Skip to main content

Full text of "The Journal of Hellenic studies"

See other formats





2rf)C Sofictj) for tfjc |)vomotion of ll)fl!rnic 5)tutiirs 







Reprinted by permission of 


A Division of 




Printed in Germany 


llules of the Society 

liist of Oflicers and Members 

Proceedings of the Society, 1897-8 

Proceedings of the Cambridge Branch of the Society 

Additions to the Library 

List of Periodical Publications in the Libraiy 

Allen (T. W.) ... 
Anderson (J. G. C.) 

5> 51 )> 


HuuuKs (E. W.) ... 

Burrows (R. M.) 

Bury (J. B.) 
Gardner (E. A.)... 

Gardner (P.) 

Grundy (G. B.j ... 

)> )» • • • 

Hill(G. E.) 
I.mhuof-Blumer (E.) 

Myres (J. T,.) ... 
Paton (W. R.) and 

Perdrizet (P. E.) 
Richards (G. C.) 
S.MITH (A. 11.) ... 
Verrall (A. W.) 
Walters (H. B.) 

White (J. F.) 
White (R. E.) 
woodhouse (w. 
Yorke (V. W.) 


The Text of the Homeric Hymns. Part V. 

A Summer in Phrygia : Part II. (Phites IV., V.). 

Some Corrections and Additions 

Excavations of the British School at Melos : the 

Hallof the Mystae (Phites L-IIl.) 

... Notes on Pylos and Sphacteria 

... The Arabs in Asia Minor (G41-750), from Arabic 

. Sources 

... Pylos and Sphacteria (Plates VIL-X.) 

Mr. G. B. Grundy on Pylos and Sphacteria 

... The Double City of Megalopolis 

... A Head in the possession of Pliilip Nelson, Esij., 

M.B. (Plate XL) 

Boreas and Oreithyia on a late Attic Vase (Plate 


A Suggested Characteristic in Thukydides' Work. 

Battles, Ancient and Modern 

A Dedication to Artemis 

... Coin-types of some Kilikian Cities (Plates XIL, 


... ,S'ee Paton (W. R.) 

Myres (J. L.) On Some Karian and Hellenic Oil- 


The Game of Morra 

Archaeology in Greece, 1897-1808 ... .;. 

Illustrations to Bacchylides (Plate XIV.) 

Death and the Horse 

... On some Black-figuied Vases recently acquired by 
the British Museum (Plates XV.-XVII.) 

Note on some Attic Stelai 

Women in Ptolemaic Egypt 

... The Greeks at Plataiai 

Inscriptions from Kastei'n Asia Minor 

Index of Subjects (Vols, xvii., xviii.) 
Greek Index (Vols, xvii., xviii.) 























1. Mosaic at Mclos. 
II. ,, „ „ (Detail; Gazelle). 

^III. „ ., ., (Detail: Cock). 

IV. Map of Pluygia Paioieius. (Inset: the Phrygo-Lydiau Frontier.) 

V. Map of the District round Synuada. 

VI. Boreas and Oreithyia. (Vase in the Ashmolean Museum.) 

VII. Views of Pylos and Sphacteria. 

VIII. Views of Pylos. 

IX. Views of Pylos. 

X. Views of Sphacteria. 

XI. Head in the possession of Pliilip Nelson, Es(p, M.B., XIII. Cilician Coin?. 

XIV. Theseus and Amphitrite. 

XV. The Sacrifice of Polyxena. 

XVI. The Departure of a Warrior. 

XVII. (1) Combat over Fallen Warrior. 
(2) Kyathos by Nikosthenes. 




Map of the Battle-field of Plataiai 

Tyclie and Plutus (from the Hall of the Mystae, Melos) 
Site of the Hall of the IVfystae, Melos : View from M cud 


«1 «• «• «■ t« ^4 11 1< *' ft 

Pavement in 

Key-Pbn ... 

Statue of Iliorophant ,, „ 

Maible Uasis ,, ., 

Bust of Aurelia Euposia ,, ,, 

Morra ; Gold Ring (Imperial Ottoman Museum) 

,, from R.F. Hydria (Dzialynski) 

„ Bronze Figure from Foggia (B.M.) .. 

Gold Ring (B. M.) 

Krater in Ashmolean Museum (Reverse) 

Plan of the TraXaiov Ipv/xa on Sphacteria 

Wall on Sphacteria 

Oil-Press at Argiuuuta in Kalymnos 

Press- Beds from Karia, Tripoli, and Kalymnos 
Hellenic Fortress and Oil-Press at Emporiu in Kalymnos 
Press-Bed from Lykastos, Crete 
Press-Bed at Klimatovi'ini in ^Nlelos ... 

Weightstones for Oil-Presses 

Croesus on Pyre ( Vase in Louvre) . . . 
Deatli of Meleager (Vase at Naples) 



























Thebe on Cadmos Vase of Assteas 

Daughters of Proetos (Vase at Naples) 

Heracles and Lion (Vase in B.M.) 

Heracles Sacrificing (Vase at St. Petersburg) 

Theseus Ileceiviug Wreath from Aniphitrite (Vase at Dologna) 
Theseus and Poseidon (Vase in Bibl. Nat. Paris) ... 

„ „ „ (Vase in Collection of Princess di Tricase). 

Ship of Theseus (Scene from Fran(;ois Vase) ... 

Corinthian b.f. Oenochoe (B.M.) 

Peloponnesian b.f. Amphora: Reverse (B.M.) 

Athenian b.f. Kantharos (B.M.) 

* Kleinmeister' Kylix, with Arming of Warriors (B.M.) 
B.F. Amphora : Heroes Playing at Draughts (B.M.) 
Herakles and Kerberos on b.f. Amphora (B.M.) 
Herakles and Geryon on b.f. Kylix (B.M.) 

Panathenaic Araphoriskos (B.M.) 

Stater of Sicyon : Reverse (B.M.) 




{kXvtottcoXo^, k\vt6<!, eXi^ etc.) 

Did the Greeks, and in particular did the Homeric poets, associate Death 
with the Horse ? The great importance, in the archaeology of art and 
religion, of all associations connected with the grave, will perhaps give interest 
to a somewhat full discussion of this question, or rather of the single piece of 
evidence, upon which, so far as concerns Homer, the question seems to turn. 
Did the poets describe Hades, lord of Death, as ' lord of the goodly steeds ' ? 
Is this what they meant by kXvtottcoXo'; ? It is the purpose of this paper to 
show that they did not, that this interpretation is involved in difficulties 
and impossibilities three-fold and four-fold, has for it neither reason nor 
authority, and must, with all that depends upon it, be given up. 

The first and perhaps sufficient objection is this. Before the epithet 
KXvTOTTtoXo^: could be referred to the horse, ttwXo?, it is plain, must have 
signified a horse. Now it is quite certain, though apparently not 
recognized, that to the composers of the Iliad and Odyssey no such word as 
TTwXo? ho7'se was known. They used, it is true, the word to which, by a 
stretch of meaning and for convenience, that sense was given by their 
imitators and successors ; but they knew it only and strictly in what seems to 
have been its primitive and etymological sense, a foal, a young horse under the 
mother. ' Chestnut horses (iTrTrou?) hundred and fifty, all mares and many with 
their foals {ttcoXoi) at their feet ' says Nestor in A 681 : and see also T 222, 
225. Against ttwXo? horse the evidence is overwhelming. If these poets had 
known at all a word so irresistibly convenient as a synonym for tTTTro? beginning 
with a consonant, they must have used it, in the extant poems, not once but 
scores of times. This estimate is no mere conjecture ; it is proved by experiment. 
The composers of the Hymns, imitators of ' Homer ', but differing much in 
language and feeling, did, like the Attic poets, know TrwXo? {young hoo'se) in a 
sense nearly equivalent to iVTro?, and accordingly with them horses are TrwXot 
twice, (those of Ares in 8, 7, and those of Selen^ in 30, 9), that is to say about 
once for ten times that the animal is mentioned. Now at this rate the Iliad 
alone should have given us ttwXo? horse about forty times or more ; ^ yet it 

^ Kiricot is found there about 400 times ; see though I may mention perhaps that I have 

Ebeling's Lexicon s.v. My references and read both /Zmk^ and Odyssey through with this 

statistics are largely taken from this book, subject in mind. 

2 A. W. V ERR ALL. 

does not once. Nor does the Odyssey. We read, it is true, in i/r 24G how 
Athena ' detained at Oceanus the golden-throned Morn, and would not let her 
yoke the swift-foot steeds that bring light to men, Lanipos and Phaethon, the 
TTwXot that draw Morn.' 

'Hw h' avTG 
pvaaT in *D.Kedv<p y^pvaodpovov, ovK ea iTnrovf; 
^^vyvvcrO' 6)KV7ro8a<;, (\>uo<i dvOpcoiroiai (^epovra^, 
[AdfiTTov KoX ^aedovd', oit' 'Hw ttojXoi dyovai]. 

If we suppose this last verse to be of the true ' Homeric ' age, we must 
translate it according to the use of that age, and must take the poet to mean, 
what is perhaps not inconceivable or unnatural, that the car of the young 
Morning is drawn by a team of foals. But it is an obvious and more probable 
supposition, that the verse is a mere note, satisfying that passion for names, 
to which poet-scholars were liable but bards were not, and that the author of 
the verse, using ircoXoi as synonymous with 'iTnroi, simply betrays thereby his 
later date. To invent for this single passage a sense of ttwXo?, which Iliad 
and Odyssey combine to reject and disprove, is not permissible ; and it remains 
therefore true that by the composers of these poems TrwXo? horse was not 
used, which in the circumstances is equivalent to ' not known '. 

If therefore in KXvTOTrcoXoi;, as used in the Iliad, ttwXo? meant horse, it is 
a case of survival. We should have to assume that ttwXo? had once borne 
this meaning, as it did again in later poetry, and that in the compound, as a 
traditional epithet, this sense held its ground, although the corresponding 
sense of the simple had suffered in the age of ' Homer ' an odd eclipse. 
Let us see whether the application of the compound admits this 

That application is extremely peculiar. It is restricted not merely to 
Hades, but to Hades in one single phase and function, as receiver of the 
warrior's parting soul : — 

' And for thee I say that slaughter and black Death shall come about 
here at my hands ; vanquished by my spear thou shalt yield to me my glory 
and thy life to Hades Idytoiwlos ' 

€v^o<; ifiol h(oaeiv, '^v')(i)v 8' "AiSt KXvroiraikai} 

Now when the poets so used ^XutottcuXo?, surviving, ex hypothcsi, from a 
time when it meant of the goodly steeds, of what sense in it, if any, were they con- 
scious? Or could they use it traditionally, without any question of the sense ? 
Surely not. Tliey may have soused, and probably did, htuKropa dpy€i(f>6pTr}<;, 
as a de-scription of Hermes. But then these words, or rather names, were free, 
for them, from any connexion of etymology. They do not, on the face of them, 
signify anything in the Greek of Homer ; they are not in appearance formed 
from any elements to which separately Homer gives a sense. But /cXuTOTrtyXo? 
is. Of one meaning in Homeric language it was manifestly capable ; it could 
mean ' of the famous foals'. How then, unless the elements of the word were 

E 654, and similarly A 41.0, n 625. 


capable of some other meaning, should this meaning be ignored ; or how 
could the compound continue to be used in a connexion where, in its 
natural meaning, it was plainly absurd ? The epithet ■)(^pvat]\(iKaro<;, ' of the 
golden arroio ', was retained, in its traditional connexion with Artemis, by the 
Homeric poets, although to them, by a restriction in the sense of rfKaKcnr), it had 
come to signify ' of the golden distaff' (S 122, 131); because the new sense 
was in this connexion, though less appropriate, at least not impossible. And 
similar was the history of Zeu? rep'rrifcepavvo'i, transformed from the liurlcr of the 
thunder into the dcliglder in it. But when 'rr<o\o<; liorse had ' come to mean ' 
foal, and fool only, then 'Ai8rj<{ /cXyTOTrcuXo"?, as an expression significant but 
now absurd, would naturally die. That it did not die is /w-ma facie proof 
that it was not connected, and was not supposed to be, with the ttwXo? which 
for Homer meant foal ; and that in attributing to this TrwXo?, by pure hypothesis, 
a use earlier than Homer, but for Homer extinct, in the sense of horse, we 
are on a wrong track. 

Now in these circumstances it is instructive, and it should not be sur- 
prising, to find that, although to the Greeks of the classic and later times no 
other word ttwXo? was known, as a term in use, except that which primarily 
meant foal and subsequently also horse, nevertheless among students of 
Homer the best tradition affirmed that the termination of K\vT67r(oXo<{ 
('AiSr;?) had an origin and meaning totally different. Aristarchus, according 
to several witnesses,^ connected it with TrwXeladai, to ran(/e, haunt, visit. The 
explanations of the epithet, which the witnesses deduce from this etymology, 
are certainly incredible, indeed preposterous. But this only goes to prove 
that the etymology itself, which they could not use, was not invented by 
them (nor by Aristarchus, if he is responsible for the explanations), but was a 
genuine inheritance from times when the language of the rhapsodists was not 
yet dead. And whether this was so or not, the etymology, as an etymology, 
is possible, correct, and Homeric. The verb TrtuXeo/xat is Homeric, and to 
irwXiofiai, the adjectival termination -irooXof stands in the same relation as 
-TToXo? (in oLOTToXo^, TpiTToXo^, htKaairoXa, afx^liroXos,) to the parallel, 
cognate, and synonymous TroXeofiai. Before therefore, in order to interpret 
«:XyT07r&)\o9, we assume a sense of -ttwXo? which Homer does not warrant, 
we are bound to try whether, with or without the assistance of Aristarchus, 
we can interpret it by the sense which he does. 

The truth appears to be, that the little group of Homeric adjectives in 
-TrwXo? (for /cXuToVtwXo? is not unique) are all connected not with ttwXo"? 
foal, and certainly not with ttwXo? horse, to Homer a vox nihili, but with the 
root TTtuX- range, which appears in ircoXiofiat,. The position in Homer of the 
nominal stem from this root, ttwXo-, is exactly parallel to that of ttoXo- ; that 
is to say, neither appears in Homer as an independent substantive, though 
TToXo? had elsewhere in Greek a long and illustrious descent ; and both appear 
in Homer as terminations in a group of compound adjectives. The particular 
use of TTwXiofiai,, from which the most familiar of these adjectives originally 
came, is that which, as was indicated (according to the witnesses) by 

' See note on j). 4. 

B 2 


Aristarchus*, survives in the compound iTnirayXiofiai, when connected with 
activity on the battle-field : — 

avrap 6 TOiv aXXcov eVeTTtwA-ctTo arlx^^f <J^vZpS)v ' then went he elsewhere 
ranging the warrior-ranks' (A 265). It refers to that rapid and incessant 
motion from place to place, which, in the loose, desultory, and undisciplined 
method of Homeric fighting, made so large a part of the fighter's power and 
efficiency. When all depended, as it does in Homer, on catching your man 
in the instant of isolation or exposure, to be quick of movenunt, nimhle in 
range was among the first of warlike qualities ; and this is the quality which is 
claimed for the Phrygians (in general), when they are called atoXoTrwXot 
(r 185, etc.), and for the Danaoi (in general), and the Myrmidons (in general), 
when they are called raxv-rroiXoi. Even if it were legitimate and Homeric 
(which, let us repeat once more, it is not), to assume for these adjectives the 
element ttwXo? horse, that assumption would still be excluded by the use of 
them. The men of Agamemnon and Achilles, as a class or people, could not 
possibly be known or noted for their swift harses ; for with few exceptions 
they had no horses at all. But as fighters they are noted for their quick 
range, their nimble movements in the field. 

From the same stem probably came ei/TrtoA-o?, the traditional epithet of 
Ilios, though here a doubt arises, which for TaxvireoXof; and atoXoTreoXo? is not 
enter tainable. It is possible to derive ei/7r&)Xo<? from ttcoXo? foal, and to 
connect it with the famous legend of the twelve foals, begotten by Boreas upon 
the mares of Erichthonios, son of Dardanos (T 220 foil.) ; and this we may 
even take to be so far true, as that the epithet, so interpreted, gave a likely 
suggestion for the legend. But that the legend produced the epithet is not 
likely, for then it would naturally have linked itself in poetic tradition with 
Dardania, which was the name of the place where the foals were born, and 
not with Ilios, which (according to the legend itself, T 216) did not then 
exist, but was built, according to the prevalent account, long after, for 
Laomedon son of Ilos. As a fact the city, which is evircoXa, is scarcely ever 
Dardania, and regularly Ilios; nor is the legend required to account for the 
phrase "IXto? evTrioXa, which meant originally just ' Ilios, the pleasant haunt ', 
from TTOiXo-, irwXeofjLat,, as otoTroXo? ^copo? ' a solitary haunt ', and signified, 
like ev vaio^ievo^ etc., that the place was 'good to visit' and 'good to 
frequent ', in short, a country agreeable for human habitation. And indeed 
the tradition of ancient scholarship preserved an obscure memory of this, 
when eu7r&)Xo9 (see Ebeling, s.v.) was translated, not incorrectly, by evyeoxi 
' a pleasant land ', and the like. 

Apart from proper names, such as 'E;T^€7ra)Xo9, which may mean anything 
or nothing, these are, I think, all the words in -ttwXo?, which Homer supplies, 
except KXvTOTraiXo^: itself. This, if it was really known and used by the poet 
or poets of the Iliad — we shall see presently the reason for the doubt — cannot 
be separated from aioXoVtuXo? and rap^uTrtuXo?. Hades, as /cXutottcuXo?, must 

' See Ebeling ».v. K\vr6irci)\oT. — 6 'Apltrrapxos iirl tow K\vToirw\cfi' aKoift KXvrijv iiriir6\7]<Tty 


be ' Death, the famous ranger (of the battle-field) ' : and since, in fact, it is 
always tlie soul of the warrior slain upon the field, which this Hades receives, 
the conception is one which we may well accept as, at any rate, a st.age in 
the liistory of the phrase. Compared with the irrelevant and impossible 
horses, it is no less superior on the poetic side than on the linguistic. But it 
seems that we ought to look yet further. 

Y or firstly, although from atoXoTrtwXo? and Ta;^u7rft)\o9 it is not hard so 
to interpret kXvtottcoXo^:, it was not perhaps equally natural and obvious 
upon these lines to invent it. Both aloXo- and ra^v- are terms of motion, 
like TTcoXo- itself. Not so kXvto-, and the coalition is thus less easy. Nor 
have we a perfectly satisfactory analogy in evvtaXo^ or otoiroXo^, which, 
strictly speaking, would justify only the rendering ' Death, famous for his 
haunt ', famous, that is, for the place which he ranges or visits, an idea neither 
so clear as might be wished, nor so much to the purpose. Secondly, how does 
it come about, that this ' famous ranger ' of the field is never so described 
when the breadth and rapidity of his range would be illustrated by the circum- 
stances, never in scenes of wide, swift massacre, such as are so often presented, 
but only at the side of the single fallen man, over whom his enemy stands 
exulting ? A ' fixed epithet' may be often misapplied, but it should scarcely 
be so always. These objections do nothing to help in the ' horses', to which 
the second applies even more strongly than to the ' range ' ; indeed it is 
impossible, as I think, to explain why, if /tXuTOTreoXo? had really referred to 
horses, it should never be linked by Homer to any of the numerous personages 
who are with him ' famous for horses ', and only to Hades, who, so far as 
appears, was not. But the objections justify a suspicion that we are not 
yet at the bottom of the matter; and since the capacities of ttcuXo- seem to be 
exhausted, it remains to see, whether anything more can be made of kXvto-, 
an examination which, as few Homeric words are more characteristic and 
important than /cXi'to?, will be interesting for its own sake. 

In general the Epic use of kXvt6<; is simple and well defined. 

(1) It is applied, according to the etymology, to persons, places, and the 
like, which are properly and literally 'heard of, /amoves, renowned. So 
Agamemnon, Argos, etc., etc. Even in this class however it appears, upon 
a more careful inspection, that some selective feeling, not apparent in the 
etymology, has affected the choice of objects. Not all renowned persons are 
in fact kXutoi, nor those chiefly, or indeed at all, who are most plainly 
renowned ; females, for example, hardly ever, neither goddesses nor women, not 
Penelope, not Helen, though more ' famous ', one would think, than all the 
male sex together; of the gods some only, and those repeatedly, but chosen, 
if ' fame ' were the question, with strange caprice. 

(2) What the selective principle is, by what association the word was 
attracted and confined, appears plainly in the things^ the objects not capable 
of personification, to which it belongs. It is said or implied in Lexica that 
kXvt6<; renowned is extended in Homer to the general sense of heauteoxis or 
goodly ; but this statement is so inexact as to be practically false. How ill 
8uch large and vague expressions correspond with Homeric feeling about the 


word, might appear sufficiently from the fact that Homer, using k\vto<! 
incessantly, knows no such expression as, for example, k\vt6<; t7nro<;. Even 
the limitation that ' Homer uses it especially of the works of human skill 
(Liddell & Scott), though mainly true, is both too wide and too narrow. 
When the word does not mean simply and literally rcnoiviied, it is applied 
solely to works of art, or rather to works of craft, human or divine, and among 
works of craft almost exclusively to a small and peculiar class. Arms (and 
more rarely clothes in general) are everywhere KKvrd, kXvto, revx^a, kXvto, 
€i/j,aTa, houses are everywhere kXvto,, kXvtu Scofiara, and so are, here and 
there rarely, one or two other things of the same kind, that is to say, products 
of craft ivhich directly manifest the poiver, digiiity, and security of the person by 
whom the craft is possessed oi' commanded. The feeling which, whether known 
to the poets by observation or divined by imagination, the word expresses, is 
the admiration, respect, and worship attaching, in the rudimentary stage of 
civilisation, to craft and its possessors, to the empire of the metals, and the 
powers which depend upon it, good smith-work, good masonry, and good 
carpentry. That is why, with rare and dubious exceptions,' males only are 
kXvtoI; why "H^ato-To? (or 'Afji,(l>iyvi]€i<;) send 'Eivvoa-iyaioq (not Poseidon as 
such), who would be patrons, one of the smithy and the other, in his suhterranean 
office, of the mine, are conspicuously kXvtoi; and lastly, why the instances of 
kXvto, T€v^ea (eifiaTa) and kXvto, hwfiara are more numerous than ail other 
kXvto. together. So also the objects, when specified, by which persons are 
entitled to the epithet, are almost always works of craft, and apparently never 
products of nature : KXvr6ep'yo<i, KXvTordxvri^, KXvT6ro^o<i, vav(TiKXvTo<i, hovpi- 
kXvto^. It is in later poetry, not in Homer, that we find such expressions as 

It is worth while, since this topic lies deep in the sources of Homeric 
feeling, to dwell for a moment upon the signal illustration of it offered by 
four pictures in the Odyssey, all intended to create wonder, and in a certain 
sense admiration, the dwellings of Calypso, of Circe, of the Phaeacians, and of 
the Laestrygons. If kXvt6<;, to Homeric ears, had signified only that 
sentiment of vague and general admiration, which belongs to the terms 
which we have to put for it, to heaictcous, nohle, goodly, glorious and the like^ 
then, among these homes and their occupants, the epithet must belong plainly 
and conspicuously, though with some difiierence perhaps in the shade of it, to 
Calypso and to Circe ; it must apply also to the Phaeacians, less strongly 
perhaps but not much less ; while to the Laestrygons it nmst be altogether 
refused. The abode of Calypso is painted as the very ideal of natural 
goodliness, that of Circe as consummate in the luxuries of magic, Phaeacia as 
exc^uisite in art ; but the land of the Laestrygons, where was no tillage, ' no 
signs of the labour of men and oxen, only we saw the smoke curling upwards 

' Even the very rare examples of a feminine "^woalyaios in the next line) for eths 'Antpirphr) 

kKvt6s are ""not beyond suspicion (B 742, or the like. In B 742 it is easy to restore a 

t 422) : kAut^ apparently does not occur, a masculine k\i;t#, and to account for the cor- 

significant fact. In « 422 the unique KKvrhs ruption of it. 
'Au*ito(t»j may be an error (sugf^ested by kAutoj 


from the land ', is <as dreary and repulsive as its people. But quite other, for 
Homer, are their claims to be kXvtoL That is a matter not of beauty, but of 
craft. Calypso is not /k\uto9, nor her cave, trees, waters, nor any of the fair 
things that belong to her. Neither (which might more surprise us) is Circe, 
no, not though she has a house, a true palace (k 210 and jmssitn), and that full of 
magnificent wonders. But this, if we have once felt the Homeric feeling 
about k\vt6<;, is intelligible enough and quite right. Magic may be superior to 
' craft ', but it is not the same thing. Houses of men, and of gods too, when 
and because they are the works of craft, are kXvto. hwfiara : but the chambers 
of a witch, who could create serving-maids out of the fountains and streams 
(k 350), need not be the product of craft at all ; and accordingly the huyfiara 
KipKr]<:, though mentioned repeatedly and adorned with various epithets 
(reTvyfieva, KoKd, even iepd or mystic), receive not once the familiar and regular 
Homeric epithet kXvtu : nor does anything which the witch possesses. The 
Phaeacians upon the same principle are of course kXvtoi, and their works 
KXvrd, kXvtoC, dyuKXvToi, and irepiKXvToc, themselves, their dress, houses, 
sanctuaries, etc., etc. ; not because they are * goodly', but because they are in 
all things artists, and their dwelling-place full of wonderful art. For the 
Laestrygons and their works, though assuredly not ' goodly ', ' beautiful ', or 
attractive in any way, arc kXvtoI antl KXvrd no less, and indeed in this 
quality have a marked pre-eminence. The whole account of them and their 
country fills but 50 verses, as the Odyssean voyagers scarcely enter it and 
barely escape. Yet the epithet occurs three times {k 87 Xifiiva kXvtop, 112 
KXvTct hwfiara, 114 kXvtov 'AvTi<f>aT7]a), and is the first note, as it were, of 
Odysseus' impressions. And the reason, upon Homeric principles, is obvious. 
It is the ' artificial basin ', with its plumb walls and projecting piers of 
wrought stone, which excites this awe in the beholders, and in Odysseus a 
salutary fear. It is the 'smooth road' and the 'high buildings' (103, 111), 
and the formidable weapons (121, 124), which show that Antiphates, king of 
the Laestrygons, commands to a supreme degree the resources of craft, and 
therefore, though cannibal, is emphatically atXi/to?. Indeed it seems more than 
probable that ' Fargate of the Laestrygons ' is, or originally was, a picture 
coloured, if not drawn, from the report of some terrified mariners, who, 
trading from lauds of pasture and agriculture, saw for the first time some 
place, on the Euxine, may;be, where metal- work was practised on a large scale ; 
a sort of black country, where ' the smoke went up from the land ', where the 
trolly, on paths of incredible facility, rolled down from the hills the wood 
for the furnace (k 102), where shifts so extended the hours of labour that 
' night and day near met in one,'^ and whence the visitor, roughly handled by 
the hard workmen and appalled by the signs of their skill and power, fled 
away to report that their figures were gigantic, and that they lived, like the 
Martians of Mr. Wells' romance, on the flesh of men. Such at all events is in 

* K 85. There is nothing inconsistent with nights of the far-north. It would be on the 
this in the current suggestion, that the ' meet- Euxine that a Greek would probably first hear a 
ing of night and day ' refers to the brief bunmier rumour of thio phenomenon. 


fact the Laestrygonian type ; and it illustrates excellently the true Homeric 
sense of KXvrot, grand ^ great, a word for us not really translatable, but 
approximating in effect to powerful or rather craftful, implying awe rather 
than mere admiration, and from all such terms as beauteous or goodly to be 
sharply sundered and distinguished. The gracious life of Aeolus, and the 
hideous life of Antiphates, are passed alike in kXvto, Soifiara (tc CO, 112), for 
this praise belongs to the ' brazen bulwark ' and the ' sheer stone ', though it 
does not belong to the fairy's paradise nor to the witch's bower. 

But against a general background of this shade, ascertained, as we must 
remember, by scores and scores of examples, three examples stand out in 
conspicuous disagreement, both with the general rule and with one another. 
Each offends against Homeric usage, and offends in a different way. They 
hare long been observed for their peculiarity, and all receive special notice, 
for instance, from Liddell and Scott. 

(1) Once, and once only, is broken the rule that natural things, products 
of nature, cannot be kXvtu. The herds of the Cyclops seem to be such 
(t 308) : Kal T0T6 irOp aveKaie koI rj/j.€\y€ kXvto, fifjXa. 

(2) Once, and strangely, tnankind as a whole seems to be a kXvtov. 
When Sleep has done his errand for Hera, he departs eVt KXvra <f>vX' dv- 
OpCOTTCOV (3 361). 

(3) Once, most strangely of all, tlie dead, universally, seem to be kXvtoi or 
kXvto.. Odysseus, at the entrance of the lower world, must address his 
prayers to kXvto, edvea veKpotv (/c 526). 

Now we have no right, until the severest scrutiny has shown that no 
other explanation is open, to assume, in the circumstances, that these three 
exceptions are genuine. The presumption against them is enormous. Take 
the first. The epic poets mention hundreds and hundreds of times domestic 
animals such as ySoe?, diy€<i, Xttttol, KVvef;, 6le<i, firjXa, alvoXia, etc., etc., and 
with many admiring epithets. The adjective kXvt6<;, expressing as it does a 
peculiarly characteristic feeling, is one of their favourite words. If such 
phrases as kXvtoI /3oe9, KXvra fiijXa, had really been possible to their eai's, 
what likelihood is there that we should be left with one single example ? 
Why should the flocks of the Cyclops be selected for this praise, and what 
does it mean ? To all the notions normally suggested by /cXi/to?, the life and 
manners of the Cyclops, a rude, easeful, sluttish simplicity without culture of 
any kind, present the extreme opposite. ' Celebrated ' they were not, neither 
they nor anything of theirs, for they were cut off from the world and unknown ; 
and as for their flocks, it does not appear that they differed from 
flocks in general. The^' are 'fat', they are 'fleecy'; but how should they 
exhibit the greatness of ^jpwc?' and craft ? Expositors have felt this so 
strongly as even to suggest that kXvto, here should mean noisy, loud ; but 
that is a counsel of desperation. 

To call mankind or the tribes of men kXvto. is so far at least more 
intelligible, as the quality so predicated is proper to beings who are men 
or manlike. But it does not belong to the type of man. It is essentially a 
trait of superiority and dominion. We are told that here it indicates the 


superiority of mankind to the brutes. But why should this conception, than 
which surely none could be more alien from the general tone of the Epos, 
suddenly force itself upon the poet's mind, when contemplating mankind in a 
relation essentially animal and common to the brutes ? In relation to Sleep, man 
is but a brute. Why then, because visited by Sleep, should men excite, for 
this once, the peculiar admiration expressed by /icXi/to?, or indeed any 
admiration at all ? 

And the dead ? The fame, lordliness, power, craft of the dead ! They 
are the silent, strengthless, forgotten, the — all which kXvtoi are not. For 
though Leosica may say that this k\vt^ eOvea vexpav refers to 'illustrious' 
dead, it does not refer to illustrious dead, but distinctly and expressly to ' all 
the dead ' (k 518), the dead in general, 'brides and grooms, long-laboured age 
and tender virginity ' (\ 38). Perhaps nothing is more characteristic of the Epos 
than the absence and repudiation of all ideas attributing power and ability 
to the dead. They are essentially helpless and craftless, and, if they may ever 
recover activity for a time, can do so only by aid and gift of the living ; and their 
intercourse with Ulysses on this occasion is especially impressed with that con- 
ception. Why then should they here for once be KXvrd, and in what sense ? 

In short, these passages are not explicable, and the presumption is that 
they are erroneous, a presumption hard indeed to prove, but not incapable of 
proof. Suppose that the error were the same in all three. Suppose there 
were a word, which, while scarcely distinguishable from kXvto's, fitted each of 
the three unconnected contexts, and supplied in each a fresh point. Could it 
be reasonably doubted, that this word, and not kXvtoi;, was the word employed ? 
Such a word is KXlroq, couched, lying down, the participial adjective from kXX- 
to couch, related to K€KXlfiivo<i couched as x^ro^, <j>61t6^, and many other words 
of this poetic and archaic type, to Ke')(yfi€vo<:, €<f>Oifi€vo^ and the rest. The 
flocks of the Cyclops, though not otherwise miraculous or marvellous, are 
remarkable in this, that tJiey share at night the home of their master. It is the 
first thing that we hear of them ; ' we saw a cave... near to the sea, and there 
many flocks and herds were used to sleep. And about it a high outer court 

was built with stones And a man was wont to sleep therein, of monstrous 

size, who shepherded his flocks alone and afar,' and so on {t 182). The males 
lay usually in the yard, but the females, ' all that he milked ', actually 
within the cave {ih. 237), the filthiness of which is noted with epic simplicity 
{ih. 329) ; and the Cyclops lay among them, kcIt hroaO ' avrpoio Tavva-ad- 
fitvo^ Sia fiijXmu {ih. 298) ; and these arrangements, it will be remembered, 
are of the first importance, not only to the colour of the tale, but to the 
incidents. It is therefore natural and to the purpose, that the narrator, his 
mind full of this picture, should describe how at morning, after Odysseus' first 
night there, the giant ' kindled the fire and milked his couehM flocks ' (ttO/j 
dvixaie xal rj/j,eXrfe KXird firjjXa, ih. 308), those, that is, who shared his bed, 
the word, more man-like than beast-like, glancing aptly at his beast-like 
habits. And it may be observed, that in the evenings, when the beasts have 
not been ' couched ', it is not the KXird ^r)Xa who are milked, but ' the ewes 
and bleating she-goats' {ih. 244, 341). 

10 A. W. VERRALL. 

So again very properly Sleep, when he has finished the special employ- 
ment for which he was summoned to Olympus by Hera, departs ' to the 
couches of mankind ' (cS^j^er' eVt k\it^ <f>v\' avdpcoTrcov), returns, that 
is to say, to his ordinary sphere and business. Where else should his visits 
be paid but to ' them that lie down ' ? 

And among those that sleep, couch, and lie down, one class in particular 
receive the name, in Homeric language as in all others, specially and dis- 
tinctively, those that have lain down for ever, kXit^ eOvea vexpcov, the ' tribes 
of the couchM dead '. 

Now one of two things : either the exact and varied applicability of the 
word K\iT6<i to these three occasions, selected upon otlier grounds and without 
reference to such applicability, is accidental, or it proves that kXit6<: was 
indeed the word there used. For myself, I hold the first alternative to be 
fantastically impossible, and therefore embrace the second, taking it as 
certain that the epic poets had a word k\it6<: couched, which was liable (this 
is obviously true) to be confused with the homophonous kXvto^, and, being 
archaic in type and replaced in later language by other equivalents, has 
actually given way to kXvto^ and disappeared. It was still alive and known, 
when these parts of the Iliad and Odyssey were composed ; and we shall do 
well to consider whether we can trace it later. 

As to the phrase from the Iliad, kXvtcl <f)vX' avOpmirwv, we have some 
interesting evidence in the 'Pythian' part of the Hymn to Apollo, an 
imitative composition dating probably from the sixth century, later at any 
rate than the Epos in general, and bearing many marks of its lateness. Here 
we read, when Pytho is being recommended to Apollo for the site of his 
future oracle (270), ' There no fair chariots shall go the round, nor shall there 
be noise of swift-foot steeds about the fair-built altar ; yet to that privacy {koI 
«U9) the great peoples of men (avdpcoTroyv kXvtcl (f)vXa) may bring gifts to 
I^-paion, and thou with glad heart mayst receive the fair victims of men 
that dwell around (TrepiKTiovcou dvOpcoTrcov).' And again, the monster snake of 
Pytho ' did many a mischief among the great peoples of men ' who came to the 
place as builders and worshippers (355). * Whoever met her, became the prey 
of his fate.' And again, ' All sacrifices,' says Apollo (537) ' that the great 
peoples of men (irepiKXvTci (f>vX' avOptoTrav) shall bring to me.' It is clear 
that the ear of this author had been caught, as well it might be, by the 
expression in this form, with kXvto, ; and he treats it exactly as traditional 
phrases from our own archaic and consecrated literature, sometimes no better 
founded or more significant, are dealt with by our own poets and preachers. 
He does his best, that is, to accommodate it with a proper context and 
meaning. With this purpose, he has changed the sense of <f)vX' dvOpconcov. 
In the Iliad this means of course simply mankind, the human species, as (f>vXa 
dewv means gods, and <f>vXa ryvvaiK&p the female sex! But in the Hymn, 
conformably to later use, <f>vXa means peoples, nations, the inhabitants of that 
earth of which Pytho was supposed the centre. And further, since it is for 
the glory of the god that these tribes are brought into view, the epithet kXvto, 
great, grand, mighty, has at least so much reflected propriety as is sufficient 


for a consecrated formula. It is plain therefore that into this, by the 
sixth century, kXvtu had already obtruded itself, though whether this was 
the form in which the phrase first attached itself to the worship of Delphi (or 
rather Pytho), is not so clear. There is reason to think (see Euripides, Iph. 
Taur. 12G2) that there, as at other sanctuaries of oracular and medicinal deities, 
prescriptions were once sought by the method of sleeping in the sacred 
precinct, and communicated by dreams. The appearance in connexion with 
the gifts, which the avOpwirwv ^v\a were to bring, of the name Paion, of the 
snake, and of the need for quiet, bears a strong suggestion of this Asclepian 
usage, and of Kkira <f>v\' avdpoiirwv, couched or sleeping men, as the primitive 
form belonging to it. 

However in the sixth century kXvto, c^OX,' dvOpconcov somewhere certainly, 
and perhaps therefore in the Iliad, had established itself. But in the Odyssey 
kXito, firjXa not kXvt^ (and probably therefore also kXito. edvea veKpoiv) 
might still be read a century later. For Sophocles read it, and copied it in 
this passage of the Ajax (372) : — 

w hva-/iiopo<;, 09 X^P'' f^^^ 
fie6r]Ka tov<; dXd(TTopa<i, iv S' eXcKeaai 
^oval Kal /c\tT049 irea-wv atTroXioif; 
ipefivov alfi ehevaa. 

' Wretch that I am, who suffered the accursed men to slip tlirough my hands, 
but fell on coiled kine and couched flocks, and made their dark blood flow ! ' 
That he has here in mind the Homeric phrase there can be little doubt, but 
that he read and wrote KXvrd, kXvtoi<;, is not easily credible. Even if such 
expressions as kXvto, aliroXia, ' fine herds ', had been familiar to the Epos 
(where in fact nothing of the sort ever occurs), they would still not have been 
suitable for transplanting into the style of Sophocles. Largely as the Attic 
dramatists use the Epic vocabulary, especially of course in lyrics, it is not 
their habit (unless I am mistaken) to adopt from the Epos the conventional 
simplicity of its ' fixed epithets'; nor do they use Epic words without regard 
to the changes and restrictions of meaning, which they had since undergone. 
As an example of the first point we may note, that this seems to be the sole 
appearance in Attic drama, perhaps in any poetry not professedly imitating 
the Epic, of eXt«e<? /36e9. And the second point is well illustrated by the 
Sophoclean use of kXvto^ itself. The use of it in Homer, as we have seen, is 
strongly affected and limited by a special association, which, so far as we can 
trace, has little to do with the etymology. In Sophocles on the other hand 
the special, archaic feeling and significance is naturally lost ; the etymology 
recovers its hold ; and kXvt6<: means simply gloi'ious, famous in the strict 
sense. Thus in Oed. Tyr. 172 the fruits of the earth (kXvtu^ x^ovoi;) are her 
glory, and spoils are gloriowi in Ai. 177. It is the same generally speaking in 
Pindar, with whom, as might be expected, the viordi famous is a favourite.^ It 

* Find. Pyth 9, 36 baia. K\vrav x*P« <"' ^poffe- irolav ; is hardly explicable by this sense of 
vfyK(7v, ^ l>a Kal in \exiuv Kfipat ij.e\ir]i4a kKvt6s, or indeed by any other. That Apollo 

ii A. W. VERRALL. 

seems then strange that Sophocles should introduce k\vt6<; here in some 
vague sense, which, even if it were Homeric, would still not be Sophoclean, 
inasmuch as it is irrelevant to the context and the thing described. There 
seems not to be, either in the nature of the beasts which the Greek army 
had collected for food, or in the situation of Ajax, any reason why he should 
speak of them with admiration. But there is much reason why he should 
speak of them as couched or sleeping, for he had massacred them in the night, 
an addition to their helplessness and his disgrace. 

It will be noticed that kXUeaai fiovai is translated above by ' coiled kine ', 
as if parallel to ' couch4d flocks '. I believe that it is, or at least that Sophocles so 
intended ; but this supposition is not necessary to a preference for AcXirot? 
over kXvtoU. In Homer IXt^c? ySoe?, whatever the first word signified or had 
signified, practically means no more than hine, and Sophocles might have 
borrowed it bodily in this sense. What was the true, original sense is a question 
80 remotely connected with our subject, that it cannot be treated here otherwise 
than summarily. It is clear (see for example Ebeling s.v.) that the Oraeco- 
Roman scholars had no information on the point, and were justly dissatisfied 
with their guesses. The conditions apparently are (1) that the word should 
describe some bovine characteristic, universal and obvious ; and (2) that it 
should be deducible from the notion curling, curled, curled up, coiled, for eKi^ 
exhibits this sense and no other, with peculiar distinctness, in all Greek from 
Homer downwards. Indeed it is scarcely too much to say, upon the facts, that 
to a Greek ear IXtf cannot have conveyed anything else, and the question 
really is. Why did the Epos speak of hine as curled or coiled ? The bovine 
horn (one interpretation) is not universally and specifically ^ , nor, if it were, 
would it make the beast such ; its hair is not more tKi^ than that of many 
other animals, nor so much ; and its ' rolling ' or rather swinging gait, due 
mainly to the great bulk of the body in proportion to the supports, is not 
^\tf at all, for the word describes shape, not movement, and the equivocal 
' rolling ' is an illegitimate bridge. The alleged rolling or turning of the feet 
might explain eiXiirohe';, but not KXtKc^ : nor can I think it likely, whatever 
may be the scientific truth, that herdsmen and poets would have chosen a 
mark which, as anyone may prove by watching, is, in the common, slow 
motions of the creature, to say the least, not conspicuous. It remains however 
very probable that the two standing epithets etXtVoSe? and eXixe^ are in 
some way connected. Is it possible — I put it only as a suggestion, which in 
any case, I believe, was favoured by Sophocles — that both were derived from 
the couchant posture, and pointed to the beast's manner and inveterate habit 
of Ij/ing down ? Certainly nothing is more obviously characteristic, both the 
thing and the way of it. Whether a cow * tucks up its feet ', when it lies, more 
completely than a sheep or goat, I cannot say, but from- the bulk of its body 
it seems to do so. It will often look, from a little distance, as if it had no 

should speak, in this connexion, of his KKirrhy to KKlyovtrav) with a meaning natural and 

X^pa, glorioua or farruma hand, has not been obvious. Aeschylus and Euripides scarcely use 

proTed intelligible ; and I believe that Pindar wXwrrfi at all, and throw no light upon it. 
Mid ttKirkp x^P* (from xXir^f, and eqaivalent 


legs at all. In stepping also, the curl of its lifted fore-leg is, for some reason, 
very conspicuous. And, as every one knows, it is always ' tucking up ' and 
remaining ' tucked up ' for hours together. Now the prefix elXi- points to a 
curling up as well as to a rolling along, perhaps more naturally. It seems there- 
fore not impossible that elXiiroSe*: originally meant this, and that i\iK€<i y96e?, 
coiled Jcine, described the same thing from a slightly different point of view. 
Probably the epic poets scarcely felt in ^cKe<i any separate significance at all ; 
but we can leas easily suppose this of Sophocles and his Athenian audience, 
who, if they took the view here propounded, had a case for it as students of 
Homer, and an excellent defence for the combination of ?\iKe<: /Soc? with 
kXit^ alnoXia. 

Returning now to our theme, we have it, as the result of this long 
excursion, that the Epic vocabulary contained the word kXito':, overlaid in 
script, as might be expected, by the familiar K\vT6<i, which indeed may be 
called a mis-spelling of it. Like hundreds of other words, like most words of 
its class, it disappeared from the fully developed language, leaving relics in 
the grammarians' erepoAcXtTo?, iyKkiriKo^:, in eKK\t,To<; avoidable (Photius), and 
perhaps elsewhere. A traditional /cXi/tottcdXo? is therefore ambiguous 
between these letters and KXtroTrcoXo^. Now we have seen already that 
KXtToi couched was a description proper to sleepers and to the dead, and further 
that it was applied to sleepers as receiving the visits of the personified 
Sleep. But further it can be shown that naXo- (TrtoXio/jLai) was a proper 
term for the haunt or visit of such personages as Sleep and Death ; for it is 
applied by Aeschylus to those of their kinsman the Dream. ' Visions of the 
night, coming ever to my maiden chamber ' {aUl yap 8yfr€i<i ^vpv^oi ttcoXcu- 
fievai e? 7rap0€vS>va<i rov^ ifiov<i.,.) says the Aeschylean lo {P.V. 672), 
adopting, as the archaic form shows, the language of some more ancient poet. 
Combining these elements, we have, in Hades KXiToirtoXo^: (quasi 6 -rrapct 
kXitov^ 7ra)Xovfi€vo^)t Death wh/) frequenteth the fallen, who visiteth them 
that lie dovm, whose haunt is among such. For the form of the compound we 
may compare aypavXo^ (6 iv aypoifi avXi^6/j.€V09:), dvBp6(rTpo<f>o^ (o iir' 
avBpa^ (TTpeifiOfievo^), diBo<f>oiTri<i (o <f>otT&v irapcL rov "AiSrjv), SiKda-woXoq 
(o iroXovfievo^i eU hUasi), etc. And since, when Death visits a person living, 
it is for the soul that he comes, it is natural that he should never appear as 
/cXtTOTTwXo? except in the act of receiving it. 

As for KXvTotraiXo^, it may have existed in the Epos in the only sense 
there possible^, famous for foals, but there is no proof of it. It might per- 
haps have been an epithet for Dardania, and it appears as such in one of the 
' Lives of Homer *, but with einratXov (already discussed) as a variant. But in 
truth it was not with such things as foals (or horses) that kXvro^j was 
associated by genuinely Homeric minds, and the balance of likelihood is 

' The only sense, that is, in which the word fifth century at least, and may even, as an 

could have been originally and deliberately alternative, be ' Homeric'. But invention does 

invented. The reading "AiSi KXvroitdiXif, with not account wholly for its origin, which 

the explanation ' Death the ranger ', must, I requires the co-operation of accident. 
Ht^ould think, go back, as an alternative, to the 


against tlieir having known /cXuTOTrtuXo? at all. To later poets it was 
perfectly natural, and in the sense famous for horses. Pindar (fr. 289) applies 
it in this sense to Poseidon, but whether he got it from his own invention, 
from Homer, or elsewhere, there is nothing to show. 

With the disappearance from Homer of Hades /cXutottcoXo? disappears 
all rea-son (see Dr. Leaf on E C54) for thinking that by the Greeks, or at any 
rate by Homeric Greeks, Death and the Horse were associated. That Hades 
the god, like any other great personage, might use horses upon a suitable 
occasion, as for example to carry off Persephond, goes without saying ; but he 
was not thought, so far as appears, to use them much ; and at all events 
between them and his function as Death, the Homeric imagination had not 
established any connexion. It is doubtful (but that is beyond our scope) 
whether the Greek imagination ever did. 

A. W. Verrall. 



§ 1. It is less easy to forgive Xenophon for telling us so little about 
the foundatioa of Megalopolis than for telling us nothing at all about the 
foundation of Messene. We would give much to know the details of the 
building of the city on the slopes of Ithome and the synoecisra of Messenia ; 
but Megalopolis, in its double character of a federate city and a federal 
capital, presented such complicated problems that the silence of those who 
could have best told us how those problems were solved is more aggravating 
than many of such silences to the curiosity of posterity. In this paper I 
propose to deal with one problem which seems never to have been quite 

§ 2, The investigation of the site conducted seven years ago by the 
British school confirmed, within less than half a mile, the statement of Poly- 
bius that the circumference of the walls was 50 stadia, and showed that the name 
Megalopolis was not so much a claim to unusual political importance for the 
new city as an appropriate expression of its unusual dimensions The circuit 
of the walls, as traced by Mr. Loring, measured 46 stadia (or 47^, if we add 
twice the breadth of the river).^ It is evident that the main reason for not 
selecting one of the older Arcadian towns as the centre of the Arcadian 
League, when it was founded in B.C. 371-0, was not, as Grote thought, their 
mutual jealousies, but rather their small size ; and, on the other hand, the 
motive of the relatively large circuit of Megalopolis was its intended position 
as capital of the League. Strategically such a large circuit was a weak point, 
not only because there was more wall to defend, but also because, owing to 
the expense of building and the necessity of building quickly, a long wall 
could not be built as solidly and well as a short one. A comparison of the 
remains of the wall of Megalopolis with those of the wall of Mantinea brings 

' Excavations at Megalopolis, 1890-1891 which 5 stades correspond to about 750 yards). 

{J. U.S. Supp. i., 1892), p. 114. Measuring the For comparison it may be mentioned that the 

circuit myself on Mr. Loring's plan, I made it out circuit of Thebes was 43 stades, that of 

to be nearly 20 stades longer. Having puzzled Corinth (not including Acro-Corinth) 40, that 

over this discrepancy, I discovered that he has of straggling unwalled Sparta 48. 
accidentally given a wrong scale for the stades (in 

16 J. B. BURY. 

into relief the second defect ; ^ while the diflSculty of defence is illustrated in 
the later history of the town.' 

§ 3. We have not sufficient data to enable us to determine the 
population of Megalopolis. A statement of Diodorus which has been used 
for this purpose contains an unknown element. In his account of the 
siege of the city by Polysperchon, the historian states that the number of 
citizens, slaves, and ^ivoi who were able to take part in the defence was 
15,000. Now (1) the ^evoi are an unknown quantity, and (2) Diodorus does 
not tell us how old were the oldest, and how young the youngest, of those 
males, citizens and slaves, who bore arms in this emergency.^ Instead of 
attempting to deduce a definite figure, it is safer to infer the magnitude of 
the population relatively to the other cities of Arcadia from the inscription 
in honour of the Athenian Phylarchus. This document* has been generally 
supposed to belong to the third century, and to prove a revival of the 
Arcadian League. But it really belongs to the first years of the League,** and 
may be fixed to the years B.C. 368-363.^ Of the fifty damiorgi of the 
Federation who are enumerated, ten are Megalopolitans and nine Mantineans.' 
The presumption is that this proportion roughly corresponds to the proportion 
of the respective populations of the two cities. Without pressing the infer- 
ence too far, we may safely say that, if the only purpose of Megalopolis had 
been to synoecize the Maenalians and Parrhasians, a city one quarter as large 
again as Mantinea would have been ample for the need, with room to spare. 
But the area of Megalopolis is nearly four times® that of Mantinea. It 
follows that the superfluous space was required for Federal purposes. 

§ 4. When the fact is grasped that the magnitude of Megalopolis was 
determined by its double character, we are soon led on to perceive some 
difficulties which must have caused anxious and serious meditation to the 
Arcadian ^ statesmen who conceived and carried out the plan of its founda- 

* Op. eit. p. 109. orgu The decree most have been prior to the 

* Polybius, 6, 93. Cp. below S 13. seceasion of MantiDea, and posterior to the ac- 
' Mr. Woodhouse [Excavatums, p. 8) uses the cession of Heraea and Orchomenos. One of the 

figures of Diodorus, and arrives at ' a popula- reasons for assigning the later date was the 

tion of perhaps 66,000 ' (both freemen and Attic dialect of the inscription. It seems to me 

slaves) ; Beloch {die BevUlkerung der yrUchisch- that this objection is answered by the inscrip- 

romiacKen Welt, p. 127) calculates 60,000 from tions of Antiochoe on the fronts of the sett- 

the same data ; both assume that (^foi = backs in the Megalopolitan theatre. 

niroiKOi. Of course, in any case, the data and ' There are only five Tegeates, and we may 

the inference refer to the population of the infer that their town had declined in nnmbers. 

town along with the district {x^fx^, Diodor. 18, Beloch (Joe. cU.) is wrong in his statement that 

70), and not the town alone. I doubt much Megalopolis sent as many delegates ' as Man- 

whether we can implicitly trust the figures of tinea and Tegea together.' 

Diodorus. • See below | 10. 

* Dittenberger, Syll. n. 167. * Epaminondas often gets the credit for M^^- 

* This has been recognised by Dittenberger, lopolis — without any evidence, I think, except 
ib. p. 661. the flourish of Pausanias, who aays he might 

* The limits are fixed by the presence of rightly be called the oecist of Megalopolis. The 
^antinean, Orchomenian, and Heraean dami- fact that he was the actual oecist of A(essene, 


tion. It was impossible to expect the Parrhasians and Maenalians, who now 
gave up their old tribal names and took the civic name of Megalopolitans, to 
undertake the responsibility of defending the whole line of fortification of a 
town which was far larger than their own needs required. And, on the other 
hand, the Pan-Arcadian League could not prudently place its buildings and its 
treasury at the mercy of one of its members. It was manifest that some 
precautions were necessary for the protection of the League, in case Megalo- 
polis were ever induced to secede. 

The interests of the League, as well as the interests of the city, demanded 
that Megalopolis should be defended not only by the Megalopolitan state, but 
also by the Pan- Arcadian state ; and the demand could be met only by the 
formation of a corps of federal troops. This is what was done. We find a 
band of 5,000 soldiers paid by the League, ready for service in any emergency, 
but quite distinct from the federal host, which gathered to march against an 
enemy when need called, but dispersed when the campaign was over. It is a 
legitimate inference that the constant duty of the Eparitoi, or a considerable 
part of that body, was to act as the garrison of Megalopolis. They were 
always available for emergencies elsewhere ; but it was the existence of the 
Federal capital that in the first place rendered the formation of the Eparitoi 

But when the necessity of a Pan-Arcadian garrison for the Pan- 
Arcadian capital had been recognised, there were many contingencies and 
dangers arising out of the double character of the town, which it was of 
great moment to foresee and provide against. 

§ 5. Megalopolis possessed one feature in common with the elder Man- 
tinea, which King Agesipolis had rased to the ground. The river Ophis flowed 
through Mantinea, and by damming it up the Spartan king had succeeded in 
taking the town. When the Mantineans rebuilt the city in the same months 
which saw the foundation of Megalopolis, they took good care to keep the 
fatal river outside their walls by digging a second channel for it, so that the 
stream divided on the east side, and, embracing the city round about, reunited 
its waters again in the north-west. Then what had been a weakness became a 
strength. In the same way the Helisson flowed through Megalopolis : but here 
there was not the same danger, since the ground was hilly, and not a dead 
flat like the site of Mantinea, Many Greek cities, perhaps most, were built 
on rivers ; but they were generally skirted or girt by them. It is no common 
thing to find a fortified city divided by a stream,^ 

combined with the support which he gave to the the building of the city proves nothing. See 

organisation of the Arcadian League, might Paus. 8, 27, 2. 

easily set afloat the idea that he was responsible ^ Pausanias cites Cnidus and Mytilene ; 8, 

for Megalopolis too. With our present evidence 30, 2. Dirce flowing through Thebes is another 

we are bound, in my opinion, to give the credit instance, but the case is somewhat different, 

of the idea to the Arcadian leaders who were There is no doubt that Dirce waa originally 

active in organising the federal state. The outside the walls ; the western extension of the 

sending of Pammenes from Thebes to protect city across the stream was comparatively late. 


18 J. B. BURY. 

§ 6. It was this river which supplied the founders of the Pan-Arcadian 
city with a simple means of solving their problem. The meaning of Megalopolis 
began to dawn on me when I stood on one of the high benches in the theatre 
and, looking northward, felt driven to ask why the city had crossed the river. 
It would have been in accordance with the design of other Greek cities if the 
circuit had been entirely on the southern side of the Helisson, stretching 
south-eastward over the site of the modern town. Strategical considerations 
would have emphatically recommended this plan ; for, if the northern wall 
had skirted the south bank of the river, the city would have been 
strengthened by an additional natural defence on the northern side. The 
inevitable inference is that there were cogent reasons of a political nature 
for disregarding the obvious considerations of strategy ; and it is obvious that 
these reasons can only have been connected with the double character of the 
place. There is no difficulty in drawing the conclusion — 

The Helisson divided the Federate city -from the Federal capital. 

§ 7. The northern half of Megalopolis was the city of Megalopolis in the 
strict political sense. For its defence the Megalopolitan citizens were re- 
sponsible, just as the Mantineans were responsible for the defence of Mantinea; 
and it was as exempt as Mantinea from Federal interference. The Agora was laid 
out on the north bank, and the Buleuterion was built beside it.^ This Hall of 
Council had nothing to do with the League ; it was for exclusively Megalopoli- 
tan purposes. The councillors who met together there dealt with the affairs of 
the city ; they were in no way concerned with the direction of the affairs of the 
Federation. When they Avent to take their place in the Federal Assembly 
and let their voice be heard in the discussion of Federal affairs, they were 
obliged to cross by bridge^ the river which divided their own city from 
the Federal capital of Arcadia. 

The southern division of Megalopolis was Pan-Arcadian ground. Here were 
all the Federal buildings and offices. Here stood the great Hall of Council 
or Assembly, called the Thersilion, in front of the theatre, which might itself 
be used for holding the meetings of the Ten Thousand. Here the 
Arcaxlian citizens, who gathered from all parts of the land to the capital of 
the League, were lodged, whether in permanent dwelling places, or in temporary 
tents, like those which served the spectators at the Olympian festival. 
Here dwelled the Federal magistrates and officers for their term of office 
here were the Pan-Arcadian treasury and the Pan-Arcadian archives. Here too 
the Eparitoi must have had their quarters ; and it was their duty, in case of 
an hostile assault, to defend the southern circuit of the walls. Here were 
ample spaces for the Arcadian throng to group themselves, the folk of each 
city, we may guess, in a quarter of its own, and to mix together, not only in 

Pausanias, 8, 30, 4. fjuestioned whether Megalopolis ever had a 

^ It is remarkable that no traces of an stone bridge. A wooden bridge seems the most 
ancient bridge have been found, and it may be probable hyjiothpsis. 


the debates of business, but in the festivities and amusements which would 
accompany the national meetings. 

The temples enumerated by Pausanias throw no light on the matter. 
Those which ho saw on the north side suggest no federal association. On 
the south he mentions seven : three of these (two to Asclepius, one to 
Artemis Agrotera) seem to have been still used, the other four were in ruins. 
Seeing this progress of decay, we cannot be surprised to find no mention of a 
sanctuary of federal significance, such as one may confidently assume to have 
existed during the federal period of the history of the city. 

ilj 8. By this arrangement the sojourners in the Federal capital, with 
those who came from time to time to attend the Assemblies, as well as the 
small number of permanent Federal officials, and the military garrison, had 
all the advantages of living in a city ; while the Federation was secured 
against the danger of Megalopolitan encroachment, against all corxfusion 
between Megalopolitan and Pan-Arcadian rights, by the clear and unmistakable 
boundary of Helisson's stream. In case a party in Megalopolis should ever 
induce the city to desert the League — and this was a terrible contingency which 
the founders of the dual town had to face — the Pan-Arcadian capital would 
indeed be in a serious peril ; but it would not without more ado pass into 
the hands of the secedors, as must have been the case if there had been no 
physical barrier corresponding to the difference between Megalopolis as a 
sovran city and Megalopolis as a Federal capital. In such an event the 
garrison of the southern town could easily maintain itself against the northern 
until reinforcements from the Arcadian cities arrived ; and northern and 
southern Megalopolis on either bank of theirriver might conceivably exist side 
by side, hostile and independent. 

§ 9. Thus the river performed a twofold function. It was a barrier 
which preserved the distinction between the two characters of Megalopolis 
against obliteration or confusion ; and it was also a military defence for the 
Federal capital against the possible revolt of the city to which it was locally 
attached. When Megalopolis was to be defended against a common enemy, 
the river was no hindrance to free communication between the Megalopolitan 
and the Pan-Arcadian sections of the garrison ; one city, and not two, was be- 
sieged, one city, and not tvi^o, was defended. But, if the Arcadian League were 
ever threatened by the hostility of Megalopolis itself, then the river would assume 
a new aspect, and become the northern fortification of the Federal capital, the 
southern fortification of the revolting city ; Megalopolis would break up into two 
adjacent towns. The Helisson served the purpose of a barrier, without obtruding 
that purpose as an artificial barrier would have done ; the innocent river need 
not suggest to the dwellers on its northern bank that the Federal government had 
ever considered the possibility of their defection or the necessity of a line of 
defence against them. 

§ 10. It has been pointed out above that a town one quarter as large 
again as Mantinea would have been of luxuriously ample size for Megalopolis, 

c 2 

20 J. B. BURY. 

if Megalopolis had not been the Federal capital Now the northern city more 
than fulfils this condition ; for it is about one third as large again as Maniinea. 
I have calculated ^ the areas of Mantinea and the two Megalopolitan towns, 
by weighing them in accurate scales, as follows : — 

Area of Mantinea * 1,471,612 square yards 

or (1,230,247 square metres) 

Area of Northern Megalopolis » 1,977,486 square yards 

Area of Southern Megalopolis 2,113,238 square yards 

Total area of Megalopolis 4,090,724 square yards 

§ 11. It is important to remember that the theatre was intended for 
Arcadia, and not merely for Megalopolis. It was a Federal building, and 
its construction must have been paid for out of Federal funds. This is 
proved (1) by its close connexion with the Federal Hall of Assembly — a 
connexion which is structural and not one of mere proximity ; and (2) by its 
vast size, compared with the little theatre of Mantinea. The Hall of Assembly, 
affording standing room for 10,000, and the Theatre, capable of seating 
20,000,* were part of the same design. The Megalopolitans of course had 
the advantage of the theatre ; when it was not required for Federal 
purposes, it was available for them ; this was one of the advantages to set 
off against the disadvantages of their close union with the Federal 
capital. The inscriptions of Antiochus on the backs of the front seats, 
which belong to the first twenty years of the history of Megalopolis, 
accord with the Federal character of the theatre. Antiochus is probably 
the envoy whom the Arcadian League sent up to Susa m B.C. 367. 
Xenophon describes him as an Arcadian pancratiast ; and he is probably the 
pancratiast of Lepreum mentioned by Pausanias.^ This hypothetical 
identity confirms the view that the benches which Antiochus dedicated in 
the theatre were a gift to the Pan- Arcadian League and not to the Megalo- 
politan city. 

§ 12. The serious disadvantage in the position of the Megalopolitan state 
was the prospect which it had to face in case the League were weakened or 
dissolved. In the latter case the southern town would be thrown entirely into 

^ The calculation depends on the plans of that the relative sizes of two cities do not cor 

Messrs. Foug&res and Loring. respond to their circuits. The circuit of the 

^ If Mantinea be treated as an ellipse, the wall of the southern town is a little less than 2j^ 

area (rob), calculated from M. Foug^res' state- miles, that of the northern a little more than 

meotof the lengths of the major and minor axes, 3 miles. The entire circumference of the 

would give 1,136,630 metres. The fact that the northern town is about 4 miles. The circum- 

ellipse is not perfect, being extremely blunted ferenco of Mantinea is somewhat more than 2^ 

at one side, accounts for the difference in the miles (3,942 metres = 21 stades, 180 feet) ; see 

results. My colleague, Mr. W. E. Thrift, Foug^res, 'Fouilles de Mantinde," 5.C.JJ. 1890, 

kindly helped me in these calculations. pp. 68-70. 

' If Polybius had known these measiuements * 19,700 : R. W. Schultz in Excav. p. 41. 

he might have used them for further illustration ^ 6. 3. 9. 

of the geometrical truth which he insists upon, 


the hands of the Megalopolitans, and they would have to defend a town twice 
too large for them. And, if the richer and more powerful members seceded, 
the treasury would be no longer able to support an adequate Pan-Arcadian 
garrison, and in this case too the city would suffer. The defection of 
Mantinea was thus a serious blow to Megalopolis ; and ten years after 
its foundation the city itself must have borne the chief burden in holding the 
League together. It was obviously to its interest to do so. The manner in 
which Demosthenes, when he advises Athens in B.C. 353-2 to support 
Megalopolis against Sparta, uses the terms ' Megalopolitans ' and ' Arcadians ' 
as almost synonymous, is highly significant.^ We do not know whether the 
Eparitoi still survived in any shape, but we may be sure that the stress of 
defending the southern as well as the northern wall fell upon the citizens of 
Megalopolis. When the League was dissolved about thirty years later, the 
Federal side of Megalopolis, which had been ever becoming less and less 
important, finally disappeared ; ^ the Pan- Arcadian town south of the river was 
left to the Megalopolitans to deal with as they could or would ; and they had 
at least the consolation of having undivided and undisputed possession of the 
great theatre and the adjoining stadion. The front seats could now be reserved 
for the magnates of Megalopolis, being no longer required for the magnates of 
Arcadia; and the wedges could be appropriated to the tribes of the city. 
We find tribal names inscribed on the backs of some of the front seat-backs, 
in letters which are ascribed to the second century B.C. ; ^ they represent the 
Megalopolitan, just as the inscription of Antiochus represents the Federal, 
stage in the history of the theatre, 

§ 13. The deserted spaces of Megalopolis must have impressed 
visitors by a melancholy sense of the contrast between the high hopes and 
ambitious designs with which Lycomedes and his fellows had gone to work in 
founding the League, and the speedy decay and disappearance of the 
institution which they had called into being. The inhabitants within their 
unmanageable girth of wall must have sometimes felt with bitterness that they 
had been sacrificed to the fond dream of a perpetually united Arcadian nation.* 
Mr. Freeman observes that, though ' the great scheme of LykomM§s, the 

' Op. Demosth. Meg. §§ 30, 31, 32, &c. city after its capture by Cleomenes in b.c. 222, 

- Hyperid., Dem. xvi. ed. Blass, where the see Polybius, 5. 93. The disaster is distinctly 

critical words are unfortunately missing. The ascribed to the size and emptiness {rh fi4yf0os 

internal history of Arcadia is obscure after the ainijs koI t^v ipnulav) of the place. But there 

battle of Mantinea. We find the Federal As- is no hint in Polybius that its population had 

sembly active in B.C. 347 and 344, hearing the decreased since the fourth century. The pillage 

pleadings of Aeschines and Demosthenes (Dem. by Cleomenes reduced the inhabitants to poverty 

F.L. §§ 10, 11, De Oor. § 79). In the war of (§2, itoW&v fxty iiriWta9ai wivrav 8i airaviCtiv). 

Agis and Antipater, B.C. 330, Megalopolis sup- One would have thought that it might have 

ported the Macedonian, and had almost all been feasible to build a new southern wall to the 

Arcadia against her (Aesch. Cles. § 165). Did northern town, along the bank of the river, and 

Megalopolis at this crisis pretend to represent pull down the fortifications of the southern 

the League, and did her opponents meet for town, thus leaving the theatre outside the 

federal purposes at some other centre ? walls. Before the time of Strabo (8, 8, 1 ) the 

' Excav. pp. 123, 124. Great City was 'a great wilderness. Cp. 

* For the proposal to reduce the girth of the Pansanias, 8, 33, 1. 


most promising that any Grecian statesman had yet designed, had altogether 
fallen asunder, his labours were far from being wholly fruitless. He had 
given a model for the statesmen of later generations to follow.' ^ But he 
had also given a warning. The ingenious experiment of a double city 
was not tried again. If the Arcadian Megalopolis had never existed, it 
is not improbable that an Achaean Megalopolis would have been founded by 

J. B. Bury. 

' Federal Government, 2nd eJ., j)}!. 161-2. 


Part V 


This Hymn, whether from the simplicity of its narrative or from 
accident, presents fewer textual difficulties than any of the four larger 
compositions. Serious corruptions there are none, and the notes it is 
necessary to write arc occasioned rather by the misplaced activity of critics 
than by real obscurities in the tradition. 

Literature since 1886 is confined to the contributions of A.' Ludwich, 
Rheinisches Museum 1888, p. 566, and R. Peppmiiller, Fhilologus 1889 p. 13 
sqg. Accounts of the Goddess (which however do not bear materially on the 
Hymn) are given by Roscher in the first volume of his Lexicon, Tiimpel in 
the new Pauly-Wissowa, vol. 1, and by Mr. Farnell, Cults of Greek States^ 
vol. 2. 

13 TTOirjaai craTiva Kal apjiara troiKLKa ')(^aKKm. 

Barnes conjectured a-ariva'i, which has been accepted, for the two other 
places where the word occurs (Eur. Hel. 1326 Orjpojv ore ^vyiov; | ^ev^aaa 
6ea a-artva^ Anacreon fr. 21. 12 vvv B' cTn^aivet aarivecov) leave no doubt 
upon its gender or quantity. It is difficult to see what cause produced the 
omission of the sigma and the (presumable) prosody adrlva. In the two 
passages just quoted there is no trace in any MS. of a neuter; Musgrave 
indeed corrected aaTiva<; from aarivav, but ^vyCov; makes the correction 

I have not kept Barnes' further suggestion re koI, seeing that the 
passages in which koI preserves its length before a vowel, though a small 
minority, are sufficient to guarantee the usage when the MSS. present it. 
They are in the Hymns, the following : 

1 Dem. 275 0)9 elirovaa 6ea /j.e<ye6o<: Kal | 6t8o9 afiecyjre 

2 lb. 424 HaXXa? t' iypefidyr^ Kal \ 'Aprefii'i lo'^^eaipa 

3 Ap. 198 ttXXa fidXa fieydXrj re iSeiv Kal | €lSo<i dyrjTTj 
[lb. 203 fiapfiapvyac re iroScov Kal \ evKXcoaroio p^trcoi/o?] 
[lb. 423 Kal Spvop 'AX<f)€ioio iropov Kal \ cvktitov AZttu] 

24 T. W. ALLEN. 

4 Aphr. 13 TTOCTJa-ai, craTlva[<;] kuI \ apfiara TroiKtXa ■^oKko) 

5 ib. 82 TrapOevo) aBfXTjTj] fjLey€9o<; koI \ eiho<i ofioirj 
[re Kal JJ plerique] cf. 1 and 3. 

6 16. 113 yXaxraav 8' v/jLeriprjv Kal \ ^fier^prjv a-d<f)a olBa 
[contra ib. IIG]. 

7 Arteviis 27. 22 avrap iyoD vfiecov Kal | aWr](i fivijaofi doiBi}<; 

8 Diosc. 33. 19 avrap iyoi vfiicov Kal \ d\Xr)<i /xvija-ofi doi8rj(i 
[contra 3fus. 25. 7, and Hest. 29. 14]. 

Ruhnken (on Bern. 274) endeavoured to make the insertion of re 
absolute, but he is justly resisted by Ilgen on Aphr. 82. Variants on the point 
will be noticed at Aphr. 82, and in two slighter cases Hcrm. 289 d\\' aye firj 
TTVfjLaTov re Kal vcnaTov virvov l(ivr]<; (re om. At D ed. pr.), Aphr. 8.5 €l86<; re 
fiiy€d6<i T€ Kal elfiara aiyaXoevra (re om. N). Outside the Hymns Ilgen 
I.e. quotes T 392 KaWet, tc gtIX^cov Kal eifia<riv, oiiBi k€ <f)aiT}<; and Theog. 66 
fieXiTovraL iravTcov re vofiovi Kal rjdea KeBvd to which I may add Z 211 
ravTrjf; tol yeverjij re Kal a'ifiaro'i ev-)(^ofiaL eh'ai, where re is omitted by 
' H Cant, schol Plat. Gorg. 449 A ' etc., Vat.jg Ven.9. 13 A Mc, Z 478 coSe 
^irjv r' dyadov Kal '\Xlov l<f)i, dvdaaetv, /Birji/ dyaOov re Kal many MSS. 
A 528 Kela iTrirovi re Kal dpfi idvvofiev, re om. ' L * and the rest of this 
family ; O 492 r)8' onvat fiivvdrj re Kal ouk eOeXrjaiv dfivveiv, re om. L7 20- 
Vat.24 Ven.g ; T 417 fiopaifiov earc 6e^ re Kal dvepi l(f)C hafirfvat, re om. 
' L Lips.' N^, Vat.jg, ^ ; H 574 ^pta? avrofieScov tJS' dXKifio^, for 7)8' many 
MSS. have re Kal, Kal alone is found in Ven.jg, Nj, Mg, ^q, Vat.^; Hes. 
0pp. 222 ^ 8' eirerai KXaiovaa rroXiv Koi rjdea Xadv ' sic M 5 Vat. 2 V 2, al. 
rroXiv re Kal! 

52. 609 re ded<i dvifit^e KaraOvrjroU dvOpcovot^!. The correction of 
Schafer, (rvveixi^e, (contributed to Matthiae's edition 1805, praef. pp. vi, vii) 
is easy palaeographically, for to the examples Schafer gave Demetrius' 
correction might have been added, Herm. 94 <^a<; a-vvetreve for <}>aalv ea-eve ; 
but it is hard to see why if the MSS. preserve avvefit^e v. 50 and a-vvifit^a 
V. 250 they should not have done so here. The probability is therefore some- 
what in favour of dvifii^e, for which in a metaphorical sense sufficient 
parallels may be found in the Lexx. 

59 — 63 = ^ 363, 45, H 169, 172. v. 63 dfi^poai^ eav^ ro pa ol reQvwfievov 
Tfev corresponds literally to H 172 except that in the Iliad we have ehav<u for 
kav<fi. The distinction between edv6<i subst. and edvo^ adj., at which Ruhnken 
scoffed, is now firmly established ; we have therefore the choice of making 
an unexampled synizesis of kdv^ dissyll., or (with Samuel Clarke, and not 
either Barnes or Ruhnken, as it is wrongly stated in different editions) 
reading khav<p as in the Iliad ; and the is singularly recommended 
by the variant eavcS on H 172 which is found in Athenaeus 688 E, schol. 
H 346 and Pap. Mus. Brit. Dccxxxii (A. S. Hunt, Journ. of Phil. 
xxvi. ]). 4S). We are to suppose that ehavM, an drra^ elprj/xevov, was 


mistaken by an early but unmetrical scribe or reader for the more familiar 
epic forms. 

62 afxfipoTtfi ola Seov<i eTrevijvoOev aiev iovra*; 

63 dfi^po(Ti<p iavQ) TO pa oi T€0v(ofiivop r^ev. 

The theory of a double recension is no less dangerous than any other prin- 
ciple that is unsupported by direct MS. evidence : the lengths to which it can 
lead may be seen in Kochly's edition of Hesiod. As a provisional measure 
however it certainly tempts application in many cases, and is always 
preferable to the arbitrary and wasteful process of bracketing one line 
rather than another. Here of these two lines editors have inclined to cut 
out 63, on no other ground but that it is the second : GemoU with much 
sense defends it. Similar pairs of lines of which either one or other is 
dispensable are, in this hymn 97 and 98, 136 and 136a, 274, 5 and 276, 7 ; 
Apollo 136-8 and 139, Dion. i. 4 and 6, 7, as previously noticed, and possibly 
Artem. ix. 8 and 9, Aphr. x. 4 and 4a, Heracl. xv. 5 and 5a. 

91 ^Ayj(^ia-r]v 8' e/ao? elXep eTrot 8i fiiv avriov ijvSa' 

Peppmiiller I.e. with unnecessary subtlety would read Td<f)o<i for B' epo? 
in this place. Apart from the marked absence of graphical support offered 
by the tradition, and the asyndeton involved in the alteration, it is surely 
better that the impression made on Anchises should be immediate. 
Aphrodite had arranged her appearance with especial regard to avoid any 
over great respect (vv. 82, 83), and the hero's address 92 — 106 is almost as 
much epic compliment as Odysseus' to Nausicaa. His afterthoughts (185, 6) 
are not to be too literally taken. 

113 sqq. Mr. T3n-rell I.e. p. 48 remarks on the modernity of Aphrodite, 
who explains her knowledge of Anchises' language from her having had a 
Trojan bonne. However difficulties of language are recognised in ancient 
literature : cf. B 804 A 437 Agamemnon 1035. 

136 ov a<f>iv deiKeXirj i/i/o? ecaofiat aXX' eiKVia 
136a €1 rot deiKeXirj yvvrj ea-aofiai ■^k koI ovkL 

Cf. on V. 62 above. Both of these lines stand in all the MSS. ; either 
makes acceptable sense, together they are incompatible, while neither seems 
derivable from the other. We have therefore a fair case for assuming a 
double recension, and the instance is parallel to Apoll. 136-8 and 139. Com- 
pare also Hes. TJieog. 690 and 591, 639-41 and 642. 

172 ktraafievT} B' ev irdvTa trepl "xpol Bla Oedwv 
€<TTrj apa KXiairj, evvoi^Toio fieXdOpov 
Kvpe Kdpt), xdWo^ Bk irapeidav dTriXafnrev. 

This passage and 266 sqq. are the two syntactical difficulties of this poem. 
Here the meaning was long obscured by the faulty tradition of the verb in 

26 , T. W. ALLEN. 

174; Demetrius' correction rjpe was accepted down to Ruhnken, (on /?cwi. 
189) who restored the obviously correct Kvpe from M. Estienne invented Trap 
lor apa, and this was long believed to be the reading of one or more MSS. 
The local dative however needs no defence, and KXiairj is not ' bed ' but ' hut.' 
' She stood up in the hut and her head touched the roof is the sense. There 
remains the asyndeton of 173, and this difficulty is real. The facile altera- 
tion evTToirjTov Sk (due to Ruhnken I.e.) is unsatisfactory since it does not 
suggest an adequate motive for the corruption ; fieXdOpov seems intangible 
and is guaranteed by Dem. 189. On the whole the asyndeton may be 
excusable if we make a longer pause after kXktitj. The case will be some- 
what similar to 267. 

179. olov B^ fie TO irpSiTov. Hermann would omit to, and La Roche 
{horn. Studicn p. 40) /xe, to avoid the ' Attic correption ' — but as the com- 
mentators point out without reason. Contrariwise Artemis ix. 8 avrap iyu> 
<T€ irpcoTa, M inserts re before irpSjTa. 

198 rat hk Koi Alv€ia<i ovo/m eaaeraL ovvexa p! alvov 
€<T')(€v a-)(o^ €V€Ka /3poTov avepo^ €p,7r€aov evv^. 

It is not surprising that commentators have doubted at evcKa in 199, for 
if it be taken as a conjunction the poverty of expression is almost intolerable. 
At the same time no one will wish to substitute Gemoll's ore re, nor the 
attempts of his predecessors, iva kcp (Barnes), evexa ^porov avipo<i kfnreabv 
evvrj(; (' that came upon me on account etc.', the too ingenious method of 
Ilgen, approved by Matthiae), ea-^ a%09 ovvck dpa (Hermann ' certa 
emendatione,' accepted by Franke), on pa (Abel). It has struck me that 
perhaps another asyndeton might be borne : ' his name shall be Aeneas for 
that a dreadful grief is come upon me — for a mortal man's sake efiirea-ov 
eifv^J If this be thought too abrupt we must with Baumeister be content 
with the MS. reading. 

224. ^vaal r atro yrjpa<: oXoiov. To I 446 quoted by everyone since 
Barnes we may add Noa-roc fr. 6. 2 yrjpa<; avo^vaaa. 

252 vvv Se Br) ovKeri p,oi a-TOPW^^^creTat i^ovofjirjpat 
TovTo fier aOavdroKTiv, 

Martini's arop^a 'x^eia-erai, both picturesque and close to the MSS., has 
received fresh support by Mr. Tyrrell's advocacy (I.e. p. 33). Of the other 
suggestions Matthiae's TXijaerat is excluded by metre, as Tyrrell and 
Ludwich (Ithcin. Mus. 1888 p. 566) remark, but Ludwich's own attempts a-T6p,a 
Xij^erac and arop dXuxrirat i^ovop,rjpap are not convincing : Buttmann's 
axva-erai, while admirably near to the MSS., introduces a doubtful form. 
ZTopar eaaerai. (Clarke), aTopM, 'xrjaerai, TretVerat, Xrja-erai (Ilgen), 
XntTiToi (Buttmann, Franke), rfaerat. (Agar), have pleased their authors. 


254 eVet fidXa iroWov adc6i]v 

a')(eTKLov ovk ovorarov, uTreTrXdyx^dijv 8e voolo. 

Martini's ovofiaa-rov has been accepted without question, owing doubt- 
less to the familiarity of the phrase ovk ovofiaarov. The Homeric usage 
however confines itself to the phrase KaKolXtov ovk ovofiaarr^v thrice 
repeated, where the meaning is literally that the ill-omened word "Wia is 
not to be pronounced. In Hes. Theog. 148 rpeL<i -rralhe'i fxeydXoi, koX o^pifioi 
OVK ovofxaa-Tot \ K-OTTOf; re Bpmpeto? re Tvyt]'; 0', the sense of literal 
' naming ' is the same : fr. 44. 7, etp^e Be Btopa \ ttuvtoV, ovk ovo^aard, 
' countless.' The sense of ' unmentionable, horrible ' does not occur til) 
Apollonius iii. 801 nplv rdSe Xcofiyjevra koI ovk opofiaara reXeaaai. 

Now ovorara is plainly a clerical error for ovoracrTd ; the omission or 
insertion of <r in these quasi-participles is universal, e.g. 123 uKTiarov for 
uKTiTov, Herm. 80 davfiaard, Oavfiard, B 592 cvktitov, evKTiarov etc. 
'OvoTa^o) is a word which occurs only in the Hymn to Hermes 30 avfi^uXov 
r]8r) fiot /j,4y' ovrjarifiov ovk ovord^o) and in Hesiod 0pp. 258 o-/coXtw<? ovordi^wv, 
therefore is appropriate in the vocabulary of such a document as this. I 
would therefore be content with ovk ovoraarov ' not to be made light of,' in 
the sense of the familiar evda kcv ovKeTi epyov dvrjp ovoaaiTo ficTeXOoyp and 
many similar phrases in the Iliad and Odyssey, and the participle ovoard I 
164. Aphrodite is not without a certain sense of the effect that her aTt) will 
produce in Olympus. 

It should be noticed also that ovofiaa-Ta is made somewhat less probable 
by the nearness of i^ovo/jurjuai in 252. I see on examining the edition of 
Samuel Clarke (1729) that ovoTaard is recommended, though not put in the 
text. I am glad of the coincidence. Clarke compares € 379. 

264 rfjai &* afi rj iXdrat i^e Spve^ vyJnKdprfvot 
yeivofiivDaiv e<f)vaav eVt ■)(jdovl ^mriaveipr), 
KaXal rrfXeddovaai iv ovpeaiv vyjrijXolcnv 
k<na<7* rjXi/SaToi Tefiivrj Be i KLKXijaKovatv 

The arrangement and correction of these lines have given trouble to 
modern editors, Matthiae and Hermann cut out one or more, to avoid the 
asyndeton of 267 ; Gemoll with the same object inserted B' after ev in 266. 
Franke however decided that all the lines were necessary to the description, 
and made a stop at vy^rjXolaiv. By this arrangement, which will probably 
commend itself to modern readers, the abruptness of v. 267 is to some 
extent excused by the parenthesis which opens there. 'WXi^aroL 267 of 
trees is certainly an extension of Homeric usage, but is sufficiently 
warranted by Hes. Theog. 483 avrptp ev rjXi^dTO), 675 Trerpa^i r)Xi^dTov<t 
a-Ti^apf]<i iv ')(€p<r\v e'xpvre^. Scut. Her. 421 ^ ore Trerpr] [TrevKij ' M 3 '] 
ijXifiaTo<; with Rzach's note. Not more violent is the use of Bvai]Xeyee<: of 
frosts 0pp. 506. 

28 T. W. ALLEN. 

274. It cannot be denied that Trptorov in 278 together with avrUa 
following in 280 makes 274, 5 incompatible with 276, 7. It is evident that on 
the first occasion that Anchises sees the boy he is to acknowledge him, and 
therefore of the two presentations one must exclude the other. These 
considerations afford some ground for holding 274, 5 and 276, 7 alternatives, 
i.e. remains of different versions. 

283 TcS 5^ (TV fiv0ec<T0ai /j,€pLpr}fiivo<! w? <re /eeXevo). 

(fyaaiv roL vvfi(f)r]<i Ka\vKO)7n8o^ cKyovov elvai K.r.\. 

By Matthiae's absurd conjecture <f>dadai, in 284. bolstered up by irrelevant 
quotations, a great deal of humour is lost. The necessary imperative is con- 
veyed by fiv0eia-dai 283 : at 284 commences the statement which the hero 
is instructed to make. The excellent Anchises, homme a bonnes fortunes, 
cannot be supposed to remember his conquests, but with a delightful fatuity 
he does not disclaim his paternity : ^aalv rot vvfM(f>r}<; k.t.X. The spirit is 
the same as in the familiar lines a 215 fMijTrjp fiev re fie ^rja-i rov efi/xevai 
avrap e'ytoye | ovk oi8', ov yap ttw rt? iop yovov avio^ aveyvw, the merit of 
which was recognised by Aristophanes. Cf. also 8 387. 

VII. Dionysus. 

See Crusius, Philologus 1889 vol. 2 pp. 193—228. 

41 ol Be ISovre^ 

fit) S' ijBr) {-eiv ajp) TOT itrcLTa KV^epvijTtjv cKeKevov 
yfj ireXdav 

If we compare Apoll. 393 tjfiaOorjv of all MSS. for what is generally 
accepted as original vfja Ooijv, we may suppose firj B' ^Btj here to represent 
NHAHAH, NHd<HAH, ie. vrf i]Br). The suggestion is Hermann's, the 
older editors down to Matthiae had taken firjBeiBrjv seriously as a patronymic. 
Little is gained by Kochly's i/^a ndXiv or Gemoll's vrjTnirj. 

55 Bdptrei Bie xaTiop tg5 ifiS KC'X^apia'fieve dvficS. 

Here I must confess to absolute impotence. Professor Ridgeway 
{Joum. of Philology 1888 p. 113) maintains xdrcop and derives it. The 
conjectures are mere midsummer madness, — ndrcop (quoted by Estienne * in 
quibusdam editionibus '), Kpdrwp ap. Barnes, aKratp fllgen), iXuTijp (Wolf), 
iKTcop and dxarcop (Baumeister : uKdrap from a.KaTO'i gives at least a sense) ; 
<f}CX€ irdrep (Kochly) which raises the just wrath of Schulze, Quaest. UjJ. 
p. 386 ; 6dpa€i /xrjBert rdpfiei (Gemoll), ddpaei Wvvrcop (Peppmtiller, 
Philologies 1889, p. 22). The termination -cop is used so sparingly in Greek 
to form agents, that it is useless to look for nouns derived from any stem 
such as Kar-, or cKd^; or eKarov, which are suggested by M.'s reading exaTfop. 


Ou the other hand the word may possibly be a proper name, and the survival 
in M be a short form of such a name as e/caTtjvayp, Fick Fersonennamen 
p. 107. In the version followed by Ovid and Hyginus the steersman is called 

XIX. Pan. 

This poem of forty-nine lines has had the advantage of being 
thoroughly discussed by A, Ludwich, Rhein. Mus. 1887 pp. 547 sqq. and 
R Peppmliller, Philologus 1889 vol. 2 p. 1 sqq. It may easily be imagined 
that the third comer has not much to glean. I pass by a certain number of 
alterations which possibly neither learned critic would press to-day, and 
attack the essential points. 

9 aWore fxev peiOpota-iv i<f>e\K6fievo<i fid\aKot(7iv. The tradition is 
sound, certainly ; we do not require e(f)€^6fi€vo<! (Baumeister) nor e(f)aW6fi€P0^ 
(Ludwich), but it may be doubted if the sense ' attracted ' applies. Surely 
a physical notion is more suited to the ungainly god : a semi-humorous term 
for floating — ' hauled, towed,' appears more appropriate. This is suggested 
by A. Matthiae. It is doubtful if peWpov ever really means a bank. 

14 ToVe h' eaTTcpof eKXayeu olov 

aKpr)<; i^apiwp' 

Olou has puzzled the critics and produced a crop of alterations from 
Martini's e^ayev oia(i to Ludwich's cKXayev otfiijp or vfipop. Hermann read 
oIo9, but that Pan has company is expressly stated v. 19. The key is given 
by Hes. Theog. 26 Trot/ieve? aypavXoL, kolk eX.ey')(ea, yaarepet olop. This 
PeppmuUer recognises, though he spoils his effect by the unfortunate altera- 
tions ore h' ea-TTcpof rj Kkdyep olov. Olop = fi6pop occurs Aesch. Agam. 136 
olop fxt] Tt9 dya deoOep KP€<j)d(rr} where the scholiast glosses it fiopop firj ; 
I 355 €p6d TTOT olop €ficfipe as Eustathius takes it, and often in Apollonius, 
ii. 634, iii. 1109, iv. 652, 1077, 1316. Theocr. xxv. 199. Tot^ for Tore (as in 
22, and cf. H 11) would improve the sense, which then is 'he often coursed 
over the hills and often chased the beasts in the glades ; and again would he 
sing, only of an evening, coming back (up) from the chase.' Pan being a 
sportsman waits till the day is over to begin his music. Pierson's correction 
dypr)<i for dKpr)<; is generally accepted ; cf. Theocr. i. 16 an dypa<t \ rapUa 
KeKfjiaKcof afMiraverac of Pan, xxv, 87 e'/c ^ordprj^; dpiopra of sheep, Apol- 
lonius ii. 938 dyprjdep ot ovpapop ela-apafiatprj of Artemis, iii. 69 6ijpr)fi 
i^apicbp of Jason. 

18 Oprjpop i'rmrpo'x^iova-a 'yk^t fieXiyqpvp doiSijp. 

The conjectures are very indecisive, for putting aside iiriTrpoiela-a ')(i€i 
and e-Trnrpox^ovaa lei as violent, Ilgen's a%eet, Ruhnken's ta;j^et and 
Gemoll's »7%eet are much of a muchness. The MS. reading is certainly hard 

30 T. W. ALLEN. 

to swallow, but I have a secret suspicion that it is what the poet gave. 
Would a somewhat poor writer not find justification in phrases like ^vt^i/ 
eVt <yalav ^"x^evav y 258, ypv^ X^°f^^^ ^ ^^' X^" TroXvri-^fea (jxovrjv r 521, Poo'i 
h' aTrehvve fSoeir}p -^ 364, e? 5' dyoprjv ayepovT ApoUonius iv. 214, vpopeeiv 
KaWippoov vhtop h. Apoll. 380 ? 

20 <f>oiT(t)(7ai TTVKva Troaalv eVi Kprjvrj fieXavvBpo) 

I hesitate to alter irvKva into TtvKa witli Ludwich after Barnes {I.e. 551). 
Granted that the change is slight (cf. Quintus xii. 219 where irvKiva and 
-TTVKa are variants on irvKva) the quantity of -kv- is too often short in the 
second syllable of the _ ^ ^ to forbid us to allow it in a poem of the uncertain 
age of this hymn. Cf Theocr. xiv. 23 kuI Xv^vov aip'a^, xvi. 49 airo ^poiof 
KVKvov eyi^o), Theoguis 910 Kal Saxvofiai yfrv^tjv, Anth. Pal. v. 133. 3 6 (70<f>b<i 
KVKvo^'. Quintus iv. 153 &><? kvkvov CKTave, v. 374 oi ol reKva 8r)mcrci)VTai ; cf. 
also Aristophanes Clouds 384 and 40G TruKvoTrjTa in anapaests, Knights 739 
a-avTov Se Xv^voTrcoXaca-i. Pseudophocyl. 158 el Be rt? ov SeSdrjKe Te-)(yr)v. 

22 Saifxcov 8' €v6a Kal evda yopwv rore 8' e? /xecrov epircov 
TTVKvd TToalv Bieirei' 

Kochly's dopcov is too violent even for the awkward movements of Pan, 
and is justly rejected by Ludwicb after Franke. Pan is now outside, on 
either side, of the ring, now inside it. The plural, to which Ludwich objects, 
surfily contains no difficulty : we can as well say ' the dances ' as ' the dance ', 
compare Artemis xxvii. 15 with 18. 

33 OdXe yap it66o^ vypo<; iireXOdav. 

Ruhnken first disturbctl OdXe, which had satisfied the earlier readers, by 
turning it into Xdde, a conjecture which from its false air of graphical 
facility, has reigned in most editions since. Ilgen and Hermann kept the 
original, but the impulse once given produced BdKe, Xd^e, KeXe, and Ludwich 
and Peppmiiller are at one over eXe. AdOe introduces a refinement foreign 
to the extremely simple psychology of Pan, and the other suggestions lack 
palaeographical probability : * desire imperceptibly came upon him ' is hardly 
like Pan ; ' the desire came upon him hot ' is quite in character. For 
OdXXeiv in this sense cf. Soph. Philoctetes 259 and other tragic examples in the 

XXIII.— Zeus. 

2. 6efj,iTi ; corr, Barnes OefiiaTi. The MS. reading as Gemoll observes 
is curious ; it is quoted by schol. Pind. 01. xi. 28 as the reading in O 87, but 
the MSS. give it no support, unless 6efiiBt, Vat.^g be considered as such 
{6efiiBo<{ also in 'J' /3 68). At T 4 and h. Ares viii. 4, there are no 


XXIV.— Hestia. 

4. €p')(€o t6v8' ava oIkov e7rip)(€0 Ovfiov €)(ovaa 
(Tvv Att firjTcoeuTt,' 

Either dvfiov must be sacrificed (as Schneidcwin with evfieviovaa, to 
support which one might bring F 411 'rTopaave-)(^ovaa L.j^, jg Vat. .37, 7p. Ven.j^ 
for TTopaaveovcra, H 342 dfi(f>l<{ iovcra and exovcra), or one must suppose a 
lacuna containing an epithet of Ovfiov, (cf. JJcm. 3G0, 1 Aphrod. 102, Dion. vii. 
49), and to this I inchno. The repetition ep;)^eo — enepx^o is singular, but 
perhaps forcible, cf. Soph. Elcctra, 850 Kayco tovS' ca-rcop vnepiaTcop, Ar. 
Frogs, 369 touto4? avZo) Kavdif aTravSco, Anth. Pal. v. 161, 3 oX^pp^ , epayrei;, 
oXtoXa hLoi'xpP'O'i-, and contrariwise Quintus ii. 314 aXX' dvaxa^eo rrjXe fiodov 
(TTvyepov re (fyopoio j p^afeo /xj] ae ^dXoific k.t.X. 

XXVI.— Dionysus. 

12. S09 S'ljfia? x^ipovraf e? cjpa<; avrit; iKeadat 
e'/c 8'av6' wpdeov ei? tou? ttoWoi)? eViafrou?. 

These expressions which have troubled the commeiitators (Matthiae and 
Franks bracket 13) have probably a general significance and no reference to 
feasts or seasons, cf. Aristophanes Clouds 562, Thesiti. 950, Frogs 381 Theocr. 
XV. 74, It was a popular formula conveying length of life or vague 

XXIX.— Hestia. 

9 sqq. After proposing in the Oxford text a somewhat more elaborate 
arrangement, I have come round to Martini's transposition of v. 9 to after 
v. 11. Although this displacement is unmotived palaeographically, it seems 
necessary since the plural vaiere, v. 9 can hardly follow the singular dyyeXe, 
V. 8. A comma must be placed after elBore^i, and d', coming rather late, 
connects the whole sentence. 

XXXII.— Selene. 

Roscher, Neue Jahrbiicher, 1889, pp. 397 sqq. 

1. M-qin/jv delheiv ravvaiirrepov ea-Trere Movaai. 'AeiSeiv and eairere 
are incompatible ; of the two deiSeiv certainly seems the sounder. Possibly 
the writer mistook the meaning of 'eo-Trere, and thought it meant ' begin,' or 
• follow.' 

XXXIII.— Dioscuri. 

15. KVfiara h^iaropeaav Xef/c^9 dXo^ iv ireXdyeaatv 
yavTai<s a-rjfJiaTa KoXd ttovov a^iaiv 

The older editions, down to Franke inclusive, put a comma after vavrai^, 
and treated <rr)fiara...a(^iat,v as a clause by itself. Baumeister (after 


D'Orville and Matthiae had suspected irovov) was dissatisfied with the 
repetition of a^iaiv after vavrai,^, and joined arjfiara KoXa with what 
precedes : a-(f)ia-iv then became corrupt, and in its place Baumeister 
proposed Kpiaip, Abel \va-iv, Tyrrell ax^acv, and I afieaiv. All these are 
evidently useless stopgaps, and failing some more convincing emendation 
of 7rovov(r<f>i<riv it seems probable we should go back to the old stopping, 
which may not have been intolerable to the poet. 

T. W. Allen. 



For the future any discussion of the problems connected with the Battle 
of Plataiai must take into account Mr. Grundy's careful survey of the field. 
In the map that accompanies his monograph wo have at last reached finality.^ 
The satisfaction of this supreme requirement is his best contribution to the 
subject. His application of strategical principles to the narrative of 
Herodotos ^ is only partial ; and his result is not clear, because he has tried 
to realize the apocryphal portions of the ancient account. It is only 
after stripping off the husk of romantic accretion that we can proceed to 
examine the details by the light of military principles. It is such 
preliminary work and such subsequent recasting of the narrative that is 
here attempted.^ 

Mr. Grundy hits the truth when he suggests that Herodotos obtained 
his information about the operations from an intelligent, but not highly 
placed, officer. Further, Herodotos himself was not primarily a military 
historian. His narrative therefore treats merely subordinate and inter- 
mediate steps as final ends ; and while events are thus viewed only from the 
outside their presentation is moulded by the epic cast of the writer's genius. 
Of perhaps still greater moment is his strong Athenian bias. In the 
recognition of these three factors, — the epic character of the narrative, 
ignorance of the true strategical issues of the situation on the part of his 
informant, and the contamination produced by the sympathy of Herodotos 
with, or his sole reliance upon, the Athenian tradition, — we hold the key to 
the entire account of the campaign of 479 B.C. Some of the details may 
have been derived from Thersander of Orchomenos, e.g. the Phokian episode.* 
It is also possible that Herodotos incorporated in his history local stories 
of the battle. Specimens of these may perhaps be seen in the description of 
the charger of Masistios,** and of the spoil taken from the Persian camp : « 
the three stories which represent the Aiginetans in so poor a light "^ were 

1 See the Battle of Plataea, by G. B. Grandy ; * ix. 16 fol. 
published among the Supplementary Papers of ^ ix. 20. 

the Royal Geographical Society, 1894. « ix. 80 fol. cf. c. 83. 

