 2. . Asclepios from Epidauros.* Gardner, fig. 95.

. ‘Mourning Athene.’ Relief. Gardner, fig. 70.

. Friezes * of Nereid Monument. Two Nereids.*

2. Nereid Monument. Slab from large frieze.*

. Mausoleum frieze. Amazons.* Gardner, fig. 91.

. Temple of Diana, Ephesos, Sculptured drum. Alcestis.* . North side of Alexander Sarcophagus.* Gardner, fig. 106. . Pergamene frieze.* Zeus. Gardner, fig. 114.

» » 7 Restored cast. Zeus.

” » ™ Athene, Victory, Giant. Gardner, fig. 115. :; » + Restored cast. Athene, Victory, Giant, Ge. + Giant next staircase.

»” ”

Sb81. 82. 83.

84.

85. 86. 87. 88. 89. 90. 91. 92. 93. 94,

Ixix

Pergamene-frieze.t Giant. 4 » + Restored cast. Hecate, Aries, Artemis. ᾿ ΜΕ 4 » selene, Helios. af nor At Mf Parthenos, Bootes, Nyx (Goddess with snake-entwined vase). » t Restored cast. Phoibe, Asteria, Dionysos and Icarios.* Terracotta panel. 2.M. Terracottas, pl. 25. Dionysos visiting Icarios.* B.M. Bacchante with kid.* B.M. Bacchic Thiasos.* B.M. Hellenistic Relief. Walls and vine. (Schreiber, pl. 41.) Bacchus in mystic basket.* Terracotta panel. B.M. Apotheosis of Homer.* B.M. Indo-Greek Relief from Malakand Pass.

” ” ” ” ”

SCULPTURE.—Statues, Busts, etc., of Fine and Later periods. [See also the

Se

—_— MH SODN ADB ot γα

section on the Parthenon].

. Marsyas * of Myron. Lateran. Gardner, fig. 51. . Bronze Marsyas* of Myron. B.M.

Discobolos * of Myron. B.M. 250.

. Diadumenos* of Vaison. B.M. Gardner, fig. 75. . Temple of Zeus. Olympia. Pediments restored.

EK. Pediment.*

E. Pediment. Aged Seer.* τ. 4 W. Pediment. Central figure.*

A W. Pediment. View in Museum.

”» ” ” ”

» 3) ᾽) ”

᾽) ;

. Head of Asklepios* from Melos. Β.Μ. 550. . Caryatid from Erechtheion. Β.Μ. 407.

12. . Bronze head of Aphrodite.* B.M. . Fragments* from Epidauros.

. Hermes* of Praxiteles.

. Hermes* of Praxiteles.

. Head of Aphrodite.*

. Satyr* from Lamia.

. Ideal male head.

. Head of ‘ Eubuleus’ from Eleusis,

. Satyr of Praxiteles.

. Hermes of Andros.

Dionysos * from Monument of Thrasyllos. B.M. 432.

23. Statue of C. Ofellius* at Delos.

. Silenos with young Dionysos.*

. Bust of laughing Satyr.*

. Bronze bust, young Satyr.*

. Aphrodite of Melos.* Gardner, fig. 119. . Boreas group Akroterion.* Delos.

. Aphrodite* from Epidauros.

. Julius Caesar.* B.M. Gardner, fig. 129. . Hecate. Arch.-Epigr. Mitt. v. pl. 1. . Athene* from Epidauros,

Ixx

. Head of Eros* from Paphos; side view. J.H.S. ix. pl. 10. . Mausolos.* B.M. Gardner, fig. 90.

. Artemisia.* ΒΜ,

. Demeter* of Cnidos. B.M.

. Persephone* trom Cnidos. B.M.

. Head of Alexander.* B.M.

. Girl fastening chiton.

. Terracotta head* from Paphos.

. Fragments of sculpture* from Paphos.

. ‘Dying Gladiator.’*

. Head of Gaul.* BM.

. ‘Paetus and Arria.’*

2. Fallen Giant* and Amazon* (Attalian offering). . Laocoon.* Gardner, fig. 116.

. Head of Apollo Giustiniani.* B.M.

. The Farnese Bull.*

wPan*

. Artemis of Versailles. Gardner, fig. 121.

. Young Pan* of M. Cossutius Cerdo, B.M.

vot =, 48),

3) 3) ” ”

VASES.—Prehistoric, Mycenaean, etc.

. Hissarlik Vases. B.D. 2008, 2023. . Mycenacan Vases from Islands. B.D. 2062, 2067. . Mycenaean ‘false amphora.’

Mycenaean Vases. Calymnos and Carpathos. J.H.S. pl. 83.

. Warrior Vase, Mycenae, rev. Schuch. fig. 284.

obv. Schuch. fig. 285. rev. ἘΞ

2) 3) })

”

. Dipylon Vase fragment, Tiryns. Schuch. fig. 131.

. Attic Amphora, 7th cent. Warriors, etc. b.D. 2079.

. Dipylon Vase. A. Ζ. 1885, pl. 8.

. Dipylon Amphora. Funeral processions. B.D. 2071.

: ᾿ i B.D. 2071(=11).

. Melian vase subject. Two warriors in combat. B.D, 2086. . Melian Vase(=13).

. Melian Vase. Apollo and Artemis. &. & C. p. 53.

16. . Male head from Archaic Vase. (Helbig, Hom. Epos.’ fig. 74.) . Aristonothos Vase. ἢ. dJ. ix. 4.

19. 20.

Bearded head from Melian Vase. B.D. 240.

Amphora of Polemarchos. Naweratis 1. pl. 4. Vase from Aegina. ᾿ Harpies, etc. -4.Z. 1882, pi. 9. Heracles and Geryon. J.H.8. v. p. 176,

” ” >

22, Fragments from Naucratis, Naueratis i. pl, 5,

Ixxi

Va23. Heracles and Nessos. Gorgons. A./). 57, 24. Rhodian plate. Menclaos, Hector, Euphorbos. (B.M.) 25. Vase of Arcesilaus, Silphium weighing. ἢ, & C. fig. 43. 26. Corinthian votive tablets. Antihe Denkmaeler, pl. 6. 27. Proto-Corinthian (Macmillan) lekythos. 07.11.5. xi. pl. 2.

VASES.—Black-figured,

Deities. Vb 1. Panathenaic Amphora. Burgon. B.M. B130. 2. εἶ Leyden. Wt. & C. fig. 62.

3. Sacrifice to Athene. 1, & C. pl. 7. 4. Athene and Bull at Altar. G.A.V. 242. 5. Poseidon and Athene. Amasis Amphora. 7), & C. fig. 56. 6. Dionysos, Ariadne, Citharist. 7. Triptolemos, bearded. Heracles. 8. Lekythos. Cheiron, Hermes and Infant Heracles. 4.Z. 1876, pl. 17. 9. Heracles and Hydra, B.D. 724. 10. 7 bringing up Cerberus. B.D. 730. 1: 3 drawing wine of Pholos. B.D. 726. 12. : binding Cretan bull. B.D. 727. 13. Heracles and Triton. Κ & C. fig. 57. 14. Heracles and Geryon of Exekias. G.A.V. 107. 15. Apotheosis of Heracles. 1, ἀ C. pl. 8.

Theseus. 16, Theseus, Minotaur, Ariadne and Chorus. (raz. A7ch, 1884 pl., 1. The Francois Vase, ete.

17. Francois Vase. General view. Rev. B.D. pl. 74.

18. " Bs Ε “ Obv. B.D. pl. 74. 19. -: ¥ Calydonian Boar.

20. » τ Friezes (small). W.V. 1888, pl. 2. 21); ¥ 4 Peleus, Cheiron, etc.

22) Ἂ f, Hermes, Zeus, Muses.

23: e = Muses, Hera.

24. Γ ᾿ Apollo, Fountain.

95. aa τ Rhodia, Thetis.

26. Ἧ ἐ- Thetis, Hermes, Athene.

27. a x Troilos.

28. " cs Antenor, Priam.

29. : a Priam, Hector.

30. . ᾿ Troilos Band. (=24 to 29.)