2 Op. cit. pp. 43 fol. ' ix. 78 fol., Lampon of Aigina urges Paus- 
» I find from Holm's Grk. Hist. ii. 75 (E.T.) anias to maltreat the body of Mardonios : c. 80, 

that Delbriick ' explains the movements of both Aiginetans buy golden spoil from Helots on pre- 

annies on the basis of correct military principles.' tence that it is brass : c. 85, pretended tomb of 

I have not seen Delbruck's book. Aiginetans at Plataiai, 




perhaps also current at Plataial The contents of chaps. 71, 72, 76 probably 
come from a purely Spartan source.^ Not one of these supplements to the 

From. Athens u Thebts 
Site of Persxajx. 


ilodjim, carruigerocLd-3hxnvn,thiiLS =:=^ llb<ferrvt7«cfc» coinrriiaot, wufc-fAose uj«L in. oJici^nttirnes, tKuj i>.u.-.-j» 

I . Rrst position of the Greeks II . Second position of the Greeks . Ill Third position of Centre 

The Battle-field op Plataiai. 
[Based on Mr. Grundy's Survey.] 

Attic core of the narrative has any bearing upon the operations preceding 
the battle. 

' Relating the fate of Amompharetos, Aristodemos, Kallikrates, etc. and the rescue of the 
concubine of Pharandates." 


The operations of the campaign resolve themselves into three strategic 
movements : — 

(1) The occupation of the lines on the slope of Mount Kithairon 
(cc. 19-24). 

(2) The advance to the Spring Gargaplria and the River Asopos 
(cc. 25-49). 

(3) The retrograde movement to the * Island ' (cc. 50-70). 

The key to these manoeuvres lies in the consideration of the roads 
running northwards across Mount Kithairon to the Boiotian capital. These 
roads and passes are clearly described by Mr. Grundy.^ 

(1) In the east there is the road running through the pass of 

Dryoskephalai, familiar to all who have travelled from Athens 
to Thebes by diligence. It enters the range under the walls of 
Eleutherai, and debouches upon the plain just to the east of 
the modem village of Kriekuki? The point at which it enters 
the plain marks the probable site of Erythrai.^ 

(2) The central road from Athens to Plataiai, with a branch to the 

right passing through Hysiai, the site of which, in the main, is 
occupied by Kriekwhi} 

(3) The western road and pass, from Megara to Plataiai. 

(4) Lastly, a road running from Plataiai to Thebes.^ This road 

probably, and the main Dryoskephalai road certainly, crossed 
the Asopos by a bridge. 

On the eastern road lay the entrenched camp of the Persians, and the 
main body of their army, barring all advance northwards. The exact 
situation of the camp is a matter of no importance. It probably occupied the 
bend of the Asopos, lying on the north bank, quite close to the bridge, the 
retention of which was of the utmost moment to the Persians. Their cavalry 
must have lain mainly on the south bank. The disposition of the Persians 
was admirable, posted as they were behind a by no means contemptible river 
in a strongly entrenched camp, covering their communications with a well- 
provisioned base. 

Mr. Grundy's description of the first position of the Greeks is probably 
quite correct.® They advanced over Mount Kithairon, their objective being 

* P. 5 fol. Cf. Leake, North. Or. iL 334 ' Mentioned in Thuc. iii. 24, a passage to be 
and map. discussed later. 

^ But modem traffic now follows the loop to 'And is an improvement apon the generally 

the left, which actually passes through Kriek'&ki. received view, in which Hysiai is put at Krie- 

3 See Grundy, pp. 6, 9. kCki or E. of it, and Erythrai still further E. 

* Grundy, p. 15. Grundy, p. 11 fol 



Thebes. As their ^oint of departure was Eleusis,^ the allies must have 
traversed the easy Dryoskephalai pass. On finding the Persians confronting 
them they threw themselves in extended order across the Athens-Thebes 
road', thus covering their own communications with the Peloponnese and 
taking up a favourable position for defence. For as yet it was quite an 
open question whether Mardonios would not advance to the attack : the 
veriest tiro could not have construed the Persian withdrawal from Attica as 
a confession of inferiority. The Greek right rested on the steep slopes of 
Mount Kithairon : the centre and left seem to have been thrown forward 
somewhat, — probably in order to take advantage of the wells and 
conveniences of the village of Erythrai. 

The success of the Greeks in dealing with the Persian cavalry ^ was so 
pronounced that Pausanias was encouraged to make a change of position.^ 
The inaction of the hostile infantry also contributed to this resolution. Of 
greater influence than either of these reasons was the reflection that for the 
Greeks to remain passive was to play the Persian game. The masterly 
inactivity of Mardonios forced Pausanias to attempt a daring coup. It was at 
least better to die free men on a well-fought field than to survive the 
consciousness that the liberties of Greece had been betrayed by sitting still.* 

The movement contemplated by the Greek commander involved two 
serious drawbacks. The hold upon the main road through the range of 
Kithairon was relinquished, and a descent was made into ground more 
practicable for the enemy's cavalry. Herodotos does not furnish any 
satisfactory answer to the inquiry as to how Pausanias justified his evacuation 
of the impregnable lines of Mount Kithairon. According to the historian, 
the change was suggested solely by convenience of ground,^ — the particular 
convenience not being revealed, with the exception of the more abundant 
water-supply, which was confessedly only one of several advantages. The 
ultimate design of Pausanias in descending from the heights must be given 
by modern conjecture. 

What then was the second position of the Greek army ? 

If we read aright the intentions of Pausanias we can put our finger on 
the line. It involved a descent (iirtKaTa^fjuai), and a movement into the 
territory of Plataiai (e? ttjv UXaraiiSa yrjv). It lay, therefore, N.W. of the 
first position. It was reached via the foot-hills of Mount Kithairon and the 
village of Hysiai {Bta t^? vTrcopirfq tov K.idaip&vo'; irapa 'Tcrta?). The goal 
of the advance lay consequently in the neighbourhood of the Asopos, as is 
clear from the subsequent history. Further than this, two points on the line 

^ ii. 19 : (Tvnniyivrfs 8« iv 'EKfvfflvi. iis 5« teides (c. 13). I cannot follow Holm {Gh-k. 

ipa irlHoyro rrjs Boiwrlrjs 4t 'Epvdpds. Hist. ii. 113) in regarding it as 'altogether 

' ii. 22 fol. : death of Masistios and repulse improbable.' On the contrary it is all of a 

of the cavalry. piece with the conduct of the Athenians during 

* ii. 26 : (io^i a<pi iirtKaraBrivai it UKaraiis. the campaign. 

* Here must be noticed the strange, but in ^ ix. 25 : 6 yap x«poJ i<palvero ■iro\\<f ii>v 
my opinion quite true, tale of Plutarch relative iirirriifeliTfpos . . . rd rt iWor Kal (vvSpdrepos, 
to the Athenian conspiracy frustrated by Aris- 


are expressly named, viz. the Spring Gargaphia and the refievof; of the hero 

Let us take first the Spring Gargaphia. 

Among the low hills on the north of Mount Kithairon there are, 
according to Mr. Grundy \ two springs, and only two, that can put forward a 
claim to the ancient name. The area of the battle-field is marked by a 
distinct depression, which runs from N.E, to S.W. up the Kriekwki brook to 
the bottom of the village, and from that point N.W. to the head-waters of 
the most westerly tributary of the Asopos (stream A^ in Mr. Grundy's map) : 
there it joins the plain, which extends northwards from Plataiai. The two 
springs lie on the line of this depression. The traditional Gargaphia is the 
more westerly of the two, i.e. the modern Apotrvpi, which lies nearly on the 
verge of the plateau, about a quarter of a mile before the Kriekiuki-Pyrgos 
path enters the aforesaid plain. Measured upon Mr. Grundy's map, the 
distance of this spring from Plataiai is 12 stades. The other spring, or 
collection of springs, is found at some distance (on Mr. Grundy's map, 
5 stades) east of Ajyotripi. Mr. Grundy follows Leake in giving the name 
Gargaphia to these last sources.^ They lie 14 stades from Plataiai. 

What data do we get from Herodotos as to the position of the Spring 
Gargaphia ? He gives us the following items : — 

(1) It was 10 stades from the ' Island' (c. 51). 

(2) It was 20 stades from the Heraion, which was ' in front of ' 

Plataiai (c. 52). 

(3) By implication we learn that it must have been about 10 stades 

from the stream called Moloeis, the Argiopian Region, and the 
temple of Eleusinian Demeter (c. 57). 

With regard to the identification of the ' Island,' it will probably be 
generally conceded that Mr. Grundy has made out his case, and satis- 
factorily established the locality to which this name was applied.* More 
valuable, however, is his identification of the temple of Demeter.* No one 
can doubt that its place is marked by the modern Church of St. Demetrios. 
Only with respect to the temple of Hera is hesitation unfortunately possible. 

How do the springs above described square with the data extracted from 
Herodotos ? 

(1) Measurement shows that the distance of the spring Apotrvpi 
from the ' Island,' as identified by Mr. Grundy, agrees more 

^ P. 16. Cf. Leake, North. Or. ii. 332 fol. to 8 etades in a calculated distance of 20.' Cor- 

"^ P. 16. Leake, North. Or. ii. 333. Mr. recting the measurement as above the error comes 

Gmndy states that this spring is 16 stades to a choice between 6 and 8 stades,— an im- 

rom Plataiai. Comparing this with the 12 material difference. Be it remembered also that 

stades of Apotripi he writes, — 'It is easy to the point to which the measurement is taken 

imagine that a mistake of 4 stades was made in (the temple of Hera) is not yet established. 

a measurement of the distance by the eye alone: it ' P. 27. 

is not so easy to suppose that the error amounted * P. 33. 


closely with the statement of Herodotos than does that of his 
Gargaphia, At 10 stades from Apotrijn we are in the centre 
of the Nesos ; whereas, measuring from Leake's (and Mr. 
Grundy's) Gargaphia, we reach a point too far up the slope of 
Mount Kithairon, or else actually find ourselves outside the 
limits of the Nesos, in the direction of the town of Plataiai. 

(2) The uncertainty with respect to the situation of the Heraion 

renders an appeal to measurement here delusive. So far as it 
goes, the result seems to point to an exaggeration of the distance 
on the part of Herodotos. 

(3) Comparison of the interval separating the two springs from the 

Eleusinion is decisive against the claims of the well to which 
Leake and Mr. Grundy give the name of Gargaphia. Measur- 
ing from the Apotripi spring, 9| stades bring us to the Chapel 
of St. Demetrios, 10 stades to the stream flowing along the 
S.E. side of the ridge on which that building stands.^ On the 
other hand, measuring from Leake's Gargaphia, the Chapel 
and stream lie at a distance of only 4J and 5 stades respec- 
tively.2 Yet Mr. Grundy accepts the above-mentioned stream 
as the ancient Moloeis and the scene of the final stniggle. 

We now turn to consider the position of the monument of the hero 

Here we can supplement Herodotos in some slight degree from Thucy- 
dides.' The 212 men who escaped from Plataiai during its investment in 
428 B.C. ran at first ' 6 or 7 stades along the road leading to Thebes, having 
on their right hand the heroon of Androkrates ' : subsequently they turned 
off to the right and fled in the direction of Mount Kithairon, towards Hysiai 
and Erythrai. We notice that whereas Herodotos speaks of a rifievo^, or 
enclosed domain, Thucydides calls it a -qpwov, or monumental chapel. It 
must have been a building standing in the midst of a sacred enclosure, 
which was probably planted with trees. That this was indeed the case we 
learn from Plutarch, who describes the heroon of Androkrates as ' sur- 
rounded with a dense grove of shady trees.' * 

Few can have read the passage in Thucydides without having been 
struck by the apparent pointlessness of his remark as to the position of the 

* Stream A5 on Mr. Grundy's map. iirrk ffrailovs ol nAaratTjj r^v iir\ rStv @r)$&y 

' Yet Mr. Grundy writes (p. 33) : 'It will be 4x<i>p-naav, lirtiff" viroffTpf\fiavTfs fjtffav r^iv wphs 

seeo on the map that the distance from the rh opos <pipovffav b^hv is 'Epvdpas koI 'Tffiks Koi 

position of the Spartans near the spring which \afi6fi(i'oi rwv bpS>v Sia<p(vyov(rtv is rtks 'Miivas. 

Leake (rightly, I think, as I have previously Grundy (p. 10 fol.) quite accurately gives the 

said) identifies with Gargaphia, accords closely route followed by the fugitives. 

with the distance given by Herodotus.' * Plut. Arist. xi. : rh rov 'AvipoKpirovs 

' Thuc. iiL 24 : ol U\arairis ix^povv &6p6ot i)pifov . . &K<Tfi vvkvuv koI avaKlwv Sivdpioy 

T^v is wiBas tpipovffav SShv iv Sffi^ ixovrts rh irfpitxifj^fyoy, 

Tov 'AvtpoKpirovs rip^oi>. . . . koI iirl fiiv l{ ^ 


monument in question. For if the hereon lay hard by the road, constituting 
a famih'ar landmark, it was surely needless to insist upon its relation to 
travellers advancing along that road in the direction of Thebes. A closer 
examination, however, removes this seeming pointlessness. In addition to 
the regular high road from Plataiai to Thebes (4), a man might cross the low 
hills in a N.E, direction ^ and ao strike either the road that issued from the 
pass of Hysiai (2), or the main road from Athens (1) issuing from the pass 
of Erythrai (Dryoskephalai). The remark of Thucydides, that the hereon 
stood on the right hand of the fugitives, thus turns out to possess consider- 
able value.^ It fixes their point of exit from the town to the northern 
section of the enclosing lines, and the route of flight to a northerly direction, 
thus indirectly eliminating the possibility indicated above, — that the exit 
was made on the N.E. of the town and the line of flight continued towards 
the same point of the compass. The corollary from this is that the site of 
the monument should be sought between the line of the Plataiai -Thebes 
road and the line of the path that runs to the north-east : in other words, it 
is an entire mistake to imagine that the hereon lay quite close to the Plataiai- 
Thehes road, i.e. in the plain itself. 

In addition to the passage from Thucydides, we are able to adduce one 
from Plutarch. It is true that, as history, Plutarch's account of the 
campaign is of small value. Nevertheless, the circumstance that Plutarch 
was a Boiotian, and the probability that he knew the ground, combine to give 
some importance to the few topographical details preserved in his Zife of 
Aristeides. It is only by the adoption of a foregone conclusion that his 
testimony is brought into conflict with that of Thucydides. We refuse to 
subscribe to the verdict of Mr. Grundy ^ when he declares that * one has to 
stretch the language of Plutarch until it cracks in order to reconcile his 
topography with that of Thucydides.' 

In describing the movement of the allied army to its second position, 

^ In other words, taking the path chosen by must refuse to acknowledge with Mr. Grundy, 

the Corinthians in their inarch from the Heraion that ' it is evident that Thucydides understood 

to the scene of action, as related in Herod, ix. the 'Hpyoc to be less than three-quarters of a 

69: St^ TTJs virwpfr)s Kul Tuv Ko\uvuv riip (pfpov- mile from Plataea. ' The outcome of this as- 

ffav &v(t> ieh rod Ipov Trjs A^^uijrpoj. Such would sumption is Mr. Grundy's hypothesis of a triple 

not of course be the usual path from Plataiai to phase of the Greek second position. Air that 

Thebes, but it might well have been followed Thucydides says, is that the fleeing Plataians 

by the fugitives, whose objective was not ran about a mile along that road to Thebes 

Thebes, as it had the advantage of bringing which lay to the left, or west, of the monument : 

them nearer the passes into Attica while avoid- that they actually passed the monument is 

ing the obviously dangerous route along the nowhere stated. 

base of Kithairon. '^ P. 35 note. An example of wrong method 

2 The words iv 8f{i$ txovres rh rov 'Avipo- adopted by Mr. Grundy from Leake, North. Or. 

Kpdrovs fipifov are inserted for no other purpose ii. 354, a passage which Mr. Grundy quotes with 

than to define exactly the preceding phrase r^r approval. Mr. Grundy makes much of Plut- 

ii &ii$as (pipovffav di6y, — a phrase which was arch's failure to mention the vfjffos. It will be 

equally applicable to the alternative path men- seen that Plutarch is in the right : the situation 

tioned by me. It is ordinarily assumed that of the tnjffos is of no momentj as we might 

the fugitives passed the monument in question. guess from the fact that not one of the Greek 

For this opinion I can see no warrant, and I ' contingents erer reached it. 

40 W. J. W00DH0USE5. 

Plutarch writes as follows, — ' near Hysiai, at the foot of Kithairon, there is an 
ancient temple of Demeter and Kore, and there hard by was also the heroon of 
Androkrates.^' The natural inference from this is that Plutarch imagined 
the Eleusinion and the heroon to have been fairly close together. Compare 
this with what Herodotos tells us about the enclosure of Androkrates ^ — ' and 
there they ranged themselves, nation by nation, close by the fountain 
Gargaphia and the sacred precinct of the hero Androkrates, partly among 
hills of no great elevation, and partly upon level ground.' What is there in 
this to support the double assumption on the part of Mr. Grundy ^ that 
Herodotos meant to give us the two extremes of the Greek line, and to 
indicate at the same time that the spring lay among the hills while the 
monument stood in the plain, i.e. on the left wing ? The conclusion to which 
both Herodotos and Plutarch point is that the heroon of Androkrates and the 
Spring Gargaphia stood (within reasonable limits) in the same area.'* What 
this area w£is we have already ascertained for the spring. What it was for 
the heroon we have already deduced from the words of Thucydides. The 
two streams of evidence guide us to one and the same point for the site of 
hereon and T€fi€vo<;. That site is marked by the modern Chapel of St. John 
crowning the height which rises immediately to the north of the Apotripi 
(Gargaphia) spring.^ 

What then do we conclude as to the second position of the Greeks ? It 
occupied the depression which Mr. Grundy describes ® as running across the 
battle-field roughly from east to west. Here the allies had the advantage of 
a supply of water in the Apotripi (Gargaphia) spring, — the sources farther to 
the east would obviously also be in their hands ; they were screened from the 
observation of the Persian main body ; they were also protected from the 
cavalry as well as was possible anywhere ofif the actual slopes of Mount 
Kithairon.^ The Greek outposts would occupy the heights to the north of 

* Plut. Arist. xi. : ruv'taiZv irKrialov {nthr'hv of Arch. vol. vi. 471). 

Ki0aip«»'o va6s iixriv ipxa'iof irivu A'fiffnrpos * The same conclusion seems to follow from 

'EKfvfftviai Kal KSprfs irpoffayoptvofifVTjs .... Paus. ix. 4, 2, where the Temple of Eleusinian 

Kirrov 8' ^y koI rh rov 'AfipoKpirous Tjp^ov Demeter, the fivrjfia of Leitos and the Spring 

iyyvs, SX(r«« ■KuKvwv koX (tv<tkIuv BivSpuy wfpif Gargaphia are apparently grouped together as 

xift-tyov. contiguous to one another. We may note here 

' Herod, ii. 25 : inriKSfitvoi 8i iriffcrovro Kar' that Mr. Grundy is altogether wrong in imagin- 

fBvfa irXytfflov t^i t« Kp^vris rfjj Fapyaipiris Kal ing (p. 34) the temple of Demeter here spoken 

ToO rtnivtos ToD 'AvSpo/cpdrtor toO ttpuos 8(i of by Pausanias to be different from that 

ox^wt' Tf oii/c iiffniKwv Kal iwiSov x^pfow. mentioned by Herodotos in his account of the 

* P. 36 note : ' I think that the words of battle. 

Herodotus . . . can only mean that the rififyos ' Taking into consideration what is told lis 

was on the left of the Greek line, for the iwiSos of the heroon by Plutarch in the passage already 

X»pof can only be the plain between Plataea and quoted, I see in the modem name Platdni 

the Thespian Asopus, on which, by-the-bye, ( = Plane tree), borne by the locality indicated, 

according to Thucydides the riiitvos must have a traditional survival of the old Hellenic 

stood.' Cf. p. 17: 'the rintvot of the hero rifxtvos. See Leake's map. The huts round 

Androcratea, which Herodotus tells us was the the chapel have apparently disappeared, but 

other extremity of the line, i.e., on the left the memory of the name remains, 

wing.' The same assumption is made by Stein ' P. 2. 

(ncU in loe.) and Grotc, (Hist. v. 19 note 2), but ' The description of the position and its 

is rightly combated by Mr. Hunt(.4wicr. Joum. advantages, as given in Died. xi. 80, 6 (liv ykp 


the position, viz. the height on which the heroon stood, and the eminence 
lying to the east, between the hereon and the temple of Demeter. 

The object aimed at by Pausanias in removing from Mount Kithairon is 
rightly stated by Mr. Grundy.^ The Greeks tried to effect a great turning 
movement by their left. They threw themselves upon the Plataiai-Thebes 
road, intending to force the passage of the Asopos and to cut the Persian line 
of communication. Mr. Grundy justly calls attention to the fact that the 
military capacity of Pausanias is universally underrated. For boldness 
of design, prudence in execution, and power of handling masses of men in the 
face of almost insuperable obstacles he deserves a high place in the list of 
Greek generals. Under the conditions of ancient warfare the undertaking 
was not as desperate as it would seem.^ In the absence of long-range 
weapons and arms of precision, it was perfectly feasible. Moreover the 
advantage in skill, discipline, and equipment was overwhelmingly in favour of 
the Gi'eeks.^ The Persians might well have been driven eastwards off their 
line of retreat. It was necessary, however, to take precautions against the 
Persian cavalry, which was massed on the Greek right flank, at a distance of 
at most three miles^ A sort of Echelon formation was therefore adopted, the 
Greek contingents being disposed obliquely from S.E. to N.W. across the roads 
leading from Plataiai to Thebe?. 

It is at this point that we begin to find the narrative of Herodotos 
interrupted and distorted by the national bias of his Athenian informants. 

Here for the first time the historian directs our attention to the 
disposition of the Greek troops. He goes off at the word Kar'iOvea (c. 25) 
and introduces the quarrel between the Tegeans and the Athenians for the 
post of danger and honour on the wing (cc. 26-28 init) The whole of the 
story must be excised, on the following grounds : — 

(1) The left wing of the Greeks in the second position lay 
on iTTTrdaifio'; p^wpos.* It might consequently expect to suffer 
from the attacks of the hostile cavalry, as was actually the case 
(c. 49 end). How then reconcile the Tegean demand for 
station on the left wing with their previous reluctance (shared 
by the whole army) to support the Megarians against the 
Persian cavalry in the first position ? ^ It is not sufiicient 
to advert to the success already gained against the cavalry 

iK fiiv Twv Bf^iuv ytw\o<poi v\iiri\6s, 4k S« tuv river Nebel and the marshes on its banks. 

fvaivvfiCDV & 'Kcrwirhs irora/xSs- rhv 5' i.vb, nfffov Compare also the passage of the Granikos by 

t6'kov iirux*'' V inpaToirtSfla, vt<ppay^(vr} tj) Alexander. 

<p{)(T(i Kol Tois rSiv rSiraiv a(x<pa\elais), is clearly ' Cf. Holm, Grk. Hist. ii. 75 (E.T.) : 'the 

simply modelled upon that of Thermopylai, and Greeks were well-handled bodies of heavily- 

cannot be pressed into service here. armed infantry.' 

1 Pp. 22, 43. * Cf. ix. 25 : Si' axfSov xop^ov ; 31 : iit\ r^ 

^ The situation finds its counterpart in the 'kffooir^; 49: 6 St ' Kawirhs iyxov. 

battle of Blenheim. The Asopos did not con- * ix. 21. 

stitute a more formidable obstacle than did the 


(c. 25) and to the confidence thereby inspired, for the service 
now demanded was much more than steadiness against 

(2) There is no evidence to support the statement made, according 

to Herodotos, by tlie Tegeans, that post on a wing was their 
prerogative. Subsequently at any rate we find the Tegeans 
occupying precisely the station finally allotted to them on the 
field of Plataiai, i.e. next to the Spartans themselves. This is 
the case in 418 B.C. at the battle of Mantineia,^ and in 394 B,c. 
at the battle of Corinth.^ 

(3) How was it that the Corinthians, 5000 strong, did not raise 

objections if they were moved from the side of the Spartans, 
presumably a post of honour, in order to make room for the 
Tegeans ? 

(4) The Tegean demand, if ever urged, must have been decided 

instantly by tactical considerations. A large compact body, 
like that of the Athenians (8000 in number), which was 
accompanied by the best light troops in the army (archers), was 
required on the wing, not the Tegean handful of 1500. 

(5) The story of Herodotos is irreconcilable with the words which 

occur in chap. 28, — ' The place next to themselves was given 
by the Spartans to the Tegeans, on account of their courage and 
of the esteem in which they held them' * These words suggest that 
their actual place in the line was assigned to the Tegean 
hoplites in pursuance of some plan not given in Herodotos. 
The nature of the plan will clearly appear in the sequel. 

(6) The quarrel, if a genuine incident, must have occurred earlier 

than is stated by Herodotos. It must in fact have broken out 
at the moment of taking position on Mount Kithairon. For 
the evidence goes to show that there also the Tegeans had not 
been posted on the wing.* 

* Thuc. V. 71. * In the second position the Megarians are 

• Xen. Bell. iv. 2, 19. And at that battle third in the line, reckoning from the left, i.e. 
of Mantineia in which Epanieinondas fell in they stand next on the right of the Plataians and 
362 B,c. the Tegeans apparently stood next to Athenians. This place apparently corresponds 
the Thebans, i. e. the leaders, in this case on the to that which they held in the first position : 
left wing. Cf. Diod. xv. 85, 2 : eit&aloi 8' for there also they occupied the left centre (cf. 
aiiTol (xkv i-K\ rb tbdvv/xoy Kipat ^TC^x^'/ffa*', ix. 21 : tUcyapies (rvxov raxOivrfi rp t« rh 
wapaffrdras txovTtt 'ApKdids, rh 8< 8((ibv irapi-<'>'TaTov ^v tov X'^P^o*' irai'Trfj, koI fi irp6ix- 
SuKav ' Apyfloii k.t.X. The point is that they oSos fidKiara ravrfi iyivero rp 7Triry, — this can 
stand shoulder to shoulder with the premier only have been on the left and left centre of the 
corps, whatever its position. line). To this we ought to add the considera- 

3 ix. 28 : irpofffx'a* ^i <r<pi cTAovto Icrdvai ol tion that, if the Tegeans had been on the wing 
3vapTi^Ta< rovs Ttyfliraj Koi rifi^i tIptKtv Koi in the first position, they would have urged 
AptT^j. that as an argument here. 


On these grounds we unhesitatingly reject the story of the quarrel. It 
is an Athenian invention designed to flatter Athens by means of a verdict 
put into the mouths of the best troops in Greece, at the expense of a 
contingent second to none in valour (c. 28).^ 

In fact, the whole account of the marshalling of the Greek troops comes 
far too late in the narrative. Their arrangement in the line must have dated 
from the opening day of the campaign. With it disappears also the account 
of the marshalling of the Persian forces. There was no such formal parataxis 
as Herodotos depicts. The place of the account, which is closely modelled 
on the epic, is determined solely by artistic reasons, without reference to the 
logic of military practice. It is inserted precisely at this point because we 
have reached a crucial stage of the campaign ; but the arrangement of the 
troops strictly belongs to an earlier moment, while the quarrel to which it is 
represented as giving rise is a pure fiction. 

Having thus adopted from his epic model a quite artificial scheme of 
events, how does Herodotos proceed to develop it ? Here we have the two 
armies ranged and described in battle array, but— nothing comes of it. Re- 
course is had to the sacrifices in order to explain the refusal of the combatants 
to finish the business, thus happily begun, in the true Homeric fashion.* 
Herodotos is manifestly quite in the dark as to the real reason for their delay. 
His assertion of the only obstacle that would appeal to his hearers, — the 
persistent veto of heaven, — involves him in difficulties, as it directly contra- 
dicts the account given in chap. 41, which relates the conference of the 
Persian officers. For if Mardonios was so eager to fight,^ why had he not 
long ago given battle ? It was surely not out of respect for the feelings of 
the Greek contingents fighting on the Persian side that he had conformed to 
the utterances of their soothsayers. Why should Mardonios summon hia Stafif 
only to insult it ? The episode of the conference is inserted for no other 
purpose than that of enabling Herodotos to contrast dramatically, more suo, 
two antithetical solutions of the situation, — on the one hand decisive battle 
for good or ill, on the other the sound policy of waiting for disaffection and 
bribery to do their fatal work upon the national forces.* 

Next there follows the account of the midnight visit of Alexander of 
Macedon to the Athenian lines.^ This also is a story full of improbabilities, 
and without any claim to retention. How did Alexander escape recognition 
at the bridge-head held by the Persians ? Or, if that is supposed to be no 
difficulty, how did his errand elude the notice of the Persian sentinels ? If 
again these imagined him to be the bearer of despatches to the Greeks, where 

^ The turn of expression in the concluding impression of delay in the action, 

sentences (chap. 28) is designedly invidious, — ' ix. 37 ; MapSovl(f> Si rtpodufitofiivif fiixv 

AOrivalovs i^toviKorfpovs ehai Ix*"' ^b Kfpas ipx*'*' o"*^ tirtriiSta iylvtro rh. tpd. Cf. chap. 

Ijirtp 'ApKiSas. Who does not recognize the 41 : MapS6vios -irtpirififKTff Tp eSpi). 

curl of the lip in this ? * Partly also Herodotos design to give 

2 Observe how skilfully the history of the expression to his own opinion on the situation, 

various soothsayers (ix. 33-38 init.) is used in ' ix. 44 fol, 
order to interrupt the narrative and to give the 


was the risk of which he makes so much ? ^ The reasons which he alleges to 
account for the Persian delay in attacking are very obviously put into his 
mouth by Herodotos himself in conformity with what he has already written 
in chaps. 36, 37. The assertion that the Persians found their commissariat 
breaking down is a manifest lie.^ The very emphatic and artistically well- 
managed revelation of his name on the part of Alexander was quite super- 
fluous to Aristeides, who must have become familiar in Athens with the face, 
figure, and tones of the Macedonian king.^ Lastly, the whole point of the 
clandestine interview was to warn the Greeks of the intention of Mardonios 
to fight a decisive battle on the morrow.* Yet, in spite of the alleged 
eagerness of the Persian general and the difficulties threatening his army, 
the following day passed without any serious attempt being made to justify 
the Macedonian's prognostications. 

The excision of the nocturnal visit of Alexander necessarily involves also 
the abandonment of the disgraceful story contained in chaps. 46, 47. Ac- 
cording to Herodotos, the near prospect of encounter with the Persians and 
Medes so alarmed Pausanias that he suggested to the Athenian leaders an 
interchange of position on the part of their respective divisions. The 
Athenians moved to the right, while the Spartans withdrew to the left in 
order to face the Boiotians and the other Greeks who fought in the ranks of 
the Great King. The exchange, however, was detected by the Boiotians, who 
at once informed Mardonios. The Persian troops were consequently trans- 
ferred to the right of their line, so as to bring them once more in front of the 
Spartans. Pausanias then for the second time changed his position, and 
resumed his post on the right wing. Finally, the Persians returned to their 
old station, and the farce was brought to an end. 

' No incident similar to this,' remarks Grote,^ ' will be found throughout 
the whole course of Lacedaemonian history.' He might safely have gone 
further and denied that any such incident ever did occur. From beginning to 
end the story must be stigmatised as a slander. 

(1) If the Spartans had contemplated the movement at all, for what 
had they delayed its execution ? They could not have foreseen 
that they would receive timely warning of the approaching 
battle, nor yet that the Persian onset would be retarded long 
enough to enable the change in position to be made. It is 
evident that the proposition was only possible upon a very 

* ix. 45 : ts 'EW^ivtDV t'lvtKtv ovrw tpyov ' On the occasion of his visit to Athens as 

wapiBo\ov fpyafffj-ai vrh wpoBvfilris. special envoy from Mardonios, Herod, viii. 136. 

- ix. 45 : 6\lywv yap afi rififpiwu Kdirtrai Alexander we there read had a compact of 

ffiTia. How is this to be reconciled with tlic friendship (npo^fvia) with Athens, 

words of Artabazos — ix. 41 : rh rfixos rh * ix. 45 : vvv St ol SfSoKrat to fiiv <T<pa.yia iav 

^TiBaitiiv, Iv6a (t7t6v ri ff<pt ^trtVTjftrxSat -rroWhf xalpfiv, a/xa T/iutpTj 5f Siai^aicrKoiJtrfj (ry/u/3o\V 

Kal x^P^o" "Tolcri i/iro^vytoiai '( Cf. Rawlinson, iroif(ff6ai. Cf. chap. 42 : i(Ti]ixT\v( irapaprifcrOai 

iv. 412 TWte 9 : 'it is evident from their whole rt -nivra koI tvKpivia ■KoiitaQai in afxa Tjfifpp rfi 

history that the commissariat of the Persians iTttovff^ ffvu$o\ris iffofi.ivi\%. 

was excellently managed.' • Eiat. of Greece, v. 25. 


general and decided feeling in its favour on the part of the 
Spartan hoplites ; hence it cannot be set down to a sudden 
nervousness depriving Pausanias of self-command. 

(2) It was surely a strange preparation for the decisive struggle, 

fraught with such grave consequences for Greece, to march 
and countermarch the best regiments of the allied army in the 
face of the enemy. 

(3) What was the effect of the Spartan cowardice upon the mass of 

the Greek troops ? The motive of the manoeuvre must, one 
thinks, have been as apparent to the rank and file of the con- 
tingents as to the Athenian hoplites. 

(4) How is it that we never subsequently hear a syllable of this 

compliment to Athenian arms ? 

(5) The genesis of the story can be traced quite satisfactorily. 

As the last of our long: series of excisions we must abandon the incident 
narrated in c. 48. Mardonios sent a herald to the Greek lines with an absurd 
challenge, proposing that the Spartans and the Persians should fight on 
behalf of all. It is obvious that if the attempted change of post is cut out 
it must carry with it the challenge. It is modelled upon similar scenes 
in Homer; but it is also not uninfluenced by reminiscences of previous 
history.^ Apart from this, the narrative is intrinsically unsound; for how do 
the words ' puffed up by the empty victory ' ^ square with the statement that 
nothing more was attempted or achieved against the Greeks than the 
usual harassing attacks of the cavalry? We expect some deed of arms to 
redeem the doughty resolves of chap. 41. 

Now that the ground has been cleared of the excrescences due to 
Athenian light-hearted manipulation of history let us resume the interrupted 
story of the Greek movements. 

We have surmised that the movement of the allies to the second position 
was based upon something more than the desire merely to obtain a better 
supply of water : for the abandonment of their main line of communication 
and the greater exposure to the Persian cavalry on the lower ground were 
attendant drawbacks too serious to be counterbalanced by the single 
advantage named by Herodotos. Pausanias had determined to make a dash 
across the Asopos by the road which ran directly from Plataiai. The second 
Greek position represents the army in the act of carrying out this manoeuvre. 
It is disposed obliquely across the field, the left wing leading upon the 

^ Cf. Horn. II. iii. 90 fol. Combat of onset waa designed to introduce the infantry 

champions was unsuccessfully used to decide attack, why did that attack not ensue in due 

the claims of Sparta and Argos to the Thy reatis, course? Confessedly (according to Herodotos) 

Herod, i. 82 (Thuc. v. 41). the cavalry were more successful this day than 

^ ix. 49 : i 5i trfpixap^s yev6fi.evos koX ivatp- ever before. 
fl«lj ^vxprf vIkt) iirrjKe r^y 'l-rnrov iit\ robs ^ See Note A on the Asopos of Herodotos. 

''%KKT\va.s. If it is argued that the cavalry 


Why then was the offensive designed by the Spartan general not 
developed beyond this point ; and why do we not find in Herodotos a 
syllable in allusion either to the scheme itself or to its collapse ? 

The first difficulty is solved by reference to the position of Pausanias. 
Tlie army under his command consisted practically of three brigades 
constituted respectively by the Spartans (with whom we must reckon the 
Tegeans), the Athenians (along with the Plataians), and lastly the general 
body of the allies. The loose structure of the Greek national levy made 
unanimity in sentiment and cohesion in action impossible beyond certain 
narrow limits. Hence the delay in accomplishing the passage of the 
river, a delay that ruined the scheme, and all but ruined the national 

The latter part of our question is answered by reference to the ignorance 
of the historian's informant, who was quite in the dark as to the strategic 
ideas of the Greek commander-in-chief. 

Another cause also is at work. It must be remembered that the 
campaign was a national affair, and it was undoubtedly a point of national 
honour to present it in the most favourable light. By tacit general consent 
the battle never became the subject of discussion. An analogy may be found 
in the medism of the Delphic oracle, which yet, by a species of national self- 
deception, did not forfeit its claim to Hellenic respect, in spite of its failure in 
the hour of trial. ^ So in the case before us, no Greek would have been so 
unpatriotic as to confess that dilatoriness and cowardice on the part of the 
national army had nearly proved fatal to Hellenic freedom. 

We must also bear in mind that our knowledge comes almost entirely 
from the Athenians, and only from a certain section of them, so that we know 
scarcely anything of the views current outside Athens.^ In spite of Athenian 
reticence, however, we clearly see that hesitation on the part of the Greek 
force, and more especially on the part of the Athenian contingent, which was 
in the van, enabled the Persians to divine the intentions of the Greek 
commander, and gave them time to perform a lateral movement in order to 
cover the Plataiai-Thebes road. Their clouds of skirmishers then eflFectually 
prevented all approach to the Asopos, and the favourable moment was 

This brings us to the origin of the story, already condemned, which is so 
discreditable to the Spartans. 

The Athenians, being on the left extremity of the line, which rested on 
the Asopos, would cross the river at the head of the column. After crossing 

* Cf. Holm, Chrk. Ui»t. ii. 60 (E.T.), lining the Asopos banks are regarded by the 
■ Can we, for example, believe that the tradition as designed to entice the Greeks 

Athenian version of the retirement of the centre across the river (chap. 40 : m^xP' M^'' f^P toS 

to the Heraion (ix. 52) passed current among 'Ao-wiroC iin/jXaav oi 0dp$apot, irtiptSfifyoi rtiv 

the states whose troops were implicated in that '^wiivwv). Such are the marks of a literary 

movement 1 battle, not the touches of a man versed in the 

* ir. 49 : ipuK6fit¥ot ii 4iri rov 'Atrteirov. By actual experiences of the field, 
a strange inversion the Persian skinnishew 


they would wheel to the right, in order to check the Persian advance along 
the bank to hinder the passage. The Spartans, who were posted on the 
extreme right, formed the rear-guard of the column, and covered the crossing 
from the Persian cavalry, — a most dangerous and responsible position, and 
one that explains why the valiant Tegeans were associated with the Spartan 
hoplites.^ When the whole Greek force had made good its footing on the far 
side of the Asopos, the Spartans would naturally form the left wing of the 
new line. It is on this reversal of position, — one suggested, but never actually 
realized, — that the Athenian misrepresentation is based.^ It contains this much 
of truth, that the brunt of the fighting, until the Spartan rear-guard efiected 
its passage of the river, must have fallen upon the Athenians, who were 
required to sustain the whole weight of the Persian attack upon the head of 
the column.^ There was surely honour enough in that to have rendered 
superfluous the sorry attempt to cast shame upon the best troops in Greece, — 
the more so as it was entirely due to the Athenians' own want of resolution 
that the Spartan valour was not put to the test contemplated by 

The warp of the tissue of these fifty chapters is the green thread of 
Athenian jealousy of Sparta.* It is a highly suggestive fact that we find both 
the Spartans and the Tegeans, — who shared the honour of the final victory, — 
more or less skilfully represented in Herodotos as inferior to the Athenian 
troops. And in each instance we have been forced to the conclusion that the 
episode is false and due to Athenian vanity. The cloven hoof is unmistakably 
displayed in the account of the events following the challenge feigned to have 
been thrown down by Mardonios. With what painful circumstantiality are 
we assured that it was to the Spartans, and the Spartans alone, that the 
thanks of the allies were'due for the destruction of the Spring Gargaphia ^ : 
as though to give point to the alleged reluctance of the Spartans to face 
Persian infantry by instancing this, probably equally fictitious, failure to stand 
against Persian cavalry.^ If these things were done in the green tree, what 

^ ix. 28 : ■npofffx^"'^ ^^ <''0' cT^octo icTivai oi is merely again the self-laudatory Athenian 

5irapT»^Tai rohs Teye^rai koI Tifxris fXvtKty Kol tradition. 

iperris. * Contrast the reiterated jubilation found in 

'■^ Possibly also the Persian change of position, our Athenian sources over the victory at Mara- 

from the left to the right wing, is a genuine thon with the silence observed with regard to 

incident : the change might very probably be the brilliant achievement of the Spartans and 

actually made in order to meet the threatened Arkadiana at Plataiai. 

advance of the Greek left. ' ix. 49 : liffav fiiv iv itari t^jv Kp4\vnv AuKf- 

^ This is the truth underlying the garbled SatfiSytoi Ttrayfiivoi ftovvoi. Here again the 

account in Plutarch of the grumbling on the phrasing is used with set purpose, 

part of the Athenians against Pausanias. ' They ® Here I may say that I sec no sort of 

thought that Pausanias carried it with a partial evidence for Mr. Grundy's laboured hypothesis 

and high hand in moving them up and down, of three 'developments' of the Greek second 

like so many Helots, at his pleasure, to face the position. His theory leaves him with 100,000 

boldest of the enemy's troops.' This surely men huddled on a single hill, cut oflf from 

alludes to the disposition of troops previous to water, harassed by cavalry, and with morale at 

the crossing of the river. Plutarch's sequel zero point. Surely this ' development ' could 

(the speech of Aristeides and consequent con- issue only in tragedy. 
sent of the Athenians to change their position) 


would have been done in the dry ! Recall also the invidious expressions used 
with reference to Spartan duplicity, so different from the manly and 
straightforward, withal modest, character of the Athenians,^ and the 
reference to the by this time threadbare theme of Spartan cowardice, so 
glaringly in contrast with the calm steadfastness of conscious valour that 
glowed in the breast of the Athenian hoplite.^ Nay, the Spartans must be 
flouted even at the price of complimenting an almost equally odious people. 
Therefore is it recorded that the Tegeans charged the Persian rampart of 
shields before that the Spartans advanced a foot^: far be it from the 
Athenians to see any other city deprived of its meed of honour for the sake of 
other than — themselves ! Lastly, what prominence is given to the Athenian 
share in the assault on the fortified camp.* Well might this be so, else were 
the hoplites of Athens like to have been but sleeping partners in that day's 
achievements. Here as so often, the Lakedaimonians were baffled by the 
combination of barricades and stout defence. Not until the invincible 
amalgam of Athenian valour and resolution (aperr) koI Xnrapirf) ^ was applied 
could any impression be made on the fortifications. Into the breach there 
rushed, not the Spartans, — alas for that national defect of ponderosity, — but 
the Tegeans. 

With the end of chap. 49 there comes a change in the nature of the 
Greek operations, — a change from ofifensive to defensive tactics. The allied 
army, having lost the opportunity of turning the Persian position, is reduced 
to its old attitude of covering the approaches to the Peloponnese, and of 
waiting for Mardonios to take the initiative.^ 

The real objective of the movement of the Greeks to the 'Island' was 
the recovery of their line of communication, upon which they had then but 
precarious hold. They were, it is true, not driven entirely off it, for, as 
Mr. Grundy points out,' the Plataiai-Megara pass (3) still remained in their 
hands. Nevertheless, according to Mr. Grundy, the character of the most 
westerly pass is such as to render it impossible to supply satisfactorily the 
wants of one hundred thousand men through this channel alone. That the 
occupation of the eastern passes by the Persian advanced posts had begun to 
tell upon the Greek forces may readily be believed, but Herodotos himself 
represents the determination to fall back as due primarily to want of water,^ 
in consequence of the failure of the Spartans to protect Gargaphia. 

' ix. 64: rh, KaKthaifiOvlaiv (ppovi)fjiarai)tA.X\a St ff<pi ol 'AOrivaioi irpoffrjKdov .... riKos Si 

^pov*6vra>v koX &XAa \ty6vriiiv. For Athenian iptrp re koX Xiirapiji i-nifiriaav 'Mrivaioi rov 

mock modesty, see chap. 46 end. rtlxfos k.t.\. 

''' ix. 56: 'ABjivoioi 8i TaxBivrts ffiffav t4 * This * bull-dog obstinacy' is precisely the 

lnira\tv fi AaKtSaiix6viot. ol /liy yh.p rSiv rt quality usually attributed to the Spartans. At 

ix^i^v i.vT*ixovTo KalTTfi \nt<jDpii}iTovKiOaipS>vo%, any rate Thucydides recognizes this, — v. 73: 

<poRt6^Lfvoi tV Xtitov, ' K6t)vaioi St Kdrw rpa<p- xpoi'foui rij A^^X"^ '^"■^ fi($alovi r^ fiivtiv 

Bivrt% ii rh irtSiov. Troiovvrat. 

^ ix. 62 : -wpot^avaffTiyTts -npSrtpoi ol TtytfjTai ® See Note B for the chronology of the events 

ix^p*ov 4s Tovt Pap^ipovs. preceding the battle of Plataiau 

* ix. 70 : ?a»j niv yhp imrtvav ol 'AOi)valoi, ol ' P. 32. 

8' iifiivovro kolI woA.A# T\ioy tlxov ruv AaKtSat- ^ ix. 50 : St« rov rt SSaros ortpi}B(i<Tiis t^i 

liovimv &aT* oiiK iiri<rraix4y«»v TC<xo/iax^«<v, us ffrpartrjs. 


This repetition of tlie water difficulty we should be inclined to reject here 
again, at any rate as furnishing the ground of the retirement. For, wherever 
we place Gargaphia, the army had still the other spring at its command ; and 
Herodotos admits that at the foot of Mount Kithairon, ten stades or so in the 
rear of the position, water was abundant.^ There was also the water supply 
of the town of Plataiai itself.^ For surely the Greeks ought not to be 
imagined as cut off from Plataiai and the base of the hills, and hopelessly 
surrounded by the Persian horsemen. There cannot have been any grave 
difficulty in supplying the needs of the troops in line on the Spring Gargaphia,^ 
as the country between that position and the mountain is by no means 
difficult. The stress laid upon the deficiency of water, if not due to the 
character of the historian's informant, has its origin in the desire to bring in 
the Spartans as ultimately responsible for a retrograde movement primarily 
caused by the Athenians themselves. 

The main features of this last act of the drama, as given by the Athenian 
tradition, are as follows.* 

The council of generals determined to execute a night movement to the 
rear, the so-called ' Island ' being given as the rendezvous of the contingents. 
It was furthei; resolved that, on the same night, half the army should be 
detached eastwards to Mount Kithairon, in order to extricate the commissariat 
train blocked up in the pass. When the appointed hour arrived the centre 
fell back, — not to the ' Island,' i.e. 10 stades, but 20 stades, finally taking post 
at the Heraion, which lay 'in front of Plataiai, Next, the Spartans were 
ordered to retire ; but the irrational obstinacy of the Lochagos Amompharetos, 
who construed the movement as a flight, detained the Spartan contingent all 
night. Meantime the Athenians, suspecting the Spartans of a desire to play 
them false, remained in position on the left awaiting definite instructions. 
As day dawned, Pausanias at last abandoned his recalcitrant captain to his 
fate, and set his troops in motion ' along the line of the hills,' The Athenians 
also retired, by way of the plain. After marching 10 stades Pausanias halted 
for the Pitanate regiment under Amompharetos on the stream called Moloeis, 
near a temple of Eleusinian Demeter, in the district called Argiopian. 
Simultaneously with the appearance of Amompharetos the Persian cavalry 
swooped down upon the Spartans and Tegeans, to be followed soon by the 
Persian infantry. 

Such is the narrative of Herodotos, deceptive in its simplicity and 
apparent straightforwardness. Closer examination reveals in it the features 
with which we have become familiar. On the one hand Herodotos fails to 
appreciate the significance of the various movements of the forces, on the 

' ix. 51 : is Tovrov Sh rhv x^pov i&ovXfvaavro ties of the Helots and light troops in attend- 

pLtravatTrrivai {sc. the ' Island '), 'iva ku.\ vhnri ance on the hoplites ? They were apparently of 

Ixwc* xpS"'**" i^<p66v(f. not the slightest use against the cavalry. We 

"^ Assuming that the town had other sources may remark here, by the way, that no one can 

of supply than the springs commanded by the take seriously their numbers as given by Hero- 

' Island.' dotos. 

3 For what other service engaged the activi- * vs. 50-57. 



other he has incorporated all that national vanity, with the double object of 
glorifying Athens and disparaging Sparta, had invented. 

Up to this point in the story the central brigade of the allies has escaped 
Athenian calumny ; its share in events is shadowy, but not actually dis- 
graceful. Its turn has at last come. Although the troops of the centre had 
borne the heat and burden of the day that proved so disastrous to Spartan 
prestige, yet now, under the cloak of night, they flee in headlong haste, eager 
only to secure themselves against the dreaded cavalry. Mark, however, the 
point wherein the narrative halts. In spite of their anxiety to put themselves 
beyond the reach of the Persian horsemen, the contingents of the centre do 
not seek shelter in Plataiai itself nor on the rocky slopes of Mount Kithairon 
(which ultimately became their refuge ^), nor yet on the ' Island,' — a position 
admittedly outside the sphere of cavalry operations,^ — but they take up their 
station, apparently in good order, 'in front of the temple, which was itself 
' in front of the town. '^ 

There are several possible sites for the Heraion.* The most probable one 
stands within the circuit of the existing enceinte of Plataiai, just to the east of 
the akropolis. The question of the site is of far greater moment than is the 
identification of the ' Island,' which was in fact never reached by any of the 
Greek force at all.^ Its importance lies in this, that, knowing the exact site 
of the temple, we should be able to decide what amount of credence should 
attach to the Athenian account of the conduct of the troops composing the 

That account can hardly be accepted as it stands. It will be observed 
that the suggested site of the Heraion lies at no great distance ^ from the 
tract of ground which is convincingly identified by Mr. Grundy as the 
' Island.' The Heraion may well, therefore, have been actually t?ie position 
which the central brigade was instructed to occwpy. Its proximity to the town ^ 
is an important feature ; it was surely of some moment for the Greeks to 
retain possession of Plataiai, which was a fortified place commanding the 
entrance of the pass to Megara. In order to carry out the project of 
Pausanias it was essential to dispose the various brigades in such a way 

> When cut to pieces by the Theban cavalry, as we can see not a man betakes himself thither. 

^^- ^^' * See the American Journ. of Arch. vol. vi. 

* ix. 51 : ^j rovTov 5^ rhv x<'>pov («c. the (1890) p. 469. 

• Island ') i&ovXfvffavro ixtravaarrivai 'Iva . . ol * Yet the identification of the ' Island ' is 

l^wwifi <T(p(as fiii fftvolaro, &ffwfp KartOv iSyrup. generally made the touchstone of theories of 

ix. 52 : ol Si i)s tKiv()di)aav, ttpfvyov ifffitvoi Plataian topography. This misconception of 

T^v l-Kwov wphs r^v n\aTai(uy ■k6Kiv, (pfvyovrts the comparative value of the two points is 

ii ^TriKvioyrai M rh 'Hpalov. The repetition strikingly exemplified by Mr. Grundy, who 

l<f>fuyof—<pf6yoyrfs seems designed to give the finds it possible to discuss the operations with- 

impresaion of panic-stricken retreat. 'Attj^cJ- out reference to the site of the Heraion, other 

fitvoi 8* iOtyTo irpb rov IpoZ rck iirKa. Kal ol than its incidental mention in a sentence or 

fiiv irtpl Th 'Hpa7ov iffrparoireidovTO k.t.\. two on p. 17. 

Apparently the town of Plataiai is not ap- « About eight stades, or one mile, on Mr. 

proached more nearly than is indicated by the Grundy'.s map. 

place of the temple. Of course only a small ^ jx. 52: i(pfvyovirphs rhv nXaraitoy irSXiv... 

fraction of the centre could have been accommo- M rh 'HpoTov. rh 5i wph rrjs ir6\i6s i<Tri t^i 

dated within the town : the point is that so far nXaraUuv, 


that they might support one another. The new post of the quondam centre, 
near the Heraion under the walls of Plataiai, was well chosen in this respect, 
to check any attempt on the i)art of the Persian cavalry to creep along the 
side of the mountain and endanger the operation in which the Spartans were 
about to engage. 

That such was the intended function of the Greek centre appears from 
its behaviour during the conflict. Herodotos tells us ^ that the Greek right 
was already pushing the enemy off the field when news was brought (d77€\- 
Xerai) to the centre at the Heraion ' that the fight had begun, and that 
Pausanias was gaining the victory.' The words of Herodotos are hero 
significant, — not in respect of any inference that may be supposed to be 
deducible therefrom as to the site of the temple, but as indicating that 
Pausanias deliberately detached a member of his force for this special 
service, and also that he knew exactly whither to send his messenger. Here 
again Herodotos has missed the real import of the fact. The message of 
Pausanias was nothing less than an urgent summons for an advance. The 
sudden development of the Persian attack caused a rapid modification of the 
combinations of the Greek general ; and, failing support from the ' Island ' 
(upon which the Athenians ought long before to have taken up their position), 
a message was despatched to the centre, then lying uselessly at Plataiai, to 
hurry it up in reinforcement. It is in the highest degree worthy of notice 
that the centre in response at once splits up into two sections. The 
Corinthians and their companions marched off through the hills, while the 
Megarians and the Phliasians with their comrades proceeded by way of the 
plain. Now, in the second position, the Corinthians stood alongside of the 
hoplites of Tegea and Sparta : the Megarians were ranged shoulder to 
shoulder with the Plataians and the Athenians. It is pretty clear from this 
that the two sections of the centre ^ hastened to join their respective wings, — 
in accordance ivith the orders transmitted from the commander-in-chief : it was 
no pell-mell scramble to be in at a battle already decided without them.^ 

With regard to the centre of the Greek line all is intelligible and free 
from complications. In opposition to the received view I maintain that it is 
almost entirely in connection with the left wing, i.e. the Athenians, that 
difficulties arise. The Athenians were evidently hard put to it to render an 

^ ix. 69 : iv 6« toi5t(^ t# yico^tVif) <p6$t(> Phlious and Megara, we shall find that it 

i77tAA«Ta( TOiffi &\\oi<Ti "EWriffi Tolfft rfray- numberod 7,300 men. The remainder of the 

fifvoiffi irepl rh 'Hpaiov Koi iiroyfvofxfyoiffi ttjs centre was 11,300 strong ; the numerical 

fidxvft ^Ti fK^xi T* yfyovf Kal yiKiffv o'l fitra strength of the central sections thus bearing an 

TlavaavUw k.t.x. appropriate relation to the strength of the 

^ And, again, these sections correspond in respective wings. The two sections also con- 
strength to the wings. The right wing (Lake- tain a nearly equal number of contingents. I 
daimonians and Tegean8) = ll,500. The left think that the mention of the Phliasians ia 
wing (Athenians and Plataians) = 8, 600. The really meant to indicate the point of cleavage of 
united contingents of the centre numbered the centre, as above suggested. 
18,600. If we take the expression in ix. 69 : ol * As it is represented in ix. 69 : ol Si i.Koi- 
ifi(pl Mtyapias rt Kal ^Kiafflovi to give the two <ravr(s ravra obSfva kSvixov Tax^ivTit K.r.X. 
extremes of that section, and thus to include Lower down, the Thebans espy the Megarians 
the contingents standing between those of i-nfiyofxivovs olitva k6<i(iov. 

E 2 


explanation of their action during the retirement to the ' Island.' It is in 
vain that with malice prepense meaningless prominence is given to their own 
march through the plain, while the Spartans fell back through the hills.^ 
Meaningless, for this reason : given the position and the objective point of the 
wings, no other route than that which is so invidiously described by Herod otos 
is possible. The map furnishes the unanswerable proof of the disingenuous- 
ness of the narrative. The historian tries to fasten upon the centre the 
imputation of deliberate betrayal of the wings ; 2 but what of the Athenian 
disobedience to orders ? For the Athenians also never reached the ' Island.' 
Could anything be more transparently false than the reason assigned by the 
Athenians themselves for their breach of discipline, — 'knowing that it was 
the Spartan temper to say one thing and do another, they remained quiet at 
their post ' ? ^ Although Pausanias had issued the order for the troops to 
fall back, an order which he knew had already been obeyed by the centre, one 
which he had a right to believe was likewise respected by the left wing, we 
are asked to allow that it was possible for him to remain in position unsup- 
ported, for no other reason apparently than to delude the Athenians at the 
cost of his own destruction and the ruin of Greece. The Spartan king 
appears in the Athenian tradition as a simple farceur. Amid all the contra- 
dictions in which the narrative of the campaign abounds no sentence is so 
preposterous ; none exhibits in a more baleful aspect the inherent vice of the 
Athenians. The lie is inserted in order to conceal their own failure to gain 
the rendezvous appointed by the council of generals, — a council in the 
deliberations of which Aristeides the Just had a voice. It was necessary in 
479 B.C., and still is necessary, to ask how it came about that the right wing 
found itself without supports when the attack opened against it. 

The root of the distorted version of the retirement of the army to its 
third position is the malicious persistence of the Athenians in depicting the 
movement as a Jlight instigated hy the Spartans. Hence they were at pains to 
minimize their own share in it, oblivious of the fact that in avoiding this 
feigned Scylla they fall into the more terrible Charybdis of confessed 
disloyalty and insubordination. 

The desperate efforts of the Athenians to represent their conduct as 
magnanimous would be amusing were it not that their tradition has won its 
way to credence as sober history. The honour of the victory belonged solely 
to the Tegeans and the Spartans. It was a bitter pill to swallow, but Hellas 
could not be befooled on so patent a fact: all knew that the Athenian hoplites 
had not contributed a single blow to the overthrow of the Persian infantry in 

' ix. 56 : 6 YlavaaviT)^ . . . aitriyf 8ia rSiv in the fictitious message from Pausanias, ix. 60. 

KoKwvwv Tous \otirovs Travrai . . . 'A0riua7oi 5* ^ ix. 54 : 'Adr)vdtoi Si iiroUvv TOidSe- flxov 

raxBivra ^iffaf ra ifMiraKiv ^ Aa«t5oi/U($fioi. oi arptfiai <T(pfas avrovs 'Iva ^t({x^'?c«"', iirKTTd/xfvoi 

Hfv yhp Toiv T« oxOoii/ &VT(ixovro Ka\ T^y vwupfris to. AaKfSatfiOviuv (ppov/ifiuTa ws &K\a (ppove6vruv 

rov Kidaipuvos, (po&(6ixivoi tV 'liiiroi', 'Adrtvalot /col &\\a KeyovTwv. ws Se (KivnOv Th (TTpOTcJ- 

6i KiriD rpa<pe(VT(i ii rh irfbiov. TrtSoc, t-ntfjLirov a<p(wv iitiria 6\t>6iJifv6i> T€ «.' 

IX. j2 : ivdavTa i.fpdti'Tis o't iroAAol oiroA- Tropfvtffdai firixetptotev ol Sirapri^Tai, etrt Kol rh 

Xiaaovro, is fiiv rhv x<^pov is rhv awtKitro ovk irapatrav fj.^ SiavoedvTai aTraWdaaeadai, iireipfd- 

if i'6(f fx<'»'T*J K.r.x. CI', the expressions used Oai re \\avaavir\v rh xpfwf ('(i\ iroUnv. 


tlie decisive slnigglc. The efforts of the Athenians were perforce confiued to 
accounting for the damaging f.ict and turning it to the national honour. The 
Theban attack at the liead of the second Persian column ' came in here very 
opportunely to prevent their carrying aid to the Spartans, who were beset in 
spite of their pusillanimous concern to be secure. What, however, is the 
value of the text of the Spartan message which bulks so largely in the 
narrative ? There is an evident anxiety to magnify the Athenian arms on 
this day :2 yet their victory over the Thcbans is not so decisive as to drive 
their cavalry from the field,^ 

From what I have written, my conception of the plan adopted in the 
council of generals* is easily gathered. The Greek force was instructed 
to retire by brigades, — the centre to the Heraion, covering Plataiai and the 
' Island,' the Athenians to the * Island ' itself. These two divisions were 
designed to support the crucial element of the entire movement, viz. the 
Spartan advance to the lelief of the convoys beset in Mount Kithairon. To 
the right wing, composed as it was of the flower of the army, this difficult and 
dangerous task was appropriately committed. In the new position, the old 
central brigade would form the extreme left, under the shelter afforded by the 
Heraion and the fortifications of the town : on the other hand, the troops of 
the new centre, being nearest to the Spartans, might anticipate heavy calls 
upon their alacrity and courage, so that they were judiciously composed of 
Athenians. As in the second position, so in the third Pausanias made the 
best possible distribution of his forces. The Spartans themselves were 
designed from the first to advance straight from their old position, near the 
Spring Gargaphia, to the pass. The locality in which the final encounter 
took place proves this, for it lies off the line that must have been followed by 
troops falling back directly upon the 'Island.' 

What, then, caused the break-down of this scheme ? To this question 
Herodotos has a ready answer. The obstinacy of the Spartan captain who 
refused to withdraw from his post was the prime cause of the collapse of the 
plan. His ill-timed punctiliousness broke the Greek force into its component 
brigades, which at the moment of contact with the enemy found themselves 
sundered by no inconsiderable intervals. The different units had all but lost 
touch of one another when the Persian squadrons held the Spartan division 
fast for the attack of their supporting column. 

* ix. 61 : ol 'Mijvatoi . . . wpufaro BwOtfii' in a previous chapter (ix. 40 end) to prepare for 

Kol ra ixa\i<Tra (irafivvav. Kai a<pi ¥)hi) (TTflxovfft this by magnifying the courage of the Thebans 

fwiTietvrai ul avTiraxOffTts 'EWitvuv . . . &are in leading the cavalry charges. When the 

firiKfTi SiivaffBai fiwdriaa'f rh yap irpoffKfiftevSi' Athenians co-operate in the assault on the fort 

ff<peas fKvirff. (ix- 70) o'lirw 5)j IffX^ph ijiyero reixonax'trj Ka\ 

^ ix. 67 : BoicDTol ' AdrivaioKn (/xaxfffavTo XP'^"'^" ^"'^ iroWSy. 
XpivQV f'lrl avxviv. oi yap fxijSi^ovTfs rSiv ^ "Which cuts to pieces the Megarians and 

@-i\^alwv, ovToi elxov TrpoOv/jiirii' ovk oKtyrtv Phliasians on their march to the scene of action, 

fiax^f^foi rt Kol OVK iBtKoKaKfOvrti, ovrw &arrf ix. 69. 

rpiriK6(Tioi ouTwi' oi TrpwTot /coi ipitrroi ivOavra * ix. 50 : ol rwv 'EAX^«'a)^ ffTparrjyol . . . . 

iitiaov vi("Mi)valt>>v. Herodotos has taken care <Tvv(\i\Br\aav . . . irapitTlavaavirtv k.t.x. 


The story about AmompLarctos is jxirhaps one oi Lhu Lnubt difficult points 
in the narrative of the operations preceding the battle. We may, — and this 
is the least satisfactory course, — accept the story, and compare the attitude of 
Amompharetos with the refusal of the polemarchs Hipponoidas and Aristokles 
to execute a tactical movement at the battle of Mantineia on the orders of 
king Agis.^ Or has Herodotos here incorporated a regimental tradition of the 
Pitanates, one derived from his Pitanate friend Archias ^ ? I prefer to 
account for the origin of the st^jry in the foUowing manner. 

The Spartans did not evacuate their position without taking the 
precautions demanded by the situation. Amompharetos and his Lochos were 
detached to occupy the crest of the ridge which concealed the Spartan 
lines : on the ridge stood the monument of Androkrates. The object of 
this was twofold, — to observe the Persian cavalry, which would soon resume 
its daily task of keeping in touch with the Greeks, and to retain as long as 
possible the semblance of the Greeks being in position.^ Amompharetos 
stuck to his post to the last minute that it was prudent to do so, and then 
rejoined the main body 'at a walk'*; the honour of a Spartan would not have 
permitted a less leisurely pace. The main body had come to a halt for him 
and his news a little over a mile in advance.^ His arrival just in time, with 
the report that the Persians were moving, enabled the Spartans to change front 
and to form for action in a favourable position on the slopes at the head of the 
stream Moloeis and the Argiopian region. Amompharetos is painted by 
Herodotos as an obstinate fool,^ the rival of his commanding officer in 
buffoonery. On the contrary he was an officer conspicuous even among 
Spartans for intrepidity, one whose tried valour gained for him the perilous 
but honourable task of screening the retirement of the main body. Not 
undeserved was the prize he won for bravery in the presence of the enemy,^ a 
prize which the Spartan purchased only with his life. Possibly there is this 
amount of truth in the story of his refusal to retire, that he may have been 
prominent at the council in urging the rejection of the combination which 
Pausanias tried to effect. The parenthetical remark of Herodotos, that 

^ Thuc. ▼. 71. Uut there the charge had on the hill : the latter indicates that by the 

already begun, and the movement may well time Mardonios made his inspection the rear* 

have been impracticable. The fact that a court guard had withdrawn, and the heights upon 

martial condemned the two commandens to which the Greek sentinels had been for some 

banishment proves nothing. days visible were deserted (cf. (n rl A€{«t« tH* 

' For Archias see Herod, iii. 55. We need bpiovrn ipvfjLa ;). 

bot enter here upon the vexed question of the * ix. 57 : &ya\a&6yTa rbv \6xov ri SxAa ^y» 

Pitanate regiment, the existence of which is QiSrii/ irphs rh &\\o <n'i<pos. 

denied by Thuc. L 20. " Jbid. rh 8* kirfKdhv Zvov t« 5«Va (rraSta 

' So Herodotos (ii. 58 inil.) correctly dis- iiyffifvf rhy 'A)xOfj.<pap(rov \6xoy, irtpX Trora/jthf 

tingxmhes between the rej>ort brought to Mar- MoXbfyra lSpvfi.(yay 'Apyi6-iri6y re x^po" Ka\(6- 

doaioa by his scouting cavalry (on iirvBtro toi/j ^tyov, rri koI ^-f^firtrpos 'EKtvaty't-ns Iphy fitrrai 

'IKKrjyas iwoixofiiyovs vnh vvKra) and tlie /cal o'l t«<pl rhy 'Afj.ofji.(pdp(Toy vapfjiyoyrd arpi^ 

evidence of his own eyes (cUt rt rhy x'^po*' i"^^ V "iitiros fi rwy $ap0dp<i>y vpofffKtero Ttacra. 

iprifioy). The fonner refers to the discovery by ' ix. 55 : 6 if [sc. Pausanias) /xaiyS/xtyoy Kal oii 

the cavalry of the true state of the caae, not- <pptyfipfa KaKiwv iK*iyov. 

ithgtanding the presence of the Greek outpost ' ix. 71, 


Amonipharetos had not been present at that council,* is scarcely credible in 
itself and has the air of a makeshift to get round what the historian himself 
felt to bo an improbability.- 

The retirement of the various divisions of the army cannot have been 
attempted simultaneously. It was an operation of much delicacy for an army 
of nearly 100,000 to fall back over hilly ground in the dark, especially if we 
accept au jned de la Icttrc the account of the demoralisation produced in the 
Greek force by the incessant attacks of the cavalry.' Not until the centre 
was on the march did Pausanias give the word to his own brigmle.* The 
retirement was evidently intended to take place by divisions. It was timed 
to begin at the second night-watch, i.e. about midnight.^ The sum tot^l of 
the retiring centre, according to Herodotos, was about 39,000 men. The 
battle was fought in the month of July-August, when day begins to break 
betwen half-past four and five. A simple calculation from these data brings 
us to the conclusion that the manoeuvre could not have been executed within 
the time allowed. The Athenian division, deliberately or not, made the 
mistake of not marching first : they were consequently delayed by the 
clumsiness and unwieldiness of the centre, and the Spartans themselves were 
surprised by daylight " as they advanced towaids Mount Kithairon. 

The failure of the scheme must be traced to the tactical unskilfulness of 
the Greek commanders. In the battles of the pre-Alexandrine age in Greece 
nothing is more striking than the absence of tactics, and this in spite of the 
brilliant success attending the combinations of the few tacticians who passed 
meteor-like across the horizon of Hellas. In 479 B.C. it is almost too early to 
speak of tactics in connection with Greek armies : their movements are still 
somewhat haphazard and capricious.^ Pausanias set his officers a task beyond 
their powers. They hmi succeeded in the advance from the first to the 
second position ; but in the retrograde movement, with all its complications of 
direction and its nice adjustment of the divisions to the work of mutual 
support in the offensive designed by the Spartan general, the commanders of 
the contingents utterly failed. The army was split up into separate bodies, 

^ ix. 53 : i6il>vfia(4 r* bpiw rh wottififvof, At* roTfft AoKtiatnoviotai. 

6b wa(>aytt>6fttvos rf vporip^ ^<^79»' ' ix. 51 : i**iiv r^i vurrij jf Ztvripr) ^i»Xa«r^, 

' The following also occurs to me» — that &>% iv ^l.^ IZolaro ol XlipSai ilopn*o^*yovs itui 

Athenian wit gave this turn to the facts in tr^cai iw6n(voi rapdnffoity ol IrrSrai. 

order to exhibit a quasi-comic rf<fuc/i'o a(f«Afi<r- ' ix. 56: robs Si iw*\ iyaKpiyonirovs wphs 

dum of the boasted Simrtan principle as laid iaivrohs ^»s KartKdn^yt. 

down by Demaratos, Herod, vii. 104: 'Law ' It is instructive to notice that it is precisely 

forbids them to flee in battle, whatever the when in conflict with /omyu troops that tactics 

numlx'r of their foes, and requires them to and strategj' are exhibited by the Gr^ek 

stand tirm and either to conquer or to die.' generals, in the earlier perioil of Greek history. 

The Athenians must have been as weary of For the whole principle goTemiug such contests 

hearing this as the Spartans themselves were was quite other than that governing tlie inter- 

of hearing alRHit Marathon (Thuc. i. 73). tril^al wars. Hence the Wttlcs of Marathon, 

' Recent Greek history aflfords an instructive Salarais, and Tlataiai stand apart in interest in 

parallel. this res{>ect. It is a difference that is not 

* ix. 63 : navffavltis 8J ipiav (rfias iiroA- explicable merely by reference to our fuller 

\turffOfi4yovs iK rov ffTparoiriiov wap4iYy*\\* Kal knowledge of the details of the operations. 


but the rare steadiness of the men retrieved the blunders of their leaders. 
Plataiai also was a 'soldiers' battle,' — one of the finest ever won by the 
* Dorian spear.' ^ 



On the Application of the Name Asopos in Herodotos. 

Mr. Grunily (p. 18 fol.) finds in the use of the name Asopos the 'real difficulty' in the account 
of the operations at Plataiai, and suggests that its solution lies in the assumption that Herodotos 
used the name in two senses : — 

(1) The main stream of the Asopos, called by Leake the The.spian Asopos. 

(2) The stream that takes its rise in the Apotripi spring (stream A} in Mr. Grundy's 

map : cf. Leake, North. Gr. 333). 

He bases his opinion upon the following arguments : — 

(1) The Greek second position was defined by the Spring Gargaphia and the monument 

of Androkrates, which lay in the plain 'less than three-quarters of a mile from 
Plataea.' Yet, at the end of chap. 30, in speaking of the same position, Herodotos 
uses the words olroi fxiv vw TaxfleVres ^irl t# 'A.<routrf iffrparoirfSfvovro. 

(2) The expression just quoted is followed in the very next sentence (chap. 31) by the 

words iraprjffav, ■Kv66fji.ei'oi Toi/s''E\\7jvos flvai iv UKarai^ai, kuI outoJ iirl rhf 'Affunhv 
rhv ravTTi ^iovra. 

From this, Mr. Grundy concludes that the reference in the first passage is to the tributary A*, 
and that the addition of the words rhv ravTTi piovra, ' not evidently referring directly to tlie 
Asopus at the end of chapter 30, but to the words iv UXarai^ai,' leaves no 'reasonable doubt that 
the stream here mentioned is the main Asopus.' 

In chap. 40 (/tf'xpi m*** 7«P '"oO 'AcwttoO iir-tjicrav ot fidpBapoi K.r.x.) the reference is again to 
the main or Thespian Asopos. 

Reading further (p. 26), we find that this hypothesis of a twofold signification of the name 
has apparently been prepared in order to surmount the difficulty presented by the statement in 
chap 51, that the 'Island' lay 10 stades 'from the Asopos and Gargaphia' (ij Bt (an a-irh rov 
'Aawirov Ka\ ttjs Kpr}vqs rrjs rapya<pirii, itr' t) iffTpaTo-ireBfvomo t6t(, SfKa araSlovs aTrt'xouffa). 

■\Ve must altogether reject Mr. Grundy's suggestion. The name Asopos is applied by 
Herodotos consistently to the main stream, and to it enly. If Mr. Grundy is right in taking 
Leake to task (p. 45) for calling the large Kriek&ki stream the Asopos, it is somewhat strange to 
find tliat he himself applies the name to the insignificant brook A^ on the ground that it can be 
seen from the walls of Plataiai, while the main river is invisible (p. 26). 

It is in the highest degiee improbable that two distinct senses of the word should have been 
so closely combined as in the two consecutive sentences (juoted from chaps. 30, 31. In so far as 
Mr. Grundy's hypothesis rests upon the locality to be assigned to the monument of Androkrates, 
it has already been refuted. It is also partly the outcome of a too great rigidity in the translation 
of the phrase tit\ T<f 'hawit^. Mr. Grundy is concerned to show that the army was literally 
astride the brook (p. 21). The preposition is used in its technical military sense, which would not 
fonflict even with the ordinary acceptation of the situation of the heroon. (Cf. chap. 38 TtJre 5' 
iwl T<fi 'Affw-wif MapSovitp fxfixi(r0wfj.(voi...4evfTo : which does not mean literally on the banks.) 
There is no mystery in the addition of the words rbc raurr; (ifovra to the name Asopos in the 
second pa.ssage. They merely indicate the change of position to another portion of the river. It 
would surely have been strange to remark simjily that the Persians also advanced to the Asopos, 

' Acsch. I'cra. 812 fol. 


sfeeing that they had been encamped on tliat river since the commencement of the operationN (cf. 
chap. 19 : Ifi.a06v t« 8^ rovs Pap$<ipovs iir\ rf '\<Twn<f trrpaToirfSfvofjLfi'ovi). The words mean little 
more than 'to this precise point.' Even adniittinf; them to have .some special signification, it 
would surely follow from Mr. Grundy's confession that they refer to the words iu TlKaraififfi 
(p. 19), that Herodotos meant thereby the stream A', wliich takes its rise in the direction of 

When we reach chap. 40, Mr. Grundy decides, on what criteria I know not, that the Asopos 
there mentioned, is 'certainly the main or Tho.spian Asopus.' This, taken in conjunction with 
the rest of his topography, necessitates the adoption of a theory as to three ' developments ' of the 
second Greek position (p. 19). It would surely have been simpler to keep to the first hypothesis, 
that the Asopos upon which the Greeks lay was the stream AS than to pile up this new hypothesis 
in order after all to bring the Greeks to the main stream. 

There remains the passage relative to the situation of the 'Island.' If we take the Asopcs 
from which the 10 stadcs are measured to be the main river, then the given co-ordinates (10 
stades from Gargaphia, and 10 stadcs from Asopos) bring us to Leake's ' Island,' a i)osition which 
Mr. Grundy has shown to be impossible (p. 23fol.). Yotjif the Spring Gargaphia is rightly 
identified with the Apotripi, it becomes obvious .y impossible to argue that Herodotos measured 
from tke stream A', as his starting-point in that case could only be either the source or the mouth 
of the stream. The source is impossible as it coincides with the spring. The mouth is equally 
impossible as that is on the line of the Thespian Asopos, which line is out of the question, as 
already remarked. I suggest that k ( = 20) has dropped out before the Ka\. We should read 
7] S4 iari oiri toC 'kawKov k' koL rfjy Kp-ftvrji rris rapyatpirjs fir' ^ iffrparoireSfvovro rorf SfKa (rraSiovs 
airixovaa. The tract of ground identified as the ' Island by Mr, Grundy lies almost exactly 20 
stades from the Thespian Asopos. 

The latest utterances of Mr. Grundy {Classical Review, April 1898, p. 161), in answer to 
Mr. Frazer, simply re-affirm his views, with the additional conjecture that in the application of 
the name Asopos to the stream A^ Herodotos has preserved the local custom of the Plataians ! 

NOTE 13. 
On the Chronology of the Operations at Plataiai. 

The views advanced in the preceding pages necessarily involve the rejection or the modification 
of the chronological items embedded in the narrative of Herodotos. 

Herodotos does not tell us how long the Greeks remained in their first position, on the breast 
of Mount Kithairon. We are informed, however, that the two armies had been encamped opposite 
to each other already eight days before Mardonios was advised to close the pass through which the 
Greeks received -their supplies (ix. 39). The pass was actually closed at nightfall of tlie same day. 
The expression of Herodotos is ambiguous : it is not clear what is the point of departure involved in 
the words rififpai 8« (T<pi i.vrtKarrjfj.fvoKri IjBri iytySveffav oktw. Are the eight days to be counted 
from the marshalling of the troops in the second position ? Such seems, to be the generally 
accepted view, but it has always appeared to me somewhat of a marvel that historians should 
credit this reflection upon the intelligence of the Persian general. If Herodotos really meant that 
the Greeks had been eight days in the second position before the pass was blocked, I .should see in 
tiie statement but one more instance of the working of national antipathy. The Greeks cannot 
allow the invaders to have possessed ordinary common sense. Obvious as was the stroke of 
blocking the main artery of the Greek communications, the tradition puts it to the credit of 
Tiniagenidas, a renegade Theban it is true, but still a Greek (ix. 38 fol.>. Mardonios, to my 
mind, was more than a match for his opponents in point of military skill, and an explanation 
more in accordance with the probabilities of the must be sought. The words quoted bear 
reference to and date from the first day that the two armies found themselves face to face in the 
first position. The pass was closed as soon as the evacuation of the lines on Mount Kithairon 
threw it open to attack. The Greeks abandoned their first position within the week. 

In precisely the same way must we interpret the words in chap. 41 : i>s 8e ivSeKdrri iyty6yef 
rinfpf] avTiKarrifiivotffi iv nXaTaiperJi The expression iv l\KaraiT,ai simply indicates the theatre of 


operations, ainl docs not restrict ua to tlic sicuini (lositioii. The ulcveu days in this case also ale 
reckoned from the opening of tlie campaign. Wlicther tliis was indeed the intention of Hcrodotos 
must be left vindecided. 

The next note of time is given in the important words at the opening of cliap. 40 : ' after this 
the armies waited two more days ' (/i*To 8c tovto rh tpyov irtpas 5uo iifj-ipas SUrpiif/ay, ovZirtpoi. 
0ov\6fJLfi'ot fidxvs *pf«<)- What arc tlic two termini involved in this expression 1 With regard to 
the event from which tlie reckoning is made no doubt is possible: it is, as Hcrodotos says, the 
closing of the j)ass. Wlien we ask to what consiiicuous event in the development of the drama the 
two days' interval I)rings us, the reply is vague and unsatisfactory. For they are followed by the 
resolution of Mardonios to end this idle delay ; and yet two days more intervene before his purpose 
is accidentally accomplished. As it stands, the sentence is meaningless. It becomes intelligible 
only upon the view already devcloi)cd. 

The eleventh day is devoted to the consultation of his Staff by Mardonios (ix. 41). 
Alexander's visit to the Greek outposts takes place at midnight (ix. 44). On the twelfth day occur 
the challenge of Mardonios, and the cavalry attacks which culminate in the loss of the Spring 
Gargaphia : the Greek generals determine to fall back upon the ' Island' (ix. 48 fol.). During the 
night the army evacuates the second position (ix. 52 fol.). In the early morning of the thirteenth 
day the final battle is fought (ix. 56 fol.). 

Now we have already seen that wc must cut out as fictitious items the consultation, the visit, 
tlie challenge, and perhaps also the loss of the spring, — that is to say, the whole of the matter 
allotted to the eleventh and twelfth days, with the cxcei)tion of the deliberation of the Greek 
generals. Tlic evacuation of the second jiosition and the final struggle must therefore be 
antedated by two days, and be assigned to the night of the tenth day and the morning of the 
eleventh day respectively. In other words, the event to which the reckoning is made in the 
sentence quoted from chap. 40 ('after this the armies waited two more days') is the final battle 
itself, which took place two days after the closing of the pass of Dryoskephalai. 

The story, as given by Herodotos, imperatively requires a somewhat protracted stay in the 
second position on the part of the Greeks. It was also clearly impossible, from their very nature, 
that the interpolated episodes of the consultation, the challenge, etc. should immediately follow 
the adoption of that position. Herodotos has consequently duplicated the interval between 
the closing of the pass and the final battle. He may perhaps be acquitted of the mistake 
already pointed out, by which a further addition of eight days is made to the time spent in the 
second position. 

My idea is that when their offensive failed the Greeks at once retired, i.e. at midnight of the 
tenth day, reckoning from their first appearance on the northern slopes of Mount Kithairon. They 
were not more than three days in the second position. 

Hence my diary of the operations is as follows : — 

1 Greeks take up First Position. 

Attacks by Persian Cavalry. 
Death of Masistios. 

7 [Evacuation of First Position probably on night of this day.] 

8 Greeks in Second Position. 
[Pass closed on night of this day.] 

9 Continued Skirmishing. Plataiai-Thebes road blocked by Persians. 

10 Meeting of Greek Generals. 

[Retirement to ' Island ' partially effected on night of this day.] 

11 Final Battle. 

12-20 Burial of Dead. Collection of Spoil. Consultation. 
21 Greeks march on Thebes. 

The Usual scheme gives in addition : — 

(1) An unknown number of days in tho first position. 

(2j Eight days in second position before the closing of the pass, 

(3) Two days of purposeless waiting after the closing of the 

(4) Two days devoted to Persian Council, the Challenge, and blocking of the spring. 


Tho result of the usual scheme is that the bailie was fouj^lit on llie ihirlccnth ilny n/lcr the 
occupation of the second ]>oxilio)i, and Thebes is reached on I he twenty-lhird day after the same 
event. If tin; same generous measure is used in metiny out Ihe time spent in tlie first position, 
the Greeks must liavc been four or live weeks on the Asopos. Could a force of one liumlrcd 
thousand men liave kept the held for that length of time in the fifth ceiilury it.c; The case is 
very different from that of a blockade, in wliich one side has an al)S(ilutc superiority. Lastly, how 
explain on the ordinary theory the arrival of the Miintineians and Eleans too late to take part in 
the battle (i.x. 77)? An e.\[)lanation cannot be found in the closing of the passes, as one at 
remained open to the end : nor, if such had been tlie reason, would the leaders of those contingents 
have been banished for Aiilurc to arrive in time. On Ihe view here presented, ten days covered 
the whole scries of operations previous to tlic final catastrophe ; and tlie two Pelopounesian 
contingents may well have found that events before Plataiai outstripped their progress to the seat 
ol war. 

W. J. \V. 



Fig. 1. — Tyche and Plutus {from a column of the Hall). 


The Hall of the Mystae. 

[Plates I.— III.]. 

The Hall of the Mystae is a Roman building on the western slope of the 
ancient town of Melos. The principal object and result of the excavation 
begun by Mr. D. Mackenzie and myself in April and extended by Mr. Cecil 
Smith in May 1896, was to put on record its fine mosaic pavement. We 
were fortunate in being able to call to our aid a skilful and indefatigable 
draughtsman. Mr. Charles Clark, architect to the School, joined us in Melos 
as soon as he could be spared from the Athens excavations, and worked upon 
the mosaic for several weeks in the full heat and glare of a Mediterranean 
summer. Of the illustrations, fruits of his patient labour, which this paper 
serves to introduce, Plate I. represents the two figured panels on the scale of 
1 : 25, and is a very faithful rendering of their general effect ; while Plate 11. 
gives part of the finest panel on the scale of 1:5, and shows the method of 
execution in detail ; it is reproduced from one of a series of rubbings' coloured 
cube by cube upon the spot, which are practically full-size facsimiles of all 
the principal figures. The spirited figure of the cock (Plate III.), supplied by 

VOL. XVIII. (1898). PL. I. 



J. H. S.. VOL. XVIII (1898). PL. III. 







' All 




another rubbing, gives a good idea of the life-like force of the design. 
For the restoration attempted in the key-plan (Fig. 4) we are jointly respon- 

The mosaic seems to date from the first half of the third century. 

The site is marked H on a sketch-plan of the ancient town which accom- 
panies Mr. Cecil Smith's account of our work in Mclos {J .IT. S. xvi. ]). 348). Mr. 
Cecil Smith has there described the mo.saic (p. 354); he has since published 
two inscriptions which we found there and inferred from them that the place 
belonged to a Society of Dionysiac Mystae (J. If. S. xvii. p. 14). 

Previous History of the Site. 

This conclusion is confirmed by the fact that other inscriptions naming 
the Mystae have been found at the same spot. An Athenian magazine^ of the 
year 1862 contains the following among other archaeological news : * In Melos 
in the ground called Tpe/xivdia near the ancient theatre in the course of an 
excavation made by private persons there was lately found a mosaic pavement 
said to be about 40 m. in length, a wall of squared stones with a door in it 
and various marble sculptures bearing inscriptions, apparently of Roman date.' 
After describing them the report goes on ' the excavation is being made 
with the knowledge and indeed under the supervision of the local authorities.' 
How little this meant is implied by the words which follow, * we hope that 
the objects found may not be dispersed, as has happened on other occasions.' 
Finally it is suggested that an ephor should be sent from Athens. From 
enquiries made on the spot, it appears that there was no official excavation ; 
the Government contented itself with stopping the enterprise of the ' private 
persons ' and securing the marbles for the Athens Museum, where they now 
are. They consist of a bust of Aurelia Euposia (Fig. 8) set up iv tw ISio) 
avT7]<; epyo) by certain Ueptficofitot (Cavvadias' Catalogue 424),^ the head of a 
young man bound with a fillet (Catalogue 459), and two columns, the ends of 
which have been sawn off for convenience of transport. On one of these columns 
is incised a figure of Athene, on the other that of the Good Fortune of Melos, 
and in each case there is inscribed a prayer that the Goddess may be pro- 
pitious to Alexander, founder of the Holy Mystae (ktio-tt} elepcov fivarSyv). 
They are fully described and published in a valuable article by Welters on 
' Melische Kultstatuen' {Ath. Mitth. xv. 1890, p. 246). The figure of the 
Tyche of Melos has acquired a certain importance in the history of art 
since Furtwangler used its testimony in support of his restoration of the 
Melian Aphrodite.^ It is reproduced in Fig. 1. 

1 ♦iXfffTwp, vol. ii. p. 274. Copied thence last volume (xvii) of this Journal. The 

into the Arch. Anzeiger, 1861, p. 234, and early notices mention a headless bust and two 

Bullettino, 1862, p. 86. See especially the heads. The second of these may have been the 

article by Wolters, Ath. Mitth. xv. 1890, p. head which is now fitted to the bust. 

246, to which I refer below. 3 Mcistcrwcrke, p. 624. English edition, p. 

- The inscription is published on p. 16 of the 381. 



The description of the site as iv tt} decree TpefiivOia tt^o? t&j eVei apx^'^V 
deuTpo) is accurate in the sense that the hill-side called Traniithia and the 
theatre are in the same part of the island ; but they lie ten minutes walk 

Fig. 2. — View from the East End. 

apart on different sides of the central acropolis-ridge. The identity of that 
site with ours, which is in Tramithia ^ but not near to the theatre, is put 

' We liavc usually followed Ehrenberg's maii tion. The form Tptfjuvdla suggests a derivation 
(Leipzig, 1889) in spelling the name Tramythia, homrpftii0os —TfpnivBos. Steph. Byz. mentions 
but Tramithia is as near to the local pronuncia- a place called TpffxiBovs {v.l. Tpffiivdovs) in 


beyond a doubt by our discovery of columns of the same diameter and material 
as those in the Athens Museum and by the local atory that parts of such 
columns had been sawn off and sent to Athons with othor marbles from this 
site. An idea of the lie of the ground may be gained from Mr. Clark's 
sketches, figs 2 and 3, and from the key-plan on the opposite page. That 
part of the site which was first pointed out to us as containing a mosaic was 
a small field just south of a nuile-path which leads from the villages on the 
heights to the Tramithia landing-place. Like most other fields on 
highly cultivated hill-sides it is a terrace bounded by higher and lower terraces, 
each supported by a massive retaining wall locally called Tpd<f)o<i (for Td(f>po<;). 
In this case there was a rising traphos to the east and a descending traphos to 
the south ; the other sides of the rectangle were formed by the mule-road which 
gradually descends from the higher level of the terrace on the east to the lower 
level of that on the south, curving round our field and cutting off its north-west 
corner. The course of the road-wall and of the terrace-wall to the east is 
shown by the dotted lines W W on the key-plan ; it was only under them that 
we found the mosaic in first-rate preservation. The field has a downward slope 
from east to west, and at the lower end the pavement had been obliterated by 
cultivation ; further east, where there was some depth of soil to protect it, 
considerable injury had been caused by the recklessness of the excavators of 
1861. It was then that a great part of the fish-panel was destroyed. They dug 
as far as the eastern terrace-w^all, and seem then to have worked down from the 
upper field and to have penetrated as far as the door in the eaist wall of the 
Roman building ; but the Tpd(f)o<i between the two terraces happened to be 
the boundary between two properties, and that fact preserved it inviolate and 
with it the whole panel of the vines which lay below. This belt of unknown 
ground had weighed on the consciences of local treasure-hunters ever since. 
We heard of at least two attempts to explore it. In one case the adventurers 
tunnelled under the road, breaking through the north wall of the Roman 
building, and worked along its inner foce as far as the corner-column A which 
is still in situ. They dared not go further, fearing that the mass of stones 
overhead would fall in on them, and retired by the way they had come, but 
not before they had found a marble head. This head, which was sold to the 
Athens Archaeological Society in 1884 and passed with their collections into 
the National Museum, has since proved to belong to the statue of a hierophant 
which we found lying on the mosaic in 1896 (Fig. 6). 

Cyprus and derives the name airh ruy irtpl rhu island. For place-names in Greece derived 

T6iroy iTftpvKviuv npixivOuv, &s Tvpiot rpe^ilOovs from trees see Tozer, Highlands of Turkey, ii. 

KaXovfft. Tlie name would be formed like that p. 107. Silithorp {Florae Graccac Prodromus, ii. 

of Vaixvovs in Attica = Po^uftJeis from jianvos. p. 256) and Fiedler {Rcise, i. p. 539) say that 

The form rpf/xidos is further attested by Nicand. Pistachia Tcrebinthus grows abundantly in the 

Ther. 844 rpetiiQoio viov Tro\vtiS(a Kapir6v. Greek islands. Both give the itKidern Greek 

The Melian 's r^v TpantOia seems parallel to name as TfTpd/xiOos. Mr. Bickford Smith gives 

'y tV 5i"c'a and 'j t^ nAcfravio in the same Tpafj.-f)0r] as the Cretan form. 


The Building. 

We demolished the eastern terrace-wall, cleared the remains of mosaic 
in the lower field, and found that we had two panels and part of a third. 
Later Mr. Cecil Smith pulled down the wall on the north and uncovered a 
long strip of mosaic, which not only proved the existence of a fourth panel, 
but also preserved just so much of the geometric design of a fifth as 
enabled us to complete it and to determine the dimensions of the whole. 

The building was a long hall running east and west, 832 m. wide and at 
least 23 m. long (27 ft. 4 in. x 75 ft.). The east aad north walls are in great 
part preserved ; the south wall has almost disappeared, but enough remains 
to justify us in restoring it on the analogy of the north ; the west end is 
wholly destroyed. The tessellated pavement did not occupy the whole width 
of the hall ; along either side ran a stylobate 1"50 m. broad, raised 027 m. 
(10| inches) above the floor, supporting a row of unfluted marble columns. 
Of the marble slabs of the stylobate only one survives ; it is under the single 
base which remains in position ; but the dwarf walls which carried the stylo- 
bate are preserved, 14 cm. high. As for the columns, A is in place and the 
position of the two adjoining columns is indicated by blocks which once 
supported the marble slabs under their bases ; they give 3*32 m. (10 ft. ] in.) 
as the intercolumniation. 

The exact length of the hall was not determined ; this might 
possibly have been done by digging for the foundations of the north-west 
angle from the field beyond the road ; but there was great risk of injuring 
some valuable olive-trees. We dug down at the only possible place, where 
there happened to be a gap in the olive-grove, and found the outer face of 
the north wall (at F in key-plan) under the roadway, some five feet below the 
level of the mosaic. This part of the wall probably dated from Hellenistic 
times and originally rose above ground, for it was better built than the upper 
part, and along its foot there was an accumulation of pottery ranging from 
third-century Greek to Roman. The evidence is slight, but one is inclined to 
infer that the Hall of the Mystae stands on the site of a Greek building, 
using its walls as foundations ; in that case the earlier floor-level may be 
some feet below the mosaic. Beyond the point where we suppose the west 
wall to have stood the ground falls away ; had the building extended further in 
that direction it would have required very massive substructures ; but of 
these no trace remains. It can hardly therefore have been longer even by 
one intercolumniation than we have shown it in the key-plan. On the other 
hand the remains of the mosaic prove that it cannot have been shorter. 
The restoration of seven columns on each side may be regarded as fairly 

We were also unable to dig as far as we wished to the east. Once 
beyond the shelter of the thick terrace-wall we found that the whole area in 
the upper field had been ransacked and filled in with stones. We cleared 
part of the little chamber at the north-east angle and worked some feet 



beyond the large door in the east wall, but there were no mosaics. As the 
debris was eight feet deep and difficult to handle we did not feel justified in 
going further for the sake of completing our plan. 

It is probable that the principal entrance was at the west end, and the 
cast doorway led from the body of the hall into a chancel-like extension, an 
ahvrov opening out of the reXea-Tijpiov. Just such an inner sanctuary may 
be seen in the plan of the Baccheion, a building much like ours in date and 
character, excavated by Dr. Dorpfeld between the Pnyx and the Areopagus.^ 
The internal arrangements of the hall are in agreement with the view that 
this was its principal end. Of the five mosaic panels the western is the 

Fig. 3. — View from the West End. 

simplest, the eastern the most elaborate. Close to this doorway in the south- 
east angle stood a square structure (D in key-plan), obviously of importance, 
for the outer border of the mosaic was compressed and cut short to leave room 
for it ; it must have been a small shrine or altar.^ In a niche on the opposite 
side (C on key-plan) stood in all probability the statue of a priest to be dis- 
cussed later, which we found fallen on the pavement. Before giving up the 
idea of exploring the supposed adytum we sank a pit 8 m. east of the mosaic, 
and found fragments of a wall covered with red stucco, its floor- level being 

1 Ath. Mitth. XX. p. 161. Taf. iv. 
"^ In position and perhaps in form it re- 
sembled the 'house-altars' of Pompeii, Cf. 

Overbeck, Pompeii, p. 268, Presuhn, Die Neuesten 
Atisgrabungen, i. Taf. iv., vi. Taf. ii. 


about the same as that of the mosaic. Somewhat further east we must 
suppose an ancient terrace-wall ; for it was at a much higher level, though 
only 20 m. further away, that we found the basis (Fig. 6) dedicated to 
Dionysos Trieterikos. The circumstances under which it was found are 
described in J. H. S. vol. xvii. p. 14. We dug round the spot, but found only 
some walls of Roman date, a flagged court, in the middle of which the basis 
stood, and a cobbled path leading towards the mosaic. It is very probable 
that these buildings were in some way connected with the Hall of the 

Before passing to the mosaics it may be noted that the walls of the Hall 
were covered with a thick coat of plaster. In demolishing the traphos wo 
came upon a quantity of plaster including many fragments of mouldings. 

The Pavement. 

The space between the stylobates is filled by the mosaic pavement 
5-35 m. (17^ ft.) wide, and as restored 22'22 m. (nearly 73 ft.) in length. The 
length that is preserved is 1920 m. (63 ft.). The whole design is framed in 
an unusually broad triple border, 1-38 m. (4 ft. 4 in.) wide. The width of the 
panels is 267 ra., just double that of the border and half that of the whole 

There are five panels ; the following measurements do not include the 
guilloche border, which is 17 cm. wide, but are taken up to its edge : 

I. Vines, birds, gazelle and hare 

II. Fish and fisherman 

III. Geometric ..... 

IV. Probably a figure-subject, destroyed 
V. Geometric, as restored 

3-28 X 2-67 m. 

2-67 X 2-67 

6-48 X 2-67 

3-07 X 2-67 

3-28 X 2-67 

It will be seen on reference to the key-plan that the places of the 
columns correspond broadly though not exactly with the divisions between 
the panels. Panel III. is twice as long as I., which again is the same length 
as V. The preservation of these proportions, as of those between the breadth 
of the panel and the border, shows how carefully the mosaic was designed for 
the building. 