31. Amphora. Berlin, 1655. Amphiaraos, Chariot race, W.V,1889, pl. 10, 32. Procession of Musicians.

Vb33.

- oo Or

Ixxii

Cyrenian Vases. [Cadmos? Prometheus? Troilos, Heracles and Pholos.] .A.Z. 1881, pl. 12.

The Trojan Cycles.

. Atalanta and Peleus wrestling. B.D. 158.

. Peleus, Thetis and Cheiron, Ajax. G.A.V. 227.

}. Achilles brought to Cheiron. Heracles and Lion.

. Peleus, Achilles and Cheiron. B.M. B620. J.H.S. 1. pl. 2. . Judgment of Paris. Cf. J.H.S. vii. p. 202.

» os G.A.V.172.

. Hermes leading Goddesses. G.A.V. 173. . Judgment of Paris. 4.2. 1882, pl. 11. 2. Achilles and Penthesileia. Memnon and Aethiopians. B.M. B209.

G.A.V. 207.

3. Achilles and Memnon. Achilles and Penthesileia. . Achilles and Penthesileia. Dionysos and Oinopion. Vase of Exekias

B.M. B210. G.A.V. 206.

. Achilles, Polyxena, Troilos. B.M. B3824.

” ” ”

᾿ς Three heroes.

. Achilles and Memnon. #.BZ. pl. 22, 1.

. Hector and Andromache. ἢ. dJ. 1855, pl. 20.

. Dragging of Hector. H.B. pl. 19, fig. 6.

. Shade of Achilles. Tomb of Patroclos. G.A.V. 198. . Death of Astyanax. B.D. 797.

” 25 ” ΓΞ 52).

. Ajax and Cassandra. J.H.S. pl. 40. . Aeneas and Anchises. #.D. 32.

” »” a) (= 5b).

The Odyssey.

. Odysseus leaving Cave. J.HLS. iv. p. 263.

” ” ” (=57).

. Companions of Odysseus with Rams. J.H.S. iv. p. 261.

VASES.—Red-figured, ete.

Deities.

. Assembly of Gods. By Sosias, Ext. of Ve 48. Miiller-Wieseler,

no. 210.

. Aphrodite on the Goose. Py Euphronios (?). BM. D2. . Birth of Athene. B.M. E410.

. Athene seated with owl. By Duris. Gerhard, Trinksch. wu. Gef. pl.

13 (but cf. 4.Ζ. 1875, p. 88).

. Athene and Hephaestos, Fragment from Acropolis. . Athene receiving Erichthonics. 2741. x., pl. 39. . Dionysiac dance, Aryballos, Sabowroff Coll., pl. δῦ

Ixxiii

. Dance of Bacchantes, Term and Altar of Dionysos, by Hieron.

R. & C. fig. 80.

. Pluto and Persephone. Overbeck, Awnstmyth. Atlas. pl. 18, fig. 11.

. Return of Persephone. Strube-Brunn, Bilderkreis von Eleusis, pl. 3.

. Mission of Triptolemos. By Hieron. B.M. E140. ALCL. ix. pl. 43. . Mission of Triptolemos.,

. The Underworld. Tarentine Vase at Karlsruhe. A.Z. 1843, pl. 11.

. The Underworld. Vase from Altamura, £.D. 2042 A.

. Ixion on Wheel. W.V. Ser. H, pl. 5.

. Gigantomachia, by Aristophanes. Gerhard, Z’rinksch αι. Gef. pl. 2, 3. . The making of Pandora. B.M. D4.

. Rhyton. Eos and Kephalos. Compte Rendu (St. Petersburg), 1872

pl. 4.

. Eos and Kephalos. Stars, Helios, B.M. E466. . Eos and Tithonos. By Mieron. ALT, 1]. pl. 48.

» =20. (Internal subject only.) Heracles,

. Alemene on pyre, by Python. B.M. F149, J.H.S. xi. pl. 6. . Infant Heracles with snakes. Amphitryon, Athene, Alemene. 271].

xi., pl. 43. 2.