The detailed execution and technique of the Melos mosaic are admirably 
exhibited in the large scale drawings (Plates I., II. and III.). Glass tesserae 
are freely used in the birds, beasts, and fishes ; all these figures are carried 
out with a skill that must have been the result of long experience. The 
glass tesserae are much smaller than the marble tesserae ; the latter are 
usually square or nearly so at the top, while the former are of all shapes, and 
seem to have been chipped off from a slab of glass as they were required. 
They are mostly blues and greens. Mr. Henry Powell, who is an expert in 
modern glass-mosaic, has been so kind as to point out that these glass 
tesserae seem to have been translucent ; some of them retain their transluc- 

















































F 'Z 


ence, others have lost it owing to the action of weather. The colouring 
matter in the blue tesserae is cobalt, in the blue-green, copper, in the other 
shades of green, iron. It is difficult to say when the practice of adding 
arsenic or tin to make the tesserae opaque first came into use. Besides glass 
and marble a local material, the lustrous black obsidian, is used with great 
effect, especially in the long geometrical panel. 

The Panel of the Vines. 

The panel of the vines is the most elaborate and the best preserved 
(Plate I.). Mr. Clark has been wonderfully successful in reproducing its 
originality, its grace of design, and its rich harmonious colouring. The 
subject is unusual ; it must have been chosen for the place of honour in the 
Hall as one especially appropriate to the society of Mystae and their patron- 
god. This special local significance may help to explain the unconventional 
character of the composition, its freedom and want of symmetry ; it is 
like the work of a man who has put aside his pattern-book and is feeling 
his way towards a fresh design. The elements which he had to group 
together were familiar ; the animals grapes and leaves are the work of a 
practised hand; but in the stiff lines and abrupt curves of the branches 
there is the irresolution, the hesitating touch, of an experiment. Strangest 
of all is the want of balance in the disposition of the birds and beasts 
among the foliage. The whole south-east corner is given up to grapes 
and leaves and tendrils with no living thing among them. The contrast 
must have been all the more conspicuous before the pedestal (B in key-plan) 
was thrust into this end of the panel ; it cuts so rudely into the design that 
there can be no doubt of its being a later insertion. We may perhaps 
connect the different treatment of the south-east quarter of the panel with 
the shrine or altar which stood close by in the south-east angle, and 
suppose that even in his glowing picture of the fruitful earth, blessed with a 
luxuriant crop that leaves enough and to spare for bird and beast as well as 
man, the artist has found means to suggest the reverence due to the god and 
his gifts. He shows us the wild creatures gleaning, but hints that the 
boldest of them spares the clusters that ripen in the shadow of the god's 
altar. The explanation may seem fanciful ; at any rate it is not unlike the 
fancy of the man who wrote /movov firj vBwp, Give them water and they will 
swim, among the fish of the adjoining panel. 

I have claimed for the panel of vines a good deal of originality. Among 
published drawings of mosaics one looks in vain for any that closely re- 
sembles it. But its general scheme, the decoration of a rectangular panel 
by means of tree-like forms springing from the corners, was by no means a 
new one. It may be traced back to the fashion of filling the spandrils of a 
square panel containing a round medallion — spaces such as in our fish-panel 
are occupied by masks — with branches issuing from a stem or vase set 
diagonally in each corner; and this fashion, which appears several times at 


Pompeii, was doubtless borrowed from the favourite vasc-and-foliage border, 
of which our scroll-border with its vase at each angle is a good, though late 
and elaborate, example. 

The nearest parallels to our design are furnished by some vine-mosaics 
which have come to light in North Africa. They seem to mark an earlier 
stage of development ; the vines spring formally and symmetrically from 
vases placed in the angles ; they are not allowed to cover the whole field, but 
orm a broad frame to a central picture-panel. It is as if their derivation 
from the scroll-border were still remembered. On the other hand the birds 
among the branches and the Cupids busied in gathering grapes show that the 
frame is in process of acquiring an independent pictorial importance. In a 
mosaic from the Arsenal at Sousse (Hadrumetum) just published by 
M. Gauckler {Rev. Arch. 1897, PL ix. p. 8 ff.) the central picture repre- 
senting the Triumph of Dionysos is surrounded by a comparatively narrow 
frame of interlacing vines. In a magnificent design which is the principal 
glory of the House of the Laberii at Uthina (published by the same writer 
in Monuments Piot, Vol. iii. PI. xx.-xxiii.) the vines have encroached much 
further, and the central picture has become subordinate to the animated 
vintage-scene. A third design of the same type, found in the baths at 
Kourba, forms part of the rich collection in the Bardo at Tunis. In each case 
the African vine-mosaics contain a central picture representing Dionysos ; 
the omission of this feature, as well as of the angle-vases and of the Cupids 
gathering grapes, is in keeping with the greater simplicity of the Melian 
panel. That the general idea of the design was commoner than the few 
instances which I have collected would imply, is made probable by its wide 
distribution in early Christian times, when the imagery of the vine, beloved 
in Jewish poetry and Christian teaching, was reproduced in every branch of 
art. The vintage-mosaic on the ceiling of the ambulatory of Santa Costanza 
at Rome, a church built about the middle of the fourth century, is a good 
instance of a perfectly pagan design adopted for the sake of its associa- 
tions.^ The amm'ini who are plucking the grapes, leading wains and treading 
the wine-press, were doubtless felt to be incongruous in a church ; they 
do not appear in the later vine-mosaics. Rather the Christian significance 
of the design is set beyond doubt by some such inscription as that of a 
mosaic-paved apse at Ancona in which every leaf has the form of a cross — 
Vinea facta est dilecta in cornum in loco uheri.^ 

It is in keeping with the relatively late date of the Hall of the Mystae 
that the panel of the vines finds its closest parallel in the pavement of a 
Christian basilica. The same symbolism is Dionysiac in the one case, 
Christian in the other. The design is essentially the same, and it is difficult 
to believe that the interval of time between the two can be much more than 

^ De Rossi, Musaici cfistiani, xvii., xviii, ' A variant from the Vulgate of Isaiah v, 1. 

A coloured paper cast and a coloured diawing Bull, crist. 1879. Taf. ix., x. Kraus, Geech, 

by Zeri are exhibited in the South Kensington der chriatl. Kunst, i. p. 298. 

70 n. C. BOSANQUET. 

a century. The pavement which so closely resembles ours is that of the 
church at Orl^ansvillc in Algeria; an inscription which forms part of the 
mosaic shows that the building was begun in 324 and completed before 340.^ 

It is commonly said that in the early Christian centuries the use of 
mosaic pavements diminished, and mosaic work was almost confined to walls 
and ceilings. Of late years however a surprising number of Christian 
mosaic pavements have been discovered in Syria and Palestine. Several of 
them have the spreading vine pattern. The best known instance is the 
pavement of a church discovered by Kenan's expedition at Kabr-Hiram near 
T3a-e and afterwards transported to the Louvre.^ The general design recalls 
the African mosaics published by Gauckler; four vines spring from vases 
placed in the corners of an oblong panel ; their branches however are quite 
formally arranged so as to encircle a series of medallions placed in rows of 
five across the design. An inscription fixes the date of the pavement 
according to Kenan's interpretation at 575 a.d., in the reign of Justin II. 
De Rossi ascribed the vine-panel on the ground of its style to the fourth 
century, but later discoveries seem to confirm Kenan's conclusion. In 
particular two mosaics of this type have been found at Jerusalem, one on 
the Mount of Olives in 1871, the other outside the Damascus gate in 1894.^ 
Both bear Armenian inscriptions ; the former can be dated with comparative 
certainty to the middle of the sixth century ; while the latter, as Mr. A. S. 
Murray has pointed out, though retaining much of the refinement of classical 
work, may well belong to the vigorous art of the age of Justinian. A simpler 
mosaic from Medaba in Moab {Pal. Fund Quarterly, 1895, p. 207) resembles 
the older type in having a single medallion containing a head as the centre 
towards which the diagonally placed trees converge. 

Our Melian vine-panel seems to be a link, geographical as well as 
chronological, between the two main groups of similar designs ; those from 
North Africa, which are at their best in the second and third centuries, 
and those of Palestine which seem to date from the fifth and sixth of our 
era. In Africa as elsewhere there has been a tendency to place the decline 
of mosaic-work too early ; a study of the mosaics from Carthage in 
the British Museum shows that good work w^as done there long after the 
time of the Antonines, and the same view is maintained by M. de la 
Blanchere {Collections du Mus^e Alaoui, 1890, p. 17 ff.) in publishing the 
spirited groups of race-horses from Hadrumetum which he assigns to 
the fourth century.* There is a rich field for investigation alike in Africa 

1 Rev. Arch. iv. (1847), PI. 78, p. 661. » Renan, Mission dc la Phinicic, PI. xlix. 

Traces of a fish-panel were found in the same p. 607. 

church. This juxtaposition of earth and sea, ^ Photograph in Quarterly Statement of Pales- 
conventional in pre-Christian mosaic, and re- tine Fund, 1894, p. 261. Cf. Mr. A. S. Murray's 
tained perhaps because to the Cliristian the note, ib. 1895, p. 126. 

fish as well as the vine liad a mystic meaning, * To the fourth century too the Britisli 

is seen in other early basilicas of North Africa, Museum authorities ascribe the Carthage pavc- 

e.g. at Tipasa, Milangei d'ArcJt. et d'Hist. torn, ment of the Months. Its decorative design of 

xiv., and at Sertei, Mdangcs G. B. de Rossi, p. cypress-like trees growing in vases and converging 

^^^- towards the centre is a very beautiful develop- 


and in Palestine, and for the present it would be premature to do more than 
indicate the general relationship of the two groups of designs. It must be 
remembered that they are not likely to have been the exclusive property of 
mosaic-workers, who often boiTowed and adapted the ideas of wall-painters 
and modellers in plaster.^ Mr. Cecil Smith has already hinted at the possible 
influence of similar textile patterns in comparing the Melos vine-panel to 
some of the older Persian carpets.^ The tree with birds in its branches, 
springing sometimes from a kantharos-like pot, sometimes from a mound of 
earth, is a favourite subject in woven stuffs and embroideries in Persia, India 
and even China. 

The birds are for the most part conventional, always excepting the cock 
(PI. III.). The gazelle (PI. II.) was perhaps intended to represent the wild 
goat of the Cyclades, which still survives on Anti-milo ; but the figure which 
the ylrTi<f>o$€TT}<i chose from his pattern-book was certainly drawn from a 
North-African gazelle — a striking proof of the North -African influence 
which we have already had reason to suspect. It may be compared 
with a reclining gazelle which is represented eating grapes from a 
basket on the Sousse Arsenal mosaic. The crouching hare of our panel 
finds a parallel in the same part of Africa.* 

The Panel of Fish (Plate I.). 

To the picture of fruitful earth succeeds one representing the harvest of 
the sea. It is less elaborate than some of the fish-mosaics in the Naples 
Museum which seem to present a side-view of a tank oi* a section taken 
through the sea, with the surface marked by a line near the top of the 
picture and fish swimming to and fro in the green water : and less ambitious 
than the great floors representing the Triumph of Neptune and Amphitrite 
which have been found in the provinces. In Roman Africa pavements 
representing fishing scenes were often the appropriate ornament of an 
imphtvium. The inherent qualities of glass-mosaic are so well adapted to 
depict the gleaming scales and iridescent colouring of fish that the subject 
became increasingly popular. 

Just as in the preceding panel the principal figures are placed upon the 
north side of the hall, so here the position of the fisherman and the motto 
over his head presuppose that the spectator stands on that side. But this 
also is a decorative composition, not a realistic picture, and is meant to be 
intelligible from whatever point of view it is seen. The throng of darting 

mcint of the older trcc-pattorns. Archaeologia, of late classical architecture. TexierandPuUan, 

xxxviii. PI. ix.-xiii. It was originally planned PI. xxxiv. 

for a dome rather than a floor. Cf. Garnicci, ^ Compare a tile-work design of vines, grapes 

Arte cristiana, iv. Tav. 255. and birds from ' the south gate of the Tope 

' The ceiling-mosaics in the sidc-chapcls of Maidan, Teheran, 17th cent. ' reproduced in the 

S. George at Salonica imitate not only the Cross Gallery of the South Kensington Museum, 

painted coffers but also the cornice-mouldings ' De k Blanchere, Mtts^e Alaoui, p. 25. 


fish follows the circular frame in an endless wheeling movement, for which a 
fixed centre is supplied by the absurdly disproportionate figure of the little 
fisherman perched in their midst.^ Owing to the injuries done to the panel 
by previous excavators no part of the boat remains ; Mr. Clark has restored 
it on the analogy of many similar mosaics, making the boat nearly as dispro- 
portionate to the man as he is to the fish. One could imagine him fishing 
from a rock like three fishermen figured on a silver patera from the coast of 
Algeria 2; but this would be unusual in a mosaic. The boat on Plate I. 
is sketched in from the fish-panel at Sousse, a picture which furnishes a 
vivid illustration of a passage in Aelian describing the four methods of 
fishing: SiKTveia or netting, K6vTco<ri<; or spearing, Kvpreia or catching by means 
of the Kvprrj (Latin nassa), a basket-trap like our " weels " and " eel-bucks," 
and a'^KKTTpela or angling with hook and line.^ Of the four methods Aelian 
considered line-fishing the most sportsmanlike, and trapping the least worthy 
of a free man. The Sousse mosaic when it was complete had a boat in each 
corner ; in one the fisherman is striking a fish with a trident, in another he is 
about to cast a net, in a third he holds a cord to which three bottle-shaped 
basket-traps are attached ; the fourth comer, in which a'yKKnpeia was 
doubtless represented, has been destroyed. The fisherman on the Melian 
panel holds a rope, the loose end of which passes under his left arm, but we 
have no means of deciding what was at the end of it. Like other boatmen 
from Charon upon the Ickythoi onwards he wears the chiton exomis. 



Fig. 5. 

The words fiovov /jlt) ^hwp picked out in black tesserae on the white 
ground above the fisherman's head (Fig. 5) have been happily explained by 
Dr. Sandys, who compares them with Martial's Epigram I. xxxv. 

Artis Phidiacae toreuma clarum, 
Pisces adspicis ; adde aquam, natabunt. 

' The idea of fish swimming in a circle is " Aelian, N.A. xii. 43. M. Gauckler does not 

used with equal eflFect, as Mr. Cecil Smith points mention the passage. I am in many ways in- 

out, on a series of red- figured plates from South debted to his full and interesting essays on the 

Italy, among the latest examples of Graeco- mosaics from Sousse (ifcv. ^rcA. 1897, (2), pp. 8- 

Itdlian paihted pottery, which were probably 22; fishing-scene, Pi. xi.) and Oudna (Afonu^ 

intended to be used as fish-plates at table. ments Piot, iii. pp. 177-229 ; fishing-scene, p. 

BrU. Mus, Vase Catal. F. 259 ff. 198). In Plato's Si>phist the art of fishing is 

' Found between Tipasa and Cherchel. Bui- resolved into fpKoOriptKi}, rpioSovria, and atrira- 

tetin Archdol, de la Comiti cki Travaux Histo- MfvriKii. 
riqute, 1893, PI. x. 


Martial is describing a chased silver bowl, an old piece of Greek plate ; when 
it was filled, the fish with which the interior was decorated appeared to swim, 
just as the ships painted in certain black-figured ky likes floated when they 
were filled with wine. 

In compressing some such epigram into three words the artist has 
made the point a little obscure. His self-praise was certainly justified 
during our excavations by the expert criticisms of local fishermen. They 
readily recognised and named most of the fish, and were never tired of 
admiring the life-like play of light and colour on the scales. One detail 
baflfled them as well as ourselves — the globular object with a slender neck 
to the right of the fisherman. It looks like nothing so much as a gourd- 
shaped glass bottle, three parts full of dark-blue liquid, the upper part being 
empty and transparent ; but this does not correspond with any known kind 
of fishing-appliance ; neither a Kvprrj nor a gourd-float would be trans- 
parent ; so we are constrained to suppose that it represents some marine 
creature. The name nlva (classical irivva, a bearded mussel) was proposed 
and rejected. 

The Geometric Panels and Border. 

The two geometric panels are typical specimens of Roman provincial 
mosaic. They are not original compositions like the two preceding panels, 
but stock designs. In skeleton, as is shown on the key-plan (Fig. 4), they 
are based on different systems of intersecting octagons, such as any one 
experimenting with regular geometric design must inevitably discover. In 
the case of the long panel the intersecting sides of the octagons bisect one 
another, and divide each octagon into a square and four hexagons. The 
next step is to subdivide each hexagon into a square and four rhomboids. By 
this device the original octagonal planning is effectually masked, and a 
cross-like form made up of eight rhomboids becomes the predominant feature 
of the design. In this form the pattern appears at Pompeii ; it is increas- 
ingly common in the provinces during the second and third centuries. 

The design of the western panel, where the octagons intersect at their 
angles, is at no time so common as the other, and is hardly found before the 
third century. Almost the only building, besides the Hall of the Mystae, 
where these two patterns occur together, is the somewhat late British villa 
of Weldon in Northamptonshire.^ 

The character of the border is a further evidence of late date. The 
6wastika-V[k.e wheel-pattern is very common in the later floors of Britain, 
Gaul, and Germany. The florid scrolls of the vase-and-foliage border find 
parallels in Christian rather than classical mosaic; and the proportion (1:4) 
of the width of the border to that of the whole mosaic is characteristic 
of corridor*pavements in the third and fourth centuries A.D.^ 

^ Lvaons, Rel. Brit. Bom. I., Pt. ii., PI. vii. PI. 41. The latter can hardly be earlier than 
2 E.g. at Silcheater, Archaeologia, Iv. p. 241, the 4th century. 
and at Halicarnassus, Newton, Histt Diseoveriee, 



The SculjHiirc and Inscriptionfi. 

The clue to the identification of the building had already been given by 
the basis dedicated to Dionysos Trieterikos (Fig. 7 below) when in demolishing 
the great traphos we came upon a headless statue which proved to be a 
portrait-herm of a hierophant, Marcus Marius Trophimus, set up by the 
Mystac (Fig. 6). One of our workmen had previously told us of a head 

Fig. 6.— Statie oy Hierophant. Height 1-80 in. 

which he had found in the same part of the 
its beard and wreath enabled us upon ou 
unpublished head which is thus described 
Athens Museum : ' 329. Portrait-head of a 
short beard and moustache and curly hair; 
times. Found in Melos, and bought by the 
Eye-brows and pupils indicated. End of 

building, and his description of 
r return to identify it with an 
in Cavvadias' Catalogue of the 
man wearing a wreath, with a 
small life-size ; work of Roman 
Archaeological Society in 1884. 
nose broken. Parian marble ? 


Mr. Cecil Smith has since taken a cast from the neck of the head in Athens, 
and tried it upon the herm in Melos ; the two were found to join accurately at 
the back of the neck ; in front the surlaces had been chipped and did not 
meet, but the identification was quite satisfactory. The marble of both 
head and body is singularly white even for Parian. 

The statue represents — or will, when head and body are united — a 
middle-aged man with broad face, full cheeks, curly hair, and clipped beard. 
The wreath on his head is of ivy and flowers, a wreath such as Dionysos often 
wears. He is dressed in a chiton which is girt up above the knee with a 
deep fold falling over and concealing the girdle, a nebris confined by a broad 
belt and passing over the left shoulder, and a mantle. Part of the mantle is 
brought forward and thrown over the right fore-arm, so as to provide the 
starting-point of a puyitdlo to support the right hand ; the stump of it is 
seen on the drapery below the break in the arm. The right hand was found, 
and when it is readjusted the arm will be complete but for some of the 
fingers ; there is evidence that the hand held a curved vase, phiale or 
kantharos. No part of the left arm, which is broken above the elbow, was 
discovered. It was sharply bent, and the hand was raised nearly to shoulder 
level and probably held some heavy attribute such as a thyrsos, to judge 
from the stump of a cross-support which projects from the upper arm. 

In accordance with a custom of ancient and more especially Oriental 
religion the priest is here represented in the character of the god. There are 
several much-restored copies of a statue which represented Dionysos as 
wearing a girt-up chiton, a nebris confined by a belt, a mantle, and long 
hunting-boots.^ The figure is usually restored as holding a kantharos in the 
right hand, and a thyrsos in the outstretched left ; the left arm rests on the 
head of an archaic idol. The type was known in the Cyclades ; it appears on a 
late Greek silver coin of Andros (B. M. Catalogue, Crete and Aegean Islands 
p. 86, No. 2, Plate XX, 10). The obverse is * Head of young Dionysos r., his 
hair long and wreathed with ivy,' like that of our statue : reverse' ANAP[inN] 
youthful male figure (Dionysos) I., wearing short chiton ; his r. is extended 
down above a tripod, his left is placed on the top of thyrsos (?)'; I have 
examined the coin itself and thought that I recognised a nebris passing over 
the shoulder. There was a famous temple of Dionysos in Andros, containing 
a fountain which ran wine at the festival called ®eoSac(Tia ; the figure on the 
coin probably reproduces the temple-statue. It recalls Callistratos' descrip- 
tion of a bronze Dionj^sos by Praxiteles, which wore an ivy-wreath and a 
nebris and held a thyrsos in the left hand. The Deepdene statue is decidedly 
Praxitelean,^ and may be directly related to the type which was worshipped 
in Andros and was adopted, as the attire and attitude of the Hierophant 
prove, by the 'lepol Mvo-rai of Melos. 

' Clarac, Vol. IV. PI. 695, Figs. 1614, 1615. paidalis .somctiines takes the place of the nebris. 

The former, at Deepdene, = Michaelis, Ancient In our figure tlie artist has compromised : the 

Marbles in Great Britain, p. 280. For other hoofs are cloven, but tlie mask is a panther's, 

instances of the type, see Roscher, Myth. Lex., * Cf. Furtwiingler, Master-pieces, p. 323, 5. 
p. 1133 {Dionysos in Art, by Thriimer). A 



The lower part of the figure consists of a plain shaft, on which the 
inscription is cut, and a larger base meant to fit into a pavement. The back 
and sides of this base arc for the most part left rough, which accords with 
the supposition that the hcrm stood in the niche (C in key-plan) near which 
it was found. In front the smoothly dressed surflxcc shows where the base 
met the marble stylobate. Three akanthos leaves spring from the pavement- 
level and clothe the foot of the shaft. 

fC3 ^ 

Fig. 7. — Mauble Basis. 

The inscription, which in the forms of its careful deep-cut lettering 
resembles that on the Dionysos Trieterikos basis (Fig. 7), cannot be much later 
than the end of the second century ; on the other hand the coarse and ugly 
workmanship of the hierophant and the mouldings on the basis make it 
difficult to date either of them as early as Hadrian's time. Both may be 
older than the Hall of the Mystae ; at any rate the niche looks as if it had 
been built to accommodate the herm. 

A later group of inscriptions, in which C and W are used for 2 and n, 
consists of the prayers to Athene and the Fortune of Melos incised on two 
columns of the Hall, and the dedication on the bust of Aurelia Euposia (Fig. 8> 
To these we may perhaps add the inscription on the mosaic (Fig. 5 above), 
which seems to have had W in the last word. The poor style of the bust, 
in particular the clumsy lines of its rectangular pedestal, and the rudeness 
of the sculpturings on the columns mark a further style of degradation ; 
they may belong to the early part of the third century, when the names 
Aurelius and Aurelia were very common. The phrase iv to3 ihim avrfj^ 
epyw implies that the Hall or some part of it had been, built or restored at 
the expense of this Aurelia Euposia. We have already seen reason, on 
grounds of style, to assign the mosaic to the first part of the third century. 
It may have formed part of the epyov in question. Alexander, who on the 
column-inscriptions is called /cTtVrt;? of the Mystae (Fig. 1), must have earned 
this honorary title by sotoe similar benefaction ; it is not necessary to suppose 



that ho was the original founder of the Society. His appeals to the 
favour of Athene and Tyche, the guardian-goddesst's who appear on the 
Roman coinage of the island, suggest that the cult of Uionysos Trieterikos 
had been newly introduced and might arouse the jealousy of the older 
divinities ; ^ and the fact that these appeals were incised in a prominent 
position on columns of the Hall may mean that he was responsible for the 

With regard to the style and date of the three heads found upon our 
site (Cavvadias, Catalogue, 329, 424, 459) Mr. Crowfoot, who has made a 
special study of portrait-sculpture, writes to me from Athens as follows : — 

Fio. 8.— Bust of Aurelia Euposia. 

' These heads all seem to belong to the same period, the early decades 
of the third century A.D. The close-cut hair of the boy is similar to that 

^ The worship of Athene, as the very archaic 
character of the xoanon on coin and column-relief 
shows, was much older than that of the Tyche of 
Melos. But the latter patriotic cult may have 
been established as early as the 4 th century 
B.C., by the remnant of the old population 
whom Lysander sent back. Cf. the Tyche 
made liy Praxiteles for Mogara, and liis ' Bona 

Fortuna,' which was at Rome when Pliny 
wrote. The people of Antioch were doubtless 
following an established fashion when they set 
up a statue of the Fortune of their city 
early in the 3rd century. See Wolters' article 
in Ath. MUth. xv. For the Melian type of 
Tyche bearing the infant Plutus, cf. the statue 
at Thebes, Pans. ix. IG, 1. 


worn by Alexander Severus and his successoi-s, and the short curls of the 
hierophant to those of the emperors at the beginning of the century. The 
coiffure of Aurelia may be related to some of the fashions which prevailed 
in Rome during the second quarter of the same century, or may be a 
modification of an earlier fashion, set perhaps by Julia Domna. (It would 
obviously be rash to say that the Mclian ladies were always- successful in 
copying the short-lived fashions of the capital.) Such a date suits the stylo 
perfectly. The bust of Aurelia is the rudest, but all three are as good as 
most contemporary Athenian works. The surface is polished and the eyebrows 
not raised but incised ; in both points this is a contrast with the treatment 
usual at Athens about the middle of this century, of which we have a dated 
example in the Kosmete, No, 388 (Archonship of Kasianos, 236 or 245 A.D. 
Cf. Dumont, Sur VEpMhie attique, I. p. 247). The Melian works are at least 
successful in portraying distinct characters, and are interesting, therefore, for 
the light which they throw upon " certain people of importance in their 
day." The sour face of the hierophant is hardly more attractive than the 
lady's expression of obstinate bigotry; combined they are sure evidence of 
the psychological atmosphere of the third century, and differ strikingly from 
the air of hlas4 refinement which is dominant among the Antonines.' 

To the inscriptions already published may be added a mason's mark 


cut on the top of a column-drum at the S.E. angle. 

The Society of Mystae. 

Associations of worshippers of particular deities had been common in 
Greece since the fourth century B.C. under the name of 6py€U)ve<i* Oiaadrai,, 
or ipaviarai} The Mystae of Dionysos Trieterikos may be compared with a 
number of other Dionysiac societies, calling themselves ol Mvarai, which 
flourished especially in Asia Minor and Thrace during the second and third 
centuries A.D. We find them at Smyrna (with a cult of Dionysos B/aeto-ev?), 
Ephesus (cult of Demeter and Dionysos <I>\€6>9), Teos (Dionysos Xr}rdv€io<;), 
Magnesia on the Maeander, Seleucia in Cilicia (D, ' Apx^^aK^of), , in Western 
Thrace (D. Bot/ji/?), and at Apollonia on the Black Sea, They had much in 
common with the associations called oi Bax^ot and to BaK)(^etov, which 
existed in the period at Athens, Megara, Cnidos, Cyzicos, Perinthos, Thasos, 
and Tomi. Our knowledge of both groups of societies is derived from 
inscriptions, of which the most important is one found at Athens which con- 
tains the statutes of the lobacchi and the minutes of one of their meetings,^ 
This curious document gives the most minute information about the 

* The inscriptions relating to these and other from Foucart's Associations rfligituses. 
associations among the Greeks have been col- ^ Published by S. Wide, Ath. Mitth. 189-1, 

lected by Erich Ziebarth, Das griechische p. 257, and by E. Maass, with fuller com- 

Vereinsvouen, Leipsic 1896. The facts which mcntary, Orpheus, p. 18, 
follow arc drawn in the main from this work and 


constitution and procedure of a Bacchic society, the election of members, 
entrance-fees and subscriptions, the duties of the officers, the meetings at 
wliich the members drank wine in the Society's banqueting hall, and the 
strict rules which were found necessary for the preservation of order. 

The Mystae of Mclos honoured their hierophant with a portrait-statue. 
About the same time, early in the third century, to UpuyraTov viov Buk^iov 
of Thasos paid a similar compliment to their hierophant. We also hear of 
a hierophant in connection with the Mystae of Ephesus and of Magnesia. 
The fact that at Cyzicos the names of the hierophant and the fivaTdp^r)<{, 
followed by those of the Mystae, appear in a list of public officers, shows 
that there at any rate they occupied a prominent position. The officers 
and many of the members of these societies were persons of good birth 
and standing. In many cases women were admitted to membership and to 
office. The ^dK')(pi of Tomi are called Ylaaovf; lepof diaao^, apparently 
after their foundress. There is nothing unlikely in the assumption that the 
rich lady whose ep^ov is mentioned on her bust (Fig. 8) was a member or 
even an officer of the Melian Mystae. 

The inscription on that bust raises a new point of some interest ; it 
reveals the existence within the society of a body called oi vepcficofiioi. We 
might suppose that these were members who had attained a higher stage of 
initiation, privileged perhaps to take part in some sacrifice or choric dance 
vepi ^(ofiov. But the inscriptions which give so full an account of the 
organisation of these societies say nothing of such a subdivision. On the 
other hand we find constant mention of a throng of functionaries, who bear 
a great variety of names. The lobacchi had six officials, the Mystae of 
Magnesia five. The BovAcoXot, a Bacchic society of Pergamon, had an 
dpxi/3ovKo\o^, a secretary, two singing-masters, three Sileni and a choragos.^ 
The height of extravagance is reached by the Mystae of Apollonia on the 
Black Sea, whose eight officers bear names suggestive of the cult of Zagreus 
and of the Trieteric festival that was celebrated on Parnassus and Cithaeron 
and in Crete. The Mystae of Dionysos Trieterikos in Melos may also have 
had their \iKva(f>6po^ and Kpar'qpiap')(p<i, their apj^t^aaa-dpa and Ki<Tra(^6po<i. 
By oi Tr€pifiu>fiioi we should probably understand the whole body of officers. 
This interpretation may help to explain the only passage where the word 
occurs in classical literature. Juvenal (ii. 16) describes a man of infamous 
life, evidently a well-known character, under the name of Peribomius. There 
is a scholion, Peribomius : nomen archigalli. If the person referred to was a 
priest of Cybele, and if, as our inscription suggests, nepijdco^iofi was a general 
title for the functionaries attached to Asiatic cults, the name chosen by the 
satirist conveys just the discreet hint which might be expected. 

The discovery of the hall in which the Mystae held their meetings is an 

' The twelve priests mentioned as conducting ciation of liirrai or ^Jlkxoi. This description 

a Dionysiac festival in Patmos, in a passage by a hostile writer shows us a Bacchic society at 

quoted by Maass, Orpheut, p. 52, from the Acta its worst, just as the rules of the lobacchi 

of John, were probably the officers of an asso- show one at its best, 


important addition to our knowledge of tliese associations. Such halls are 
mentioned in inscriptions under the names oUia, oIko<;, or iepov. Its 
resemblance in general plan to the Hall of the lobacchi at Athens (p. 65, note 1) 
confirms the view already expressed as to the similar character of the societies 
of fxixTTai and ^aK-)(oi. 

The building remained in use for a considerable time, so long that in 
several places the mosaic became worn ; instead of being repaired or renewed 
it was roughly patched with bits of marble wall-lining. Judging from the 
fact that on Roman provincial sites it is not uncommon to find traces of 
throe or more tessellated pavements one above another, we may estimate the 
probable life-time of such a floor at from 100 to 150 years. There is no 
reason to suppose that the building was ever converted to other uses ; had 
that been the case the statue of the hierophant would not have remained 
unmolested in its niche, still less have been left in fragments on the floor. 
It looks as if in the course of the fourth century the meeting-place of the 
Mystae was first neglected, then deserted, lastly stripped of its marble 
fittings. The removal of the stylobate slabs which formed its socket would 
naturally occasion the fall of the statue ; it was pushed on to the pavement 
and lay there, broken by careless hands, but not mutilated by the spite of 
fanatics as were the torsos discovered in the Three Churches excavation 
{J. H. S. xvii. p. 131). Later the collapse of the roof buried it in fallen plaster. 
Last of all the construction upon the ruins of a broad cultivation-terrace 
preserved to our own day both the statue and the finest part of the 

The head and body of the hierophant are still separated. It is to be 
hoped that the authorities of the Athens Museum will not neglect the 
opportunity of securing the body, which remains at Melos in the warehouse 
of the proprietor of the site. By so doing they will double the value of the 
head which they already possess and add to their collection a new type of 
the highest interest. 

I have to thank Mr. Duncan Mackenzie, my colleague in the excavation, 
for a number of valuable suggestions. 


J.H.8.V0L.XVIII ( 1888) PL. IV. 


THE PH KY GO - J,M n A N 1 11 ( ) M IE K . 

^2 .-W 

W. Shcuwe F.RG.S. ISMaretonTerroJoe.FanUcoS'W. 


ShaveFR/iS ISMarttonT^raee.PimhccSW 

J.H 3 VOL.XVMI I898)PL.V. 


[Plates IV., V.] 


After exploring the north side of the Lycos valley, and before pro- 
ceeding down the right bank of the Maeander towards Ortakche (see Part I. 
init.), we spent a short time in tlie Phrygo-Lydian borderland, seeking for 
some evidence to clear up the uncertainties attaching to the situation of 
Sala, Tralla, Aetos, Kallatebos, and Apollonos-Hieron. (See Plate IV.) 

In discussing this district, I enjoy the advantage of having before me the 
work of three previous explorers, Prof. Ramsay,^ Prof. Radet, and the late 
Dr. Karl Buresch. I have been careful to quote the opinions of these critics, 
though I must sometimes dififer from them, because it is convenient for the 
reader to have before him all the views that are, or have been, held in regard 
to each town. Amid a few differences, I find myself in general agreement 
with Prof. Ramsay and Dr. Buresch. In the development of their views 
from the earliest to the latest stage there is, happily, a steady approximation 
towards the same results ^ ; and yet, by a strange accident, this approximation 
was quite unconscious, for Dr. Buresch knew only Prof. Ramsay's earliest 
views, while Prof. Ramsay in his latest work had only a very imperfect know- 
ledge of Dr. Buresch's earliest opinions.^ I regret that I am obliged to 
differ more widely from the results reached by M. Radet, and, as some readers 
might put a wrong construction on such expressions of dissent, I would here 
record my appreciation of the value of his fresh and suggestive work. When 
a subject is in process of growth, every suggestion and every fresh point of 
view are valuable, and it is only by due consideration and open criticism of 
all proposed combinations that we can cherish the hope of ultimately reaching, 
or approximating to, the truth. 

Sala. — Starting from a point opposite Tripolis, we proposed to keep 
along the left bank of the Maeander towards Tchindere keui, and then cross 

^ The obligations expressed to Prof. Bamsay come to regard it as ' anf richtigem Wege ' 

in Part I. have to be repeated here. {Beiseber., p. 103 = Aiis Lydien,^Y>. 201) : only 

- This is the more striking when we consider he would place the town nearer Blaundos, while 

that Prof. Ramsay had explored only the fringes Ramsay in the meantime has also found reason 

of this district. to bring it nearer Blaundos {CB. vol. ii.). 

' We may take as an example the case of Again, in regard to Apollonos Hieron their 

Mysotimolos. Buresch at first rejected Ramsay's earliest views diflFered greatly, while they arc 

view in the strongest terms, but recently he has now (as we shall see) practically agreed. 

H.S. — VOL. XVIH. a 

82 J. G. C. ANDERSON. 

over to Geune. A few minutes ride brouglit us opposite tlie hot springs 
(Ilidj'a), mentioned by Arundel ^ and Hamilton,'^ which are built over in the 
conventional Turkish style, and used as a hammam (bath). In half an hour 
more we came to another very hot spring (likewise on the right bank), which 
has formed around itself a rocky incrustation, over which the water flows 
down into the river. About this point we entered the great gorge in the 
Mossyna mountains, through Avhich the Maeander forces its way amid pic- 
turesque scenery into the low-lying valley of the Lycos, and we travelled up 
and down its steep sides for nearly two hours to Dere keui^ (850 ft. above 
the Lycos valley), and thence for two hours more to Tchindere keui (1100 ft. 
higher), near which we copied inscr. 14. On the opposite side of the canon, 
almost on the same level as Tchindere keui, lies the large village Geune, a 
governmental centre, and seat of a mudur. Our chief object at Geune was 
to find and examine the extensive ruins seen by Hamilton east of the village, 
between it and the Maeander. ' While crossing this flat country,' he says 
(II. 371), ' my attention was arrested by several square blocks of stone in 
the fields on the right ; and on proceeding to examine them I found myself 
on the site of an ancient city. The ground and walls between the enclosures 
contained many similar blocks, some of which were still in situ, others were 
pedestals, but without inscriptions, while broken pottery and tiles lay 
scattered about in all directions. The most remarkable feature was what 
may be called a street of tombs, extending in a north by east direction from 
the town. All of them had been much injured, but the foundations of many 
were still perfect. The whole area of the city had been ploughed over, but 
the remains of walls of houses and other buildings were everywhere visible. 
... A little to the south-west of the tombs were the foundations of a small 
building, with several broken columns five or six feet high still in situ. . . . 
The ruins extended on both sides of the road, and were in places much over- 
grown with vegetation. . . . The Turks call them Kepejik. . . .' AiTived at 
Geune, we naturally expected to have no difficulty in finding a guide to show 
us these ruins, which were so conspicuous in 1837; but no one in the village 
seemed to know anything about them. This extraordinary fact is confirmed 
by the experience of the late Dr. Buresch, who visited Geune some years ago^ 
but failed, notwithstanding the assistance of the ' courteous mudur,' to 
discover any one who knew anything whatsoever about the existence of 
Hamilton's city. Unable in the poor state of his health to undertake the 
task of searching for a site which was unknown to the natives, he naturally 
concluded, that ' like numberless other ancient towns, it had vanished from 
the face of the earth.'^ Such, however, is not the case. When Ave set out 

* Tfie Seven Churches, p. 227. in the full account of his researches, Aus Lydien, 
^ Researches i. \>. 526. epigraphisch-geographische Reiscfriichte hinter- 
' Not Dede keui, as in CB. i. p. 194. lassen von Karl Buresch (published just as this 

* At Geune I heard of the previous visit of an paper was going to press). I have added refer- 
archaeologist wlioni I inferred to be Ur. Buresch, ences to the latter. 

but unfortunately I liad not seen his /fcwc6cric^<, ^ Aber ich glaube vcrsicltcrn :.u diirfcn, doss 

which appeared in Bcr. dcr Kgl. Sachs. Ges. d. sic wic \ingczaliUe andrc vovi Erdbudcn vcr- 
JViss., 1894, p. 88-128, and has been reprinted schwundcn isl, p. 108 { = Aii,s Lydien, p. 205). 


in despair to search for the ruins, we found that they were not unknown to 
some of the peasants in the fields, and though nothing would induce them 
to leave their work and show us the spot, we ultimately discovered the site 
just where Hamilton placed it, on the level plateau fully three miles east of 
Geune. The ruins have become much more ruinous since Hamilton's visit. 
The stones have been thrown into ignominious heaps to make room for 
vineyards and cornfields, or used to build huts and outhouses ; yet a few 
blocks still remain in situ, and the foundations of a large rectangular building 
(with portions of one or two courses of stones) are still visible. The ' street 
of tombs' has entirely disappeared, but some fragments of columns and 
innumerable squared blocks extending over a large area on both sides of the 
road attest a city of considerable size. Not a single inscription is to be seen ; 
Hamilton found none when the ruins were more numerous and better 
preserved, and our search of two hours resulted in nothing but the 
discovery of the ' Constantinian ' monogram within a circle, neatly 

carved in relief on the end of a large rectangular block. 

Without excavation we cannot hope for documentary evidence to fix the 
name of this city.^ But the choice is small : it must be either Sala or 
Tralla. Now Sala was the more important of the two, and these ruins are 
by far the most imposing of the unidentified ruins which exist in this district. 
Moreover, as Prof. Ramsay points out {CB. i. p. 179), Sala, like Blaundos, is 
assigned to Phrygia by Ptolemy (v. 2, 27) and by numismatists, to Lydia by 
the Notitiae : hence both towns must be looked for in the same direction, and 
no site west of the road (or on the road) from Tripolis to Sardis could possibly 
be assigned to Phrygia. These considerations lead us to adopt the opinion of 
Prof. Ramsay (I.e.) and Dr. Buresch ^ that the site is to be identified with 
Sala. The perfectly defenceless character of the site, which may be compared 
with Bria (see Part I. p. 415), suits the view expressed in CB. p. 179 that it 
was a Pergamenian foundation established as a counterpoise to the Seleucid 
city Blaundos,^ Prof. Radet places Tralla here, as Prof. Ramsay did in his 
first essay ; Sala he puts far away at Gobek (En Phrygie, pp. 107-109). 

Aetos *. — 'Aero?, a place of importance in the Middle Ages, is mentioned 
by Nicetas in his account of the march of Frederick Barbarossa (1190 A.D.) 
from Philadelpheia to Laodiceia by way of Tripolis,^ and also in the Notitiae 
where it is conjoined in one bishopric with Apollonos-Hieron.^ Tomaschek 
has acutely pointed out that the modern village Aidos ^ on the upper waters 

^ Some evidence might be found if all the ' Nic. Chon. p. 539 ed. Bonn : 8<eb rov 'AtroO 

blocks could be turned over and examined : but x<^po^ Xtyofiivov Tropeufltj/rej, which means that 

our efforts were unrewarded. Frederick marched through its territory (not 

"^ In Anbclracht ihrcr Bedeutung bin ich . . necessarily through the town). 

geneigt in ihr Sala zu erkennen {Rciseb. p. 108 •* Not. X. 232 and XlII. 92, ed. Parthey and 

= Aus Lyd. -p. 205). Pinder. 

3 The Epirotename MOAOZOS occurs on 'Tomaschek, Zur hist. Topogr. ron Klein 

its coins (cp. Part I. p. 408). Hamilton could ^^^^^ ^^^ Mittelalter, p. 98. He falls into error, 

find no traces of fortification (II. p. 371). however, when he says 'der antike Name von 

* Cp. Hist. Gcog. p. 124, CB. I. pp. 194, 'A«tJ>s war 'AirdA.Xa>fos Uphv (Not. ep.),' mis. 

J 97 etc, understanding the meaning of <Jtoi. Aidos is 


84 J. G. C. ANDERSON. 

of the Kogamis, two hours WSW. of Geune, retains tlie ancient name. The 
old site however is, as usual, at some distance from the modern village. 
About two miles lower down the Aidos Dere, on its right bank, there rises a 
conical hill which the natives call Assar, isolated on all sides except the 
north, where it is joined by a low ramp to the ridge which bounds the JDere. 
On the hill there are distinct traces of ancient life : we saw a few blocks of 
stone lying about near the foot of the slope (including one or two marbles), 
the remains of a flight of stej^s leading up the SW. side, the foundations of a 
small Byzantine chapel on the summit, and small fragments of tiles and 
pottery strewn about over the hill-side. Amongst these fragments I picked 
up a few painted pieces which Mr. Cecil Smith has been good enough to 
examine : most of them may be late, but one fragment showing a series of 
concentric circles on a light-red background, a pattern so common on Cyprian 
ware, is (he says) genuine Greek work and may go back as far as the fifth or 
sixth century B.C.^ 

A glance at the annexed map (PL IV.) shows that Assar is a point of 
strategical importance, commanding not only the road from the Kayster Valley 
by Kirk Tchinar Devrent to Sala (Geune) and the East, but also, to a certain 
extent, the path along Karindjaly Dere ^ which Frederick Barbarossa followed 
in his march from the plain of Philadelpheia to Tripolis, avoiding the direct 
route through the pass Devrent (or Derbent) Boghaz.^ His army was attacked 
after leaving Philadelpheia, and by making this detour round by Aetos he 
avoided all necessity of fighting his way through the pass. 

A different identification of Aetos is proposed by Dr. Buresch. He 
would place it beside Kirk Tchinar Devrent (' Devrent of the forty plane- 
trees '), a village at the entrance to the pass, where there is an old site called 
Devrent Kalessi* similar to Assar: and he also suggests that Kallatebos 
(Herod, vii. 31, see below) may be placed there. He argues that in the two 
days' march from Philadelpheia to Tripolis Aetos was probably the inter- 
mediate station. Now, apart from the fort at Devrent, it seems impossible to 
find a site which will suit the dates of the march ; and this situation ' possesses 
the characteristic (a very important one for our identification) which 
Herodotus attributes to the position of Kallatebos, viz. irapiivat iraa-a avdyKrj 
fyiverai : that is, as tlie Persian marching from the Maeander plain to the 
Hermos valley ^ in 481 B.C. must necessarily pass the fort of Devrent, so the 
Crusaders in 1190 a.d. could not avoid passing it,'" But in estimating this 
theory we must bear in mind that his exploration was incomplete : he did 
not visit the site at Assar or observe that Aides retains the ancient name. 
Of Kallatebos we shall speak presently ; but the preceding exposition has 
shown that Frederick could and did avoid passing the fort at Devrent. 

now corruptlj- pronounced Aidas. Tomaschek's name). 

identification of Aetos with Aidos is accepted ' We followed this route in going from Aidos 

both by Prof. Ramsay and by Prof. Radet. to Baharlar : it is au easy road. 

^ So far, of course, as one may judge from a * Or Assar. 

email fragment. ' I have corrected the oversight vo)ii Rermos- 

• So called in Kiepert's map, but, I think, thalc nach dcr Maiandrosebcne. 

not rightly (tliough I am not sure ot the correct « llcisebericht, p. HI ( = Aus Lydicii, p. 207). 



Tralla, — But we^caii approve cveiythiug else tliat Dr. Burescli says about 
Devrent Kalessi. The hill (which is not very large but somewhat steep) is 
situated a quarter of an hour W. of Kirk Tchinar Devrent. Lying as it does 
at the point where the great road through the Kogarais valley emerges from 
the long narrow pass and crosses the road leading from south-east Lydia into 
the Kayster valley and thence to Ephesos, it is obviously of 'extraordinary 
strategical importance.' 'It would be strange' (as he says) 'if this 
important point had remained unoccupied... ; it might be said that Nature 
had expressly formed it to bear a castle and set it as a sentinel in this 
important place. Remains of late fortifications and foundations of houses on 
the not very roomy ^ summit, fragments of pottery and tiles scattered in rich 
profusion over the summit and slopes, a sarcophagus accidentally uncovered 
on the S.E. slope, and finally the Roman and Byzantine coins found here,^ 
show that right into the Middle Ages men have lived and watched here ' 
(p. 110 = Aus Lydien, p. 207). 

What name, then, is to be assigned to this site ? I think we must 
identify it with Tralla, following Prof Ramsay's first suggestion (in CB. p. 
180). Ho rightly says that 'the very name would suggest its origin in a 
settlement of Thracian mercenaries, who under the name of Tralleis or 
Traleis,^ served under the Pergamenian kings along with Mysian, Paphla- 

gonian, and other troops and Tralla perhaps lay near the Derbent- 

Boghaz, commanding the important road from the Hermos to the Lycos 
valley.' In later passages {e.g. pp. 200 n. 2, 580 f, 688) he inclines to 
identify Tralla with Aetos (which also occurs as a Thracian name, Niceph. 
Bryen. p. 149), taking Aetos as the name given by the Thracian Tralleis to 
their own town, while Tralla was the name used by others {GB. ii. p. 580, 
n. 5). But the fact that Tralla and Aetos are both mentioned in Not. x. 228, 
232, and Not. xiii. 88, 93 seems conclusive against this view.^ 

An important confirmation of our identification is to be found in the 
Peutinger Table, which places Tralla at a point where a road from 
Philadelpheia forks, one branch going to Peltai and Apameia, the other to 
Laodiceia. It is represented thus : — 


Laudicium pilycum 

' My note says, 'a fine space on the top.' 
"^ We may add (1) traces of steps up one side 

(2) quite near the foot remains of building in 

much better style than those on the summit. 

The cemeteries of the village contain a very few 

old stones. 

* Tpa\t~ii Insckr. Ferg. no. 13 : Hesych. 
s.v. TpaWtii (op. Tomaschek Die alten Thraker 
ii. p. 44). 

* His map (vol. i. ) rightly indicates Tralla 
somewhere in the Devrent Boghaz (in accord- 
ance with the view expressed on p, 180). 

86 J. G. C. ANDERSON. 

Prof. Ramsay, following a hint of M. Radet's, has shown that there are 
two roads mixed up here, Philadelpheia-Tralla-Peltai-Apameia, and Phila- 
delpheia-Tralla-Tripolis etc., which in reality coincide in jmrt (viz. as far ((s 
Tralla). Now Devrent Kalessi is just about 25 Roman miles from Philadcl- 
pheia, and the situation therefore suits excellently his proposed restoration, 
Philadelpheia xxv Tralla xxviii Atyokhorion, etc.^ {CB. ii. p, 580). In spite 
of the jumbling, the Table still rightly indicates the fact that at Tralla there 
is a parting of the ways, one going to Laodiceia, the other to Peltai and 

The road from the Kayster valley joined the Kogamis valley road at 
Tralla (beside Kirk Tchinar Devrent). In the upper part of the Kadi keui 
Deressi remains of the pavement of this road are still visible, and in an old 
cemetery hidden amongst trees, 25 minutes southeast of Kadi keui (which 
is about four miles from Devrent keui), I copied the following milestone, 
which makes Tripolis the caput viae. The stone is unfortunately in very 
bad condition. 

23. D fSj D(omino) N(ostro) 

rCC^C I A NO Fl. Cl. [lovjiano 


p . . . . 

A V Gs^_^ A . Aug. 

ATRIPOCTS a Tripolis 
Vl l.vl M X I Mi(Ha) 

There is a milestone of Jovian at Apameia {G.I.L. III. Suppl. 7054), 
otherwise we might restore [Iul]iano. 

At the bottom of the left-hand side there is a fragment of a Greek 
inscr. engraved the reverse way in small characters, which seems to refer to 
a different emperor (? Gratian), 

The form Tripoli is curious. 

Kallatebos. — From Aides we crossed over the hills to Baharlar, a 
village in the Kogamis ^ valley south-east of Ine Giol, and recopied there the 

* He ingeniously explains S[ocra]tu as a Latin (Vienna, 1888). 
corruption of 'j x'^P''"'''Ati;oj through the form ^ So on a coin (CB. p. 196 h. 3) ; Cogamns 

s-cor-atu. The reading is not certain, as may in Pliny, 
te seen from the photographic reproduction 



inscription in which M. Radet bcHeved lie hatl found documentary evidence 
for the situation of Kallatebos (Herod, vii. til).^ 

In the epigi'aphic copy I have trie<l to represent the inscription exactly 
as it is on the stone. The shaded part is a hole cut in the marble, which lias 
destroyed the important portion of the text. As I have already said,^ the 
present state of this stone is remarkable : some letters have partially, others 
wholly disappeared, and yet the polished surface remains intact. I have 
therefore not used the shading which is generally employed to indicate breaks 
in the stone, except where there'are actual breaks. 





EAYTOriAi. ~l>. 




"EtJou? . . . T»}9<Tapo<i 
z/t]«[7;9 /iir)(v6';) U]avi]fiov B', ol kcl- 
To]t/c[ot eV . . ]X)9ot9 eTetfirjcrav 
®[€vBdv ? S]€o8copov @€6<f>l[X- 
5 o]v ■^[pcoa Tov^ kavTwv evep^e\T- 
ri\v ape\Tri<i\ eveKev Kol evvoi- 
a]<? rr}\j; elsi] eavrov'i, d[vBpa 
a]<ya[66v], vfieiv <TTo[i]^'i [kuI ^ovX- 
evlrijpiov Kal eh a7ro[So^etoi/ 
10 v]8payQ)yiov eV [t]o[{} 6pov<i ajrepy- 

Notes. — L. 1, My impression was that the left edge of the stone was not 
broken. Buresch says that before the K the whole | [probably part of n] 

1 Radot, B.C.ff. 1891, p. 373 flf. ; cp. CB, i. 
p. 199 f. Buresch, Reisehericht, p. 112 f., 

(Aus. Lyd. p. 208) restores the first three lines. 
2 Athenaeum, Oct. 23, 1897, p. 566, 

88 J. G. C. ANDERSON. 

is preserved hart am Bruchrande : if so, the edge must have got rubbed down, 
and I have arranged the restoration on this supposition. L. 2, probably M , 
as Buresch says, the M being spread out. L. 4-5, tlic restoration of the 
name seems doubtful. L. 7, dvSpa suits the traces [the P is squarely cut] 
but the introduction of the second person in v/xeiv is very odd. L. 8, there 
is space for a letter between and A : o-rom? occurs C.I.G. 2483. L. 9, 
perhaps d7ro8o-)(eiov, i.e. 'reservoir* ( = €k86xI'Ov of CJ.G. 3454). 

The restoration of 7-10 is, of course, not certain : but the crucial line 
is 3. M. Radet wishes to make the name of the KaroiKia Kallataboi [or 
better Kallataba], assimilating it to Kallatebos, the town which Xerxes 
passed on his march between the Maeander and Sardeis. Unfortunately, 
M. Radet's restoration is quite impossible. The number of letters lost 
between I K and A B 1 2 cannot be more than six, allowing for at least one 
narrow letter ^ : seven occur in an equal space in 1. 9, but there the letters 
are slightly smaller.^ oi /ca[To]tAc[ot oi iv KoWarJa/Sot? or oi Kdp€<: oi iv 
KaWard^oK; (which is in itself most unlikely and does not suit the letters) 
are therefore both impossible, even if ol be omitted. It was audacious to say 
that ' Kallataboi, being given by epigraphy/ ought to be preferred to 
Herodotus' form Kallatebos {B.C. If. I.e. p. 376) ! Another suggestion is that 
of Prof Ramsay in CB. ii. p. 673 n. 5. He has come across a reference to 
'ApBaffav €v t^ kutcL rrfv ^pvyiav Mva-ia,^ which he thinks ' may very well 
indicate the Mysian country that lay south and south-east from Philadelpheia 
on the Phrygian frontier ' (ii. p. 573) : and he suggests oi /ca[To]t«[ot iv 
'ApB]d^oc^ as a possible restoration of our inscription. This is, however, 
slightly too long : and moreover the termination is most probably — X/9ot9,* — 
an objection which would also apply to the late Dr. Buresch's Kd[To]iK[oi oi 
iv] "AyQot?, giving a name "AySa, which is a possible form (cp. Steph. Byz. 8. v.). 
We must reluctantly conclude that the name of the KarocKia is lost beyond 

As to the provenance of the inscription, Dr. Buresch says that all 
accounts assigned it to an old site N.E. of Baharlar on the other side of the 
Kogamis, stretching from the foot of the hills nearly down to the river and 
with its N.W. edge not far from Bahadyr keui. I received the same account ; 
and the spot whence it is said to have been taken (beside an old cemetery 
beyond the river) was shown me by a villager who was present at the 
ceremony. The ruins at this site are extensive ^ but characterless. Now 

' This nearly agrees with Buresch : ' 4 um above statement is more exact. 

faivgreiche (als MfFH) oder zur Nolh i Buck- * Eusebius, Hwt. Eccl. v. 16 (reference to 

staben von Durchachnittshreite und 2 schmalc Montanus). 

(d. h. I ) verschlungen haben. My views were * I made a note to the effect that the letter 

formed at Baharlar before I knew of his paper. before B is apparently A, although it might 

' Allowing for two narrow letters (like I or possibly have been A (considering the way in 

P) and consideriug that the letters are not which parts of letters have disappeared), 

always of quite the same breadth, nor equally " But I did not think they extended nearly 

•paced, I said in Athenaeum that • the space ^o far as Bahadyr keui. 
cannot contain above seven letters': but the 


though we cannot restore KaWdra^a in tliis inscription, can we suppose 
that these ruins represent Kallateboa, considering that alter the foundation of 
Philadelpheia the older ttoXk? dwindled into a mere KaroiKui ? That it did 
so dwindle is most probable, and the site is not unsuitable, though one rather 
nearer Ine Oiol would be preferable. It is at least certain tliat Kallatebos 
was near Ine Uiol. Wiiat Herodotus says is, that in marching from Kydrara 
[Hierapolis] to Sardis it was absolutely necessary fur the Persian army to 
cross the Maeander and pass by the city of Kallatebos, Sia^rjvat rov Mat'av- 
hpov TTOTafiov irdaa dvdyKt] yiverai Kol iivai irapa KaWdTTjjSov ttoXip, iv rfj 
dvSp€<; 8r)fj,ioepyol fieXc €k /xvpiKr]<; re Kal irvpov Troievai, i.e. there waa only 
one possible road, the road which goes through the Devrent Boghaz and 
along the river-valley, passing by Ine Giol and Ala Sheher (Philadelpheia). 
And Hamilton observed that ' the tamarisk does not grow in the mountain 
passes, but occurs in great abundance in the valley of the Cogamus, near 
Aineh Ghieul ' (ii. p. 374). The old site opposite Baharlar, therefore, may 
very well represent Kallatebos, for the plain round Ine Giol would in any 
case be part of its territory ; but it is possible that the site of the city was 
nearer Ine Giol and has completely disappeared.^ 

Apollonos-hieron or Apollonieron has evidently been found by Dr. 
Buresch, who places it at an old site 2^ kilometres E.S.E. of B63 Alan. This 
confirms Prof. Ramsay's suggestion (C£. i. p. 194 f.) that it should be looked 
for at or near Bulladan, which is the important town of the district and a 
governmental centre (seat of a kaimmakam). I agree with Dr. Buresch that 
Bulladan is a purely modern foundation ; but to it have passed the heritage 
of Apollonieron and the name as well, for Bulladan or BuUandann is not a 
Turkish word, and Prof. Ramsay is clearly right in saying that it ' retains the 
ancient name Apollonieron, just as AbuUiont in Mysia retains the old name 
Apollonia.' ^ M. Radet places ApoUonos Hieron at Erziler, north of Aidos. 


On our way eastwards, we may be permitted to stop again for a moment 
in the Lycos valley to note an interesting discovery made by my friend M, 
Weber of Smyrna, who desires me to publish it here. Last autumn M. 
Weber found the ruins of the bridge which carried the road from Laodiceia to 
Hierapolis over the Lycos. This bridge was evidently a solid structure built 

* We cannot therefore approve Buresch's idea Herod, 'a words, which simply state the fact that 

(quoted above, under Aetos) that Eallat. might the only available road was that on which 

be placed at Devrent Kalessi. Speaking of the Kallat. lay. The Baharlar site is, therefore, not 

site near Baharlar, he asks KOnnte man nun a whit more ' unumganglich ' than Ine Giol or 

etwa diese fUr die Stelle des herodoteischen any other point on the road. 

KaWiri)fiosy der unumg&nglichen Wege- * For the loss of the a, see 6'^. p 195 7io<el. 

station, anspreehen t Ocwiss ist, dass man jener Compare also Bulaw6din or Biilaw^din, the 

Stdtte sehon weit eher als Ine Oj6l die Eigen- modern name of PolybStos. The sense in which 

schaften eines als 'unumganglich' gekenmeich- a modern town may be said to represent an 

neten Punktes zusprechen kOnnte (p. 117 = Aus. ancient one is clearly defined in Hist. Oeog, 

l/yd. 212). This entirely misses the point of p. 83. 

90 J. G. C. ANDERSON. 

of fine large blocks, with three arches, of which the central one is still 
complete. It is situated about half a kilometre to the north of the point 
where the river Asopos now joins the Lycos, the ground between it and the 
present course of the river being now an impassable marsh. The discovery 
is important as showing that the Lycos has here clianged its course 


Two inscriptions of Colossae have to be added to the small list we 
already possess. Tlie first is a dedication to Hadrian by a trihumcs militum. 

25. In a field near the ruins. 

AY TO K PA kvroKpd- 

T0PIKAI2A -^^p' K"'*^"- 

TJTPAI AWAA ^J' Tpami^oJ ['A- 

1\PIANA2E hpcav^te- 

nia AM A TTtVA.Ma- 

KEAHNXP// K^ro^vxV^- 

26. In another field. 

'AyaOrj Tu^j;. 
7ri'7/u-a^n79 roB' aeOXov iyoD Kdarcop aTreBiyfirjv 
veiK'^aa^ a6evapal<i 7raXa/u-[a]t9 d^V(TTO<{ d/jia>fio<i' 
ovBe yap rjv 6efii<; aXXo) dBe\(f)€iS TIo\vB[€v]kov<; 
^eipa^ dvacr')([efji]ivai 7rvy/jLri<i %ap[At'] dv[Ti]7rd\oLO, 
d6XodeT[i]^] B' e/u.€ Trat? [(name) Kpei-nov eveifie ? 

L. 4, XAPHAN" .InAAOlO. Nothing seemed to have been engraved 

after HAIZ. 

In the inscr. published by Waddngton, no. 16936, from a copy by M. 

Renan, read in 1. 1 [ dve6r]Kev Trj\v deav Tv-)(r]v Trji irarplBi. Cp., for 

instance, an inscr. of Antioch, Sterrett, E.J. no. 97. 


Three inscriptions of Sanaos, which had to be omitted in Pt. I. on 
account of the strict limitations of space imposed, are added here. 

27. On a large architrave block, in the cemetery below Sarikavak. 

^(,oB(iipov Kal KaWnrrpdrov t&v KaWnTToaTov o totto^. 



The two sons are the same who are nientionetl as liaving taken part 
with their father Kallistratos, son of Dioiloros, in presenting a ^ovXevTijpiov 
to Sanaos (M. Weber in Ath. Mitth. 1893, p. 207 = CB. no. H'Sy 

28. Used to cover a hciju in the watercourse of the fountain : dug up 
and replaced for a consideration. 

'Kv T]e9 - 7' ravTai^ aopoc<i 7rpoKe[Ky]- 
Sev]fievoc elalv irpo'yovoL Avp. Mo[i'- 
creou /3' ToO 'AttoXXcoviov, rrpoaeiri- 
K\rihev6y]aeTe he ktj b Mofo-eo? iv 
5 rfj Kara Bvcrfia<i {a)opS, iv y kol rj fxi}- 
Ttjp avTov 'AraXt? kuI rj /yvvi] 
avTOv 'A(f}(f)ia TrpoKeKrjSev- 
fi€[v]ai elalv el he Ti<i erepov kij- 
hevai iv Tai<i cropol^ Taurat?, drjcri 
l<; Tov (piaKov hr]v. irevraKia\i\t,a. 

L. 8 'A<f)(f>ia, more usually 'A7r(f>ca. On the name, which is probably a 
Lallnanie (a by-form of Appa), see Kretschmer Einleitung in d. Gesch. d. 
Oriech. Spr. p. 347. On kt) for Kal see CB. II. no. 678, p. 742. 

29. At Appa (Yokaru), near a well : small lettering. 

;7/HA€yeHC£T( ACK HOM OYC 6 OCf N 
'''"^A Y TC Y A 'jj <!> I A n P K f K H A 6 Y~^ 

M0¥ ^'A<t) 

Ovvofia fiev koXov 'E/j/a- 
iwi evddhe Kelrai \ avv TaT[i- 
a ya/j,€TT) (f)iKdvhpq) he fid- 
Xca-ra | kuI Tpo<f>ifiov tov v- 
5 lov- ovhevl he i^eaTiv dX- 
Xorpitp a-QifiaTi el<7€X[d- 
Iv,^ eX Ti<t he direiOrja-et rj d\X- 
Xo TI, Trpd^c, Oijaev 7rpoaT[ei- 
fiov hrjv. ,(i<p' . 

The first two lines are very rude attempts at hexameters. 


The plain east of Lake Anava, now called Taz Kiri, was probably 
Apameian territory, and it does not seem to have contained any town. There 
are however considerable remains, including many large blocks of stone and a 

' The letters have apices, and in 1. 2 the stone Ramsay's reading, CB. ii, p. 525, no. 367, 

has iarov. where M. Paris read (la[Ba'\Ke'iv {B.C. If. 1884, 

2 Cp. €V ris iLfiTTfKois, J.H.S. 1887, p. 393. p. 251). 
' Compare latXQlv at Eumeneia, ace, to Prof, 

92 J. (J. C. ANDERSON. 

few unimportant inscriptions, at Basmaktchi (where a weekly market is held) : 
and although they are mostly to be found in the cemeteries and have 
probably been largely carried from Sanaos, it is quite possible that there was 
a village here.^ A small settlement existed at Basmaktchi Yaila, a refugee 
village high up on Yan Dagh, three hours from Basmaktchi, on the path to 
Buldur. I copied there the following inscription which is engraved on an 
enormous block (now broken into two pieces) forming part of a heroon, the 
foundations of which have been uncovered by the peasants. 

30. In the epigraph ic text the two pieces are placed together. 

:y///////////////^//////////////^^^^^ M BQ M iM 

E T E I C I FM H H C rr POJ^r Vzl E H I M E N T 1 r E • 



. . . Ev/eX^9 rov KaWtAcXeof? ? tJov fiatfiov 
KaX\ TT)v <T[o]pbv Kol tA Trepl rov ^cofiov ^tov 
eavT^ Kal rfj yvvaiKl Tdra Karea-Kevaa-ev 
erei <r[^]y', fi'r)vo<i TrpcoTov ovSevl /x^vroi ye 
5 i]^ov yevqa-erai fierh rr}v Tekevrtjv 
Tov EvKkiov<i dvv^ai rrfv aopov 
KoX iireiaevevKeLv awfia, iirel o tovto 
To\^['^]<Ta<i vTrevOvvo^i yevrjaerai 
t? rov (f>ia'Kov tuv avT0Kpar6pa>v trpoa- 
10 Tci/Jbo) Srjv. TrevraKia'yjELKloi,^' rovrav Be to 
dvriypa(f>ov dTreriOr) h Tct dp-yela. 

The date is 178-9 A.D. Eukles and his wife Tata evidently belong to 
a family of Maximianopolis, whose stemma is traced in CB. I. p. 333. The 
Record Office is probably that of Apameia, unless Maximianopolis possessed 
one. Basmaktchi Yaila was probably an outlying settlement of the 
great Imperial Estate near the southern and western end of Lake Askania. 
Beside the village there are some rock-cut sarcophagi. 

' That the name 3iTo«57roXn (Hierocles) which as KovioiiroXn is of ^lowaoiiroXn, was showq 
M. Radet places at Basmaktchi (map in En long ago by Prof. Ramsay. 
Phrygie) is a corrupt form of 'AyaaraatoiiroXis, 



.31. Of another inscription on a cornice piece, now forming the lintel of 
the mosque door, I could make out little more than [top /3&>/a]oi/ aitu toU 

€<f)€<TTQ)<Tl KcloatV Kai . . . 


The Siblianoi arc placed by Prof. Ramsay in the marshy Maeander 
valley which stretches south from Ak Dagh and forms a corner of the great 
plain of Peltai. He thinks they possessed no proper TroXt? but retained the 
old village-organisatiqn, having three centres («w/Attt), one at Vicus (Tchaudir 
Tchiftlik), one at Boz-eyuk, and one at Khoma on the slope of Ak Dagh {CB. 
i. p. 222 fF.). Yet the coinage shows that at least ' in the early third century 
the tribe must have become more closely organised... and selected one of these 
villages as a TroXt? or city centre' (p. 225). The change seems to have 
occurred earlier: for the following fragmentary inscription, which can hardly 
be later than the early second century, mentions the Record Office of Siblia.^ 
This inscription, which is the only epigraphic evidence we have for the name, 
is built into the platform of the Railway Station at Evjiler ^ (close to the site 
of Lampe) ; it has been cut on all sides. 










[oySei/09 aWov 

eypvTo^ i^ov(T\iav KaTaOia-dai €T€po[v Trrw/ia* 

el 84 Tt9 T]oXfii]a-€i, vvevdvTiov t[c Trotrja- 
ai rj dWo Ti TTpd^ec], diroTeia-ei h to iep(OT[aTov ra- 

* Yet it may be held that this fact ia not in- 
consistent with the want of a real irixts ; for 
though the relation of the villages in a KO)ti6- 
■KoXis is an unsolved problem, we must infer 
that one Kw/irj was more important than the 

thers and formed a sort of centre. 

"It has been carried from one of the villages 
in the plain, no great distance. See the map 
in CB. vol. i. 


.1. G. C. ANDERSON. 

fielov 'ATTLKa^: S]iax^i\ia<;' tovtov dv[Tiypa<f)- 
5 Of aireTedr) l<i to] 'Si/3Xiav(t)v dp'^eia, €to[v<; Bca- 
Koaiocrroi) ? Ke\ SeKUTOV /xt](v6^) e' , ?'• e'crTa[t Se ovBevl 
i^ovaia ..].... et [yu.]^ TeKV0L<i fiov rj iyy[6voi^, el Be 
Tis eVepo? dvv^et toutJo to ri[p'\o)ov, virevOvvo^ e[crTat tcS 
TOV vo^ov^ Kii'hvvw Kal T&5 7rpo8r]\[ovfX6V(i) 

10 . 7rpO(TT€lfl(p. 

For the use oi' Ait'ic <h'((rh)ii(ir m this district, see Part I. No. 15. In 
the troublous Byzantine times Khoma with its strong fortress (kali) became 
the great centre. On a hot August morning I started up the mountain to 
examine the kaU and reached it after an ascent of fully an hour and a half, — 
alone, my attendants having fainted by the way ! It is a bare, isolated rock, 
standing out from the mountain-side in solitary grandeur, about 2,000 feet 
above the village. There is little to be seen now except part of a late wall 
on the narrow summit and a cistern (Jiammam) lower down on the further side 
(which is not visible from the top, but was reported to me after I had climbed 
doAvn again).! 


33. In the village of Evjiler. 

' O 

^V O, 













10 AEY0HNAIEHMO {sic) 
15 TICEniXHPHCEl//// 

09 Kal AvprjXia Ilp€l<TK[a 
rj avp/3io<; avTov ku- 
5 T€cr<C.<r'^Keva(Tav to [rj- 
pmov €avTul<i Kat 
To2<; TeKvoi<; avT[(t)v, 
€(f)' M ovK €^ea\Tat 

dXXoV €49 aVTO Kt)' 

10 hevdrjvat, e(l /jl)t) /xo- 
VOV<i TOP 'PoL'0[et- 
VOV Kal TTjV Up- 

ela-Kav Kal to. i^ a[y- 
Twv T€Kva' el B\e 

15 Tt9 inTLyr^prjo-ei [iV 

* Another kale Mas reported fvutlier to S.E. 
(nearer Diner-Apameia). 

See CB. p. 227 f., 347 etc. 




TO Tjpaiov TLva dell- 
vat, drjo-et ei<i ro l- 
€p(i>TaTov rafiel- 
ov ovofiari 7r(p)o(r[T€C 
20 fi]ov Srjudpia htayjiiXL- 
a TTivjaKocna. 

On the right-hand siilc is OYCpa), for which see CB. II. p. 395, no. 280. 
The date is 214-15 A.D., which suits well the use of Aurelia as 

34. Ibid. 


n. 'IovX{io<i) Tt/3(e/Jt09) 




A visit which I paid to Apollonia resulted in the discovery of some new 
inscriptions and the improvement of some already published.^ I failed, how- 
ever, to find the important inscription given by Prof Sterrett, W.E. no. 548, 

35. On a large rectangular block now used as a fountain-trough, in the 
vineyards below Ulu-Borlu. A large cross was afterwards cut on the stone 
where the breaks are shown. The block had to be displaced before the 
inscription could be read, and it was no easy matter to move it. 


KOA55NS2N ^ 

K.aicra[p]a M. Avpi](\iov) 

' Av[T(OV€t]vOV 

%e^a[(T]T6v, vi- 
5 ov Av\T\oKpdro- 
po<i<i A. 
X€7rTi/jLi,o[v] %ev- 
ovrjpov T\.\e\pTiva- 
KO<i %€^{a(rrov) rj ^ovXrj 
10 Koi 6 8r]fio<i 
' ATToXkcoviaTMV 
AvKicov ®paK<av 
K.6\<opcov ^ 

^ The koppa is slightly blurred on the stone, 
but it seems quite certain. 

» Inscr. of Apollonia CJ.G. 3969 ff. ; Le Bas- 

Wadd. 1192 ff. ; Sterrett, 
B.C.H. xvii. pp. 255-59. 

W.E. 517-554; 

96 J. G. C. ANDERSON. 

LI. 11-13. The title used by the people of Apollonia in the second and 
following centuries in inscriptions ^ and on coins ^ creates some difficulty .^ 
The name Apollonia, and the existence of Thracian colonists would seem to 
point to a Pergamenian foundation.* But, as Prof. Ramsay points out to me, 
an inscription published by Sterrett ( W.£. no 589), recording the erection of an 
aya\/j.[a 0eoO] Net^aro/ao? Kara KeXevcriv tov A[t09] by a priest of Zeus 
Eu[rydamenos], seems to show that Apollonia was a foundation of Seleucus 
Nikator. The Thracians, then, may have been settled in the city at a later 
time, for we have already found (Part I. no. 10) that the Attalidae 'actually 
introduced into Seleucid foundations bodies of new citizens likely to be faithful 
to themselves.' It would appear from an early inscription, no. 40 (below), 
that the two classesof colonists maintained (for some time at least) a 
separate existence in reality as well as in name, for there a bequest is made 
elf evtoyiav iv t[^] 7r6X[e]t ®paKO)v Siv \^av e]i?7 hlKai\a>'\<i. The use of the 
term koKoovoI is not easy to understand : was it adopted in a spirit of rivalry 
with the Augustan KoXcoveiat, such as Antioch, Lystra, etc.,^ in place of the 
usual term kutohcol, which was by this time synonymous with Ktafir) ? 

36. In the wall of the Kishla, 
////APICNENOCAIEYPYAAM-N//// ...Kex]api<Tfj,ivo^ Al Evpv8afir)v[^ evx^v- 

The form At occurs also in an unpublished inscription at a village 
between Apia and Aizanoi At ^povrcopTi ev^vv, and in an inscription at 
Karadilli (see Khelidonia-Diniae). 

The cult of ZEYZ EYPYAAMHNOS is mentioned in two other inscrip- 
tions of this district, (1) atBoyuk Kabadja (Sterr. W.E. 589), where we should 
restore iepeix; Aeto? 'Ev[pv8afxr)vov] and (2) in an inscription of Gendj Ali pub- 
lished from Ramsay's copy in Eev. Arch. 1888, ii. p. 223 (= Ghron. d^Or. 1883-90, 
p. 500), where a monument is erected by a priest of Z€u<;EvpvBafirjvo<i and his 
wife 'jrp(i>rav\o<t Ato? Ovpvhap,r}vov. D/jfOTauXo?, ' chief flute player,' ' first 
flute,' implies that music and dancing formed part of the religious ceremonial ; 
on this subject, see CB. ii. p. 359. Evpv-8a/u,r]vo<i is doubtless a Hellenized 
form of a native name.® We may compare Evpv-^d\i,vBo<;, a name of 
Dionysos (Hesych. s.v.), where l3aXiv8o<i is certainly the same word as Phryg. 
fiaXriv ' king,' which occurs in Thrace as a name of Dionysos (B a \ t v^ top 

* C.I.O. 3969 etc. ; Sterrett no. 517: Hist. ' Possibly Ovpvhafirjvhs is not an engraver's 
Geog. p. 172. error but a fonn really nearer the original. 

' From Antoninus Pius to Gallienus {Head, ^ Skr. bala, 'power,' balin 'powerful': cp. 

Hi3t. Num.). Tomaschek, Die alien Thraker ii. p. 41, (in 

» Cp. Wadd. on 1195; G. F. Hill, B. M. Cata- Sitzungsber. Wien. Akad. 1894), where this is 

logue of Lyeia etc. , p. civ. given as one of several explanations of the name 

* As Prof. G. Hirschfeld thought {Oott. Oel. 'Ba\(oy oder BoXkJs' which 'soil Dionysos bet 
Anz. 1888, p. 692) : so Ramsay, Hist. Geog., p. den Thraken geheissen haben EM.' Gaisford's 
400, who shows that /ivKiwv 0po«a)»' means edition of Et. Mag., however, gives neither 
\vKioDv Ka\ epflncif (as given in an insrr. ib. 172). BaXlas nor Ba\i6s, but only Bd\tv (without any 

' Cp., for example, Sterr. Jf'.E. no. 352, variant), which is clearly the same as Phryg. 

E.J. 97 etc. fia\i)v. 



t^iovvaov, ®pdK€^, Etyvi. Mag. ed. Gaisford) and in Bithynia, which is 
Thracian (Strabo 541), as an epithet of Zeus, Ati Ba\»;^ {Ath. Mitth. 1894, 
p. 373). Evpv- was taken by Fick^ as the Greek adjective, in which case Eupv- 
y9aXti/8o9 would be equivalent to Eupu-zieSeoz/, Evpv-dva^ {-dvaaaa). 

37. By the side of a field below XJlu- 

5 ////EZBEY2ANTA 5 

10 ////VEZANTAKAT^//// 10 



15 //V/ZAPETHZENEKE//////// 15 

Borlu : letters rather broadly cut. 
[17 7€pou<rta] 

7rtx]ou, t€/3€a 'Poi- 
/^»7]9 7€i'6/iei'o[i', 
7r]po? Tov Se/Sacr- 
t]oi/ 8t9 hoipehv K\al 
dy]opavofii] aavra 
KaX\ yvfiva<Tiap')(^iap 
TeXjeVai/ra Acar^ 
hi\a6riK'qv 'OXvfnrll- 
X\ov [A]a^a toO ai/€['^- 
t]oi) \[a]/Li7rpa)9 /cal 
<^t\]o8ofG)9, «al 7r[d<r' 
t)]^ dp€Tfj<i ^veK€[v. 

The last phrase xal irda^iq^ dperrif evexev is co-ordinate with the 
participles. D[emetri]os is honoured for his public spirit in undertaking 
these munera and * for his merits generally.' 

The mention of a priesthood of Rome shows that this inscription belongs 
to the early times of the Empire (see CB. nos. 199, 302, 345, and p. 365 on this 
cultus at Eumeneia and Apameia). The main function of the Oymnasiarch 
in Roman times was to superintend the distribution of oil and help to provide 
it : ' nothing could better illustrate the deterioration in moral fibre of the 
Graeco- Asiatic cities than the transformation of the director of education into 
the purveyor of oil ' {CB. ii. p. 443-4, where the office is described). 

The person honoured in this inscr. belonged to one of the most powerful 
families of Apollonia. The following genealogy can be traced from Sterr. 
no. 518 (better in Wadd. 1195a 2) and our next inscription : — 

* Die griech. goUerbeinamen in Bezzen. Bet' 
trdge xx. p. 160. 
' Wadd.'s reading 'AiroAA. 'AiroWwvlov rov 

'OKvfiirlxov rov "Aprtfion'os IB the correct one 
iu 1. 4 I read T A T I A (complete). 





Olympichos /3' 


ApoUonios = Tatia 

Alexander = Daughter (unnamed) 

ApoUonios /3' 

38. In the KaU wall, upside down : 


' A\\€\^av\hpo<i '0\vfnri')(pv 
Toii] 'OXvfiirly^ov ' AttoXXooviov 
To]v eavTov Belov Ka\X\ rrevOepov. 


'AXe^avBpo^ '0\v{p)'iri')(pv 

Tov '0\vp,'iri-)(pv rrjv ^ <Cjhv^ 

eavTOv irevdepdv. 

39. Of the following inscription two diiferent examples exist, one in the 
KaU wall ( = Sterr. 530) and another in the wall of a house at the opposite 
end of the town. In both cases the lettering is rude. 


AAr-YnOAi ii- 



HATf (Ki onyioi///// 

AAEg A- • ApoyE ///// 
4)VAHC ivC/'NeibuN erro // 
H C A M L' N M N H M H C Vac 

Avp. FiVTv^Vi '^^'' TlaTpiKtot; vloi 'AXe^dvSpov iniKXrjv 
A . YKYC TToXiTeivofiivov) <f)v\rj^ Bei/ertBv ivoirja- 
afxev fipijfirjt; (^dpcv). 

The words <^v\r] Bei/eTwi/ mean perhaps the ' Blue Faction,' not a tribe 
called by the Thrako-Illyrian name BeVerot, though the latter would be quite 
suitable. W. M. R. informs me that he copied the second stone in 1888 and 
read A P YKYC, marking the p as very doubtful. 

40. The copy which I made of the badly defaced inscription published 
by Sterrett (no. 539) helps us to complete 11. 10-20. 









El2:TeANAA^ 'I >/T or 

Ml El^^ AfEYA 
X(ANEN".:.nOA .1 



10 . . . . dv[a]0i]- 
<To[fiac S]e [e«] t[^]9 ef 
avTri<t Trpoaohov 
et<i T€ dvd\T)[-\lr]iv tov 
fji[vrf/ji]etov [/c]ai evco- 

15 ;)^tW eV t[^] 7roX[e]t 
^pciKcov wv \av ejt?/ 
St/cat[o)]9* 8ia Twy 
Sr]/jLO(Tca)v dviypa- 
yfra iv tc3 7' /cat /Lt' «at 

20 p €T€l. 

The style of this inscription is eariy. The monument was erected in the 
year piJ-y', 143, but the era is doubtful. In Studia Bihlica IV. p. 54, Prof. 
Ramsay suggests that the Galatian era may be 189 B.C., the era of ' freedom,' 
which would give 47 B.C. as the date of our inscription. L. 9, the third 
letter seemed not to be Y; AAAO is doubtful. L. 11, after the first E there 
is room for two letters. L. 16, ' The Thracians who are entitled to it,' 
implying a specified number or body ; evidently the Thracians were still 
distinct from the Lycians (see on no. 35). After HN there is room for two 
or three letters. L. 17, I have a note to the effect that the letter after 
A I K A I is apparently . 

41. In the wall of the Greek church, on two blocks : restored wrongly 
by Prof. Sterrett (Nos. 520, 521), who did not observe that they are parts of 
the same inscription. 





////////y///H M E N 


(/B'YNHl .■ 

0] Brjfio^ [irifirja-^ev 
MeXrivlrjv 'AA,e]^[a]i/- 
Spov, yvvaiKa 8e Ar}- 
jjLrjrplov TOV Mei/e\a- 
ov, dp€TT](i iveKev 
Koi <r(u0[poa]uz/r;9. 

The first block contains also Sterr. 519, the second, 522. 
inscriptions we get the stemma, 

From these 

H 2 




Demetnos = Meltine, d. of Alexander 

Meuelaos = Tati[a], d. of [Aristo ?]macho3 (no. 522). ^ 

42. In the wall of a house : 

'ApT€fiiB(opo[<;] 1 AiofiJ]Sov<: [Te]|^i/?> ihitp fiv['n]\ M': X^P^^' '^'*t^l 
'n.dvTapBpo<! I 'AprefxiScopov. 

43. A fragment in another wall : small letters. 

We can see in 1. 2 Tdiov K[\avBtov] ; in 1. 4 eT[&)i/] Trivre koI B[eKa] ; 
in 1. 7 iTT'jrel'i Be 'FQ}\jialoi]; in 1. 9 ix rS>v tJStcoi/ KaT\€<TK€vaarav\. 
I subjoin corrections to Sterr. 629 and 632. 

44. In No. 529 Prof. Sterrett's restored reading of 11. 1-9 is confirmed 
by the stone (which is complete from beginning to end), except 1. 6, where we 
should read dvaXoy^crav\Ta.^ LI. 10-14 I read iv dtraaiv evvoia, Bi\a to koX 
TOP veaviav \ Trdarjq dp€Tfj(; reXeiloraTov iaxv^^^vac \ ^ijXov. For ^ijXov MM. 
Legrand and Chamonard read 7r6\ti^ {B.G.H. 1893, p. 258.)^ 

45. The beginning of No. 532, where Prof. Sterrett has 'A[^77i/]a9 
lJ€i>Kr]<t>6pov, must be corrected. My copy, which was made by means of a 

glass, reads 

i.e. [ Trpea-^evTTjv] 

Xeyewi/Jo? T[p]4a«o[<rT^9] 0[v\- 
TTta? ^eiK7)(l>6pov. 

This confirms a suggestion made by Dr. Brandis in ITermes, 1896, p. 164. 

' 11. 3-4 in Sterrett are wrongly spaced. 

Read TaTl[av 'Kpiaro ?]/;t(ix<»'i 

yvva{iKa 8i M«]ctX<£oi». 

^ My copy has in 1. 8 TONSaNAZIA 

THTOY ; but koI is necessary, whether 

actually on the stone or not. The inscription 

lies on its side at the very top of the KaU wall, 

and I read it with a glass from the top of the 
wall opposite. 

» In B.C.ff. no. 36, 1. 6-7, read /^lowaiov 
Vlv''o[<pd]fovs, and in 1. 9 APTEMIAOZ. 
In no. 35, 1. 8, I seemed to see 

/rAIE"/ANEni, etc. 



In the last chapter of CB., vol. ii., the author gives some notes on the 
trade-route to the East, reserving a fuller discussion in view of further 
exploration. I spent the best part of a month in traversing the country 
from Apameia (Diner) to Tchai (where Phrygia Paroreios begins), and from 
Tchai to Tyriaion and the south-east comer of Phrygia. The results, at 
least for the former half of the journey, are more meagre than we could have 
wished ; but this is hardly surprising, just because this district was the great 
artery of communication between east and west,^ and was therefore exposed 
more than others to the destructive inroads of successive invaders. Especially 
is this true of the country at the head of Sultan Dagh and along its eastern 

Roads from Metropolis to Synnada. — The Campus Metropolitanus 
has been so often ransacked that little new was to be expected there. I 
copied, however, an interesting fragment of a Latin inscription which seems 
to mention a village Polynta, but meanwhile I reserve it pending a re- 
examination of another stone which would appear to bear a companion 

But there is one controverted question on which a thorough exploration 
should have something to say, I mean the line of the Roman road between 
the Campus Metropolitanus and the plain of Synnada, along which the huge 
mondlithic columns of Dokimian marble were carried to the coast.* The 
modem waggon-road from Synnada crosses the hills by way of Uzun Bunar, 
joining the Eastern Highway at the lower village {mahalla' quarter ') of Kara- 
dilli. In 1881 Prof. Ramsay decided that this must have been the line of 
the Roman road.^ In 1888, however, he discovered another road passing 
Baljik-Hissar (or Baghtche-Hissar), beside which it ' crosses a lofty ridge by 
a finely engineered path, the cuttings and curves of which can still be 
observed,' * and then wends its way over the hills to Ginik and Metropolis. 
Though he was prevented by circumstances, which the archaeological traveller 
can appreciate, from exploring this route completely, he was convinced by 
the evidence of engineering skill that this was the line of a Roman road, 
and in 1891 he discovered the Termini (No. 693) by the side of this road, 
opposite the village Yiprak (see CB. ii. p. 751 fF.). His final suggestion is 
that there were two roads, a trade-route, and a horse-road " to carry at least 
the lighter trade " {CB. II. p. 762-3). M. Radet has recently adopted the 
former view, dismissing the latter route (which he has never examined) 

* Koiv4\ ris bths rirpiitrai iiraai tojj iirl tAj Kal wXdKts, 

ivaroXiiS dioiiropovffiy c( 'Eipiirov, Strabo, * J.H.S. 1887, p. 481. 

p. 663. * Hist. Oeogr. p. 170. 

* Strabo, p. 577, niovti noy6\i6oi (ttyd\oi . . . 

102 J. G. C. ANDERSON. 

with the words ' un effroyable chaos de rampes, de gorges et de precipices ^ ' ; 
but this does not settle the question, for scientific method does not accept as 
proved a theory that refuses to take account of negative evidence. 

The maps of this district are very inade([uate, and tliough I cannot 
guarantee the absolute accuracy of the one which I give (Plate V.), I claim 
that it is nearer the truth than any of the others. Let us briefly describe 
the routes. (1) The road by Uzun Bunar naturally passes by Atli Hissar, 
and enters the hills a very short distance to the south of the village Tchoban 
Kaya, (1 hr. 20 min. from Bedesh). After a short ascent of 350 feet, it 
goes along the level summit for fully an hour, and then descends 250 feet 
into an ova. Traversing this ova, we come to undulating ground, and de- 
scending by the very slightest of gradients pass Uzun Bunar 1 hr. 35 min. 
from the point at which we entered the hills, and reach the edge of the Kara- 
dilli plain in about half an hour more.^ Here we are only from 150 to 
200 feet lower than the Synnada plain, and the whole road is so easy that 
we are hardly aware that we are crossing hills at all. At Kara-dilli we 
join the Eastern Highway (to Metropolis). 

(2) There is another road to Metropolis which diverges from the former 
below Atli Hissar and enters a long glen in the hills, appropriately called 
Uzun Dere (' the long valley '), running in a SW. direction. Fifty minutes 
after leaving Atli Hissar, we pass Alaka (which lies above us on the left) and 
travel down the dere for an hour until we come to a point at which it bends 
round to the left. Just at this point it is joined by the road which crosses the 
hills vid Baljik- Hissar.^ Starting again, we reach in twenty minutes the foot 
of a Bel (a low broad ridge with hills rising on both sides), whose summit lies 
400 feet above us : and after crossing it we make an easy descent of forty -five 
minutes to the brow of the hills looking down on Ginik, where stands the 
boundary-stone already mentioned. From here there is a long easy descent 
of an hour or so to Ginik (Euphorbion). This road also is remarkably easy 
for a hill-path : the only climbing we have to do is in getting over the Bel. 

» En Phrygie (1895), p. 123. which lies Atli-Hissar. The 

" From this point the village is perhajis about descent is rather steep towards 

an hour distant. Uzun Bunar seems to be M. the foot, which is reached (4,800) 

Radet's Fontaine (see his route-map) ; but if so, at 

it does not seem to be correctly placed, as the 10.8 ,, Cross the ravine at its extreme 

above description shows. Cf. Ramsay, CB. p. limit, and begin the ascent over 

752 n. 4. the next ridge, reaching the top 

3 Here is the itinerary from Baljik-Hissar to (5,200) at 

this point. It will show the erroneousness of 10.32 ,, Thence an easy descent, reaching 

M. Radet's description. [It will be noted that the foot (5,100) at 

this route cannot pass Alaka.] At the village 10.52 ,, Ascend again till 

the aneroid read 4, 700 ft. The road winds by 11.5 ,, (summit 5,350). Thence a winding, 

a tine curve round the- KaU and then over the undulating path, and a final 

ridge. descent to the Uzun Dere, which 

9.33 A.M. Leave the Kali. is reached at 

9.42 ,, Brow of the ridge (5,200 ft.). 11.25 ,, Level is 5,200. Here we join the 

Reach the summit (5,300) at Alaka-Ginik road. At 11.5 we 

9.49 ,, Begin to descend towards the broad were still passing fields belonging 

?-avine, opposite the mouth of to Baljik-Hissar. 


Now, though we can testify to the very clear traces of cuttings and curves 
on the Baljik-Hissar road, we are obliged to conclude that no Roman or other 
engineer making a road for heavy traffic from Synnada to Metropolis could 
choose this route : and this conclusion I am sure Prof. Ramsay would at once 
have reached had he been able to examine the whole line of the road. It 
is not conceivable that an engineer would select this line for a waggon-road in 
preference to the easy route via Atli Hissar and Alaka, so clearly marked out 
by nature, if he wished to cross the hills to Metropolis in this direction. 
What then is the explanation of this engineered road ? Baljik-Hissar is 
an old site.^ Its inhabitants are called Swvapi'Sai in the metrical inscr. 
set up by the township, SoyfiaTi koivm ySouA,?}? kuI Syjfiov, in honour of 
Demetrius, the Asiarch (Ramsay, Rev. Arch. 1888, p 220). Thynnaros is a 
local hero of Synnada, mentioned on its coins,^ and Thynnaridai is obviously 
the poetical equivalent of Synnadeis^ \ that is to say, the settlement at Baljik- 
Hissar was a village* in the territory of Synnada, which had many subject 
K&fiac. The lists of the 'Bevot TcKfjiopeioc give us the names of several of 
them, e.g. Kandroukome, Koumalettos, etc., and prove that the full designa- 
tion of their inhabitants was SvwaSeiii otKovvTe^; iv K.av8povKci)fir} , iv K.ovfj,a- 
\eTTft), etc.^ The raison d'^re of the road might, therefore, be found in the 
existence of this village and its need for a direct road to the plain of 
Metropolis on the one side and to Synnada on the other.^ This, however, 
does not seem to constitute a sufficient reason for such a carefidly made road ; 
and doubtless Prof. Ramsay has given the true explanation when he suggests 
that this was the direct horse-road from Dokimion and Synnada to Metro- 
polis, made by the Romans to carry the lighter trade. 

The line of the great Roman road, then, is limited to two possible routes, 
that by Alaka and that by Uzun Bunar. Which are we to accept ? On con- 
sideration, I think we must regard it as certain that, while there always 
existed a road vid Alaka, (used at least as a horse-road), the road by which 
the great blocks of Dokimian marble were transported took the line by Uzun 
Bunar, joining the Eastern Highway at Kara-dilli. The mutilated milestone 
discovered by MM. Radet and Ouvr^ at Atli Hissar {B.G.H. 1896, p. 115, En 

' On the rounded hill above the village, et qu'il honorait le h6ros Thynnaros comme 

roand which the road runs, a very few traces of iponyme (p. 123). [I have since come across an 

the old settlement remain. The natives call it excellent parallel in a metrical inscription of 

the ' Castle ' (KaU). The highest point of the Temenothyrai, where the inhabitants of that 

kaU is 375 ft. above the village. city are called Tr^fj-eviSai, i.e. descendants of the 

2 Drexler, Num. Zft. 1889 p. 177; Imhoof- hero Temenos (mentioned on their coins), as the 

Blumer, Griech. MUnzcn in Abhandl. d. Bayer. late Dr. Buresch rightly explains it, quoting 

Akad. d. Wiss. (I. CI.) 1890 p. 748. ©wvapiZai as a parallel (Aus Lydien, p. 164)]. 

^ So it is rightly explained by Ramsay in * The road is still used by the villagers of 

Hist. Oeog. p. 36 note. Baljik-Hissar in going to Tchul Ova {Camp. 

* It may be melissa (see below). Metropol.). At some future time the boundary- 

6 Sterr. W.E. no. 366, 20 and 376, 33, 42, stone mentioned in CB. ii. p. 752 will be dis- 

etc. See G. Hirschfeld, Qbtt. Oclehrte Anzeigen covered to give us a fixed point, but the 

1888, followed by Ramsay H. G. p. 409. We peasants will have to re-discover it first. Only 

cannot at all accept M. Radet's explanation one man seems to have seen it : he guided u.s to 

that ce dime [de Synnada^ s' appelait Thynnara tlie spot, — but it could not be found 1 

104 J. a. C. ANDERSON. 

Phrygie p. 124) supplies no decisive evidence, for it would suit either route ; 
but the Uzun Bunar route is easier than the other and only very slightly 
longer. On this view, the Roman road will pass near Ginik, beside which 
M. Radet and Prof. Ramsay agree in placing Euphorbion ; and this is probably 
all that is meant by the Peutinger Table's route, 

A Synnadc luforbio* mil. xxxvii 

■ I Euforbio Al) etiforbio 

' I Ab amca mil, xxxvi 


when we remember that in the Table the distances are reckoned from city to 
city, and the cities often lay a little apart from the direct line of road.^ In 
order, apparently, to conform to the appearance of the Table, M. Radet makes 
his road climb up among the hills on the north side of the Kara-dilli Ova 
and high above the Kiz-Kapan pass to Ginik and thence to Metropolis. Can 
any one who has seen the country between Ginik and Karadilli believe that 
there is the slightest probability that the road followed such a line ? The 
distances in the Table are, of course, quite wrong. Synnada xxiiii (or xxv) 
Euphorbion xxvii (or xxviii) Apameia would be nearer the truth. 

Khelidonia-Diniae. — On our view of the roads, there are two other 
important points, viz. Kara-dilli (lower village) and Atli Hissar. At the 
former there are considerable remains, including a carefully defaced inscr. 
of about ten lines, of which I could decipher little more than .... kuX 
'Ayava dS€X<f>fj avTov . . . . ; while in the upper village there are the two 
inscriptions published by F. Sarre, Arch.-Epig. Mitth. 1896, p. 31, Nos. 7 and 
8. No. 7 is inscribed on an altar-stone, bearing reliefs on the top and four 
sides,^ and reads Ko/37ro^|6/J09 ^iLa\6oiTi)<i 'H\t|y koI Al ev\)(riv. * To this 
site should probably be assigned also CB. II. No. 707 a and h. There seems 
little doubt that we should place here XeXtSovm, mentioned by Strabo (p. 663), 
on the trade-route between Metropolis and Holmoi, and identify with it the 
town Diniae ^ which Manlius passed on his march between Metropolis and 
Synnada (Li v. xxxviii. c. 15). After emerging from the Kiz-Kapan pass, the 
road naturally converges on Kara-dilli and then turns away towards the hills. 

SiBlDOUNDA. — The other important point is Atli Hissar. The neigh- 
bourhood of this village is the best situation for an ancient city in the whole 
district of Synnada (after Synnada itself). Lying in the plain at the point 

' The I perhaps belongs only to the word * For the form Al see no. 86 above. In Sarre 

PHRYOiA written across this route. no. 8 inrtpl is exceedingly common in such 

' Whence it follows that 'the sum of separate Byzantine inscriptions : read also k\ TK} with 

distances [is decidedly greater than the whole a twirl]. 

ength of theroad,' J.J?.,S. 1887 p. 463, where » So identified by Ramsay, Hist. Oeog. pp. 

the principle ia stated. 142 and 171 and Radet, Map in En Phrygie 

• They are much worn, but one is the head (where it is placed at Kara-dilli). 
of an ox. 



where the roads from Metropolis converge, surrounded by fertile lands and 
supplied with water, it might naturally be expected to correspond to an 
important ancient site; and we find as a fact that the remains here are more 
considerable than any other unidentified ruins anywhere near. Now there is 
an independent city belonging to the country around Synnada wliich has not 
been plausibly located, viz. Sibidounda. It struck coins at least from the 
time of M. Aurelius to that of Gordiau, and is given as a bishopric of Phrygia 
Salutaris under the metropolis Synnada by the Nutitiae, where the name 
occurs under forms which are only shght variations ^ of the one form Sibindos 
or Sibindon, as the following Table shows : — ^ 

Not. VII. 