. Heracles and Nemean Lion. Munich Amphora. £&.D. 723. . Heracles and Eurystheus, by Euphronios. Klein, Huphr.? p. 89. . Heracles and Kyknos, Diomede and Aeneas. B.M. E73. Journ. of

Philol. vii. pl. B.

5. Heracles and Antaios, by Euphronios. Klein, Huphr.? p. 118. . Apollo and Heracles contending for tripod, by Andocides. Gerhard,

Trinksch. αι. Gef. pl. 19.

Theseus.

. Aegeus and Themis. G.A.V. 327. . Theseus and Aithra. Odysseus, Diomede, and Palladion. Conversation.

By Hieron. ᾿ ἃ]. vi. pl. 22.

. Theseus and Kerkyon, by Euthymides. Annali, 1870, pl. O.

. Labours of Theseus, by Chachrylion.

. Labours of Theseus, by Duris. B.M. E48. G.A.V. 234.

. Theseus, Athene, and Amphitrite, by Euphronios. Klein, HZuphr.

p. 182. The Trojan Cycles,

. Peleus wrestling with Thetis. Kylix of Peithinos. Gerhard, 7'rink-

schalen, pl. 9.

. Peleus and Thetis, from Camiros. B.M. E424.

. Peleus and Thetis. Nereids. B.M. E73. Jowrn. of Philol. vii. pl. A. . Peleus and Thetis. Poseidow. Nereus, by Duris. W.V. Ser. vii. 2. . Peleus, Thetis, and Cheiron. HB. pl. 8, fig. 6.

. Judgment of Paris by Hieron, JV.V. Ser. B, 5.

. Judgment of Paris. Toilet of goddesses. M.dZ. iv. pl. 8,

Ixxiv

. Judgment of Paris (late). ΙΓ΄ Κ΄. Ser. A, pl. 10, 3. . Judginent of Paris. Sabouroff Coll. pl. 61.

» = 59 (subject only).

. Judgment of Paris. Sabowroff Coll.

2. Paris and Helen. Compte Rendu, 1861, pl. 5, fig. 1.

. Leading away of Helen, by Hieron. W.V. Ser. A, pl. 5.

. Ajax, Teucer, and Telamon. B.D. 743.

. Achilles and Briseis. J.H.S. pl. 3.

. Achilles and Briseis of Euxitheos. G.A.V.187. B.M. E 258.

” (= 66).

. Achilles and Diomede, by Euphronios. Klein, Huphr.? p, 241.

Patroclos bandaged by Achilles. (Int. of Ve 1). By Sosias. Miiller- Wieseler I. no. 210. εν ἘΞ.

. The taking of Briseis, by Hieron. B.D. 776. . Parting of Hector and Andromache.

” (= 72).

. Odysseus in tent of Achilles. B.D. 781. Hypnos and Thanatos. . Embassy to Achilles. .4.Z. 1881, pl. 8.

. Odysseus, Diomedes, Dolon. B.M.F 157. 1.8. pl. 17, fig. 4.

. Murder of Rhesos. W.V. Ser. C, pl. 3, fig. 2.

. Thetis and Hephaestos. Gerhard, 7rinkschalen, pl. 9.

. Nereids bringing armour. Heydemann, Nerezden.

. Nereids with arms of Achilles. JfdJ. iii. pl. 20.

Nereids with arms of Achilles. 1.8. pl. 17, fig. 1.

2. Walls of Troy. Achilles and Hector, Priam, Hecuba, Athene. G.A.V.

203.

. Combat of Diomedes and Aeneas. B.M. E 78. Journ. of Philol. vii. B.

(Part of Ve 34).

. Trojan farewell scene. Priam and Hector. G.A.V. 189.

. Ajax (?) and Hector separated. H.B. pl. 15, fig. 4.

. Hypnos and Thanatos with body. MdZ. vi. pl. 21.

. Warriors arming, by Duris. £B.D. 2207.

. Trojans arming. (Ext. of Vc. 98). Klein, Zuphr.? 215.

. Achilles, Nestor and Iris. W.V. Ser. D. ἃ. Miiller-Wieseler, no. 207. . Achilles and Hector in combat. G.A.V. 202.

. Sacrifice of Trojans at pyre of Patroclos. M.dJ. ix. pl. 32.

2. Hermes and Achilles. G.A.V. 200.

. Priam in the tent of Achilles. B.D. 791.

. Achilles at dinner. Body of Hector. 216]. viii. pl. 27.

. Priam as suppliant to Achilles. G.A.V. 197.

}. Redemption of Hector. 714]. v. pl. 11.

. Achilles and Penthesileia.

. Achilles and Troilos. By Euphronios. Klein, Huphr.? p. 220. (Cf.

Ve 88, 99.) Achilles seizing Troilos. By Euphronios, Klein, Zuphr? p. 214, (Ext. of Ve, 98.)

Ixxv

Vc100. Eos and Thetis before Zeus. H.B. pl. 20, fig. 10.

101.

102.

103.

104. 105.

106. 107.

108,

109.

110.

Prt.

112.

120. 121: 122. 123. 124.

125. 126.

127.

128. 129.

130.

131.

132. 133. 134. 135.

136.

137.

Tai; 8.

Mal.

2,

Memnon and Achilles, Hector and Achilles. B.M. E 468, G.A.V. 204.

Weighing the souls. 1.8. pl. 22, fig. 9.

Kos and Memnon, by Duris. W.V. vi. pl. 7.

Winged beings with a corpse, by Pamphaios. B.M. E 12.

Odysseus and Diomedes with Palladion, by Hieron. 21]. vi. pl. 22.

The Trojan Horse. G.A.V. 229, 230.

Iliupersis by Brygos. W.V. Ser, B. viii. 4.

Tliupersis at Naples. Miiller-Wieseler, no. 202.

Tliupersis. B.M. F 100,

Orestes slaying Aegisthos. H.B. pl. 28, fig. 10.

᾽ν » ᾽

The Odyssey.

Odysseus and Companions tied to rams, 7.1.8. iv. fig. 3a (facing p. 252). Scenes of Daily Life, ete.

Ships by Nicosthenes. J.H.S. pl. 49.

Marriage procession. Stackelberg, Gracber, pl. 42.

Revellers. By Euphronios. Burlington Fine Arts Coll. no. 8.

School scenes, by Duris. 1. & C. fig. 72.

Athletes practising (Panaitios Kalos). Klein, Huphr?, p. 284.

Discobolos (Panaitios Kalos). Klein, Huphr*. p. 285.

Athletes practising (jumping etc.) 4.2. 1884, pl. 16.

Athlete hurling spear.

Youths with horses. A.Z. 1885, pl. 11,

Horsemen (“rothemis Kaios), by Euphronios and Diotimos.

Alcaeus and Sappho. Mus. Ital. 11. pl. 4.

Dionysiac Chorus. B.D. 422.

Tricoupi Kylix. Youth pouring libation. J.H.S.x. pl. 1.

Three figures and shade, at a tomb. Athenian lekythos. “ὦ, & C. fig. 87.

White Lekythos. Deposition of a woman, by Hypnos and Thanatos. Dumont and Chaplain, i. pl. 27.

White Lekythi. Charon and his boat. Antike Denkm. pl. 23.

Charon and girl, Lekythos. Antike Denkm. pl. 23. (fig. 3 only.)

INSCRIPTIONS.

3-7. Inscriptions from Epidaurus. Heading of treaty between Samos and Athens, with relief of Hera and Athene.* C.J.A. iv. ii. no. 1b, Collignon 11. fig. 56.

MISCELLANEOUS SUBJECTS (Mycenacan and Early Periods).

Tiryns fresco. Bull hunt. B.D. 1901. Gold Intaglio. Female figures. Schliemann, Mycenae, fig, 550,

Ixxvi

. Gold Intaglio. Schhemann, A/ycenae, figs. 354, 335.

. Gold mask. Schliemann, Mycenae, fig. 474.

. Two-handled cup with birds (Mycen: 10). Schuch. fig, 240.

. Dagger blade. Lion hunt. Schuch. fig. 227.

. Dagger blades. Horses, Ducks, and C: fig! Schuch. figs. 270, 271. . Fragment of silver bowl (Mycenae). Defence of a walled city.