Later editions of Not. VII 

Not cd. De 



Not. III. ■ 
and X. 

Not. IX. 

Not. VIII. Not. Baailii. 




"iiKvdiov 1 2(/3ti'8oD 





That the city belongs to the district around Synnada is clearly shown by 
the order of names in the Notitiae which is as follows : — 

4. Ipsos 

5. Prymnesso3 

6. Meros* 

7. Otroua [De Boor] 

8. Sibindon (-os) 

9. Polybotos 

10. Phyt€ia(= Beudos*) 

11. Hierapolia 

12. Eucarpia 

13. Lysios 

14. Auguatopolis 

15. Br ouzos. 

Now if this list be compared with that of Hierocles, we find that 
corresponding to Sibidounda is the corrupt name Ae^aXcKia, which is given 
after Augustopolis [KXtJpo? 'Opimjq and KXjJpo? no\iTiKrj<;, Hist. Geog. p. 178, 
CB. p. 635] and before Lysias, Synnada, Prymnessos, etc. This fact was 
pointed out by M. Radet {En Phryg. p. 115) and is accepted by Prof. Ramsay, 
who explains the corruption as due to a copyist's error, s passing into a and 
A into A. I would suggest, then, that Atli Hissar with its fine situation and 
numerous remains^ is far the most suitable site for a city of the importance of 
Sibidounda ; and this situation suits perfectly the order of Hierocles ' when 
we recognise,' to use Prof. Ramsay's words, 'that Lysias was in Oinan Ova [see 
below] and that thus the three cities were closely connected by a line of road ' 
{GB. p. 753).^ This identification would accord excellently with the Termini 

^ Excepting the corruption "XntvShov (k for &). 

' Not. Basilii [ — Not. i. in Par they- Pinder] 
and Not. Leonis edited by Gelzer {Georg. Cyprii 
Deieriptio Orb. Rom. ). De Boor's Not. in Z/t. 
f. Kirchtngeich. xii. p. 528. 

* The intrusion of Meros is probably to be 
explained by the circumstance that it was 
formerly subject to Pryninessos (see Part L 
p. 424). 

* 'Boudeia and Phyteia are perhaps other 
forms of the name Beudos Vetus ' {H.O. p. 
143). This is clearly right 

' Inscriptions are published by Ramsay, 
B.C.H. 1883, p. 299, by MM. Legrand and 
Chamonard, B.C.H. 1893, p. 289-90, and by 
MM. Radet and Ouvre, B.C.H. 1896, p. 109 f. 

* We might compare, for example, Hierocles' 
order Siuethandos, Laodikeia Eatakekaumen^, 
Tyriiuon; orHomonaJeis,Ili8tra(/7i>ra), Laranda 
{Karaman), Derbe {Ovdeliarin). A protest 
must be made againit the exaggerated stress that 
is often laid on the precise order of names in 
Hierocles (or the Notitt.) We have a good 
specimen in M. Radet's reasoning about KA^pot 


{CB. No. G93 and p. 751 ; see above), if the name Sihidounda (in any form) 
is mentioned there, but the stone is so badly worn that the central letters of 
the name will never be read with any certainty.^ Prof Ramsay himself 
suggests Baljik Hissar or Bedesh, where M. Radet determines an ancient site, 
as suitable for Sibidounda (p. 753) ; but the former we have seen to be 
a village subject to Synnada, not an independent city : whence it follows 
d fwtiori that Bedesh was a kwixt] too.''* Our localization at Atli Hissar, on 
the direct lines of communication with Pisidia, suits also the fact that the 
coins of the city show a Pisidian type, the goddess Helena between the 
Dioskouroi,^ and are similar in fabric to Pisidian coins.* It is probable that, 
if the lists of the Xenoi Tckinweioi were complete, we should find mention 
of Sibidounda, considering how largely the district round Synnada figures 

According to this identification of Sibidounda, the village MELISSA, on or 
near one of the roads from Synnada to Metropolis, where Alcibiades was killed 
and buried and where Hadrian erected a statue to his memory in Parian 
marble, must be placed at Bedesh or at Baljik Hissar. The only evidence is 
Athenaeus xiii,, c. 34, ' and we too saw the tomb of Alcibiades at Melissa, on 
our journey from Synnada to Metropolis*^; and it is obvious that these words 
do not fix the precise situation. It is probable that they were travelling by 
the direct horse-road, but even if they were taking the waggon-road, and the 
tomb of the famous Athenian were some little distance oflf it, at Baljik Hissar, 
they would certainly turn aside to visit it and then continue their journey. 
In either case they would have seen it ' on their journey from Synnada to 
Metropolis.' ^ 

'Oplvi^i, KKripoi TloKiriKTis, ^t$a\tKia, Avalas, Prof. Ramsay's copy shows. sib[inden]os 

HvaSa (Hier. 677, 3-7). Synnada is fixed. A seems quite probable. 

town ' Orin6 ' is placed at Tchukurdja in a * Which agrees with the insignificant 

mountainous district (on evidence for a criticism character of the ruins there, 

of which it is suflBcient to refer to CB. ii. pp. ^ CB. p. 755 note 1 ; Irahoof-Blumer, Monn. 

635, 687, etc.), whence it follows that K\rjp, grecq. p. 345; Reisen in Lyhkn etc. ii. p. 168 ff. 

noA«T. Me Domaine Urbain' designates the (reliefs): G. F. Hill, B.M. Catal. of Lycia, 

neighbouring plain Kutchuk Sitchanli Ova. Pamph., and Pisid. -p. \\\i. 

Le classement du Synecd^me ezige qu'on y place * For example, coins of Ariassos, Andeda, etc. 

Tum seulement KA^p. HoA., mais enco7'e I'Mchi A cursory glance at Hierocles' list of towns in 

qui lui fait mite dans le catalogue A($akiKla Pamphylia might lead the reader to think that 

[ = Sibidounda] . . qui se place tout naturelleinent Sibidounda is really to be found there under the 

d Karadirek, dans I'angle oriental du Kutchuk- form S/vSouvSa. But if he takes the trouble to 

Sitchanly-Ova. Lysias, qui vient ejisuite, tombe compare the Notitiae, he will see that 7,iviavvla 

d tlfek-keui [Effe Sultan]. The topography of corresponds to twlilmv, 'ZavliZov, or Kai-SfSou, 

Asia Minor would soon be settled, if such i.e. Andeda. 

reasoning were admitted ! ' Here is the whole passage : . . knoBaviiv 

' I examined the stone on two occasions. iKrfifvQr] iv MtXlffarj ku/xt) t^j *pvylas iiri0ov 

CIB is certain (the first letter is not so rounded ktvOds virh *apvaBd(ov. ttSofxfv 5^ Ka\ r]n(7i rh 

as the c above and may be a mis-cut for s) : os 4v MfKlaar) tov 'AKKt&tdSov nvrifxa, 4k l,vviSaiv 

is also certain. There are remains of a letter below fis Mr]Tp6Tro\iv &<piKvovnevor iv ^ kut' Itoj 

B (in the line above), which on the first exam- eifrai Bovs, SiaKtAfuffanffov toCto rod irivra 

ination I took for a badly-fonned i, but the kpiarov'h^piavov BaaiKtui- %s koI hviaT-qafv iirX 

second time it seemed to be the tail of an s. t# ^i/^jyuon napiov \l0ov (lK6ya rhu 'A\Kt$iiSv*'- 

There is room for other four letters, for the « May there not be something more than a 

sjiace bet wren Lin and riioc is greater than mere coincidence in the occurrence of the name 


Lysias. — Everything that is known about Lysias, which was probably a 
Seleucid ^ foundation, will be found in CB. ii. p. 754 f."^ The name is there 
assigned to the ' city whose ruins are seen on a mound between Oinan and 
Aresli.' In support of this localisation Prof. Rani.say quotes a passage from 
the Acta S. Abcrcii, where it is said that one day while engaged in his 
apostolic labours in the Pentapolis, Avircius found himself athirst and without 
water on a lofty mountain, oVep eVrtj/ uvrcKpu Trj<i TroXeco? Avaia<i ; where- 
upon, K\iva<i ra •yovara irpoarjv^uTO Koi dve^Xvaep irr^iyr) Kadapov vafiarot. 
Kul 7rdvT€<i oi BcyjrcbvTe^ i^ avrov eKopia-drjaav. 6 Be totto? e'^ cKeiPov rov 
Kaipov TovvKKiaia eireKXrjdr). The sacred fountain and ' Place of Knee- 
bending,' 3 as Prof Ramsay says, arc evidently on the mountains bounding the 
Pentapolis on the east, and south of the Sandy kly hills, which lie over against 
Synnada (p. 755). It seemed well worth while to make a search for this 
fountain. We crossed the mountains twice, from Karghyn to Baljik-Hissar 
and from Yiprak to Karghyn, without discovering anything. But at Karghyn 
and afterwards at several other villages we heard of a fine bunar in the 
mountains S.E. of the village Mingile (Mingine, wrongly, in Kiepert's map), 
called Giaour Oluk, i.e. ' Giaour spout.' This unique name promised well, 
and we set out on a journey of three hours up Gumular Dagh to examine it. 
Arrived at the spot, we found a copious spring surrounded by the ruins of 
what had been a square building, as we judged from the foundations, for 
hardly any of the numerous squared blocks remain in their original 
position, though a few have been roughly thrown together again to enable 
the water to flow through the stone spout into two rectangular troughs. It 
is probable that both the spout (which has given its name to the bunar and 
to the yaila around) and the troughs are ancient work, and that we have 
here a genuine old fountain, — a fact recognised by the traditional Turkish 
name Giaour Oluk.* Its position is marked approximately in the map 

Baljik-Hissar ' Honey Castle ' at a site which, ^ The name Lysias is connected with Seleucid 

as we have seen, may be M<X«r«ro ? The sugges- history : Lysias, a general of Seleucus Nikator 

tion (which occurred to me independently) was in 216 B.C. (Polyaen. iv. 9, 5), may have been 

made in Hist. Ocog. p. 36 n. I do not, how- the founder of the city, 

ever, mean that the Greek name was translated * Cp. also J.H.S. 1887, p. 497. 

by the Turks. M. Radet asserts {En Phryg. ^ See CB. ch. xvii. p. 714 ; Church in R.E. 

123) that rien n'est plus friqiient, dans Vonom- p. 436 n., 'At the source of a stream among 

asiiqiie de VAnatolie, que ce remplacement du the mountains between Synnada and Hieropolis 

terme grec ancien par un mot turc offrant une was a place called Gonyklisia — i.e. where the 

signification identique (La Lydie etc. p. 36 n. early rite 7ov£{t&))' »cX(<rij was held. This remote 

2) : but he shows no proof. There is some place was clearly a secret meeting-place ; and 

doubt as to whether the real name is Baljik or after the meetings had ceased, and the archaic 

Baghtche ('garden ') Hissar [or Assar]. From term was no longer understood, a foolish legend 

the rude pronunciation of the peasants which grew up to explain the name ; see Expositor 

runs the words together, it is impossible to 1889, p. 262.' 

decide. If you ask them which is the correct ■• The name itself is sufficient proof, for it 

form, they answer ' The two are one ' ! One of states in so many words that this is a ' uon- 

the Hodjas assured me that the former was Turkish fountain': compare, for example, 

more correct, but they do notseem really to know. Giaour Oren, the name of the ruin's of Trajano- 

W. M. R. was corrected for saying Baljik and polis. The Turkish number |(AA which is 

fissured that Baghtche was the real name. scratched on one of the fallen blocks is a mere 

108 J. G. C. ANDERSON. 

(Plate v.). Now if it be remembered that the whole apace between the 
Pentapolis and Oinan Ova is occupied by Gumular Dagh and its spurs, it is 
clear that this situation corresponds very well with the description ' over 
against Lysias.' I have not the slightest doubt, therefore, that Giaour Oluk 
represents VovvKXiaia, and that we have found an important confirmation 
of the site of Lysias in Oinan Ova. 

One or two inscriptions of Lysias may be added. 

46. At Oinan — 

I //// MIONKAICAPOCAOYAI //// | "AJ^toi/ Kaiaapo^ hov\[r)- 

Relief Relief 

TOHPWONAIAAOYMENOJ to r^ptpov AiaSovfiivtp 

-COIAICOANAPIKAISAPOCAOY vac. t]^ i8i<p dvSpl Kat<rapo<i Bov- 
/IO\Un''///l///hr///ln-//ll \^] . . . 


47. Ibid : carved on a rectangular block. 

+ "ttr^p ev'yrjf; rov ')(aiplov. 

48. Ihid. Fragment of a sepulchral dele : on pediment, relief of eagle 
standing with outstretched wings. 

. kavrm Koi rfj crvfi^i(p fivrjfie %a/Jti/ %(oaTpaTo<i. 

At Karadja Oren in the S. corner of Oinan Ova there is a fragment of a 
similar stele yfii\i...\jir}\Tp\ K^aX] iraTpL . . 

KiNNABORiON. — After traversing Oinan Ova, the Eastern- Highway 
crosses a ridge which bounds the ova on the north and enters the great plain 
called Karamyk Ova, the lower part of which has been transformed by cen- 
turies of neglect into one enormous marsh, extending from below Geneli well 
on to Karamyk. In this large plain, which runs right up to the edge of 
Phrygia Paroreios, there is room for several cities ; but only two can be cer- 
tainly assigned to it, Kinnaborion and Holmoi. Kinnaborion is discussed in 
J. H. S., 1887, p. 495 (No. Ix). It is first mentioned in the Tekmorian Lists 
of the third century ^ : by 451 a.d. it had been raised to a bishopric, a rank 

recent graffito: I saw it again on a tcheshme at and 46 { = J.H.S. 1883, p. 23 ff.) J Kivvd^optis, 

Utch Eyuk, in the valley of the Tembrogios. no. 374, 2 and 15, no. 378, 5 and 9 ; Kivva- 

It means 1871 a.d. ffop^ffvot, no. 366, 32. 
> Kii'i'a8of)idrr}!, Stcrrett fr.E. no.^ 36G, 33 


which it holds also in the earlier Notitiac (dating from ca 700), where it is 
placed under Synnada.^ In accordance with these indications, Prof". Ramsay 
rightly assigned Kinnaborion to the lower side of Karamyk Ova {Cli. ii. 
p. 748 and Map, J. H. S. l.c.y The exact site lies in front of the village 
Armudli, where there is a low naound which clearly conceals ancient ruins, 
and beside it a TiXrbe ^ built of old blocks. Neither here nor in the village 
did we find any inscriptions, but a mile or two further on, where the marsh 
runs up to the mountain side, we discovered a Latin inscription cut on the 
face of the rocks. As ill-luck will have it, it is now almost entirely obliterated, 
but it seems to contain the name C. Carist[anius Fronto], and in the mean- 
time I reserve it, pending a re-examination. 

The omission of Kinnaborion in Hierocles* list is to be explained by the 
fact that it was for a time conjoined in one bishopric with Lysias (6 Avaidho<; 
rjTOi K.ivva^opiov). 

The name Kt,vpa^6piop is connected by Kretschmer* with the Lycian 
personal name p^ntabora, KcvBd^vpif; (Beisen in Lyk. i. 82) : for the assimi- 
lation he compares Tpefievparayv and Tpe^evSai, and we may add 'Opoav- 
v€v<i, the ethnic of 'OpoapSa in B.G.H. iv. (1880), p. 401 (= Loewy, Inschr. 
griech. Bild., no. 305) and in an inscription of 208/206 B.C. from Egypt {Class. 
Rev. 1898, p. 276 fif.). This seems probable : a similar relation between place 
name and personal name is seen in Kihpajxo^ — }Lthpap,ova^ (No. 66), 
Tottaion — Tottes, etc. 

HOLMOI. — The villages of Karamyk and Ak-kharim both contain a 
certain number of remains, but they have probably been all carried. An 
ancient site, however, undoubtedly existed beside Karadja-Oren (' Blackish 
Ruins '), the most important of the villages in Karamyk Ova at the present 
time. At a short distance east of the village the natives still point out an 
old site {eski orcn), part of which is occupied by a deserted cemetery, 
while the village itself is full of remains of all kinds built into the mosque, 
Tiirbe,^ fountains, and walls. Inscriptions unfortunately are almost non- 
existent; the demand for good building stone has doubtless caused the 
destruction of many within comparatively recent times.^ Only two fragments 
were found : one has been published in Heberdey and Wilhelm's Eeiscn in 
Kilikien p. 163 (No. 272) ; the other is possibly a fragment of an honorary 

49. ////iCKAlCCO//// KriaTri\i kov o-a)[T^p t^? TroXeox? ? 

^ rhv Krjua$oplov Not. vii. 170, ix. 353 ; <5 307. He wrongly calls it Ktwafiopa, not 

Kivvafiwplov viii. 443; d Kiva^copiov Not. Basil. observing the occurrence of the city name in the 

389 (ed. Gelzer) ; 6 K.ivv($oplov Not. De Boor. Notitiae. 

"^ A site ' perhaps near Geneli ' is proposed in ' Cp. Kinnaborion (above). 

J.n.S., but the few remains that the village « The basement of the fine mosque-minaret 

contains have probably been carried. is composed entirely of old blocks re-faced : a 

' For the significance of this fact cp. above, small ' door-stone ' may be seen built in near the 

Part I, Vol. xvii. p. 400. top. 

* Einleitung in d. Gesch d. Oriech. Spr. p. 

no J. G. C. ANDERSON. 

It seems clear that tliis site is to be identified with Holmoi mentioned 
by Strabo (p. 663) on the eastern trade-route eVt rr^i/ ap^rjv t?)? Uapaypeiov. 
The name may be a descriptive epithet, like Trapezopolis, and not the 
Grecized form of a native name, and o\fio<; would indeed be a neat description 
of the site of Karadja-Oren, as one sees it coming from Tchai. In CB. ii. 
p. 748—9, Bazar-Agatch, two miles N.E. from Karadja-Oren, is taken as the 
site of Holmoi, but the remains there are inconsiderable — some squared 
blocks, Byzantine pillars, and a few fragments of ornamental work — and I 
came to the conclusion that an ancient site could hardly be placed there. 


For the definition of Phrygia Paroreios, a district so clearly marked off 
by nature, see Hist. Geoy. pp. 139-140. It may be roughly described as a 
long plain running N.W. to S.E. between the parallel ranges of Sultan Dagh 
and Emir Dagh, as far as Ilghin (Tyriaion), where it is bounded by the hilly 
country which stretches between these two ranges (see the Map, PI. IV.). 

Iulia-Ipsos.— See J.H.S. 1887, p. 490 (cp. Hist. Ocog. p. 434). Amongst 
the critics Tchai and Ishakli (Sakli) dispute the claim to the heritage of 
Julia-Ipsos. The actual site will probably never be found, for most of the 
remains have been used up by the Seljuks for the fine buildings whose ruins 
are still to be seen at Tchai "^ and more especially at Ishakli. But it seems to 
me that the probability is all in favour of a site quite near to the latter.^ Both 
Tchai and Ishakli are market towns, but Ishakli is the more important of the 
two, and it is the governmental centre (a mudurlik). Tchai appears to have 
been selected as a site in post-Roman times on account of the copious supply 
of good water which comes down from the mountain beside the village, but is 
not to be found anywhere else in the neighbourhood.* Moreover, it is a most 
striking and important circumstance that the modern governmental arrange- 
ments in this district repeat the ancient facts. Not only is the boundary 
between the vilayets of Broussa and Konia the same as the ancient boundary 
between Salutaris and Pisidia (see Ramsay, J. H. S. I.e. ; cp. Cuinet, Turquie 
d'Asie), but the centres of government are now Ak Sheher (corresponding to 
Philomelion), Bulaw6din (corresponding to Polybotos) and Ishakli (likewise 
in all probability corresponding to Julia-Ipsos).^ Lastly, the most important 
remains, both ancient ° and Seljuk, are at Ishakli. A fragmentary inscr. has 

■ Some inscr. from the north side of Paroreios flowing down from the Sultan Dagh nearer than 

(along Emir Dagh) are published by Hogarth, Deresinek, a village lying up a glen far off the 

J.H.S. 1890, p. 158 ff. road. 

' The site at Karadja-bren, however, was " Synnada, PrymnpRsos (Kara Hissar,) and 

probably the chief quarry for the buildings at others might be added. 

Tchai. 6 Amongst these is a large building of Byzan- 

' The Pent. Tab. gives no help, as one of its tine work. At Tchai, apart from the Seljuk 

two numbers is wrong. ruins, there are only some Byzantine blocks in 

* The water in the plain is not good for fountains, etc. 
drinking and there is no other copious stream 



been published by Mr. Hogarth J. H. S. 1890, p. 1(11 (No. 7). I have two to 
add, the first copied at Ishakli, the second at Yaka Sinek, a village about 
three miles to the North-West. These arguments seem to show distinctly 
that the importance of Ishakli at the present day is not a new fact, but only 
a continuation of the conditions which existed in Graeco-Roman and Seljuk 

50. Ishakli : a square slab with large cross carved across the field, and 
on the upper margin, 

I ////AnOAct)ANoVnAnA//////// I 'A7ro\{o)<l>dvov{s) UuTra . . . 

51. Yaka Sinek : doorstone with inscr. on either side. 

"Vac . 
i/Iin r O C £ N e A A E K E I If (Hi I 
IllllJ/lin, il C EN E Y T Y X I A I/////// 
////////// CE'BAINAf'E ////////////// 
/-////////. E M 1 P/i n I K P f^fllKKli 
li/lflUn ACPNAKfMXl qiiKiim 
(((I I ill o N A f e y K loHiff/m/f//, 

/////////SOC ECTiwAN///////////// 
fi/f iiiiiiO C N £ A\£ C I ^11 1 HI Kill' 
/////////Ai 4 BA £ n e //////////////i; 


A P CO T OJ ( A / CO A Ae Af (it/^ 
AANHA\e I0N6TT0IH(#/'/ 



KoiX 'AttttS? Koi TaT6i[9] 'AXfe^ai/- 

fivrj/jbelov i7roir](r[av 
fiv^fir)^ '^dpLv ■)^alpe ? 

The metrical part is hardly worth the trouble of an attempt at 

L. 1. 'E,vri]6ri<i Tt<f dvr)p * A\i[^av]Spof! iv0dSe Kel[Tai. 

^]rj(T€v (ev) ei/Tv^^ia [ac^] e^a Xva y' elXSero Bvfio^'l (A. Souter). 
cCre 8]e fioipa iriKph [^0)^9 viv d<f)'qp]'Tracr€v aL(f)vco^ 
7roT/u,]ov a(f)€VKTOV icrriv dvd[yKrj 

Philomelion.^ — While passing through Ak Sheher, I made a copy of an 
epigram published in C.I.G. 3982 from a very fragmentary copy of Hamilton's 

^ Inscr. of Philom. and vicinity, C.I.O. 3982 
ff.; Le Bas-Wadd. 1704-5 ; Sterrett E.J. No. 

166 and perhaps 168-4 ; Ramsay in Kuhn's Z/t. 
f. vergl. Sprachf. viii. p. 391 (Phrygian) ; 

112 J. G. C. ANDERSON. 

and repeated in an improved form by Kaibel {Epigr. Graeca e lapid. conlecta, 
No. 248), who noticed that it is modelled on Anthol. Pal. vii. 164. The 
stone is much worn and the letters are very faint. If I had nut recognized 
from 1. that the inscr. was already published, a closer examination might 
possibly have furnished a complete restoration of 11. 9-10. 

52. Xtdakot 'FiXaTTjt rrjt 
kavTov yvvaiKC (f)t\oa- 
TOp'yia<i Koi fji,vi]fir]<; 

alcoviov ■^(^dpiv. 
5 A. ^pd^e, yvpac, yeverjv ovofia ydova, ttw? he. davovaa 
Tj\6e^ BecXaia 8va-yafM0<; ei? ^Kihav, 
OTTTTCo'i ol TrapdyovT€<i dvayvaxiiaiv oSecrai, 
Tr)v <TT}[v \v]Trpo[Td]TT}v hvafiopov rjXiKirjv. 
B. ISilfjLt fi€v [ovK dyevr]<i ?], yeverj 8e fJi'0{v) icrTC Svdreipa (-a? ?) 
10 ovvo\jia fioL 8' 'EiXdTTjv ol ]<^t\otOe^ei/TO Tpo<f>[r]€]<;. 

A. '^ijfia 8e [rt? ro\8' [e;j^&)o-e]i^ ; B. ['E/a]o9 Tr6aL<i 6 irplu aOiKra 

rjfieTipr}<i \vaa(T<Ca^'^ dfifiara 7rapd€virj<;. 
wXecre S' ov to[k]€t6<; fie \vy[p]6<i, Molpac Se poTrfj fiot 
eh voaov eh 7r€vdr)<Ci^ koL fiopov rjVTiaa-av. 
15 B. 'H KoX aTrat? ; A. Ov, ^elve' \e\o{i)'Tra yap ev veorrjrc 
T[/3t]<T<rou9 dpriyevel<; TratSa? ev 6p<^avir]. 

B. EZev ev oX^ia-rrji TroXcrjc rpi^c. A. Kal <t6v, oSeira, 

tvScov evdvvoL iravra Tu^i; jStoTop. — 
"OffTt? ifxev aTtjWau 0a\^ec \i9ov ovk dSiKrjOeh, 
ovTo^ rav avrav fiolpav ifiol Xa'^eraX^i'^. 

LI. 8-10. 

E I Ml M £ N/;5////////////////r £ ,VJ gH^EMQElTI By AT £ / PA 

L. 13. P0nHM0l:P0niM0l Hamilton.— The epigraphic copy of 1. 10 
seems to show that this line was made a hexameter : such irregularities 
sometimes occur in these epigrams. On \vyp6<i, see A. Souter in Class. Eev., 
1896, p. 420, and 1897, p. 31. The composer retains the Ionic dialect of his 
model, and adds a little Doric in the imprecatory formula ! 

53. Geurness keui : ornamented stele with standing bird in the tri- 
angular pediment. 

Mei/€\ao9 Kal 'A[X,]e^ai/8/309 Kal M/ .... 
Ylairla KaXXtirTTOV tw eavTa)^[v varpl 
fivi]fir]ii Ixdpiv. 

Heberdey-Wilhelm, Iteisen p. 163, no. 271. In editors. The inscr. was apparently not corn- 
er. I.G. 3984 { = Wadd, 1704) Hamilton's copy of pleted. 
11. 7-11 is correct and is wrongly altered by the 


53 his. Gedil kcui : a fragment of a Christian inscription. 


KrrM/m i pa 

ce leCTAI AY 

[. . . €1 Ti<: . .] 
• • x]^*P^ 7r[poaoi]- 
<T€i ecnai av{T(fi) 7r[/3o]9 
TOP deov. 

On the contraction av cp. no. 88 below. 

GisZA. — In the Tekmorian Lists we find the ethnic Vi.(T}^7)v6<i or Til^r)v6<i 
(Sterr. W.E. 366, 19 and 75), giving a village name Gisza or Giza, which is 
clearly the Carian r^iaaa ' stone.' ^ The following inscription, which lies in a 
street of Ak Sheher, proves that it was a village subject to Philomelion. 




//r-NM A N H TTATT A ToKAeK^^/ 


^1 A/ o) K Ai >a6r-^^ A ' " """"■'" 



. . C^ovhr)^ yievdvhpov 

Tecr^eavr) ^axra eavrfj 
Kol '\]vfidvr] Yiaira rov Mev- 
dv'\hpov Tea-^eavo) dveyjnw 
ISitp Kol ^evdv\hp((> 
^rf\vo(^C\.ov l^KaX 
^K\a-K\r)'mdhrj . . . 

55, AzARA or EzARA, another subject Kmfir], mentioned in the same lists 
(Sterr. 382, 5 and 366, 28) retains its name as Azari keui {Hist. Geogr. 
p. 411, see Map). I copied there the following rudely engraved inscription. 

H Bon €Te^| t 
A A^n ApTi^o)" V 
T6kAUCC( \A0 

a/AxOoivAti Ac a 

A/ IKA/Vy'eiKAlOinA 

a)t^t6 Aecc An/oc 

£I//Ty K80N6CT 

1 Steph. Byz. s.v. Monogissa. Cp. Eist. Oeogr. p. 412. 

114 J. G. C. ANDERSON. 

. . . ft)?] eirpeirev /iep67re[cr]o-fv 
Kal eirl tv^^ov erev^e SufxaprL ao[p6]v re /cdaea-a-i, 
AaoBi'/CT) T^9 KvBo<i dva yOova iraaav CKuvei 
Kal 01 irdvr ireXeaaav oaa OvrjTola-iv eoLKe{v) 
rvfi^ov T €a-Ti][a-eu 

The T after rvfju^ov has been accidentally omitted in the epigraphic text 
(the Y M being too widely spaced). 

Before we discuss the district south-east of Philomelion, a few words 
must be said about its geographical character. As the traveller leaves 
Philomelion, his eye wanders over what seems to be (and, roughly speaking, 
.is) an enormous plain stretching in front of him for many weary miles. But 
this great expanse of country is not one dead level. The plain proper extends 
only a little beyond the Ilan Yusuf Tchai,^ the river which rises behind the 
village Kara-Agha and falls into the Ak Sheher Lake : and on the side of 
Sultan Dagh it is broken by a succession of low mountain-spurs, between 
which numerous rivulets run down to join the main stream. Beyond this 
river the ground slop6s gently up to an undulating plateau, diversified by low 
sand hills above Arkut Khan and extending as far as Ilghin and the hilly 
country which bounds the Balki Deressi on the East. This plateau is drained 
by the Balki stream and the river that rises at Doghan Assar and flows past 
Arkut Khan into the Ilghin Lake.'^ From Philomelion to Iconium two roads 
are available. One takes the route by Arkut ^Khan and Ilghin, coinciding 
with the Eastern Highway as far as Laodiceia Katakekaumene, where it turns 
southwards and crosses the mountains to Iconium.^ This is the line of the 
modern waggon-road. The other crosses the plain in a south-easterly 
direction to Balki-keui, whence it turns southwards to Tchigil and then 
eastwards over the mountains to Konia. This road passes Hadrianopolis and 
Kaballa (below). 

Pisa and Selinda. — Between Philomelion and Kara Agha we discovered 
two new sites. (1) The first of these is Pisa, which retains its name to the 
present day. It was situated beside the village Bissa in the plain under the 
shadow of the mountain, less than half-an-hour (about a mile and a quarter) 
from Aghait, the Byzantine Gaita (Cinn. p. 42, infra). The village contains 
numerous remains, but many of the marbles have been destroyed to build a 
new mosque : the process of destruction was going on when we visited the 
village. This town has to be distinguished from another Pissa, likewise 
retaining its name, situated on the hills on the north side of the valley of 

1 So called, at least in its lower course, from of Ilghin. We crossed the latter about three- 

a Circassian village on its banks. quarters of an hour after leaving Ilghin. 

'^ I enquired particularly at different places There is no stream from Doghan Assar joining 

about tlie course of these streams and all accounts the Ilan Yusuf Tchai at Kotchash (so far as I 

agreed in saying that the Doghan Assar stream heard or could see). 

Hows past Arkut Khan into Ilghin Lake, while ^ There is also a hill-path from Kuuderaz, 

the Ayaslar-Urus-Yendin (miscalled Kendil) described by Hogarth, ^c. p. 153 f. 
stream falls into Balki Deressi a little to S. 



Apollonia Pisidiac, and mentioned in the wars of Alexius ComncnuR (Nicetas, 
p. 540, see CB. i. p. 18G-7). The ancient name of our town is attested by 
the following inscription, recording the presentation of a statue of Sept. 






^W/////////0l\ E C TT OTI-Ncj)! A AIOCMA 


Nniad ^HMnnAP^AYTOY 


AvTOKpuTopa Kaicrapa Kovkio- 
V ^eTTTifiiov %€Ovr)pov Tlep- 
TivaKa 'S.e/SacrTov, 7179 Kal 6a- 
X[a<7cr77]9 /cat 7rdar}<; tt)? oIkov- 
/U.e[i/T;<j] SeairOTTjv, ^iXaio<i Ma- 
plwvot ISiVfiipT) r<a Yleiaea- 
v(!)\y\ 8i]fiq) Trap' eavrov 

Evfiivr) in 1. 6 for Evfiivrji^) : cp. 'Epfioyevrj, correction to Sterrett 
No, 169 (below) ; it is less likely to be a corrupt form of the genitive, and it can 
hardly be the adjective, agreeing with Xeovijpov. The form of the ethnic 
Ileccreavol supplies a parallel to {T)oT€ava)v proposed in No. 20, 1. 7. 

(2) Selinda has left its name to the modern village Selind, which lies 
further along Sultan Dagh to the south-east. In the cemetery of the village 
there are some old stones (Phrygian ' doorstones ' and other sepulchral 
slabs bearing traces of inscriptions, architrave fragments, pillars, etc.). 
Judging from the reports of the natives as to the provenance of the two 
inscriptions given below, we shall place the ancient site a short distance up 
the hill side behind the village of Ellesler,^ which lies on the lower slopes of 
the mountain, about a mile SSW. of Selind. Here therefore we have another 
example of the common rule that the modern site is generally some little 
distance from the ancient one. The site is fixed by the following inscription. 

57. Lying by a house beside the mezarlik of EUesler. 


hrjv '\ep<o- 
vo<; 6 "Zei- 


' They were turned up by a villager while ploughing his field behind the village : the 
place was shown us by this man's brother. 

I 2 

116 J. G. C. ANDERSON. 

The township is also mentioned in an inscription now at the village 
Eregiz or Regiz (pronounced Ereiz or Reiz)^ in the plain below Bissa 
( = Sterrett, Epigr. Jmtrn. No. 163), where a tombstone is erected by a 
husband to his wife, the daughter of Menemachos, son of Charidemos of 
Selinda.2 It is possible that this village Seilinda or Selinda is the Silindo- 
kome mentioned in the Acta S. Theodori Syheotae, see Hist. Geog. p. 246.^ 
But more probably two Phrygian villages bore this name, one in the Paroreios, 
another in the territory of Juliopolis (compare Silandos in the Katake- 
kaumene). Both Pisa and Selinda are called merely hrjfia in the inscrip- 
tions. It is probable that both were dependent townships, subject to 
Hadrianopolis, in the same way as a considerable town like Orkistos was 
subject to Nakoleia until A.D. 331, or Takina, kc, to Apameia {GB. p. 29G). 

58. Ihid. A small sepulchral stcU : 

AYPHAfUfiMlXne Ayp,;\/a 'A/.^m Me- 

tOANAPOYONHDnW vavhpov Oi/r;o-i/i[ft) 


MHCXAPIN fit}^ X"P''^' 

The stones have all been carried away from the site, but the village of 
EUesler contains nothing : it is probably of more recent foundation than the 
other which retains the old name. 

Thymbrion — Hadrianopolis Sebaste. — These cities are discussed by 
Prof. Ramsay in J.ff.S. 1887, p. 491, and Hist. Geog. p. 140. According to 
Xenophou's Itinerary, Thymbrion was situated midway between Kaystrou 
Pedion (probably Ipsos), and Tyriaion (probably near Kolitolu Yaila in 
Xenophon's time),* 10 parasangs or nearly 35 statute miles from either town.^ 
This points to a situation in the neighbourhood of Kotchash. The Fount of 
Midas (Yassaghan or Yassian Bunar, midway between Ishakli and Philomelion) 
was apparently in the territory of Thymbrion, and Prof. Ramsay is clearly right 
in thinking that Thymbrion was the great city of the plain until the foundation 
of Philomelion by the Pergamenian (or Seleucid) kings. The last mention of 
Thymbrion occurs in Pliny N.H., v. 95, where the Tymbriani are in the conventus 
of Philomelion. We are therefore led to infer that 'the city wasrefounded by 
Hadrian under the name Hadrianopolis' {J.H.S., I.e.). Hadrianopolis, which 
comes between Tjn'iaion and Philomelion in Hierocles' list, is mentioned in 

* Anatolian pronunciation tends to convert g in his Myrjixtia 'AyioKoyiKi. 

between two vowels into a y sound and finally * J.H.S. I.e., Ath. Mitth. 1889, p. 180-1. 

to let it drop altogether : cp. Tchigil, now pro- Xen. Anab. i. 2 : 'Evrfvdtv (from Kaystrou 

nounced Tchiyil (below). Regiz is a diflferent Tedion) i^fXavvd (rraOfiovi Svo, irapa<rayyas 

village from Egrigioz, which is quite close to Sina, tls B^n&piov, ir6\n> o\Kovnivy)v. 'EvravOa 

Ak Sheher. ^v irapk r^v bZhv Kpi]vrj r] WlSov KaXovfifvri rov 

- The correct reading is C€AINA€UJC. *pvyuv fiatriKtus 4<p fj \iytrai M(5oy rhv Tiirvpov 

* The correction Silindo-kome, there sug- S»?p«C(rai oXv(f Ktpdaas avri)v. 'ZvrevBtv 
gested in a footnote in place of Silindiconense, i^thavvd aradnols hvo, rapavxyyas StKu, tls 
ia proved to be right by comparison of the Tvpidfwv it6\iv oiKovniyriv, 

Greek original, published by Joannes Theophilj 


an inscription now at Kara Agha, (Sterr. E.J. 160), which, liowover, does not fix 
the site; for it is obvious that the inscriptions of the village, which are all 
built into the mosque and engraved on the same rough reddish sandstone 
blocks,' have been carried, and there are no other remains in the village. 
A passage in Cinnamus, p. 42, describing the operations of Manuel I. against 
the Turks in 1145, gives us some help. The Turks after a defeat at Philo- 
melion retired to a place called in Turkish 'Av8pa')(/idv, whereupon Manuel 
started in pursuit, iroXiv re ' ABpcavoviroXiv vTrep^a<i (hiafialvei, yap Ka\ e? 
avrrjv AvKaovlap to ovofia tovto) ev rivi x^PV ratTa ovofia rrjv irapefifioXrjv 
iiroiija-aTO, i.e. crossing into the territory of Hadrianopolis (which includes 
the whole southern part of Paroreios), he encamped at Gaita, which is still 
called Agha-it.^ An examination of all the villages in this plain led me to 
the conclusion that Hadrianopolis should be placed beside Kotchash, where 
there are numerous remains built pell-mell into the Tilrbe-mosque or lying 
beside it (see Sterrett, U.J. Nos. 165-173, and below No. 59). In that case 
the river Karmeios mentioned on its coins (Ramsay in A then. Mitth., 1883, 
p. 76) will be the Ilan Yusuf Tchai. 

The importance of the ancient city has passed in some degree to the 
mitdurlik Doghan Assar, situated on the slope of Sultan Dagh (like the 
majority of the villages in this district^, and a site at or near the village has 
been suggested for Hadrianopolis (J.H.S. and Hist. Geog. II. cc.) ; but Doghan 
Assar lies away in a comer, off the line of the direct road to Iconium, and 
does not suit the conditions nearly so well as Kotchash. It was probably a 
K(i}iir) subject to Hadrianopolis, like Gaita, Pisa, and Selinda. 

To Hadrianopolis belong the inscriptions of Kotchash (Sterrett Nos. 165- 
173) and Kara Agha (156-161), Tchetme (162), and Doghan Assar (174 and 
Sarre, A.E. Mitth. xix. p. 37). Some improvements on the published copies 
may be given here : 

No. 156, 11. 3 and 5 are corrected in Hist. Geog. p. 178 

No. 157, 1. 4 read [^]wi/ : space for only one letter. 

No. 158, I. 3 BACIAICH is clear. 

No. 159, 1. 1 ABACKAk//^(., i.e. 'A^daKavrof;. 

No. 160, 1. 4 restore t[^ tota] : the H is under A and the C under N- 

No. 162, 1. 1 read t^9 M.avlho^, and Tet/ioXaw (as required). 

No. 166 apparently reads AvprjXio^ "^ovalov], Av^dvoav, 0oa)[i/]^ Tarct 
[fivYitir}^ Xa-ptv. 

No. 168, I. 4 [n]a7rtW, and in 5 // /El CO, i-e. [e]ei<p (with angular ^), 
since it cannot be [uJetciJ. 

No. 169. The stone is rough both above and below the engraved part^ 
which shows that nothing more was inscribed. 'Epfioyivr) is the nomin, case 
like EvfievT) in No. 56 (above). 

^ The lettering is poor and late. ' There is hardly space for more than one 

^ ' Agha-iit, Herrenpassage ' Tomaschek p. letter. The last three words are under the 

108. Gaita is also the name of a village near relief. 

Nicaea, E.G. p. 201. 

118 J. a. C. ANDERSON. 

No. 170, 1. 1 AMn, which gives 'A/iTre'pwf- 

No. 174. My copy begins uC/Z/Z/aM (^ aivi) and continues v 

Kuov/jL\/jLavec 8ok€ [apparently] [aTT]ca8 etrov. 

59. Kotchash, in the mosque. 

AYP€YCTA0IC Ai/p. EvaTaSc^ 

IMCNOC "Ifievo^ 

AYP 3KYPeiAMeN€KAeOJC Avp. Kvpeia MeveKXeco^ 

THCAYTOYCYNBKx) rfj kavrov avv^ito 

AANHMHCeNCKeN tivrip,rj<i eveKev. 

For the form Eva-rddiq ( = Kv<TTddco<i) cp. HarpUfi and 'Hpa/cXt? (No. 91) 
and CB. ii. No. 264, where reference is made to J. H. Wright, Harvard Class. 
Shed. 1895, p. 59 f. 

"I/iav (gen. "I/Aeyo?), with prothetic l like 'la-KVfivo^ for Xuvfivo^,^ etc., is 
one of a group of names (so common in the Pisidian inscriptions) derived 
from the name of the god Manes-Men. The frequent occurrence of this group 
and of other native names in Paroreios and the vicinity is a significant indica- 
tion of the vitality of the native civilisation, 'l/iav occurs also in No. 63 
(cp. 54), Hogarth I.e. No. 19; MAp7]<i in 65, Hogarth 9, 20, 21; Mavt? 
Sterrett 162 (see above); Mavia, Hogarth 17 and 20; Mai/ocrd<? Hog. 16. 
Similar names are 'Aa-ia in 73, 'Aoreu? in 83 (Chr.). 

60. Yendin keui. 

In another fragment occurs the name Koi/wv (see No. 85). 

61. Doghan Assar, a fragment in the mosque. 

TYMBCJEniC T//// Tu/i/So) eTrt <tt- 
HAHNANYCI0¥¥I ifK-nv^Avvalovvl- 
EOCAYTOYEZET €o<i avrov €[^] ir■ 
E(^)H>AHHKK||||EC^[H(: ecov fxvr]p,[r]vl] €(rTr](T[€- 
HF¥r/l// V . . . . 

Balki dere and Kaballa. — One inscription of Balki keui, and some 
from other villages on the road between it and Konia, are published by F. 
Sarre, A.E. Mitth. xix. (1896), p. 35 f. I traversed the road from Ilghin by 
Balki to Tchigil and copied the following inscriptions. 

^ So '^vas for Nay, etc. 


62. Getclud keiii (in tlio Dere, 2 hr.s. 5 min. from Il^hiii) 



in the 

Met/309 Na8t rfi eav^Tou) 
yvvaiKl yXvKVTUTT] 
K€ kavrw ^oiv Ke (f)p- 
ovwv avearr^crev. 

Na<?, Na (No. 05), 'Era?^ belong to the class of Lallnamen which appear 
in all languages aiul of which Kretschmer, Einhititng, p. 334 ff., has collected 
numerous examples. This class of name is * very common in Isauria 
(Heberdey-Wilhelm, Reisen p. 128) and Pisidia. 

63. Balki keui (J hr. from former), in the bridge. 

ACI.H 6M ©el 
TH^ AYT 0^ 

QI//KO I CO m 


Aifp. M.€vve- 
a? "I[/*]ei/09 
rfj eavTov 
a[v]v^iQ) yX[v- 
KVTarrj Aou- 
8a Koi Tu> vw 
Avp. "l[fl]€Vl fJ'v[f')- 
fir]<; xa]/3ti/ 

The name Avp. M€vv€a<; "I/*ei/o9 occurs twice in Sterr. W.£. 373 (30 
and 39). 

64, Tchigil (Ashagha) : poor lettering. 

pecTiNJ4^Tco,;,A| LOAN 

C A-AN//////////N H 
MHC- • Y^AP iN 

AvprjXla ^68a k6 'O- 
pecrrlva tS iSio) av- 
Spl irodtvoTaTU) 
M[o]i;a ? avea-Tt)- 

A68a also in Sterr. H.J. 202. AoSa, AouS?;?, AowSa? are all by-forms 
of DADA : see Kretschmer, op. cit. p. 337. 

^ Cp. CB. no. 91, p. 269, where 'Ec^ t^ 7i/»'ai(cl is rightly read, Rev. Univ. Midi, 1895, p. 362. 

120 .). G. C. ANDERSON. 

65. Tchigil (Yokaru), 20 min. East ; near the mosque. 

'/WP/VAjUANOYT ^""P- ^^ Mai/oi^ T- 

ANO/WCUAAANE ^^°^ [A]&>Sa(? or 'loiSa?) ave- 

CTK^iN^NHAHC aTTjaev fx . x ■ 

66. I hid. on a pillar. 

ATTAAOC ^KrraXo'; 

eAYTOJKA kavT<fi Ka\). 

THTYNAIKIK 5l 'rfl ywaiKi K[al 

T(JL)YI(x)Z(x) ^ TCS Ut'o) fft)[l^ KOl (f)po- 

NCONMNH ^ va>p fipi][fir)^ x^^- 

P I N e n 1 H piv iirolT)[aev. 

Besides these there is another fragment and an inscription on a tall 
homos ; the latter was so faint and worn that I failed to make an intelligible 

These inscriptions and the numerous remains at Tchigil and Balki, 
which were said to have come from a site called Bel Oren on the easy hill- 
road between Balki and Tchigil/ attest an old settlement of considerable 
importance. Prof. Ramsay ^ would place at Tchigil the town kaballa, birth- 
place of Constantine V. Copronymos,^ an important fortress on the road 
between Hadrianopolis and Iconium, the Caballu-come of the Peut. Table's 
false road Laudicia Catacecaumene xxiii Caball. xxxii Sabatra. The impor- 
tant passage for the topography is Cinnam. p. 42 fif. already quoted in part. 
In 908 it was held by a certain Andronicus and is described as ox^pov rt 
(ppovptop, ov irdvv ri p,rjKo6ev rov ^Ikovcov ScaKetfievov (Zon. xvi. 14). Dr. 
Tomaschek gives some additional references : in 822 Choireas held (f)povpiov 
rov 'AvaroXiKov t) Ka^aXa (Theoph, Cont. p. 72) ; ip to) Ka^dXa Xeyofiipcp 
aarec {Vita Euthym, ed. De Boor, c. xi. 8: cp. xiii. 20).* 

I think that an examination of the passage in Cinnamus shows clearly 
that Kaballa was situated much nearer to Konia than Tchigil. Here is the 
course of events. After a defeat near Gaita, the Sultan Masut fled to 
Iconium; but not desiring to be shut up in the city, he divided his army 
into three detachments, leaving one to guard the capital, placing the second 
on a steep behind the city and the third (with himself at its head) eV he^ia 
' relying on the strength of the mountain which stretches between Iconium 
and the fortress Kaballa.' Manuel now reached Kaballa and a battle ensued 
(on the right of the city). Part of the Sultan's army was routed and fled, 
pursued by the Romans. Meanwhile the remainder 'of the Roman army 
was attax:ked by an ambush reinforced by the guard left in Iconium (who 
sallied out, taking courage from the fact that Manuel was being carried by 

^ The ordinary road keeps along the Dere. ^ Mich. Glykas p. 528. 

' Hist. Ocog. p. 359. * Tomaschek, op. cit. p. 103. 



the ardoiir of pursuit fi\r from Tcoiiium) and by the detachmont posted on the 
steep behind the city (t^? iroX. oiria-to). On the return of the pursuers, the 
Romans succeeded in repelling the attack wlicn night came down. In the 
morning Manuel ' set out thence and encamped before Iconium ' * : but he 
gave up the idea of besieging it and after burning ra^ irpo t^9 TroXeox? 
olKo8o/jLa<i he began to retire again towards Lake Pasgousa (p. 58, i.e. Lake 
Karalis, Beysheher Giol). Meantime the Sultan was reinforced by troops 
from beyond Iconium ^ and hastened to attack the Roman army at the 
difficult pass Tchivrili-tchemani ^ (' the sinuous declivity,' Tomaschek, 
p. 101) ... 

This account shows clearly that Tchivrili-tchemani is the pass a few 
miles west of Iconium, beyond which the road to Vasada and Lake Karalis 
diverges from the Iconium-Hadrianopolis road,* and not the pass just east of 
Tchigil. It is equally clear that Kaballa is quite near Iconium, as Zonaras 
says (xvi. 14). Dr. Sarre (p. 35) would place it at Kavak keui, where there 
are remains (especially many fragments of Byzantine sculpture). I have not 
traversed this part of the road and therefore cannot offer an opinion about the 
site, but the situation suits the conditions. Whether the Gleichlaut der Worte 
Kawak mit Kaballa is more than an accident is doubtful, for Kavak (' poplar ') 
is a common village name : but it is quite possible that the Turks in taking 
over the old name gave it a form which had a meaning in their own 

Tyriaion, — Some inscriptions of Tyriaion and the district between it 
and Laodiceia Katakekaumene have been published by Mr. Hogarth in 
J.H.S. I.e. p. 162 ff.^ There has recently been a great destruction of marbles to 
obtain good stones for the new government buildings at Ilghin, but I 
succeeded in adding a number of inscriptions to the small list already known. 
Individually they may not always be very interesting, but in the mass they are 
not unimportant, and I give many of them in cursive only rather than 
omit them altogether. The first will interest philologists. 



^ r N ANAlAMaXfiPOKE^il 



1 Cp. Nicetas, p. 72, ed. Bonn. 
^ ol avuTiTOD ir6\fa)S (fKovvro 'IkovIov, p. 46. 
' T^t^priXirCrifiavl X'^P"^ SviTirp6(To5os etirfp ris 
K.T.\., p. 47. 
* Cp. Hist. Oeog. p. 359 note t. 

6 In no. 13, 1. 2 Jin. I read TAlOY ; 
no. 14, the epigi'aphic text is correct. One 
inscr. of Tyriaion in Heberdey-Wilhelm, Reisen, 
p. 162, no. 270. 



For a discussion of tho Pliiygian inscriptions reference may be made to 
Ramsay in Knlin's Zft.f. vrrfil. Sprrtchf., N.F. viii. pp. 381-410 (where all the 
inscriptions arc collected^) ; notes by Frck in Bezzenberger's Beitrdge z. Kunde 
d. indg. Sprachrn, xiv. (1889), p. 50 f., and Ramsay, ibid. p. 308 ff. ; A. Torp, 
Zu den phryg. Inschr. aus rom. Zeit (Kristiania, 1894) and Zum Fkn/gischen 
(Ibid. 1896). 

LI. 2, 6. SevvT] or B.evva, a fern, name, occurs several times in N.W, 
Phrygia: at Nacoleia (Ram.say, I.e. Nos. xv., xvi), at Apia and in upper 
Tembrogios valley (unpublished). 

L. 3, fxavKav lav earaet; . . . ' the monument which . . . set up.' 

68. Mahmud Assar, in the cemetery : letters rather faint. 

CAAeiTOY vac. 

to? (Te/jLOv[v K^VOVfi.- 

apec Ka[K]€v[v] a8aKe[T 

TlTT€TtKfJ,€VO^ aTTl- 

eaS etTov. 

KUKevv: cp. kukcv, B.C.H. 1893, p. 289 (cp. B.C.H. 1896, p. 111). 

aTTt,€ah : so in C.I.G. 3986 (near Ilghin) according to Seetzen's copy, 
while Hamilton's has AniCAA; ariah, Hogarth, No. 3; aariav, Ramsay, 
No. xiv., -TTtaS, No. xi., -laS according to my copy of Sterr. F.J. 174. 

69. Arkut Khan : 


Xovaov 'A/36- 
Xia yvvaiKt 

')(dpiv Trap' e- 
ai;[T]oi) Kal 
^a)u iav- 

"Zova-ov occurs frequently in this neighbourhood, Sterr. U.J. 156 (better 
in Hist. Geog. p. 178 n.) and 166, Hogarth I.e. Nos. 17, 25, 27. On the name, 
which is perhaps another Lallname, see Kretschmer, <yp. cit. p. 352. 

' Except the two fragments added by Sterr. 
E.J. 174 (see above) and 186 ; three added by 
Hogarth l.c nos. 1-3 ; and one by MM. Legrand 
and Chamonard, B.C.H. 1893, p. 289 (better in 
B.C.H. 1896, p. Ill) : a copy of this inscription 
which I made in 1896 reads AAAAKE 

TOPAEHC etc., the last letter being slightly 
blurred in the inside. In B.C.H. 1896 l.c. 
MM. Radet and Ouvre collect a few of those 
previously published, in ignorance, apparently, 
of the articles quoted above. 


70. Eldesh, in a cemetery : 

KOCXIICONKVPIOYKAIC// Korrfiuou Kvpiou Kaiaa- 

*?OCOY€PNAC6lP»-< pjo? ovepva<i ilpi^- 

It seems somewhat strange to find the Eirenarchate held by an Imperial 
vema. Was he connected with an Imperial Estate ? The Eirenarchate is 
discussed by Prof. O. Hirschfeld, Berl. Akad. SUzungsber, 1891, p. 808 ff. ; cp. 
CB. i. p. 68. The relations between the Eirenarch (with his gens d'armes, 
StfoyfiiTai), the Paraphylakes (heads of the village police, see above No. 14), 
and the '%rpaTr)<yo^ eVt tt)? x^P^"* cannot as yet be definitely determined ; 
but it seems probable that the Eirenarch (who was responsible to the muni- 
cipal authorities) was charged with the maintenance of public order in the 
city and its territory as a whole, and was therefore the superior of the local 

71. Ilghin : in large cemetery. 

BATAKHCMAI(t)ATeiT Bara/c?/? Maicfxirei t- 

HMHTPlMNHMHCe fj firjrpl fxv7]fir]^ e- 

NCKeN veK€v. 

BaraAcr;? or BaTTayK-;;"? was the name of a family which held the priest- 
hood of the Great Mother of the Gods at Pessinus, at the time of the campaign 
of Cn. Manlius (Polyb. xxii. 20, 5 irapayiyvovrai FdWoi irapct, "XTrthot koI 
BaTTUKov) and in the time of Marius (Plut. Vita Marii, v. 17). 

Maic^aret? is an interesting name. The masc. form occurs as the name 
of a Galatian slave [but Phrygian by race] in an inscr. at Delphi, Mafc^ara? 
TO yivoii TaXdrav (Wescher-Foucart, Inscr. rec. a Delphes, No. 189) and at 
Tokat in Pontus Galaticus (C.I.G. 4184, better Ath. Mitth. xiv. p. 316) along 
with Zapar}To<; (gen.) for which Dr. J. H. Mordtmann aptly quotes Hesych. 
ZaprjTiq ' ''ApT€fjLL<}, U^paac. The first element of the name is seen also 
in Mac-^ov^dvr)<i (Cataonia, B.G.H. 1883 p. 130), as is clear from Midpo- 
^ov^dvq^, the name of the Cappadocian general at the battle of the Granicus 
(Diod. Sic. xvii. 21 : cp. xxxi. 22) : also in Mathdra^ at Cos (Paton-Hicks, 
No. 10& 73, and No. 4i'^ = B.G.H. v, p. 225), and perhaps also in MatKidvr}'; 
at Cibyra {B.G.H. ii. p. 595) and Mat'/cuo? at Myrina {B.G.H. vii. p. 211). 
Just as 7tapariTo<i goes back to Zap7}Tt«?, and MiOpo^ov^dvrj'i to Mithras, 
so the Mai- group is derived from the goddess Ma (' the mother,' as every 
child knows !), worshipped especially at Comana (Capp.) but also in Rhodes, 
probably in Cos, and in other parts of the Asiatic mainland. 

72. Ibid. 

Taret? K.iBpafiovr} dv8pl \ fivqvri<i eveKev. 

124 J. G. C. ANDERSON. 

KiSpa/xovafi is related to the city-name Ki'Spafio^, as Attcs to Attouda 
Tottes (Tatas) to Tottaiou (Tataion), and many others quoted in Hist. Geoy. 
p. 241. The name occurs in Pamphylia {Ki.hpa^iva<i, KvBpafiova<;, Lanckor. 
Nos. 98, G5), in Pisidia (KiSpafid-i C.l.G. 4366mO. and the Phrygo-Pisidian 
frontier {Kt,hpa^l6a^-Kt,hpa^la^ Sterr. E.J. 39, 1. 30 and 46, 1. 13, KtSpo/ia? 
CB. 127, p. 310 improving Sterr. 45). Cp. Kretschmer, Einlcitnng, p. 333. 

73. Ilghin : now lying beside the new Depot. 

'Acria and 'Ao-ev? (No. 83) are interesting old names : they are clearly 
connected with the old divine names AC€IC, the title of the native Laodiceian 
deity, which was perhaps of Syrian origin (see CB. i. p. 33), and 'Aaia, 
which remained as an epithet of Athene in Colchis and at Las in Laconia 
(Paus. iii. 24, 7). 

74. Ihid. M.€V€KpdT\ri<i 'Kmria \ dv^arpl \ fipi]fir)<{ \ p^apti'. 

75. Ilghin : large cemetery. 

AiBvfio<i I 'AKvX€ivr) 6\v<yaTpl dp€\[<ryr\r}]<r€v /*i'|[»7]/a»?9 X'^P^^- 

76. Ilghin : in a tcheshme. 

AtXla Md^cfia xal Avpr]{\co<i) Nt[/c]a)i/ eavroi^ {jAj^VH'V^ X^'P''^' 

77. Ilghin : in another tcheshme. 

Et'p^apt[o-]T09 ^v\<^poa-vvr] yvvatKi \ Kot kavT^ ^ofVTi \ fivi)fir}<i x^P^^- 

78. Tchaiishji keui (head of Ilghin lake) : ' door-stone ' with pediment 
ornamented with floral designs. 

'ApxeXao^ Taret rfi yvvai\Kl IBia (f)i\oa-Top<yia^ ^vckcp. 

79. Mahmud Assar : in the Mosque. 

////OAcIjAAAKIBIAAH^ . ]7ra'- ^\. 'AXKi^idBr)- 

//////CTPATKTDTHC^^ ^ l](rTpaTiQ)T'n<; 

////ineVCKYPIAAH^ tVJTrev? KvpCXXrj 

////CYNBI<;^rYKY^ Tjj] avv^itp y(\)vKv- 
llll TATHMNHMHMHC rd-rri fivvfirj </i*»/> ? 

//// C^XAPIN o^ X^P^^- 

This inscription may be Christian. For the date [*]'n-a' expressed by 
letters alone without erov^, see CB. ii. p. 479, no. 350 and no. 645. 

80. Mahmud Assar : below the inscription is a relief representing two 
women, the left holding a hammer (?). 

Avp. "ZeXevKia Kal A[i]\\co<i kuI AeKfio^ reS | IBi^ 
Trarpl yXvKv\Td{T([)) fi.x- 

81. Eldesli : in cemetery. 

'ATTTra? Tarel [t]/} elSia | yvvaiKi fi. e. 
In this village there are other two defaced inscriptions. 


82. Ibid. : in large letters. 

////HCTONANATOAHKO//// K6fx]7]<; rbv 'XvaTo\r)Kb{vt. 


83. Ilghin : beside the Depot. 

'E\ta Bao-t|X,to-o-a dv\yarr)p Et\a|poy 'Trpe<T^v{Tepov)\ 
[tJoO 'Ao-e'tu? |[o']i"^ T(p i'|&> t^ov WaWr^piKiov a\ve<TTr)aa\v 
Tft) y\vK[v]\TdT(p fMOV d[v\8pl Tla]T\pt/c[iov P'X-]- 

Basilissa at Hadrianopolis, Sterr. 168 (see S7qjra) ; Hilara in Sterr. 163 
(read AiXdpa, cp. alavra, for kavrw); on 'Atreu?, No 59 (above). 

84. Ibid. 


+ A.vpv\io<i 'A- 
Ke Mt/30? vet- 
V ^eohovKov d- 
€v rfi fjbrjrpl 
T[pL€iVTO<t rrjv 
((T)TJ]Xr)v rav- 
rr)v fivr]/jb7)(':) 

In 11. 8, 9 the stone has E twice forC. FIPIEIYToC seems to be another 
Phrygian or native name. 



85. Ilghin : in a mo-sque. 

+ Avpikio(f:) I Kovcov d\v€<TTT)<r€v I /j,vi]fir]v Tov I TiKvov av\- 
Tov Tlav\\ov. 

The name Kovcov was common among the Christians : cp. on No. 60, and 
Hogarth No. 18. ITauXo? also in a fragmentary inscr, at Tchaiishji keui, 
UavXeivoi; at Hadrianopolis (Sterr. 160) : while 'Ituai/i/Tj? occurs in No. 90. 

86. Ilghin: in large cemetery. 

+ Avpi]\io(; I M.a^ifio<i I av(e)<TTr](T\€ tov elSiov tck- 
v\ov ' Epfiolyevov \ ^.\x- 

87. Ihid., in large letters. 

. . Si^ov, Kvpu, tO? €')^0p[v<; . . 

88. Tchaushji Keui. 

yiaTp(ovr]<; roBe arjfia iTTKTKOTrov Be dv- 
<yarpo<i yivrjaideov, tov 7rdvTe<; eTet/xcov, 
&)<? yap eoLKev la-TrjWrjv h' eaTtjaev 7r6crc<; koI <f>- 
iXTUTa T€Kva av{T)ov ' KpeWiavo<i ttj I8ia yvveKt 
Koi avijov) T^Kva 'Afifiia Koi Eip/JLLavo<i tt) Ihia fj,7]Tpl 
dvecTTrjcrav fiv^fir]<; ev- 


Hi). Mahmud Assar. 

A eujN H^hA 
M ATA <^ A A 

T HCA H(:N TOii 


iTf.; T A N M N H 
M H C X A P 1 N 

+ Tvv/3o<; eiairov- 
Sewv p-aKci- 
p(i)\v (OV TO, 6v6- 
yLiara <t>X. 'A- 
5 Xe^avSpo<i 
Ke 'A/j,cr]<i 8ia- 
Kovrjaris. dvea- 
T^(Tap,€v To(y) 

Tt'T[\]oi' flVT]- 

10 fir}<i ^dpiv 
K€ Met I/Of + 

LI. 1-3 Prof. Ramsay acutely detected ehirovhimv [i.e. airovhaiwv) 

L. 11 perhaps Mei[p]ov. 


90. Tekkc keui, ^ hr. E. of Mahmud Assar. 

AvptXia H^j/cXa dv€a-T\rjaa tov €fj,o\v reKvov 'lco\dv- 
ov /j,i/T]\fir]i; ■^dp(iv). 


91. Elciesh. 

EvddSe KaTd\KiT€ 'H/3a|/c\t9 Ke UaTpiKii; \ kc 
Tl[o]\vKap7ro<{ \ irpea^vrepoL \ /x.^. 

92. Arkut Khan. 

Pxor Tiu 

<{> LU N + 

-j- l^v-yrj Aeov- 
Ttou Kal 11- 
p')(OV tS){v) 

We have now reviewed the country along the great commercial highway 
of the Roman period from Apameia to the south-east corner of Phrygia. For 
five centuries or more, a constant stream of traffic passed along this road, and 
flourishing cities with numerous subject villages were to be found on it at 
short intervals. Here then, if anywhere in the interior, we should expect to 
find that the Graeco-Roman civilisation struck its roots wide and deep, absorb- 
ing and transforming the old native half-Oriental civilisation. Yet nothing 
is clearer than its failure to make any lasting impression. The Phrygian 
language lived on, and the native spirit retained its vitality and ultimately 
prevailed. In the plain of Metropolis the native population, the Euphorbeni,^ 
maintained its existence side by side with the Greek city. Lysias was planted 
amongst the Oiniatai : now only a low mound marks its site, while the 
Oiniatai have left their name to the village Oin^n and the plain around. 
The Roman city Julia flourished and died, and the old name Ipsos reasserted 
itself. In the plain of Philomelion the villages Azara, Pisa, Selinda live on 
as Azari, Bissa, Selind. In the remaining part of Paroreios we have found 
numerous indications of the persistence of the native element. Tyriaion 
clearly retained its native character, with a mere veneer of Greek civilisation, 
till the establishment of Christianity. The Hellenization of the interior 
(apart from the great cities) as a whole was due to the spread of Christianity ; 
but the Hellenism it brought was of a pithless, stagnant type, which was too 

1 Cp. CB. ii. I). 750-1. 


easily absorbed and assimilated when the great wave of Orientalism over- 
spread the land with the Turkish invasion. 

J. G. C. Anderson. 

March, 1898. 

Note on Kaballa and Tchivrili-tchemani (p. 121). 

Pro£ Kamsay has sent me the following communication : — " It is very probable that 
you are right in placing Tzibrilitzemani, etc. nearer Iconium than I have done. I hesitated 
long between your view and the one I have taken ; and at last it seemed to me that here, 
as in other cases, the Byzantine accounts exaggerate Manuel's progress and speak as if he 
had gone further than he really did. The literal interpretation of the Greek words is with 
you ; but I am not sure that we can take them literally. Similarly, historians speak of 
Loulon as near Tarsus, though it is nearly seventy miles distant." It is, of course, true 
that no stress can be laid on a mere phrase like ov ndw n yLy\ii66ev rov 'Ikov'iov (Zon. xvi. 14), 
which may (in a Byzantine historian) be perfectly vague, or on a mere solitary statement 
that a certain event took place ; but if we discredit a detailed account such as Cinnamus 
gives, do we not thereby raise a very large question ? 

Note on the Maps. 

The map of Phrygia Paroreios (PI. IV.) is based on Dr. Richard Kiepert's map attached 
to Dr. Sarre's book "Reise in Kleinasien," but I have introduced several alterations and 
additions based on my own observations and compass readings. The small inset map 
follows Prof. H. Kiepert's map in Dr. Buresch's .4 ms Lydien, etc., with a few alterations 
and some additions. Plate V. is based on Prof. Kiepert's large-scale map of Westlichea 
Klein Asien, but numerous alterations are made from my readings and observations (while 
in one or two points Prof. Radet's maps in En Phrygie have been laid under contribution). 




The Imperial Ottoman Museum has recently acquired a very valuable 
and interesting gold ring (Fig. 1) which was found in 1894 or 1895 in a 
tomb at Lampsacus. The Museum authorities subsequently undertook further 
excavations in the necropolis of which this tomb formed part, and it is a 
matter for great regret that no detailed report of the results was drawn up ; 
we are therefore forced to content ourselves with the somewhat meagre 
information given by the late Baltazzi-Bey to M. Salomon Reinach/ according 
to which the necropolis yielded fragments of red-figured pottery and specimens 
of silver autonomous coins of Lampsacus. Both these details are of importance 
in fixing the date of the ring; for on the one hand silver coins of this class 

Fig. 1.— Gold Ring in Imperial Ottoman Museum. (Twice actual size.) 

belong almost exclusively to the fourth century, and on the other, the manu 
facture of painted vases was not continued after that date. When we add 
that the evidence of coins ^ and inscriptions proves that this was the most 
flourishing period in the history of Lampsacus, we have strong a priori 
reason for assigning the ring to this century, while a consideration of the 
style of the intaglio may help us to fix the date within narrower limits. 

The ring has been already published by M. Salomon Reinach,^ who thus 

^ Eevue ArcMologique, 1895, ii. p. 363 = 
Chroniqtua d'Orieni, ii. p. 471. 

2 For coins see B. M. Catalogue of Greek Coins, 
Mysia, PI. xix. For inscriptions, B.C.H. xx. 
p. 553 (Proxenia of Epidaurus). 

3 Bad zincotype in the Eevue Arch. {loc. cit.). 
Enlarged design in the Chroniques d' Orient, 
which gives a very inadequate idea of the deli- 
cate execution of the design. The drawing in 
the text is by Mr. F. Anderson, from an im- 



describes the subject—' Seated draped Venus, holding in lier hand a long 
wand with which she is threatening a Cupid who stands facing her.' As a 
matter of fact Cupid is holding the wand quite as much as Venus, and Venus is 
not threatening her son ; she has extended two fingers of the right hand, the 
first and the middle finger, as if she were counting, from which we may 
conclude that she is playing a game well known to every traveller in Italy, 
the game of morra — (il giuoco della morra). Venus lifts two fingers; Cupid 
replies by a recognised move and flings forward his clenched fist. Additional 
proof is furnished by the wand, which each party clasps with the left hand. 
The game of morra is played with the right hand and at such a rapid pace 
that it would be impossible to count the moves if the attention of the players 
were distracted by any movement of the left hand, which is therefore ruled 

Fig. 2. (From R.F. Attic hydria in Dzialynski Collection.) 

out of the game. Nowadays this is achieved by putting it behind the back, 
but in Greece, as vase paintings show, it was done by making the players hold 
a wand. There is however this difference between the vase paintings and the 
intaglio, that in the former each player holds one end of the wand, and in the 
latter they both hold it by the same end. In order to facilitate comparison 
between the two methods of holding it we furnish (Fig. 2) a drawing of the 
scene depicted on a red-figured Attic hydria in the Dzialynski collection — the 
earliest known representation of the subject.^ 

presaion given to me by M. Mylonas. My thanks 
are due to H. E. Hamdy-Bey for kind permission 
to study the original. The dimensions of the 
bezel are 25 mm. x 21 mm. 

^ To Panotka is due the credit of having been 
the first to recognise that this scene depicts the 
game of morra. He detected it on a Graeco- 
I talian hydria in Munich, No. 805 (Bildcr ant. 



I only know one anciont monuniGnt which ropresonts the game of morra 
played as it is played in the present day, with the left hand behind the back. 
This is a large bronze (Fig. 3) in the British Museum.^ It is said to have 
been found at Foggia in 1869. In any case it is Italian, and not Greek work, 
and of comparatively recent date, being very similar in style to the large 
bronze Cupids found at Pompeii. It represents Eros standing holding up his 
left hand with an animated gesture, wliile the right is concealed behind his 
back. It may be presumed that this statue formed part of a group which 
represented Eros playing at morra with his mother, or more probably with 
his friend Ganymede. He plays three, raising the thumb, index and second 

Fio. 3. — Bronze Figure from Foggia. 
(British Museum.) 

fingers. From this we see that the game of morra was played in Italy in 
antiquity as it is to-day. It was only in Greek countries that it was played 
with the wand. 

Leben, PL x. 6), and on the Dzialynski hydria 
{Arch. Zeit. 1848, pp. 246, 7), published for the 
first time by Otto Jahn (Annali, 1866, Tav. 
d'agg. U). Both vases have been frequently 
figured, the latest publication being by Schrei- 
ber-Anderson, Atlas of Class. Ant., PI. Ixxix. 7 
and 10, with a bibliography of the subject. 
Heydemann found and published another repre- 

sentation on a cup from Ruvo (Naples Museum), 
No. 2574 {Arch. Zcitung 1891, PI. 56, 1), but 
it is doubtful whether the game is depicted on 
a Berlin hydria. No. 1953 (Jahn, Annali, 1866, 
Tav. d'agg. V and p. 328). 

1 No. 826. Ht. 2 ft. 6i in. Acquired from 
Piot. Unpublished. 

K 2 


The Lampsacus ring is a masterpiece of fourth-century goldsmith's 
work ; nothing but actual study of the original can enable us to realise the 
absolute perfection of technique displayed in rendering the folds of Aphrodite's 
chiton ; but some idea of it may be gathered from a comparison with the 
intaglio of another gold ring now in the British Museum/ which represents 
Aphrodite amusing Eros with a bird (Fig. 4). The type of Aphrodite is the 
same, costume and chair are identical, but the execution of the design 
wants the exquisite finish of the earlier work and proves that it must be 
assigned to a later date. This difference of date is confirmed by a 
comparison of the different types employed for Eros. In the one design he is 
a youth with great strong quivering wings meant for use, in the other a mere 
cupid decorated with a pair of useless winglets. 

Fig. 4. — Gold Ring in British Museum. (Twice actual size.) 

The composition of the design of our ring is no whit inferior to 
the rendering of it. If we study those vase paintings which represent 
this subject, we see that the adversaries are of equal size and are seated 
facing one another, each holding the wand by one hand; but our engraver 
has varied the design so as to make it fit better into the space at 
his disposal. Aphrodite sits and Eros stands before her; thus the artist 
obtains an upright design better suited to the narrow field of a seal than the 
horizontal one required for a vase painting. The design is so admirably 
suited for the decoration of a mirror case that it is there perhaps that we 
may most reasonably expect to find it some day. 

Paul F. Perdrizet. 

1 Unpublished. Acquired at the sale of Lord Vernon's antiquities, 1885 {Arch. Anzeiger, 
1886, p. 128). 



I SHOULD like to call attention to an attitude in some of the Attic 
stelai which has attraced little, if any, notice, while its meaning, so far as 
I know, has met with no consideration. 

The best example of the attitude is to be found in the beautiful Plate 
XXIV. of Mr. Percy Gardner's ' Sculptured Tombs of Hellas ' (Conze, PI. 
LXXVIII). Mr. Gardner describes the sitting figure as ' a lady stretching 
out both hands towards a matron who stands before her,' certainly an inade- 
quate description of a peculiar attitude. Conze, in his great work ' Die 
Attischen Grah^eliefs' says that the standing figure with the left hand takes 
hold of the right arm of the sitter on the under side (unterfasst). But this 
seems to me to miss the point. It is not the right arm but the right wrist 
of the sitter that is laid hold of. The forefingers of the standing figure are 
extended along the forearm and the thumb is raised at the wrist, so that the 
sitter's right hand lies softly in a sort of couch. From an artistic point of 
view the attitude is a remarkable one, being unlike that of any of the other 
best known stelai. 

In the nine parts of Conze's work I have found but one other example 
of the same attitude (Part I., PI. XLIII., fig. 150), and in it there is a slight 
modification, viz. that, while the left hand of the standing figure still lays 
hold of the right wrist of the sitter, the fingers of the latter twine softly 
round the standing figure's left wrist, from the under side, so that the two 
hands are linked together closely at the wrist, just below the bracelet which 
lies on the forearm. 

In the almost unique example of a ' Dying Woman ' treated in a 
realistic manner (Gardner, fig. 66), we have the same locking of the hands 
together at the wrist, but with this difference from the previous illustration 
(Conze, fig. 150), that the fainting lady extends her right hand, which is 
received by the standing figure's right hand at the wrist. The latter lady is 
described by Mr. Gardner as the ' Mother whose extended arms signify 
sympathy and grief,' the peculiar attitude being entirely overlooked. 

In Conze, Part I., Plate XXVI. there is a beautiful relief of a sitting 
lady between whose knees a naked boy presses forward with loving eagerness. 
Conze says, ' She embraces him with both hands.' But this is a weak 
description of a beautiful attitude. Her right hand lies tenderly under his 
left wrist or forearm, her fingers being extended to his elbow, a change of 
position from that of our former illustrations. But from the relation of the 


figures to each other any other arrangement would have been impossible. 
We have however the same soft, gentle touch as in the former cases. 

Other two variations complete the list of illustrations [which I have 
been able to find in Conze. 

In Part VIII., Plate CXCVIII. an old man with his right hand clasps the 
right hand of a girl (in the usual manner of the Be^ioxri^), but Conze fails to 
observe that the left hand of the old man is laid on the right wrist and 
forearm of the girl, his fingers appearing below, clearly a mark of the closest 

The same attitude is found in Conze, Part I., PI. XCVIII. ' Corallion ' 
clasps the right hand of ' Agathon ' with her right in the usual manner, while 
her left lies under his right wrist and forearm. This is noted by Conze. 
Here again this action shows an affectionate tenderness. 

Last of all in the beautiful mural bas-relief of Naples (P. Gardner, PI. 
XXIX., ' Orpheus, Eurydice, and Hermes ') we have a fine example of the 
attitude in its most expressive form. Here it is Hermes, the Psychagogos, 
that lays his hand most tenderly on the right wrist of Eurydice, twining it 
round in what seems to us a forced action, as by this gentle touch he per- 
forms his sad duty of leading again to the Shades the newly-lost wife. The 
mourn fulness of the scene is heightened by the position of Orpheus and Eurydice 
towards each other. Turning to him with her last fond gaze she places her left 
hand softly on his shoulder (another mark of loving regard), while Orpheus 
gently lays the fingers of his right hand on the left wrist of Eurydice. The 
pathos and beauty of this bas-relief and of the one in the Louvre have long 
attracted the admiration of all beholders, but my special point has been 

Now, have we any allusions in Greek literature to the significance of 
these rarely found illustrations, specially in connection with the right wrist ? 
Doubtless scholars will discover others, but I find in Homer two very distinct 
references. In a most tender passage (Od. XVIII. 258), Penelope tells 
Eurymachus that, when Odysseus left for Troy, whence he might not return, 
as the Trojans were great warriors, he bade her an afifectionate farewell and 

he^LTeprjv iirl Kapir^ iXwv ifie X^^P^ Trpoa-rjvSa. 

This seems exactly to describe the attitude and its pathetic significance. 

Another passage (II. XXIV. 671) also throws light on the subject. 

Achilles, yielding to the commands of the gods, and conquering his implacable 

hate, agrees to give up the dead body of Hector on the prayer of the aged 

Priam, and 

'n<; dpa (f)Q)v7](Ta<i iirl Kapirtfi %etpa <yepovTo<i 
eWa^e Be^iTeprjv, fMTjirco'i Set'cret' ivl dvfio). 

In his note on this passage Mr. Leaf says that this attitude is a mark of 
kindness. But it surely indicates a deeper and stronger feeling, a desire to 
give courage and confidence (as Homer says) to the aged king amid the 
dangers to which he was exposed in the camp of his enemies. Coming fi-ora 


the resentful Achilles it has a deep significance. The two passages taken 
together show that the attitude was used on occasions of intense emotion or 
of deep passion. Carl Sittl {Die Gehdrdcn der Griechen und Romcr) mentions 
the attitude in p. 314, note 2, and on p. 315, note (5, but does not attach any 
special meaning to it. With regard to the hand on the shoulder (two figures 
each with the hand on the shoulder of the other), lie says that the position 
somehow, yet still clearly, indicates affection. In art, however, he connects it 
with a late period, but it must be borne in mind that we find it in the 
' Orpheus and Eurydice ' bas-relief, which is generally ascribed to the end of 
the fifth or at the latest to the beginning of the fourth century. 

The aspect of the Greeks and Romans towards the wrist seems worthy of 
careful examination. While we speak of kissing the hand, Homer says 
Xa^wv Kvae %etp' eVt Kapirw (Od. XXIV. 398), doubtless the right wrist in 
connection with 8€^icoai,<{. From the more delicate skin and the more 
sensitive touch of the wrist, one would feel inclined to say that the ancients 
showed a finer appreciation of the motive. 

Then, further, while our young men and maidens dance hand in hand. 
Homer, in his picture on the shield of Achilles, represents them as dancing 
iiri KapTTO) '^elpa<; exovrc'i. 

John Forbes White. 



[Plate VI.] 

The vase which is the subject of the present paper was acquired a few 
years ago in Italy for the Ashmolean Museum. The find spot was given as 

The vase is a bell-krater, in height 20 1 in. (m '523) in diameter 19f in. 
(m •498). There is a wreath above, and a line of maeander pattern beneath 
the figures. The sides of the vase have a decoration of unusual richness. 
An elaborate pattern of palmettes rises to the handles, the roots of which are 
surrounded by the so-called egg-moulding (Eierstab). The reverse type is one 
of the conventional groups usual in this class of vases. Three youths, their 
long hair bound with wreaths (white), and wrapped in himatia, are standing 
together in a building which is indicated by a column rising in the midst. In 
the back-ground are hung up two square frames (dedications?); two tall 
curved leaves like notes of interrogation rise from the ground. One of the 
youths, to r., holds a patera ; another, to 1., holds a strigil ; the third 
is wrapped wholly in his cloak. In the field to 1. is a flower. 

The interest of the vase resides wholly in the painting of the obverse, 
which I take to be a new and probably unique representation of the carrying 
off of Oreithyia by Boreas. Boreas, who is represented as a dignified bearded 
man, clad in Phrygian cap, chiton with sleeves, chlamys and boots, is seizing 
by hair and right arm Oreithyia, who has flung herself violently on the 
ground, and raises her hands beseechingly to a richly draped matronly figure 
who is seated on a rock (hair in kerchief, under- and over-garment). An 
Eros, in a curious attitude, as if he were also perched on a rock rather than 
floating, rises beside the seated matron, holding some white object (wreath or 
fillet ?) in both hands. Behind the matron stands a female figure, her hair 
bound with a sphendone, holding in her left hand the end of her veil. 
Behind Boreas, his horse advances to r. The drawing is very good for the 
period, which I take to be not very late in the fourth century. White colour 
is used for the face, arms and feet of the standing womap, the border of the 
cap of Boreas, and the mane of the horse, as well as for ornaments and 
accessories. Plate VI. is from a very faithful drawing, made by Mr. F. 

That the vase is Attic will probably not be disputed. The subject in itself 
points clearly to Athens. The elaborate devices under the handles, which are 

J.H.S. VOL. XVIII (1898) PL. V(. 




Kbater in Ashmolean Museum (Rbveese). 

almost identical with those on an Attic vase of earlier style from Gela, 
recently acquired by the Ashmolean Museum, indicate the same origin. 
Vases of this class are in the Berlin Catalogue (Nos. 2641-8) and in the 
British Museum Catalogue (vol. iv) assigned to Athens. 

Representations of the carrying off of Oreithyia by Boreas are by no 
means rare in Greek vase-painting. They are especially common on Attic 
vases of the fifth century. Sometimes the scheme is one of flight and 
pursuit : sometimes the girl is represented in the arms of her suitor. On the 
chest of Cypselus Boreas has the serpent legs of Typhon ; on a red-figured 
vase ^ he is double-headed like Janus ; Oreithyia like Thetis is often seized 
in the presence of her companions, who hasten away to tell the tale. The 
most important vase of this class is at Munich.^ It ia a fine red-figured vase. 
Boreas winged (BORA 5) has seized Oreithyia (ofJEIOVia), who stretches out 
her hands towards Herse (eP^E) who follows the pair with arms outstretched 
as if to aid : Pandrosos, Aglauros, and a third young woman not named, fly in 
terror to Cecrops and Erechtheus to tell them what has happened. A very 
similar vase at Berlin is published by Gerhard,^ where the same figures 

» Ann. d. Inst. 32 (1860), PI. L.M. 
^ O. Jahn's Catalogue, No. 376. cf. Welcker, 
Alte Denkmaler, ill. p. 144. 

' Etniak. u. Kampan. Fasenb. PL 26-29. cf. 
Berlin Catalogue of Faaea, No. 2165. 


reappear, though only the names of Boreas and Oreithyia are given ; and 
where again Herse -seems disposed to attempt a rescue. On one red- 
figured vase Athena is present, and is by no means out of place, since 
according to one version of the myth Oreithyia was carried off while engaged 
as a canephoros in her service. But according to more generally received 
accounts she was surprised while picking flowers by the Ilissus, or filling a 
vessel at Callirhoe, or carried off from the rocks of the Areiopagus. 

A distinctive representation dating from the fourth century is found in 
the well-known akroterion of Delos. Here, Boreas, winged, clad only in 
drapery which falls behind the shoulders, and in boots, raises aloft Oreithyia, 
whom he has seized in the presence of two of her sisters or friends.^ At the 
feet of captor and captive, a small horse springs to the right. This horse, or 
mare, is taken by Miss Harrison to represent a transformation of Oreithyia, 
whom she thinks to have been, in origin, a sea-nymph, and so to have 
possessed, like Thetis and the Old Man of the Sea, the power of assuming 
various shapes. 

In Roscher's Lexikon, under Boreas, p. 811, citation is made of a vase 
whereon a young man, wearing a Phrygian cap, bears away, in a quadriga, a 
struggling girl. This vase was published originally by Welcker,^ who took it 
for an abnormal representation of the carrying off of Oreithyia, and the writer 
in the Lexikon (Rapp) accepts the attribution. But it cannot be upheld ; 
Stephani observes^ that all save the horses is modern painting, and the 
figures in the chariot have neither a genuine appearance, nor any likeness to 
Boreas and his bride. Probably, the real subject of the vase, which is of the 
Panathenaic class, is a victorious racing chariot. 

Passing from other representations of the wooing of Boreas to our vase, 
we are struck by the many points in which it varies from the accepted version. 
To begin with, Boreas is not winged. The pattern on his cap and the scales 
on his sleeves serve to mark his northern origin, and his affinity to Scythians 
and Amazons. He wears tall hunting boots closely like those in the Delian 
group. The horse which accompanies him seems clearly meant to bear away 
the captive, and does not lend itself readily to the view of Miss Harrison 
above quoted, that it belongs rather to Oreithyia than to her suitor. 

Of the three figures on the left of the picture one is undoubtedly Eros, 
and one must almost certainly be taken for Aphrodite. But it is possible 
to hesitate whether Aphrodite is the seated or the standing figure. We 
have, indeed, here, an interesting problem. Three views deserve con- 
sideration : — 

(1) That the standing figure is Aphrodite, the seated figure a relative 
of Oreithyia. 

(2) That the standing figure is Aphrodite, the seated figure an 
impersonation of locality. 

(3) That the standing figure is Peitho, the seated figure Aphrodite. 

^ See Furtwingler's restoration in Arch. The presence of the horse is a certainty. 
Zcilung, 40, 339, Roscher's Lexikon, p, 813, Miss ^ Alte Denkmdler, v. pi. 21. 

Harrison, Mythology and Monuments, p, Ixxvii. ' Boreas, p. 11, 


We must consider these possible interpretations in turn ; and first, the 
view that the seated lady is related to Oreithyia. 

The name of the wife of Erechtheus, Praxithea, is recorded, but according 
to all accounts, she was not present at the scene of abduction. We may, 
perhaps, suppose that the idea of separation from home has introduced the 
mother into the scene as the representative of home. 

Greek vase paintings of all ages have, as every one knows, a strong tend- 
ency to fiill under definite schemes, and the presence of a mother at a scene of 
abduction certainly adds a touch of pathos. Instances of the addition in 
vase-paintings of persons whose presence is rather ideal than actual are 
very common. We may cite, on black-figured vases, the father and mother 
of the Cercopes witnessing the capture of their sons,^ and Peleus and 
Neoptolemus present at the arming of Achilles^; on red-figured 
vases, Timandra present at the abduction of Helen, on the vase of Hieron ; 
Apollo present at the duel of Achilles and Hector,* and Cheiron present at 
the surprise of Thetis. Such instances might be multiplied, but it is 
unnecessary. Perhaps it is more to the point to observe that Demeter is 
sometimes, on vases, present at the seizing of Persephone,* although, according 
to the received legend, she was absent at the time, and unaware, for a while, 
what had happened. It is true that the majority of archaeologists, from 
Millingen to Overbeck, have seen, in the vase paintings mentioned, not the 
first violent abduction of Cora, but her more peaceful annual departure to 
the world of shades ; but this interpretation is doubtful, and even if it be 
adopted, Demeter should scarcely be present. An objection to the identi- 
fication of our seated lady as Praxithea is that she is perfectly quiet and 
self-contained, evidently in no wise disturbed by the terror and the appeal of 
the girl. 

Of one of the sisters or companions of Oreithyia we can scarcely think. 
Herse, daughter of Cecrops, is, as we have seen, present at the deed on some 
early vases, and even tries to interfere. But our seated lady can scarcely be 
Herse, nor can she be Athena, since none of the characteristic features of 
Athena appear. 

A second interpretation is to find in the seated lady some personification ' 
of locality, perhaps Callirhoe, present, as Eleusis is sometimes present at the 
sending out of Triptolemus. The rock-seat would, certainly, very well suit 
this interpretation. On late vases, and especially in Pompeian paintings, 
such personifications are frequent in mythological scenes. And they take an 
interest in what is going forward, expressing sympathy by attitude and 
gesture. But they do not usually take an actual part in the action. It is 
difficult to think that Oreithyia could appeal for protection to Callirhoe, or 
any impersonation of locality. 

The third interpretation remains, that the seated lady may be Aphrodite, 
and the standing figure behind her Peitho. In favour of this view is the 

* Ashmolean Vases, PI. 8. * Overbeck, Kunstmythol. pi. xvii., 26, 26a, 

- Heydemann, Griech. Vasenhilder, vi. 4. cf. text, p. 597. 

^ Gerhard, Auserl. Vasenb. 202. 


position of Eros, close to the seated figure. But an argument against it may 
be drawn from the attitude of Oreithyia, who is evidently appealing with 
vehemence to the seated figure. Aphrodite, if present, would be invisibly 
present ; and an appeal to her would scarcely be in place. 

Since there lie such serious objections to all interpretations which most 
readily occur, I am greatly disposed to adopt one which is more far-fetched, 
and has, as far as I know, no support from ancient writers, but yet has 
something in its favour. Can the seated figure be Gaia ? Gaia was, in a 
sense, the mother of Erichthonius, who is scarcely to be distinguished from 
Erechtheus, and appears in vases at his birth. She is closely connected with 
the early dynasty of Attica. The figure on our vase, in its dignity, would 
stand well for Gaia. Her rock-seat is appropriate. And Gaia, knowing the 
future well, might probably regard with complaisance a deed of violence of 
which the results were so auspicious to the Athenians, who more than once 
during the Persian wars had cause to rejoice that Boreas was their son-in-law. 
On the other hand, it would be very natural for Oreithyia to make appeal to 
her ancestress, Gaia, if she were present. 

I am quite alive to the dangers of giving far-fetched or fanciful 
interpretations of vase scenes, which in ninety-nine cases out of a hundred 
fall into classes and schemes. But, occasionally, a vase of unusual character 
makes its appearance, in interpreting which one is at liberty to assume more 
originality and more definite purpose in the painter. Our vase seems to me 
to belong to this small class. In any case, whatever the true interpretation 
of the picture before us may be, it is certainly a rare and important repre- 
sentation of an interesting Attic myth. 

Percy Gardner. 










[Plate XL] 

The head reproduced upon PI. XI. has recently been acquired by 
Philip Nelson, Esq., M.B., and we are indebted to him both for his courteous 
pernoission to publish the head and for the photographs, taken by himself, 
from which our illustration is derived. The head is of Parian marble, and is 
clearly of Greek workmanship; it is also evidently derived from an original 
of the very highest artistic merit. It is in excellent preservation, except 
that the end of the nose and a part of the lips on the right side have been 

Dr. Nelson has kindly supplied me with the following information as 
to the history of the head and as to its dimensions and present condition. It 
was acquired by him in Bath at the sale of the collection of the late Captain 
Maignac, who inherited it from his father-in-law, an artist named Walton, a 
contemporary of the painter Barker of Bath, 1769-1847. This Walton in 
all probability brought the head from Italy, where he is known to have 
travelled and collected pictures, &c. ; but there seems to be no more exact 
record as to its origin. The head seems to have remained practically 
unknown to archaeologists until its acquisition by Dr. Nelson, who, appreci- 
ating its importance, sent photographs to the British Museum in July, 1897. 
In addition to the photographs now reproduced, he offers to have a mould 
made and casts prepared, if there is sufficient demand.* 

The left side shoAvs the bend of the neck and the beginning of the 
shoulder ; on the right side this is broken away ; the cleavage marks on the 
under surface show head and body to have been in one piece. Dr. Nelson 
also sends me the following dimensions : — 

Length of nose 

,, ,, eye with lachr. canal 
,, ,, ,1 without ,, 
,, ,, mouth 

Height of forehead 

Length of ear 

Width of ear 

Tip to ala of nose 

,, ,, edge of lower lip 

Upper to lower edge of lips 

63 mm. Between inner angles of eyes without 

31 ,, lachr. canal 

27 ,, Aperture of eye 

43 ,, Ala to ala of nose 

47 ,, Ala of nose to angle of mouth ... 

63 ,, Nose to tip of chin 

31 ,, Length of face 

35 , , Width of face at zygoma 

37 ,, Inner corner of eye to chin 

20 ,, ,, ,, ,, ,, hair 










^ Dr. Nelson's address is 14, Princes Road, Liverpool. I feel no doubt that his offer will 
be widely accepted. 


Dr. Nelson also observes the resemblance of this head to the works 
attributed by Professor Furtwangler to Cresilas, and suggests that we should 
recognise in it the Doryphorus recorded by Pliny among the works of that 
master; this is an identification we must now consider. 

The head, which appears to represent a youthful athlete, is not an 
exact replica of any known type,^ but it has an obvious affinity with the 
series that has been grouped together and discussed by Professor Furtwangler 
in his Masterpieces of Greek Sculpture under the name of Cresilas. And, 
if we were to accept this grouping and identification without discussion, all 
that would now be needful would be to describe the new head, and to record 
that another example — perhaps the most beautiful and characteristic of all 
— has now been added to the series.^ But the subject is so complicated and 
difficult that even those who have followed and accepted Professor Furt- 
wangler's brilliant arguments will not be sorry to reconsider the evidence 
on which they are based ; while there are probably others who are still 
unconvinced, and will be glad to investigate an alternative theory. 

The head which shows the closest resemblance to that reproduced on 
PI. XI. is the well-known Amazon of the Capitoline type.^ Now the 
relation of this Amazon to the other well-known statues of Amazons has 
always been a puzzle. Our views upon this matter depend to a great extent 
upon the amount of weight we assign to the tradition recorded by Pliny, 
that four* of the most celebrated artists of antiquity made statues of 
Amazons in competition for Ephesus, and that when the relative merit was 
decided by the votes of the competitors themselves, Polyclitus was placed 
first, Phidias second, Cresilas third, and Phradmon fourth. In the various 
attempts that have been made to fit this story to the various statues of 
Amazons that are extant, there is practically only one point on which all 
authorities are agreed ; this is that one of the types, which represents the 
Amazon as leaning her left elbow on a pillar, and resting her right hand on 
her head (Furtwangler, op. cit. PI. VIII.), is to be attributed to Polyclitus. 
One may feel inclined to discredit the rest of the story as a mere fabrication 
to glorify the Argive school ; but we know from Lucian, the most trustworthy 
and intelligent art-critic of antiquity, that there was an Amazon by Phidias, 
and a wounded Amazon by Cresilas is mentioned elsewhere by Pliny. 
Professor Furtwangler says that Pliny's statement 'should rather be con- 
sidered as confirmed by the fact that copies of precisely four statues of 
standing Amazons still exist, which on the one hand are clearly to be 

' This is a matter on which I speak with the posed. But surely, if so, the style is that of 

more confidence, as I understand from Dr. the copjdst, not of the original. It can hardly 

Nelson that Prof. Furtwangler has expressed the be disputed that the line originates in the bronze 

same opinion. casing of the inserted eye-sockets. 

* This new head shows a peculiarity in the ^ Furtwangler, Af aster pieces, Fig. 54 (Eng\ 

clearly marked line bordering the eyelids, which edition). 

occurs elsewhere, e.g. on the head of the * Pliny's confusion in making Cresilas' nation- 
Mattel Amazon. It is there claimed by Furt- ality, Cydon, into a fifth, does not affect the 
wangler, op. cit. p. 134, n. 2, as a point of style, value of the evidence ; it simply shows he was 
not of 'bronze technique,' as is commonly sup- cop)dng unintelligently. 


referred to four different artists, and, on the other, are evidently closely 
connected by identical measurements, by a general similarity of conception 
and dress, and by their belonging to the same period of art.' Even if we 
grant — what is by no means beyond dispute — Professor Furtwiingler's 
distinction of the four types and his assignment of all of them to the same 
period, we are faced here by a coincidence which is at least improbable. 
Any one who is acquainted with the history of the monuments of Greek 
sculpture, with the strange freaks of fortune that have led to one being lost 
and another preserved, and with the utter disproportion of what is now in 
our museums to what once enriched the shrines of Greece, will appreciate 
how unlikely it is that such a set of works should be preserved to us in its 
entirety. And, moreover, the assumption that all these Ephesian Amazons 
were represented as standing, not on horseback, and that they were identical 
in size, rests on no better evidence than the assumption that they were all 
wounded. The story of the competition itself is probably not taken seriously 
by any one ; but if we give it up, what is left to be deduced from Pliny's 
story except that there were four statues of Amazons at Ephesus ? There is 
no proof that they were either contemporary, similar in pose, or identical in 
size. It follows that there can be no compulsion for us to assign these four 
given types of Amazons to just the four masters mentioned by Pliny. Of 
course it is perfectly open to an archaeologist to prove, upon purely stylistic 
grounds, that any one of these types does belong to any one of the four 
masters; and in the case of the Polyclitan type, there is a general agree- 
ment, based upon adequate evidence for comparison. But, in the case of 
the others the evidence is slighter, and under the circumstances there is no 
necessity to assume that any particular one of the types goes back to one of 
the four sculptors mentioned by Pliny, though of course there is a probability 
that the work of some other beside Polyclitus may be preserved. 

It is desirable to keep before us these conditions as to identification, 
because, if we admit that the Capitoline type and the Mattei type must be 
by Cresilas and by Phidias respectively, the room for argument and con- 
jecture is very closely circumscribed, though probably some authorities would 
be inclined to reverse the attribution, and to go back to the old view that 
the Capitoline type must be assigned to Phidias. But if we do not admit 
that these two masters necessarily made these two statues, the field for com- 
parison is widened. Now the attribution of one of these statues to Phidias, 
though, it has been maintained in either case, has never been proved to 
conviction. To Cresilas Prof. Furtwangler has made a most brilliant attempt 
to affiliate the Capitoline type. 

The monumental evidence from which his argument starts is the statue 
of Pericles made by Cresilas, of which we most probably possess some copies. 
These copies differ a good deal among themselves; and while there is enough 
in common to all of them to give us some notion of the style of Cresilas 
there is also a good deal of variation, especially in technical details, which 
may be due to the copyist rather than to the original artist. Now if we look 
at the general conception and style of the head, apart from such technical 


details, and compare it with that of the Capitohne Amazon, or the similar 
head on the Mattei Amazon,^ I venture, for my part, to think that there is a 
difference far more essential than any points of resemblance that can be 
traced. The treatment of the hair is decidedly different, so far as one can 
compare a male head with a female ; and, even if this difference be due to 
the copyists, its evidence must tell against rather than for the identification ; 
and there is a considerable difference also in the shape of the eyes. That the 
eyes differ considerably in the various copies of the Pericles is an indication 
in itself; for the extraordinarily definite and clear-cut eyelids that we see in 
the Amazon and in Dr. Nelson's head could hardly have given rise to such 
variations. A comparison with the Pericles does not then compel us to 
attribute the wounded Amazon to Cresilas. Prof. Furtwangler argues indeed 
with great persuasiveness from Pliny's description of Cresilas' Amazon as 
wounded that the Capitoline type must be his. But the Polyclitan type is 
wounded too, though there is no record of it in literature, and though 
the motive of the wound is not worked out with equal skill. The fact is that 
we have do deal with a succession of probabilities, rather than with any 
definite proof, and so the question may at least be regarded as open enough 
to invite further discussion. 

In the case of the Amazons we meet with a phenomenon which is 
repeated with remarkable exactness in another case, and whatever explanation 
we accept in the one instance we shall probably be justified in applying to the 
other instance also. And curiously enough we have to deal in both cases with 
a group of works of which the most characteristic is universally recognised as 
belonging to Polyclitus. The other group comprises the statues which show us, 
in several variations, representations of a Diadumenus.^ In the case of the Dia- 
dumenus we have not, it is true, any tale of a competition by several well known 
sculptors, as in the case of the Amazons ; but we have a Diadumenus by Phidias 
mentioned, and so we have, in the two cases, the same two sculptors recorded 
as responsible for the most famous examples. And we find, as we should 
naturally have expected, that in the case of the Diadumenus as of the 
Amazon, modern authorities have recognised one type as decidedly Polyclitan, 
while another has been identified, though with less consensus of opinion, as 
derived from Phidias.^ There is however, another complication introduced 
in the case of several copies of even the Polyclitan Diadumenus ; for they 
show a strange contrast with the copies of the Polyclitan Doryphorus. Prof. 
Furtwangler, fully recognising this contrast, attributes the difference to a 
change of style on the part of the sculptor, the Doryphorus being an earUer 
work of purely Argive character, while the Diadumenus represents the work 
of his later years, under Attic influence which may well have been conveyed 

^ The Mattei Amazon, it will be remembered, while the two most characteristic athletic statues 
has lost its head, which has been replaced by of Polyclitus were the Diadumenus and a Dory- 
one properly belonging to the Capitoline type. phorus. 

' This coincidence is peculiarly interesting, ' Furtwangler regards the Farnese Diadume- 

when we remember Dr. Nelson's suggestion that nus as Phidian, following Gerhard and Bbtti- 

his head represents the Doryphorus of Cresilas, cher. 


through the channel of Cresilas. The only alternative is to recognise in the 
much mutilated head of the Vaison Diadumenus the most authentic copy of 
the work of Polyclitus himself, and to see in the more perfect and much softer 
work of such heads as those at Dresden and Cassel a modification introduced 
by an Attic copyist. Such a modification, of a quite Praxitelean character, 
must be recognised in the terra-cotta statuette from Smyrna published in the 
Hellenic Journal for 1885, PI. LXI., unless indeed we regard this Praxitelean 
modification as the work of a modern rather than an ancient copyist. But, 
whichever explanation we adopt as to the Dresden and Cassel heads — to 
which must be added another recently acquired by the British Museum, and 
the new Diadumenus from Delos — we must sec in them a modification of 
Argive severity under the influence of the softer Attic style, whether that 
influence was exerted upon Polyclitus himself in his later years or only upon 
the copyists who reproduced his work. 

Now if we turn to the Amazons again and remember that the 
explanation which applies in the one case must in all probabihty apply to 
the precisely similar phenomenon in the other case also, we may with great 
probability infer that here also we see the modification, under Attic influence, 
of the severer Argive type which is most characteristic of Polyclitus, at least 
in his younger years. If this explanation of the relation of the softer and the 
severer type be accepted for both cases alike, it will follow that we cannot 
while attributing the severer type in both cases to Polyclitus assign the softer 
type to Cresilas or to his influence. For Cresilas was the contemporary of 
Pericles and of Phidias, and belonged to an older generation than Polyclitus. 
It is not indeed impossible that the artistic activity of the two may have over- 
lapped to some extent. But it seems more natural to suppose that what is 
generally recognised as the characteristic Polyclitan type is the original, and 
that the other softer — one is almost inclined to say sentimental — type is a 
later modification ; and any such relation implies that we must not attribute 
the second type to Cresilas, but rather to some pupil or follower of Polyclitus 
who had fallen more or less under Attic influence. If, merely for the sake of 
clearness, and without any notion of introducing useless conjecture, we wish to 
bring any names into the matter, such a work as this is what we should expect 
from a sculptor like Naucydes or the younger Polyclitus ; or, if we take the 
Amazon into consideration, we might well attribute the new head which so 
closely resembles it to Phradmon, who appears to have been another artist of 
the same school, and then identify the Capitoline Amazon as derived also from 
Phradmon, the fourth of the sculptors quoted by Pliny for the Ephesian 
Amazons. This, however, is a conjecture which might be upset by the 
discovery of new evidence as to Phradmon's style ; and, in any case, we know 
so little about Phradmon that we cannot speak of him with any confidence. 
But it certainly seems to me that, if we are to assign both the Amazon and 
Dr. Nelson's new and beautiful head of an athlete to any particular sculptor, 
Phradmon's name suggests the right associations. 

There is a whole series of later Polyclitan works, among them such well- 
known examples as the Idolino at Florence and the Westmacott athlete in 


the British Museum ; and it is among these, though far above most of them 
in artistic merit, that the head we are now considering finds its natural place. 
This head is certainly the work of a sculptor who was a pupil of Polyclitus, 
but who introduced an Attic grace and power of expression into his master's 
severer style. Viewed in this light, the new head is not only a great acquisi- 
tion in itself, but it also supplies us with the earliest and finest example of a 
series already recognised and widely represented in the museums of Europe. 

The object of the present publication is to make accessible to archaeolo- 
gists, in an adequate reproduction, a head of very great beauty and interest. 
Before its exact place in the history of sculpture can be established, there i.s 
need of a far more lengthy and elaborate discussion than the mere sugges- 
tions that are given above. They will serve, however, to indicate the 
direction in which its affinities are to be sought, and to open a discussion in 
which others besides myself will doubtless take part. 

Ernest Gardner. 

J H S VOL. XVIII '1898), PL VI 



¥" -^ 

<j .^...x^-r:^ 

■6-- " i<^.- ^ 



J H S. VOL. XVm (1898' PL. Vlil 




•^ A.' 




J. H. S. VOL. XVIII. (1898) PL. IX. 



J.H.8. VOL. XVIII. '1696 PL. X. 




"- "WW^-^S^ggfe/^^^^ ..^^^2 





[Plates VII.— X.] 

In my first article on Pylos and Sphacteria ^ I made the raali promise 
that in an early number of this Journal I would support my theories by 
docimientary evidence. It is with shame that I realise that this is now two 
years ago. Various circumstances have delayed me. I have been unable to 
visit Greece again myself, and the friends who were kind enough to do the 
work for mo were constantly baulked by the storminess of the place. Not 
only was it often impossible to set up a camera orrore Trvevfia ex ttovtov eirj, 
but even to reach Sphacteria at all. Of the Pylian boatmen, as I know from 
my own experience, it cannot be said that d(f>ei8rj<; 6 KaraTrXov^ KaOearrjKc. 
It is only as a patchwork of the results of three different expeditions that I 
am now in a position to publish a plan of the irakaiov epv/xa and a fairly 
complete collection of photographs. In the present article my business \vi\\ 
be to act as showman to this series ; I have little new to add, and, happily, 
no fresh opponent to meet. My collaborators have, I think, on practically 
every point on which they have expressed an opinion, given their support to 
my views.2 The British School at Athens has been good enough to send 
down a representative on two separate occasions. Plate VII. Fig. 3, Plate VIII. 
Fig. 4 and Plate IX. Fig. 6 are from photographs taken by Mr. R. C. Bosanquet, 
who is publishing some of his own observations at the end of this article. The 
plan of the nakaiov epvfia — Fig. 10 — was made by Mr. J. W. Crowfoot, and I 
shall be able in dealing with it to quote in extenso from his notes. The bulk 
of the photographs, Plate VII. Figs. 1 and 2, Plate VIII. Fig. 5, Plate IX. Fig. 
7, Plate X. Figs. 8 and 9 and Fig. 11, were taken jointly by an old Glasgow 
pupil of mine, Mr. A. Lindsay, now Scholar of University College, Oxford, 
and his father, Prof. Lindsay, of the Free Church College, Glasgow. I cannot 
be grateful enough for the zeal with which all these scholars have thrown 
themselves into the details of this intricate question, and the kindness with 
which they have placed their results at my disposal. 

The illustrations naturally bring out many of the points which formed 
the subject of my controversy with Mr. G. B. Grundy.^ To Mr. Grundy's last 

' J.H.S. vol. xvi. pt. I. p. 55. (c) Classical Review, Nov. 1896. Criticism 

^ Mr. Lindsay's one exception is noticed p. 151. by G. B. Gnmdy. 

Mr. Bosanquet thinks Thucydidcs never visited (d) Classical Review, Feb. 1897. Answered 

the spot. For the extent to which I attach im- by R. M. Burrows, 

portance to this point, see C.ii., Feb. 1897, p. 9. (c) Classical Review, April 1897. Further 

» The bibliography of the subject is as criticism by G. B. Grundy, 

follows: (/) Classical Review, Dec. 1897. Further 

(a) Athenaeum, April 11, 1896. criticism by G. B. Grundy. 

Paper read before the Hellenic Society by A lucid, and to me very gratifying, summary 

G. B. Grundy. Criticism by R. M. Burrows. of my position is contained in vol. v. of Mr. 

(b) J.H.S. vol. xvi. pt. L, April 1896. Frazer'sPoiMantas, pp. 608-613. (Addendum to 
Articles by G. B. Grundy and R. M. Burrows. vol. iii. ) 

L 2 


words in this controversy I do not mean to reply in detail. The argument 
would largely resolve itself into the question not whether he is right or I, 
but whether in our mutual criticisms we have misrepresented each other. 
This is an unfruitful subject, and it is better to leave readers who are 
sufficiently interested to judge for themselves. They have all the data before 
them.^ I can only say that if I have misrepresented Mr. Grundy in any point 
I am extremely sorry for it. 

Let us turn to the illustrations, and begin with Plate VIII. Fig. 4, the 
general view of Pylos. It is taken from Sphacteria, which is just seen in the 
immediate foreground. The prominent position in the foreground is occupied 
by the Sikia Channel. Behind it is Pylos and the Sand Bar. The western 
of the two existing outlets is seen on the extreme right ; one which is now 
blocked up, further to the left. On the extreme left are the rocks where 
Brasidas tried to land. On the summit of Pylos are the remains of the 
Venetian Castle, with the Bay of Boidia Koilia and the circular spit of sand 
enclosing it to the right. On the further side of Boidia Koilia can be seen 
rather dimly Hagio Nikolo and the mainland to the north. On the horizon 
to the left is the Island of Prote.^ Plate VIII. Fig. 5, gives in detail the 
laud side of the rough ground on the left of Fig. 4. If instead of this photo- 
graph I had chosen for publication another of Mr. Lindsay's, taken nearer the 
water's edge, I could have shown still more indubitably the impossibility of 
beaching ships at this point.^ I wished however to call attention to the 
curious groove formation of the rock, noticed by Mr. Lindsay, which can 
be seen slanting from the centre of the photograph to the right foreground. 
It looks as if the rock was at some time or another roughly cut so as to form 
a groove for a wall.* I should not imagine however that this was done at the 
time of Demosthenes' occupation. It may be remembered that I noticed 
traces of Messenian and possibly Athenian work a little further inland,^ and 
Mr. Lindsay's remarks point to a Venetian date for the groove. He says 
' The width of the groove varies from two to five feet. It runs just within 
the rough rocks, usually about two yards outside the smooth ground. It 
begins at the south-west corner and runs for about 170 yards, at one place at 
a distance of about 70 yards from the sea. At its northern extremity it is 
joined by the remains of a wall running inland. This wall is not at all like 
any of the others of early date, and, so far as one can judge from its scanty 
remains, is Venetian work.' 

Mr. Lindsay also reminds me of an interesting point about the rough 
ground where Brasidas tried to land. The jagged rocks near the water's 
edge lie detached, one behind the other, and between them there is standing 
ground. When Demosthenes and his men came outside their proper line of 
defence they could stand trap aitr^ rfj pa')(^ia^ with their feet in the water, 

' They may compare for instance J.H.S.vol. * Cp. J.H.S. xvi. pt. I. pp. 17 and 25, C.Ii. 

xvi. p. 47 line 6 with ibid. p. 42 line 29, C.R. Feb. 1897 p. 4 and C.R. April 1897 p. 157 
Feb. 1897 p. 9 note 3 and C.R, April 1897 p. * It is possible that this groove was seen by 

158 bottom of Grst column and top of second. Bory de St. Vincent (Relation, p. 147). 

- Thuc. iv. 13, 3. « J.H.S. xvi. pt. I. p. 64. 

« Thuc. iv. 9, 2 and 3. 10, 4. 


and yet have ready-made walls to give them cover. It is little wonder that 
they won the shield of Brasidas.' 

The only other points to notice in Fig. 5 are the fine view of the 
Venetian Castle in the background to the right, and the glimpse of Prote to 
its left. 

Brasidas' rocks naturally suggest the exact position of Demosthenes* 
southern line of defence. Fig. 4 shows clearly enough that it would never 
have done for Demosthenes to leave outside his line of defences so much 
good level ground as Mr, Grundy's wall BB would make him do.^ My point 
however has been not merely that the southern wall — marked G.G. in my 
original plan ^ — must have been built as close to the edge as possible, and 
turned inland only where Brasidas' rocks compelled it, but that a short cross- 
wall — marked I in the plan — must have been built from north to south at 
the south-east corner. Here it was that, according to my theory, Demosthenes 
beached his ships, and here where the Spartans meant to make their attack 
with siege engines when they had failed on the south-west. I pointed out 
that I myself had walked without difficulty into Pylos this way, and that the 
Spartans would have done the same if there had been no wall to stop them. 
I suggested too that to the east of this wall, of the direct north to south line, 
there must have been then, as now, a slope where men could beach ships and 
use siege engines.* The documentary evidence is interesting. If we had 
only Fig. 4 to judge from, a view taken from the south, it would be difficult 
to believe that the rise in the ground at the south-east corner was so gradual. 
Mr. Grundy's heart, hardened against me by his survey, might become harder 
Htill. ' I do not know, of course,' he says,^ * what Mr. Burrows means by 
" never approaching the perpendicular." I see that at this south end of the 
east cliff, the summit of the clifif rises to a vertical height of 60 feet above 
its eastern foot, which is only at a horizontal distance of 81 feet from 
that summit. This slope moreover is not continuous, but in part much 
steeper than that implied by these general measurements; in fact, if I 
recollect aright, the lower part is perpendicular cliff, with a slope from the top 
of the cliff to the 60 feet level. Anyone who realises what this really means 
in nature will understand that Mr. Burrows' remark is highly misleading. 
Mr. Burrows then proceeds to talk of survey defeating it own object if it 

supersedes observation How can survey supersede observation when it 

is itself nothing else save the record of observation aided by instruments of 
accuracy ? ' 

I must ask Mr. Grundy to look at Plate VII. Fig. 1. It is this same 
slope taken in profile from the east, from the sand-bar itself. The ground 
running down from left to right in the background is of course part of 
the north slope of Sphacteria, the Sikia Channel lying unseen between. 

^ Thuc. iv. 12 1. clearness sake refer to the lettering of this plan. 

" See J.H.S. vol. xvi. pt. I. pi. II. and C.R. * J.H.S. vol. xvi. p. 64 and C.R. Feb. 1897, 

Feb 1897 p. 3. PP- 2' ^- See too Mr. Bosanquet's notes, infra, 

» J.H.S. vol. xvi. pt. I. p. 57. I shall often for » C.R. April 1897 p. 156. 



Demosthenes' would, I imagine, follow the same line as the Venetian wall, a 
tower of which we see to the right. Comment is needless. 

As for the slope to the east of the wall, it is clear that it is older than 
the alluvial deposit of the immediate foreground, and probable that it is of 
different formation. There would be ample room on the slope itself for the 
operations involved in my theory, even if the sand-bar had not yet begun to 
form to the east of it.^ 

While we are near the Sikia Channel it may be worth while con- 
sidering from a new point of view the question of the blocking of the 
channels.^ I have not laid sufficient stress on the wild improbability of the 
blocking ever having actually taken place under any circumstances and in 
any position. When the Athenians arrived and found no anchorage, they 
sailed north-west to Prote, Even if the Spartans could have thought it 
possible that they should abandon Demosthenes without an effort, without 
even an attempt to land stores and reinforcements, the direction in which 
they sailed and the nearness of their anchorage^ would have made them 
hesitate in forming any such idea. Are we to believe that when the 
Athenians sailed away the channels were blocked, but that when they came 
back the next morning the passage was clear ? We may allow a good deal 
for the stupidity of the Spartans, but are we to believe that they went 
through the difficult and elaborate operation of mooring light ships across 
channels exposed to wind and current,* and that then, just as they had proved 

» I have argued— C.Ti. Feb. 1897, p. 4— that 
the sand must have drifted to the S. E. corner 
before what we may call the West Centre of the 
Sand-Bar filled up, and that the present state 
of the two emissaries confirms this view. Mr. 
Grundy — C.R. April, 1897, p. 157 — answers 
that they are artificial. It would be more 
strictly accurate to say that they are weak 
points in the sand-bar, artificially turned into 
regular openings. Mr. Grundy himself gives 
the reason why they are where they are. It is 
because if made close under the cliffs they would 
become choked by the sand forming on the 
inner side of the Sikia Channel. This is what 
has actually happened, he proceeds to say, with 
the emissary marked on Plate VIII., Fig. 4, as 
running half way through the sandbank near 
the South-East corner of Pylos. His arguments 
are surely all for me. The causes which operate 
now may have operated then. None the less 
the movements of currents are an insecure basis 
for argument, and I am glad that my theory 
can, if necessary, dispense with it. See J.H.S. 
Vol. xvi. Pt. I. p 69. 

It is scarcely necessary to remark that the 
wall attacked by tlie Pelopounesians after the 
armistice was not wall 1 at all, but that which 
the land army had from the very first attacked 

on the north. Mr. Grundy {C.R. Dec. 1897, 
\). 448) should contrast Thuc. iv. 9, 2, 11, 2 
and 23, 2, with iv. 13, 1. See also C.R. Feb. 
1897, p. 4, col. 2, on rh Kara rhy Xt/xtva ruxos. 

2 C.R. April, 1897, p. 158, cp. C.R. Feb. 1897, 
])p. 8 and 9. Though still feeling that the argu- 
ments against Mr. Grundy's revised theory are 
overwhelming, I grant that I spoke too strongly 
when I said he had not one argument for it. 
He has pointed out that the land forces would 
thus have been kept in touch with Sphacteria. 
See, however, J.H.S. Vol. xvi. pj>. 74, 75. 
Sjihacteria was safe so long as the Athenians 
could not anchor. 

^ Prote was near enough for it to be quite 
clear if a fleet were making for it, even if the 
actual anchoring could not be seen. 

* It might also be argued that the time 
they could have had at their disposal for doing 
this was limited. For some part of the day 
immediately preceding the arrival of the 
Athenian fleet they were still attempting to 
efl'ect a landing on the S.W. (Thuc. iv. 13, 1). 
For this they would not only need all their 
ships, but also a free passage through the Sikia 
Channel. Besides it is evidently part of t\ 
diff'erent policy. 


the value of their policy by baffling the enemy's fleet, they promptly set to 
work in what remained of the same evening to loose their cables and beach 
their ships ? Men do not build elaborate fortifications and then throw them 
down on the same day that they have repulsed an attack. 

It is, after this, hardly necessary to point out again that Thucydides' 
words are not only not in favour of the blocking ever having taken place, 
but directly against it.^ 

In regard to the fortifications on the north side of Pyloa there is less to 
be said. Plate VII., Fig. 3, is the Cyclopean wall or tower behind Plate IX., 
Fig. 6. This, which I conjecture to be Demosthenes' line of defence, was 
marked L on my original plan ; to the Cyclopean wall I gave no special 
lettering. In regard to these two photographs I have nothing to add to my 
original remarks, except that Mr. Bosanquet agrees that" L may well be rough 
fifth century work.^ Plate IX., Fig. 7, however, which is wall L taken from 
a distance, brings out a point which I had overlooked when arguing as to the 
strength of the position it occupied. Mr. Grundy questioned the defensibility 
of this wall because it lay on a slope and the lower portion of it would therefore 
be exposed to enfilading.^ I answered the argument * without disputing its 
main premise, that the enemy would overlook the lower part of the wall. I 
forgot that immediately in front of the upper part there is a deep fall of the 
ground. Mr. Lindsay has brought this clearly out in his photograph. An enemy 
would not be at an advantage over any section of the defenders of the wall.° 

In regard to the a •priori question as to whether, apart from any 
consideration of the wall, Mr. Grundy's line of defence or mine is the more 
probable, I have only to quote some remarks of Mr. Lindsay's. He does not 
agree with everything that I have said on the subject. He could see not that 
there was any possibility of landing on the north shore of Pylos behind Mr. 
Grundy's wall AA..^ He does agree, however, that south of A A much ground 
would have to be defended, or at least guarded. His conclusion is: 'The 
line of LL therefore gives a much stronger and shorter line of defence. The 
line of AA is very easy to approach and has no such natural advantages.' 

We now turn to Sphacteria. The first point to notice is Figure 10, the 
plan of the TraXatov epvfia made this spring by Mr. J. W. Crowfoot. He 
writes to me that it was made with a prismatic compass and measuring 
tape only, but that each of the walls was measured separately and with 
care.'' The contouring on the other hand was done by eye, and is quite 

' Thuc. IV. 13, 3. See C.JR. Feb. 1897, i>. 9. ' Mr. Crowfoot sends me a note to say that 

* J.H.S. Vol. xvi. pp. 67, 68. See, liuwuver, he thinks he has rather underrated the bulge on 
Mr. Bosanquet's notes, infra. the north-west corner of the wall round the 

* C.R. Nov. 1896, p. 373. summit. This agrees with my own remem- 

* C.R. Feb 1897, p. 5. brances of the subject. See my original plan, 
^ It is only at this fall of the ground that J.H.S. vol. xvi. pt. i. p. 57. On this point, 

there are the number of stones of Cyclopean and in regard to the position of the S. wall 

size that can be seen in Fig 6. of the Hollow (p. 165), I have thought safest to 

6 Jf.H.a. Vol. xvi. Plate JI., cp. ibid. p. 66. leave the plan as Mr. Crowfoot made it. 



In regard to this plan Mr. Crowfoot writes as follows : — 
'The walls which compose the fort are in some parts difficult to trace; 
in others there is no regular line now visible : these I have marked on the 
plan with dotted lines. The island has probably never been much nif>re 

Fio. 10.— Plan of the ira\aihv Ipu^a-oN Sphacteria. I?y J. W. Crowfoot. 

Of the Walls marked \\\ /// no certain traces were seen. 

Where there are dotted lines - - - there is now no regular line of wall. 

The other jiarts of the Fort are drawn as at j)rcsent standing. 

inhabited than at present, and the stones forming the walls have not been 
carried away : except at one point on the western side they can be seen 
covering the slopes beneath the fort. I saw none which had been shaped or 
faced either on the ground or in the remaining courses, which were often set 


in rough lines. Only in one place, near the north end, could I find both sides 
of the wall : it measured 2 metres. Of the four bastions, only the most 
southerly is at all difficult to trace : behind this and behind the western 
bastion I could follow the line of the connecting wall, which was not, so far as 
I could see, "bonded" with the walls of the bastion. But when .so few 
courses remain, it would not be safe to lay much weight upon this fact. 

' Descending into the hollow I found the northern wall fairly easily, and 
measured it for more than 37 metres: its breadth in the middle is 8 metres. 
I am inclined to think that a line may have run at right angles to it, con- 
necting it for some distance at least with the northern point of the upper fort, 
but there are no distinct traces of this. About the southern wall of the 
hollow I cannot speak so confidently. The ground was covered with such a 
thick growth of brushwood, etc., that though I saw several lines of stones, I 
could not be sure of the exact position of the wall, but I have no doubt that 
it lies approximately where I have marked it. 

' Those who have visited the spot will see at once the importance of the 
Hollow. The upper fort is, except at the southern end, a mass of rock with- 
out an inch of soil anywhere upon it and the incline everywhere is consider- 
able. The Hollow is much more attractive : it is sheltered more or less on 
all sides, and there is, as I found to my cost, sufficient soil to support a very 
sturdy underwood. But unless the summit of the hill was defended, the 
hollow would have been exposed to an attack from above. 

' One question remains at present undiscussed. The fort was ancient 
in the time of Thucydides : what then can have been its original purpose ? 
The walls of Giannitza ^(i;. Pernice, Athen. Mittheil. 1894, p. 359) prove that 
the sub-Mycenaean folk of Messene were driven to higher points than those 
usually chosen by their predecessors, but the fort at Sphakteria can hardly 
have ever been the Akropolis of a flourishing community. It is more likely, 
one may conjecture, to have been a nest of pirates, who would have found 
plenty of spoil in the rich lands north and south of Pylos {v. Thucyd. i. 5).' 

We may now illustrate the plan by the photographs. The walls round 
the summit of Mount Elias — marked BB on my original plan — are repre- 
sented by Plate VII. Fig. 2 and Plate X. Fig. 8. In regard to them I have little 
to add.^ More than one of us who have worked at Pylos and Sphacteria will 
be glad that our dragoman, Charles Papadopoulos, is immortalized in Fig. 2. 
He was with Mr. Grundy and Mr. Lindsay, as well as with me. Speaking for 
myself I can bear witness that nothing could have been more enthusiastic, de- 
voted, and capable than the way in which he threw himself into the spirit of 
my work. I know too that it was not a little owing to his knowledge of the 
ground that Mr. Lindsay and his father were able to take such apposite 
photographs. Plate X. Fig. 9, for instance, represents the South Wall of the 
Hollow — marked D in my original plan — and it is particularly fortunate that 
Mr. Lindsay was able to photograph it. So hidden is it by brushwood that 

' See J. U.S. vol xvi. pp. 58, 59 ; also Mr. Bosanquet's notes, infra. 



Mr, Crowfoot could not discover it.^ As it was, a thick tangle had to be cut 
away before the camera could be got into position, and this was finally 
effected at so near a distance that the size of the stones is altogether out of 
proportion to those in the other photographs. 

Mr. Lindsay, on seeing Mr. Crowfoot's plan, writes to me as follows : 
' I think you are quite right in saying that the conjectural position of the 
southern wall on Mr. Crowfoot's plan is too far north. The wall is just where 
the hollow begins to slope down into the gorge. It was very hard to see 
owing to the dense growth of bush all round it, but the condition of the 
greater part of it was much like that of the north wall of the Hollow. 
I managed to get under the bashes at several points and made out the line of 
the wall. It only differed from the north wall in the fact that only a single 
row of stones was standing in most places and not so many stones were 
lying round. Where the Hollow slopes down into the gorge the rock goes 
sheer down on the west side a little way, say about 10 feet at first. This 

Fig. 11.— Sphacteria. "Wall CC. 

height gradually gets less as we move further east and when it is about 
8 feet high we begin to find wall D built against it on the outside. The bit 
we have photographed is situated at this point, built against a low line of 
rock. Where the rock stops the wall goes on, over the level ground, at the east 
side of the top of the gorge, just in front of the south end of the Hollow.' 

Our next photograph is Fig. 11, the north wall of the Hollow, marked 
CC on my original plan. Mr. Crowfoot writes to me that it appears to be at 
a curious angle, but that he was struck by the fact at the time, and cannot 
have exaggerated it by more than a foot or two, if at all. I can quite believe 
that the angle is exact. Of that part of CC which connects with the upper 

^ Mr. Bosanquet did not see it, because he left his search for it to the end, when the weather 
became unfavourable. 


fort, I thought I found traces, but grant that they were not so certain as the 
other parts of the fort.^ 

The only point that remains to be discussed is the last struggle round 
the Fort. All there is left for nic to do is to (piote Mr. Bosanquet and 
Mr. Lindsay. Mr. Bosanquet writes: 'Your theory of the surprise ^ held 
good when read over on the spot.' Mr. Lindsay confirms my views on two 
essential points. The Messenians could not have surprised the Spartans 
except from the south-east, because any movements to the north-east must 
have been detected. 'Standing behind wall BB,' he says, 'and looking 
north-west, I could see every foot of ground right down to the sea. It would 
be absolutely impossible for any one to pass from the west to the north with- 
out being seen by the defenders of wall BB.' But not only was it impossible 
for the Messenians to enter the hollow from the north, but it was possible for 
them to do so from the south-east. ' The gorge ' he says, ' is quite climbable. 
The chief difficulty would be the dense growth of bushes and trees.' In 
regard to the way by which the foot of the gorge was reached, Mr. Lindsay is 
inclined to believe with Mr. Tozer that a path may have existed along the 
foot of the cliff. ' We sailed close along the cliff,' he says, ' several times. It 
is sheer at the top, but slopes more towards the foot. There is still a con- 
tinuous line of bush from a gap in the cliffs north of the Panagia landing 
right to the foot of the gorge. Where there is bush there is some foothold. 
There is only one place which would be difficult to pass now, and that from 
the look of the rocks has changed recently. In fact the boatmen said the 
rocks had been shattered there by lightning.' 

With this quotation I bring my argument to an end. The need for 
constant reference to previous articles in this journal and elsewhere will 
probably irritate the reader. I must plead that the only alternative was to 
repeat much that is easily accessible and to run to excessive length. The 
time has not yet come to go over the ground again and give anything that 
should purport to be a final account of the whole matter. To do so at this 
stage would not have advanced knowledge, nor indeed have been courteous to 
my critics. I hope that before long I shall feel at liberty to banish polemics 
to footnotes and describe in direct narrative form what in my opinion actually 
took place in the affair of Pylos and Sphacteria. 

Ronald M. Burrows. 

Notes by Mr. R. C. Bosanquet. 
The Wall on Pylos. 

Professor Burrows quoted Blouet's opinion that this wall was modern, 
but ought in fairness to himself to have mentioned that one member at least 
of the expedition to the Morea held it to be ancient. ' Ce mur cyclopden,' 

' For my argument that the Spartans must ^ J. U.S. vol. xvi. \A. 1, j)}). 61, 62; C.H. 

haveheldwallCCaswellaswallBB, seeJ'.jy./S'. Feb. 1897, p. 2. Cp. C.R. April 1897, pp. 
vol. xvi. pt. 1, pp. 60, 61. 165, 156. 


says Bory de St. Vincent, ' est (jvidemment ce qui restc de plus antique dans 
tout le canton, et cependant on a affect^ d'y voir una batisse moderne sans 
importance/ an allusion to the scepticism of the architectural section.^ 

A careful examination of this wall satisfied me that it contains no 
squared stones, no stones with mortar adhering to them, and no tiles, such as 
would probably be found in a wall hastily constructed for defence in recent 
times. The fact that it is built of undressed stones is the explanation of its 
survival. No one takes such stones to build with when he can get blocks 
ready squared a few yards away. The supply of more attractive mcaterial on 
Koryphasion is not yet exhausted, although an enormous quantity has been 
carried off to build the modern town at the south end of the bay. Bory saw 
the process of destruction going on in 1829. Vischer in the fifties remarks 
that the ruins are being used as a quarry and verschwinden immer mehr.^ 
Many of the walls marked on Bory's plan (of which, thanks to Professor 
Burrows' foresight, I had a photograph with me upon the site) have been 
demolished, but the wall of unhewn stones still stands, eight feet thick, its 
top level with the ground on the west and rising from four to nine feet above 
that on the east. The 'tower' behind it is fully ten feet high (Plate VII., 
Fig. 3). 

Gell speaks of Sphacteria ' famous for the defeat and capture of the 
Spartans in the Peloponnesian War, and yet exhibiting the vestiges of walls 
which may have served as their last refuge,' ^ This is his only remark on 
the subject, and he gives us no clue as to what he saw or heard. 

Leake paid a hasty visit about the same time (1805) as Gell. ' Of the 
fort,' he says, ' of loose and rude construction, on the summit, it is not to be 
expected that any remains should now exist ; but there are some ruins of a 
signal-tower of a later age on the same site.' * This tower probably stood 
not as Professor Burrows conjectured (J.H.S. xvi. p. 63) at the S.E. angle 
of the Greek fort, but on the foundations of the best preserved of the western 
towers, in and around which I noticed a quantity of tiles. There were none 
elsewhere in the line of wall, and their abundance at this point is, explained 
if a mediaeval building of stone and tiles at one time stood here. It was 
probably a signal-tower, as Leake inferred, intended to warn the garrison of 
Palaeokastro of vessels making for the southern entrance to the harbour. The 
iraXaiov epvfia was no doubt occupied as a look-out station in Hellenistic and 
Roman times ; near the walls I picked up bits of Greek black-glazed ware, 
and part of a common ' pseudo-Arretine ' plate. If the tower stood as late as 
the fighting on Sphacteria in 1825, its squared stones n[iay have been requi- 
sitioned for the Greek batteries. The Turks too may have used them in 

^ J.H.S. xvi. 66. Of. Expid. de Moree, Griechenland (1857), l). ^38. 

Architecture i. p. 4, and Bory de St. Vincent, ' Sir W. Gell, Journey in the Morea, p. 5. 

Rilntiwi p. 154, (!f. p. 63 and his plan, PI. iv. * Leake, Travels in the Morea i. p. 408, 

' Vischer, Erinncrungen u. Eiiulriicke aiw 


1827, Had it survivetl until 182*J the members of the French expedition, 
who looked in vain for Greek remains on the island, would probably have 
mentioned it. 

However, not only the mediaeval tiles, but the rude blocks of the Greek 
foundations were left, and in 1865 Captain Mansell marked them as 
Cyclopean Ruins on the Admiralty Chart. 

In 1888 Schliemann came here in quest of Homeric remains, and 
employed men, as I learned on the spot, to clear away the debris which 
obscured the line of the western wall.^ His brief report attracted little or no 

In 1895 Professor Burrows independently discovered these walls, and 
was the first to discover those across the hollow. His observations have 
since been verified by Messrs. Lindsay, Mr. Crowfoot, and myself. 

As to the early date of the little fort there can be no question. It was 
perhaps a stronghold to which the fishermen and shepherds living round the 
bay retired upon the approach of pirates. In recent centuries the bay was a 
favourite resort of the Corsairs.^ 

The Question of the Sand-bar. 

My visit to Pylos in November 1896 chanced to coincide with the 
exceptionally heavy rains which caused disastrous floods in many parts of 
Greece. At Athens the Ilissus rose as it had not done since the spate 
recorded in Dodwell's drawing, destroyed houses and gardens, and inundated 
the Piraeus. Owing to wind and rain it was hardly possible to use my 
camera, but I was able to watch the process of lagoon-formation under the 
most favourable circumstances — a privilege for which, I afterwards paid a heavy 
penalty in the form of fever — and to verify the theory independently put 
forward by Mr. Grundy in these pages, and by Dr. Philippson in his study of 
the Peloponnese. On the morning before the heaviest rainfall the Jalova 
brook was barely a foot deep when I crossed it near its mouth. In the 
evening after some hours of rain it was a swollen torrent, discoloured with the 
sandy soil of the plain and quite unfordable. The main stream was certainly 
emptying itself into the bay, not into the lagoon. There was a gale blowing 
from the south, and much of the solid matter brought down by the flood 
must have been added to the sand-bank which separates the lagoon on the 
north from the bay. Two days later, when I again rode to Old Pylos round 
the shores of the bay, the trunks of several trees had been thrown up on the 
south slope of the sandbank, and the wash of the sea was fast covering them 
with sand and shingle. Within the lagoon a similar process hatl been at 
work. The water was much discoloured and had risen to the level of the sea 
outside. Evidently the soil brought down by the streams which discharge 
into the lagoon is deposited on the bottom, tending to fill it up, while the soil 

1 Ath. Mitth. xiv. 132. * Randolph,, Present UlcUe of the Morea (1686), p. 25. 


brought down by the Jalova and other streams flowing into the bay is 
deposited on the bottom in calm weather, but goes to increase the sand-bank 
whenever the wind is from the south or west. Before the partial obstruction 
of the Sikia Channel the flow of the sea from the west must have tended to 
keep open a passage through the sand-bar at its western end. But I see no 
reason to suppose that the slope, which undoubtedly exists to-day at the 
south-east angle of Koryphasion, did not exist two thousand years ago. 
Professor Burrows' view, that this was the place where Demosthenes drew up 
his ships and where the Spartans proposed to use their firj-^aval, seemed to 
me both possible and reasonable ; and I spent some time on this part of the 

May it not have been here, tiear the present well, that the Athenian 
sailors Bia/noofievoi rov KU'^XrjKa eVi rfj 6d\d<r(rTj eirivov olov eiKo^ v8ojp ? 


A useful bibliography relating to Pylos and Sphacteria will be found in 
Mr. Frazer's Pausanias. It seems worth while to add a note on the maps of 
the area under discussion. 

\l. Leake (1806). Travels in the Morea i. PI. iv., 1830. 

A rough sketch-map. 
II. Stanhope and Allason (1814). 

Cf. J. S. Stanhope, Topography of Plataca, etc., p. 27, 1835. 

Perhaps never published. Gail, who names Allason, Gell, and Hobhouse, as 
authorities for his map of Plataea, would probably have used AUason's work, had it been 
accessible, for his very inaccurate map of Pylos and Sphacteria, but does not profess to 
have done so. See J. B. Gail, Cartes relatives d la geographic de Herodote, Thucydide, 
etc., 1824, PI. 26, 47, 63. 

III. Capt. W. H. Smyth, R.N., Bay of Navarin {Admiralty), 1820, and later revisions. 

[Hence Bursian, Qeogr. von. Griechenland ii. Taf. V. 1872.] 

IV. The same. Two maps in Arnold, Thucydides ii., 1832. 

V. Bory de St. Vincent (1829) Exped. Scient. de Morie, Atlas, PI. iv., 1835. 

[Hence Curtius, Peloponnesos ii. Taf. viii., 1852.] 
VI. Blouet (1829) Expdd. Scient. de Moric, Architecture i. PI. v.-vii., 1832. 

Unimportant map of Pylos, plan of Palaeokastro, plan and section of Nestor's cave, 
detail of Hellenic wall, plan and elevation of Hellenic pier. 

VII. Capt. A. L. Mansell, R.N., Bay of Navarin (Admiralty), 1865. 
VIII. G. B. Grundy (1895) Jotmial of Hellenic Studies xvi. PI. ii., iii., 1896, 

Bory de St. Vincent's map will always be of value as recording walls and 
ancient remains, some of which have since disappeared. For topographical 
and historical purposes Mr. Grundy's admirable survey has superseded all 
others. Since it is certain to be largely copied, I feel sure that he will 
pardon me for pointing out some trifling errors of nomenclature, due for the 
most part to the recent Admiralty map, which is not so safe a guide as it 
should be. 

J.H.S. xvi. PI. III. Lyhos (cf. p. 5 'The alhivial plain of Lykos'), should 
be Levko (AevKo). Gadaro Point should be Gaidaro. In both cases the 


riglit spelling is given in the Admiralty map of 1820. Tortori rocks, (ik. 
TovpTovpi. Pylos Island, a map-maker's name for the rock south of Sphac- 
teria, is not known locally. Cf. B. de St. V.'s protest, Rdation, p. 48. 
Marathonisi : the real name is Chelonaki, 'tortoise,' which appears undc.T 
the form Kilonaki in the map made by Smyth for Arnold. Smyth's 
Admiralty map gives Marathonisi or Kuloneski, and these names have passed 
into all the later maps and into the text of the Expedition to the Morea. 
Possibly the island was once called Marathonisi, but the name has long been 
obsolete. Kuloncski is suspiciously like an engraver's misreading of Smyth's 
Kilonaki. Enquiry at Pylos failed to elicit anything but Chelonaki, with 
a possible variant Chclonitza. Finlay gives the right spelling in his account 
of the Battle of Navarino, 

The elders of Pylos also denied that the name Boidio-Kilia had anything 
to do with the cave. Leake is the first and perhaps the only authority 
for the statement that the cave gave its name to the harbour. In the same 
year (1805) Gell saw the cave 'which some Frank has taught the two or three 
Greeks who ever heard of Nestor to believe was the stall where lie kept his 
cows.' Both Blouet and Bory in 1829 speak of the cave as bearing the name 
of Nestor. It is possible that Leake or his informant was mistaken. A 
last century writer mentions that part of the harbour of Zea (Keos) was called 
' le cul de bceuf.' 


[I am grateful to Mr. Bosanquet for more than one reference, including 
that of Bory de St. Vincent's Eolation, which I had read, but overlooked at 
the time of writing. Vischer, it should be added, saw the Tower behind L 
(Erinnerungen, p. 435) and considered it Cyclopean. In reference to Boidia 
Koilia, it may be worth while to point out that the connection of the name 
with Thucydides' Bov<f)pd^ (J.H.S.xol xvi., pt. i., p. 19) is as old as the French 
Expedition, where it appears in the Recherches G^ographiques of M. Boblaye 
(1836, p. 114). 

To Mr. Bosanquet's list of Maps I would add an interesting plan of 
Modon and Navarin, published in Venice, 1572 (Brit. Mus. Cat. S 132 (41)). 
It is based, I think, on local knowledge, gained, perhaps, immediately after the 
battle of Lepanto. The lagoon is treated as an inner harbour with a narrow 
entrance. Sphacteria (the name is not given) lies in front of the outer 
harbour, through not quite in the right position. Prote (Prodano) is placed 
rather too near. ' Navarino,' without the qualification of ' Vecchio,' is the 
name given to our Pylos. New Navarin was either not yet built when the 
information on which the map is based was procured, or was not firmly 
established enough to be given a name of its own by the Venetians. It was 
built at any rate by the end of 1572. 

Ronald M. Burrows.] 


... :■ ■ i / ( 



Polygraphisches Institut A.G. Zurich. 

J.H.S. V0L,XViri.a698) PL.XIH 



Polygraphisches Instilut AG. Zurich. 

[Plates XIL, XIII.]. 


1. JE 32.— AVTOKPATOPA I, TPA[lANON KAI . . .] r. 
Head of the Emperor, laureate, to right. 

Hev. AITEAinN 1., EKKAH[CI A ?] r., EMP (145) 1. in field. 
Goddess in double chiton seated to left, patera in her right hand, 
the left resting on the chair. In the exergue is the emblem of the 
town, a goat lying to left. 

Dr. H. Weber, London. PI. XII. No. 1. 

The era of the city begins in the autumn of 47 B.C. Hence this coin, 
which is struck in high relief, dates from the second year of the reign of 
Trajan (autumn 98-99 A.D.). The head is not a likeness. The inscription on 
the reverse identifies the goddess represented with the 'E k kXtj a t a of the 
Aigeaians, a personification which has not hitherto, as far as I know, occurred 
on coins. 

2. ^ 26.— MAP . lOV .ct)IAinnON EHI . KeC. Bust of 
the Caesar, laureate, to right, wearing cuirass and draped. 

iiev. ev . ni . ee . MA . Air l., etON NC r., BSC l. in the 
field. Tyche seated to left, wearing a turreted crown and veil. 
In her right, a small temple represented in profile, with a statue 
under the arch of the facade and an eagle on the pediment. At the 
feet of the goddess a goat lying to left, its head reverted. 

Coll. Giel in St. Petersburg. PI. XII. No. 2. 

em . K€C . stands for eTn^aveararov Kaiaapa} eV . fll . 9€ . MA 
for evyevwv wiarcov deo<^tXS>v MaAceSoycoi/.^ 


3. iE 21.— AYTO . KAI 0E . Yl . AOMI r, TIANOC CE . 
PEP. 1- Head of the Emperor, laureate, to right; behind it 
a star. 

' Cf. Griech. MUnzrn, ]>. 142, 422(.S;iloniiius). ^ Cf. luc. cit. \k 180, 549. 



Bcv. KAICAPEfiN rft i'rrpo'i) 1-, ANAZPABH {sic) r., 
ETOIyC B I P in two lines in the field. Elpis as goddess of the 
city stands to left, wearing turreted crown, double chiton, and cloak. 
In the right hand, which is raised, she holds a flower, and in the 
left, which is lowered, a fold of her robe. 

Inv. Waddington, No. 4111, PI. IX. 26. PI. XII. No. 3. 

Cf. the erroneous description of this specimen in V. Langlois, 
Revue Num. 1854, p. 9, 3, PI. I. 2, and A. de Longperier's correction, 
loc. cit. p. 137. 

A similar piece with the date | P published by Babelon in the Annuaire 
de la Soc. de Num. vii. 1883, p. 25, PI. II. G apparently also bears Elpis as the 
type of the reverse; instead of [(|)AAY]|0Z we should probably read 
[0E . Y]I0Z. 

Elpis again appears as goddess of the city, wearing a turreted crown, on 
cains of Alexandria in Egypt.^ 

4. iE 23.— AYT . KAI . NEP. TPAIANO . . . Head of the 
Emperor, laureate, to right. 

Rev. KAICAPEHN r., ^^ ANAZAPB[n] 1. Bust of Zeus (?) 
to right, laureate and draped, in the background the Acropolis rock, 
crowned by two buildings, one on right, one on left. Between them, 
above the bust, the date [ej] | oc KP (126). 

Berlin Museum. PI. XII. No. 4. 

A similar representation is seen on a coin with the portrait of Claudius.^ 
The only era of Anazarbos begins in the autumn of 19 B.C. Thus 

the date 126 corresponds to the year beginning autumn 107 and ending 

autumn 108 a.d. 


5. M 26.— AVTOKP . KAI . NEPYI (sic.) t., TPAIANOZ 
ZE . PEP . AA. 1. Head of the Emperor, laureate, to right. 

Rev. AYrOYZTANU)N r., ETOYZ oc n (86) 1. Bust of youth- 
ful Dionysos wreathed with ivy to right, on the breast robe or 
nebris, over the left shoulder thyrsos. Behind, kantharos. 

My collection. 

Similarly in Cat. Greppo No, 1093, with erroneous description, 
and Babelon, Annuaire de la Soc. de Num. vii. PI. II. 10 with the 
date n (?). This date is inadmissible for the. reason that Trajan did not 
assume the title ' Dacicus ' until the end of 102. As the era of Augusta 
begins in the autumn of 20 a.d., the year 86 runs from autumn 105 to 106. 

1 Poole, Alexandria, PI. VIII, 1620. ^ Monn. grecqucs. p. 349, 10, PI. F, 20. 



G. JE 28.— AY . K . no . Al . 1., OYAACPIANOC r. Bust of 
the Emperor with radiate crown, cuirass and drapery, to right. 

Ecv. AAM.MHTP.Tl., HCAAMW (tiBo^) r. Apollo standing 
to left, with chlamys on hia back and quiver on the right shoulder. 
In the left hand he holds a bow, and in the right, which is hanging 
down, a branch. 

My collection. PI. XII. No. 5. 

Hitherto only two coins of Lamos, bearing likenesses of Severus and 
Caracalla, have been published.^ The town lay on the river Lamos, somewhat 
to the east of Elaiusa Sebaste.^ 


ea. M 37. — AVTO . KAIC . MARK . 00 . . . Bust of 

Macrinus with laurel wreath, cuirass and cloak to right. 

Eev. MAA . lEP . TOY | GCOY AMc{)IAOXOV, in the field 1. 

€T.., r. ^ (year 284). Tyche seated to 1. with turreted crown 

and veil. In her right hand are ears of corn (?), the left rests on 
the rock. At the feet of the goddess are two river gods swimming 
one to left, the other to right. 

Mim. Chronicle 1897, Proceedings p. 6, where the date is given 


This is the first dated coin of Mallos. The era of the city may be the 
same as that of Mopsuestia, which begins in the autumn of 68 B.C., or it may 
be the one beginning a year later. The date 284 would thus correspond 
either to the year 217 (i.e. autumn of 216), the first year of Macrinus' reign, 
or (counting from 67 B.C.) to the autumn of 217 to 218. 

The mention of the Hieron of the deified Amphilochos, the founder of 
Mallos, also occurs here for the first time on coins. This sanctuary was the 
seat of an oracle, famed till late Roman times, which Alexander the Great, 
on his march to Issos, distinguished by ofifering sacrifice.^ Amphilochos as 
hero and seer is represented on various other coins of Mallos* as well as 
on a coin of Tarsos (v. No. 53). 

The two river-gods doubtless denote the two arms into which the 
Pyramos divides from Mallos to its mouth.^ 

^ Nouv. Annales de V Inst. ArcMol. II. p. 349, Numisinatiqibe, 1883, p. 95-98 and 126. 
PI. E. * Loc. cit. p. 118, 59 ; 119, 62 and 63 ; 120, 

2 Heberdey and "Wilhelm, Reisen in Kilikien, 66 and PI. VI, 38, 40 and 43. 
Wien, 1896, p. 47f. « Loc. cU. p. 94. Heberdey and Wilhelm, 

' Imhoof, Mallos, AnnxMire de la Soc. Fr. de Reisen in Kilikien, Wien, 1896, p. 9. 

M 2 


6b. JE 35.— AYT .... MARK . OHEA . Al A[A0VMENI- 

ANJoN I Ce. Bust of Diaduraenianus to left, draped. 

Ecv. M[AA . l]eP . nOA . | eeOV AM4)IA0X0V. Youth- 
ful Amphilochos nude, wearing boots, staading to left, with a branch 
in his right hand, and in left, drapery and sceptre. At his feet in 

front is a boar to left. In the field 1. [CT .], r, r --i (284). 

Lobbecke. PI. XII. No. 6. 

Cf. Inv. Waddington, No. 4369, with MIAlHPnOA (?) GEOY 
etc., eT . An . (the robe is described as a snake). 

The numeral sign C in the date appears to have been missed in the 
striking, either from some damage to the die or from carelessness on the part 
of the die-cutter. I have no suggestion to make for the completion of ffO A, 
which in Lobbecke's specimen takes the place of TOY on the previous coin. 
If A stands for A, TroXt? would be a possible reading. 

Selinus Traianopolis. 

The earliest known coinage of Selinus is that with the portrait of Queen 
lotape.^ With Trajan begins the coinage bearing portrait-heads of the 

7. M 32.— AVTO.KAI.MAP. r., AVPH. below, ANT... . 1- 
Bust of the youthful Caracalla r., laureate and draped. 

Rev. TPAIANO r. nOAeiTUN CCAI . THC leP L, AC r. 
Youthful god standing to front, with long hair, chiton and girdle, 
and a cloak hanging over his back. In the right hand, stretched 
out sideways, is a patera, and the left, raised, rests on a sceptre, on 
the top of which there seems to be a bird sitting to left. 

My collection. 

8. M 29.— AV. KAI . M. AV . ce. 1., AACXANAPOC r. 
Bust of the Emperor to right, laureate, and wearing a cloak. 

Rev. TPAIANO . C6AIN 1-, OV . THC ICPA r., C in ex- 
ergue. The same god to front, with patera and sceptre. On the 
right, at his feet, is a bird sitting to right, with head turned round 
and flapping wings. 

My collection. PI. XII. No. 7. 

Coins of the same type with portraits of Caracalla (?) and Philip are 
described by Boutkowski (who identifies the god wrongly, once as an Amazon 
and again as Diana,^) and also by Babelon in Inv. Waddington No. 4486, PI. 
XL 15, with the portrait of Macrinus. The type on the reverse undoubtedly 

• Griech. Munzen, p. 190, 580 ; Lcibbcckc, ^ j,jy Waddington, No. 4485. 

Zeitschr. f. Num. xvii, j.. 17, PI. II, 8. » Serrnre's Bull, dc Num. 1895, p. 3. 


rcprcscuts Apollo, and seems to be a replica of the Sidctic god with the 


0. tE 24. — Aegis with winged gorgoncion in the middle. 
Border of dots. Kdgc sloping. 

Rev. ZOAEfiN, below. Goddess with turreted crown riding 
on a bull which gallops to right and winch she holds by the horns. 
In the field 1., above g, r. N or K. 

Gr. 8. 30. 

My collection. — Of. V. Langlois and A. de Longpdricr, Revue 
num. 1854, p. 23 and 142. PI. IV. 27. 

10. M 25. — The same, 1., above [g] and behind the riding 
goddess an eagle standing to left, head reverted. 

Brit. Museum. T. Combe, PI. X. 17. PI. XII. No. 9. 

11. M ,26. — Similar, I., above fIJ, in front of the riding 
goddess an owl sitting to right. 

Gr. 14, 51. 

A.'Lobbecke. PI. XII. No. 8. 

A design similar to the 'rider' type of these coins may be seen on the 
silver coins of King Stasioikos. One of these a stater (gr. 11. 10) in my 
collection, is represented in PI. XII. No. lO. It is usual to interpret this 
Kyprian goddess as Aphrodite or Astarte^ and undoubtedly the riding goddess 
of Soloi (who is identified as goddess of the city by her turreted crown) 
should be interpreted in the same way, and not as the Greek Europa.' 

To these coins of the period of the Seleucidae correspond half-pieces of 
the same date : 

12. M 20. — Head of Athena to right. Border of dots. Edge 

Rev. ZOAEflN r- Bearded Dionysos with hull's horns, 
standing to front in a long chiton, kantharos in his right hand, the 
left on a thyrsos. In the field 1. (g) and |;^. Border of dots. 

Gr. 6. 40. 

My collection. PI. XII. No. 11. 

Copenhagen, Ramus i. p. 271, 1 described as Bacchus simply. 
Munich, Mionnet iii. 611, 344 as Zeus. 

^ Cf. Num. Chrcmiclc, 1897, PI. IX, 6 ; Achtmdnidcs, p. cxlvi. and Inv. AV.addington, 

Zcitschr. f. Num. -x.. (1883), PI. I. 2-4. No. 4840, describes tlie riding goddess as 

2 De Luynes, Num. Cyjtr. PI. V. 1 and 2 ; Artemis. 
Stephani Compte rendu, 1866, p. 101 ; Six, ^ pg Longperier ^c. c?'<. Balielon Inv. Wad 

Series Cypr. p. 348, ff. l)iit Babelon, Rois dington, No. 4501. 


Cambridge, Leake, Num. Hell. p. 123 called ' Bacchus in 
pointed cap ' and with two different monograms. 
Berlin, on 1,, ^ and | . 

The Bull-Bacchus, who, according to Inv. Waddington, No. 4524, seems 
to occur also on a bronze with Gordian, is a rare type on coins, and has 
hitherto been known only from coins of Skepsis.^ 

After the restoration of the ancient Soloi by the Kilikian emigrants who 
returned from Armenia, the city took the name Pomjjeiopolis, and a new era 
began in autumn 66 B.c.^ The reading ' Solopolis ' on coins given by Khell 
and Allier is founded on an error, that is to say, on arbitrary restoration of the 
initial letters of imperfectly preserved inscriptions. 

On the other hand it appears from the following rare and rudely executed 
copper coins, which were probably struck while the city was being rebuilt, 
that its name was at first, but only for a short time, IIofM7n]'ioi or Uofjurrjia. 

13. JE 22. — Head of Pompeius to right; behind AN. Border 
of dots. 

Rev. TTOMnHIANriN in a straight line r. Nike moving to 
right, with wreath and palm-branch ; in the field r. | A | N, 1- [S] 
Gr. 7. 37. 
My collection. 

14. M 23.— Similar; in the field r. A, NB, ^ (?). 
Gr. 8. 65. 

My collection. PI. XII. No. 12. 

These coins show the usual types of Pompeiopolis, but the name of the 
inhabitants takes the form Ylo^nr'qlavoL 

During the empire, down to the middle of the 2nd Century, Pompeiopolis 
appears to have coined very little. But afterwards, the year 229 (autumn 
163-4) is marked by a numerous and peculiar coinage, which includes, besides 
Concordia coins of the two emperors, Marcus and Lucius, a series of coins 
without portraits of the emperors. Of the latter kind I bring together the 
following : 

15. M 26.— TN . nOM 1-, TTHIOC r. Head of Pompeius to 

Bev. nOMrrHIOnO r. AeiTHN 0KC 1. Tyche seated to 1. 

with turreted crown and veil, a swimming river-god at her feet. The 

chair is ornamented with a sphinx. 

Paris. Mionnet IIL 612, 354. PI. XII. No. 14, 

Milan. Mus. Sanclementi, II. p. 1, PI. XIIL, 1. 

16. M 21. — Head of Pompeius to right. 

Bev. nOMfTHIOnO | AeiTHN and in the field r. 0KC. 
Bearded figure standing. 

1 Griech. Miinzen, p. 104/, PI. VIII, 6-9. 

2 Monnaies grecques, p. 362 ; Zeitschr, f. Num. x. (1883), p. 296 ff. 


Berlin, v. Prokesch-Osten, Coni2>ics rcndas dc la Sue Jr. de Num. 
vi. 1875, p. 245, 41. 

17. M liS. — 0KC 1. Nike witli wreath and palm-branch 
moving to lelt. 

Bcv. nOM[nHIO] 1, nOACITHN '•• Bearded figure standing 
to left, the upper part of the body nude, tlie right hand raised, the 
left at the side. 

Rollin and Fcuardent. PI. XII. No. 13. 

18. M 33. — Zeus Niltcphoros seated on throne to left, his left 
hand raised and resting on sceptre. 

Rev. nOMnHIOnO l; AeiinN OKC r. Bearded figure 
standing to front, the head, laureate, turned to the right, the right 
hand at the side ; wears boots ; the robe leaves tlie upper part of the 
body and the right arm nude, while a fold of it falls over the left 

My collection. PI. XII. No. 15. 

Paris. Mionnct III. C12, 851 ; Sestini, Lett. vii. p. CI, PI. IV. 7- 

Cat. Allier, p. 1)7. 

Inv. Waddington, No. 4515, where the statue is described as 

19. M 29.— nOMnHIOnOAeiTHN r-, OKC 1. Bust of 
Athena to r., with helmet and aegis. 

Rev. Apollo nude, standing to front, the liead to left, the legs 
crossed. In the right hand he holds a laurel branch, in the left the 
lyre, resting on a column. 

Florence. PI. XII. No. 16. 

Paris. Mionnet III. 612, 352. 

20. M 26.— GKC 1- Bust of the Stoic Chrysippos (?) to 
right, with cloak, the left hand touching his beard. 

Rev. nOMnHIO 1-, nOAeiTHN r. Bust of the poet and 
astronomer Aratos (?) to right, wearing cloak, looking upward. 

Paris. Mionnet III. 612, 353. 

Mus. Hunter, PI. XLIII. 23 (Vi.sconti, Icon. gr. PI. XXIII.) 
where the date, half obliterated, is wrongly given as OP. 

Brit. Mus. Zeitschr. fur Num. ix. p. 118 and 127, PI. IV. 12 and 
13, where the names of the portrait heads are interchanged. 
Imhoof, Portrdtkopfe, p. 89, PI. VIII. 31-32, cf. Schu.stcr, Fortrats 
der griech. Philosophcn, 1876, p. 22, PI. IV. 2. PI. XII. No. 17. 

Mus. Basel. 

From the correspondence of date and from the types in coins 15-20 it 
is safe to conclude that they were all struck to commemorate some great 
festival in the year 164, perhaps the dedication of some grand building, new 

IfiS F. IMIIOOF l!|J'.Mi:i!. 

or ncwly-rcstorcd, and adorned witli statues of famous men. Marcus is so 
well-known as a lover of learning and its representatives and as a generous 
patron of the sciences, that we may further conjecture that he himself 
provided the funds for the structure or its adornment. 

Similar examples of coins struck for special occasions or festivals 
arc described elsewhere^; it would be easy to add to the list. 

No commentary is necessary on the portrait-Jieads in No. 20, except to 
say that they have sometimes been called Chrysippos and Aratos, sometimes 
Aratos and Chrysippos.- I am inclined for the ])rcsent to prefer the former 
interpretation, because in anticpiity, as now, every unbiassed person must 
have recognised the meditative philosopher in the design on the obverse, and 
in that on the reverse the poet and astronomer with liis eyes raised to 

The standing figure in No. 18 is probably not, as Babelon supposes, 
meant for Chrysippos. The presence of the laurel wreath makes it more 
likely that the head represents Marcus Aurelius as Emperor and Philosopher. 
The statue on No. 17, on the other hand, may very well represent Chrysippos, 
for there was in the Kerameikos at Athens a statue of him, seated indeed, 
but with the hand stretched out.*^ 

21. M 32.— AVT . K . M . ANT. rOPAIANOC CeB., in 
the field H . | TT . Bust of the Emperor to right, with radiate 
crown and cuirass. 

Rev. nOMnHIOnO 1., ACITHN qT (306) A and in the 
field 1 ?". Apollo laureate, nude, standing to left, with chlamys, 
bow over the left shoulder, and boots. In the lowered right hand 
is a laurel branch, and in front of him a flaming altar. 

My collection. PI. XII. No. 18. 

Brit. Mus. Similar, without the altar. 

22. M 33.— Obverse, similar. 

i2fi7. nOMnHIOnOAeiTHN, in the field r. A^ and :^T (306). 
Bust to right, beardless, draped. 

Inv. Waddington, No. 4525, PI. XI. 18.^ PI. XII. No. 19. 

Babelon describes the bust as a portrait of Chrysippos ; in reality it 
represents neither Chrysippos nor Aratos, but some other celebrity of Soloi, 
perhaps (if we may rely on references like Strabo 671) Philemon. 

23. iE 32.— AVT .... ct)IAinnON eVCCCB., in the 
field n . I n . Bust of the emperor to right, with radiate crown 
and cuirass. 

' Griech. Miinzcn, p. 56-58. Scba.stopoli.s in ^ Cicero dcfin. bon. ct mal. I, 39. 

Pontos, No. 62-71. ♦ According to Arigoni II, PI. 40, 3G4, tliert 

■^ r!or<ko, Jnhrh. d. Arrh. husl. V, 1890 ; i.s a .similar coin with Gctu. 
lirll. J,. .-)6 — 58. 


Rrv. nOMnHIOnOA . IAT (Sll), klow 7. Bust of 
C'k)//sipj)t>s (?) <lr;iiHvl to ri^^lit, llic rij^lit liaud loucliiiif,' his cliiii, 
which is bearded. 

A. L()bbc-ckc. PI. XII. No. 20. 

Inv. Waddiiigtoii, N<.. 4r)2S, Tl. XI. 1!), where llie dale is 
iiicoricctly read as LAT and the portrait described as Arato.s. 

24. /E :}(). — AY . Ke . r . OVIBI . TP€Bn . TAAAON 
[Ceii.], in the field fl. | O. Bust of t^dhis to riglit, with 
radiate crown, cuirass, and cloak. 

Jiev. nOMnHIO 1., nOAITHN r., in the (ieid A|(;. 

Aiwllo as on No. 21 without the altar. 

My collection. 

Paris. Mionnet III. G15, 365, described as ' fcmmc dchmit iamnt 
unc hrdnchc' 

The signs A 7 do not stand for the date, but probably indicate the value 
of ' C assaria.* 


2-5. M 28. — Head of Antiochus IX. with diadem, slightly 
bearded, to right. Fillet-border. 

Eci: [BAJZIAEHZ | ANTIOXOV r., 4)IA0nAT0P0Z | 4* 
Al 1. The so-called ' Monuuicut of Sardanapalos ' with an eagle on 
the summit. 

Gr. 16. 50. 

My collection. PI. XIII. No. 1. 

Among the various examples of the tetradmchm of Antiochus IX. lately 
come to light, this one is distinguished by the execution of the portrait, 
which is unusually well modelled for the period. 

26. M 24. — Bust of Tychc, with turreted crown, veil, and 
earring, to right. Fillet-border. 

Bev. TAPZEflN r., A 1. Asiatic god to right, standing on the 
back of a winged lion with horns. 
Gr. 8. 70. 
My collection. 

This piece is overstruck on an example of the following coin of Adana. 
Ohv. Head of a goddess with veil r. 
Rev. AAANEHN r- Zeus Nikephoros seated 1. 

27. M 25. — Bust of Tyche to right, with turreted crown and 
veil. Border of dots. 

Rev. TAPZEHN I'-. AZK|rAY 1- The so-called 'Monument 
of Sardanapalos,' with an eagle on the top. Border of dots. 
Gr. 9. 05. 
My collection. PI. XIII. No. 2. 


Similar pieces have AlO. GEO . MAP. XAP . and CAN . ct)IAI . 1. in 
the field/ while as a rule coins with the types of Nos. 2G and 27 have only 

I have elsewhere given more detailed descriptions of the design on the 
monument,^ and two additional ones will be found under Nos. 28 and 29. 
This figure was formerly called Sandan, the Asiatic Hcrakles ; Babelon 
considers that it ought to be identified with Zeus of Doliche, whose worship 
was somewhat widely spread in late Roman times.^ But this new suggestion 
is not entirely satisfactory. 

The god whose image persisted almost unchanged on coins of Tarsos 
from the period of the Seleucidae to Gallienus is certainly a local divinity 
associated with the city from the most primitive times, and there is no 
evidence to prove his identity or even relationship to the Syrian god in 
Doliche. It was not until shortly before tlie breaking up of the old religions 
that the cult of Jupiter Dolichenus seems to have gained much ground. 
Hence we are not justified in giving to a purely Tarsian representation the 
name of a composite divinity of late Roman date with characteristics foreign 
to the ideas of the ancient Kilikians. Rather we must be content for the 
present to use a purely general name such as ' Asiatic god.' For the figures 
on old Asiatic monuments which are most nearly approached by the type of 
Tarsos, as, for instance, those on the Babylonian cylinders * and on the rock- 
reliefs of Jasilikaia near Bogazkoi (north of Tavion) ^ have not yet been 
explained with any certainty. 

The ' Dolichenus ' of Roman times, a figure of Zeus usually standing on a 
bull, should probably with more reason be referred back to one of the Syrian 
gods which appear with various attributes, and sometimes seated or standing 
between two bulls, on coins of Antiochos XII.,° ofRosos,^ Hieropolis,^ Gabala,* 
and Dion.i^ 

28. M 20.— AVT • KAI . 0E . TPA . OAP . VI . 0E . NEP . 

1 Revue mim., 1854, PI. IV. 29; Inv. "Wad- « Monn. grecquci, p. 437, 121, PI. H 15. 

dington, Nos. 4609 and 4610 ; Brit. Miis. Strangely enough, Leo Bloch, in Reseller's 

^ Monn, grecques, p. 366,54-57, PI. F 23-25; Lexikon, Kora p. 1314, maintains that this 

433, 96; 435, 112-113; PI. H 14 ; Tier- unci bearded figure is female, and represents a draped 

PJlanzenbilder, p. 70, PI. XII. 7-9. Isis. 

3 Lcs Rois de Syrie, p. clvi.ff. Cf. E. Meyer, ^ j^y^. cit. p. 440, 8 PI. VII. 223. 

Roscher's Lexikon, I. p. 119111. According to « Gricch. Miinzcn, p. 235, 772-775, PI. XIV. 

Meyer, the cult of Dolichenns can only be 7. 

traced back to the middle of tlie 2nd century ' Nouv. flal. Myth. p. 89, PI. XIV. 16 ; La- 

A.i). and was of ephemeral significance. jard, Cultedc Vinus, PI. V. 5. 

* Raoul-Rochette I'Hcrcule Assyriev, 1848, '** Lajard, loc. cit. PI. IV. 6 ; de Saulcy, 

PI. IV. 16 and 17 ; Lajard, Cnltc de V^nus, Num. dc Terre-sainte, PI. XIX. 9. Cf. 

1849, PI. IV. 11-12. also Woltcrs, Zeux IleliopoUfcs, Avier. Journ. 

5 Perrot and Chipiez. Hist, de VArl, IV of Arch, vi, 1890, p. 65-8 (illustration); 

(1890), PI. VIII. E, and No. 6.37, where one of ^i\\Ainczk2L,BildwerkeausCarnuntum,Archaeol, 

the gods who stand on lions ajipciars with a -c^ngr. Miith. aics Oester.-Ungarn, viii, p. 64, 

sword at his side and a doulde axe anil staff in I'l. II. 
his hand. 


Bust of Hadrian, laureate, with cuirass and drapery to right. 


Bearded god standing to right on the back of a horned lion. 
He is dressed in a short chiton and cloak ; on his head is a Persian 
tiara crowned witli a top-piece ; behind the left shoulder is a quiver 
in front of which is a projection like a bow. At the left side is a 
sword, and in the left hand a double axe and wreath. The right 
hand is stretched out. Countermark with the head of an emperor, 
laureate, to right, and A | M ( K uniler it. 

Gr. 9. 75. 

My collection. 

This type, which has just been discussed, and occurs in different varieties 
of die on silver coins of Hadrian, is well illustrated in Imhoof and Keller's 
' Tier- und P/Janzenbildcr; PI. XII. 8.i 

The head in the countermark appears to represent Caracalla ; the three 
letters are the initials of the well-known titles of the city, irpcorr}, fieyicrTtj, 

As a variety of the ' Monument ' figure the following is also remark- 
able : — 

29. M 32.— AY . KAI . r . MCCC . KYIN . AeKlOC TPAIA- 
NOC, in the field FT | TT. Bust of the Emperor to right, with radiate 
crown, cuirass, and cloak. 

Hev. TAPCOV h, MHTPOnOAenC r., in the field r. 
A . M . K . r . B . The type is similar to that of No. 28, but instead 
of standing on the lion, the god is mounting the animal from behind 
by placing his left foot on its back. 

Coll. Gonzenbach, St. Gall. PI. XIII. No. 3. 

To judge by coins of the Empire, Apollo Lylceios (or Tarseits) and Perseus 
were two of the divinities whose cults enjoyed most prestige in Tarsos. Tiiey 
are often represented together. The figure of Apollo is usually of archaic 
style. He is nude and stanjds to front with the legs close together. His long 
hair falls sideways over his shoulders, and his head is adorned with a laurel 
wreath. The god stands on the Om2')halos, on either side of which lies a bull ; 
in his hands he grasps the fore-legs of two wolves that are standing up on 
their hind legs on either side of him. Sometimes a high column occurs as a 
basis for this group. 

The creatures which the god holds by the fore-legs are not deer or 
antelopes, as was formerly assumed, nor are they greyhounds, as Babelon ^ 
conjectures, but ivolvcs. This is placed beyond a doubt by the form of the 

^ Cf. also de Luynes, Niom. des Satrapies, PI. Bull, de Corr. Hell, vii. (1883), p. 283 ff. 
VII. 1. 3 Inv. Waddington, Nos. 4638, 4659. 

' Monn, grecques p. 351 ; Waddinyton, 


\on<f tail represented, as a rule, not turned upwards as in a dog, but hanging 
down.^ Add to tins that the wolf is one of the well-known and wide-spread 
symbols or attributes of Apollo, while the dog is not. 

The cultus image of Apollo occurs on coins from Hadrian to Gallicnus. 
Sometimes the omphalos is indicated merely by a small arch, without tlie 
attendant bulls, and on some late coins the archaic character of the statue 
has been missed through lack of skill on the part of the die-engraver. 

As the following list will show, the statue of Apollo often appears 
erected before Perseus sacrificing, or as an attribute of Perseus. Perseus was 
represented in various ways as founder and hero of the city,- and was honoured 
as ^orjdo^ (No. 41 and 42) and irmpwo'i (No. 48). 

I. — Apollo. 

1. The O'i'ltns iinrigc on the column. 
See below, Perseus, Nos. 45-47. 

2. The Cultus imafje vnthout the column. 

30. ^ 31— (j)AYCTeiNA 1, CGBAC FH r. Bust of the 
younger Faustina to left. 

image with the two wolves to front standing on the omphalos 
between two recumbent bulls. 


Mionnet iii. G27, 435 ; Lajard, licchcrchcs sur Ic cnltc dc VSnus, 
p. 70, PI. V. 1 and Archacol. Zcitung, 1854, p. 215; Overbcck, /oc. cit. 
Coin-plate I. 30. 

31. M S2.—0hv. Macrinus. 

liev. [ceVH. MAKPeiNIANHC | MH]TP0 . TAPCOY and in 
the field [A .] M . K . The cultus image standing on the omphalos, 
head to left. 

Berlin. Zeitschr. filr Num. viii. p. 10, PI. II. 6. 

32. iE 29.— AVT . KA. M . AVP . ANTHNeiNOC. Bust of 
Elagabalus, laureate, to right with cuirass and drapery. 

Rev. TAPCOV I MHTPO. The cultus image on the omphalos 
to front; in the field two stars. 

Library of Bologna. 1*1. XIII. No. 4. 

Num. Chronicle, 1873, p. 35 (incorrect). 

' Of. our illustrations and Ovcrbeck, Apollon, p. 29, PI. L 30 and 3L 
^ Cf. Eckhel Num. vet. anccdoti, p. 80. 


33. yE 3.S.— A . K . M . A . CCOY . AAeiANAPOC CGB and 
in the field n . i FF . Pmst of the Emperor to riglit, huueute, 

llcv. AACXAHAPAM. C |eO . AA . MHT . TAPCOV. and in 
the field r. A. M . K., 1. T . FF B .' The callus image to front, liead 
to left. 

Rollin and Feuardent. PI. XIII. No. 5. 

Overbeck he. cit. Coin-plate I. 31. Paris of. Mionnet iii. 038, 
49G and 497 (with ' deer.') Brit. Museum. 

34. i^ 37.— AYT. K. r . lOV. OVH . MAXIMeiNOC and 
in the field n . | n . Bust of the Emperor to right, with radiate 
crown, draped. 

Eev. TAPCOV THC | MHTPOnOAeU)C and in the field 1. 
A.M.K., r. r.B. The cultus image to front. 

Vienna. Frolich, Qnatuoo' Tcntamina, p. 318; Cat. Mus. Caes., 
129, 11 ; Mionnet, Siqipl. vii. 276, 481, with 'dogs' or 'deer.' 

35. M 37.— Similar. 

Rev. Inscription the same. The cultus image on the 
omphalos to front, head 1., in the r, hand a wolf, in the 1. hand 
a bow. 

My collection. PI. XIII. No. 6. 

Cf. Mionnet, iii. 640, 509. 

Brit. Museum, which also possesses the same type on a coin 
with Balbinus. 

36. M 30.— ANNIAN AITPOVCKl AAAN CC. Bust of 
Etruscilla to right, with crescent at shoulders. 

Rev. TAPCOY MH|TPOnOAeaC and in field 1. A.M.K., 
r. r . B . Cultus image on the omphalos, head to r. 

Brit. Museum. Cf Haym, Thcs. Brit. ii. Pi. XLVII. 4, 
' Artemis with two stags ' ; Mionnet, iii. 653, 595, ' with dogs.' 

37. ^ 30.— AVT . K . n. A . OVAAePIANON CC and in the 
field n . I n . Bust of the Emperor to right, laureate and 

Rev. TAPCOV MHT | POnOACnC, in the field 1. A . M .K ., 
r. r.B. The cultus image on the omphalos turned slightly to 
left, head to right. 

Mn the first word of the inscription either I ""ev. (elsewhere unusual) stands for irpoKa- 
has dropped out after P or we should rea.l ^^f-M*",,. v. Waddington, Bull, de Corr. Hell. 

' iLx t i I y\ ^^ I i\ ■ c vii. p. 285. Whether this initial has the 

h\i^av6p(iavfi) \v{Twveiviavi)), as on a coin of • •/. , . . , 

fKc \i^\¥ ivr,,„„ -fi <.u same signihcance here IS uncertain, because the 

the lint. Museum with the same emperor ° 

and the type of Pallas. On eontein- letters T . €17 (apx'^") are absent, 
porary coins the letter fF i» the field of the 


Vienna. Mus. Theup. p. 1081/2; Mionnet Sujjpl vii. 287, 531 ; 
the animals are called dogs or deer. 

Cat. Greppo, No. 1106, described as nude Hekate with dogs. 

3. The Emperor sacrificing hcforc the cuUus image. 

and in the field n . I n.. Bust of Caracalla to right, laureate and 

licv. ANTHNeiNI ANHC C6VHP . AAPI . TAPCOV and in the 
field A . M . K . The cultus image standing to front on the 
omphalos between bulls' heads, with the two wolves ; beside it the 
Emperor in toga, standing to left before a flaming altar, holding a 
patera in his right hand. 

Palis. Mionnet, iii. 632, 465. PI. XIII. No. 7. 

Cf. De Witte, Cat. Greppo, p. 151, No. 1106, 'Apollo with 

4. Other scenes of sacrifice before the cultus image. 
See below, Perseus, No. 45-47. 

5. The cultus image as attrihUe of Perseus. 

See below, Perseus No. 41 and 42, No. 48 and 49. 

II. — Perseus. 

1. Perseus with harpe. 

39. M 26.— Obv. Hadrian. 

Rev. TAPCEnN MHTPOnOAEnC. Perseus, nude, with 
winged sandals, stands to left, the harpe and drapery in his left. 
hand ; he holds out his right hand to Apollo, who is nude and 
standing to right with crossed legs, leaning on the tripod. A laurel- 
branch is in Apollo's left hand, and between the legs of the tripod 
rears a snake. 

Gr. 9. 68. 

The Hague. Imhoof, Zeitschr.f Num. iii. p. 333, 1, PI. IX. 3; 
Overbeck loc. cit. Apiollon ; Coin-plate V. 16. 

2. Perseus with harpe and Gorgoneion. 

40. M 35.— AVT . KAIC . A . KAIA . BAABeiNOC CCB . 
and in the field TT . | TT . Bust of the emperor laureate to right 
with cloak. 


Rev. TAPCOY M|HTPOnOA€nC and in the field 1. A . K ., 
r. M . B . r . Perseus as on No. 39 except that he holds the winged 
Gorgoncion in his lowered right hand. 

Cat. Gruau, No. 1053. Sabatier, Revue Num. Beige 18G5, 
PI. XVIII. 25. 

Similar with Maximinus, Mionnet iii. C40, 510, and with 
Gordian, Mionnet iii. 644, 534. 

3. Pcrse^ts with harpe and Cultus-image of Apollo. 

41. ^ 27.— AAPIANHC TAPCEnN. Head of the bearded 
Herakles crowned with oak-leaves to right ; club behind the 

Rev. MH|TP 1,, OnOAEnC r. Perseus, nude, with winged 
sandals, standing to left, harpe and drapery in his left hand, and on 
his outstretched right hand the cultus image of Apollo (with the 
two wolves) standing to front on the omphalos. In the field 1. 
BOH I 00 Y, and below, at the feet of the hero, a bull to left 
attacked by a lion from the side. 

Gr. 19. 52. 
. My collection. PI. XIII. No. 8. 


Cf. Mionnet iii. 623, 417; Leake As. Gr. 129, 1; Inv. 
Waddington No. 4625-7 PI. XII. 6, the animal group not 

42. M 29.— Similar, with AAPIA|NHi:TAPCEnN and 

Gr. 11. 65. 

My collection. 

Brit. Mus. T. Combe PI. X. 18 = Mionnet Sup)pl vii. 259, 405. 

The attribute in the right hand of the Perseus and the animal group at 
his feet seem hitherto to have escaped notice on all known coins of this type.^ 
The representation of the fight of the lion and bull is identical with that on 
the silver coins of Hadrian ; ^ on later coins the bull appears seized from 
behind.^ The fight is certainly meant to be symbolical, and must be brought 
into connection with Perseus as Helper (^orjOof;). 

43. iE36.— [AVT.] K . M . ANT . TOPAI ANOC CCB. and 

in the field TT . | IT . Bust of the Emperor to right, with radiate 
crown, shield and spear. 

1 Cf. Monatsblatt d. num. Oesellsch. Wien, xiii, No. 9 (Gr. 10. 65, iny collection). 

1896, No. 156, p. 379/80, Inv. Waddington, » Examples with Gordian, Mionnet III. 645, 

No. 4625, where the cultus-image is recognised. 543-47 ; with Decius, Mionnet III. 652, 590-1, 

^ Mionnet Suppl. vii. PI. VII. 4 ; de Luyncs Rev. 7mm. 1854, PI. VIII. 50. 
Num. dcs Satrapies, PI. VII. 7 ; and here PI, 


Rev. TAPCOV MHT|POnOAenC and in tlic field 1. 

M . A . I K ,, r. B . I r . Perseus Avith the cultus image as on 

No 41 ; in addition to the haipe he holds a fishing basket in his 
left hand. 

Berlin. PI. XIII. No. lO. 

44. M 35.— AV KAI . r . OVIBION TPIBH . TAAAON and 
in the field FI . | TT . Bust of Gallus to right, with radiate crown, 
cuirass and cloak. 

Rev. TAPCOV MH | TPOnOAenC and in the field 1. 
A . M . K ., r. B . r . Perseus with the cultus inmge as on No. 41. 

My collection. PI. XIII. No. 11. 

Paris. Mionnet Siqipl. vii. 286, o28 (and 529 after Sestini) 
whei-e the cultus-image is described as ' two small figures on a 
prow ' or as ' bow inverted ' and the drapery is mistaken for a 


4. Perseus sacrificing before the Cidins image. 

45. JE 38.— AVT . KAI . A . Cen . CGYHPOC nCP . and in 
the field H . | FT . Bust of the Emperor, laureate, to right, with 

in the field above . f . B . The cultus image of Apollo with the 
wolves stands to front on a column on the omphalos between 
recumbent bulls. To the right of the image is a flaming altar, in 
front of which stands Perseus to left, with curly hair, drapery round 
the hips and left arm, patera in his right hand and harpe in his left. 

Waddington. PI. XIII. No. 12. 

Cf. Inv. Waddington No. 4G38, where Perseus is wrongly 
described as an emperor. 

46. M 40.— AVT. K . ANT . rOPAIANOC CeBA. and in 
the field n . I FF . Bust of the emperor to right, with radiate 
crown, cuirass and cloak. 

Rev. TAPCO|V MHTPOnOAenC and in the field, A.M. K.B. 
r., r . 1. An altar with a zebu-ox lying to left in front of it. Behind the 
altar are visible the upper parts of two draped figures with heads 
to right, between whom towers a high column bearing the cultus 
image of Apollo and the wolves. At the right side of the altar stands 
Perseus to left, wearing drapery on his hips and arm, and winged 
sandals on his feet, and holding in the left hand harpe and in the 
right patera. At the left side of the altar stands the goddess of the 
city to right, in a long robe and wearing a turreted crown (?) ; both 
her arms are raised to the cultus image. 

Milan, Brera, Mus. Samlementi III. p. 85, PI. XXXII. 337. 

PI. XIII. No. 13. 


Nam. Ckr,Ni. I.S7:}, :U;, ( f. Iiiv. \V;ul(liii<;t()M, No. UuW, I'l. XII. 

Vuiiciiits of this rcprcseiitiiLion (up lill now wupublisln'd) iii;iy 
bo seen in Inv. Wadilington, No. 405'), with the lu-.-ul of Sev. 
Alexaiiclcr, and in the following example. 

47. /K sri.— AV . Ke . r . MCC . KOV. AGKIOC TPAIANOC 
€V . eVCee . -'ukI in the field fl . | ff . Bust of Decius t(j right, 
with radiate crown, ruirass and cloak. 

Jtcv. TAPCOY MHTPOnOAenC and in the field r. A . M . K ., 
in the exergue T . B . An altar with a zebu-ox lying to left in 
front of it; on the left, beside it, a high column supporting the 
cultus image of Apollo. Behind the altar is a standing figure, the 
head to left ; tlie right hand rests on a spear and there seems to be 
a shield on the left arm. To the right of the altar stands Perseus 
to the h.'ft, nude, holding in the left hand a harpe and drapery and 
in the right a patera. To the left of the column Tychc with 
turreted crown stands to right with both arms raised. 

My collection. PI. XIII. No. 14. 

Cf. Num. Citron. 1873, 3G, with a fanciful description of the 

The g(Kldess of the city standing before a sanctuary in the attitude of 
prayer occurs again on a coin of Tyros.^ 

5. Perseus and the Fisherman. 

4S. M 38.— [AVT . K . M .] AVP . CEV. AAEIANAP 

and in the field FT. | Ff . Bust of the Emperor to right, laureate, 

P^iv. AAeiANAPlAN . A | N . AA (?)... . In the field 
above A . M . K ., at the sides r . | B . and in tlie middle 
nAlTPnlOC. Perseus, nude, with winged sandals, standing to 
right. His long cloak is fastened round his neck and hangs down 
behind, covering his back. In his right hand he holds the harpe, 
and in his left, which is raised, the cultus-image of A})ollo and the 
wolves. Opposite the hero stands a bearded fisherman in a short 
chiton. The figure is turned slightly to right and the head to left. 
A fishintr rod is in his outstretched hands, with a fishing basket 
hanging at the upper end and a large fish at the lower. 

Coll. Waddington. PI, XIII. No. 15. 

Cf. Inv. Waddington, No. 4654, PI. XII. 10. 

49. JE 37.— AVT . k . ANT . TOPAIANOC CGB . and in the 
field n . I n . Bust of the Emperor to right, with radiate crown, 
cuirass and cloak. 

» Imhoof, Aniike Miin-Mldcr'm tlie Jahrb. d. archdol. Inst. iii. (1888), p. 286, (I'l. 9, 4). 
U.S. — VOL. XVI 1 1. N 


Mev. T APCOV I M I H 1., TPOFTOAenC r., and in the field, in tlie 
middle, A . M . K ., 1- B . andr. r .. Similar group, only the fisherman 
standing to left is beardless and of relatively smaller stature than 
Perseus. The little cultus image has no distinguishable attribute. 

My collection. PI. XIII. No. 16. 

Brit. Museum, with Bf on left and the small cultus image with 
distinct wolves. 

Paris. Mionnet III. 647, 561, described as with two fish ; cf. 
Mionnet Suppl. VII. 283, 512 (after Vaillant) and 513 from a 
bad illustration in Gessner, Imjpp. CLXXIII. 24, and p. 730, 206 
bis ; Leake, As. Or. 130, 1. 

50. iE 36.— AV. KAI . r . MeC. KVIN . A€KI0C TPAIA- 
N C and in the field n . I FT . Bust of Decius with radiate crown 
cuirass and cloak to right. 

Rev. TAP|COV MHTPOnOAenC, in the field above A. M . K. 
and in the exergue r . B . Perseus nude, with winged sandals, 
standing to left, in an attitude of surprise, raising his right hand to 
his mouth, and holding in his left harpe and drapery. Opposite the 
hero stands a bearded fisherman to right, in short chiton and boots. 
He holds in his left hand, over his shoulder, a fishing-rod and basket, 
and in his right a large fish. 

My collection. PI. XIII. No. 17. 

Cf. Mionnet III. 652, 587 and Leake SuiJi^l. 100 ; both call the 
object in Perseus' right hand a gorgoneion. Cf. also Sabatier in 
Rev. Num. Beige. 1860, PI. V. 5, where the K in the field is mistaken 
for a small Nike, the fish for an altar, and the fisherman for the 

Cavedoni's view is that the type of the coins No. 48-60 represents a meeting 
of Perseus and the fisherman Diktys, who drew Danae and her son from the 
sea with his net, and was afterwards made king of Seriphos.^ It is, however, 
obviously much more likely that the scene represented is some local legend 
of Tarsos, but as this legend has not been otherwise handed down, attempts 
at explanation of the types would be useless. It is worthy of notice that the 
figure of Perseus on No. 50 seems to express by his attitude surprise at the 
offering of the fish, and that in No. 43 he has taken over the fisherman's 
basket as an attribute. 

The Kronos type on coins of Tarsos, of which the following piece is an 
example, is easy to distinguish from Perseus. 

51. ^ 33.— AV. KAI . n. Al . OVAACPIANON C€ . Bust 
of the Emperor to right, with radiate crown, cuirass, and cloak. 

Rev. TAPCO I V MHTPOnOACa | C, in the field 1. A . M . K ., 
r. r . B . Kronos bearded, walking to left. His r. breast and arm 

^ Spicilegio. num. p. 211. 


are bare, and on his lieacl (which is turned to right) is a small crown 
with three peaks. His robe is ilrawn over the back of his head, in 
his outstretched right hand he holds the harpe, and on the left hand 
is an imperfectly defined object, 

Miis. Athen. No. 5800. PI. XIII. No. 18. 

Cf. Ro.scher's Lexikon ii. p. 1558, Fig. 8 where the drapery and 
crown on the head of Kronos have passed unnoticed. 

The worship of Kronos in Kilikia is proved by Stephanus Byz. a.v. 
" Khava, and by coins of Mallos and Flaviopolis.^ 

52. iE 33.— AV . KAI . noV . Al .GVAACPIANGC CC 
and in the field FF . I FF . Bust of the emperor to right, with radiate 
crown,, and cloak. 

-Rev. TAPCO |V MHT I POnOA and in the exergue enC . 
A . M . K . r . r . Three nude beardless male figures standing side 
by side to front. The two to the left have the head to right, the 
one to the right has the head to left. Each figure with one hand 
places a crown (turreted crown ?) on his head. The middle figure 
holds in the left hand a palm -branch which rests on his shoulder, 
the two others also hold each a palm branch in the lowered hand. 

My collection. PI. XIII. No. 19. 

Cf. Mionnet iii. 655, 611 and Suiypl. vii. 289, 543 = Sestini, Mus. 
Ucd. ii. 299, 44; Cat. Moustier, No. 3163. 

The attitude of the three men is identical with that of the victors in 
festival games. As other coins of the time of Valerian show three prize 
vases with the inscriptions KOPAIA, AVrOVCTIA and AKTIA,^ it is likely 
that the three figures represent the three victors in the games. 

53. M 33. — From the same die as the preceding. 

Rev. TAPCOV M|HTPOnOAenC, in the field 1. A . M . K., 
r. r . r . Amphilochos, in a short tunic and boots, standing to left, 
a branch in the right hand, and in the left a sceptre and cloak. In 
front of him a boar walking to 1. 

Lobbecke. PI. XIII. No. 20. 

The explanation of this type is to be found in representations of the seer 
Amphilochos on coins of Mallos.^ 

MHTPOn|OAIC. The Tyche of Tarsos seated to left with 
turreted crown and veil. In her right hand are two ears of corn 

^ Imlioof in Roscher's Lexikon, ii. 1572 — ' Imhoof, Mallos (Annuairc de la Sac. fr. tU 

1573. Nxim., 1883) p. 118, 59 ; 119, 62 & 63, PI. vi. 

■■' Mionnet, iii. 656, 615 and Suppl. vii. 290, 38 & 43, and in the present article above, No. 
545 & 546 ; Pelleriu, Ecaceil, iii. p. 260 (illustr). 6b. 

N 2 


and a poppy head ; her left hand rests on the scat, which is adorned 
witli a griffin standing to left. At the feet of the <;oddess is the 
upper part of a figure (the river Kydnos) swimming to left, and 
turning his head, wliich is wreathed with sedge, towards Tyciio. 

A wreath or bandeau with hanging fillets, and set round the 
outside with eight heads. Of these three male and one female are 
turned to left while two male and two female are turned to right. 
The male heads are apparently all bearded and without crowns. 

Gr. 14. 80. 

My collection. PL XIII. No. 21. 

This example (similar to Mionnet Sitppl. vii. 257, 394 = Babelon 
Anniuiire de Num. vii. 1883, p. 24, PI. II. 5, where Tyche is called Kybelc 
and the eight heads are conjectured to be tliose of divinities, perhaps 
goddesses of the cities belonging to the kolvov) was first made known by 
^dhdiiiHY in t\\e Rcvae Num. Beige 18GU, PI. V. 4 ( = Cat. Greau, No. 1945), 
and this notice was plagiarized by Boutkowski in his Didionnairc Num. i. 
p. 1487, No. 2484. Sabatier in his illustration represented the male heads 
as laureate, and called them (counting from left to riglit) Sabina, Hadrian, 
Pius, M. Aurelius, L. Verus, Commodus, Faustina and Crispina. Cohen (on 
the other hand), in the Cat. Grean, supposed them to be Pius, M. Aurelius, 
L. Verus, Commodus, Severus (with radiate crown), Faustina, Crispina and 
Domna. The new phototype shews that the bearded heads are not crowned. 
The coiffure of the female heads is that which is known from portraits of the 
younger Faustina, Lucilla and Crispina. The series as a whole gives the 
impression of being intended for portraits of the Antoninc family. Though 
the size is too small to admit the possibility of a real likeness, yet certain 
individual characteristics can be traced. 

Another coin of Tarsos with the wreath, of which a good example is 
preserved, makes it evident that the heads represent neither goddesses of the 
city nor other divinities.^ 

55. ^ 32.— AVT . KAI . M . AVP . ANTHNeiNOC . Bust 
of Elagabalus, laureate, to right, with cuirass and cloak. 

Rev. TAPCOV THC MHTPOnOA€nC . and in the exergue 
A . M . K . An altar wreathed ; over it a large wreath ; to the 
right beside it a bandeau with fillets thus adorned : I. two boys' 
heads to right ; then T, female (?) bust to left, beardless head 
laureate to left and similar head laureate to right ; then B, and 
two male busts draped, to right, in all seven heads. 

My collection. PL XIII. No. 22. 

Cf. Mionnet iii. 637, 491 with nine heads; 492 with two bandeaux, 
each with six or seven heads; Suppl. vii. 274,408, with seven heads; 

' Sec Waddingtou Bull, de Corr. IJell. vii. 1883, p. 28G 7. 



Inv. W;i(l(liiigtoii No. 4046 PI. XII. 1) with six heads, ('f. also tho 
coins with Rlaxiimis and si.x heads, Inv. Waddington No. 4001, 
PI. XII. 11; with ({ordian and two circles, cacli with seven licads, 
Mionnot iii. 04(1, 54.S (Pellerin M></. ii. I'l. XXXI. 5), Inv. Wadding- 
ton No. 4(i()(S and Tnany others. 

The head in the niiddK; of No. 55 may possibly be nu;ant lor Elagabaliis ; 
lor tlie others no names can be conject\jred. 

On a coin of the time of Volnsianns lettci-s stand above ;ind between the 
eijfht small busts of the bandeau ; ^ but they do not help to interpret the 
heads as they seem to be the usual series of initials A.M.K. r.B.r.€. 
(the two last possibly standinj^ lor y' iTrap-^Loiv) and partially to repeal, tlu; 
inner inscription of the bandeau. 

F. Imiioof-Blumek. 

irinlr.ii.hur, May, 18!»8. 


A?" — Poinjjuiopolis. No. 21-24 

Aegis witli gorgoiicion — Soloi 9-11 

Akropoli.s — AnazarLos 4 

Ain])liilochos — Mallos 6^ Tiirsos 53 

Antioclios ix. — Tarsos 25 

Aphrodite on a bull — Soloi 9 IF., Stasioikos PI. 

XII. 10 
Apollo, standing — Lanios 6, Solinus 7, 8, Poni- 

pciopolis 19, 21, 24 
Lykfit).s or Tarseu.s (iniltus image) — Tarsos 

30 ff., 39 
Aratos, ])ortrait — Pompoiojiolis 20 
^01)66^ (Perseus)— Tarsos 41, 42 
Bull Bakilios v. Dionysos 
Chrysij)pos, statue — Poiupciopolis 17 

jiortrait — Ponipeiopolis 20, 23 

Countermark — Tarsos 28 
Dionysos, bust — Augusta 5 

liearded and horned — Soloi 12 

Doliehenus — Tarsos 27 I". 

'EwKArjcn'o — Aigeai 1 

Elpis — Anazarbos 3 

"E.iri<pav(aTaToi Kdlaap — Aige;ii 2 

Era of Autumn C8 or 69 B.C. Mallos 6' 

,, ,, 66 li.c. Pom|ieiopolis 12 

,, ,, 47 B.C. Aigeai 1 

,, ,, 19 B.C. Anazarbos 3 f. 

,, ,, 20 A.D. Augusta 5 

i.h-yivu)v TTtaTuv 6to<piK(iv MaKtSouuv — Aigeai 2 

Festival coins — Pompeiopolis 15-20 

Festival games, Victors in — Tarsos 52 

Fisherman and Perseus — Tarsos 48-50 

r . E ("■"PX''''*') — Tarsos 55 

i'epek ttSKis (?) 0eov ' Afj.(tii\6xov — Mallos 6'' 

Upov Tov Btov ' Afj.<pLK6xov — Mallos 6* 

Kronos, standing — Tarsos 51 

Lion and bull — Tarsos 41,42 

Marcus Aurelius (?), statue — Pompeiopolis 18 

fj.7iTp6Tro\ii rrjs AajucuTi5os — Lamos 6 

Overstriking- Tarsos 26 

irarpipos (Perseus)— Tarsos 48 

Perseus— Tarsos 39 ff. 

and fislierman — Tarsos 48-50 

Philemon (?), portrait — Pompeiopolis 22 

no/uTrrjiof 01— Pompeiopolis 12, 13 

no/uir->7tos— Pompeiopolis 15 

Pompeius, portrait — Pompcipolis 13-16 

nipoKadfCofiffri ?) — Tarsos 33 

Sacrifice, scenes of — Tarsos 38, 45-47 

Sardanapalos, so-called monument of — Tar.sos 

Stasioikos — 11 f. 