Ephimeris, 1891, pl. 2, fig. 2

. Gold cups (Vaphio). Bulls. Hphemeris, 1889, pl. 9. 10. 1: 12. 19. 14. 15. 10.

Axehead. Ephemeris, 1889, pl. 8.

Bronze razor. B.D. 238.

Gold cup (Aegina). B.M. J.H.S. xiii. p. 196. Gold pendant (Aegina). B.M. J.H.S. xiii. p. 197. Gold pendant (Aegina). B.M. J.H.S. xiii. p. 201. “Homeric Warrior, fully armed.” Side view. “Homeric Warrior, fully armed.” Front view.

Slides 17-48 are a collection of Mycenaean and cognate Egyptian

17. 18. 69: 20. 21. 22. 23. 24. 5. Spirals and lotus, Egyptian. P. 85.

subjects formed by Prof. W. M. F. Petrie.

Tell el Amarna, spirals on columns. Petrie, Zell el Amarna, pl. x. head see = bull, fresco. * cS δ ΗΠ: sakes calf ks i in ΝΣ ΕΝ ᾿ bull and lion, fresco. Unpublished.

τ ‘ canal, fresco. Unpublished.

Gein Aahhotep and Mycenaean. Orig. and Schuch. fig. 186. Neferhotep ceiling, spade spiral. P. 81. Spirals, Egyptian. P. 83.

iy P86;

27. Cat fresco, Egyptian. Brit. Mus. no. 170.

29. Cat dagger, Mycenae, 2 slides. A.M. vii. 8.

. Disc of Sarobina (Berlin). From photo.

. Gold cup with rosettes. Schliemann, Mycenae, fig. 344. . Fluted cup. Schliemann, Mycenae, fig. 342.

. Silver vase of Kefti (Rekhmara). P. 100.

. Wavy band dise. Schuch. fig. 189.

. Spiral disc. Schuch. fig. 191.

. Diadem, three rows. Schuch. fig. 158.

. Diadem (half only), large bosses. Schuch. fig. 149.

. Spirals from gold breastplate. Schuch, fig. 256.

. Nubian pots with spirals (photo).

. Nubian vase with boat . Scarabs and Cretan stones (Evans). J.H.S. xiv. p. 327. . Phoenician patern. Perrot and Chip. iii. 546.

. Orchomenos ceiling. Collignon, fig. 19.

. Tiryns alabaster frieze. Collignon, fig. 26.

5. Octopod dise. Schuch. fig. 190.

}. Boeotian gold band. “Ep ’Apy. 1892, pl. 12.

”»

Ixxvii

Ma47. Spirals, Mycenaean stele. Schuch. fig. 146.

48

49. 50.

51. 52. 53.

Cow’s head. Schliemann, Mycenae, tig. 327.

Phoenician bowl (Amathus). Siege. Helbig, Hom. Epos, pl. 1.

Phoenician bowl. Combats with gryphons and lions. Clermont- Ganneau, L’Imag. Phén. pl. 4.

Phoenician bowl. (-Ξ 0).

Phoenician bowl (Praeneste). Egyptian subjects. J/.dJ, x. pl. 32, fig. 1.

Shield of Achilles, restored (Murray).

. Figure from cuirass. Helbig, Hom. Epos, no. 48. . Priam redeeming Hector (Olympian bronze), Au/s. L. Curtius gewidm.

pl. 4.

. Bronze statuettes. Warriors. B.D, 2190-1. . Painted tablet. Armed Warrior (Acropolis). . Athene with spoils of Gorgon. Intaglio (Cyprus). Murray, Handbook ,

pl. facing p. 152, fig. 9.

. Terracotta plaque. Funeral Procession. Rayet, Monuments, pl. 75.

MISCELLANEOUS SUBJECTS.—Later Periods.

. Relief, with votive wreath.

. Votive relief. Surgical Instruments (Epidauros).

. Votive relief. Ears. (Epidauros).

. Incised mirror. Corinthos and Leucas. /J.H.S. ix. p. 62.

. Head of Athene. Silver coin of Athens, fifth cent. B.c.

. Contest of sAthene and Poseidon. Bronze coin of Athens. J.H.S,

pl. 75, Z xiv.

. Athene with shield and thunderbolt. Bronze coin of Athens. J.H.S.

pl. 75, AA xiv.

. Head of Apollo. Gold stater of Philip 11. of Macedon. . Triptolemos in snake-chariot. Bronze coin of Eleusis. J.H.S. pl. 77,

EE xx.

. Aphrodite with shield, and Eros. Bronze coin of Corinth. JJLS,

pl. 53, αὶ exxi.

. Aphrodite in temple. Bronze coin of Corinth, JH.S. pl. 58,

G ecxxvi.

2. Roman medallion. Arrival of Asclepius at Ins, Tiberina. . Roman coins, showing temple of Aphrodite at Paphos.

. Diagram, shewing Doric Chiton. B.D. 419.

. Model oxhead (modern) in a garden, as a charm.

. Bradfield. Greek Theatre.

. Bradfield. Scene in the Agamemnon.

Ixxvili

LOAN COLLECTION OF THE TEACHERS’ GUILD.

Arrangements have been made by which members of the ‘Hellenic Society are enabled to borrow the slides in the collection of the Teachers’ Guild.

The slides are lent for one night at 1d. each, 10d. a dozen.

All orders respecting this collection’ should be sent to Messrs. G. Philip and Son, 32, Fleet Street, E.C.

The full catalogue may be consulted at, or borrowed from, 22, Albemarle Street, and the slides can be seen at Messrs. Philip and Son’s.

The collection is classified as follows :—

iiviii. Maps and Plans... τὰ oh toe ae sas 61 slides. xi. Views. Oriental i as me ie τῶν ἀν. eh 0m, εν Asia Minor... a oa He nee ΟΣ Kills, 55 Northern Greece aan = Le of deer BD i, as XV. Ὁ; The Islands... ¥ [Ω de at es | αν ὦ»; Attica (outside ‘Aahiens) Je ee ans BEV. gs EVI Peloponnese ... ie ἊΣ ΝΣ dae 59: Ὁ Ὁ UH. ἡ τε Western Greece ay suf a. 16 fg. α΄ 4 ον Italy and Sicily MY baa ταὶ PKs xxi. Prehistoric Greece. Homer and Μηδ δ ἐν if 45. δ, xxil. Persia and the Persian Wars ... ἐδ 8 Ds py ae xxiii. Athens. Topography .. shit 52) δὲ XXIV. 95 History ἐπὶ ἘΠΕ, coins, ete), 9... xxv. Greek Portraits.. 33 96 τς xxvi. Coins (taken τ the most ae! As the ienfid in Duruy’s History of Greece).. a. ὖο SEZ OF xxvii. Social Life (scenes of ΠΣ life, on medial som rae 29. Sh xxviil. Arts and Manufactures.. eee or or wed οι. xxix. Warfare... bof Po» ἢ, κα oh ΝΣ 4

χχχὶ. Sculpture. (A complete series of the subjects in Prof. EK. A. Gardner’s Handbook of Sculpture. The reference numbers are the same as those of the

figures in the handbook) ... i. We je, OOSE

xxxii. Religion and Mythology ἽΝ δ 99 xxxill, Greek Drama. (The ancient theatre; ἀρ scenes from Greek plays—Antigone, Agamemnon at Bradfield,

Ion at Cambridge, Frogs at Oxford.) ay ms 43

THE J. PAUL GETTY MUSEUM LIB

GETTY CENTER LIBRARY

wut

0098 3615

Saree

Pee

ἜΞΩ ὝΕΣ Ὁ τ ούτς

srt ~ eS

2529.

᾿ 7 , vor ὅ oes τς ἊΣ : : a : “ 2256 π 2) vy τ ; , , ΣΤ: Jon x : : ᾿ : ; "4 mw, ¢ δ + “« >.

wy

G os

δ᾽

vos

ΠΩΣ τς

τ,

Zi ,

IR} a =a)

a

peed

Lavenncdt ao) r ΠΣ ae ye β : 4 7 < Δ Aon naiei dade giatrghest

ae ae ΖΟΣ ΄ ?