Tyche, seated — Aigeai 2, Mallos 6-^, Pompeio- 
polis 15, Tarsos 54 

adoring — Tarsos 46, 47 

Wreath adorned with human heads — Tarsos 
54 ff: 

' Imhoof in ZcilRchr. f. Num. iii. 342, 23. 


Considering the attention now paid to the geography ot Asia Minor, it 
has struck me that a collection of the notices relating to the Arabic invasions 
of that district, which are scattered here and there in the Arabic annalists and 
must be sought through thousands of pages of Arabic print, would serve a 
very useful purpose. These extracts not only throw light on geography and 
the Arabic nomenclature of the localities, but, when compared with the 
accounts of the same events in Greek and Syriac writers, are of great value 
for the study of chronology. 

The writers from whom extracts are given under years are the following : — 

(1) The chronicler known as Ibn Wadhich or Al Ya'kubi, who wrote about 
900. (ed. Houtsma. Leiden, 1883). 

(2) Al Tabari d. 923. (ed. Barth and others. Leiden, 1879, &c.). 

(3) The Khitab Al ' Uyun (Book of Springs).^ (ed. do Goeje. Leiden, 
1871). This work, though dating not earlier than the middle of the 11th 
century, preserves several valuable notices relating to this period. 

(4) Ibn Al Athir (d. 1232). (ed. Tornberg. Leiden, 1851, &c.). This 
author generally copies Al Tabari, but occasionally has notices not found in 
that writer, and is useful for the period before AH 40, for which Al Tabari's 
text is not extant. 

Much valuable information is also to be found in the work of Al Baladhuri 
(d. 893) (ed. de Goeje. Leiden, 18(33), who gives a connected narrative of the 
conquest of each district ; but, since his work is not arranged in aunalistic 
form, I have not given the extracts from it with those of the other writers, 
but separately at the end. Notices derived from the same source as those 
of the Mohammedan writers are also to be found in the bilingual chronicle 
of Elijah of Nisibis (written 1019), most of these being quoted from the work 
of Mahomet the Khawarizmi (circ. 835) ; ^ but, since this portion of Elijah's 
chronicle has been translated into German by Dr. Baethgen (Abh. fltr 
die Kunde des Morgenlandes Bd. 8), there is no need to repeat the notices 
here, but it will be sufficient to give references to them in the margin. The 
authority most frequently quoted by the Arabic writers is Al Wakidi 
(d. 823). Most of the notices are merely annalistic entries; but sometimes, 
especially in Al Baladhuri, longer accounts are given. These I have 

^ The extant portion of this work begins with ^ A few are also quoted from the Chronicle of 

the accession of Al "Walid I. (705). the Arab Kings, a work uf tliu 10th century. 


been obliged from considerations of space to yhortcn ; but, however important 
for Arabic life and character the omitted passages may be, nothing essential 
to the purposes of this article is lost by their suppression. Only the long 
and interesting narrative of the expedition of 71G — 718 in the Khitab 
Al ' Uyun I have been obliged to pass over altogether. 

In the margin of the annalistic notices I have given references to notices of 
the same events in other writers (not necessarily derived from the same 
source), including, besides Elijah of Nisibis, in Greek Theophanes and Nike- 
])horos, and in Syriac Michael the Syrian,^ the chronicle of 775 falsely 
attributed to Dionysios,^- and the Chronicle of 840 {Zfitsrhr.d.dnUsch.morgcnl. 
Gesellsch. vol. 51, p. 5C9). In the extracts from Al Baladhuri, to avoid repeti- 
tion, I have generally referred only to the preceding annalistic extracts. To 
nvoid possibility of misleading, I have given all geographical names in the 
first instance in the Arabic form, placing the usually received names in 
brackets following, wherever they can be identified. I have added a few 
notices relating to Armenia and Syria, which are so closely connected with 
those referring to Asia Minor that it appears unreasonable to omit them. 

Caliphate of 'Umar I. 
A. H. 20 (Dec. 21, G40-Dec. 9, 641). 

Ibn Al Athir. And in this year, I mean the year 20, Abu Bachriyya 
'Abd Allah, the son of Kais, made a raid into the land of the Romans ; and 
he was the first who entered it, as it is said (and it is also said that the first 
who entered it was Maisra, the son of Masruk, the 'Absi), and he carried off 
prisoners and spoil. 

Caliphate of 'Uthman. 

25 (Oct. 28, 645-Oct. 16, 646). 

Ibn Al Athir. And in this year Mu'awiya made a raid upon Roman 
territory and reached 'Ammuriya (Amorion) ; and he found the fortresses 
between Antakhiya (Antioch) and Tarsus deserted, and he stationed in them 
a large number of the men of Al Sham (Syria) and Al Gazira (Mesopotamia), 
until he returned from his raid. Then after that he sent Yazid, the son of 
Al Chur, the 'Absi, upon a raid in the summer ; and he gave him orders, 
and he acted accordingly ; and, when he went out, he destroyed the for- 
tresses as far as Antakhiya,^ 

28 (Sept. 25, 648-Sept. 13, 649). 

Ibn Al Athir. And in this year Chabib, the son of Maslama, made a 
raid upon Suriya,* in the land of the Romans. 

^ In the Arabic version in the British the name rather points to the Syrian city. 
Museum MS. Or. 4402. * ^■«•, Syria : the name seems to be used by 

2 Published, with translation, by the Abbe the Arabs to denote Euphratesia and Cilicia. 

Ohabot (Paris, 1895). But ])erhai)s we should read Sauriya (Isauria) : 

* This seems to show that Antioch in Pisidia cf. p. 194, note 3. 
is here meant, though the previous mention of 

ISi K. \V. |;H(M)KS. 

32 (Au<r. 12, ()-)2-An<r. 1, (;:,.S). 

Spbcos 3, 36 Ibii Al Atliir. It is said tliat in this year Mii'awiya, the son of Ahii 

Snfyan, made a raid uj)nn the straits of Al Knstantiniyya rConslantinoplc) ; ' 
and with liim was liis wife; 'Atkha, the dan^ditcr of Karaza ; and it is said 
also that liis sister was wilh him. 

33 (Aug. 2, f)o3-July 21, (irA). 

Tbn Al Athir. In this yoac was the raid of Mu'awiya njtoii the 
fortress of Al Mara, in the laml of the Romans, in tjie neighbourhood of 
Malatya (Melitcne). 


41 (May 7, (10 1 -Apr. 25, 6G2). 

Ibn Wadh. . lie sent Chabib, the son of Maslania ; and tli(; Roman 
commander made peace, and did not care to engage with liini. 

42 (Apr. 2G, 062- Apr. 14, G63). 

Tlicojili. AM Al Tab.. And in tliis year the Moslems made a raid upon the Romans 

^^^^ and iuHictcd a severe defeat upon them, as men record, and killed many of 
their j)atricians. 

43 (Apr. 15, 663-Apr. 3, GG4) 

Kl. Nis. 43. Ibn Wadh. . Busr, the son of Abu Arta, made a raid into the land of the 

" 6157 ^ Romans, and wintered there. 

Al Tab. adds : Until he reached Al Knstantiniyya, as Al Wakidi asserts ; 
and some of the authorities deny this, and say that Busr did not winter in 
Roman territory at all. 

44 (Apr. 4, GG4-Mar. 23, 6G5). 

El. Nis. 14. Ibn Wadh.. 'Abd Al Rachman, the son of Khalid, the son of Al Walid 

^ ''f I'^'er'f *' made a raid until he reached Akluniya (Koloneia). 

Al Tab. . Among the events of this year was the invasion of the Roman 
tenitory by the Moslems under 'Abd Al Rachman, the son of Khali'd,'-^ the son 
of Al Walid, who wintered there,^ and the sea expedition of Busr, the son of 
Abu Arta. 

45 (Mar. 24, 665-Mar. 12, GGG). 

Tlieoph. AM Ibn Wadli. . 'Abd Al Rachman, the son of Khalid, the son of Al Walid, 

6156(0 made a raid and wintered in the land of the Romans, and reached Antakhiya 
(Antioch in Pisidia). 

* This expedition is elsewhere rccoiflcd only Tah.'s text, and I in.scrt it from Ilm Al Athir. 
hy the Anncniaii Hiheo.s, who makes Mii'awiya '* Tlic Hyriac fragments jmlilishcd by Dr. 

iiiareh to (Jhalkcdon in the 13th ofConstantine 'Nuhlckv {Z.Jj.Mj/. '2'.t, ].. 7011'.,) .i.t^nvo in tin' 

(653 4). date, giving A.S. 1175 (Oct. 1, 603 -Sipt. ;'.(>, 

- 'Son ol Khalid' has diojijicd uut of Al 664). 


Al Tab.. And in tliis year was the winU'iiii^^ nf 'Alxl Al Raclimaii, tlio 
son of Klialid, the son of Al Walitl, in the land of the Romans. 

46 (Mar. 1.'^ GGG-Mar. 2, G()7). 

Ibn Wadli. . JMalikh, tViu .son (jf 'Alxl Allah, the Kliatli ami made; a r.iid ; 
and it is said that it was Malikh, the son of Ilnhaira, the Sakhuni ; and he 
wintered in the land of the Romans. 

Al Tab.. And anion^^ the events of this year was the winterin^f of 
Malikh, tlio son of 'Abd AUaii,^ in the land of the Romans; an<l it is .said also Kl. Nis. 16 
that this was 'Abd Al Rachman, the son of Ivhalid, the son of Al Walid ; and it 
is said also that it was Malikh, the .son of Hubaira, the Sakhuni.- And in this 
year 'Abd Al Rachman, the son of Klialid, the son of Al Walid, returned from 
the land of the Romans to Chims (Kmcsa) ; and Ibn Utlial the Ansari gave 
him a poisoned drink, as it is said, and he drank it, and it killed him, 

47 (Mar. 3, G67-Fcb. 11), GG8). 

Ibn Wadh. . Malikh, tlic son of Hubaira, the Saklmni, made a raid and K'- N'^- 17 
wintered in the land of the Romans. 

Al Tab. , And in this year was the wintering of Malikh, the son of 
Hubaira, in the land of the Romans, and the wintering of Abu 'Abd Al 
Rachman the Kaini at Antakhiya. 

48 (Feb. 20, GG8-Feb. 8, G69). 

Ibn Wadh. . 'Abd Al Rachman the 'Atbi made a raitl and reached 
Antakhiya the black.^ 

Al Tab, . And in it was the wintering of Abu 'Abd Al Rachman tlic 
Kaini at Antakhiya, and the summer expedition of 'Abd Allah, the son of 
Kais, the Fizari, and the raid of Malikh, the son of Hubaira, the Sakhuni, 
by sea, and the raid of 'Ukba, the son of 'Amir, the Guhani, by sea, with the 
men of Misr (Egypt) and the men of Al Madina ; and over the men of Al 
Madina was Al Mundhir, the son of Zuhair, and over their combined forces 
was Khalid, the son of* 'Abd Al Rachman, the son of Khalid, the son of Al 

49 (Feb. 9, GG9-Jan. 28, G70). 

Ibn Wadh. . Fudhala, the son of 'Ubaid, made a raid; and by his hands 
God made captives and carried off many prisoners. 

Al Tab,. And in this year was the wintering of Malikh, the son of 
Hubaira, the Sakhuni, in the land of the Romans. And in it was the raid of ei. Nis. 49 
Fudhala, the son of 'Ubaid, upon Garabba; and he wintered at Garabba, and ^''gi'go^f*^ 

^ Text ' Ubaid Allah ' : we may correct from Isauria is perhaps intended. 

Ibn Al Athir and Ibn Wadh. . * The words ' Khalid, the son ol',' are not in 

'^ MSS. 'Fizari': we may corr(^et from Ibn the MSS., but are supplied by conjecture in 

Al Athir and Ibn Wadh. . Thorbeckc's text. Otherwise we should have a 

"•* I do not know any other authority for tliis glarin-,' contradiction to the statement oi'Al Tab. 

epithet ; it it is meant to distinguish this sub ann. 46. 
Antioch tiom tliat mentioned above, Autioch in 

186 E. W. BROOKS. 

it was captured by his hands, and he made many prisoners in it. And in it 
was the summer campaign of 'Abd Allah, the son of Khurz, the Bagli. And 
in it was the raid of Yazid, the son of Shagara, the Rahawi, by sea ; and he 
wintered at the head of the men of Al Sham. And in it was the raid of 
'Ukba, the son of Nafi', by sea ; and he wintered at the head of the men of 
El. Nis. 51 Misr. And in it was the raid of Yazid, the son of Mu'awiya, into Roman 
6159' territory, till he reached Kustantiniyya ; and with him were Ibn 'Abbas, and 
Ibn 'Umar, and Ibn Al Zubair, and Abu Ayyub the Ansari. 

Instead of the last sentence Ibn Al Athir has : In this year (and 
the year 50 is also mentioned) Mu'awiya sent a powerful force upon a 
raid into the territory of the Romans ; and he appointed Sufyan the son of 
'Auf to the command, and ordered his son Yazid to join the raid ; and he was 
disinclined to do so and made excuses, and his father abstained from pressing 
him. And during their raid the men were attacked by famine and grievous 
disease. . . } And, when Mu'awiya heard of his verses, he enjoined him 
to join Sufyan in the land of the Romans, in order that whatever befell the 
men might befall him. And he went, and with him was a large body of men, 
whom his father sent with him ; and in this force were Ibn 'Abbas and Ibn 
'Umar and Ibn Al Zubair and Abu Ayyub the Ansari and others, and 'Abd Al 
'Aziz, the son of Ruzara, the Khilabi. And they advanced into the territory of 
the Romans until they reached Al Kustantiniyya ; and the Moslems and the 
Romans fought for some days, and the battle was severe between them. 
. . . Then Yazid and the army returned to Al Sham.^ 

50 (Jan. 29, 670-Jan. 17, 671). 

Ibn Wadh. . Busr the son of Abu Arta made a raid ; and Sufyan the 
son of 'Auf wintered. 

Al Tab. . And in this year was the raid of Busr the son of Abu Arta 

Theoph. AM and Sufyan, the sou of 'Auf, the Azdi, into the land of the Romans. And it is 

said that in it was the raid of Fudhala, the son of 'Ubaid, the Ansari, by sea. 

51 (Jan. 18, 671-Ja». 7, 672). 

Ibn Wadh. . Mahomet the son of 'Abd Al Rachman made a raid ; and 

Theoph. AM Fudhala, the son of 'Ubaid, the Ansari, wintered. 

Al Tab. . And among the events of this year were the wintering of 

El. Nis. 51 Fudhala the son of 'Ubaid in the land of the Romans, and the raid of Busr the 
Theojih. AM 

6163 son of Abu Arta in the summer. 

52 (Jan. 8-Dec. 26, 672). 

Ibn Wadh. . Sufyan the son of 'Auf made a r^^id ; and he died and 
appointed 'Abd Allah, the son of Mas'ada, the Fizari, to take his place. 

^ I omit personal anecdotes which have no deke (see p. 184, note 3) places the expe- 
bearing on the expedition. ditioii of Yazid in A.S. 971 (660) ; but, as that 

'^ The Syriac chronicle published by Neil- was a time of peace, the date is clearly wrong. 


Al Tab. . And Al Wakidi states that in this year was the raid of Sufyan, Kl. Nis. m 
tlic son of Anf, the Azdi, and his wintering in tlie hmd of the Romans; and '^'giesio'^^ 
that he died (hiring the year and appointed 'Abd AHah, the son of Mas'ada, tlie 
Fizari, to take his place. And otlier authorities say : No, the man who 
wintered in the land of the Romans this year at the head(jt the men was Busr 
the son of Abu Arta, and with him was Sufyan, the son of 'Auf, the Azdi. An<l T1iim.|,1i. am 
in the sunnner of this year a raid was made by Mahomet, the son of 'Abd ^ i *'!*'* .... 
Allah, the Thakafi. • ' ' 

53 (Dec. 27, ()72-Dec. 1.5, G7.S). 

Ibn Wadh. . Mahomet, the son of Malikh, made a raid ; and it is said 
that Tarsus was taken this year, its captor being Gunada, the son of Abu 
Umayya, the Azdi. 

Al Tab, . And among the events of this year was the wintering of ' Abd ei. Nis. r.i(?) 
Al Rachman, the sonofUmAl Chakham, the Thakafi, in the land of the 
Romans. And in it Rudus (Rhodes), an island in the sea, was taken ; and its 
captor was Gunada, the son of Abu Umayya, the Azdi ; and he settled the 
Moslems in it, as recorded by Mahomet the son of 'Umar ^ ; and they sowed 
seed and acquired flocks and herds in it, which they pastured all round it ; 
and, wlien men approached, they took them into the fortress ; and they had 
watchmen who gave them warning of anyone upon the sea who wished to 
make war upon them, and they were on their guard against them. And they 
were the greatest annoyance to the Romans, and they attacked them on the 
sea and cut off their sbips. And Mu'awiya supplied them plentifully with 
provisions and pay; and the enemy were afraid of them. And, when 
Mu'awiya was dead, Yazid, the son of Mu'awiya, removed them. 

Ibn Al Athir adds : And it is said that it was taken in the year 60. 

54 (Dec. 16, 673-Dec. 5, 674). 

Al Tab. . And in this year was the wintering of Mahomet, the son of 
Malikh, in the land of the Romans, and the summer campaign of Ma'n, the 
son of Yazid, the Sulami. And in it, as Al Wakidi states, was the capture by 
Gunada, the son of Abu Umayya, of an island in the sea near Kustantiniyya, 
called Arwad.^ And Mahomet, the son of 'Umar, records that the Moslems 
remained in it for a space, as he says, of seven years, and the commandant 
was Mugahid, the son of Gabr. 

There follows in Al Tabari a long personal story, the substance of which 
is expressed by Ibn Al Athir in the sentence : 

And, when Mu'awiya died, and his son Yazid succeeded to the govern- 
ment, he ordered them to return, and they returned. 

^ Michael the Syrian records wliat seems to * This seems to be a mere duplicate of the 

be the same expedition under the 2nd of Con- occupation of Rhodes recorded under the pre- 

stantine = AS. 982 (Oct. 1, 670— Sei)t. 30, 671). vious year. 

2 i.e. Al Wakidi. 

}HH K. W. r.KOOKS. 

55 (Dec. (), r)74-Nov. 24, 075). 
Kl. Nis. [,(\ Ih„ Wadli. . Malikli, tho son of 'Abd Allali, the Khath'ami, made a raid 

and wintered in the land of llio Romans. 

AI Tab. . And ainon<( tlie events of tliis year was tlie wintering,' of Snfyan, 

the son of 'Auf, the Azdi, in the land of the Ronian.s, as Al Wakidi s;»ys ; ' and 

Kl. Nis. r>5 some of tlie authorities say; No, the man who wintered in the land of the 

(;](;^ Romans this year was 'Abd Allah, the son of Kais, the Fizari ; and some say ; 

No, it was Malikh, the son of 'Abd Allah. 

50 (Nov. 25, G75-Nov. 13. 07G). 
Kl. Nis. r.7 Ibn Wadh. . Yazid, the son of M<i'awi3'a, made a raid and reached Al 

Knstantiniyya ; and Mas'ud, the son of Abu Mas'ud, wintered ; and the com- 
mander by land was Yazid, the son of Shagara, and by sea 'lyadh, the .son of 
Al Charith.''^ All these thin<,'S are also .said to have liappened in the year 57. 

Al Tab. . And in this year was the wintering of Gunada, the son of Abu 

Kl. Nis. 54(?) Umayya, in the land of the Romans; and it is said that it was 'Abd Al 

Rachman, the son of ]\Ias'ud ; and it is said that this year Yazid, the son of 

Shagara, the Rahawi, made a raid by sea, and 'lyadh, the son of Al Charith, 

by land. 

57 (Nov. 14, G76-Nov. 2, 077). 
TlH^opli. AM Ibn Wadh.. 'Abd Allah, the son of Kais, made a raid. 

Al Tab.. And this year was the wintering of 'Abd Allah, the son of 
Kais, in tlie land of tho Romans. 


58 (Nov. 3, 077-Oct. 22, 078). 
Kl. Nis. 56(?) Ibn Wadh. . Malikh, the son of 'Abd Allah, the Khath'ami, made a 

Kl. Nis. 58 laid ; and it is said that 'Amir, the son of Yazid, the Guhani, did so ; and 
Yazid, the son of Shagara, was killed at sea. 

Al Tab.. And this year Malikh, the son of Abd Allah, the Khath'ami, 
made a raid into the land of the Romans. And in this year Yazid, the son 
of Shagara, was killed ^ at sea on a .ship, as Al Wakidi says. He says : And 
it is said that 'Ainr, the son of Yazid, the Guhani, was * the man who 
wintered in the land of the Romans; and it is said that the- man wlio 
made the raid by sea this year was Gunada, the son of Abu Umayya. 

Ibn Al Athir. This year Malikh, the son of 'Abd Allah, the Kliath- 
'ami, made a raid into the lantl of the Romans, and 'Amr, the son of Yazid, 
the Guhani, by sea; and it is said that it was Gunada, the son of Abu 

' It is hard to reconcile this with tha state- that Y., tlic sou of p., (made a raid) by .sea" ; 

ment under A.H. 52 tliat Al Wakidi jdaccd hut by the change of a point ('kutila' for 

Sufyan's death in that year. ' kila ') we get tlie same as in Al Tab. . 

- MS. Al Chaib. Al Charith is an obvious * lomit tlie eojiula before the verb with MS. C. 

corrnctioii of Houtsma ; cf. Al Tab. . The name Tlie printed text must be rendered ' it was 'Amr 

' C'harb ' does not take tlie artielc. . . . and he was the man who wintered.' 

* According to Houtsnia's text, ' It is said 

THE AKAP.S IN ASIA iMINoK, KI{(>M .\l!.\l5in Sf)lTl{(M:s. ISO 

.')0 (Oct. 2:j. (;7«-0ct. 1 2, 07J)> 

11)11 Wadli. . 'Auir, tliu son of Mmra, tlic (luliaiii, ' iii.uli' a laiil hy laml, Kl. Nis. .'i8(f) 
ami tluTc was not that year any raid by sea. 

Al Tab.. And that year was the wintcriiiff (»!' 'Anir, the son'of Miirra, 
the (Juliaiii, in the land of" the Romans on land. Al Wakidi says: Tiicre 
was iKjt that year any raid by sea. And others say: Not so; Onntuia, tlie Kl. Ni>. fi'J 
son of Abn Umayya, made a raid by sea. 

GO (Oct. 13, G70-Sept, 30, (iSO). 

Al Tab. . And this year was the raid of Malikh, the son of 'Abd Allah, 
upon Sauriyya (Isauria), and the entry of Ounada, the son of Abn Umayya, 
into Iludus, and his building of the city there, as Al Wakidi say.s.^ 

Caliphate of Yazid I. 

01 (Oct. 1, 000-Sept. 19, G81). 

Ibn Wadh. . Malikh, the son of 'Abd Allah, the Khath'ami, made a 
raitl in the summer ; and this was a raid upon Sauriyya.^ 

Caliphate of 'Abd Al Malikh. 

70 (June 23, G80-June 14, C90J. 

Al Tab. . And in this year the Romans rose up and assembled together 

against the Moslems in Al Sham ; and 'Abd Al Malikh made peace with the Tlicopli. AM 

king of the Romans on condition of paying him a thousand denarii t^very ju^.li j-^'j qgq, 

a.ssembly-day,'* fearing danger from him to the Moslems. (w itliout ilau-) 

C'hron. of 840 

75 (May 2, G94-Apr. 20. G95;. (without date) 

Ibn Wadh. . Mahomet, the son of Marwr.n, made a raid in the snmmer ; Tluoi-li. AM 

and the Romans came out against Al A'ttiak,"' and they were slain by Aban, 

' .^ •'El Nis. 7.5 

the son of Al Walid, the son of 'Ukba, the son of Abu Mu'ait, and Dinar, the Mich. AS 

son of Dinar. ^^^^ 

Al Tab. . Among the events of this year was the raid of Mahomet, the as ioo6 

son of Marwan, in the summer, when the Romans came out from before 

Mar'ash (Germanikeia). 

^ There can be little doubt that this is the * i.e. Friday. Theoph. 'every day': so 

same as ' Ami', the son of Yazid, the Guhani. Michael. 

Under 58, where the Arabs have Ibn Yazid, El. ^ i.e. ' the valleys,' the name of a place bc- 

Nis. has Ibn Murra. tween Germanikeia and Antioch. The MS. has 

- Al Tab. has probably confused Al Wakidi's Al A 'man, but Houtsma's correction is no doubt 

date for the occupation with that for the cvacu- right, and Al A'mak is tlie name given by Al 

ation ; cf. ann. 53, 54. Baladhuri (see p. 207); cf. also ami. 112. The 

•* This must be thrown Ijack to 679, since Syriac writers call the place ' the valley of 

l)eace was made before Mu'awiya's death (Apr. Antioch.' 
C, 680). 

190 E. W. BROOKS. 

76 (Apr. 21, 695-Apr. 0, 006). 

Ibii Wadli. . Yacliya, the son of Al Cliakham, nmdc a rni(i in tlie summer 
at Marg al Shacliani between Malatya and Al Massisa (Mopsouestia).^ 
Tlicopli. AM Ibn Al Atliir. And tliis year Maliomet, the son of Marwan, made a 


raid upon the Romans in the region of Malatya.^ 

77 (Apr. 10, 696-Mar. 29, 697). 

El. Nis. 77 Ibn Wadli. . Al Walid, the son of 'Abd Al Malikh, made a raid upon 

'^6189 Atmar ; and his raid was in the region of Malatya. And Chassan, the son of 
Al Nu'man, made a raid by sea.^ 

Al Tab, . And this year Al Walid made a raid in the summer. 

78 (Mar. 30, 697-Mar. 19, 698). 

Al Tab. . And 'Abd Al Malikh sent Yachya, the son of Al Chakham, to 
make a raid this year. 

El. Nis. 79. 79 (Mar. 20, 698-Mar. 8, 699). 

Theoph. AM i . i i • 

6192 Al Tab. . And this year, as it is said, the Romans fell upon the men of 

flithoutdatel Antakhiya. Ibn Al Athir adds : and defeated them. 

80 (Mar. 9, 699-Feb. 25, 700). 

Al Tab. . And 'Abd Al Malikh sent his son Al Walid upon a raid this 

81 (Feb. 26, 700-Feb. 14, 701). 

Al Tab.. I was informed by 'Umar, the son of Shabba; he said: I was 
told by 'Ali, the son of Mahomet ; * he said : 'Abd Al Malikh sent his son 
'Ubaid Allah ^ upon a raid in the year 81 ; and he took Kalikala (Theodosiou- 
polis-Karin in Armenia). 

82 (Feb. 15, 701-Feb. 3, 702). 

Ibn Al Athir. And this year Mahomet, the son of Marwan, made a raid 

upon Arminiya (Armenia) and routed them. Then they asked him for peace, 

and he granted it to them ; and he appointed Abu Shaikh, the son of 'Abd 

Theoph. AM Allah, sfovernor over them, and they acted treacherously towards him and 

killed him. And it is said also that they killed him in the year 83. 

^ According to Yakut Marg Al Shacham was Al W. made a raid;' After this several lines 

near Aniorion. Similarly El. Nis. (ann. 23) are missing down to AH. 83. 

makes Mu'awiya take Ankyra and advance to * Better known as Al Madaini, a writer of 

Marg Al Shacham. the early part of the 9th century. 

^ Armenia IV. according to Theoph. . "* We should probably read 'Abd Allah, since 

^ This is Houtsma's correction. The MS. has no such name as ' Ubaid Allah appears among 

' Al Bachr ( = the sea) the son of Ch. the son of the sons of 'Abd Al Malikh. 


83 (Feb. 4, 702-Jaii. 23, 703). 

Ibu Wadh. . 'Abd Allah also ' made a raid and took Al Masaisa, and El. Nis. 88, 

built a small fortress in it.^ Theoph. AM 

Mich. A3 

84 (Jan. 24, 703-Jan. 13, 704). 1015, 1017 

1 /.All Chn.n. of 846 

Al Tab. . And in this year was the raid of 'Abd Allah, the .son of 'Abd a.S 1015 

Al Malikh the son of Marwan, into Roman territory ; and in it he took Al 

Massisa. Such is the record of Al Wakidi. 

85 (Jan. 14, 704-Jan. 1, 705). 

Ibn Al Athir. And this year Mahomet, the son of Marwan, made a raid TheopU. AM 
into Arminiya and passed summer and winter in it. 

Caliphate of Al Walid I. 

86 (Jan. 2-Dec. 22, 705). 

Ibn Wadh. . Maslama made a raid and took two fortresses. El. Nis. 86 

Al Tab. . Maslama, the son of 'Abd Al Malikh, made a raid into the 
land of the Romans. 

87 (Dec. 23, 705-Dec. 11, 706). 

Al Tab. . And in this year Maslama, the son of 'Abd Al Malikh, made 
a raid into the land of the Romans; and with him was Yazid, the son of u am 

Gubair, and he met the Romans with a great force at Susana (Sision) in qi^'q 
the neighbourhood of Al Massisa.^ Al Wakidi says: This year Maslama met i avt 

Maimun the Gurgani ^ (and with Maslama were about a thousand fighting 5201 
men of the men of Antakhiya) at Tuwana (Tyana) ; and he killed many men 
among them, and God took the fortress by his hands ^ (and it is said that the 
man who made a raid upon the Romans in this year was Hisham, the son of 
'Abd Al Malikh) ; and God took by his hands the fortress of Bulak and the 
fortress of Al Akhrim and the fortress of Bulas (Pylai?) and Kamkim; and he 
killed of the Musta'riba^ about a thousand fighting men, and carried their 
children and their women into captivity. 

^ This word seems to show that Ibn Wadh. 6 tov Xowei). 

recorded an expedition by 'Abd Allah in the * This mean.<» Hyrcanian ; but we should 

previous year. no doubt read ' Gurgunami, ' the Arabic name 

2 Theoph. and the Chronicle of 846 mention for the Mardaites, which is the title applied to 

the building only. El. Nis. and Mich, record Maimun by Al Baladhuri (see p. 203). 

the capture and the building under separate ^ This seems to be an anticipatory statement, 

years. since Al Tab. afterwards records the capture 

^ Al Tab. omits to mention the result of the under 88, to which year it is also assigned by 

meeting, which according to Theoph. was a Ibn Kutaiba, the earliest extant Arabic historian 

great defeat of the Arabs. A comparison with {d. 884). 

Theoph. makes it probable that the subject of " Arabs not of pure birth, 
the last clause is not Maslama but Yazid ("AC'Soj 

192 Tv W. IIKOOKS. 

88 (Dec. 12, 7U(i-Nuv. ;i(), 7U7). 

Ibii Wadh,. Maslamu and Al 'Abbas, the son of Al Walid, niadc a laid 
and took Suriya (Isaiiia ?), and Al 'Abbas took Ardaluniya.^ 

Al Tab. . And among the events of this year was (iod's captnie by the 

Tlieoph. AM hands of the Moslems of one of the Roman fortresses called Ttiwana in 

Nikcph. [.p. Guniada II. (May O-June 0),- and they wintered at it ; find over tlie army were 

Mieh^AS *^'^^^^^^^' ^^^^ SO" of *Abd Al Malikh, and Al 'Abbas, the son of Al Walid, 

loig the son of 'Abd Al Malikh. And Mahomet, the son of 'Umar, Al Wakidi, 

records that Thur, the son of Yazid, told him on the authority of his 

masters : he said : The capture of Tuwana was effected by the hands of 

Maslama, the son of 'Abd Al Malikh, and Al 'Abbas, the .son of Al Walid; 

and the Moslems routed the enemy that day, so that they went to their 

church ; then they returned, and the men*^ were routed until they thought 

they should never recover from it. And Al 'Abbas remained, and some 

men with him, among whom was Ibn Muchairiz, the Gumachi ; and Al 

'Abbas said to Ibn Muchairiz, " Where are the men of the Kuran who are 

seeking Paradise ? " And Ibn Muchairiz said, " Call to them to come to 

you." And Al 'Abbas called out, " Ye men of the Kuran ! " And they 

came all together ; and God routed the enemy, until they entered Tuwana. 

And in it Maslama also made a raid into Roman territory, and by his 
hands three fortresses were taken, the fortress of Kustantin and Ghazala 
(Gazelon) and the fortress of Al Akhrim, and he killed of the Musta'riba 
about 1,000 men, besides carrying their children into captivity and taking 
possession of their property.^ 

Khitab al 'Uyun. And in the year 88 Maslama and Al 'Abbas, the son 
of Al Walid, made a raid upon Tuwana and wintered at it. And the Romans 
assembled against them ; and they met, and God IMost High routed the 
Romans, and 50,000 of them were killed. And God Most High took Tuwana 
and another fortress near it with prisoners and spoil. 

89 (Dec. 1, 707-Nov. 19, 708). 

Al Tab. . The Moslems in this year took the fortress of Suriya ; and over 

the army was Maslama, the son of 'Abd Al Malikh. Al Wakidi states that 

El. Nis. 89 Maslama made a raid into the land of the Romans this year, and with him 

was Al 'Abbas, the son of Al Walid ; and they entered it together ; then they 

Theoph. AM separated, and Maslama took the fortress of Suriya, and Al 'Abbas took 

^'' Adhruliyya ; and he encountered a force of Romans and routed them. And 

others besides Al Wakidi say : Maslama went to 'Aminuriya (Amorion) and 

encountered the Romans there, a large force, and God routed them : and he 

* This is the MS. reading. Houtsma wovild siege of nine months. 

read Adruliya ; cf. Al Tab. ann. 89. s n^j^ ^\ Athir ' the Moslems.' 

^ MS. B adds 'the (irst of it' (May 9), wliile ■» The last clause is clearly a diqdicate of the 

Ibn Al Athir lias Gumada I. (Apr. 9— May 8). notice under the preceding year. 
Mich, places the capture in Mar. 708, after a 

Till': Al!Ai:s IN ASIA MINOK, FIJO.M \I;AI'.IC S( )I' H( 'KS. 103 

took Hirakla (Hfnikieia) and Kainiidiya (Nikoint'dcia ?),' and ATAbbas made 
a smiiiiicr cain])aiL,ai in the noi^ddjourliooil of Al liiidaiidun (Podaiidos). 

1)0 (Nov. 20, TOS-Nov. S, 70!»). 

Ibn Wadli. . 'Abil Al 'Aziz, t.lic son lA' Al Walid, made a raid and took a Kl. Ni-^. JtO 

Al Tab.. And in tliis year Maslania mado a raid into tlio land (»f the 
Romans, as Malioniet, tin; son of 'Uniar, iec<»rds, in tlu; n(;i^ld>ourliood of 
Snri}a, and took the live fortresses in Snriya. And in it Al 'Abbas, tlic son 
of Al Walid, made a raid, some say, till be readied Al Arzan,'- and otiiers say, Tli<..|.li. AM 
till lie readied Suriya. And Malioniet, tlie son of 'Uniar, says: tlie account 
wliicli says ' till be readied Suriya ' is riglit. 

01 (Nov. 9, TOD-Oct. 2.S, 710). 

Ibn Wadb. . 'Abd Al 'Aziz, tlic son of Al Walid, made a raid. 

Al Tab. . And in this year, as Mabomet, tlie son of 'Ulnar, and others 
record, 'Abd Al 'Aziz, the son of Al Walid, made a raid in the summer ; 
and over tlie army was Maslania, tlie son of 'Abil Al Malikli. 

'J2 (Oct. 29, 710-Oct. 18, 711). 

Ibn Wadh. . Mahomet, the son of Marwan, made a raid. 

Al Tab. . Among the events of the } ear was the raitl of Maslamn, the 
son of 'Abd Al Malikli, and 'Umar, the son of Al Walid, into the land of the 
llomans ; and tlirec fortresses were taken by the hands of Maslania; and the 
people of Susana migrated into the interior of the land of the Komans. 

93 (Oct. 19 711-Oct. 6, 712). 

Ibn Wadh. . Al Abbas, the son of Al Walid, and Marwan, the son of El. Nis. 92 
Al Walid, and Maslama made a raid and took Amasiya (Amaseia) and the Thco].li. AM 
fortress of Al Chadid. Mif.f AS 

Al Tab. . And among the events of this year were the raid of Al 'Abbas, 1023 
the son of Al Walid, into the land of the Romans, and God's capture of ^''Xsm^^ 
Sabastiyya (Sebasteia) ^ by his hands. And in it was also the raid of Marwan, 
the son of Al Walid, into the land of the Romans ; and he reached 
Khangara.* And in it was the raid of Maslama, the son of 'Abd Al Malikh, 
into the land of the Romans ; and he took Masa (Amaseia) and the fortress 

* vvll. Kuliya and Kamuliyah. Kamouliana Ilui Al Atliir, who adds 'and Al Marzbanain 

in Cappadocia I. may be meant. Ibn Al Athir and Tus ' (of. aim. 95). It is possible, however, 

lias 'Kamuniya. ' The mention of Heraklcia that Mistheia is meant (cf. Thcoph. A.M. 

(Ponton ?) points to Nikomcdeia, but it is 6204 ; Nikeph. p. 48 ; Chron. of 846 A. S. 

strange that its capture should not be recorded 1021). 

by the Greek writers. " vl. Oangra. Mich, records the capture of 

^ Arzan should be Arzancne or its chief town, ' Gargarun ' in A.S. 1022. On the other hand, 

but this is clearly out of place here. Yakut mentions Khangara, ' a district in the 

3 The MSS. have Samastiyya : I ciuend from territory of the Komans.' CI. also ami. 109. 


194 E. W. BROOKS. 

of Al Chadid and Ghazala and Tarchamah ^ in the neighbourhood of 

94 (Oct. 7, 712-Sept. 25, 713). 

Ibn Wadh. . Al 'Abbas and 'Umar, the sons of Al Walid, made a raid. 

El. Nis. 94 Al Tab. . And amonofst the events of this year was the raid of Al 'Abbas, 

^6205 ^^^® ^^^^ ^^ Walid, into the land of the Romans, and it is said that in it lie 

Mich. AS took Antakhiya. And in it, as it is said, 'Abd Al 'Aziz, the son of Al Walid, 
1023 . 

made a raid into the land of the Romans till he reached Ghazala ; and Al 

Walid, the son of Hisham, the Mu'aiti, reached the land of Burg Al Chamam ;2 

and Yazid, the son of Abu Khabsha, reached the land of Suriya.^ 

95 (Sept. 26, 713-Sept. 15, 714). 

El. Nis. 95 Al Tab. . And this year was the raid of Al 'Abbas, the son of Al Walid, 

the son of 'Abd Al Malikh, into the land of the Romans, and God took three 
fortresses by his hands, as it is said, and they were Tulas, and Al Marzbanain, 
and Hirakla.* And in it Al Wadhdhachi was killed in the land of the 
Romans, and about 1,000 men with him. 

96 (Sept. 16, 714-Sept. 4, 715). 

Ibn Wadh. . Bishr, the son of Al Walid, made a raid. 
Al Tab. . And this year, as Al Wakidi says, was the raid of Bishr, the 
son of Al Walid, in the winter ; and, when he returned, Al Walid was dead. 

Caliphate of Solomon. 

Theoph. AM Ibn Wadh. . Maslama made a raid and took the fortress of Al 

Tvr^h^^AS Chadid, and wintered in the lands of the Romans ; and 'Umar, the son of 

1026 Hubaira, made a raid by sea; and they occupied all between Al Khalig^ and 

Al Kustantiniyya, and they took tiie city of the Slavs ; ^ and Solomon sent 

them reinforcements under 'Amr, the son of Kais, the Khindi, and 'Abd Allah, 

the son of 'Umar, the son of Al Walid, the son of 'Ukba. 

Ibn W. also has a duplicate account as follows : And Solomon went out in 
the direction of Al Gazira and took up his abode at a place called Dabik,^ in 

^ This is the reading of the MSS. Guidi places.' He has already recorded the capture 

would read Bargama, which differs only by of Al Marzbanain and Tus (Tulas ?) under AH 

points. Bargama, however, is the Arabic name 93 (p. 193, note 3). The last name might stand 

for Perganios, which seems quite out of place for Doara, which would go well with Sebasteia, 

here. The capture of Pergamos is recorded by but not with Heraklcia, unless Herakleia-Ky- 

Michael and the Chronicle of 846 under A.S. bistra is meant. Another reading is Tunas. Al 

1027 (716). Theophanes also records it under Marzbanain = the two marzbans. 

716 (A.M. 6208), but, as he makes it contem- ^ i.e. the canal. The name covers the Helles- 

poraneous with Leo's accession, he must mean pont, Propontis, and Bosporus, 

to place it in 7l7. ® Prof. Ramsay [Hist. Gcog. of Asia Minor, 

^ i.e. the tower of the pigeon. p. 351) identihes the city of the Slavs with 

^ Since Suriya (Syria, see p. 183, note 4) and Loulon, near the Cilician gates : but the city 

Sauriya (Isauria) differ only by a pointy it is here mentioned would seem to have been near 

often impossible to say which is meant ; cf Constantinople, 

ann. 90. 7 ^g. Phauik. 

* Ibn Al Athir, ' he took Ilirakla anJ other 


the province of Kinnasrin (Clualkis); and he sent Maalama, the son of Abd Al 
Malikh, upon a raid into the territory of the Romans, and told liim to go to Al 
Kustantiniyya, and remain before it till he took it. And Maslama went on 
till he reached Al Kustantiniyya, and remained before it till he had sown 
and eaten of what he had sown ; and he entered and took the city of the 
Slavs. And the Moslems were smitten by scarcity, and hunger, and cold ; and 
Solomon heard of the condition of Maslama and his men, and sent them 
reinforcements under 'Amr, the son of Kaia, by land ; and he sent 'Umar, the El. Nis. 97 
son of Hubaira, the Fizari, to make a raid by sea ; and that because the 
Romans had made an attack upon the city of Ladikiya (Laodikeia), in the 
province of Chims, and had burnt it, and had carried away some of what was 
in it. And 'Umar, the son of Hubaira, reached the canal (khalig) of Al 

97 (Sept. 5, 715-Aug. 24, 716). 

Al Tab. . And among the events of this year was the equipping by Theoph. AM 
Solomon, the sou of 'Abd Al Malikh, of the armies intended to march to Al Miches 
Kustantiniyya, and the appointment of his son David, the son of Solomon, 1027 
to conduct the summer expedition ; and he took the fortress of Al Mara, 
And in it, as Al Wakidi records, Maslama, the son of 'Abd Al Malikh, made a 
raid into the land of the Romans ;^ and he took the fortress which had been 
taken by Al Wadhdhach, the chief of the Wadhdhachiyya. And in it 'Umar,^ 
the son of Hubaira, the Fizari, made a raid by sea upon the land of the 
Romans, and wintered in it. 

98 (Aug. 25, 716-Aug. 13, 717). 

Al Tab. . And among the events of this year was the sending by Solo- El. Nis. 98 
mon, the son of 'Abd Al Malikh, of his brother Maslama, the son of 'Abd Al '^^^^fig'*'*^ 
Malikh, to Al Kustantiniyya ; and he told him to remain before it till he Nikeph. p. 53 
took it or an order from him came to him. And he passed winter and as 1028 
summer there I was told by Achmad, the son of Zuhair, on ' Dion.' AS 

• 1028 

the authority of 'Ali, the son of Mahomet : he said ; When Solomon assumed 
the government, he made a raid upon the Romans ; and he stationed him- 
self at Dabik, and sent Maslama in front ; and the Romans were afraid of 

him ; and Leo appeared from Arminiya And the patricians 

said to Leo, ' If you deliver us from Maslama, we will make you king ' ; and 
they made a covenant with him. And he came to Maslama and said, ' The 
]»eople know that you will not make serious war upon them, but will give 
them a respite, as long as the corn lasts with you : and, if you burn the corn, 
ihey will submit.' And he burned it : and the enemy remained, and the 
Moslems were straitened until they nearly perished And 

1 Though in the text these events are ascribed each year in the narrative, but all together at 

in the summary to 96, we should very probably the end of each Caliphate, 

read 97, since Ibn W. has already recounted the ^ Ibn Al Ath. 'the land of the Wadhdha- 

events of 96 under the Caliphate of Al Walid. chiyya.' 

This author records the campaigns not under ^ fext ' 'Amr.' I correct from Ibn Al Athir. 

o 2 

196 E. W. DROOKS. 

that happened to the force which had never happened to an army before, 
until a man was afraid to go out of the camp ahnie ; and they ate the 
beasts of burden and skins and the trunks and leaves of trees and everything 
except dust. And Solomon remained at Dabik, and continued tlicre through 
the winter; and he was not able to send them help till Solomon died.^ 

And this year the city of the Slavs was taken. Mahomet, the 

son of 'Umar, says ; ' The Burgan (Bulgarians) made an attack in the year 
08 upon Maslama, the son of 'Abd Al Malikh, and he had few men with 
him ; and Solomon sent him help under Mas'ada or 'Amr, the son of Kais, 
witli a military force ; and the Slavs made a treacherous attack upon them ; 
then God routed them, after they had killed Shurachil, the son of 'Abda. 

And in this year, as Al Wakidi states, Al Walid, the son of Hisham, and 
'Amr, the son of Kais, made a raid, and some of the men of Antakhiya were 
cut to pieces ; and Al Walid attacked some men in the outlying districts of 
the Romans, and took many prisoners from among them. 

Chron. of 846 And this year David, the son of Solomon, the son of 

AS 1028(?) <^J3j ^i Malikh, made a raid into the land of the Romans, and took the 
fortress of Al Mara near Malatya.^ 

99 (Aug. 14, 717-Aug. 2, 718).3 

Ibn Wadh. . Solomon, the son of 'Abd Al Malikh, sent his son David to 
the land of the Romans, and Maslama was remaining quiet before Al 
Kustantiniyya ; and David took the fortress of Al Mara, in the neighbourhood 
of Malatya. 

Caliphate of 'Umar II. 

During his government in the year 99 'Amr, the son of Kais, the 

Khindi, made a raid in the summer. 

Theoph. AM Al Tab. And in this year 'Umar, the son of 'Abd Al 'Aziz, sent to 

-,., ^^^ ,_ Maslama, who was in the land of the Romans, and told him to return from 
Nikeph. p. 55 . . ' 

it with the Moslems who were with hmi : and he sent him some high-bred 

horses and a large quantity of corn, and he urged the men to go to his 

assistance. And the number of high-bred horses which he sent to him was, 

as it is said, 500 horses. 

100 (Aug. 3, 718-July 23, 719). 

Al Tab, . And in this year 'Umar, the son of 'Abd Al 'Aziz, sent Al 
Walid, the son of Hisham, the Mu'aiti, and 'Amr, the son of Kais, the Khindi, 
of the men of Chims, to make a raid in the summer. 

1 There is a much longer account of the expe- fortress of ' Antigun ' in A.S. 1028 (Oct. 1, 716 

(lition against Constantinople in the Khitah Al — Sept. 30, 717). For Al Mara of. ann. 33. 
'Uyun ; but it would take too much .space to ' We should perliaps read 98, since Ibn W. 

translate it here. mentions an expedition of 99 under the reign of 

^ The Chronicle of 846 makes David take the 'Umar. 


Ibn Al Athir. In this year 'IJinar, the son ..I 'Al)cl Al 'Aziz, ordered 
the men of Tarandu (Taranton) to withdraw from it to Malatya; and Tar- 
an(hi is in Ww lloman territory, three days' journey from Malatya; and 'Abd 
Allah, the son of 'Abd Al Malikh, had settled the Moslems in it after he 
had made a raid upon it in the year 83;' and Malatya was at that time 
dosertetl : and he introduced among them a military force; from Al Gazira, 
to be stationed among them until the snow came down and they returned to 
tlieir district. And this state of atilairs went on until 'Umar succeeded to the 
government; and he ordered them to return to Malatya and left Taranda 
unoccupied, through fear of injury to the Moslems from the enemy : and he 
left Taranda deserted, and appointed as governor of Malatya Ga'wana, the 
son of Al Charith, one of the sons of 'Amir, the son of Sa'sa'a. 

Caliphate of Yazid II. 

102 (July 12, 72()-June 30, 721). 

Ibn Wadh. . Under his government in tlie year 102 'Abd Al Walid, 
the son of Hisham, made a raid at the head of the men into the land of the 
Romans, and encamped at the ford near Antakhiya. And 'Umar, the son of 
Hubaira, attacked the Romans in Fourth Arminiya, and routed them and 
took 700 prisoners from among them. 

Al Tab. . And in this year 'Umar, the son of Hubaira, made a raid upon 
tlio Romans in Arminiya, and routed them, and took many men prisoners 
from among them — it is said 700 prisoners.^ 

Ibn Al Athir adds: And in this year Al 'Abbas, the son of Al Walid, 
the son of 'Abd Al Malikh, made a raid upon the Romans, and took Dalisa 
(Dalisandos ?).^ 

103 (Jul. 1, 721- June 20, 722). 

Ibn Wadh. . Al 'Abbas, the son of Al Walid, made a raid; and the men El. Nis. i03 
were cut to pieces in detachments. And 'Abd Al Rachman, the son of 
Solomon, the Khalbi, and 'Uthman, the son of Chayyan, the Murri, made a 
raid, and encamped against a fortress, and took it. 

Al Tab. . And in this year Al 'Abbas, the son of Al Walid, made a raid 
upon the Romans, and took a city called Rasala.* 

104 (June 21, 722-June 9, 723). 

Ibn Wadh. . 'Abd Al Rachman, the son of Solomon, the Khalbi, made a 

* According to Theoph. the atta(>k of 'Abd The dilFerence between ' kila ' (it is said) and 

Allah on Taranton in A.M. 6193 (701) was un- ' katala ' (killed) is only one of jwinting. 

successful. The occupation of Taranton is =* See next note. 

placed by Michael in A. S. 1022(711), and by the « vvll. Gha.sla and Wasala. Ibn Al Athir 

Chronicle of 846 in A.S. 1021 (710). According 'Dasala.' Perhaps Ouasada is the jdace meant ; 

to both these authorities the captor was Mas- l)ut it seems ])rohable that it is the same as that 

lama. mentioned under the previous year under the 

'•* Ibn Al Athir 'and killed 700 iirisoncrs.' name of Dalisa (the vowels are doubtful). 

198 E. W. BROOKS. 

raid on the south in the summer ; and 'Uthman, the son of Chayyan, the 
Murri, made a raid upon the north in the summer. 

105 (June 10, 723-May 28, 724). 

Ibn Wadh. . Sa'id, the son of 'Abd Al MaUkh, the son of Marwan, made 
a raid ; then he returned and made a raid upon the regions of the Turks. 
Theoph. AM Al Tab. . And in this year was the raid of Sa'id, the son of 'Abd Al 

^^'^ Malikh, into the land of the Romans ; and he sent out a detachment of about 
1,000 fighting men, and, as is recorded, they were all cut to pieces. 

Ibn Al Athir. And this year Marwan, the son of Mahomet, made a raid 
upon the south in the summer and took Kuniya (Ikonion) in the land of the 
Romans and Khamkh (Kamachos).^ 

Caliphate of Hisham, 

Khitab Al 'Uyun. And this year Marwan, the son of Mahomet, made a 
raid at the head of the forces of Al Gazira and the forces of Al Sham (and he 
was governor of Al Gazira in the name of Hisham), and with him was Sa'id, 
the son of Hisham, at the head of the forces of Al Sham ; and he entered by 
the road of Malatiyya and took a fortress called Muwasa by storm, after he 
had beseiged them and assaulted them with engines. And they asked him 
to grant a capitulation, and he refused to grant them anything but a sur- 
render at discretion. And, when he had taken it, he decided to kill the 
fighting men and carry the children into captivity ; and he divided them 
among the Moslems, and destroyed the fortress. 

106 (May 29, 724-May 18, 725). 

Ibn Wadh. . And during his government, in the year 106, Mu'awiya, the 
son of Hisham, made a raid at the head of the men ; and he sent Al Wadh- 
dhach, the chief of the Wadhdhachiyya, and he burnt the crops and the 
villages, because the Romans had burnt the pasture lands. And Sa'id, the 
son of 'Abd Al Malikh, made a raid upon the north in the summer. 

Al Tab. And in this year Sa'id, the son of 'Abd Al Malikh, made a raid 
in the summer, 

107 (May 19, 725-May 7, 720). 

Theoph. AM Ibn Wadh. . Mu'awiya also made a raid.^ 

6218 j^i rp^^^ _ Maslama, the son of 'Abd Al Malikh, made a raid by land. 

El. Nis. 107 Khitab Al 'Uyun. Maslama, the son of 'Abd Al Malikh, made a raid 

62lT^ Dion. ' upon Kaisariyya (Kaisareia), and that is between Malatiyya and Khamakh 
AS 1040 (Kamachos), and took it. 

cf. Mich, and 

Cliron. of 846 — 

AS 1037 
(Neokaisareia) There is nothing to show whether the author ^ Al Tab. makes this a raid upon Cyprus, 

would place this expedition before or after tlie which does not come within the limits of the 

death of Yazid (Jan. 724), article. 


108 (May 8, 72G-Apr. 27, 727). 

Ibn Wadh. , Ma.slama, the; son of 'Abd Al Malikli, niado a raid in the 
summer on the south ; and 'Asim, the son of Yazid, the llilali, made a raid in 
the summer on tlie north. 

Al Tab. . And in this year was the raid of Maslama the son of 'Abd Al 
Malikli, until he reached Kaisariyya, a city of the Romans on the borders of 
Al Gazira;^ and God took it by his hands. 

And in it also Abraham, the son of Hisham, made a raid and took also 
one of the fortresses of the Romans. 

109 (Apr. 28, 727-Apr. 15, 728). 

Ibn Wadh. . Mu'awiya, the son of Hisham, made a raid, and with him ei. Nis. los 
was Al Battal in command of his advance-guard, and he took Khangara.^ '^'"V.^'Jo^"'^^ 

Al Tab. . And amongst the events of this year was the raid of 'Abd 
Allah, the son of 'Ukba, the son of Nafi', the Fihri, at the head of a force by 
sea, and the raid of Mu'awiya, the son of Hisham, upon the land of the 
Romans ; and he took a fortress in it called Taiba,^ and some of the troops of 
Antakhiya in his company were cut to pieces. 

110 (Apr. 16, 728- Apr. 4, 729). 

Al Tab. . And in this year, as is recorded, Mu'awiya, the son of Hisham, Tlieopli. AM 
made a raid upon the land of the Romans and took Samala. And in it 'Abd 6219(?) 
Allah, the son of 'Ukba, the Fihri, made a raid in the summer ; ami over the 
sea forces, as Al Wakidi records, was 'Abd Al Rachman, the son of Mu'awiya, 
the son of Chudaig. 

Khitab Al 'Uyun. Mu'awiya, the son of Hisham, made a raid in the 
summer; and he sent 'Abd Allah Al Battal in command of his advance- 
guard, and he took a fortress in the territory of the Romans, and in it some 
men were cut to pieces by them ; and Mu'awiya, the son of Hisham 
besieged * 

111 (Apr. 5, 729-Mar. 25, 730). 

Ibn Wadh. . Mu'awiya, the son of Hisham, made a raid in the summer 
upon the north, and Sa'id, the son of Hisham, made a raid in the summer 
upon the south. 

Al Tab. . And among the events of this year was the raid of Mu'awiya, 
the son of Hisham, in the summer upon the north, and the raid of Sa'id, the 
son of Hisham, in the summer upon the south until he reached Kaisariyya. 
Al Wakidi says: In the year 111 'Abd Allah, the son of Abu Maryam, made 
a raid at the head of the sea-forces ; and Hisham appointed Al Chakham, the 

* Ibn Al Ath. adds : ' and that is a celebrated also ann. 93. 

city.' He also records under this year the raid ' vl. Taina ; possibly to be identified with 'rb 

recorded under 107 by Al Tab. . kActtpov 'Artovs (Theoph. A.M. 6219). 

2 As there are no points in the MS., the name * The name has fallen out. Perhaps it is 

might also be read ' Gangra. ' Mich, records Nikaia, the siege of which is recorded by Mich, 

the capture of Gangra under A. S. 1042 (731). Cf. under A.S. 1042 (731). 

200 E. W. BROOKS. 

son of Kais, the son of Makhraina, the son of 'Abd Al Muttalib/ the son of 
'Abd Manaf, to command all the nujn of Al Sham and Misr. 

112 (Mar. 26, 730-Mar. 14, 731). 

Ibn Wadh.. Mu'awiya, the son of Hisham, made a raid upon the 

Romans; and lie did not succeed in entering their territory, but remained at 

the frontier at Al 'Amk,^ in the district of Mar'ash. 

T!mo]>1i. am Al Tab. . And among the events of this year was tlie raid of Mu'awiya, 

iMi<li AS the son of Hisham, in the summer; and he took Kharshana (Charsianon) 

1042 and burnt Farandiyya in the district of Malatya. 

Khitab Al 'Uyun. And this year 'Abd Al Wahljab, the son of Bukht, 
was killed while in company with Al Battal, in the land of the Romans ; and 
that because the men were scattered from Al Battal and put to Hight ; and 
'Abd Al Wahhab . . advanced towards the enemy . . and mingled 
with the host and was killed, and his horse was killed. 

113 (Mar. 15, 731-Mar. 2, 732). 

Al Tab. . And among the events of this year was the death of 'Abd Al 
Wahhab, the son of Bukht ; and he was with Al Battal 'Abd Allah in the 
land of the Romans. And Mahomet, the son of 'Umar, records on the 
authority of 'Abd Al 'Aziz, the son of Umar, that 'Abd Al Wahhab, the son 
of Bukht, made a raid with Al Battal in the year 113, and the men were 
scattered from Al Battal, &c. (the rest as in Kbit. Al 'Uyun, ann. 112). 

And among the events was the raid of Mu'awiya, tht son of Hisham, 
into the land of the Romans ; and he stayed on the frontier in the district of 
Mar'ash and returned. 

114 (Mar. 3, 732-Feb. 20, 733). 

Ibn Wadh. . Mu'awiya, the son of Hisham, and Maslama, the son of 
'Abd Al Malikh, made a raid. 
El. Nis. 114(0 Al Tab.. And among the events was the raid of Mu'awiya, the son of 

6224 Hisham, upon the north in the summer, and the raid of Solomon, the sou of 
Hisham, upon the south in the summer ; and it is recorded that Mu'awiya, 
the son of Hisham, smote the suburbs of Akrun (Akroinon), and that 'Abd 
Allah Al Battal and Constantine met with their forces ; and he roiited them 
and took Constantine prisoner. And Solomon, the son of Hisham, reached 

115 (Feb. 21, 733-Feb. 9, 734). 

Ibn Wadh. . Mu'awiya and Solomon, the sons of Hisham, made a raid, 
and over the advance-guard was 'Abd Allah Al Battal ; and he met Con- 
stantine and took him prisoner and routed the Romans.' 

Al Tab. . And among the events of this year was the raid of Mu'awiya, 
the son of Hisham, upon the land of the Romans. 

* The text of Al Tab. lias Al Mnttalib. I * i.e. the valley : see ami. 75 ami note, 

insert ' 'Al)d ' from Ibn Al Athir. 


Kliit.'ib Al 'Uyun. Mu'awiya, tlic son of Ilisliam, in:uU- a laiil in the 
sujuiiier, and with him were tlic men of Al Shain ami tlic men of Al (iazira 
and 'Abd Allah Al Battal. And, when tho Moslems and the Romans nnot, 
and over the forces was 'Abd Allah Al Battal . , . . , the Romans 
wore routed, and the Moslems fell upon them and made great slan<,diter, and 
took many captives, and took possession of their camp and made spoil of 
their property. 

IIG (Feb. 10, 734-Jan. 30, 735). 

Ibn Wadh. . Mu'awiya, the son of Hisham, made a raid. 

Al Tab. . And among the events of this year was tho raid of Kl. Nix. no 
Mu'awiya, the sou ot Hisham, in the summer, upon tiu; land (d' the "0220(0 

117 (Jan. 31, 735— Jan. 19, 73G). 

Ibn Wadh.. Mu'awiya and Solomon, the sons of Hisham, made a raid.* El. Nis. 117 
Al Tab. . And among the events of this year was the raid of Mu'awiya, '^^'^Qo-mn 
the son of Hisham, upon the north in the summer, and the raid of Solomon, 
the son of Hisham, the son of 'Abd Al Malikh, upon the south in the summer 
in the neighbourhood of Al Gazira; and he scattered his detachments over 
the land of the Romans. 

118 (Jan. 20, 736-Jan. 7, 737). 

Al Tab.. Among the events was the raid of Mu'awiya and Solomon, El. Nis. 118 
the sons of Hisham, the son of 'Abd Al Malikh, upon tlie land of the Theoph AM 

_^ *^ 6228 


119 (Jan. 8-Dec. 28, 737). 

Al Tab. . Among the events was the raid of Al Walid, the son of Al 
Ka'ka', the 'Absi, upon the land of the Romans. 

120 (Dec. 29, 737-Dec. 17, 738). 

Al Tab. . Among the events was the raid by Solomon, the son of Hisham, Theoi-h. AM 
the son of 'Abd Al Malikh, in the summer, and his capture, as is recorded, of 
Siudira (Sideroun). 


121 (Dec. 18, 738-Dec. G, 739). 

Ibn Wadh, . Maslama, the son of Hisham, reached Malatya, 
Al Tab. . Among the events was the raid of Maslama, the sou of Hi.sham, 
the son of 'Abd Al Malikh, upon the Romans ; and in it he took Matamir. 

122 (Dec. 7, 739-Nov, 25, 740), 

Ibn Wadh. . Solomon, the son of Hisham, made a raid upon the district Theoph. AM 
of Malatya. 6231 

1 From this point down to 121 tJie text of Ibn Wadh. is defective. 

202 E. W. BROOKS. 

Al Tab. . In this year 'Abd Allah Al Battal was killed with a force of 
Moslems in the land of the Romans. 

Khitab Al 'Uyun, Al Battal, the son of Al Chusain, (his name was 'Abd 

Allah) and Constantino met with large forces ; and God Most High routed 

Theoph. AM them, and Constantino was taken prisoner. And Al Battal advanced with 

the captives, and he was attacked in the rear and killed, and with him was 

killed Malikh, the son of Shu'aib. 

Ibn Al Athir, In this year Al Battal (and his name was 'Abd Allah 
Abu'lChusain, the Antakhi) was killed with a force of Moslems in the land 
of the Romans ; and it is said also that it was in the year 123. 

123 (Nov. 26, 740^Nov. 14, 741). 

El. Nis. 123 Ibn Wadh. . Solomon, the son of Hisham, made a raid in the summer. 

124 (Nov. 15, 741-Nov. 3, 742). 

El. Nis. 124 Ibn Wadh. . Solomon, the son of Hisham, made a raid, and he met Leo,^ 

"^6233 ^'^® Emperor of the Romans, and Artiyas (Artavazd) ; and he returned, and 
there was no battle between them. 

Al Tab. . And in this year Solomon made a raid in the summer, and he 
met Leo, the king of the Romans, and carried off captives and spoil. . 

125 (Nov. 4, 742-Oct. 24, 743). 

Theoph. AM Ibn Wadh. . Al Ghamr, the son of Yazid, the son of 'Abd Al Malikh, 

^23^ made a raid. 

Al Tab. . Among the events was the raid by Al Nu'man, the son of Yazid, 
the son of 'Abd Al Malikh, in the summer. 

Caliphate of Al Walid II. 

Ibn Al Athir. This year the Romans came out to Zibatra,^ and that is 
an ancient fortress ; and it had been taken by Chabib, the son of Maslama, 
the Fihri ; and the Romans demolished it at that time ; and it -was rebuilt 
without strength ; and the Romans demolished it again in the days of 
Marwan, the son of Mahomet, the Ass.^ . . . And in this year Al Walid sent 
his brother, Al Ghamr, the son of Yazid, to make a raid.* 

1 El. Ni.s. ' the son of Leo ' ; and this is Melitene and in lat. 36° 50', long. 61° 20'. 

obviously rij^ht, since Leo died in June 741. ^ ' The ass of Al Gazira ' was a nickname of 

^ Between Melitene and Samosata and Al Marwan IL 

Chadath (see p. 208) according to Yakut. Per- * Ibn Wadh. is therefore wrong in ascribing 

haps it should be identified with Deba (mod. this raid to the reign of Hisham, who in fact 

Tshebat). Abu'l Fida (Tab. Syr. pp. 28, 30) died in Feb. 743. 
places Zibatra two days' journey south of 


Extracts from Al Baladiiuri. 

The Affair of the Guragima} 

And in the days of Ibn Al Zubair, after the death of Manvan, the son of Cf. j.. 189 
Al Chakham, when 'Abd Al Mahkh was seeking the succession to the 
Caliphate, . . . and was calling for the help of the men to go to Al 'Irak to 
fight against Al Mus'ab, the son of Al Zubair, a Roman army went out to the 
mountains of Al Lukham (Amanos) under one of their generals ; then they 
went to Lubnan (Lebanon), where was collected a large force of the Ouragima 
and Nabataeans and runaway slaves of the Moslems. And 'Abd Al Malikh 
was compelled to make peace with them on condition of paying 1,000 denarii 
every assembly-day ; and he made peace with the Emperor of the Romans for 
the amount which he was to pay him in order to prevent him from fighting 
against him, and because he was afraid he would go out to Al Sham and 

conquer it. . . . And this was in the year 70 And Maimun the 

Gurgunami was a Roman slave belonging to the sons of Um Al Chakham, 

the sister of Mu'awiya, the son of Abu Sufyan, and they were Thakafis ; and 

by birth indeed he came of the Guragima, so that he joined them and went 

out to Mt. Lubnan with them. And 'Abd Al Malikh heard that he was a 

man of prowess and valour ; and he asked his masters to set him free, and 

they did it ; and he gave him command of a military force and sent him to 

Antakhiya ; and he made a raid upon Al Tuwana in company with Maslama, cf. p. 191 

the son of 'Abd Al Malikh, (and he was at the head of 1,000 of the men of 

Antakhiya), and he was martyred after showing distinguished courage. And 

'Abd Al Malikh sent a large army to make a raid upon the Romans in order 

to exact vengeance for him.^ 

The Frontier of Al Sham {Syria). 

I was informed by some elders of the inhabitants of Antakhiya : they 
said : The frontier of the Moslems in Al Sham in the days of 'Umar and 
'Uthman (God be gracious to them) and the succeeding sovereigns, was 
Antakhiya and other cities, which Al Rashid called 'Awasim ^ ; and the 
Moslems used to raid the country beyond just as now they raid the country 
beyond Tarsus. And between Al Iskhandaruna (Alexandria by Issos) and 
Tarsus the Romans had fortresses and armour-stores like the fortresses and 
armour-stores by which the Moslems pass at the present day. And some- 
times their inhabitants left them and fled into the territory of the Romans 

^ i.e. Mardaites. aTpirevat ViaaaX^iMs «ol 'A^SasiT^j/ Tvavov Sia 

- This account is at variance with that of Al t^v naviav tov anoKTayOfVTos aTparov avy r^ 

Tab., at least according to the most obvious hlaiov/j.q. xnth Maptavov) uccorda with Al Bala- 

meaning of that writer's words, for he certainly dhuri, 

seems to represent Maimun as being on the ' i.e. defences. 

Roman side. The account of Theophanes {ine- 

204 E. W. BROOKS. 

in fear ; and sometimes Roman fighting men were moved into them to 
occupy them. And it is said that Herakleios brought men with Ijim and 
stationed them in those cities, when he retired from Antakhiya, lest the 
Moslems should come and colonize the land between Antakhiya and the 

territory of the Romans. And God knows And 

there is a difference as to who was the first to pass the Gates (these are the 
Cf. !>. 183 Gates of Baghras (Pagrai) ). And some say : They were passed by Maisara, 
the son of Masruk, the 'Absi, who was sent by Abu 'Ubaida, the son of 
Al Garrach ; and he met a Roman force accompanied by some Musta'riba from 
Ghassan and Tanukh and lyad, who were going to join Herakleios ; and he 
attacked them and slew a large number of fighting men from among them. 
Then he was joined by Malikh Al Ashtar, the Nakha'i, with reinforcements 
from Abu 'Ubaida, who was at Antakhiya, And others say : the first who 
passed the Gates was 'Umair, the son of Sa'd, the Ansari, when he was sent 
on the matter of Gabala, the son of Al Aiham. 

And Abu'l Khattab the Azdi says: I have heard that Abu 'Ubaida 
himself made a summer raid and passed by Al Massisa and Tarsus ; and the 
population of these places and the neighbouring fortresses emigrated : and he 
passed through the Gates, and his raid extended as far as Zanda.^ And 
another account says : he sent Maisara, the son of Masruk, and he reached 

I was informed by Abu Salich Al Farraa, who had it from a man of 
Dimashk (Damascus) called 'Abd Allah, the son of Al Walid, who had it 
from Hisham, the son of Al 'Az, who had it from 'Ubada, the son of Nusa, 
Cf. p. 183 as Abu Salich thinks ; he said : When Mu'awiya made a raid upon 'Ammu- 
riyya in the year 25, he found the fortresses between Antakhiya and Tarsus 
deserted ; and he stationed in them a force taken from the men of Al Sham 
and Al Gazira and Kinnasrin, until he returned from his raid ; then a year or 
two years afterwards he sent Yazid, the son of Al Chur, the 'Absi, on a 
summer raid ; and he gave him orders, and he acted accordingly, and the 
officers did his bidding. And this man said ; And I found in the book of 
Aug. 24, 651- the raids of Mu'awiya that he made a raid in the year 31 in the district of 
Aug. 11,652. j^i Massisa, and reached Darauliyya^; and, when he went on the expedition, 
he did not pass by any fortress between him and Antakhiya \^jithout de- 
stroying it. 

And I was informed by Mahomet, the son of Sa'd, on the authority of 
Cf. p. 191 Al Wakidi and others : he said : In the year 84 'Abd Allah, the son of 'Abd 
Al Malikh, the son of Marwan, made a raid in the summer, and he entered 
by the Gates of Antakhiya ; and he came to Al Massisa and built its fortress 
upon its old foundations. And he planted in it a colony taken from the 
army, among whom were 300 men, whom he had selected from among those 
possessed of valour and distinguished courage ; and the Moslems had not 

' Yakut mentions Zandan near Mopsouestia ^ Perhaps we should read Adhruliyya or 

and quotes Khalifa, the son of Khayyat, as re- Ardaluniya (see p. 192). Dorylaion seems 

cording a raid upon it by 'Abd Allah the son of impossible, though that is the name usually 

Sa'd the son of Abu Sarcli in the year 31. represented by Darauliyya. 


colojiizcd it before that time. And lie built a mos<iue in it close to the hill 
of the fortress. Then he went on with his army till he made a raid up«m 
the fortress of Sinan and took it; and he sent Yazid, the son of Chunin, the 
Tai, the Antakhi ; and he made an incursion and then returned to him. 
And Abu'l Khattab the Azdi said : The first in Al Islam who built the 
fortress of Al Massisa was 'Abd Al Malikh, the son of Marwan, actinj^ 
through his son, 'Abd Allah, the son of 'Abd Al Malikh, in the year H4 upon 
its old foundations : and the building and garrisoning were completed in the 

year 80 He said : And 'Umar, the son of 'Abd Al 

'Aziz, journeyed till he came to the granary of Al Massisa; and he wished to 
destroy it and to destroy the fortresses between it and Antakhiya. And he 
said, " I am afraid of the Romans besieging the inhabitants of it." And the 
men told him that it had been colonized in order to keep the Romans who 
were in it away from Antakhiya ; and, if he laid it waste, there woulil be 
nothing to stop the enemy until they came to Antakhiya. And he gave up 
the idea and built a general mosque for the inhabitants in the district of 
Khafarbayya. ... He said : Then Hisham, the son of 'Abd Al Malikh, 
built the suburbs ; then Marwan, the son of Mahomet, built the booths on 
the east of the Gichan (Pyramos), and round it he built a wall, and set up a 

wooden gate in it and dug a trench 

They (the elders of the frontier) said : And the man who fortified Al 
Muthakkab^ was Hisham, the son of 'Abd Al Malikh, acting through Chassan, 

thesonofMahuwaih, the Antakhi And Hisham built the fortress 

of Katarghash '^ by the instrumentality of 'Abd Al 'Aziz, the son of Chayyan, 
the Antakhi; and Hisham built the fortress of Mura by the instrumentality of 

a man of Antakhiya And Hisham built the fortress of Buka ^ 

in the territory of Antakhiya ; then it was restored and renewed 

. . . . And Abu'l Khattab says : The bridge on the road to Adhana 
(Adana) from Al Massisa (and that is 9 miles from Al Massisa) was built in the 
year 125, and it was called the bridge of Al Walid ; and that was Al Walid, 
the son of Yazid, the son of 'Abd Al Malikh, the murdered 

The Frontier of Al Gazira {Mesopotamia). 

They said : When 'Uthman, the son of 'Affan, (God be gracious to him) 
became Caliph, he wrote to Mu'awiya, appointing him Wali of Al Sham ; and 
he made 'Umair, the son of Sa'd, the Ansari, Wali of Al Gazira ; then he 
superseded him, and united Al Sham and Al Gazira and their fortresses in 
the hands of Mu'awiya. And he ordered him to make a raid upon Shimshat 
(Samosata), and that is in Fourth Arminiya, or send someone else to make a 
raid upon it. And he sent thither Chabib, the son of Maslama, the Fihri, and 
Safwan, the son of Mu'attal, the Sulami : and they took it some days after 

* See Tomaachek, Zur Historischen Topogr. ' Near Mopsouestia according to Yakut. 

V. Kleinasien, p. 71 (Wiener Akad. Sitzungsber. ^ From Al Bal. p. 159 it appears that this 

Bd. 126). place was close to Mt. Amaros. 

206 E. W. BROOKS. 

they had encamped before it on the same terms as the capitulation of Al Ruha 
(Edessa) ; and Safwan remained in it, and there he died at the end of the 
Cahphate of Mu'awiya. And it is said : No, the man who made the raid 
upon it was Mu'awiya himself, and Hadhan with him ; and he made Safwan 

Wali of it, and he settled in it and died there 

And they said : Chabib, the son of Maslama, made a raid on the fortress of 
Khamkh after the capture of Shimshat, and could not take it. And Safwan 
made a raid upon it, and did not succeed in capturing it. Then he made a 
raid upon it in the year 59 ; and that is the year in which he died ; and with 
him was 'Umair, the son of Al Chubab, the Sulami ; and 'Umair mounted the 
wall and never ceased fighting upon it alone until the Romans retired, and 
the Moslems climbed up and took it for 'Umair, the son of Al Chubab. And 
he gloried in this and was glorified for it. Then the Romans recovered it, 
and Maslama, the son of 'Abd Al Malikh, took it ;^ and it never ceased being 
taken and being recovered by the Romans 

Malatiyya. And they said : 'lyadh, the son of Ghanm, sent Chabib, the 
son of Maslama, the Fihri, from Shimshat to Malatiyya, and he took it ; 
then the gates were shut. And, when Mu'awiya became Wali of Al Sham 
and Al Gazira, he sent Chabib, the son of Maslama, thither, and he took it by 
storm : and he settled a colony of Moslems in it with an administrator. And 
Mu'awiya came to it when he wished to enter Roman territory ; and he 
garrisoned it with a force taken from the men of Al Sham and Al Gazira and 
others. And it was on the road of the summer expeditions. Then its in- 
habitants migrated from it in the days of 'Abd Allah, the son of Al Zubair, 
and the Romans came out and pidled it down ; then they left it, and some 
Armenian and Nabatean Christians settled in it. 

And I was informed by Mahomet, the son of Sa'd, on the authority of Al 
Cf. p. 197 Wakidi in his tradition ; he said : The Moslems settled in Taranda after 'Abd 
Allah, the son of 'Abd Al Malikh, had made a raid upon it in the year 83 ; 
and they built houses in it : and it is about 3 days' journey from Malatiyya, 
in the territory of the Romans; and Malatiyya was at that time deserted, 
there being no one in it except some of the subject-peoples, Armenians and 
others. And some scouts from the army of Al Gazira used to come there in 
the summer and remain in the town until the winter came on an4 the snow 
fell ; and, when this happened, they withdrew. And, when 'Umar, the son of 
'Abd Al 'Aziz, (God be gracious to him) succeeded to the government, he 
removed the population of Taranda against their will ; and this was because 

he feared danger to them from the enemy Then he settled them 

in Malatiyya, and left Taranda deserted ; and he made Ga'wana, the son of 
Al Charith, one of the sons of 'Amir, the son of Sa'sa'a, Wali of Malatiyya. 

They said : And 20,000 Romans went out in the year 123 and encamped 
against Malatiyya; and the inhabitants shut their gates, and the women 
mounted the wall with turbans on their heads, and fought. And a messenger 

^ Theoph. records its capture under AM 6203 (711). Another capture in 723/4 is recorded by 
Ibn Al Athir (above, p. 198). 


from the inhabitants of Malatiyya went out to ask for help, and the courier 
rode on until he came to Hisham, the son of 'Abd Al Mahkh, wiio was at Al 
Rusafa (Resapha) ; and' Hisham despatched the men to Malatiyya. Then 
the news reached him that the Romans had withdrawn from it, and he called 
the messenger and told him ; and he sent with him some horsemen to keep 
guard in it. And Hisham conducted a raid himself; then he came down to 

Malatiyya, and stayed in it until the building was completed 

And they said : Abu 'Ubaida, the son of Al Garrach, when he was at Manbig 
(Hierapolis), sent Khalid, the son of Al Walid,.to the district of Mar' ash ; and 
he took the fortress upon condition of the people migrating. Then he left it 
deserted ; and Sufyan, the son of 'Auf, the Ghamdi, when he made a raid 
upon the Romans in the year 30, started from before Mar'ash, and marched Sept. 4, 650- 
through the territory of the Romans. And Mu'awiya built the city of ^"^' ^^' ®^^ 
Mar'ash, and stationed a military force in it. And after the death of Yazid, 
the son of Mu'awiya, the Roman attacks upon them increased, and they 
withdrew from it ; and 'Abd Al Malikh made peace with the Romans after 

the death of his father Marwan, the son of Al Chakham And in Theorh. am6i84(?); 

the year 74 Mahomet, the son of Marwan, made a raid upon the Romans, and Mich.''foi,'26o tv) 

broke the peace. And in the year 75 Mahomet, the son of Marwan, also made 

a summer raid ; and the Romans came out from before Mar'ash to Al A'mak^ cf. p. i8» 

in Gumada I.,^ and the Moslems overcame them ; and their commander was 

Aban, the son of Al Walid, the son of 'Ukba, the son of Abu Mu'ait, and with 

him was Dinar, the son of Dinar, a mauli^ of 'Abd Al Malikh, the son of 

Marwan, and he was governor of Kinnasrin and its territory. And they met 

in the valley of Mar'ash and engaged in a stubborn fight, and the Romans 

were routed, and the Moslems pursued them, slaying and taking prisoners. 

And this year Dinar met a Roman force at the bridge of Yaghra,* which is 

about 10 miles from Shimshat, and defeated them. Then Al 'Abbas, the son 

of Al Walid, went to Mar'ash, and stayed there and fortified it, and removed 

the men into it And in the days of Marwan, the son 

of Mahomet, when he was occupied in fighting against the inhabitants of 
Chims, the Romans came out and besieged the city of Mar'ash, until its 
inhabitants capitulated on condition of being allowed to migrate. And they 
went towards Al Gazira and the province of Kinnasrin with their families. 
Then they destroyed it. And Marwan's governor over it at that time was Al 
Khauthar, the son of Zufar, the son of Al Charith, the Khilabi; and the 
Emperor at that time was Constantine, the son of Leo. Then, when Marwan 
had finished the affair of Chims, and had destroyed its wall, he sent an army 

^ See p. 189, note 5. about 11 miles from Samosata, which may 

^ Aug. 28-Sept. 26, 694. The Syriac writers perhaps he meant. The lake Al Yaghra near 

place the battle in AS 1006 = AH 76, in which the Syrian Gates (Tomaschek p. 74) is of course 

Gum. I. =Aug. 17-Sept 15, 696. out of the question. Abu'l Fida (Tab. Syr. p. 

' i.e. slave or freedman. 153) makes the river Al Yaghra a tributary of 

* Perhaps the bridge over the Singas. This, a river which flows into the Lake of Antioch, 

however, according to Kiepert's map is 25 Roman but no such river passes any\^•here near 

miles from Samosata. There is a smaller river Samosata. 


to build Mar'ash ; and it was built and re-founded. And the Romans came 

out during the civil war and destroyed it 

They said: And the fortress of Al Chadath^ was among those that were 
taken in the days of 'Umar, its captor being Chabib, the son of Maslama, in the 
name of 'lyadh, the son of Ghanm ; and Mu'awiya restored it after that. And 
the sons of Umayya called the gate of Al Chadath 'Al Salama Al Taira,'^ because 
the Moslems were cut to pieces in it ; and that was Al Chadath, as some men 
say. And some say : A young (chadath) lad with his companions, met the 
Moslems at the gate, and fought against them ; and it was called the gate of 
Al Chadath. And in the time of the civil war of Marwan, the son of Mahomet 
the Romans came out and destroyed the city of Al Chadath, and removed the 

inhabitants from it, as they did at Malatiyya They 

Cf. p. 185 said : And Malikh, the son of 'Abd Allah, the Khath'ami, who was called 
' King (malikh) of the summer raids ' and was one of the men of Filastin 
(Palestine), made a raid upon the territory of the Romans in the year 46, and 
carried off much spoil. Then he retired ; and, when he was about 15 miles 
from the gate of Al Chadath, at a place called Al Rahwa, he stayed there 
three days and sold the spoil and divided the captured arrows : and that Al 
Rahwa was called Rahwa Malikh, They said : And Marg 'Abd Al Wachad 
was a pasturage reserved for the horses of the Moslems. And, when Al 
Chadath and Zinatra^ were built, they had no need of it, and it was sown. 
They said : And Zinatra was an old Roman fortress ; and it was taken at the 
same time as the old fortress of Al Chadath, its captor being Chabib, the son 
of Maslama, the Fihri. And it stood until the Romans destroyed it in the 
days of Al Walid, the son of Yazid ; and it was rebuilt without strength ; and 
the Romans encamped before it in the days of the civil war of Marwan, the 
son of Mahomet, and razed it to the ground. 

E. W. Brooks. 

1 Between Melitene and Samosata and Ger- * So the MSS. : de Goeje would substitute 
manikeia according to Yakut. ' Zibatra, ' which differs only by a point and is 

2 i.e. ' the unstable security.' the form given by Ibn Al Athir (see p. 202). 


P. 208, Note 3. — Zibatra is no doubt the Sozopetra of Kedrenos 
(2, p. 130) ; but, as there seems to be no earlier authority for this name, 
it is perhaps only a Hellenization of Zibatra. 


This paper is an attempt to interpret certain stones, wliich have come 
to light recently on ancient sites in Karia, as parts of ancient oil-presses, on 
the ground that they are well adapted to fulfil certain purposes which are 
still essential to the modern native process of oil extraction in that part of 
Asia Minor and in the adjacent islands. The inference is that the ancient 
process closely resembled the modern in the principal features which are re- 
counted below. 


The Modern Method of extracting olive oil consists of the two processes 
of grinding and pressing. 

In the most primitive mode of grinding which is still in use, the olives 
are crushed either on a flat stone by a roller,^ or in a stone trough by a mill- 
stone rolling on its edge. In more modem grinders two mill-stones are used, 
which revolve in a circular trough, as in the grinding of kaolin or cement. 
The process of grinding seems never to have varied, except as regards the 
power which is employed ; horses having been substituted for men, and steam 
for horses. In Algeria and Tripoli the circular trough goes back at least to 
Roman times.^ 

The crushed olives are called Trvprjva, a term which properly refers to the 
broken kernels. The Trvprjva is exposed to the pressure in bags of hair-cloth, 
which in large presses are piled one upon another, after being well drenched 
with hot water. In the ordinary process, each bag is pressed twice, and is 
drenched again with hot water in the interval, to facilitate the extraction of 
the oil. Owing to the supply of water which the process demands, oil-presses 
are generally established near a well or a spring. 

The simplest oil-press, which is still in use in many of the Turkish 
villages in Anatolia, consists only of a stone or wooden trough in which the 
bags of TTvpfjva are placed, with a wooden plank above them, on which men 
stand to press out the oil. The trough is of oblong form, and is furnished 
with a spout by which the oil runs into a wooden tank. Similar trough- 
presses, cut in the rock, are common in many parts of Palestine, and almost 

' E.g. in the Tripolitan oil-mills described published as a 'laver' by H. SwAiNSON 
below, J). 215. CowPER, Antiquary, Feb. lBi^&.=Hill of the 

2 For Algeria : cf. TissoT, VAfrique Graces (1897), p. 151. 
Romaine I. p. 288 : A Tripolitan example is 




indistinguishable from the wine-presses. There is a good example among the 
old olive-trees at Gethsemane.^ 

Much more effective and elaborate presses, however, are in use in the 
islands and in some parts of the mainland ; and several stages of advance can 
be traced, leading to the steam and hydraulic presses which have lately 
been introduced in Lesbos and elsewhere to meet the increased crops of 

In the common screw press which may be seen in most parts of the 
Levant, ten or twelve bags of Truprjva are piled between two wooden plates, of 
which the lower is fixed while the other is brought down upon the pile by the 
screw. The lower plate, or press bed, is covered by a shallow iron trough, and 
in some cases is itself of solid iron. The screw is still often of wood, though 

Fig. 1. — Oil Press recently in use at Auginunta in Kalymnos. 
[ The press-bed and weight-stone are unshaded. ] 

iron screws are being introduced for large presses. The power is applied at 
first by means of single levers set in the screw-head ; afterwards, as the work 
proceeds and more power is required, by auxiliar}^ levers which are attached 
to the first by a rope, and turn on a second shaft at a little distance from the 
press. In a Cretan press with iron screw, seen by J.L.M. at Ag. Theodoro in 
Selino Province in 1893, a regular tackle was used instead of an auxiliary 

In the presses above described, the head of the screw bears directly upon 
the upper plate of the press, as in a hand printing press. But a simpler and 
probably very archaic example is still in use, in which the screw acts on the 

* Of. MiCAii vi. 15 ; Benzinger, Hehrdische 
ArchdoJogie, p. 212; Tiustkam, The Land and 
the Book, p. 207 ; and for the more elaborate 

])rocesses now in use in Palestine, Tristram, 
I.e. p. 338-9. 


plate indirectly, and with increased effect, by means of a long lever. An 
example (Fig. 1) seen at Arginunta in Kalymnos will serve to illustrate this 

The press is set near the wall of the lionse, and consists of a large stone 
block, with a circidar channel in its upper surface (of smaller diameter than 
the ancient press-beds described below) by which the expressed oil runs off 
into its tank. In the wall is a large s(|uare hole, to serve as fulcrum for the 
smaller end of a squared tree-trunk, which forms the beam of the press. The 
upper side of the beam-end is fashioned so as to offer a good surface to the 
upper side of the hole, in every position of the moving beam. The larger end 
of the beam is traversed vertically by a screw-hole, in wliich the great wooden 
screw travels, with free point upwards, and lever-fitted shank downwards. 
The wide head of the screw carries a perforated board, from wliich a large 
block of stone is suspended by wooden tenons dovetailed into mortised sockets 
in its sides. The head of the screw revolves freely beyond the board in 
a cavity in the upper surface of the stone. 

Consequently, when the apparatus is adjusted above a pile of bags of 
Trvpfjva, a forward turn of the screw lifts the stone from the ground, and 
brings its full weight into play at the end of the long lever. This arrange- 
ment, though the maximum pressure is reached at once, when the stone is 
lifted, and can never be very great, is free from the danger of excessive strain, 
to which a timber-frame press is liable, in which unlimited pressure can be 
applied directly by the screw. 

Sucli oil-presses were formerly common in Kalymnos, but only two 
survive. In the figure (Fig. 1) the whole of the woodwork is drawn from 
that of the modern press of Arginunta ; but the bed of the press is that from 

An almost identical press is described by Carsten Niebuhr {Reisen 1. p. 
151, PI XVII. D.) as being in use in Egypt: the only difference being that 
the solid beam is replaced by a substantial box (practically a hollow girder), 
which is filled with stones to give additional weight. 


The Ancient Method closely resembled that which is still in use, as may 
be inferred from the literary sources, and from the following fresh data. 

(1) Monolithic troughs for grinding the olives are to be seen in the 
Milesian territory, e.(/. two seen by W. R. P. (22 Sept., 1893) on the road 
between Yeronda (Branchidae) and Akkeui (v. map ^). But they have not 
been noted near the oil-presses of Karia and Kalymnos which are described 
below, though they are an essential part of the apparatus for every known 
process of extraction. It is curious also that no presses have been seen 

1 W. R. P. writes from Kalyiimos (31 Oct, verted into a more modern type. 
1896) that even this example has been con- ^ J.H.S. xvi. PI. X. 

p 2 



associated with these troughs, but it is possible that some of them may 
represent the prototype of the Anatolian tread-press described above ; on the 
other hand, the press-beds mai/ have been of wood. These Milesian troughs 
are found in a region which is now quite denuded of its olive-trees, on sites 
which are so thickly strewn with pottery that they must have been occupied 
during a long period. Unfortunately no pottery of characteristic styles has 
yet been noted on these sites, and much might very well be medieval. 

(2) Actual oil-presses have been observed on several sites : the most im- 
portant are as follows : — 

(a) On a roughly fortified summit in the Menteshe valley on the north- 
east side of Latmos, a large flat stone was found by W. R. P. in siht at a short 
distance from the wall of the building or enclosure. The dimensions of the 
block, which is represented in figure 2, were L. 2.5 m. x B. 24. m. In the 
middle of one of the shorter sides a rectangular projection was left, level with 
the top of the block, about 0'4 m. broad, and standing out 2 m. from the 
side. In the top of this a deep channel of 0'15 m. diameter was cut, so that 
the projection served as a spout. The channel was continued to meet a 

Fig. 2. — Press Bed from the 
Menteshe Valley in Karia. 

Fig. 3. — Press Bed from 
Senam el-Ragud in Tri- 

Fig. 4. — Press Bed from 
Emporio in Kalymnos. 

circular channel of the same dimensions cut in the top of the block, nearly 
in the centre, but a little towards the side of the radial channel. The spout 
and channel were directed away from the wall of the enclosure above men- 
tioned ; and in this wall, exactly in the same line with it, was a horizontal 
hole 0"60 m. high, 0*40 m. wide, and 0'36 deep from the face of the masonry. 
Its structure shows that it was intended to hold a large beam, and that the 
beam was intended to resist a thrust either from above or from below. 
In Fig. 3 is represented an essentially similar stone from a Roman site 
Senam el-Ragud in the African Tripoli ; figured by H. Swainson Cowper, 
Antiquary, Feb. 1896. = The Hill of the Graces (1897), p. 149, and described 
by him as an ' altar ' of prehistoric date. 

It can hardly be doubted that we have here the press-bed of an ancient 
olive-press. The bag of crushed olives, which must in this instance have been 




single, and very large, would be placed within the circular channel, and the 
receptacle for the crude oil beneath the spout. The hole in the wall would 
receive the fulcrum end of the great lever ; while the power would be applied 
to the free end, beyond the oil vessel. In this instance, however, it is not clear 
by what means the power was applied. 

(/3) A similar press-bed was noted by W. R. P. in the same neighbour- 
hood, and here also there was a hole in the wall, in the same position. 

(7) The press-bed at Azajik near Myndos, indicated in our plan {J. H. S. 
xvi. p. 206, Fig. 6) lies, not as at Menteshe and Emporio inside the fortified 
enclosure, but outside the gate. It has one large circular channel with a 
spout, and closely resembles those from Menteshe. 

(8) A more elaborate oil-press, approximating even more closely to the 
modern type, was found by W. R. P. in a fortified^ enclosure at Emporio in 
Kalymnos, of which the plan is given in Fig. 5, and a drawing of the press- 
bed in Fig. 4. The building (c) which contains the press is oblong, with the 

Fig. 6. 

door in one of the long sides. The press stands at the further end of the 
opposite side. The bed of the press (d) is a heart-shapetl stone, with a pair 
of circular bases like that of the Menteshe press, but surrounded by channels 
which widen and then converge into a common spout. Below the spout is 
a permanent tank (e) for the oil, oblong in form, and bounded on its further 
side by a strong wall of masonry, which probably served some purpose in 
connection with the press. The whole press presents the closest analogies to 
the Menteshe presses above described ; and there is no reason to believe that 
it is of other date than that of the fort, which is of the fourth or third 
century B.C. 

(e) A press-bed of very simple type, Fig. 6, was recently excavated by 
peasants on the acropolis of Lykastos (modern Astri'tza) in central Crete, and 
examined in 1895 by J. L. M. In this case the upper surface of the stone is 
completely hollowed into a shallow flat-bottomed trough of ovoid outline, 
with the spout at the pointed end : small channels are cut in the bottom to 

' We know no special reason why these sites. That at Enipori6, as the plan shows, can 
presses are so fre([uently associated with fortified hardly have been a mere place of residence. 



direct the oil towards the opening. The absence of a central platform for the 
bag of olives seems to indicate an earlier, at all events a less advanced, type 
uf press, even than that of the Menr,ushe press.* It is difficult U> c'xj)lain both 
the large diameter, and the circular shape of the press-beds from Menteshfe 
and Emporiu : that at Argiminta is oblong, and this is the common modern 
form. The facts also, that they are of stone, and so much larger in area than 
the modern press beds, and that no grinding troughs have been found near 
them, perhaps indicate that they were employed also, with a cylindriral 
roller, for the preliminary crushing. This view was also exjiressed to W. U. P. 
by the experienced \V(jrkman who remodelled the y)ress at Arginunta ; and in 

Fio. 7. 

the modern domestic oil-mills of the African Tripoli a raised cylindrical 
pedestal of G — 8 feet diameter built of rubble, with a slightly concave upper 
surface paved with hard stones, is used in a similar fashion, the olives being 
bruised under a roller, which is usually a fragment of a granite column. 

(^) Mr. Cecil Smith has kindly permitted the publication of yet another 
type of press-bed (Fig. 7) photographed by him at Klimatovimi in Melos, near 
the site of the Hellenic town. The stone is much damaged, but appears to 

1 Cf. the even ruder, and spoutless troughs, 
figured in Prof. Flinders Petrie's Tcll-el-Hesy, 
p. 55 (from Somerah), p. 58 (from IVady 

DCmeh) ; and descriKud by him as washing 



have been rectangular; the trougli is circular, and well worked, with a flat 
bottom and no central pedestal. The drain is in the rim at one side. The 
stone bears, round the rim of the trougli, in well-cut sixth century Melian letters 
the inscription EvpvavaKTiSdv, doubtless the name of the family to whom 
the press belonged. The inscription has been published by M. Holleaux, 
B.C.H. II. (1878), p. 521 ; who, however, failed to recognise the purpose of the 
block, and defends the erroneous reading KvpvafiaKTLSdv. It will be found 
under No. 3G in Hiller von Gaertringen's forthcoming volume of the Island 

(3) The mechanism by which the power was applied is not indicated in 
any of the above examples. 

(a) A stone with two sockets (Fig. 5, No. 4) which lay near the oil-press at 
Emporio, may perhaps represent the clamped stone which weights the screw 


1 MMMi 


Fig. 8. — Weiqhtstones. 

1. a. — c. from Kalloni in Lesbos : 2 a. — c. from Argindnta : 3 a. — c. from Amorgos. In each case 

a = top view, 6 = longitudinal section, c = end view. 

at Arginiinta : but it is without close parallel, either ancient or modern, and 
looks much more like a common type of door-sill. 

(/9) A stone, however, photographed by J. L. M. in Amorgos in 1893 
(Fig. 8, No. 3) corresponds very closely with that at Arginunta. The dovetailed 
sockets for the suspension beams are clearly visible ; the size and general 
proportions are nearly the same as in the Kalymniote example ; and it was 
found close to a building of similar appearance and date to that at Emporio, 
one section of which seemed to have been used as a vat or cistern. The 
press-bed, however, was not to be seen, and the walls w^re too far destroyed 
to leave any sign of a beam-hole. 

It still remains to be shown whether the Amorgine stone was attached to 
a screw, or was raised by tackle so as to bear upon the end of the beam : but 
the close likeness between the two stones suggests that the whole mechanism 
was of the same type in both cases. 


Au essentially identiciil slonc, t'roiu Kasr .Sciiuina in Tripuli, is lij^'urcd by 
Mr. Swainson Cowpur,' and others are I'reijuent on the sites of the Kon'ian 
oil-factories throughout that district. 

(7) An even closer parallel is given by a stone found by W. K. P. on the 
west shore of the Gulf of Kalloni in Lesbos (Fig. 8, No. 1). In this the grooves 
in the sides are not mortised, but run with parallel edges down the whole side; 
so that the stone must have been set in a regular frame. On the other hand, 
the hollow in the upper surface, to receive the screw-head, is a detail which is 
wanting in the example from Amorgos. This example is unsymmetrical, a 
peculiarity which still needs explanation ; but perhaps the stone has been 

(8) The stones figured in Koldewey's Leshos, p. 35, etc. and pronounced 
by him to be parts of wine-presses, are perhaps also from oil-presses of the 
same type. 

W. R. Paton. 
J. L. Myhes. 

1 Antiquary, Feb. 1896.=IIill of the Graces (1897), p. 150. 


The matter of this paper lias been a subject of consideration witli me 
for some time past, and I venture to put forword tlie conclusion I have 
arrived at, not because I consider it to be a certain one, but as possibly 
affording a working hypothesis providing an explanation of what has been to 
me, and may have been to others, an obscure and difficult point. That the 
subject demands the earnest attention of those who study Timkydides will, I 
think, be generally admitted, and this, together with the fact that I have 
formed the conclusion on a certain amount of first-hand experience, may afford 
some excuse for the publication of my views. 

The vast majority of the incidents in the Peloponnesian War are treated 
by Thukydides with great brevity, in some cases with a brevity dispro- 
portionate to their importance. There are, however, according to the ordinary 
acceptation, three incidents into which he enters with a peculiar and striking 
amount of detail. 

(1) The Siege of Plataea. 

(2) The operations at Pylos and Sphakteria. 

(3) The Siege of Syracuse. 

I say 'according to the ordinary acceptation* advisedly, because I venture 
to think that there are really /o?«' narratives, viz. : — 

(1) The Siege of Plataea. 

(2) The Siege of Pylos. 

(3) The Siege of Sphakteria. 

(4) The Siege of Syracuse. 

i.e. thus (2) in the original list consists of two different stories. 

I came to that conclusion in 1895 after an examination of the region of 
Pylos and Sphakteria, on the intrinsic evidence of the story as compared with 
the site, but I had not then had time to take a comprehensive view of the 
general problems which the story of Plataea, which I examined in 1892-3, 
taken with the Pylos-Sphakteria narrative, presented. 

The first consideration that naturally suggests itself in reference to the 
matter is that these incidents upon which Thukyjiides enlarges in so 
noticeable a manner are all of them narratives of sieges. It would on the 
face of it seem likely that there was some special reason for this. Furthermore 
there is at least one noticeable omission from the list — the siege of Potidaea, 
which, though of such importance and magnitude, is dealt with in his history 
with far less detail. 


A second consideration on the general ([ucstion is that Tliukydides was 
an historian contemporary with the events which he describes. It must 
ahnost necessarily be the case that the interests of such an historian should be 
less wide than those of one who is writing, like Herodotus, for example, of 
events which are past to him. He must necessarily be atiected by the 
interests of his audience, an audience, in the first instance, contemporary 
with the events which he is describing. To every audience the interest must 
He mainly in that which is novel to it, meaning thereby everything which 
differs from the wonted circumstances of their life. The criticism may not be 
original, but I think that it is to this that we must ascribe the pecuHar 
limitation which Tliukydides places upon the subject matter of his story. 
His history is a narrative of incidents rather than of institutions, whether 
political or social, because in the latter the contemporary historian would find 
little or nothing save what would be perfectly well known to a contemporary 
audience. Regarding his work as a military history some of his most noticeable 
omissions must, I think, be attributed to this fact. He tells us, for example, 
practically nothing of the Athenian army system, and but little of the naval 
organisation, simply, I take it, because tliese things were institutions so well 
known to the readers for whom he immediately wrote, that the account of 
them was not likely to excite much interest, nay, would rather add an 
unattractive feature to his work. 

Even if we knew nothing of the military history of the period preceding 
that in which and of which Tliukydides wrote, we might then perhaps suspect 
from the elaboration of detail with which this contemporary historian deals with 
these four cases of siege operations that there was something in this 
department of it which was peculiarly novel : that the operations relative to 
the attack and defence of fortified places had entered on a new phase of 
development within the limits of the historian's own personal experience. 
Fortunately we possess evidence of this being the case.^ 

To the student of Greek history, there are, I venture to think, few 
questions which so frequently and persistently call for consideration and 
solution, as the contrast which is presented in the military history of the 
fifth century between, on the one hand, the peculiar strength of the natural 
positions which the character of the country afforded for the Acropolis of its 
towns both great and small, and, on the other, the peculiar incapacity 
which the typical Greek army displayed in the attack on such places. The 
dilemma becomes more striking still when we consider the most prominent 
individual case among Greek armies, the Spartan, whose reputation for 
incapacity in this respect was notorious. And yet, in spite of this, this very 
army was able to maintain the hegemony of its country over a large part of 
Greece, thickly sown with fortifications of great natural strength. In 
attacking these, its only method was blockade. Nor were the other prominent 

^ The earliest example in the Greek woilJ, at the siege of Samos, v. Diod. xii. 28 and 
and this is by no means fully authenticated, Plut. Perik. 27. Both passages are from 
is the reported use of siege engines by Perikles Ephoros. 

220 G. B. GRUNDY. 

Greek armies, at any rate until tlie time of the Peloponnesian War, really in 
advance of the Lacedaemonians in this respect. The Athenians had indeed a 
reputation that way, but it was evidently the reputation of the one-eyed 
among the blind, of those who know little, among those who know nothing. 
Such details as we have of the siege of Potidaea show that at the beginning 
of the Peloponnesian War the Athenians were not far advanced in the science 
of attack on fortified places. The old passive system of blockade is the one 
adopted, and though it is in the end effective, the cost is enormous. 

The question naturally arises — how is it that the Greeks, after a long 
and frequent experience of warfare with one another, had never carried this 
special branch of the art to a higher pitch of development than that at wliich 
we find it at the beginning of the Peloponnesian War ? Furthermore how 
did it come about that a state like Sparta, in spite of its notorious incom- 
petence in this department, was able, in face of what was at times the most 
serious opposition, to exercise the strong political influence which it exercised 
over neighbouring states whose towns were provided with all but impregnable 
Acropolis ? The answer to this question seems to me to lie in the special 
nature of the land of Greece south of the Kithaeron-Parnes line. 

It is hardly necessary to say that the major portion of the area of this 
part of the country consists of mountains incapable not merely of cultivation 
but of affording aught but meagre pasturage. Interspersed among these 
mountains there are indeed plains of great fertility, but of very small extent 
compared with the area of the uncultivable land, and on the produce of 
these plains the population of those states which had not facilities for foreign 
trade was absolutely dependent. It is quite certain that in that part of 
Greece south of the Kithaeron-Parnes line the amount of cultivable land, rich 
though it was, was not more than enough, if enough, to support the then 
population of the region, and in the case of Attica the inhabitants had long 
been dependent on the supply of foreign corn. This special characteristic 
must of necessity be true of every mountainous country, but it is peculiarly 
true of Greece. 

It was this fact that rendered it easy for a force unskilled in siege 
operations to keep control over the country south of Kithaeron, and more 
especially over the Peloponnese. City states such as Tegea and Mantineia 
for example, if disposed to kick against the pricks of the Spartan 
hegemony, could always be brought to order by defeat in the field. 
And Sparta could force them to take the field. It is perfectly true 
that, had they shut themselves up in the Acropolis of their towns, the 
Spartan army could have done them but little personal injury. But they 
could not afford to do this. In the first place there can have been no 
appreciable surplus of food supply, especially in the. early summer, the 
beginning of the campaigning season, wherewith to provision an acropolis 
against a prolonged blockade. But more than this. They could not afford to 
leave their year's crop to be ravaged and destroyed by an army in the 
country. That would have meant possible starvation in the coming winter, 
without even the prospect of being able to obtain some sort of supply from 


their neiglibour.s, who themselves woulti require all the food stuff they could 
grow. The average Greek state, then, when invaded by a hostile army, had to 
take the field with its own force and match hoplite against hojjlitt! in the open, 
and in this department of tlie art of war the professional army of Sparta was 
infinitely supeiior to the unprofessional citizen soldiery of her neighbours. 

North of Kithaeron the natural circumstances of the country arc 
different, though not in direct contrast to those south of that line until the 
great plain of Thessaly is reached. Still, what is true of Thessaly is partially 
true of Boeotia. The typical Greek army in Thessaly was like a fish out of 
water. It had to meet circumstances for which it was wholly unadapted. 
The extensive fertile plains of that country permitted of the use of cavalry, 
in which arm the typical Greek army was wholly dt ficient, and furnished .so 
large a supply of. food stuff, and so large an area of cultivation, that its 
fortified places could be provisioned against a long blockade, and could afford, 
too, to see a part of their land ravaged. The form of pressure which Sparta 
could employ in Arkadia would be ineffective in Thessaly, for the devastation 
of a large area of country is a work of months, not days, nor even weeks — a 
fact not always recognised, but of which the history of the invasions of Attica 
during the Peloponnesian War supplies conspicuous proof. We can see this 
cause, amongst others, at work in determining the history of such districts as 
Boeotia and Elis. The natural circumstances of these countries placed them 
in respect to a food supply in a position more approximating to that of 
Thessaly than to that of the poorer territories. Consequently, though not 
wholly safe-guarded against external pressure, they were much less open to 
it in the form in which, for instance, Sparta exercised it. 

The reputation of Athens among the Greeks at the time of the Persian 
War in the attack on strong places, though evidently but ill deserved, was 
very possibly due to the fact that in her frontier wars with Boeotia she had 
been brought face to face with circumstances in siege operations in which 
passive blockade would be more or less ineffective, and she may well have 
been forced in attacking Boeotian border towns to employ something of the 
nature of active operations. 

But by the time of the Peloponnesian War a new factor had arisen in 
Greek warfare which had, in the circumstances of the time, very seriously 
curtailed the effectiveness of the old method of blockade. The genius of 
Themistokles had devised, and the foresightedness of Perikles had brought 
to completion, the great linked fortress of Athens-Piraeus. Blockade of so 
extensive a fortification by land would have been almost impossible, and 
would be in any case useless unless the highway of the sea could be closed. 
No amount of devastation of their territory could force the possessors of such 
a stronghold to take the field against their will and to risk all on a battle in 
the open. Even among the Spartans there were men of ability who were 
quite able to appreciate the extent of 'the possibilities which this new 
invention, thus carried out, opened in Greek warfare. Archidamos' speech as 
reported in the first book of Thukydides is sufficient evidence on that point. 
Here was Sparta's great rival occupying a home territory in which, owing to 

222 G. B. GRUNDY. 

the poverty of the land, the old system . of blockade should have been 
peculiarly effective, in possession of an engine which would render that 
system null and void. Others might well follow her example, and if so, the 
days of the Spartan hegemony were numbered 

In Attica itself, the Athens-Piraeus fortress rendered it possible more- 
over for Athens to maintain adequately the defence of other strong places in 
the country, such as Oinoii, Phyle, and Dekeleia, since, owing to their 
nearness to the great central fortress, it would not be difficult to supply them 
also with the sea-borne provisions, the obtaining of which the possession of 
the great fortress, backed by a powerful fleet, rendered so easy. The details 
which Thukydides gives us of the attack on Oinoii in the first invasion of 
Attica by Archidamos seem to suggest that Periklcs' original plan in the 
Peloponnesian War included the occupation of certain stuong places o\itside 
Athens-Piraeus itself. There is no doiibt that this part of his design, if it 
ever existed, was subsequently modified. I would suggest, however, that this 
did form part of the original design, and that the subsequent modification 
was due, partly, at an^ rate, to the fact that Pcriklos and the Athenians 
suddenly discovered that the eyes of the Spartans had been opened to the 
necessity of adopting a plan of siege operations in accordance with the new 
conditions which the fortification of Athens-Piraeus had imposed on the 
enemies of Athens. It must have been a severe shock to the military 
authorities in Athens itself to find that slow-moving Sparta had stolen a 
march in military science on their own enterprising state. That is evidently 
what had taken place. If Thukydides' evidence as to the war debate at 
Sparta be reliable, and he had ample opportunity of discovering in later 
years the outlines of what then took place, we must conclude that there 
did then exist in Sparta itself an able minority, led by Archidamos, who 
were capable of forming a just estimate of the chances and necessities 
of the coming war. They apparently saw that under the then present 
conditions Athens was practically invulnerable, if merely the old methods 
were employed, and so they introduced into Greek warfare in Greece 
an innovation, which, rightly learnt and rightly employed, offered a 
possible prospect of a decisive result. It was not rightly learnt, nor was it 
rightly employed, as the details of the siege of Plataea show us, .but never- 
theless it was an innovation which offered great possibilities. Whence the 
Greeks and especially the Spartans learnt the new system of attack we 
cannot perhaps say with confidence, but there is reason to suppose that the 
idea of it was introduced into Greece from Sicily. 

The point, however, on which I wish to insist is that active siege opera- 
tions in any form were a novelty to the generation in which Thukydides 
lived, and as such were just the sort of matter which would be seized upon 
by the contemporary historian as providing a subject which could hardly fail 
to interest his immediate audience. 

It is to this cause that I would attribute the fact that Thukydides chose 
these four incidents of the Peloponnesian War as subjects for detailed 


But there is a further point that is noticeable with regard to the subjects 

Tlic sieges described are in no two cases of the same kind. They 
resemble one another in being all of them active attacks on or defences of 
places fortified by nature, or by art, or by both, but there this resemblance 

Plataea is a case of a siege of an inland town of considerable strength of 
position with permanent fortification. 

Pylos is an example of the siege of a position of great natural strength 
defended by improvised works. 

Sphakteria of the siege of an island without artificial defence, but, owing 
to the peculiar difficulty of landing upon it, and therefore of attacking it, a 
naturally strong position. 

Syracuse is an example of the siege of a great maritime town. 

In respect to diversity the examples are well chosen. There are, however, 
two other characteristics of these narratives which may or may not be taken 
as significant. 

(1) In the three cases chosen in which some artificial form of defence is 
employed, the attack is a failure. 

(2) In all four cases the superiority of the attacking party either in 
respect to numbers, or, in the case of Syracuse, in respect to military prepara- 
tion, equipment, and experience is represented, either by implication or by 
direct statement, as being most marked. 

As I have already said, I have had the opportunity of examining three of 
these narratives on the spot, viz. those of Plataea, Pylos, and Sphakteria. 
The reports of those examinations I have published, and with the exception 
of the modification of one view expressed with regard to the blocking of the 
harbour entrances at Pylos [which has also been published], I have not 
since seen reason to change my views in any one particular. 

The result of these examinations is that I have not found a single state- 
ment of fact in either narrative which is not either wholly or partially 
confirmed by the circumstances observable at the present day. I do not 
mean to say that Thukydides was infallible. What I mean is that he did get 
good information, though I have had occasion to point out that owing to his 
not being personally acquainted with the gi'ound of which he was writing, he 
was liable to misapply his information. 

Thus far I am merely stating personal views which I have already 
expressed. I now pass on to other considerations of a more widely reaching and 
therefore more difficult kind on which I did not at the time of the publication 
of my papers venture to state an opinion. 

The impression of truth of detail which these narratives conveyed to me, 
was accompanied by an impression, which I could not dismiss, that the 
historian had deliberately, without any departure from truth of fact, intended 
to convey to the reader a general idea in all these cases, of operations on a 
much greater scale than might be deduced from a close and critical considera- 
tion of the narrative of events which he gives. 

224 G. B. GRUNDY. 

This idea is conveyed with marvellous art. It is impossible to point to 
any single detail in any one of the narratives to say ' This is a distinct case of 
exaggeration.' I believe that in any single instance in which a reader 
attempted to bring such a charge against any single passage, it would be 
seen that the historian might fairly retort ' The exaggeration is not in what I 
state, but in the meaning you attach to my statement.' 

I will take as an example the narrative of Plataea. T know that the 
impression Avhich that tale creates in the mind of the ordinary, and not 
perhaps critical, reader of Thukydides is that it is an account of siege 
operations of considerable magnitude conducted, at any rate on the side of 
the attack, by a considerable force. One need only read Miiller-Strllbing and 
those who take similar views on this question to see that this impression is 
by no means confined to the imcritical student, 

I have already said that I fully admit that it is conveyed. But the real 
questions with regard to it are, as it seems to me : — 

(a) How is it conveyed ? 

(6) Is it deliberately conveyed ? and if so, why ? 

Thukydides tells us that he aimed at getting hold of the facts of 
history and at stating them truthfully. Taking him as we find him, it is 
exceedingly unlikely that a writer of such marked intellectual acuteness 
would, after making such a positive statement, have allowed, in cases in which 
he wished to go a little beyond the truth, the fact of his so doing to be plain 
on the face of it. This is noticeably the case in the Plataea narrative. 
Thukydides, if impeached on the score of exaggeration in that narrative, 
would have a terribly strong defence, and would have no reason to fear to 
enter the witness box for examination and cross-examination. Could such a 
charge be proved in face of the defence that the historian expressly states 
that Plataea was ' not a large place ' and that the number of defenders was 
480 in all ? The exaggeration is too artful, too impalpable, to be laid hold of 
by any ordinary means of extracting evidence. Yet the mere fact that the 
story has conveyed a wrong impression to generation after generation of 
readers is sufficient proof that the exaggeration is existent in it, though in a 
form hardly susceptible to literary analysis. It is the ' Krypton ' in the air of 
the story. 

Its effectiveness is due to the natural tendency of readers in all ages all 
the world over to take the most obvious conclusion as being the only con- 
clusion possible. Into this literary trap the writer could reckon on his readers 
falling. The demonstrable exaggeration is in what is implied, not in what is 
stated, and yet after all the demonstrability of even this form of exaggeration 
rests on the strictly unsound basis that the most obvious implication is the 
most true or even the only true one. To take the most noticeable instances 
of this in the Plataea story. Thukydides never informs us of the number of 
the army which Archidamos took with him to the siege of Plataea. It is 
true, of course, that he does not tell us the actual number of the Pelopon- 
nesian armies which in previous years had invaded Attica, though he does tell 
us that in the first instance the army consisted of two-thirds of the effective 


of the various allies, i.e. its numbers must have amounted to many thousands. 
Does Thukydides mean to imply by his language in ii, 71. — ['The following 
summer the Peloponnesians and the allies ditl not invade Attica, but marched 
against Plataea. Archidamos the son of Zeuxidamos, king of the Lakedae- 
monians, was in command.'] — that this army was equal in numbers to the 
invading armies of pievious years? I am inclined to think that he docs and 
he does not. He is quite willing to give the impression of magnitude without 
in any way committing himself. That he has succeeded in so doing is 
evidenced by the interpretation which modern writers have put, not upon his 
words, but upon his silence. Numbers varying from GO, 000 to 1 20,000 have been 
seriously put forward as the probable size of the besieging force, and, it need 
hardly be added, that on such premisses most destructive arguments have been 
founded as to the credibility of the historian. It is perhaps unnecessary to say 
(I) that such numbers are absurd on the face of them, and (2) that the 
historian says nothing which could in any strict sense be taken to imply 
numbers so huge, or indeed numbers in any way resembling them. Let us 
consider for one moment what the attack and defence of Plataea meant 
to either side. 

From a negative point of view its possession by the Peloponnesians 
was of the utmost importance to them. It all but blocked the land com- 
munications of the allies of the League north of Kithaeron with those to the 
south of that line, save when Attica was in occupation of the Peloponnesian 
army, I have elsewhere had occasion to point out that there are only two 
passages across this range from the north which do not debouch on Attic 
territory, viz. that very difficult route round the west end of the range to 
Aegosthena, and the pass on the Plataea-Megara road. The latter was 
blocked, practically, by Plataea : the former was too difficult to be really 
useful. This fact must have been apparent from the very beginning of the 
war to the leading commanders of the Peloponnesian League. There can be 
no doubt that the Thebans saw it, and we may attribute the attempted 
surprise of Plataea to their apprehension of the part the town might play in 
the war which was imminent. Archidamos seems to have been aware of it, 
too, and the very first measure he took in the war aimed at its reduction. 
In Thuk. ii. 18, the historian describes the first invasion of Attica. He tells 
us that Archidamos first led the League army against Oinoe and besieged the 
place. Why of all places in west Attica should he attack Oinoe first ? It 
lay right out of the way of an army invading the country from the side of the 
Megarid, via the route which would naturally be taken, i.e. via the Thriasian 
plain. But, as those who know Attica will recognize, as indeed may be 
recognized from a map of the country, it commanded the only road from 
Athens to Plataea. He did not take the place, and therefore he had for 
strategic reasons to postpone the attack on Plataea, until experience had 
shown him that nothing would induce Perikles to risk his land force beyond 
the walls of Athens. Then and then only could he afford to ignore Oinoe, a 
small place in itself, which could only become formidable to a force besieging 
Plataea in case the Athenian main army chose to march out and use it as a 


226 G. 13. GRUNDY. 

base of operations against those engaged against the town. When an unmis- 
takable indisposition to do this was displayed by the Athenians, then, and 
not till then, did he undertake the siege. 

'It would be difficult, then, to overestimate the value of Plataea to either 
side, and it is unlikely that Archidamos would employ a force inadequate for 
the siege of it. But it was not a large place, as Thukydides tells us, and the 
defending force amounted to only 480 men. It is exceedingly unlikely that 
large contingents would be drawn from the allies for an operation which 
could in reasonable expectation be accomplished by a comparatively small 
percentage of the whole disposable force of the League. There were very 
forcible reasons for not employing more men than were absolutely necessary. 
For several years past large percentages of the able-bodied male population 
of the Peloponnese had been taken away at a time of year when they could 
very ill be spared from the harvest, a most serious disadvantage, we may 
say an insuperable disadvantage, to a population dependent entirely on the 
products of their country. The disadvantage was all the greater at a time 
when a more or less effective blockade of the Peloponnese rendered the 
importation of foreign food stuffs difficult. On the hard probabilities of the 
case there is far more reason for supposing that the force which Archidamos 
commanded at Plataea was a small one, amounting at most to a few 
thousands, than that it was one of considerable size. There is even some 
reason for supposing that the enterprise may have been but little appreciated 
at Sparta, and the support accorded to it but lukewarm. The authorities 
there seem to have been just as incapable at this time as at others of forming 
any just estimate based on large views of policy or of strategy. It is, for 
example, impossible to fail to see the correctness of the estimate of the 
chances of the war as sketched in Archidamos' speech at Sparta, yet his 
views, in spite of his high position, found apparently but few supporters. His 
object in the attack on Oinoe was completely misunderstood, and the attack 
itself much blamed, so much unappreciated indeed that Thukydides in his 
inquiries into the Peloponnesian side of the war some years later seems not to 
have heard aught of the design which Archidamos entertained, manifest 
though it is. Archidamos appears to have been of a capacity and insight far 
above the average of that of his race. 

Thukydides, and indeed, the Athenians of his time, must have had a 
pretty clear idea of the magnitude of the army which engaged in the siege of 
Plataea, viz. that it was not a huge host such as has been imagined by later 
writers on the subject, but a comparatively small force suited to the demands 
of the situation, both as regards the requirements of the attack itself, and 
also as regards the circumstances of those states from whom the army was 
drawn. Nor is there a word in Thukydides' narrative of the siege which 
would necessarily convey any other impression. So far the story was for the 
immediate audience, the contemporary audience. But there was another 
audience which the writer could not forget, the audience of the future : Krrjfia 
re 69 del fiaWov r) dycoviafia e? to Trapa^prjfia uKoveiv ^vyKeirac. For them 
the case wa.s different. The novel feature in the art of Greek warfare 


would not be merely an interest to contemporaries, but, if properly set 
forth, an interest to future generations as first examples of a new intro- 
duction into that always interesting department of history, military history. 
To be effective the examples must be varied and striking, and to be so, must 
give an impression of scale. With tlie sole exception of Syracuse the examples 
which lay to the historian's hand were not great in actual scale, and the 
impression had to be created artificially. And .so it was created. A certain 
amount of artistic reserve was introduced into the narrative : the historian 
abided by the strict truth in every positive statement he made, but by a 
certain amount of negative repression created a general impression of magni- 
tude which the positive statements did not in reality support. Had he not 
said : ol yap avOpwirot ra^ aKoa<; twi/ Trpoyeyevrjfiivcov, kuI rjv €7rt)(^(i)pia 
<T(f)icnv r/, ojxoLOif; d/SacraviaTO)'; irap' aWi^Xeov Bi'^ovrat ? 

In what relation, then, does the story stand to the strict truth ? In some 
such relation, I take it, as the type stands to the individual. The historian 
wished to create a type of this special kind of siege conducted after the new 
fashion. His very aim necessitated that the actual incidents should be placed 
])efore his readers in Avhat he judged to be their typical form ; his art consists 
in having done this rather by the creation of impression, than by deviation 
from the actual trutli. 

This characteristic of the method employed by the historian is not less 
striking when the details of the Plataea story are taken into consideration. 
An examination of the theatre of events must convince the inquirer that there 
is a solid basis of truth at the bottom of every one of them, and yet in each 
case there is an impression of magnitude conveyed which a strict examination 
of what is stated does not support. Even the tale of the mound, and of the 
works undertaken by the besiegers to counteract it, is peculiarly supported by 
the natural circumstances of the ground. It is the impression which it 
conveys which has led to discredit being cast upon it by modern criticism. 
But whatever the scale on which the operations were conducted, the general 
character of them, and the variation of method employed are peculiarly 
interesting at this stage in the history of Greek warfare. There was a 
combination of the old and the new fashion of attack. Of the new fashion I 
have no doubt that Archidamos was the champion, and that his advocacy had 
to meet a stubborn opposition from the preponderant conservative element in 
the army. Hence the new method never got a real trial, and in so far as it 
was employed seemed a failure. At the same time it is impossible to read the 
military history of this time without seeing that the opposition to the new 
system was not necessarily the only cause of its failure in this instance. We 
must take into account the inexperience of those who were now putting 
it into operation for, what was to them probably, the first time. Furthermore 
the Greek soldier of this period, unless led by a Brasidas, was not conspicuous 
for dash, and dash was required for the taking by assault of a place such 
as Plataea, manned by a garrison animated with the courage of despair. 
The new method was the first employed, the stockades being, no 
doubt, a concession on the part of Archidamos to military conservatism. 

Q 2 

228 G. B. GRUNDY. 

The new method was a faihirc, and the old method of circumvallation was 
resorted to. 

For my present purpose I have said all I wish to say about the Plataea 
narrative, save this in summary. As a narrative, it seems to me a model of 
artistic acuteness. Tlie writer, great artist as he is, has draAvn for us a word 
picture which conveys, and which he intended should convey, an impression 
greater than his subject. 

The Pylos-Sphakteria narratives partake of the character of the Plataea nar- 
rative, though they are not identical with it. I have already stated my belief 
that they are two stories, not one, and were really regarded as such by the 
historian, though they were contiguous in place and time. I expressed this 
view of the separate character of the narratives in a paper which appeared in 
the Hellenic Journal some two years and a half ago (Vol. xvi.), but I am 
afraid that the arguments by which I supported this theory were formed on 
bases of which the cogency would not be apparent to the classical scholar, 
whose attention is seldom directed to questions of physiography, and to 
whom, therefore, arguments founded on that basis do not appeal. Dr. Frazer, 
for example, in his recent edition of Pausanias, is quite severe with me on this 
question, evidently for the sole reason that he is unacquainted with the 
problems which physiography presents. 

In so far as I am able to learn, those who are interested in ancient 
history have as a rule accepted the view that the lagoon of Osmyn Aga was 
existent in some form or other at the time of the events at Pylos and 
Sphakteria. Furthermore the view has been stated, though unsupported by 
evidence, that this lagoon was an integral portion of the bay, i.e. that the 
sand-bar did not exist to any practical extent. So simple a statement, which 
everybody can understand, is sure to find acceptors. It has the further 
advantage of leaving the criticism on the historical character of the narrative 
much where it found it, viz. Thukydides made certain gross errors, as Arnold and 
others have pointed out, and, as to their explanation, you are free to adopt 
any you like. This last-mentioned method of working out the problem has the 
merit of simplicity and freedom, and is therefore attractive. For myself, I am 
convinced that those who attempt to work out the problem on these premisses 
are labouring at lost labour. I am also convinced that this fact must be recog- 
nized by the majority of students of ancient history, if only the matter be stated 
in a comprehensible form, avoiding, as much as possible, pure technicalities. I 
will therefore state the opposing view, and put in as clear a form as possible what 
seem to me the overwhelming reasons for it, and against the contending theory. 
The view is simply this : that at the time at which the events at Pylos 
took place the lagoon* was much as it is now, save (1) that the sand-bar had 
not yet extended quite up to Pylos itself, i.e. there was still a channel at the 
west end of it, and (2) that the water, at any rate in the west end of the 
lagoon, was deeper than it is at the present day. 

I am re-stating the reasons for this view, because I feel that, in my 
original paper, I did not perhaps put the matter in a form which would be 
comprehensible to every student of the subject. 


The reasons arc as follows : — 

(1) The area in the neighbourhood of" the bay where detritus has been 
deposited is lor all practical purposes co-tcnninous with the present site of 
the lagoon and its innnediate shores, [with the exception of the plain of 
Xerias on the east side of the bay, which has nothing to do with our present 

(2) On this area the detritus brought into the north part of the bay by 
the action of rain and streams has been deposited. 

(3) The streams arc small, and practically non-existent save in rainy 

(4) The formation of a sand or mud bar in the case of such deposits 
occurs, as can be seen all round the coasts of the Mediterranean, at a very 
early stage in the process of deposit. 

(5) This process of deposit must have been going on ever since the bay 
of Navarino was formed. 

Now I ask, is it conceivable that in the countless ages that intervened 
between the formation of this bay and the events at Pylos, this process had 
proceeded so slowly that this north end of the bay was still an integral part 
of it, whereas in the (geographically speaking) brief period which has 
intervened between those events and the present time, the lagoon has become 
a lagoon, i.e. at least half the work of deposit has been accomplished ? 

There can be only one answer to this question : it is impossible, even 
absurd to suppose such a thing. The other theory is more simple, but it will 
never lead to any elucidation of the Pylos narrative, since it supposes a state 
of things which neither did exist, nor could have existed at that time. 
There were two harbours at Pylos at that time, if physiography can tell us 
aught on the subject, and there are two narratives in the history of the events 
at Pylos and Sphakteria, as given by Thukydides, and these two narratives 
refer, though the author never apprehended the fact, to two different harbours. 
For the purpose of the historian, writing, as he was, an essentially military 
history, the contiguity of the narratives in respect to time and place, was 
purely accidental. His aim was to create typical early examples of a new 
kind of siege operation, and events offered him the two examples in close 
proximity to one another. Pylos is practically ignored in the Sphakteria 
narrative. The two series of events were really separated in his intention. 

It is not necessary here to point out in detail the effect which this ' two- 
narrative ' theory has on the criticism of the whole ' Pylos-Sphakteria ' 
story. I have already done this in the previous paper on the subject. It 
does, however, clear up much that is otherwise obscure in Thukydides' Pylos 
narrative, and the error in the story becomes attributable to a very natural 
mistake made by the author as to the identity of ' the harbour ' of which his 
informants spoke, and not to his stupidity, tLe last defect, perhaps, of which 
those who know his work could suspect him. I cannot but be glad that the 
' two-harbour ' theory does elucidate the narrative, but, in any case, I say 
with conviction that whether it did or did not do so, it is, leaving the history 

230 G. P.. GRUNDY. 

aside altogether, in reality not merely a theory, but a solid i'aci which 
students of Thukydides will have to take into account. 

It is very difficult to say whether wc have in these two narratives 
instances of suppressed exaggeration similar to those which are present in the 
Plataea story. The historian had better, i.e. larger material on which to erect 
his typical examples ; and the desirability of increasing tlie scale of action for 
the AVTiter's purpose, was not so evident. There is one feature present to 
which the Plataea story affords no parallel, namely the colouring of the tale 
in order to enhance the services of an individual, in this case Demosthenes. 
Thukydides had evidently a weakness for men of his type, who were ready to 
play for big stakes in the war game without reckoning the possible cost, but 
it need hardly be said that he carried his weakness at times too far, when he 
detracted from the merits of a rival in order to favour the object of his 
admiration. It seems to me that this is the only means of accounting for the 
way in which the historian treats what was perhaps the most brilliant incident 
in the two sieges, the surprise of the Peloponnesian fleet by Eurymedon and 
his colleague. 

In the Pylos story there is one incident the account of which does not 
carry conviction with it ; I refer to the tale of the attack from the sea side 
by Brasidas and his fleet. I have pointed out that of the two simultaneous 
attacks by the Peloponnesians, viz. this one by Brasidas, and the one on the 
north] wall, Thukydides can only give us details of the former. He tells 
us that the total number of assailant vessels was forty-three, whereas Demos- 
thenes had only sixty men with him. Even taking into account the difficulty 
of landing, I am strongly inclined to suspect some exaggeration here, either 
in the form of over-estimate of the numbers on one side, or under-estimate 
of those on the other. Perhaps the real fact is that the ships, though their 
number is correct, had only small complements on this occasion, a fact the 
author has studiously suppressed. At any rate it is noteworthy that we have 
here a numerical disproportion strangely resembling that at Plataea, and 
success is on the side of the infinitely weaker number in both cases. Was 
there an implied moral to the two stories ? Was Thukydides a conservative 
in military matters ? 

There is, too, in the Sphakteria narrative a very possible example of the 
exaggeration of impression. Had all the ships which Kleon used for his 
disembarkation on the island had anything like their full complements the 
force landed would have been enormous. 

In the case of Syracuse the need for exaggeration did not present itself: 
the operations were on a sufficient scale to provide an impressive example of 
the type of siege the author wished to create, though here the novelty was 
rather in the defence. 

One word before I close this article. I have, as it will be seen, sought 
to explain two points in Thukydides' work : — 

(1) The cause of his choice of subjects for peculiarly detailed 

(2) Certain peculiarities in three out of the four stories. 

The explanation seems to mo to be : — 

(1) He Avas ambitious to create typical examples of a novelty in the 
military art. 

(2) To achieve his object satisfactorily he had certainly in one, possibly 

in three, out of the four instances to create an impression of scale where he 

did not find it existing. 

G. B. Orundy. 

Professor Burrows on Sphakteria. 

In the last number of the Hellenic Journal Professor Burrows has once 
more attacked Pylos and Sphakteria, this time with the aid of certain alHes 
from the Peloponnese and a whole battery of photographs. I am, I confess, 
loth to dye this ink-stained field a deeper hue, and I have certainly no 
intention of sending friends of mine to Pylos with a view to reviving a 
controversy which has gone far enough. At the same time I do not wish to 
appear to undervalue the evidence of Messrs, Carr Bosanquet, Crowfoot, and 
Lindsay, though it is not of a very decisive character. I should there- 
fore like to point out, as briefly as possible, its true bearing, and in doing so 
I have the satisfaction of not having to treat it as hostile, though brought 
into court by the other side. 

It is rather difficult to deal with Mr. Lindsay's evidence, as given by 
Professor Burrows, relating to the south end of Pylos. Had we had Mr. 
Lindsay's first-hand account of the matter, it would no doubt have been 
easier to understand. I do not understand it sufficiently in its present form 
to attempt to discuss it in detail. I see, however, that Professor Burrows 
now not only imagines a sand-spit at the south-east corner of Pylos, but 
even makes Demosthenes draw up his ships there ; and further, refers to this 
imaginary sand-spit [against whose existence the physiographical evidence at 
present obtainable is very strong] the expression diro^daeox; Be fiaXiCTa ova-r)<;. 
How could these ships have escaped destruction in such a position, when the 
Peloponnesian fleet was in undisturbed possession of the harbour ? Surely 
the fact that these ships remained undestroyed shows clearly that they were 
drawn up at some point where landing was not easy, i.e. at the south-west 
corner; and the fact that they were outside the wall, shows that the latter 
must have been back from the shore at this point. 

The photographs which are appended to the article justify to an 
eminent degree the distrust with which I have always r.egarded such aids to 
topography. If it be merely a question of the appearance of an object which 
can be taken at a distance of a few yards, then, no doubt, the picture is of 
value, but to found arguments dealing with minute detail on such photo- 
graphs as those numbered Fig. 1 and Fig. 4 in Prof Burrows' article is simply 
misleading. Fig. 1 (of the south-east angle of Pylos) has at first sight all the 


appearances of a photograph of a gradual slope. It is not until you examine 
it closely that you see how steep the slope is. No photograph taken from 
the front of a slope can ever give anyone who has not seen it a true apprecia- 
tion of its angle. The foot of the slope appears relatively much nearer to 
the eye than the top of it, and consequently the slope itself appears much 
more gradual than it really is. The angle is, as a fact, 36°. Moreover I have 
used the term ' slope ' for want of a better word. It is really in parts 
precipitous, and perhaps the word ' cliff ' may be taken as more nearly 
describing it. Can Professor Burrows quote any example from the fifth 
century of a Greek force attempting, in face of opposition, an assault on such 
a slope, or on anything resembling it ? Fig. 4, which gives the same in profile 
from above, exaggerates somewhat in the opposite direction, and as it is out of 
focus the detail is lost in a distance. Still either photograph will justify, in the 
eyes of those who are not accustomed to imagine the Greek hoplite as making his 
way over ground which a deer stalker would with difficulty climb, my assertion 
that the practicable approach at the south-east angle of Pylos was so narrow 
that, even if landing there had been easy, which does not seen to have been 
the case, a few men could have defended it against immensely superior 
numbers, and a wall was neither required nor built by Demosthenes there.^ 
The existence of the Venetian tower and wall at this point dates, of course, 
from times when the sand-bar had reached the foot of the cliff, and there- 
fore affords no argument for the necessities of the case 2,000 years before. 

The value of photographs in topography depends, as I have said, alto- 
gether on the use which you are content to make of them. They are of 
value in giving those who have not seen ground a general impression 
as to its appearance. For minuter details, unless taken of objects a few 
yards from the camera, they are absolutely misleading to those who have 
not first-hand knowledge of the ground they represent. 

The question of the blocking of the channels is dependent on a far larger 
question, the condition of the lagoon at the .time the events on Pylos took 
place. I have expressed my views with sufficient clearness on this point in 
another article in this number of the Journal. I have not the slightest 
doubt that Thukydides had very good grounds for his repeated assertion with 
reference to the intended blocking. The difficulty to him was, and to us is, 
that he failed to recognize the existence of two sheets of water, both of which 

^ Prof. Burrows on p. 149 of his article has though he has quoted iny actual words, he has 

quoted a long passage from a previous paper of misread them. He says the photograph shows 

mine, in which occur the words : 'at this south in profile the slope to which I refer. It does 

end of the east cliflf, the summit of the cliff not. It shows in profile the slope to the 

rises to a vertical height of sixty feet above its southern, not the eastern foot. To this southern 

eastern fopt, which is only at a horizontal dis- slope I referred in the Classical Review of Nov. 

tance of eighty-one feet from that summit.' 1896 in the words 'The cliff is sixty feet high 

Professor Burrows disposes of this remark by within fifty yards of the Sikia.' But it is the 

an appeal to the photograph, and says 'Com- east slope which is of importance to the ques- 

ment is needless.' The effect of this criticism tion between us. As to comment we are in 

is unfortunately negatived by the fact that agreement. 

234 O. B. GRUNDY. 

he designated as ' the harbour.' The vagueness of his assertions on this point 
is all in accord with the caution of a cautious man who is not quite clear as 
to his information. 

Mr. Lindsay's evidence with regard to the position of the north wall is, 
considering the fact that he had the opportunity of ascertaining in detail 
Prof. Burrows' views before he went to Pylos, and went there, apparently, at 
the latter's request, not unfavourable to the views I have held. As to the 
fifth century (sic) wall, Mr. Carr Bosanquet speaks with the characteristic 
caution of an experienced archaeologist. He says ' it may well be rough fifth 
century work.' Of course it may be, but is it in the slightest degree likely 
that it is ? It may just as equally well be the work of 5, 10, 15, or 20 
centuries later. There is no special characteristic to determine its age. The 
Greek and the Cumberland shepherds at the present day build the same kind 
of wall (v. Figs. 6 and 7, in Prof. Burrows' article) in constructing a sheep 
shelter ; so would anyone else constructing a wall with similar material. The 
survival of such a wall on a site which has had such a subsequent history as 
that of Pylos is so improbable as to render identification valueless for 
historical purposes. 

I doubt whether anything can be gained by attempting to determine the 
date of this wall. Though the importance of Pylos is at present almost nil, 
it stood to the Venetian trade with the East in mediaeval and modem times 
in much the same relation that Gibraltar does to our own Eastern trade. 
It was the scene of repeated attack and defence, and even so recently as 
1825 was maintained by the insurgent Greeks for six weeks against the 
assaults of Ibrahim Pasha's force. On ground such as this, which does not 
afford soil for entrenchment, the rough stone wall is the only form of defence 

The opinion of the professional architects of the French expedition, 
quoted by Mr. Bosanquet on p. 155 of the J.H.S. Vol. xviii. must surely be 
of greater weight than that of the archaeologist, M. Bory de St. Vincent. 
Archaeologists are not exempt from the unfortunate tendency to believe 
what they wish to believe. 

Of the identity of the site of the traXaLov epvfia there can be no 
doubt. I am in full agreement with Professor Burrows on this' point. I 
was also inclined to think that he had found traces of the work above 
ground which I myself had missed. His original account of the remains 
he had discovered led me to believe that they possessed definite character- 
istics pointing to an early origin. But if these photographs, Figs. 2, 8, 9, in 
Professor Burrows' article, and the print Fig. 11 in the text convey a correct 
impression, these wall-remains present no such definite characteristics. That 
marked Fig. 9 I have not seen. Those marked Figs 2 and 8 I have. I have 
never seen walls elsewhere in Greece which at all resemble them, except 
such as are notoriously modem (e.g. the remnants of the breastworks erected 
by the Greeks at the siege of Tripolitza), roughly put together of the 
material lying at hand. Nor must it be imagined that this site has not 
passed through vicissitudes in mediaeval and modem times which must 


have led to the construction and reconstruction of numerous rough walls 
upon it. 

The north end of Sphakteria is the key to Pylos and Navarinu, as 
Hussein Djeritli pointed out to Ibrahim Paslia. Apart from the ([uestion of 
artillery fire — (and it was here that the Egyptians estabhshed their batteries 
in 1825 in the attack on Pylos) — it was an essential 2>oint cl'oppui for any 
force attacking this peninsula, affording as it did a bird's-eye view over three- 
quarters of it, and what more natural than that such a force should guard 
the summit with a rough wall, since the soil for entrenchment is conspicuous 
by its absence. Compare, too, the wall of Fig. 2 alleged to have been built 
by the Messenians with the so-called fifth century wall of Fig. 6. The latter 
is more perfect than the former, but they are the same in character. Either 
might belong to any age. Mr. Crowfoot's careful plan tends more towards 
conviction, but presents this difficulty. How is it that the wall-foundations 
in the hollow show not merely a want of connection but, in the case of the 
northern one, a manifest disconnection virith the fortification of the summit ? 
The exigencies of the ground, if Mr. Crowfoot's plan be, as it seems to be, 
absolutely correct, demand no such thing. And yet it might be supposed 
that people who were, as the plan shows, acquainted with the use of flanking 
towers, would not have constructed their work in this form. In fact it does 
not seem as if the wall on the summit and the walls in the hollow were part 
of the same design. If so, which is the remnant of the iraXaiov epvfia, and 
which is not ? 

I see that Mr. Lindsay, in deference, no doubt, to Thukydides, suggests 
an improvement on Professor Burrows' design of taking the Messenian 
captain and his men in boats to the bottom of the cliff beneath the summit 
of Sphakteria. The remodelled theory rests on the insecure foothold of an 
imaginary path from the Panagia to that point. How insecure the foothold 
is will be seen by reference to the Admiralty chart, where the cliff is shown 
as going down sheer into a depth of between thirty and sixty feet of water 
for three-quarters of the distance between the Panagia and the bottom of 
Professor Burrows' chimney. 

Mr. Woodhouse on Plataea. 

Those who are interested in Greek history should certainly read Mr. 
Woodhouse's able article on Plataea. I think he exaggerates the differences 
betweeft our views, though absolute concord on so comphcated a question is 
not to be expected. He also seems to forget that I had to take up the work 
of inquiry practically ab initio and to construct a practical basis on which to 
found an explanation of the battle. 

The points of difference between us are (1) as to the position of the 
Gargaphia spring, (2) as to the site of the Hereon of Androkrates, (3) as to 
the identification of stream A 1. 

(1) Gargaphia. — That the identification proposed by Leake and myself is 

236 G. B. GRUNDY. 

capable of dispute, I admit : but Mr. Woodhouse's main arguments against it 
appear to me to be badly founded. 

(a) He, identifying it with tlic Apotripi Spring, says that the Gargaphia 
of Leake is nearer the ' Island ' than the Apotripi. The distance is almost 
exactly the same. 

(b) He further says that the Gargaphia was by implication 10 stades 
from the R. Moloeis and the temple of Eleusinian Demeter. He evidently 
assumes that it was within the second position of the Greeks, and that, as 
they moved 10 stades before arriving at the Moloeis, etc., the spring must 
have been 10 stades from the latter. 

It seems to me, however, that the detail of the taking of the Spring by 
the Persian Cavalry, while the Greeks were still in the second position, shows 
pretty clearly that the spring was not within that position, though, of course 
it must have been near it. The position was the top of the Asopus ridge : 
the spring was at the bottom of it. 

(2) The Heroon of Androlcroies. 

Mr. Wood house would place it on the Asopus ridge, on the site of the 
Church of St. John. I cannot see how Thukydides' language can be so 
interpreted. He says, iii. 24, ' The Plataeans started from the ditch and took 
in a body the road leading to Thebes, having on their right the Heroon of 
Androkrates, thinking the beseigers would be less likely to suspect their 
having taken the road towards the enemy's country, and seeing, too, the 
Peloponnesians with torches going in pursuit along the way towards 
Kithairon and Dryos Kephalai which leads to Athens. For 6 or 7 stades the 
Plataeans went along the Thebes road, and then, turning, took the way 
leading towards Erythrai for Hysiai, and having taken to the hills, escaped 
to Athens.' 

All the ways mentioned can be identified with ease, and are, in 
two of the three cases, existent. 

The fugitives took the Thebes road, which must have been more or less 
identical in line with the present route from Kokla, which stands just above 
the ruins of Plataea, to Thebes, having on their right the Heroon. After 
going 6 or 7 stades they turned, evidently with intent to reach the 
trackway which leads from Thespiai via Pyrgos to the pass of Dryofe Kephalai 
and the sites of Hysiai and Erythrai, and, without ascending to the pass, took 
to that high bastion of Kithairon which projects into the plain east of 
Erythrai. It seems to mc that Thukydides implies that the Heroon was in 
the angle through which they turned, and I am inclined to think that the 
remnants of it will be found at one of the two stone heaps marked on my 
map, about half a mile north-east of the remains of the 'Acropolis ' of Plataea, 
They must have left the site of the Church of St. John'far away, not on the 
right, but on the left. 

(3) ^1 and the Asopus. 

In a note at the end of his article Mr. Woodhouse repeats the charge 
which Dr. Frazer makes in his Pausanias. He accuses me of inventing this 
identification of Al with the Asopus with a view to the subsequent identify- 


cation of the ' Island.' This charge is utterly unwarranted by what I have 
said in my original paper, and I may add that at the time the difticulty with 
regard to Herodotus' use of the name ' Asopus ' and its solution occurred to 
me, I had not identified the ' Island,' which came in the very last part of my 
survey. Dr. Frazer has withdrawn the charge in a letter which he has given 
me leave to publish, if I wish, wherein he admits that what I liave said 
affords no justification for it. 

There is in the article much beside what I have mentioned, but it is 
matter which I should not like to discuss without further consideration. I 
shall certainly take it very seriously into account in dealing with the wars of 
the 5 th century in book form, as I hope shortly to do. 



Mr. Mahaffy in liis preface to The Empire of the Ptolemies states 

among other problems raised by Ptolemaic history the following : ' How 

Mah. L'mp. of far does the observation, that we only know of one crown-prince with a 

p.V.° ■ wife (Soter II.) account for the divorce of that wife after his accession, and 
for the other apparent heartlessnesses in Ptolemaic history ? Is the hereditary 
title recognised in the princesses, which no doubt led to their marriages with 
their reigning brothers, a relic of Pharaonic ideas, or a mere imitation of the 
successful experiment of Philadelphus ? ' This article is an attempt to show 
that the former hypothesis is the true one, and that the marriages of the 
Ptolemies were dictated by their policy of conciliation, and were based on 
deeply rooted native prejudices. No doubt the difficulties in accepting such 
a hypothesis are very great. Why, for instance, should such a survival have 
come into comparatively greater prominence under a late dynasty ? I have 
endeavoured to sketch a possible explanation of this in the relations of that 
dynasty with the priesthood of Osiris. The hypothesis seems to clear up 
several "dark spots in Ptolemaic history and to lend the justification of 
diplomacy to actions that otherwise stand condemned by their arbitrary 
egotism and unmeaning cruelty. 

As the question seems to me to be closely bound up with the relations 
subsisting between the priesthood and the State and with the social and 
political aspect of the country generally, I shall begin by a brief examination 
of the condition of the priest-class at the time of the Ptolemaic occupation. 

It is plain that if the Ptolemies were to gain any hold over the popu- 
lation of Upper Egypt, they could only do so by conciliating or subduing the 
priest-class. The most striking characteristic of Egyptian monarchy in 
Upper Egypt and in Aethiopia had always been its entire dependence on the 
priesthood. The influence of the priest on the people was unbounded. 

Erman, Under the New Empire the priest-class had risen to a temporal and spiritual 
p. 104, seq. power unknown to earlier times. The old nobility and nomarchs had disap- 
peared : the soldier-class had dwindled owing to the employment of mercen- 
Mah. Gk. Life aries : the fid'^ifioi or military landowners under ^ the Ptolemies were 
and Thought, impoverished and had but a small and insignificant role to play ; they had 
ceased to be landed proprietors. In fact the land was entirely in the hands 
of the king or of the priests. 

Ennan, Now this rise of the priesthood as a class under the new Empire is 

p. 293. especially traceable in one place in Egypt — at Abydos, that i."? to say in the 


cult of Osiris. From this time onwards we find Osiris-worship more 
prominent throughout Egypt, and this prominence increases with the degra- 
dation of the prestige of Thebes and tlie Amon cult as a national religion, 
and also with considerable internal modification of the nature of the priest- 
hood. Thus, as Erman points out, the lay element disappears almost entirely Erman, 
from the priesthood. As the priests grow in temporal power they guard it P- ^*^- 
more jealously, and it was by the formation of this close oligarchy that they 
were enabled to bid defiance to the Pharaohs and finally to thrust them aside op. cit. p. 105. 
and absorb the supreme power into their own order. 

The removal of the seat of government to Lower Egypt left the priest- 
hood a powerful and discontented body with almost unlimited influence over 
the laity, ready to foster any signs of disaffection to the crown in the hopes 
of regaining something of their lost prestige and restoring to Thebes its 
former sovereignty. This of course applies mainly if not entirely to the 
priest-class of Thebes, centring round the cult of Amon-Ra ; and it was thus 
in the treatment of Thebes and in combating the influence of its priesthood 
that the main difficulty, I believe, of the Ptolemaic government lay. Subse- 
quent history shows how frequent revolts in this quarter were. 

To overcome the sullen defiance of the Theban priesthood seemed 
impossible, so the Ptolemies gradually adopted the policy of giving their 
special state patronage to the cult of Osiris and elevating its priesthood to a 
national importance which was justified by the wide-spread influence of the 
cult. There were obvious reasons why this cult lent itself to their policy 
better than any other. It was more generally diffused and less local in Erman, 
its connections than any other. The Osiric cycle, belonging originally to P" ' '^' 
Abydos, had lent itself to the tendency towards amalgamation which was 
so marked a feature of the religion of the New Empire. It was less rigid 
and more human, and perhaps less exclusively native than the rival 
cult of Thebes ; its mythology adapted itself to more modem requirements 
and appealed to the human sentiment of its worshippers. Osiris in his 
aspect of God of the Dead was accepted throughout Egypt as one of the great 
gods, and continued to absorb and assimilate other gods and cycles of gods 
after they had developed beyond the strictly local stage, till finally he was M^. Emp. of 
closely connected with Ptah of Memphis and worshipped in combination " P" 
with the latter as Osirhapi.^ On the other hand the Amon cycle and its priest- Erman, 
hood resisted this new anti-polytheistic tendency and remained comparatively ^' 
impervious to the influences of the age. That its predominance therefore 
remained local and centred chiefly round Thebes is not surprising. The ^™*J' 
development of its mythology seems to have been arrested ; the texts do not pooie, 
enlarge upon it as they do in the case of the Osiris-saga. • ^gyc*'! 

It was doubtless this adaptability of the Osiris cult that recommended it '^'^fng"'' 
to the Ptolemies and enabled them to graft upon it the Hellenistic elements 
of the Serapis cult. Moreover, its marked Semitic character brought it into Poole, op. cit. 
closer affinity with the religion of the foreigners— Phoenicians, Jews, P" 

Similarly Solcaris of Memphis and the pillar of Ded were identified with Oairia. Erman, 

•' p. 260. 


Samaritans, Syrians — who were so much encouraged by the state to settle in 
Egypt at this period, probably from commercial reasons. 

We can easily understand that the Osiric priesthood gladly lent them- 
selves to a policy which seemed to promise a temporal aggrandisement new 
to them. They found, however, that this temporal power was to be of a 
strictly limited nature, that the firm rule of the Ptolemies did not admit of 
a dual power in the state. The priesthood indeed was treated with deference 
and generosity ; their temple-lands were secured to them, their cults were 
elaborately honoured : in return they were expected to give religious sanctity 
to an upstart race of king-gods, and to use their influence in Upper Egypt to 
secure the loyalty of the subject and to counteract the uncompromising 
hostility of Thebes, which found vent in the frequent rebellions of Upper 
Egypt under the rule of the later kings. 

It was naturally of paramount importance to Ptolemy I to secure the 
loyalty of Upper Egypt for the sake of the trading interests of his empire. 
Probably here at the outset he came into conflict with the policy of the priest- 
class, who saw that any commercial opening up of the country would tend to 
draw away still further the life of the country to the sea-board, and strip them 
more completely of their former power. Perhaps it was this same dread on 
the part of the priests in early times that led to their intervention when the 
energetic and progressive Necho II. was engaged in constructing the proto- 
type of the Suez Canal — a canal connecting the Nile with the Red Sea — and 
thus opening up the Red Sea for commercial enterprise. At any rate they 
seem to have had the king well under their control, and according to 
Herod, ii. 158. Herodotus he gave up the undertaking owing to their representations : Ne/cw? 
/iey vvv fiera^v opvaawv iiravcraro jxavrr^iov ifxirohiov yevofiivov roiovBe, tc5 
^ap^dpo) avTov irpoepyd^ecrOai} In the New Empire the country east of the 
Delta, being on the highway to Syria, became opened and Tanis developed 
into the powerful capital and consequently a formidable rival of Thebes. 
Hence we see that the fears of the priesthood were by no means unfounded. 

It is in the Osiric cycle alone of all Egyptian cults, as they have come 

down to us on the monuments, that the female divinity plays a prominent 

Poole, part. Ra is rarely associated with any consort, as Lepsius points out, whereas 

EncI^^Br O^iris figures constantly with Isis. Isis appears, moreover, as the wife of 

p. 715 scq. Chemmis, the Egyptian Pan, the productive principle. Hathor herself is 

Brugsch, ijj later times identified with her. She is ' the goddess who is wiser than all men, 

212 sq. than all gods and spirits.' She finds out the secret name of Ra, so that the 

Turin Pap. sun-god loses his vigour and even mankind becomes hostile to him and begins 
ap. Ermaii, i n- 

p. 265. 3- rebellion. 

Now as the toundation and supreme tenet of the Osiric cult we find the 

1 It may have been owing to similarly higher level of the Red Sea ! But the 2nd 

fanatical and anti-progressive pressure from the Ptolemy, as we should expect, overcame the Diod. 

same quarter that Darius likewise desisted imagined physical difficulty by his (piXSrfX"^*' 

before completing this canal. More definite tj Sii<ppayfjia, the spiritual one by his firm 

scientific objections had to be brought to bear treatment of the priesthood, 
on the Persian king — tlie fact of the dangerously 


holy marriage of the god with his sister Isis. But Isis is not merely wife and 

queen-consort of Osiris and after his death queen-regent : even during the 

lifetime of her husband she enjoyed more than the honours of a queen-consort. 

Thus we find tov "Oaipiv rrjv JOiv oXtov rjyefiovi'au "laiSc ifj yvvatKi irapa- Diod. i. 17. 

hovra : moreover Diodorus states expressly that this illustrious precedent 

established the custom of marriages between brother and sister in Egypt ^ and 

led to the greater prestige of women generally : No/MoOerPjaat, 8i <^acn rovt Diod. i. 27, l. 

AlyvTTTiov^; Trapa to kolvov tOo<i tcov dv6pco7rQ)v yafielv dSeXcpaf Bid to 

yeyovo^: t>}? "IcrtSo? iiriTevypia . . . hid he TavTa<; Td<; atVta? KaTaSei^drjvat 

/j,€i^ovo<; €^ovcria<i /cat Tifirj<i TVy^dvecv ttjv ^aalXiaaav tov ^acriXeco'i kol 

irapd Tol<; iBcwraif Kvpieveiv ttjv yvvatKa TdvBpo<;, iv Tt] tt]^ 7rpoi,Ko<i avy- 

ypa(f>r] Trpoo-ojJLoXoyovvTWV twv yapiovvTOiv diravTa ireiOap^ijaeiv ttj yafjbov- 

fiivT]. This surprising statement Mr. Mahaffy dismisses in a foot-note as Mah. Emp. Ft. 

' probably too strong,' though he admits that the remark gives a true general ^^' '*'*'*' "^**^ ^" 

impression. And indeed recently found papyri point to a remarkable 

degree of legal and social freedom of women. Egyptian women (as the two 

names — Greek and Egyptian — prove them to be) appear frequently in the 

papyri concluding bargains, stating accounts, making petitions, lending money, 

even selling land. The Egyptian woman is legally 'capax' — a fact which no 

doubt horrified and perplexed the Greek conquerors — and, in deference no 

doubt to these scruples, we find in some documents that the woman's husband 

or a male relative is added as her legal Kvpio<i in transacting business : but Grenfell, 

this is not till the time of Ptolemy Philopator, and the innovation is intro- ■•• -P' ,' 

•^ . . xvni., XIX. ace. 

duced by a royal rescript {irpoa-Tayixa) during the earlier part of the reign of 

that unpopular and anti-nationalistic king, It seems then literally true that Herod, ii. 35. 

al yvvatKe'i dyopd^ovcri koI KairrjXevovac, and the remark is not due merely 

to tlie bewilderment of the simple Greek traveller in the land where the 

inhabitants tu iroXXd TrdvTa efiiraXiv Tolai dXXoiac dvOpdiiroiai iaTrjaavTO 

TjOed T€ Koi v6/jiov<;. The maxim ascribed to Osiris that 'it is a virtue in 

woman to let neither her person nor her name cross the threshold' (Synes. de 

prov. i. 13.) obviously belongs, as Wiedemann points out, to a late date when Wiedemann, 

Hellenic ideals were attributed by an anachronism to early Egypt. I4g"' 

At any rate we find women figuring in a prominent position in the 
Grenfell Papyri. In one (xviii) we find Apollonia, or Semmonthis, the wife Gienfell.Alex. 
of Dryton, lending wheat without interest to Apollonius and his wife Herais ; *'''°^'*^ ^''^S- 
in the next (xix) the same Apollonia appears lending money to Nechoutes ; Gren. Gk. 
and again in xx to Saeis and Harmais and their mother.^ In the third will ^fx 
of Dryton (Pap. xxi) we find it expressly stated that his wife is to retain her xx. 
earnings : oaa B' dv (f)acvr]T[at eVJt/cT'' e'^ouaa rj ^efj.fjb(ov0c^ ovTa avTrjt 
crvvovcra Apvrcovc, KvpieviTco avTOiv. In xxvii Sebtitis cedes to her daughter xxvii. 
half an aroura of corn-land ; in xxxiii we again find women selling land. In xxxii-'. 

^ It is needless to point out that it must works out its divine prototype, 

liave been the other way about. The custom * In all these cases, however, Apollonia acts 

gave rise to the myth : the myth did not give with her husband as Kvpios according to the 

rise to the custom. Such * aetiological ' myths rescript of Philopator mentioned above, 
are of universal occurrence. Eveiy social fact 

^.S. — VOL. XVIII. R 


xvii. xvii two women claim to be reinstated in the possession of property inherited 
by them from their father. The testimony of this papyrus is exceptionally 
interesting. From it it appears that the property had devolved on the 
daughters as the natural heirs-at-law — rov Trarpb^ rj/j,a)v KarakLirovTO'i rjfuu 
TO, virdpxovra avTOi ah l dd er a, while the nearest male relative apparently 
took advantage of the introduction of Greek ideas, and violently took possession 
— Kara to avyyevLKov i'jTe\d6vre<i [^7%t ?]o-Tetai/ dTroypayjrd/xevoi ovre Kara 
Bi.adi]Kr}v. And this, too, though the plaintiffs had duly paid the successsion 
duties to the queen, i.e. the fiscus — evrfkiKOL hk \r]fiel<i <yevo\n,evaL ra KadrjKovja 
TeXtj deal ^epeviKrji Kvpiau iScoKafiev. The demotic marriage-contract of the 

liecorda of the reign of Ptolemy Philadelphus translated by R^villout in the Records of the 
Past, vol. X. p^^f^ shows how carefully the wife's interests were safeguarded. — ' Thy pocket- 
money for one year is apart from thy toilet-money ; I must give it to thee 
each year, and it is thy right to exact the payment of thy toilet-money and of 
thy pocket-money, which are to be placed to my account. I must give it to 
thee. In case I should despise thee, in case I should take another wife than 
thee, I will give thee twenty argenteus. The entirety of the property which is 
mine, and which I shall possess, is security of all the above words, until I 
have accomplished them according to their tenor .... The writings which 
the woman Tahet, my mother, has made to me concerning one half of the 
entire property which belonged to Pchelchons, my father, and the rest of the 
contracts coming from her, and which are in my hand, belong to thee, as well 
as the rights resulting from it.' 

Herod, ii. 35. Such being the Egyptian woman's legal privileges we need not wonder 

at her legal obligations so puzzling to Herodotus : rpecpeiv Tov<i ro/cea? rotcrt 

fi€P Traicrt ovSefiia dvdyKr) /xr) ^ovXofxivoiai, rfjai 8e dvyarpda-t Trdaa dvdyKTj 

KoX fiT) ^ov\ofievr]<Ti. Similarly contracts have frequently been found binding 

Wiedemann, a wife to bury her husband and maintain his tomb, and one which seems to 

Her. u. p. 153. g^act this tribute from a daughter towards her father. This seems to indicate 
rap. hJiilaq. ° i i • i 

X. 8. that the family tomb may have in some cases been the special property of a 

Robertson woman-heir, a fact which is paralleled amongst the Nabataeans, in which 

Smith, people the position occupied by women is very high, as we know from coins. 

Marriage, The tomb, as Robertson Smith points out, is one of the sacra of the family 

p. 313. g^Q^j ^gg practically entailed ; that such sacra could be transmitted in the 

female line is very significant and points to an old law of female kinship. 

op. cif. p. 316. ' The old family system,' he adds, ' obsolete in political life, prevails in the 

op. cit. p. 94. grave.' Again he says, ' in the system of marriage with female kinship, 

there is no object to be served by excluding women from rights of property. 

The woman remains with her brothers, and her children are their natural 


The independence of women in Egypt, therefore, socially, legally and 

politically, seems to lend some justification to the theory that it may be a 

survival based on female kinship. 

Coming now to the history of the individual Ptolemies, I shall try to 

illustrate these principles from the different reigns. 

Ptolemy L had come to Egypt fully imbued with Alexander's views ; he 


had already participated in the scheme of Oriental matrimony, having 

married Artakama, daughter of Artabazus. At first he ruled merely as satrap 

under the boy king Alexander IV. He did not assume the title of king till 

about the year 305 B.C. according to Lepsius, though Alexander IV. died in Lepa.^W. 

314 B.C. At one time he entertained the idea of rendering his sovereignty "^aea. '' 

legitimate after the fashion of the Diadochi by marrying a member of Diod. xx. 37. 

Alexander's family, but this was before he had dared to believe in the possi- 

bihty of founding an independent sovereign line. His marriage with 

Alexander's sister, Cleopatra, was frustrated by Antigonus, but doubtless Mah. Emp. Pt. 

Ptolemy had already begun to recognise that nothing was to be gained by ^^ 

alliance with the house of Alexander. He had no intentions of laying claim 

to the Empire, and he had every reason for avoiding the jealousy of the 

Diadochi, certain to be evoked by such a marriage. His object, therefore, 

was to establish the divine right of the Lagid family and to get the priests 

to recognise the foundation of a new sacred line of kings. For in Egypt, more Lepe. Abh. 

than in any other ancient country, it was the unbroken chain of succession alone '^^ \qq P* 

that constituted the sacred nature of sovereignty.^ Hence Alexander the 

Great and his nominal successor Alexander IV. are set aside by the cults us 

irrelevant.2 Soter could not succeed Alexander IV. ; the only way to 

establish his. divine right was to glorify or discreetly conceal the pedigree o 

the Lagidae ^ and to ignore the line of Alexander, when they had once made 

up their minds to abandon the theory of their descent from Philip, naentioned 

as a current Macedonian belief by Pausanias. I am tempted to believe that Pans. i. 6, 2. 

Ptolemy's third marriage with Berenike was a diplomatic move in some way 

connected with this policy, and that there were reasons of state, probably 

urged by the priesthood, for accepting her as the foundress of the future line 

of Lagid kings rather than Queen Eurydike. The marriage with Eurydike, the 

daughter of Antipater, in the early part of Ptolemy's reign (321 B.C.) hadbeen 

dictated entirely by foreign policy. Four years later (317 B.C.) we hear of his 

marriage with Berenike, a grand-niece of Antipater, who had come to Egypt ^l&h. Emp. Pi. 

in Eurydike's suite. As Mr. Mahaflfy says, ' Polygamy was now the rule among ^' 

the Diadochi, but so distinctly political were their marriages, that a new 

alliance did not imply even a divorce of sentiment between the husband and 

his previous wife. In the present case there is no evidence that Eurydike 

was divorced, neither do we hear of any domestic conflicts between Eurydike 

and Berenike.' Mr. Mahaffy, however, does not allege any political reasons 

for this new alliance, nor does he try to explain why Eurydike so meekly suflfered 

herself to be set aside. Now the only extant statement with regard to 

* How strongly the idea of forming a dynasty ^ Alexander is not e*hs, and at Alexandria 

took hold of the Ptolemies is seen, as Lepsius only does he have a priest, 

remarks, by the unvarying repetition of the ^ It is significant that Porphyrius, in giving 

1. name Ptolemy with each king, as well as by the parentage of Soter, puts his mother's name 

the similar tendency to repeat the same names first : — TlroKfficuos 6 'Apaiv6r]i nal SAyov vUt. 

in the case of the eldest daughter of the rojral (Miiller Fr. h. Gr. 8, 719). 


V, 2 


Theocr. xvii. Berenike's parentage is that of the scholiast on Theocritus,^ who says that she 
Avas a daughter of Lagus, hence a step-sister of Ptolemy. This, Mr. Mahafify 
points out, is probably a misconception arising from the formula ' wife and 
sister ' applied to Egyptian queens ; but it seems to me possible that the 
statement, whether true or not, indicates the beginning of a deliberate 
attempt to patch up a Lagid family-tree in accordance with Egyptian notions. 
According to Maspero (Comment Alexandre devint dieu en Egypte, Anmiaire 
de I' ^coh pratique des hautcs etudes, 1896-7) ' La noblesse de chaque membre 
d'une maison pharaonique et ses litres k la couronne se mesuraient sur la 
quantity de sang divin, qu'il pouvait prouver : celui qui en tenait de son p^re 
a la fois et de sa m^re prenait I'avance sur celui qui n'en avait que par son 
pere ou par sa mere seule. Mais la, une des lois egyptiennes qu'on observait 
avec le plus de rigueur intervenait pour etablir des distinctions qui ne peuvent 
plus etre observ^es dans nos civilisations modernes. Ze mariage entrefrtre et 
socur 6tait le mariage par excellence, et il acqiUrait un degri de saintet6 ineffable 
lorsque le frhe et la socitr qui le contractaient dtaient nes eux-memes d'un frh'e et 
d'une soeur issus d'un mariage identiquc au Icur. Cette particularity des 
moeurs Egyptiennes, qui nous parait un raffinement d'inceste, avait produit 
des consequences importantes pour I'histoire du pays, et tout un ensemble de 
dispositions legales ou de fictions religieuses ^tait destine a en assurer I'effet 
dans les questions de succession royale, ou k remedier aux insuffisances de 
l^gitimitE qu'elle entrainait souvent parmi les heritiers males.' Thus in the 
traditional way Ptolemy Soter was to marry a kinswoman who was to be the 
ancestress of the royal line, and it was only when such a line had been founded 
that divine honours could be paid. Hence, possibly, the delay in the 
Schol. Theocr. deification of Ptolemy I., which seems not to have taken place until after his 
Id. xvii. 17 ff. ^jgath in 271 B.C. That Soter was not fully acknowledged in the cult till 
Mah. Emp. Pt. late in his son's reign, is shown by the Aswan stele on which the series of 
p. 489. ^Q^^ begins with the gods Adelphi and omits all mention of Alexander and 
the gods Soteres ; on the other hand this late recognition may be explained 
by the fact that the stele was connected with an Amon foundation, in which 
cult the Lagids may have been less readily acknowledged. Similarly we are 
Lep3, Abh. not surprised to find the cult of Soter at the colony of Ptolemais at a time 
^^^\^^' ^' "^^^6" i^ w^s not yet established at Alexandria. At any rate it seems clear 
that the divinity of the Lagid line was not duly recognised till established in 
the second generation, and that Alexander and his house were studiously set 
a-side. Kaerst (Die Begrlindung des Alexander- und Ptolemaeer-Kultes in 
Aegypten, Rhein. Mus., 1897) connects the deification of Ptolemy Soter closely 
with that of Alexander. It is, according to him, no ' reiner Aegyptismus ' but 
has an Hellenic basis, though ' wir finden insbesondere in Bezug auf die 
Ptolemaeer eine fortschreitende Aegyptisirung des Konigskultes, sodass 
zuletzt das lagidische Konigthum als ein verjiingtes Abbild der alten 
Pharaonenherrschaft erscheint.' On my theory the conscious ' Aegyptisirung ' 

^ Theocritus himself in unhellenic fashion calls her 'Pi.vTi'y6vas Ovydr-qp and avoids all mention 
of her father. 


begins earlier and is already discernible iu the matrimonial policy of 

In 285-4 B.C. Ptolemy Soter definitely decided the succession question 
by associating his younger son, Ptolemy, in the government, practically 
abdicating in his favour.^ It is apparently only at this point that Eury- 
dike and her children withdrew: hence we may infer that the question was 
an open one till this date. On the assumption that there were no diplomatic 
reasons for considering Berenike more suitable to be the royal ancestress of 
the new line, this act was a mere arbitrary solution of the difficulty. Ptolemy 
Philadelphus seemed the most capable successor ; Ptolemy Soter felt himself 
secure enough to enforce his choice ; so the natural heir was set aside and 
his more competent younger brother preferred. On the other hand if there 
is any truth in the theory put forward above, the accession of Philadelphus 
was due to an undercurrent of political and priestly intrigue. I shall try 
later to show how such a theory bears on the actions of the disinherited 
children of Eurydike, and how they availed themselves in the next reign of 
party politics to further their personal claims. It is well to note in passing 
that Arsinoe, the eldest child of Berenike, who plays such an important part 
in the following reign, was at this time outside the range of Egyptian home- 
politics, having been married in 300 B. c, to Lysimachus of Thrace. 

This abdication of Ptolemy I., to whatever motives it was due, was well Mah. E7np. Pi. 

received, and the coronation of the new king was an occasion of great public P- ^^^• 

rejoicing. That the step, liowever, was an experiment prompted by political 

expediency seems likely from the testimony of Porphyry and Diogenes 

Laertius, who imply that the old king continued his kingly functions in op. cU. p. 106, 

partnership with his son. ^?}^ ^■. „„ 

^ f . , . op. at. p. 488, 

The idea that a king should abdicate voluntarily while in full possession note on p. 106. 
of his faculties is a very usual one in primitive forms of civilisation, as Mr. 
Frazer points out in The Golden Bough, and is in such cases probably based 
on animistic conceptions of the function of the king. He is the sacred 
receptacle and guardian of the aggregate vitality of his people, and this vitality 
he must transmit intact to his successor while his faculties are still im- 
impaired ^ : in fact the death of the emeritus monarch was usually held to 
be essential to the preservation of divine kingship. Now there are various 
traces of this primitive custom in Egyptian and Aethiopian records down to 
a comparatively late date, whence I argue that the action of Ptolemy I. would 
be not only comprehensible but even fraught with a religious significance to 
the less civilised portion of his subjects. I do not mean for a moment to 

^ In a similar way Seti I., son of the usurper rwv Upiccv aWh «al trvWitfiSriv firaj'Tf j oi kut' 

Ilamescs I., having strengthened his claims hy AtyvirTov ovx ourai ywaiKwv «ol riKvwv «al rwv 

marrying a granddaughter of Anienophis III., 6.Woiv tuv vvapx^vTwv avrols hya&wv i<pp6vri^ov 

associated as his colleague in the sovereignty wi tjjj tcDi' fia<Tt\i<Dv ia<paKflai. 

his son, the legitimate heir, Ramesea II. As It is to be feared that the motives which 

this position was ignored owing to its vagueness, actuated this solicitude were less strictly 

Seti finally abdicated altogether. altruistic and loyal than the historian 

" Cf. Died. i. 72, 4 ov yap fji,6vov rh ovaTqixa, imagined. 


imply that he was consciously influenced by any consideration of the kind ; 
but I think it might possibly enter, as an additional advantage of the step, 
into the calculations of his advisers the priests, who were well versed in such 
popular superstitions, and knew how to work them for their own advantage. 

We may infer from the unflinching severity with which this otherwise 
humane king treated his dangerously near relatives, and from the actions of 
these relatives themselves, that the new king's succession was by no means 
Paus. i. 7, 1. left unchallenged. At the beginning of his reign he killed his elder step- 
brother, Argaeus, and perhaps, according to Pausanias, another brother as 
well. His full-sister, Arsinoe, as I have pointed out, was at this time safely 
disposed of — one of a series of diplomatic marriages uniting Egypt at 

liih.Emp.Pi. this period with Thrace. In the same way a half-sister of the king's, 

op.^it. p. 37 Lysandra, was also wisely relegated to a distant land. Similarly Theoxena, a 
daughter of Berenike by a former marriage, had been disposed of by her step- 
father, Ptolemy I., in 307 B.C. 

But naturally the chief danger was to be expected from Keraunos, the 
dispossessed son of Eury dike. If Philadelphus found it necessary to deal so 
harshly with his other brothers, it is probable that Keraunos was not without 
supporters, and we may be sure that this crafty and unscrupulous prince 
knew well how to work party politics for his own advantage. He never lost 
sight, I believe, of his ultimate object — the throne of Egypt — and his schemes 
in Thrace were merely stepping-stones towards this end. On being ousted 
from the succession he withdrew to Thrace, where he espoused the cause not 

ll&h. Binp.Pf. of his full-sister, Lysandra, but of his half-sister, Arsinoe. In fact, he 

p. 113. recognised that, since Berenike had been accepted as the queen-mother, 

Arsinoe, her eldest child, would according to Egyptian notions have claims on 

Justin, 24, 2. the throne in her own right. Hence, as Justin fully describes, he persuaded 
" ^''^* the reluctant Arsinoe II. to marry him, at the same time craftily writing to 
assure Philadelphus that he has ' laid aside all resentment at being deprived 
of his father's kingdom.' No sooner had he gained his point than he murdered 
Arsinoe's children before her eyes. Having been discredited in Thrace he fled 
to Seleukos in the hope of stirring him up against Egypt, but his career and 
further claims were here abruptly cut short by his death in battle against the 

Another troublesome relation was Magas of Gyrene, Philadelphus's half- 
brother. Being a son of Berenike, the acknowledged queen, Magas would 
not consider himself so entirely outside the line of succession as modern ideas 
Leps. Abb. would lead us to suppose. To quote Lepsius : ' nach Ptolemaischem 
BerL M. Erbrechte scheint es, dass nach dem Tode des Konigs zunachst seine Wittwe 
den Thron beanspruchen konnte und nur fgenothigt war, den miinnlichen 

op. cit. p. 504, Thronfolger zum Mitregenten anzunehmen,' and again, 'das Recht der Thron- 
note 1. folge musate auch das der Uebertragung der Mitregentschaft auf einen 
Gemahl oder einen Sohn einschliessen.' With this compare Justin xxx. 3. 1, 
'inter has regni Syriae parricidales discordias moritur rex Aegypti Ptolemeus 
(Euergetea II.), regno uxori et alteri ex filiis, quem ilia elegisset, relicto.' 
It is thus possible that the queen-mother may have had some choice in the 


matter; we find Cleopatra III. apparently preferring and dismissing her sons 
at her own pleasure. At any rate we can imagine that the rights of the 
([iieen-mother would naturally often lead to family intrigues. Therefore the 
revolt of Magas may have been based on these to us shadowy claims, especi- 
ally as Berenike is said to have favoured this rebellion, i.e. to have been wish- Sliiupc vol. i 
ful to single out Magas as her co-regent. This disaffection in Cyrene may be !'■ 
the reason of the omission of this province from the list of Egyptian posses- 
sions given by Theocritus ; while the similarly puzzling omission of Cyprus Thci.ei. .wn 
may be connected with Pausanias's remark, aVeVreti^e Be Kal dWov dBeXt^ov Pauu. i. 7, l. 
yeyovora e^ ^vpvBifcyji;, KfTrpt'ou? d(f>iaTdvTa alad6^evo<i, and point to a 
similar state of disturbance there. The claims of Magas were finally silenced 
hy the betrothal of his infant daughter Berenike to the c7-oion 2Jrince Euergetcs, Justiu, 26, 3. 
which united the claims of the brothers. 

This betrothal was nearly broken off owing to the plots of Demetrius the 
Fair, another pretender, I believe, to the throne of Egypt. This Demetrius 
was the son of Demetrius Poliorketes and Ptolemais, a daughter of Eurgdihc Pint. Dan. 47. 
and Ptolemy I. Apame, the mother of the young Berenike of Cyrene, or }>[.a.\\.Einp.Pt. 
rather the anti-Philadelphic party in Cyrene, conceived the plan of uniting P' ^^^' 
the claims of the rival branches by marrying the young princess (a grand- 
daughter of Berenike I.) to Demetrius (a grandson of Eurydike) and thus 
strengthening and combining the opposition to Philadelphus. Justin gives Justin, 26, 3. 
an account of this scheme and of its frustration owing to the fickleness of 
Demetrius : sed post mortem regis mater virginis Arsinoe (Apame) ut invita 
se contractum matrimonium solveretur, misit qui ad nuptias virginis 
regnumque Cyrenarum Demetrium a Macedonia diTce?,^QTeni,q^idet ipseexfilia, 
Ptolemei {Soteris) procreatus erat. His subsequent intrigue with the queen- 
mother having been discovered, Demetrius was put to death, the anti- 
Philadelphic party was discredited, and the former betrothal of Berenike to 
Euergetes was confirmed. Itaque versis omnium anirais in Ptolemei filium 
insidiae Demetrio comparantur. 

Such, then, were the difficulties which beset Philadelphus during the 
early part of his reign. There is another member of the family worth com- 
menting on, not from the difficulties she occasioned, but from her apparently 
meek submission to state exigencies, — I mean Philotera, the king's full-sister, lla-h. Emp. Pt. 
who seems to have remained uiimarried. In return Philadelphus paid her |,'" ^35' 
more than the usual honours of a royal princess : she accompanied the king Stiabo,i6,4,5. 
and queen on royal progresses, and cities were called after her. This, as 
Sharpe points out, was no idle compliment; the princess probably received Sharjie, vol. i. 
the crown revenues from these cities, just as we know that Arsinoe II. ^' 
received the revenues of the Arsinoite nome.^ Indeed Letronne goes so far Letronne, 
as to assert that all the colonies founded by Philadelphus were named after ^'^'^"^i^^, i'- 

his second wife or his sister Philotera, — his two full-sisters. Mr. Mahaffy, how- Mah. E77ip. Pt. 
p. 135. 

1 However, since there is reason to believe, likewise profited personally less than we might Mah. Emp. Pt. 
as we shall see below, that this was merely a suppose from the revenues of these foundations. Chronol 
convenient state fiction, perhaps Philotera lable, p. xix. 


ever, denies this, stating that he has found other village-names mentioned in 

the Petrie Papyri, such as Lagis, Lysimachis, etc., founded by this king; 

though this perhaps scarcely refutes Letronne's statement regarding the 

establishment of colonies. I fail, moreover, to understand why Mr. Mahaffy 

cites the colony of Philadelphia (Rabat Amon) as an exception to the rule. 

Mah. Emj). Pt. Surely that name more than any other stamps the colony as a foundation in 

^'' . 'o c . j^Qjj^^^jj. q£ }jjg .^j^g Arsinoe II., who alone at this period bore that name. 

op. cif. p. 116. Philotera was further honoured as a goddess with a shrine at Memphis, a cult 

Leps. Al)]i. established according to Mr. Mahaffy by Philadelphus, though Lepsius assigns 

p 500 ^^^ foundation to Eiiergetes. 

This bnngs me to what is the crucial point in my argument — the second 
marriage of Ptolemy Philadelphus. The date of this event is uncertain, but 
Lepsius and Mr. Mahaffy place it about 277 B.C., that is, a few years before the 
second visit to Pithom. It seems possible that the step may have been 
advised by the priests during the consultations of the previous visit soon 
after the death of Lysimachus, and its accomplishment may have been post- 
poned by Arsinoe's meantime falling a victim to the schemes of Keraunos. 
We have seen above the numerous difficulties which beset Philadelphus and 
prevented his full recognition as monarch. On my theory these difficulties 
Uah.Emp.Pi. could be finally settled only by this alliance with his sister Arsinoe. This 
Table' p'^xvii Arsinoe was the eldest child of Ptolemy Soter and Berenike, and, if the rights 
of Eurydike's children were to be set aside, might claim to be the legitimate 

The reasons usually alleged for this marriage seem to me inadequate 
Droysen, and unconvincing. Droysen (and with him Strack) believes that it was due 
^ seq? *^ *^^ proprietary claims which Arsinoe had on certain cities of the Euxine 
such as Cassandrea and Pontic Heraclea. But if it is true, as we hear, that 
Philadelphus had later to carry on an elaborate naval war to bring these 
cities into submission to Arsinoe, one is tempted to fancy that he might have 
helped himself to them, even without the sanction of matrimony, had they 
been his sole object. 
Koepp, Koepp likewise connects the step with foreign policy. According to his 

"^^ uTted^"'^'' ^^^^ Philadelphus hoped by this union with the widow of King Lysi- 
Mah. p. 132. machus, to attach the old subjects of the Thracian King to his cause in the 
war against Syria. We must remember, however, that Arsinoe, not long- 
after the death of Lysimachus, had consented to marry the usurper Keraunos, 
her step-brother. Besides, immediately after this marriage she had been 
Mah. Emp. Pt. banished by Keraunos to Saraothrace, where she seems to have lived in 
p. 122. retirement during the few years ^ before her return to Egypt. It does not 
seem likely then that an alliance with her would have greatly strengthened 
the bond between Thrace and Egypt. 

Mf we arc to accept the datft 277 B.C. as the died in 281 B.C.: presumably she did not Ju 

approximate time of her marriage to Phila- marry Keraunos quite at once, as Justin dwells 

delphus, though Wiedemann (Philol. N. F. i. on her reluctance. 
81), puts it as late as 273 B.C. Lysimachus 


Holm, on the other hand, believes tliat the marriage was tluo to personal Holm, Or. 
affinity of character : — * ich glaube die Hauptsachc war die Uebereinstimmung ' '^ ' ^g^ 
derCharaktere. Boiden war das Hiichste Herrschalt, Intrigue, Lebensgennss, 
Sie verstanden, halfen sich gcgonseitig und verzichen einander ilirc Fehler.' 
Even accepting this estimate of their cliaracters, one does not feel the 
cogency of the reason : the history of the period parades before our eyes a 
long succession of personages of both sexes sufficiently endowed with such 
qualities, and one cannot help believing that Philadelphus might 
easily have found 'a congenial consort, without shocking his Hellenic 
subjects by such a serious departure from Greek customs. Besides, he 
cannot up till close on the time of the marriage have seen much of this 
sister : we must remember that she was married to Lysimachus in 300 n.c, 
when Philadelphus was only eight years old ! Nor can she have been long 
in Egypt immediately before the marriage (if we accept the date 278 — 277 
B.C. for that event ^), for after 281 she married Keraunos and lived some time }>u\\.Emp.Vt 
in Samothrace, long enough indeed to become imbued with Samothracian 1^' 
religion, as she built a temple there to the Kabeiroi. p. 139. 

Mr. MahafFy also inclines to the theory of personal attraction, thougli he is p. Hi 
anxious to discount the flatteries of court poets : he dwells on her remarkable 
intellectual ability and the tact with which she adapted herself to her 
position ; he believes that she was by no means good-looking. We must 
remember that she was about forty, considerably older than the king, and 
that her life had been a troubled one. 

But the most serious objection, which applies to all these theories alike, 
seems to me the fact that they treat the marriage as an isolated instance 
and do not take into account the subsequent brother-and-sister marriages in 
Ptolemaic history. Everything, therefore, seems to point to the conclusion 
that the marriage was due to diplomacy. There is no trace of any violent 
rupture with Arsinoe I., who seems to have lived afterwards in semi-regal 
state at Koptos. It is true that she is reported to have plotted against her Mah. Emp. Pi. 
husband's life, but no definite evidence of this is forthcoming, and it seems !'• ^^'^' °°^^ ^ 
very unlikely from what we know of her subsequent position at Koptos. It 
is just the sort of rumour that would inevitably arise to explain her retire- 
ment to the Hellenic world. On the other hand, Arsinoe II., being herself Theocr. 17 
childless, adopted the first queen's children. On the stelae of Pithom and of Qj.g^^'j-g^ '^j,,} 
Mendes Euergetes is actually represented as the son of Arsinoe 11.^ ; so too Mah. Revenvc 
on the Canopus stone he figures as the son of the 0eol a8e\(f>oi. Strack ^"^j^' ^^^^^ • 

^mp. PL ^ From Mr. Mahaffy's own argument there the marriage and deification were quite 

138. seems to be no special reason for dating the recent. But, on the other hand, ia there ' 

man-iage much before 273 B.C., the date of the anything conclusive to prove that it was not? 

second visit to Pithom mentioned above, Kaerst (die Begriindung des Alexander- und 

though he fixes on 278-277 B.C. This seems to Ptoleraaeerkultes in Aegyptcn, Ilhcin. Mm. vol. 

crowd too much into the years 281-278 B.C. and 52) prefers the year 274 B.C. 

I should feel much more satisfied with a later ' It is impossible, as Mr. Mahaffy points out, Revenue Pap. 

p. 139. date. Mr. Mahaffy points out that there is to believe that the Pithom stele rciacsents an Intrud. p.x.xii, 

nothing in the Pithom inscription to show that unknown child of Arsinoe II. 



Canopus {die Dynastic der Ptohmacr p. 88), admits that this suppression of liis 
^' ■ ■ mother's name looks like an attempt to establish in the eyes of the priesthood 
the legitimacy of his succession. He considers, however, that the 'Beriihmt- 
heit ' solely of Arsinoe II. was a sufficiently strong motive to influence the 
priests. Arsinoe seems to have immediately taken up a position of influence 
in the country, somewhat perplexing unless based on her own heiress-rights. 
It is she who is deified with the title Philadelphus which later historians 
extended to her husband ; in the Revenue Papyrus she is referred to simply 
as 77 <f)i\dB€X(f)o<i, whereas Ptolemy is merely one of the Oeol a8eX<f>oi} Thus 
he appears to owe his divine honours mainly to his wife, as indeed the repre- 
sentation of Arsinoe conferring honours on her husband at the top of the 
Pithom stele would lead us to conclude. 

The testimony of coins, too, is somewhat similar. To quote Strack (op. 
cit., p. 17) ' steht ein Flirst an der Spitze des Staates, so beweist sein Name 
und sein Bildniss auf den Mlinzen seine Souveranitiit. Die Ptolemaer haben 
des wenig acht gehabt.' In fact we find on their coins ' Kopfe und 
Aufschriften von verschiedenen Koniginnen.' It is rare to find Philadelphus 
alone on them ; he appears frequently with Arsinoe II. and she frequently 
alone. On one struck by Philadelphus we find on the one side the head of 
Arsinoe, on the reverse the double cornucopiae and the inscription 'A/xxtvo?;? 
<t>L\aB€\cf)ov (Poole, PI. VIII. 1) : on the reverse of another, which likewise 
has Arsinoe's head on the obverse, is the royal eagle and the same inscription 
(Poole, PI. VIII. 3). A noteworthy point concerning the coinage of this 
reign is the fact that Ptolemy in the first half of his reign does 
not date his coins from the year of his accession. In Sharpe's 
Sharpe, words : 'it is not till the nineteenth year of his reign, sooqi after the 
vol. 1. p. 335. death of his mother, that he made an era of his own and dated his coins by the 
year of his own reign.' Thus we have coins with the heads of Soter and 
Philadelphus on one side, on the other the head of Berenilcc. Now if 
Berenike died some time before 266 B.C., the nineteenth year of the reign, 
and if it is possible that the marriage with Arsinoe II., as I have tried to 
show, may have taken place rather later than 277 B.C., in fact, immediately 
before the second visit to Pithom in 273 B.C., it seems not impossible to 
connect the two events and to find in them the immediate re^^on for the 
marriage of Ptolemy at that particular date with Berenike's heiress-daughter, 
and for the beginning of his own coinage. This argument would apply 
equally well, however, even if the marriage took place shortly before Berenike's 
death : as Berenike grew old Philadelphus would no doubt see the advisability 
of strengthening his position by the new alliance ; there may even have been 

Grenfell, ^ In Grenfell's Greek Papyri, No. XII. (date 

Alex. Erot. Fr. c. 148 B.C.) Ptolemy II. is mentioned as Phila- 
P- 2^- delphus. This, Mr. Grenfell observes (note 7, 

p. 31), is the earliest known reference to him 
with this title ; thus ' there can be little doubt 
that it was used in the list of kings among the 
priesthood of Ptolemais, when the priesthood of 

Ptolemy II. was .established.' We should 
naturally expect that the title would first be 
extended to the king in the Greek city 
Ptolemais ; similarly, as we have seen, it is in 
this place that the cult of the Soteres is first 
recognised. The colony was under the special 
protection of both Soter and his son. 


a party in the state ready to put forward the independent claims of Arsinoe, 
a party whicli may have had a section of the priesthood at their back. At Huh. Emp. Pt. 
any rate the same year probably, the nineteenth (2GG B.C.), in which Philadel- xablo'^p^xix 
phus began his coinage, saw the accomplishment of the climax of Arsinoe's 
deification — a Canephorus of Arsinoe Philadelphus is established at Alexan- 
dria.^ Two years later a further concession is made — we have the first 
mention of the gods Adelphi. In the same year the third visit to Pithom op. di. 
took place and along with it we hear of fresh religious endowments — a quid Tabk n xir 
pro quo, one is tempted to believe. 

But the culminating point of Arsinoe's political importance is still to be 
mentioned. I mean the transference of the airo^ioLpa, or tax of one sixth on 
wine and fruits, from the temples to the queen in the year 262 B.C., a detailed 
record of which transference has been preserved in the recently found 
Revenue Papyrus, This transference, as Mr. Mahaffy observes, puts a very Emp. Ft. 
different complexion on the attitude of Ptolemy towards the priesthood. On ''" '"^'^' 
the priestly monuments we hear only of fulsome votes of thanks to the king 
for the generosity of his endowments. We know now that his munificence 
was to a large extent merely a cloak to cover this great revolution in taxation, E^np. Pt. 
which reduced the clergy to the condition of state-pensioners and diverted P- ^*^- 
this great source of revenue into the public treasury. The money went 
to the fiscus, but it was claimed in the name of the deified Arsinoe. As a 
goddess identified with Isis she might claim the eKjr) without outraging o;j. ci<. p. 13 o. 
national scruples. 

As Mr. Grenfell says, ' It is hardly necessary to point out that the eKT-q Rev. Pap. 
was collected and paid et? to ^aaiXiKov like any other tax. The Ova-iai koI ^°^' 36, note 4. 
(TTTOvtal was an ingenious but transparent fiction to cloak the disendowment 
of the temples.' The result is that it is one and the same thing for all 
practical purposes whether the payment is made ei? to ^aaiXiicov or to the 
goddess Arsinoe — in fact we now know from the newly recovered fragment of Mah. Jihen- 
the Mendes stele that the queen had died some years earlier, in 270 B.C. " ^897" ' 

This brings me to the most difficult and intricate point in the reign — 
the reclaiming of Lake Moeris, and the connection of this event with the 
queen. Mr. Mahaffy, writing before the early date of the queen's death was Einp. Pt. 
discovered, is much exercised to decide whether Arsinoe was dead or alive at ^'' ^^^" 
the time, whether the renouncing of her rights to the district was a gift or a 
bequest. But taking this settlement of the Fayyum in close connection with 

^ According to Mr. Mahaffy : ' We know from process was not complete till the King's 

independent sources that the deification of twenty-third year, when she absorbed one of 

Arsinoe Philadelphus was gradual; that she the great revenues of all the Egyptian gods.' 

attained divine honours, first at one, then at As proof that this was the high-water mark 

another of the Egyptian temples. The estab- of her deification one may remember that it is 

lishment of a Canephorus or eponymous always in connection with this Canephorus of 

priestess in her honour at Alexandria, which Alexandria that her name is mentioned 

dates back as far as the year 19 of the reign, specially (as well as with her husband) as one of 

according to demotic documents, appears to the Oeo'i aSf\(pol in the date formulae of subsc- 

be the climax or consummation of this gradual quent reigns, e.g. Grenfell, Papyri X., XII., 

apotheosis. We now know that practically tliu XXV., XXVII. 


the policy of the preceding years, I would suggest that the explanation is to be 
found in the same fact, that the use of Arsinoe's name is again nothing but a 
state fiction,^ that the queen personally gave up nothing, but that the so-called 
renunciation of her rights merely cloaked the re-organisation of the crown- 
property, formerly jJ Xc/jlvt}, now reclaimed land to a large extent. It seems 
likely from a remark of Diodorus that the proceeds of the fisheries of 
the lake may have belonged according to ancient tradition in quite a special 

Diod. i. 52, 5. way to the queen for her personal expenditure — 7rpb<i /jLvpa koI top aWov 
KaWcoTTia-fjiov. At the same time Herodotus in referring to these same 
fishery-revenues says that they brought annually two hundred and forty 

Herod, ii. 149. talents €? TO /3acriki]iov and again he alludes to them as part of the state 
Ifcrod. iii. revenues in Persian times. Therefore probably the reclaiming of the land 
^ ' ''' meant not that Arsinoe renounced her personal property for the benefit of the 
state, as Mr. Mahaffy assumes, but rather that the government to suit its own 
ends chose to give up at least part of these revenues and by reclaiming the 
lake gained fertile land which was wanted at the time for special political 
purposes. Just as the uTrofioipa was diverted into the fiscus in the queen's 
name, so here the extension of the Lake province and its better irrigation was 

Mali. Emp. PL attributed to her : we hear of the 'ApaivSrjf; x^H'^ ^^'^ hereafter the district 
p. 178. ^y^g officially known as the Arsinoite nome. 

The position which Arsinoe held from the time of her return to the 
country points strongly to the recognition of her rights as heiress. At the 
top of the Pithom stele the queen is represented as a deity conferring honours 
on the king her husband. On a tablet recently discovered at Tanis (Egypt. 
Explor. Fund, Tanis pt. ii., Petrie p. 30, No. 165), Arsinoe is called ' the Net, 
the regent of the two lands, princess, lady of thrones ' : another tablet from 
the same place (op. cit. p. 32) represents Ptolemy IL in Egyptian dress offering 
land to Khem and Arsinoe. There was a cult of Arsinoe Philadelphus at 
Alexandria as early as 267 B.C. ; and even earlier, with the title Philadelphus, 
she had been associated as avvvao^ Bed in the cult of various Egyptian gods. 
For the deification at this time of her husband in his own right there is no 
evidence : he merely figures along with Arsinoe as one of the ' Gods Adelphi,' 
On the other hand the deification of the queen was carried out step by step, 
as Droysen (Berlin Sitzb. for 1882) points out, leading up to the final stage 
late in the reign, when a very practical meaning was given, as we 'saw, to the 
sovereign lady's divinity. In the face of all these facts and with no single 
piece of evidence that Ptolemy II. bore the title Philadelphus in his life-time, 
is it possible to beheve with Strack {op. cit. p. 116 flf.) that the title was 
transferred to Arsinoe from her husband ? We have, he admits, examples in 
later Ptolemaic history of the transference of similar titles from a queen to 
the king associated with her in the crown : e.g. Soter II, receives from his 
mother the title Philometor, and later from his daughter the title Philadel- 
jjhus. He admits, too, that in public documents of the reign Ptolemy II. 

' As we uow know that the queen was already how nominal her part in the transaction must 
dead when the reform took place, we may judge have been. 


bears only the dynastic name, and shelters himself behind the rarity of such 
documents. But surely his theory that ' Philadelpluis ' was the proper name 
of Ptolemy II. 'den er nach Erhebung zum Thronfolgcr mit dcm dynastischen 
vertauscht habe, der dann nicht aus dem Gediichtnis geschwundcn und spiiter 
wieder hervorgeholt sei, als es sich um einen Namcn handelte unter dem er 
verehrt werden konnte' is hardly convincing. And we are asked to believe, 
that this ' Individualname voll guter Vorbedeutung ' was bestowed on 
Ptolemy II. in his infancy 'dem Stiefbruder (Keraunos) gegcniiber, der 
spiiter die Krone tragen wUrde ' ! If on the other hand, as I believe, 
Ptolemy II. bore the name of Ptolemy in his childhood, the fact that he was 
given the dynastic name usually bestowed on the crown-prince seems to point 
to the conclusion that the marriage of Soter with Berenike was a diplomatic 
move and that Berenike's son was from the first destined for the succession. 
The deification of Arsinoe as ' Philadelphus ' indicates, I believe, her identi- 
fication with Isis. Other names of the Ptolemies suggest a similarly close 
connection with the Osiric cult. Dionysus is applied to Philopator and to 
Auletes ; Isis to Cleopatra VII. : Philadelphus again to Auletes and to 
Ptolemy the son of Cleopatra VII. Possibly Euergetcs belongs to the same 
cycle, for Plutarch {de Iside et Osiride, 42) says : to 8' trepov ovofia 
rov Oeov (Osiris) zov "O/x^lv ev e py irr] v 6 'Fipfiat6<i (^rjat hrjXovv 

Coming now to Euergetes we are first struck by the date of his marriage, 
which took place according to Callimachus in the year of his accession. We Ellis, 
know that he had been betrothed to his cousin Berenike for years ; we know, 2% ^^^' 
too, that he must have been about thirty-three (taking Lepsius's date, 
281 B.C., for the first marriage of Philadelphus) at the time of his accession, Leps. Berl. 
while Berenike must have been of a marriageable age before this date, if she ' ^' 
was old enough at the time of the crisis at Cyrene in 258 B.C. to play the 
part described by Justin.^ Why then had the marriage been delayed so 
long? Mr. Mahaffy suggests that 'there must have been some law or Justin,xxvi.3. 
tradition of the old Pharaonic royalty on account of which the wife of a Mali. Emp. Ft. 
prince royal could not be elevated to the dignity of reigning queen.' He ^" 
goes on to point out a fact most important for my argument : ' It was not the 
habit of Ptolemaic crown-princes to get married before they ascended the throne.' 
Elsewhere he says : ' In most other monarchies a suitable bride is found for 
the crown-prince as soon as he is of age ; in Ptolemaic Egypt I have observed Mali. Inirod. 
with surprise that this is against the practice of the court, though the reigning p^ p^'^xx^v. 
Ptolemies marry as soon as possible. Philadelphus, though grown-up in 
290 B.C., does not apparently marry until his assumption of royalty — in the 
opinion of some critics, not till his father is dead. Euergetes, though long 
grown-up, seems to have no wife till his accession ; Philopator, succeeding at 
about the age of twenty-four, has no wife till some years later. We hear of 
no wife of Euergetes II. till he succeeds in middle life and marries the widowed 

■ -P^ ^ To explain this discrepancy it has been seven at the time of the disturbance, which 
® ^' assumed that Berenike was a child of six or seems scarcely likely. 




Hint. d'4cov. 

pol. sous Ics 


l>. 179. 

queen. So it is (with one exception) down to the case of Caesarion, who 
would doubtless have been married before his early death, hut for this curi&iis 
court tradition. A satisfactory explanation of it I have not yet found.' Mr. 
Mahaffy suggests that only the child born in the purple was legitimate; but 
with the remark of Lepsius (already quoted) we may perhaps find a sufficient 
ground for this ' court tradition ' in the full rights of the late king's widow to 
nominate her successor and the necessity for that successor to form an alliance 
in accordance with Egyptian notions with a view to increasing ' la quantite 
de sang divin.'^ 

To apply this solution in detail. Philadelphus apparently did marry 
before his mother's death. Hence the comparatively insignificant part played 
by his first wife, and her repudiation perhaps at the time of the queen- 
dowager's death, leading to the marriage with the great queen Arsinoe II., 
and the beginning of the personal coinage of Philadelphus, If on the other 
hand this second marriage took place before the death of Berenike, it is 
probable that the first important step in the deification of Queen Arsinoe 
(266 B.C.) at least followed closely on the death of the queen-mother. 
As the deification did not begin till four years after her own death, it can 
hardly be directly dependent on that event. Hence I argue that her 
recognition by the priesthood was connected with the queen-dowager's demise. 

What then can have delayed the marriage of Euergetes to his cousin 
Berenike ? We know now that his mother, Arsinoe, had died years before, 
and Euergetes himself had been already associated for years in the govern- 
ment. There seems, therefore, no reason for the puzzling delay. We know, 
however, that his betrothal to Berenike, the daughter of Magas, was 
occasioned by State emergencies. May we not suppose, therefore, that, the 
emergency having been tided over, there was a party in the State which 
expected the heir apparent to ally himself duly with his own sister Berenike ? 
Such dissensions may serve to explain the change in the formulae in public 
documents. The name of Euergetes appears in them frequently in the 19th, 
21st and 24th years of the reign of Philadelphus (all subsequent to the 
death of Arsinoe). In the 27th year the name of Euergetes disappears. Now 
in the 27th year, 258 B.C., the plot of Demetrius the Fair at Cyrene was 
discovered, and the betrothal of Euergetes to the daughter of Magas was 
ratified. At this time the other Berenike was not disposed of: and if 
there was a party which advised a marriage between Euergetes and his 
sister, it may have been strong enough to bring about the withdrawal of the 
Crown Prince's name from public documents when he was betrothed to his 
cousin. In the year 248 B.C., probably, the policy favouring the marriage 
with the Cyrenaean Berenike once more triumphs. Berenike, daughter of 
Philadelphus, is married to Antiochus II., and in the folloiving year — the year 
of his accession — Euergetes at length marries his cousin Berenike. Ptolemaic 
history hardly justifies us in regarding these events as a final solution of 

^ Of. Lumbroso, 'II arrirait toutefois que le roi constituftt son successeur ou confi&t le choix 
a son fepouse' (Justin 39, 3\ 


the difficulty. This solution comes only with the death of Berenike in Syria 
in the following year, an event entailing the dissolution of the party 
favouring her interests and perhaps connected with the ' domestic revolution ' 
which, according to Justin and S. Jerome, recalled Euergetes from his 
campaign in Asia. These disturbances were finally settled in 238 B.C., 
when the priests in assembly at Canopus conferred on the king and queen Canopns Dec. 
' well established monarchy ' — oi 6eol BeScoKaaiv avroh exxnaOoixrav ttjv 
fiaaCKeiav Koi hoiaovaLv tolW dyada Trdvra iU rov del '^povov. To under- 
stand the queen's position we must remember that she was the granddaughter 
of Ptolemy I. and Berenike I.^ It is the queen's head that appears on the 
obverse of six copper coins quoted by Svoronos (Coll. of Joh. Demetrius), Joum. InUr. 
while on the reverse is the inscription UroXefiaiov fiaat\€(i)(;. Hence on the ^^^ j^p^ 2,' 
king's coins appear the insignia of the queen.^ 1898. 

It is not known whether the child Berenike so exceptionally honoured in 
the Canopus Decree was the eldest child ; but this assumption alone seems to 
account for the supreme honours conferred on her at her early death, and it is Canop. Dec. 
corroborated by the evidence of the heraldry on her crown : — elvat Bk ttjv 
€7nTi6€fiivr]v ^aatXeiav rfji eUovi avTTJ<; Bia<f)€pov<rav t^9 iirnidefiivr}'; raU 
eUoaiv tt)? firfTpcx; avrri<i ck aTa^voov Bvcov, b)V dvd fiiaov ear at r/ da-ni- J^eps. Ban. 
SoeLBr)<i ^aaiXeia (Hieroglyphic version translated by Lepsius — ' seiend eine 
Urausschlange zwischen ihnen '). Strack (op. cit., p. 5) bases his argument 
that the title ^aacXca-aa does not in itself imply association in the goremment 
on the fact of its application to this young princess. He argues that it is 
impossible to suppose that the title implied anything of the kind in her case. 
We know, however, that Epiphanes was associated (and possibly also 
Philometor) while still an infant. Strack admits, moreover, that 
Cleopatra II. and Cleopatra III. were queens regnant and yet they contented 
themselves with the simple title ^acriXicra-a. 

Ptolemy IV. Philopator must have been grown up on his accession in 
222 B.C., for demotic scholars say that he was formally associated with his Mah. ^mp. P/. 
father in the sovereignty and probably did some of the official work during 
his father's decaying activity. He was not married, however, nor did he 
marry for many years after his accession. We know that his mother 
Berenike II. survived her husband : even assuming that she was thirteen or I^ps^- -^J*- 
fourteen at the time of the crisis at Cyrene, she may quite well have lived up p, 503. 
till the date of her son's marriage in 213 B.C., as she would only be about 
sixty then. She may, however, have died some years earlier, when Philopator 
was too much occupied with his foreign campaign to think of matrimony.' 
Hence the delay of probably nine years between his accession and his marriage Mah. .Bmp. Pt. 
Table, p. 20, 

5127. ^ In the Adulitan inscription in the list of before Berenike's marriage, and that Euergetes 

ip.rt. territories which Euergetes received from his appears ou them merely as the betrothed of 

lote 3 father 7rapa\a0iiv vapi, tov irarp6s there is natur- Berenike. 

ally no mention of Cyrene which came to him 3 In fact there seems reason to suppose that ^ah. Emp. V. 

from his wife. Philopator caused his mother to be put to p. 248&note4, 

2 I cannot believe with Svoronos that these death. Polybius, xv. ch, 25, 

coins are Cyrenian, belonging to the perio4 


with his sister, Arslnoe IIL, in 213 B.C., in which same year the king and queen 
ivere deified as gods Philcypatorcs, — an event which, I beheve, marks a change 
in his poKcy and a compromise brought about by the native revolt of the same 
year. The revolt may have been largely caused by this very delay in his 
marriage : his mother being now dead, national prejudice saw no reason for 
loyalty to the son, or perhaps allegiance may have even been transferred to 

Mah. Emp. Ft. the late queen's heiress-daughter : in fact we know what an important part 

Polvb V 69 ^^^^ princess played in the Raphia campaign. 

With regard to Epiphanes the evidence is neutral. His mother, 

Arsinoe III., wife and sister of Philopator, was dead before his accession and 

he himself was an only child. If we accept the testimony of Polybius xv. 25, 

the murder of Arsinoe III. was doubtless instigated by Philopator because of 

strack, op. her presumptive royal rights, and in the same year (according to Lepsius) 

*^ii'ote 1 ' Epiphanes was associated in the crown. In the three dedications quoted by 

Strack (p. 11), the queen's name occurs, as Strack points out, in the two 

first, while in the third it is omitted, and the epithets and titles of Philopator, 

p . ,. her husband and murderer, have become more fulsomely flattering. We 

Gk. Pap. X. know that Arsinoe III. was honoured with a cult at Alexandria.^ She 

^" ^* appears frequently on coins (Poole, PI. XV. 6, 7). Her death left her only 

child, Epiphanes, in the exceptional position of sole claimant to royal rights. 

Hence his marriage was unessential to his full recognition, and at his 

Anacleteria he received at the age of fourteen complete divine honours, as we 

know from the Rosetta Stone. There he is already treated to a two-fold 

divine'title — 0eo9 'E7rc(f)avr]<; Fjv^dpiaTo<;. Indeed as Mr. Mahaffy says,^' These 

honours are far more extravagant than those given to Euergetes, and are to 

be compared to the hono2irs assigned to the dead child Berenike! 

Mah. J??rtp. P<. Of the mysterious Eupator's brief reign (in the year 182 B.C.) and 

P' ■ matrimonial prospects we know nothing. If, however, he was the elder son 

of Epiphanes and Cleopatra I., as seems certain from his position in the nine 

Leps. Ahh. hieroglyphic lists mentioned by Lepsius, where he appears as ' der von einem 

464 sfff. p. 465. grossen Vater entsprossene Gott,' in each case immediately preceding 
Philometor, the absence of all records of this short reign may be due to 
preference on the part of the queen-mother for her younger son. One is 
tempted to see traces of two rival parties in the names conferred on these 
two brothers. In any case, the unobtrusive manner in which the child-king 
(he cannot have been more than ten years old) appeared and disappeared is 
easily accounted for by the ascendency of the queen-mother. The govern- 
ment centred in her and the death of one son and the assumption of another 
as co-regent -were matters of little importance. The younger brother, 
Ptolemy VII. Philometor, succeeded in the same year, the queen-mother, 

Mah. JFmp. /'<. Cleopatra I., surviving till 174 B.C. In the following year Philometor married 

^ Grenfell, Pap. X. mentions a I'epefa ' kpaiv6r\% date-formulae with the Athiophorus of Berenike 

*(Xoir<£Topoi ^f 'AAelafSpt/aj. So too, Pap. XII. and the Canephorus of A. Philadelphiis, — all 

1. 5 and XXV. col. 2, 1. 6, and XXVII. col. 2. these princesses having, I believe, claims on the 

This priestess of Arsinoe fref|uently appears in throne in their own right. 


his sister Cleopatra II., though he was at that time only fifteen years old.* 
It seems likely that this early marriage was prompted by reasons of state. 
As Epiphanes and Cleopatra I. had married in 193 B.C. and Philometor was 
not born till 188 B.C., Cleopatra II. as well as Eupatormay have been an elder 
child (Strack, op. cit. p. 197 note 19). Strack however considers Eupator 
to be a son of Philometor and Cleopatra II., and believes that Cleopatra 
reigned jointly with him for a few weeks after Philometor's death (Strack 
op. cit. p. 182). 

It is interesting to observe that the history of Egyptian royalty for a 
whole century from this time onwards practically means the history of the 
three queens, Cleopatra I., II. and III., whose ascendency forms an all but 
unbroken chain. We hear of the ' wise regency' of Cleopatra I. during her ^Vsii.Emp.rt. 
son's minority. One is tempted to believe that the title conferred on ''' 
Ptolemy VII. may have been a public recognition of her good government ; 
just as, I believe, the title Philopator may have been a national expression of 
gratitude to the nationalistic Euergetes perhaps forced on his reactionary son. 
Of this queen, Strack says (p. 3) 'sie hat sich mit den Rechten eines 
Vormundes begniigt und sich nicht zur regierenden Fiirstin gemacht.' We 
must remember that she was a Syrian princess and not a blood-relative of 
the late King Epiphanes. ' Erst ihre Nachfolgerin (Strack, p. 3) Kleo- 
patra II. hat diese Stellung fUr sich in Anspruch genommen.' And 
Cleopatra II. was the sister and wife of Ptolemy Philometor. 

On the death of Cleopatra I. she becomes the prominent figure in the 
state, and continues to be so in the troubled times that follow, only sharing 
her power towards the end of her life with her daughter and rival 
Cleopatra III. Before the year 172 B.C. (Strack p. 183) we find Cleopatra II. 
married to her brother Philometor. It is she who reconciles the rival 
brothers Ptolemy Philometor and Ptolemy Euergetes II., and induces them 
to reign jointly for a time (from 170 B.C.). On a Theban monument we find 
the three — the two brothers and their sister Cleopatra II. — oflfering to Leps. Abh. 
Amon-Ra as the three Philometores. ^ 487^ 

On the death of Philometor in 146 B.C. the exiled Euergetes II. 
returned to claim the throne and the hand of his sister and brother's widow 
Cleopatra II. He carries his point, and, according to Justin, on the very day Justin, 38, 3. 
of the wedding murders his bride's probably already grown-up son. This ^^^*p Iq-/^ 
son is no doubt the second Koxpov irpoffanrov of the dynasty — Ptolemy VIII. 
Philopator Neos. Here again the title seems to indicate that he may have 
been the candidate favoured by the party opposed to the queen-mother's 
claims. Mr. Mahaffy doubts the fact of his murder, and believes that Philopator Mah. Emp. Pt. 
may have died a natural though opportune death : but party-politics seemed ^' ' 

. Abh. ^ Letronne and Grenfell and others, however, death of the mother : hence in it the name of 

Ak. p.' assign the marriage to a later date, 165 B.C. Ptolemy occurs alone— according to Mr. Gren- Grenfell, 

^^' Lepsius is uncertain. fell's restoration [BacnKfjovros UroKfuaiov tov Gk. Pap. x 

In either case the Grenfell Pap. X. (174 B.C.) iK nroAfnalov nal KKtoirirpcis 9fS)v iiri\<pavS)v 1- 1» °ot* 1- 

belongs as it were to the watershed between trous oyi6ov, 
the two Cleopatras, immediately after the 

H.S. — VOL. XVIIl. S 


to necessitate the death of Philometor's son, and from what we know of 

Physkon's character it is not likely that he would hesitate to take the proper 

steps to secure his authority. Accordingly, no sooner had Cleopatra IL in 

the following year (145 B.C.) borne the king a son, called Memphites because 

Justin, 38, 8. he was bom at the time of the coronation in Memphis, than the king divorced 

^kXfm. ^^^ ^^^ married her daughter Cleopatra IIL— ' vielleicht nicht bloss ihrer 

grosseren Jugend wegen, sondern auch, weil sie als Tochter seines Bruders 

Philometor 7ia€h Aegyptischen Erfolgerecht, welches die weibliche Linie nicht 

ausschloss, fur sich oder ihren Gemahl die Krone hdtte heanspruchen konnen,' as 

Lepsius says : ' diesen Zweifeln kam er durch die neue Verheirathung zuvor, 

nahra aber dennoch bald darauf deren Mutter, seine erste Frau, wieder zu 

sich.' Kakergetes was determined to secure the right of succession ; and in 

this way he made it doubly sure by being married simultaneously to both 

generations of heiresses. We may judge to what extent this extraordinary 

Leps. Abh. coalition worked from the evidence of inscriptions. According to Lepsius 

^'^^\fo' ^' the triad appear together as gods Euergetae on the monuments in the years 

141 and 136 B.C. ; after the latter date Ptolemy appears with Cleopatra IIL 

alone in the years 126, 125, 124; in 124 again and in 118 (that is, to the 

death of the king) the king appears once more with both queens, the 

precedence being invariably given to Cleopatra the elder. In the years 

liah. E7np.Pt. 130 — 129 Ptolemy seems to have been in exile and Cleopatra II. reigns. It 

P' ^^^' may have been at this time that the king in revenge murdered his son (by 

Justin, 38, 8. Cleopatra II.) Memphites. Nevertheless the three seem to have been once 

^^*p ^392!' ^^' ^^^^ associated in the government — probably till the death of Cleopatra IL 

Leps. Abh. On the king's death in 117 B.C. Cleopatra III. is thus left in undisputed 

fol^"nott' I authority, being, as daughter of Philometor I., sole heir. 

This queen begins her long supremacy by associating first one son in her 

regency, then, some years later, she deposes him and chooses another. The 

elder son, Ptolemy Lathyrus, seems to have rebelled against his mother's 

authority, and she on her part tried to undermine his position. She forced 

op. cU. p. 472. him to divorce his elder sister and ivife, Cleopatra IV. and marry the younger 

^mp. PL sister, Selene, whose name does not appear on inscriptions.^ As Mr. Mahaffy 

^' ■ says, ' We can hardly doubt that by this arrangement she meant to avoid the 

association of the young queen with her son's and her own name in public 

acts, as had been the case when she herself had been the younger Cleopatra, 

for there was probably some strong Egyptian sentiment against giving these 

peculiar royal and divine honours to the younger members of the family.' 

Leps. Abh. The queen-mother and her elder son reign together till 107-6 B.C. as 

^Z^\t% S*^^^ Philometores Soteres. In inscriptions the queen takes precedence : e.g. 

p 492. Grenfell Papyri xxvii, xxv, Greek Protocol of Paris Papyrus of Osoroeris, 

Demotic Papyrus of Berlin, No. 13 (Leps. Berl. Ah. p. 493). Lepsius calls 

p. 494. attention to the fact that in this demotic papyrus (Berlin 13) the king does 

not receive the title ^eo? : he only does so in the Greek and in conjunction 

vol. ii. p. 16. — — 

Viscoiiti IcoTt 
grec, PI. 54 ' ^ Queen Selene's head, however, is found coin of this period with the customary royal 

17 Porphyr. (if the coin has been rightly read) on a eagle and the name of Ptolemy on the reverse. 

ap. Scaliger. 


roith Ms mother. This subordinate position was no doubt galling to Lathyrus ; 

he divorces Selene and tries to free himself from his mother's control, but in 

vain. In 110 B.C., indeed, we find him reigning alone, but the queen-mother 

is too powerful : Lathyrus is exiled and Cleopatra IIL recalls her younger son 

Alexander I. to be co-regent 107 B.C. In 99 B.C. we find the latter married 

to his niece, Berenike III., the daughter of Lathyrus and, according to Lepsius 

and Poole, of Cleopatra IV., hence the direct heir. In the same year the 

queen-mother, the king and queen, appear in an inscription in the following 

order — Ptolemy Alexander I, Cleopatra III., Berenike III. In 90 B.C. P- 605. 

Ptolemy Alexander likewise conspired against his mother's ascendency and 

put her to death. On the death of both Alexander I, and Lathyrus, Mah.-B»i;>./'<. 

Berenike III.* succeeded and reigned alone for six months ; but'soon after her ^" 

accession the son of Alexander I. and his first wife, Ptolemy XII. Alexander 

II. returned to Egypt from Rome and immediately married and murdered 

Berenike III., his step-mother and the legitimate ruler, whereupon he himself Leps. Ahh. 

fell a victim to the household troops. p. igo. 

If this king's will bequeathing the kingdom of Egypt to Rome is 
genuine, we may perhaps believe that he meant something different by this Mah. Emp. Ft. 
coup ^ from the usual conventional policy of state murder : he may have P' * " 
recognised that only by thus extinguishing the line of succession and 
entrusting his unhappy country to the firm control of Rome could he put an 
end to the hateful intrigues of his house and the miseries entailed by them on 
the kingdom. There is nothing, it is true, to prove this except the alleged 
will and the fact of his stay at Rome, where he may well have learnt to loathe 
the traditional policy of his ancestors. Moreover, his guards killed him, as Appian, 
Appian tells us, as aTOTrcoTepov (T(f>(ov, ola ^vWa weTroidoTa, i^rfyovfievov. 

But the plan, if such it was, was frustrated by Egyptian national feeling. 
An heir to the vacant throne was found in Ptolemy XIII. Neos Dionysos 
Philopator III. Philadelphus II., ' an illegitimate son of Soter II.' This king 
is looked on as immediate successor to his step-sister Berenike III., and the Leps. Abh. 
two Alexanders are omitted from the official lists of this time, as the claims p^ '^^q. ' 
of Neos Dionysos go back to his father Soter II. and thus exclude these kings 
from the succession. 

Now I would suggest that the so-called ' illegitimacy ' of Neos Dionysos 
was due to a misconception on the part of Greek and Roman historians of 
Egyptian rights of succession. Pausanias says of Berenike III. : — fj fiovrj Paus. i 9, 3. 
yvrjaia ol {i.e. Ptolemy Soter II.) r&v TratSltop ^v. Similarly Strabo implies 
that the great Cleopatra was illegitimate: — tovtov fxlv ovv (Auletes) oi Strabo^p. 796. 
AX€^avOpet,<} e^epaXov, rpirov o avrut dvyarepfovlova-fav, tav fna yvrj<Tia r/ p 429. 
Trpea-^vraTr}, ravTrjv aTriBei^av ^acriXia-aav.^ Lepsius objects to this state- Leps. Abh. 
— — p. 479. 

' Berenike III. bore the title Philadelphus — nothing of the fact. 
a title which seems to have been transferred to ^ Champollion-Figeac and Letronne take this op. cit. p. 479. 

her father Soter II. on his return from Cyprus. to mean 'one of whom (who was) legitimate 

(Strack, Dyn. dcr Ptol. pp. 4 and 63.) and the eldest was proclaimed queen.' But 

* It does not seem to be absolutely certain this, as Lepsius points out, ia linguistically 

that he murdered Berenike : Appian says impossible. 

s 2 


ment on the ground that it is highly improbable that such an important fact 
about the great Cleopatra should only receive this casual mention. He 
ascribes the misstatement about Berenike to a confusion on the part of 
Pausanias between Berenike III. and Berenike IV. I believe that in neither 
case does it mean that the other children were illegitimate, but that the 
Egyptian idea of the heiress-rights of the eldest daughter confused the Greek 
mind and led to the misconception that they were so. Besides, it is 
manifestly absurd to say that Berenike III. was ' the only legitimate ' child of 
Soter II. when we know that he had two children by his second wife and 
Leps. Berl acknowledged queen, Selene.^ 

Ak. p. 505. Hence I take both passages as a Greek mistranslation of the Egyptian 

idea the eldest daughter {and child) i.e. the only legitvnate heir. And while on 
the other hand it seems not unlikely that the claims of the eldest daughter 
in Egypt (if she was the eldest child) conveyed the idea to the Hellenistic 
mind that she alone was yvrja-ia, it seems possible that the very idea of 
illegitimacy was foreign to the Egyptians. We have the express statement of 

Diod. i. 80, 3. Diodorus to this effect : — yafMova-L Se irap Klyvirriot^ ol fji>ev tepet9 ficav, to)v 
8' dWcov 6aa<i av €Ka<TTO<i TrpoaLprjraf Kal tu yevvcofieva Trdvra rpit^ovaiv i^ 
dvayKT)^ ivexa rrj^ TroXvavOpcoTria^. . . . v66ov 8' ovSiva t&v yevprjdivTcav 

I would suggest, therefore, that Neos Dionysos was not illegitimate, 
but that he was a son of Soter II. and his second wife, Selene, the younger 
sister of Cleopatra IV., and was one of the children repudiated along with her. 
Lepsius says of Soter II.'^ ' er verstosst Selene mit zwei Kindern.' We do not 
hear what became of the children. What so likely then as that the 
Egyptians, on the extinction with the death of Berenike III. of the older 
branch descended from Cleopatra IV., reverted to the children of the younger 
sister, queen Selene ? 

Berl.Abh.Ak. O^ ^^^ wife of Auletes, Tryphaena Cleopatra V., Lepsius says: 'sie 

p. 476. heisst in den Inschriften zugleich Schwester des Konigs ^ and scheint daher, 
wie er selbst, ein illegitimes Kind des Soter gewesen zu sein.' It is not 
impossible that this is the other child of Selene, whom he immediately pro- 
ceeds to marry in orthodox Ptolemaic fashion, and thus the succession is duly 
handed on in the younger line. That Selene considered herself the legitimate 
heir on the death of her sister is proved by the fact that we hear of her even 
Cic. claiming the throne for her sons by a later marriage with Antiochus Grypus. 

in ^*^'^y "• Hence it seems to me extremely unlikely that the two children of her first 
marriage with her brother Soter II. should not be claimants, if still alive, for 
the kingdom, and it is to them that one v^ould a priori expect the Egyptians 
to turn on the extinction of the older branch of the family. 

Mali. Ei)ip. Pi. ' Mr. Mahaffy says : ' I cannot but think that appear from history as if they had no right to ^^^'\- ^ 

p. 427. the constant assertion of the illegitimacy of the throne, unless indeed Auletes was one of ° 

Egyptian princes and princesses was an inrcn- them, and he is always spoken of as ille- 

tion of Hellenistic historia^is in the interest of gitimate.' 

the Romans.' ^ One of the king's titles is Philadelphus. 

* Mr. Mahaffy says : ' These two children dis- 


In 58 B.C. Neos Dioiiy.sos (Anlete-s) is banished, and his wife and sister 
Tryphaena CMeopatra V. reigns witli her eldest daughter Berenike IV, as co- 
regent. The following year Cleopatra V, dies and Berenike IV. reigns alone 
for a year, during which time she selects and rejects a first king-consort and 
marries a second. But in 55 b,c. Auletes is restored and puts his daughter 
to death. On his death in 51 B,c, he left the throne to his daughter Cleo- 
patra VI., his eldest surviving child, and to her brother Ptolemy XIV. Not 
long after Cleopatra was driven out by the supporters of her brother, but on 
his death she was once more established as queen with Roman assistance, 
this time with her younger brother Ptolemy XV. as co-regent and nominal 
husband, and on the death of her boy-husband she assumed her son Caesarion 
as co-regent. We possess no inscriptions with the name of Cleopatra and her 
three successive co-regents, but as Lepsius remarks: — ' wahrscheinlich g\ng Berl.Abh.Ak. 
ihr Name als des altesten Kindes dem ihrer Briider voraus, und die letzteren ' 

wurden, wie spater ihr Sohn Ptolemy XVI, Caesar, nur als Mitregenten 

On this theory we see the same stereotyped principle of succession at work 
throughout the whole of the Ptolemaic period down to the extinction of the 
race, even Roman intervention conforming to it, and one consistent explana- 
tion is found for the most unhellenic feature of Ptolemaic history : whereas 
Strack (whose book on the Ptolemies appeared after the main part of this 
paper was written) is obliged to adopt three separate explanations of the Strack, op. «< 
brother- and-sister marriages : — firstly, in the isolated case of Philadelphus the ^" 
marriage is explained by Arsinoe's rights to certain cities of the Euxine : 
secondly, in other cases the marriages were due to a desire to avoid dangerous 
alliances with foreign states : thirdly, from Cleopatra II. onwards the prin- 
cesses had emancipated themselves and were really queens regnant, and the 
marriages represent an adjustment between the two claimants. Can we 
believe that the first two shadowy inducements were sufficient to cause this 
purely Hellenic dynasty to embark on a course so strangely at variance with 
Greek sentiment ? The marriage of Philadelphus alone seems a clear indica- 
tion that the reaction had already set in, that Philadelphus,whether he would 
or not, could not stem the advancing tide of Aegyptisirung which closed 
over his succcs.sors. And yet we are told (Strack op. cit. p. 104) that in the 
matter of succession the first half of the dynasty ' sind griechischen Sitten 
getreu geblieben.' Can the persistent mention in public acts of both parents • 
of the sovereign be looked on as a Greek custom? Strack assumes two 
distinct periods : in the first period down to the time of Cleopatra II. the 
royal princesses had no rights of succession : in the second period they 
emancipated themselves and ' es war durch diese Gleichstellung der Konigin Strack of?. cit. 
der naturliche Gang der Erbfolge gestort, der nicht besser wieder hergestellt P- ^^• 
werden konnte, als wenn die zwei Gleichberechtigten durch Heirath ihre 
Anspniche vereinigten." Is it more unnatural to assume that the same 
principle was involved all along, a principle which became more defined 
certainly in the later half of the dynasty but which was none the less surely 
at work in the background from the beginning ? It is true that the early 


queens of the dynasty were not queens regnant in the sense that the Cleo- 
patras were queens regnant. But from the point of view of the divinity of 
the sovereign their position, I hold, was paramount, and on the divinity of 
the sovereign depended his recognition as king. Till he was recognised by 
the priesthood the loyalty of his Egyptian subjects was not worth much. 
And as Strack says (p. 128) 'eine Consekrierung als Landesgott kann nur 
von ihnen {i.e. the priests) ins Werk gesetzt werden, nur durch sie voile 
Giiltigkeit erlangen.' That such an idea was no new invention of the Hellen- 
istic dynasty there seem sufficient indications on early Egyptian monuments 
to testify : by a thorough investigation of the principle in Ptolemaic times 
much light, I believe, might be thrown on Pharaonic history social and 
political, and the imperfection of the record to some extent supplied. 

Erman, p. 73. Of ancient Egyptian royal matrimony Erman says : * There was only one 

legal wife, the queen ; she was of royal or of high noble birth, and indeed she 
may have been the " daughter of thefgod " i.e. of the late king, and therefore the 

Erman, p, 74. sister of her husband.' Again : — ' The queen appears as a rule to have been 
of equal birth with her husband ; she took her share in all honours.' ' After 
the death of her husband the queen still played her part at court, and as 
royal mother had her own property, which was under special state manage- 
ment.' The queen of the Old Empire is called : ' She who sees the gods 
Horus and Set ' {i.e. the possessor of both halves of the kingdom) ; under the 
New Empire she is called : ' The Consort of the god, the mother of the god, 
the great consort of the king,' and her name is enclosed like that of her 
p. 152. husband in a cartouche. Though polygamy is the exception Erman points 
out that royal double marriages frequently occur ; in these one of the two is 
apparently due to political reasons. Such double marriages are found too in 
the case of private individuals, ^ for, as Erman adds, ' many daughters of 
rich men in Egypt possessed valuable rights of inheritance in their father's 
p. 155. According to Erman again : ' The esteem which the son felt for his 

mother was so great that in the tombs of the Old Empire, the mother of the 
deceased is as a rule represented there with the wife, while the father rarely 
appears. On the funerary stelae of later times also, it is the usual custom to 
trace the descent of the deceased on the mother's side, and not as we usually 
do, on that of the father.' Moreover, the maternal grandfather Avas considered 

Erman, p. 156. the natural protector and guardian of a young man. When a youth gets an 
appointment, then ' the father of his mother thanks God.' In the New 
Empire a post is conferred on a young man ' for the Sake of the father of his 

^ The case of Amony, ' the great man of the certainly three daughters and a son. ' A 

South, ' who probably died at the beginning of curious circumstance shows us, ' says Erman, 

the reign of Araenemhet II., reminds one of ' that the two wives were friends, for the lady 

Philadelphus and of the relations subsisting Nebet-Sochet-ent-Re called her second daughter 

Erman, p. 151. between the two first Arsinoes. Of Amony's Hunt, and the lady Hunt carried her courtesy 

two wives, one, Nebet-Sochet-ent-Re, may have so far as to name all her three daughters Nebet- 

been his niece ; she bore him two sons and five Sochet-ent-Rc. 
daughters ; by the other, Hunt, he had 


mother ' ; and when he goes to the wars he ' gave his property into the charge 
of the father of his mother.' 

Such prominence of the maternal male relatives is, as we know, a 
marked feature of the ' maternal system.' Tylor objects to the term Tylor, 
Matriarchate : ' The term matriarchal,' he sayH, ' takes it too much for century 
granted that the women govern the family. It is true that in these com- July 1896 
munities women enjoy greater consideration than in barbaric patriarchal life, 
but the actual power is rather in the hands of their brothers and uncles on 
the mother's side.' In the same article he alludes to the custom of the 
heiress-husband ' where the incoming husband marries the daughter of the 
house to which he succeeds in his wife's name : ' and again ' from Africa may 
be quoted Livingstone's account of the Banyae in whose country the wives 
are masters.' ' It is in Africa,' says McLennan, * that beenah marriage is now McLennan, 
most prevalent ; there are parts of Africa in which it is quite commonly met xh^oiTj^^p 43 
with — usually alongside of, and in some sense contending with, a system of 
marriage by purchase — the two systems, indeed, being generally in use even 
among the same people, the one preferred in some cases, the other in others.' 

It would be outside the scope of this essay to examine the traces of 
similar survivals in other countries. I should like to mention, however, an 
instance of what appears to me a Hellenic idea based on this Egyptian point 
of view : I mean the Libyan Medusa (MiBova-a), ' the reigning lady,' in whose 
snake-girt head I would trace a Greek representation of the royal uraeus- 
snake on the reigning queen's crown.^ 

We know from Manetho of a legendary law which under Binothris, a Petrie, 
king of the second dynasty, was passed establishing the lawfulness of female ^^^l' "^^23 
succession to the throne.^ The rule of women, however, as Poole remarks, Poole, 
seems to have been disliked, and the queens' names are omitted in the lists ^/T^* i^* 
made under dynasty XIX. ' when the royal family seems to have been i'.' 719. 
affected by Semitic influences.' It is in the immediately preceding dynasty, 
however, that we find the most striking instances of queens regnant and of 
powerful queens-consort married to a brother. As Erman says : 'In the royal Eiiiian,p.l54. 
family of the XVIIIth dynasty, we find that A'hmose-nefert-'ere married her 
brother A'hmose ; a lady named A'hmose ^ was consort to her brother 
Thothmes I,, and A'rat to her brother Thothmes IV., and so on.' Similarly 
the bitter rivalry between the brothers Thothmes II. and Thothmes III. and 
their sister the great queen Hatasu would not strike us as out of place if it 
were recorded on a page of Ptolemaic history. Hatasu reigned first with the 
elder of her brothers, Thothmes II., as her husband's co-regent ; on his death Erman, 1 . 43. 

1 It would thus be significant that the head rather to indicate that female succession was 

of the slain gorgon is depicted on the shield of ceasing to be looked upon as an undisputed 

the motherless Athene, 'the symbol of the fact — that Egypt was gradually turning its 

overthrow of motherdom and of gynaikocracy ' back on the matriarchate and tending towards 

as she has been called. Cf. McLennan, Primi- a system of agnation. 

live Marriage, p. 258. ^ It is surely significant that this queen- 

"^ I cannot regard such a law at this early consort had the name of a former king. 
stage as a ' progressive measure ' : it seems 


— of which she was perhaps the author — she became sole niler, though the 

younger brother, Thothmes IIL/ was nominally co-regent. After a reign of 

twenty years she was succeeded by Thothmes III., who forthwith erased her 

p. 43. name from the monuments : ^ ' again we cannot help,' says Erman, ' suspecting 

violence to have been the cause of the change of government.' ^ Hatasu seems 

Rawlinson ^0 have tried to evade the prejudice against petticoat government by having 

Egyp^, P- 178- herself frequently represented in masculine attire and with a beard. In some 

of her inscriptions she is called ' the king,' though the personal pronouns 

referring to her remain feminine, such jumbles arising as 'His Majesty herself.* 

Petrie.Fw^.o/ According to Petrie, Hatasu was the sole legitimate heiress of Thothmes I., 

pf,' 96-96 Thothmes II. being his son by another and not royal wife, and Thothmes III. 

a nephew of Hatasu. Of Hatasu he says : ' Her father about five or six 

months before his death associated his daughter with him as she was the 

heiress in the female line, in which royal descent (like that of private 

p. 66. families) was specially traced.' ' It appears that on failing health the king 

placed the power in the hands of his eldest child, who had the sole right to 

it by the female inheritance, and then, just a few weeks before his death, 

married Thothmes II. to her, perhaps to secure his receiving some respect for 

his position if not for his character.' 

Two other queens, both belonging to the XVIIIth dynasty, seem to 

deserve special notice, — Tyi, the wife of Amenhotep III., and Nefertiti, the 

wife of Akhenaten (Amenhotep IV.). On the question of these ladies' rights 

Petrie ^® have the conclusive testimony of Mr. Petrie. Of Tyi he says : ' her titles 

Hint, of Eg. are noticeable: she is called "princess of both lands" and "chief heiress- 

vo . 11. p. . pj^j^ggg q£ j^ij lands." These titles seem to imply hereditary right ; indeed 

it is very doubtful if a king could reign except as the husband of the heiress 

of the kingdom, the right to which descended in the female line like other 

property.' This Tyi was the mother of the heretic king Akhenaten. Petrie 

says again of her : ' There can be little doubt of the influence of Queen Tyi ; 

she appears closely associated with the king on his monuments, her figure is 

seen side by side with his on scarabs, her name appears along with the king's 

on innumerable objects, a temple was built in her honour and she acted as 

Petrie, Hist, of regent for her son during his minority.' * The beginning of the reign of 

Eg. p. 207. Amenhotep IV. is obscure. That Tyi for a brief time held the -power at 

Tell el Amarna, is indicated by her name appearing alone in a quarry at that 

Petrie, place,' — and this though Amenhotep IV. had probably already been associated 

•Hiat. of Eg. as co-regent with his father in the years before his marriage. 

' Mr. Petrie, however, calls Thothmes III. a indebted for the continuance of her memory 

nephew oi H&t&sM. Petrie, Hist, of Eg. vol. i., among mankind to the accident that the stone- 

p. 66. masons employed to erase her name were too 

' According to Mr. Griffith it was the later careless or idle to catrry out their work com- 

Eameses II. who caused her name to be pletely. It does not seem altogether fanciful, 

Eg. Ex. Fund replaced by that of Thothmes II., 'not con- therefore, to believe that the record of other 

Rejort sidering Hatasu a legitimate sovereign of queens may have perished more effectually 

1895-6, ji. 7. Egypt.' By the time of this king's reign Semitic owing to similar outbursts of hatred on the part 

influences were strongly marked. of their male relatives or of prejudice in their 

' As Rawlinson remarks, this great (jueen is successors. 


Of Queen Nefertiti Petrie says : ' That Nef'ertiti liad a liereditary claim 
to the Egyptian throne, is sliown by )icr titles. She was " the great heiress- p. 209. 
princess of all women" and " princess of Soutli and North, the lady of both 
lands." These titles, like those of Tyi, imply an hereditary right to rule 

It is unnecessary to point out the similarity of these titles to those 
applied, as we have seen above, to Arsinoe Philadelphus. 

Thus there seems sufficient evidence to show that the position of 
Arsinoe II. and of subsequent Ptolemaic queens was by no means anomalous 
in Egypt, but was based on Pharaonic precedent. 

It is certainly more difficult and hazardous to maintain that such a 
position of women was a survival from the Matriarchate, and to assign to its 
place in the development of society in the direction of Agnation such a 
custom as brother-and-sister marriages.^ On the other hand the assumption 
that these rights of women, which, as we have seen, go back to the beginnings 
of Egyptian history, were due to a huge progressive movement, is surely far 
more incredible when one remembers how slowly the current of change 
flows — or rather stagnates — in all other aspects of the history of that nation.^ 

It seems to me from all these facts given above that brother-and-sister 
marriages in Egypt may have been a relic of some primitive system of 
marriage based on female rights of property ; that such marriages point to a 
stage of growing prejudice against women's rights of inheritance and to an 
attempt at compromise ; ^ that these rights and this compromise were 
supported by the priesthood and notably by the Osiric cult (with its supreme 
tenet, the holy marriage of Osiris and Isis), in deference to the less civilised 
portion of the Egyptian nation : hence that it is this controversial and 
transitional stage that is most prominent under the New Empire and in 
Ptolemaic history. 

This, I think, is corroborated by the fact that it is the divine rights of 
the queen that are most insisted on. All political rights claimed for her 
were based emphatically on her rights as a deity, as we see in the case of the 
transference of the dirofiocpa in Ptolemaic times ; and hence, too, the 
importance of the different stages of her deification. We may almost look on 
the exaggeration of the spiritual claims made for her as a sort of compensation 
to her for the loss of temporal power, the salve with which the troubled 

1 To do so would involve the question also Lewy — de Civili Condicione Mulierum Grae- 

whether primitive Egyptians were endogamous carum (Vratislaviae, 1885), p. 59 : Si cui neque 

or exogamous. Ii seums probable at any rate filii sunt legitimi nee a defunctis filiis nepotes, e 

that the Egyptians did not practise infanticide. lege iam Mosaica hereditatem filiae consequun- 

■■' Traces of the former independence of tur ; quarum quidem nubere intra gentem est, 

Egyptian women seem to have survived down ut conservetur res familiaris. In Graecia de 

to the present day. Mr. Lane says of modern filia hereditaria praecipuae leges scriptae ex- 

f -5 and Egypt : ' I believe that in Egypt the women stant, ut nomen familiae ne exstinguatur. 

■0 of the j^^g generally under less restraint than in any Quanquam filiae non ipsae heredes esse, sed 

nians other country of the Turkish Empire.' una cum heredio proximis genere obtingere 

cljvi. ' ' Cf. Numbers, eh. xxvii. 1-11 and xxxvi. ; videntur. 



Erman, p. 295 
Cf. Wiede- 
mann, Herod 
ii. p. 151. 

op. eit. p. 152, 


Citiet and 

Bishoprics of 

Phrygia, p. 7. 

op. cit. p. 94. 

public conscience soothed itself while gradually depriving her of actual 
political power. 

This applies, I believe, to the rights not only of the queen but of women 
in all ranks. Thus we find that in the New Empire women assume a new 
and important role in connection with temple ritual. In all temples we find 
. female singers or musicians in great numbers. Indeed Erman says : ' we 
scarcely meet with one lady under the New Empire, whether she were 
married or single, the wife of an ecclesiastic or of a layman, whether she 
belonged to the family of a high-priest or to that of an artizan, who was not 
thus connected with a temple.' ^ Wiedemann says that at this time almost 
every god and goddess had a priestess. 

It is precisely this sacred side of woman's functions that we should 
expect to find preserved among the changes of developing civilisation.^ 
Professor Ramsay shows how in the religion of Phrygia traces of the earlier 
Lydian substratum of the Matriarchate are preserved, while socially the 
Patriarchate established by the conquering Phrygians has all but blotted out 
the native system. He says : ' long after a higher type of society had come 
into existance in Phrygia, the religion preserved the facts of the primitive 
society, but it became esoteric, and the facts were only set forth in the 
mysteries.' Similarly, whatever .the incoming patriarchal element was in 
Egypt — whether Semitic or not — it is, I believe, in the religious side of 
native life that we must look for the most certain traces of the older state of 
society, and I have tried to show that it is chiefly the Osiric system which 
affords such survivals ; and hence the greater development and diffusion of 
that cult and its adoption in a fashion as the state religion by the Ptolemaic 
government would explain an apparent revival of principles which seem to 
have formed a dim and half-realised background throughout Egyptian history 
— 'the dust of antique time' which has lain unswept in the most conservative 
of all lands, inhabited by a people ^eoo-eySee? Trepto-o-w? i6vT€<; fidXiara 
irdvTav dvdpcoTrcov. 

Rachel Evelyn White. 

^ Cf. Can. St. 11. 64-73 for these clerical and 
lay (toTs iWais irapdivoa raii ^ovKofiivais) 
female temple-nnisicians in Ptolemaic times. 
By this decree the wives and daughters of the 
priests come iu for a share of the rpo<pa\ iK twc 
lepS>v as priestesses of the dead princess 

^ Cf. the Greek woman's power of manu- 
mission : Lewy — de Civili Condicione Mulienmi 
Graecarum (Vratislaviae, 1885), p. 25 : 

manumittendi potestatem propterca feminae 
habent quod antiquitus in libertatem vindicari 
servus non poterat nisi alicui deo consecratus 
qui hierodulus fieret (Curtius, Anecd. Delph. 
p. 10 sqq.) : consecrare autem ipsi mulieri 

For a similar religious survival cf, the 
institution at Rome of the rex sacrificulus on 
the abolition or rather disintegration of the 

V-: ' 




[Plate XIV.] 

The Editors of this Journal have reason to think that a considerable 
body of its readers will be glad to be furnished with reproductions of the 
monuments, mainly vase paintings, that are discussed in connexion with 
Bacchylides. Even those to whom the published illustrations are readily 
accessible will probably find it convenient to have them brought together for 
reference, in a collection made from this point of view.^ 

III. The story of Croesus. — In the story as told by Bacchylides, Croesus 
voluntarily ascends his pyre, with wife and daughters ; he invokes the gods, 
and more especially Apollo ; he orders the pyre to be kindled ; Zeus ex- 
tinguishes the flames, and Apollo takes Croesus and his children to the land 
of the Hyperboreans. According to Herodotus, Croesus with his companions 
is placed on the pyre by order of Cyrus. After Cyrus has changed his 
mind, and his servants have made ineffectual attempts to extinguish the 
pyre, Croesus invokes Apollo, who extinguishes the flames. Fig. 1 ^ shows 
the well-known vase in the Louvre, representing the subject. It had already 
been interpreted, before the discovery of Bacchylides, as evidence of an 
alternative version of the story, in which the sacrifice was voluntary.^ 
Croesus sits enthroned, and makes a solemn libation, while an attendant 
Euthymos is busy with the pyre. Some of the commentators interpret the 
objects that he holds as torches, but they are quite unlike torches, as usually 
represented (cf. J.H.S. xi. PI. 6), and resemble more nearly the whisks for 
sprinkHng lustral water. If this is the correct interpretation their use 
further emphasises the ceremonial character of the scene. 

The vase is an early red-figured one, dating from the close of the 6th 
century or the first years of the 5th century B.C.* 

^ M. Theodore Reinach's edition {Poimes ^ Monuvienti del/' Iiutt. i., PI. .^4 ; Welcker, 

Choisis de Bacchylide . . .par Engine D'Eichthal Alte Dcnkmaeler, PI. 33 ; liaumeister, Denk- 

et T. E.), which has appeared since this article maeler, p. 796 ; Reinacli, p. 25, etc 

was put in hand, has to some extent covered the ^ E. Curtius, Qriech. Oesch., Gtli cd. , i., 

same ground, for those parts of the poet with p. 574. 

which he deals. The same monuments appear, "• Cf. notes by H. Stuart Jones and Miss 

in several cases, in both collections, but this Harrison, Class. Rev. 1898, pp. 84, 85 ; Jebb, 

is inevitable in connexion with the less-freijuent Mdanges Henri ^Vcil, p. 237. 



The extinction of a pyre by tlie influence of Zeus is strikingly illustrated 
by the vase of Python,^ in Vol. XI. PI. 6 of tlio Hrllrnic Journal. There 
Alcmene is seated on an altar, before which the ])yre is piled,^ and Antenor 
and Amphitryon apply the torches. Zeus listens to the appeal of Alcmene, 

Fig. 1, 

and has cast his thunder-bolts. Copious rain is poured on the pyre by the 
Hyades, and also falls from a rainbow-like cloud. 

* A red-figiired crater in Brit. Mus. , No. F. 
149. Formerly at Castle Howard. Late fourth 

century B.C. 
2 Murray, J.H.S. xi.,p, 226. 


The poet does not say in what way the god conveyed Croesus and his 
family to the country ot" the Hyperboreans, but one may guess that on this 
occasion, if the poet had been more explicit, Ikj would have given him the 
swan rather than the Gryphon or the Tripod. 

The Gryphon was ])rimarily an artistic type, inherited by the Greeks 
from the East, and thence introduced into literature, and not in the first 
instance, with special reference to Apollo. In literature, it curiously happens 
that the earliest passages that have been pointed out, which definitely refer 
to the Gryphon as an attribute of Apollo, are in Claudian ^ and Sidonius 
Apollinaris.' In art, the association of the Gryphon with Apollo is not 
common before the 4th century. Apollo also travels on his Trip(xl, but not 
on a Hyperborean expedition. 

On the other hand, Alcaeus^ had sung how Apollo had travelled in a 
chariot, drawn by swans, from Delos to the Hyperboreans, and from the 
Hyperboreans to Delphi. When Pindar describes how Apollo carried otf 
Cyrene to Africa, he says that he bore her in a golden chariot.* A late gem, 
at St. Petersburg,^ shows the rape. Apollo stands in a chariot, drawn by a 
pair of swans, and clasps Cyrene by the waist with his right hand. 

V. Heracles and Meleagcr. — This episode has already been the subject of 
much discussion.* I need only indicate the artistic types involved. 

Heracles atid Cerberus. For this type, see the drawing published else- 
where ^ in the present volume of the Hellenic J&urnal from a late sixth- 
century black-figured vase, recently acquired by the British Museum.^ No 
monument has been identified with the conversation between Heracles and 

The Death of Meleager is vividly represented on a large Aniphoru from 
Armento, in the National Museum at Naples," of about 400 B.C. (Fig. 2). 

The young Meleager is seen in agony, supported by his brother and sister, 
Tydeus and Deianeira, while a third figure, who has been variously named, but 
is probably Althaea, approaches in haste from the left. Oineus stands as a spec- 
tator. Above him is another group of Aphrodite and Eros. Near Eros, where 
his name might be expected, is the inscription (f>06vo<;. It would be out of 
harmony with the usual direct simplicity of vase inscriptions, if we look for any 
conceit, such as that Love and Envy are near akin. It has been suggested 

^ De VI. Cons. Honor. 30. Auserlesenc Vasenbilder, Pis. 129-181. For 

" Carm. ii. 307 ; xxii. 67. See Stephani, red-figured vases, see Wiener Vorlcgehldtter, 

Compte Rendu, 1864, p. 57. Series E, Pis. 1-4. For a list of forty-nine 

' Bergk, Poetaciyr. (?r., 4th ed. iii., Alcaeus. representations of the subject, see Waltei-s, 

2-4. J.E.S. xviii., p. 296. 

♦ Pindar, Pyth. ix. 6. » Arch. Zeit. 1867, Pis. 220, 221 ; pp. 33, 97. 
' Overbeck, Oriech. Kunatmythologie iv., Heydemanii, Vasensammlungcn dcs Mus. 

p. 495. Nazionale zu Neapel, Sautangelo, No. 11 

^ Robert, Hermes xxxiii., p. 151 ; Croiset, (where see references to earlier literature). 

Melanges Henri Weil, p. 73 ; Reinach, p. 4. Roscher, Lcxikon, ii., p. 2620 ; Reiuach, 

^ J.H.S. xviii., p. 295. Bacchylidc, p. 5. 

* For black-figured vases, see Gerhard, 



as a possible explanation that the inscription refers to a personification that 
has been omitted in the transcript from a larger composition. 

Peleus and Theseus sit below in the attitude of mourners. They are 
probably introduced as two of the most noted of the companions of Meleager 
in the Boar hunt. 

The reverse of the vase has a scene in the lower world with Heracles 
leading Cerberus — but unfortunately without Meleager, 

Fig. 2. 

IX. Death of Archemoros. — This subject occurs on several monuments,^ 
but in no case in such a way as to contribute to the interpretation of line 13. 

The serpent is attacked by several of the heroes, while the child lies 
near, or encircled by it, but there does not appear to be any representation 
of the moment before the death of the child. 

1 Overbeck, Gallerie Eeroiacher Bildwerke, Pis. 8, 4 ; White Athenian Vases in Brit. Mus. 
PI. 18. 



The funeral rites are shown on a large vase now in the National 
Museum at Naples, and frequently published.^ In the foreground, the body 
of Archemoros is laid out on a couch, tended by several persons, of whom the 
Pedagogue alone is named. Above, within a tetrastyle Ionic building, is 
Eurydice in conversation with Hypsipyle and Amphiaraos. To the right are 
Parthenopaios and Capaneus, to the left Euneos, and probably Thoas, the 
sons of Hypsipyle. In the upper tier are Zeus and Nemea (on the right) 
Dionysos and a Satyr, restored, (on the left.) 

Thebe (ix. 54 and x. 30) is seen personified on the Cadmos vase of the 
late Italian painter Assteas ^ (Fig. 3). She is shown as a female figure 
seated above the spring which is guarded by the dragon. 

It is probable that she also occurs on a kindred vase now in the Louvre^ 
as a richly dressed maiden, who watches Cadmos making his onslaught on 
the dragon. 

XI. The healing of the daughters of Prpetos. — This legend has hitherto 
been known in two principal forfns : — 

(1) Melampus, with the aid of a band of vigorous young men, chased 
the Proetidae to Sikyon. Iphinoe, the eldest of the daughters, died on the 
road, and the others were cured.* 

(2) Melampus cured the Proetidae at Lusus or Lusi.* 

Bacchylides places the cure at Lusus, but altogether omits to mention 

' Heydemann, No. 3255 ; Overbeck, I.e. S. Reinach, Peintures de Vases antiques, re- 

PI. 4, Fig. 3. ctieillies par Millin, etc., ii. 7. 

2 Millingen, Ancient Unedited Monuments, * Apollodorus, Bibl. ii. 29 ; cf. Paus. ii. 

PI. 27. 7, 8. 

8 Millin-Dubois Maisonneuve ii., PL 7 : " Paua. viu. 18, 7. 

272 A. H. SMITH. 

the agency of" Melampus. The bringing of Melampus to Lusi seems to be 
due to a confusion of two stories, but that it was current in late times is 
shown e.g. by the epigram over the fountain near Lusi — 

<f)6vye 8' ifiTjv Tnjyrjv fjnadfiTreXov €V0a MeXayLtTrou? 
Xva-dfievo'i Xy<TO->;9 ITpotTtSa? dpya\er]<; ^ k.t.X. 

The subject is believed to be represented on a fourth-century vase in the 
National Museum at Naples ^ (Fig. 4). Three maidens arc grouped in 
humble positions round and near a xoanon before which is an altar, and a tripod 
on an Ionic column. The xoanon is probably that of Artemis.^ 

On the left are an elderly figure with a sceptre, and a rustic old man, 
who has been called Silenus, with a thyrsus. On the right is Dionysos. 

On the version of the legend which brings Melampus to Lusus, the 
bearded man has been so named. One of the three maidens, the wild 
figure behind the column, has been called Lussa or Madness by Wieseler,* 
on the ground that Iphinoe is already dead. 

If, however, we study the vase in connexion with the text of Bacchylides, 
the bearded man would be Proetos, who comes to Lusus, and makes a prayer 
to Artemis on behalf of his daughters. 

The presence of Dionysos as a spectator may be due to the fact that 
according to Hesiod (so at least we are told by Apollodorus ^) the madness of 
the daughters was due to their not accepting the mysteries of Dionysos. 

It may be supposed that there was already a cult of Artemis at Lusus, 
when Proetos made his prayer there, but in any case it is a very easy 
prolepsis for the artist to show us the altar, xoanon, tripod and votive tablets 
indicating the temenos which was established by Proetos in gratitude for the 

A cameo,^ formerly in the possession of M. de Witte, was thought by its 
owner to represent the same scene. In this instance the supposed Melampus 
holds up a young pig, which was specially employed for rites of purification 
in the case of persons recovering from insanity. If, however, it is correctly 
interpreted the cameo evidently represents a different version from that of 

XIII. Heracles and tlie Nemean Lion, — The invulnerability of the Lion, 
upon which the poet lays stress, was not a fixed point in the story as told by 
the early vase painters. 

^ Vitr. viii. 3, 21. xoanon of Hera.' 

2 Millingen, Vases Antiqites, PI. 52, and S. * In Miiller's Denkmaeler, I.e. Compare the 

Reinach, op. cit. Millingen 52 ; Heydemann, introduction of Lussa by Euripidea in the 

No. 1760 ; Miiller-Wieseler, Denkmaeler i., Hercules Furens and of Mania in the vase of 

PI. 2, No. 11 ; De Witte, Gaz. Arch, v., p. 126 ; Assteas, Mon. dell' Iiutt. viii. 10. 

Frazer, Paiisaniaa iv., p. 259. ' Bibl. ii. 26. 

' It has also been thought to be an image of ' De Witte, Gaz. Arch, v., PI. 19, Fig. 1 ; 

Hera. According to Acusilaos, the Proetidae of. De Witte, ibidem. 
had gone mad because they ' disparaged a 

TT,r<rsTR \Tro\M to ii.\('('iivi,ii»ks. 






The scheme of the strangHug occurs in the majority of the black-figured 
vases ^ and in the more occasional red-fignred representations of the 
scene. But the sword also occurs, e.g. on the black-figured amphora in the 
British Museum, No. B 160 (Fig. 5), and elsewhere.- In literature, the invul- 
nerability of the brute is suggested by Pindar (Isthm. v. 47) and explained by 
Bacchylides and Theocritos (xxv. 274). This, however, may fairly be regarded 
as a case in which the current artistic type gives the lead to the poets. The 
wrestling scheme was predominant, and was accounted for by the tale of 

Fig. 5. 

XVI. The last sacrifice of Heracles. — A scene of preparation for the 
sacrifice on Mount Kenaion by Heracles, in the presence of Lichas and 
Hyllos (?), is represented on certain fragments at St. Petersburg,^ derived 
from a sort of Monte Testaccio, near Kertch (Fig. 6), Heracles appears to 
have put on the robe (we know from Sophocles that he had time to sacrifice 

1 Cf. the collected list of types, "Walters, 
B. M. Catalogue of Vases, ii., p. 13. 

2 Cf. the vases quoted by Reisch, Athen. 
Milt. 1887, p. 123. 

In the group dedicated by Hippotion of 

Tarentura at Olympia, Heracles used the bow 
Paus. V. 25, 7. 

3 Stei)hani, Compte Rendu, 1869, PI. 4, Fig. 1 
and (more complete) ib. 1876, PI. 5, Fig. 1. 



the first twelve of his hecatomb before the poison began to work) and holds 
out witli both hands the fillet for the adornment of one of the victims. 

Bacchylides introduces a new incident in his account of the events on 
Mount Kenaion, when he includes victims sacrificed in honour of Athene and 
Poseidon. This raises once more the question of the interpretation of certain 
fragments already published in this Jmtrnnl} Heracles assisted by two youths 

Fig. 6. 

Li[cha8] and Philoctetes (?) is sacrificing at a stone altar, before a draped 
xoanon, while Athene herself stands and watches the ceremony. 

1 C. Smith, J.H.S. ix. PI. 1, p- 1 ; again in 
Cat. of Vases in the British Museum, iii., PL 16, 
No E 494. In the later publication the frag- 
ment at the right of J.H.S., PI. 1 is shown 

as part of the Heracles ; the Athene is dis- 
connected from the other fragments ; A I 
(part of Licbas ?) is preserved above the youth. 

T 2 

276 A. H. SMITH. 

The subject has been called, by several interpreters, Heracles sacrificing 
at the altar of Chryse, and by Mr. C. Smith, writing before the discovery of the 
A I , a sacrifice on the Acropolis. 

The presence of Lichas strongly suggests the sacrifice on Mount 
Kenaion, while Philoctetes might be introduced in place of Hyllos, through 
some confusion between the sacrifice on Mount Kenaion, and the subsequent 
self-immolation on Mount Oeta.^ That Bacchylides should include a sacrifice 
to Athene, removes a part of the difficulty of finding her so conspicuous, 
where Zeus was the deity to be honoured. The presence, however, of Philoc- 
tetes at the sacrifice, shows that the fragments cannot be made to agree 
plainly with any known form of the story in literature. 

XVII. Theaeus and the Ring. — The story of the descent of Theseus in 
pursuit of the ring and the wreath, has been closely analysed by Prof Robert, 
in successive papers. 

He has pointed out that the story consists of two elements, namely the 
giving of the wreath by Amphitrite and the story of the ring. Both incidents 
are mentioned by late authorities, by Hyginus and by Pausanias describing the 
picture of Mikon in the Theseion, but we do not know that both were 
represented by Mikon, since Pausanias expressly states that Mikon did not 
tell the whole story, though he does not say what part was omitted. 

Judging from the vases first known (nos. 1-3 below), the incident of the 
ring seemed to be comparatively recent, and its invention was formerly 
attributed by Prof. Robert to Euripides. The Tricase vase (no. 4), if it in fact 
contains a representation of the ring, points to an older source than 
Euripides for the ring incident, and this is now proved by Bacchylides.^ 
The fact, however, that Bacchylides has nothing to say about the recovery of 
the ring seems to show a want of homogeneity in the story. Also, it seems 
to indicate that it is not he who devised the incident. A poet who conceived 
the story would probably make it complete, and would hardly omit the 
conclusion of the finding of the ring on the ground that beside the present 
of Amphitrite it became quite insignificant.^ 

The illustrations that follow are taken from the four red-figured vases 
at present known, which deal with this subject. In two, the principal action is 
between Theseus and Amphitrite, and in the other two it is between Theseus 
and Poseidon. 

(1) Cup of Euphronios/ in the Louvre — a very fine red-figured vase 
found at Caere, of about 500 B.C. (Plate XIV.), 

Theseus, who is supported under his feet by a Triton, greets Amphitrite, 

* See the observations by Mr. Murray, prefixed * We owe the new illustration of the vase to 

to E. "194, in the Cat. of Vases, iii. the kindness of Mr. A. van Branteghem. See 

'^ Robert, Arch. Anzeiger, 1889, p. 141 ; also Mon. Orecs. de I'Ass. d'J^ludes Grecs, 1872, 

Marathcr/ischlacM in derPoikile, -p. 50 ; ffermes, PI. 1 ; Klein, Euphronios, p. 182; Reinach, 

xxxiii., p. 132. Cf. Jebb, Milan jts Henri Weil, PI. 4, etc. The new drawing by M. Devillard 

p. 235, Miss Harrison, Class. Review, 1898, gains greatly in force and effect as compared 

P- 85. with the older, but excellent, engraving, by 

' Robert, Hermes, I.e. p. 140, Jiaving the internal blacks rendered as solid. 

Illustrations to bacchylides. 


m7 "1 



in the presence of Athene. Tliree swimming dolphins mark the sea. The 
wreath is not shown. 

(2) Crater, in the Museo Civico at Bologna ^ — a fifth centur3 red-tigurcd 
vase (Fig. 7). 

In this vase, Theseus, supported by a Triton, clasps in suppliant manner 
the knees of Amphitrite, who holds out the wreath in both hands. Below, 
Poseidon reclines on a couch, like one who is in his own house — on the right 
an Eros is pouring out wine for him — and watches the scene. On the left 
we see the stern of the ship of Theseus, and Helios rising from the waves. 
Here also the ring seems to have no part in the story. 

(3) Vase from Girgenti,^ now in the Biblioth^que Nationale at Paris — 
an early fifth century red-figured crater (Fig. 8). 

Fio. 8. 

Poseidon, enthroned, clasps the hand of the young Theseus who stands 
before him. Poseidon is identified not only by his trident, but also by the 
decorative row of dolphins on his foot-stool. Behind Poseidon stands a 

' Ghiiaidiiii, Musco Italiano di Ant. Class. 
iii., p. 1, PI. ].; Furtwaengler and Robert, Arch. 
Anzciger, 1889, p. 141 ; Reinach, p. 66 ; Robert, 
Nekyia des Polygnot, p. 41 ; Hermes xxxiii., 
p. 135. 

2 Mon. delVInst. i., Pis. .52, 53 ; De Luyiies, 
Descr. de quelqucs Vases peints, Pis. 21, 22 ; 
Welcker, Alte Dcnkhiacler, PI. 25, Reinach, 
p. 61 (reverse ibid. p. 61). 



Nereid, or perhaps Arnphitrite, holding up the wreath. Hero also there is no 
indication of" the ring, though the main action is with Poseidon. 

On the reverse of tlie vase, a seated figure, probably a Nereid, seems to be 
twisting the wreath. She sits between a figure with oinochoe and patera, 
ready to pour a libation, and a figure with hand extended as if she is 

(4) Vase of the middle of the fifth century, tound at Ruvo, and now in 
the possession of the Princess di Tricase ^ (Fig. I)). 

Theseus and Poseidon clasp hands. The other figures are Nereus, a 
figure prepared to pour a libation as in the last example, and a figure with 
the wreath. 

Fig. 9. 

In his left hand Theseus holds what has beeu described as a box or 
shell, and assuming that the draughtsman has correctly understood his 
vase, this may be, as Petersen suggests, a receptacle for the ring. It 
looks, however, in the drawing as if it might be a fold of drapery brought 
over the girdle, and it would be strange if the ingenuity of the vase painter 
could not approach nearer to a representation of the ring, than a case to hold 
it. It is noticeable that in the figure on the right the hand and drapery 
have evidently been wrongly drawn. In any case, however, this is the only 
attempt that the vase painters make to represent the ring incident. 

^ Petersen, Rocmischc Mitthcilungcnix. ,V\.% ; 
Keinach, p. 79. 

The ie(l-fi"ured vase in the Brit. Mus. E 264 

( IFiener Vorlcgehl. 1890-91, PI. 3), interpreted 
as Theseus recognised by his parents, is in many 
respects parallel to the Tricase vase. 



•x ^ 


A scene from the Francois vase ^ is also quoted 
by Mr. Kenyon, at the instance of Mr. van Bran- 
teghem, as having reference to the incident (Fig. 10). 

In this we have the ship of Theseus close to 
the shore to whicli it is drawn up, stern first. One 
nude figure swims ashore, while the occupants of 
the ship express emotion and surprise in various 
ways. Theseus, as a citjjarist, leads in set array 
the seven youths and seven maids, who walk altern- 
ately, hand in hand, led by Epiboia or Eriboia. 

It is at first sight an attractive suggestion that 
the swimming figure is Theseus, but the objections 
adduced by Prof. Robert seem conclusive. The action 
takes place close to the shore, while Bacchylides, 
Pausanias and the Bologna vase represent the action 
as taking place at sea. Also the whole band pro- 
bably represents a single incident, and the fcstiil 
procession is most appropriate to the subsequent 
landing at Delos. . 

XIII, Theseus. — The Theseus cycle is already 
well represented in this Jmirnal by several vases, 
to whicli it is only necessary to give a reference. 

(1) Kylix in the British Museum, No. E. 84 ^ 
(J.H.S. ii. pi, 10, p, 57, for the interior. The same 
scenes are repeated on the outside of the vase.). 

(2) Kylix, formerly in the collection of Mr. 
Tricoupi {J.H.S. x. pi. 1, p, 231.). 

(3) Fragment of a kylix from the De Luynes 
collection in the Bibliotheque Nationale {J.H.S. x. 
pi. 2, p. 234.) 

(4) Kylix at Vienna, with Theseus and Skiron 
{J.H.S. ix. p. 272.). 

For further lists of Theseus vases, see Milani, 
in Museo Italiano di Aniichitd Classica, iii. p. 201), 
pis. 2-4. 

A. H. Smith. 

* From Afon. dclV Ind. iv., 1*1. 56, supplemented with notes 
BUpplied by Mr. Cecil Toir. 

* Reproduced by Reiuacb, p. 45. 

J.H.8. XVIII. (1898). PL XV. 

±8. XVIII, (1898). PL ) 


J.H.8. XVIII (1898). PL. XVI. 

IVVtosV A^..- ^.- •'^i^ tSSk 


J.H.8. XVIll. (1898). PL. XVII. 




[Plates XV.— XVII.] 

Since the publicaiioii of tlic official catalogue (Volume II.) in 1893 the 
British Museum has been enriched by several black-figured vases of consider- 
able interest and importance, which I propose to describe and discuss in this 
paper. Excluding the Odysseus and Kirke vase which I published in vol. xiii 
(tf this Journal, the total number amounts to eight, one of which bears an 
artist's signature, while another is a unique example of a very interesting class. 
Three others again are interesting from a typological point of view. I will 
take the vases in a roughly chronological order. 


Corinthian oinochoe, 8 in. high, from Aegina (Fig. 1). It luis a trefoil 
mouth and squat neck round which is a moulded ring. The handle does not 
rise above the mouth of the vase, and is quite plain, with cylindrical section. 
The vase is in good condition, except that the foot is somewhat chipped, and 
the black varnish is dull and frayed on the lip. It has been imperfectly fired, 
and the varnish has turned to red in some places. The clay ground is of a 
buff colour, and the clay itself appears to be rather gritty in texture. 

The design presents no very remarkable features. In the centre stands 
a Siren to the right, of a type frequently occurring on Corinthian vases. Her 
liair is long, and falls in masses on the neck, and on the top of her head is 
what appears to be a small fillet. The wings are recurved, and are spread out 
on cither side of the body, the left one being advanced at an impossible angle, 
in that false perspective of archaic art which arises from the desire to render 
as much as possible of an object visible at once. 

On either side of the Siren is a panther turned in her direction, but 
with face towards the spectator; their tails are curled over their backs. The 
field is decorated with ten rosettes of the usual type, but instead of being 
scattered about promiscuously they are arranged symmetrically along the 
upper and lower edges of the design. The rest of the vase is varnished except 
for a narrow band of clay left visible towards the bottom of the body. 

The style of the decoration is rather like that on a lebes from Naukratia 
in the Museum (B 101), which appears to have Corinthian affinities. 




Aiiiplioia (Plate XV. and Fig. 2) of tlie class formerly known as 
Tyrrhenian, and now usually called Corintho- Attic, but which may perhaps 
be more conveniently styled Peloponnesian.^ It was probably found in Italy. 
The vase stands lo in, in height, and presents in its shape the usual features 
of this class, a slim neck separated from the shoulder by a plastic ring, plain 
handles, and slim egg-shaped body not marked off from the shoulder. Tiie 
mouth, handles, and foot are covered with black varnish, which, however, 
is much frayed ; two rings of purple have been painted round the foot with a 
brush while the vase was on the Avheel. The decoration of the body is 
arranged in three friezes, the upper one being considerably wider than the 

Fig. 1. ^Corinthian Oinochoe. 

other two ; on the latter are bands of animals. On either side of the 
neck is an ornament consisting of a combined lotos-flower and palmetto, from 
which extend tendrils, those on the one side meeting those on the other 
except where they are interrupted by the handles. Tlje lotos-flowers have 
three petals, as always on Attic vases, opposed to the Corinthian and Chal- 
cidian vases, on which they have only two. Round the shoulder is a roughly- 
painted 'tongue-pattern' in alternate purple and black, and round the foot 

See Locsclicke in Arch. Zcit. 1876, p. 108. 



tajRiiiiig rays slioot up. Tlic- louL lias bi-eii ic pa net I and apparently rupaiuted, 
but otherwise the vase is in good condition. 

The ' Peloponnesian ' vases liavo been collected and (liscussed by 
Holwcrda in the Jahrhu eh d. Arch. Ins^f. ISOO, p. ;i37ft". ; an interesting one 
has since been added to the list by Hauscr (o;?. cit., LSDIi, p. U3). Holwenla 
gives a list of fifty, of which forty-six are amphorae; two are in the British 
Mnseuni, (B 47 and B 4.S), and twenty-six in the Louvre. of their 
typical features are illustrated by the vase here dcscribcil, ;is i.s pointed out 
incidentally in the following discussion of it. Their style is for the most 

Fig, 2.— Peloponnesian Amphoka : Reverse. 

l)art coarse and clumsy, but often rises to a higher standard of merit. The 
lines are often mechanically drawn and lifeless, a result of the slavish imita- 
tion of Corinthian prototypes. Details of drapery are seldom shown ; although 
the dresses are often richly decorated, yet the folds are never indicated. The 
date of this series is probably not later than the middle of the sixth 
century e.c. 

The inscriptions are a very interesting and important feature of the class. 
Like those on our vase they are generally in the Attic alphabet, but from 
time to time a Corinthian or Chalcidian letter occurs,^ which may be dis- 

^ CAj, the O ami ^ on Beilin Cat. 1704, ;uid tlic Chalcidian < on liiit. Mus. B 47 (sec 
Jukrb. d. Arch. lust. 1890, p. 243, No. 34). 

284 H. B. WxVLTERS. 

tinguished even in the meaningless imitations of inscriptions which occur on 
Ko many examples. Tliis is of course another result of copying; and a 
curious development is the occurence of two fornis of the same letter side by 
side as in the 9k and ^E of the Berlin vase (17U4). As regards the meaning- 
less inscriptions, the same combinations of letters have been noted as occurring 
on different vases, and the collocations seem at times to be due to something 
else tlian mere chance. 

The ordinary scheme of decoration on these vases is as follows : On the 
obverse, a mythological subject, on a wide band extending from the neck 
half-way down the body ; this is balanced at the back by a genre-scene of 
some kind, combats, riders, or dancers, or even animals. The painter 
appears to have devoted all his energies to the mythological scene, and for 
the other to have employed only stock types from his repertory. Sometimes 
these seem to be merely decorative. Or again, we get a single figure taken 
from a large composition (see Loeschcke in Arch. Zeit., 1870, p. lOS ff). 
Below these arc almost invariably two friezes of animals. 

The range of mythological subjects is not very extensive, anil all the 
subjects are characteristic of early b. f, vases. Commonest are: Nessos 
carrying off Deianeira ; Combat of Herakles with Amazons ; Birth of Athena ; 
and Calydonian boar-hunt. Other subjects which only occur on isolated 
examples are: Herakles and the Hydra; Perseus and the Gorgons ; Theseus 
and the Minotaur; the Niobidae ; Prometheus; Combat over the body of 
Troilos. Dionysos appears once, accompanied by Satyrs and Maenads ; but 
Bacchic scenes are always rare on early b. f. vases. The chief subject of 
the vase now under discussion occurs also on the specimen published by 
Hausor in the Jahrhcch {loc. cit). 

Our vase is no exception to the general rule ibr scheme of decoration. 
The chief subject, as is plain at a glance, represents the Sacrifice of Polyxena. 
This we will now proceed to describe in detail. In the centre of the scene, 
on the level of the ground, is seen a mound-shaped object, with a Hat top. It 
is not easy to say whether this is intended for the tomb of Achilles or an 
altar; probably the former, although the usual type of tomb on black-figured 
viises is in the shape of a conical tumulus {c.y. B 239, B 543 in British 
Museum). It is decorated with a diaper pattern of alternate plain purple 
squares and black squares on which crosses are incised with white dots between 
the arms. On or behind the tomb is a sort of stand or table, on which a 
fire burns ; this of course may be intended for an altar, hke that on the 
B.M. vase, B 80 ; but for a mound-shaped altar, compare the /8cu/u.09 on 
another vase of this class, similarly decorated (Munich 124 = Gerhard, 
A. V. 223). 

On the left of the altar, Neoptolemos strides forward to deal the decisive 
blow to Polyxena. In his right hand he holds the sword, which he plunges 
into her neck ; a purple stream of blood gushes out from the wound. His 
left hand he places on Polyxena's head with a view to steadying himself. 
He is armed in the usual Greek fashion, but one or two small details of his 
costume call for notice. On the cheek-pieces of his helmet and on his 


groaves arc borders of small wliitc dots, which appear to be intended for the 
stitch-holes generally seen in bronze helmets or greaves.^ His short, close- 
fitting chiton is painted white and is apparently of fine crinkled linen; his 
cuirass is coloured ])urple. He is appanMitly intended to be bearded, but the 
hairs are not indicated. 

The body of Polyxena is carried up by three men, and is held in a 
horizontal position, quite straight and rigid, the chest downwards, and the 
head slightly raised towards Neoptolemos. One is at once remii\ded of the 
type of Odysseus and his companions boring out the eye of Polyphemos; the 
pole is carried in very much the same fashion by the three men. That type 
does not occur on any vases of this, but as it is found as early as the 
Aristonofos vase,- the composition under notice may well be a reminiscence 
of it. It is a type that belongs almost exclusively to the early black figure 
period, and must have been well established by this time. 

The flesh of Polyxena is painted white where it is visible, with a thin 
incised line on the neck to indicate a necklace; her hair curls over the fore- 
heaifand fails in a long thick wavy mass down the back. She is dressed in a 
long chiton which is adorned with incised crosses and purple spots ; the dress 
lies stiff and devoid of folds. We are reminded of the line in the Hecuha 
(5G9) : iroWrjv irpovoiav el-)(ev eva-y^tjfiox; ireaelv. Her name is inscribed 
above her : 9l/l3+^V>IOn. The alphabet, as in the case of the other inscrip- 
tions, is purely Attic ; it may be noted that the -|-^ arc treated as one letter, 
and therefore not written retrograde as in the name of Phoenix below. 
Her three bearers are Amphilochos, Antiphates, and Ajax son of Oileus, 
who hold respectively the upper part of her body round the breast, her 
waist and thighs, and her feet. They also have their names inscribed : 
^0+0>ll*v^A, ^3TAc}>ITMA, AIA^ IHAAE[^^ The three warriors are 
all armed with swords, and wear helmet, cuirass and greaves; these are 
ornamented with patterns in purple and white. 

The names of Amphilochos and Antiphates call for some remark not 
merely on account of their epigraphical form. Neither name is to be found 
in the Iliad, but Antipha,tes is mentioned by Tryphiodoros, 180, and Tzetzes 
{Post-Horn. 648) as one of the warriors who were inside the Wooden Horse, 
and Amphilochos the son of Amphiaraos by Quintus Smyrnaeus (xii. 322) 
in a similar list.* One is tempted to see in the name ' Av<f>cXoxo<i an error 
for 'AvTi\oxo<i, and the son of Nestor (Quint. Smyrn. ii. 244) would be 
appropriate here in the company of his father (v. infra). In the description 
by Quintus Smyrnaeus of the sacrifi_ce of Polyxena (xiv. 257 ff.) no mention 
is made of any of these heroes, nor do their names occur elsewhere on vases ; 
.so that it is difficult to account for their selection here. 

' E.g. B. M. Cat. of Bronzes, No3. 74, 249, * In QuiiU. Smyrn. xiv. 3C6, Auj])hilochos 

2821, 2828, &c. remains with Caldms at Troy after the do- 

"^ Moil. (lelVlnst. ix. 4; cf. also ihid, i. 7, parturc of the Oiecks. lie was a seer, iind wouM 

llg. 2 ; Brit. Mus. B 154 ; Berlin 2123. therefore he appropriately [irescut at a saeri- 

* The forms 'Av<p{\oxos, 'AfirKpirrif seem to lice, as Calchas on the Tahnla Iliacu. There is 

suggest a reciprocal confusion between the /n of another Antiphates mentioned in Od. xv. 242 

auipl and the v of avri ; but see infra. as grandfather of Amphiaraos. 

286 H. B. WALTERS. 

The title A7a«? ']XidB7j<i (for 'OiXidSrj^;) is of course unusual. As a rule 
where the name of Ajax Oiliades or Telamonios occurs on vasos, the patronymic 
is omitted, but it is a common feature of this class of vases to add 
surnames, (.^. 'E/3/i^9 Kv\X7]vio<i ; and so we have Nea-royp IluXto? below. 
As regaids the form, 'IXeiOva (on B 147 and E 410 in Brit. Mus. and Mon. 
dell' Inst, vi.-vii. 5G) and 'EiXeiOva on the Berlin vase No. 1704, which belongs 
to our group, are not really analogous, as there the vowels | and E represent a 
diphthong. The form '\Xidhr}<i does not appear to be quite unknown in 
literature. In II. xiii. 203 the scholiast Zenodotos reads it for the 
vulg." 'O'CXidhr)^, which form also occurs in Quint. Smyrn. vi. 556 and 

To resume our description of the scene, three figures still remain for 
discussion. Phcenix moves away on the extreme right, with clenched right 
hand and spear in the other; his name is written ^ + |V\I04^. Beyond him 
is a folding stool. On the left stand Diomede (^EAEMOIA) and Nestor 
(NE^TOP [""VHO^), the former holding two spears and shield. Nestor is 
not represented as of any great age, but has black hair and beard. He wears 
the long white linen chiton of the charioteer, and holds a spear ; his figure 
exactly balances that of Phoenix. 

The death of Polyxena is not a common subject in Greek art. It occurs 
in several ' Iliupersis ' scenes, but in those cases we have merely the figures of 
Neoptolemos and Polyxena, and sometimes there is nothing to identify them, 
or the type is approximated to that of Ajax seizing Cassandra, as on the Brit. 
Mus. vase F 278. Pausanias alludes to it among the paintings in the Pina- 
kotheke of the Propylaea (i. 22, 6) : rov he 'A^iXXe<o<; Td(f>ov rrXrjaiov fieX- 
Xovad ian (r<f>d^€(T6ai, HoXv^ivr]} He tells us (x, 25, 10) that he had seen 
another painting of the subject at Pergamos on the Ka'ikos. Overbeck ^ gives 
a list of vases, gems, and Etruscan urns, on which the subject is to be found * 
to these may be added : 

(1) ' Peloponnesian ' vase, Jahrhuch d. Arch. Inst. viii. (1893), PI. I, 

(2) Robert, Homerische Becker, p. 73, with figures of N. and P., Odysseus, 
Agamemnon, and three unnamed heroes; subject taken direct. from Eur. 
Eec. 558 ff. 

' See Hesyehius ; Find. 01. ix. 167 and known Towneley cista in the Brit. Mus. (Cat. 

Schol. ; and Pape, Gr. Eigennamc n^ wwAav cSiCh 743) can only be regarded as a doubtful one. 

foim respectively. There is no certain indication that the figure of 

- Overbeck (Arch. Misc. 1887, p. 10) has the victim is feminine, and the proportions 

suggested the jiossible identity of this painting Mould suit equally . Avell for a boy, while the 

with one by Po!ygnotos (or, according to another whole design is somewhat indistinct. On the 

reading, Polykleitos) described in an epigram other hand, the connection of the scene with 

(Anth. Plan. iv. 150). the death of Neoptolemos on the other side of 

' Her. Bilduxrke, p. 661 ff. the cista is an argument in favour of the re- 

* It may be noted that the most important ceived interpretation, 
of the examples given by Overbeck, the^well- 


As part of an Iliupersis scene the subject occurs on tlie followinpf 
examples : 

(.S) Brit. Mus. : F 160. 

(4) Naples : 2422 (the Vivcnzio vase). 

(')) Lo\ivre : Kylix by Brygos (Heyrlemann, liinpersia, PI. I.). 

It will be interesting to compare with our vase No. (1) in the above list, 
an amphora in the Bourguignon collection at Naples, published by Hauser 
{he. cit.) 

The types are startlingly different. The painter of the Bourguignon 
amphora appears to have adopted another 'Iliupersis' type, that of Priam's 
death on the altar of Zeus, for the figure of Polyxena.^ The moment 
represented is not quite the same ; Neoptolemos has already accomplished the 
deed, and is hastily mounting his chariot (this again is a borrowed motive, 
from the Amphiaraos vase in Berlin). None of the figures are inscribed ; 
Polyxena has fallen, not on an altar, but on a tumulus which represents the 
burial-place of Achilles. 

The subject on the reverse of our amphora will not compare with the 
obverse in interest, but yet calls for some attention. It represents a revel of 
four men dancing in somewhat grotesque attitudes, flanked on either side by 
a cock. This is a common type on Corinthian vases and on more than one 
class of imitations of Corinthian wares, where they supply the place of 
the Satyrs on Ionic and Athenian fabrics. A list of instances is collected 
by Korte in Jahrhiich d. Arch. Inst. viii. (1893), p. 90, note 58 ; to which we 
may add : 

(1) Berlin Cat. 1662 (Corinthian). 

(2) Brit. Mus. B 42 ^ (subject : return of Hephaistos). 

(3) Brit. Mus. B 44 (imitation Corinthian). 

(4) Furtwaengler, Coll. Sahouroff, i. PI. 48. 

(5) 'E<^. 'Xpx. 1885, PI. 7. 

(6) Amphora at the Hague (Jahrhuch, v. (1890), p. 244), of ' Pelopon- 
nesian ' fabric. 

(7) Vase in Athens, published by Loeschcke, Athen. Mittheil, xix. (1894), 
PI. 8, p. 519 ff. (subject : return of Hephaistos). 

The meaning of these figures has been subject to some discussion. 
Furtwaengler, von Rohden, and formerly Loeschcke,' have regarded them as 

' Hauser ad loc. \k 99. Duemmler, who pronounced the vase to be 

2 In reference to tliis vase, which Loeschcke Sicyonian, and classed it, on stylistic grounds, 

(Ath. Mitth. 1894, p. 516) claims to be 'echt with the Berlin vase No. 1147 which bears a 

Korinthisch,' and which in the catalogue I have Sicyonian in.scription. 

included among ' Imitations of Corinthian fab- ' Ann. dell. Inst. 1877, p. 450 and 1878, p. 

rics,' it may perhaps be worthwhile to mention 301 ; Baumeister, Dcnkmdler, iii. p. 1962. 

that my view accords with that of the late F. 

288 H. P.. WALTERS. 

terrestrial votaries of Dionysos. Dueminler ^ first raised the question whether 
they should not be regarded as Sat'/xoz^e? of some kind, a question which now 
seems to be solved by their appearance on Nos. (2) and (7) accompanying 
Hephaistos and Dionysos. Locschcke, in publishing the latter vase, lias 
definitely pronounced for this view. He points out that their correct title is 
XuTvpoi, and that they represent the Satyrs as they appeared in the oldest 
Satyric dramas of the Peloponnese. 

The attitude in which these Xdrvpoi, are usually depicted, with one leg 
raised and pointing outwards, appears to illustrate the word piKvovadai,"^ 
interpreted by Pollux (iv. 99) as to rr)v 6<T<f)vv (popriKox; Trepidyeiv, and by 
Sophocles (apud Phot.), to Kap,7rvXoit ylyueaOai d<T')(i]p,6p(0<i, zeal kuto, 
a-vvovaiav koX {kut) op'x^'qaiv, Kdp.'jrrovra rr)v 6<T(f)vv. 

The remaining decoration of the vase, as already indicated, consists of 
two friezes of animals arranged in heraldic fashion : (1) on the upper row, a 
pair of Sirens confronted, with a pattern of two lotus-flowers and two palrn- 
ettes between ; two groups of a panther and a ram divided by a swan ; (2) on 
the lower row, two groups of a ram between panthers. 


Kantharos, 10| in. in height, 7| in. in diameter, with sharply -pointed 
handles bent round to form an inverted semi-circle with the lower edge of 
the body (Fig. 3). The vase is said to have been found at Athens, near the 
Pnyx, and was originally found in fragments ; these were put together, not 
without a considerable amount of restoration, which has now been removed. 
The lavish use of purple and white for details gives a pleasing appearance 
to the vase, while its shape is not ungraceful. The white pigment has faded 
to a considerable extent, and turned to a bluish colour on the black varnish. 
The ornaments consist of a tongue-pattern round the rim, and a row of small 
black dots along the bottom of the design, and round the upper part of the 
.stem is a chequer pattern in squares. Underneath the cup, above the stem, 
is an interlacing lotos-and-honeysuckle pattern, resembling that on the 
neck of the Polyxena vase just described, a small detail which indi- 
cates that the vase, although by an Athenian painter, is by one who has 
not yet freed himself from Corinthian influences. We shall see that the 
choice of subjects and their treatment also recall us to Corinthian archetypes. 
The handles, and all the foot, except the lower edge, are covered with black 

The kantharos is not a common shape before the fifth century B.C. Its 
form being essentially suited to metal, it was never popular in pottery at any 
time, in spite of the fact that it is one of the most beautiful shapes con- 
ceivable ; but black-figured examples may be counted on the fingers. One 
other exists in the British Museum (B 370), and three in Berlin (1737, 4012, 
4013); of the latter, No. 1737 is early Attic work (with inscriptions), and 

» jtnn, dell. Inst. IS.'i.''., p. 129 (and see pi. D). "■' Soph. Frag. 297 ; Lucian, Lf.riphanes, 8. 



is grouped by Klein, Enphronioa, p. 73, witli B 147 in Brit. Mus., and other 
vases of the school of Klitias and Ergotimos. On the other hand, the 
frequent occurrence of the kantharos as an attribute of Dionysos or as a 
device on shields ^ implies that the form was (juite familiar at this period. 

On either side of the cup is a subject, the meaning of which is clesu: 
enough, but there is room for doubt as to whether or no it is of mythological 
import. The one side represents the departure of a warrior in his chariot, 
the other, a combat over a ftillen warrior. One is tempted to see a connection 
between these two scenes, and if they are mythological, we must look for 
some known type to which they correspond. Now, as regards the first, the 
well-known Corinthian krater in Berlin (No. 1665), with the departure of 
Amphiaraos, gives a well-defined type ; as regards the second, in order to 

FiQ. 3.— Athenian Kantharos. 

arrive at a connection with the other side, we are at once reminded of the 
combat over the body of Polyneikes. Thus the two sides of the vase may 
represent two episodes from the story of the Seven against Thebes. This 
interpretation must, of course, be received with caution ; its correctness 
cannot be proved, but it is at least a permissible suggestion.^ 

' On vases in Brit. Mus. as shield-device : B 
267; in Berlin: 1790, 1865; as attribute of 
Dionysos : B 149, 153, 178, 179, 180, 195, 198, 
&c. in Brit. Mus. 

'^ See for these two representations in ancient 

art, Overbeck, Her. Bildw. pp. 91-135. He 

gives several inscribed vases with the departure 

of Amphiaraos, and other conjectural instances, 


in regard to which he points out that if they 
have any mythological meaning it is probably 
this. But he denies the existence of any repre- 
sentation of the ' Briiderkampf ' earlier than 
the Etniscan sepulchral urns. He omits, how- 
ever, the inscribed mirror in the Brit. Mus. {Cat. 
No. 621} which is not later than the Srd cent. 
B.C. and probably earlier. 


290 H. B. WALTERS. 

We proceed to describe the two scenes more in detail. 

(1) The departure of Amphiaraos (Plate XVI.). The main portion of the 
scene is occupied by a fonr-horse chariot standing to the right. The 
charioteer (Baton) is of the usual t}^e, clad in a long chiton of crinkled white 
linen, and holding a goad in one hand and reins in the other. His hair and 
beard are painted purple. The hero, or Amphiaraos if he may be so termed, 
mounts the chariot from the further side. Of his attire all that is visible is 
a short chiton, with ornamental border. On the near side stand three figures, 
a woman and two warriors. The first of these holds up a helmet in his 
hand for Amphiaraos to put on, the other holds up a bow, and carries a 
shield, with device of a snake ; but there is no authority from which we may 
derive names for them. Their position on the near side of the chariot is an 
unusual one ; in scenes of this kind the ' Nebenfiguren' generally stand in a 
line on the further side of the chariot, as in such instances as Gerhard's 
Attserlesene Vasenbilder, Pis. 136-140, 249, 252. The horses all have top- 
knots ; at their head stands a bald, bearded old man, wearing a chiton and 
himation, both embroidered, but the patterns are faded away. Behind him 
is a woman drawing forward her himation like a veil over her head ; this 
attitude is familiar as indicating a bride in the presence of her husband ; it 
is possible that she may be intended for Eriphyle, but there is no necklace to 
characterise her, as on the Berlin vase. The scene is closed by a bearded 
man wearing a chiton and himation, whose figure has almost entirely dis- 
appeared, and had been much restored. 

On the left side of the scene are three figures balancing the three on the 
right : a woman wearing an elaborately decorated chiton and a purple and 
white wreath ; a warrior fully armed, whose shield bears a star of seven 
points as device ; and a bearded man who appears to be in conversation with 
the warrior. 

On the whole there is so little characterisation of the figures, and so little 
to differentiate this from other compositions of the kind that I am inclined to 
regard it as merely a ' departure of a warrior ' scene, in spite of the possibility 
of a connection with that on the reverse. 

The reverse (Plate XVII. Fig. 1) is in worse condition than the obverse ; 
it represents, as has been said, a combat over two fallen warriors, which, 
like the other scene, is a very familiar type on black-figured vases. There 
is nothing to indicate that either of the warriors is slain ; they merely 
appear to be temporarily rendered hors dc conibat. One leans on his right 
elbow ; the other is fallen on his knees. The latter carries a Boeotian shield ; 
of his person the legs alone now remain. Over them stand a pair of warriors 
fighting, with spears, of whom the right-hand one is now lost, and on the 
right were no doubt two other pairs of combatants, but this part had been 
entirely restored. On the left is another pair fighting with spears, but the 
left-hand warrior runs away while he turns back to thrust at his pursuer ; 
his shield-device is a tripod. If the interpretation suggested above is to be 
accepted, the central figure would of course be Eteokles, and the fallen one 


Throughout this scene the work has been very careful, with minutely- 
incised lines and a lavish use of purple and white pigments ; all the armour 
is treated in elaborate detail ; but the effect has been much marred by the 

The exact position of this vase and its relation to the masters of Athenian 
black-figure vase-painting is not easy to determine. We have seen here and 
there a suggestion of Corinthian sources, but the general tone and style is 
Athenian. Though free from the mannerisms and affectations of the Kiitias 
and Ergotimos period (represented by the Francois vase and the Brit. Mus. 
vase B 147, with the birth of Athena), it yet comes near to this class ; while 
in the crowding of figures and elaboration of detail we see an anticipation of 
Glaukytes and Nikosthenes. Nevertheless it would be rash to assign this 
work to any one painter or school ; we can only say that it must be dated 
about the middle of the sixth century B.C. 


Kylix of the ' Kleinmeister ' type, from Aegina, 7^ in. in height and lOf 
in. in diameter. The style is that of Glaukytes or Nikosthenes, and in general 
appearance the vase is very similar to one signed by Glaukytes and Archikles, 
published in the Wiener Vorkgebldtter for 1889, PI. 2, fig. 26. We may also 
compare the kylix by the former artist now in the Brit. Mus. (B 400). 

The designs are painted on a red band round the cup, which is of the 
shape characteristic of the period (see B.M. Cat. of Vases, ii. PI. 5, fig. 16). 
The rest of the vase is covered with black varnish, with the exception of a 
circle of 3^ in. diam. in the centre of the inside, a red band round the bottom 
of the bowl, and the foot. 

The subject on either side is the same, with little variation : the prepara- 
tion and departure of warriors for battle. On the obverse (Fig. 4) are four 
warriors : one mounts his chariot accompanied by a small groom ; the next is 
moving away on foot ; the third is putting on his greave (a familiar b.f. vase 
motive) ; while on the extreme left is a warrior donning his helmet and 
accompanied by his dog. This last motive appears to be a new one ; it does 
not occur on any other b.f vase in the Brit. Mus., or in the Berlin collection. 
Usually the warrior's helmet lies on the ground, or else it is handed to him by 
a woman. Interspersed with these figures are twelve draped ' Nebenfig^ren,' 
one of whom is a woman. 

The reverse is almost identical, but the warrior putting on his helmet is 
omitted, and one of the bystanders, making a total of fifteen figures in all, 
against seventeen on the obverse. 

In the field on either side are various collocations of letters representing 
inscriptions, but all quite meaningless. The cup is much broken, and has 
been restored in places; the colours are faded, and the work is rather 

u 2 

292 H. B. WALTERS. 

Fragments of a kyathos, about half remaining (Plate XVII. Fig. 2). 
Shape as B.M. Cat. of Vases, ii. PI. 7, fig. 1 ; height 2J in. These fragments 
probably come from Italy, but no information has been received. They 
have a special interest inasmuch as they bear the signature of the potter 
Nikosthenes, thus adding another to the already long list of his known works. 
Klein ^ gives two vases of this shape as signed by Nikosthenes, but neither 
has been published, nor are their present possessors known. Judging from 
his description the first exactly resembled our present example : ' Kelle mit 
einem Henkel .... Tanz von fiinf Silenen und vier Manaden. Dariiber 
die Inschrift.' 

The handle is covered with black varnish ; the cup itself is covered with 
glaze of a rich bufif-red colour, and both purple and white pigments are 
employed. The decoration consists of Satyrs and Maenads dancing in pairs, 
the latter wearing long girt chitons with white spots. One wears a j^^itcov 
<rxi,<TT6<i, another kicks up her leg behind. 


Amphora from Aegina, 11 \ in. high, with designs in panels on either side 
(Figs. 6, 6). This vase appears to be of comparatively late date, as is shown by 
(1) the shape, which is characteristic of amphorae of the Andokides and Euthy- 
mides schools, (2) the almost complete absence of purple and white pigments, 
(3) the borders of ornamentation surrounding the panels, which are also seen 
on early r. f amphorae, and on hydriae of both methods. The handles are 
quite plain ; there is a rather sharply- marked division between neck and 
shoulder, but it is curvilinear, not angular, as in the red-bodied amphorae. 
The black varnish is rather worn, but on the whole the vase is in good 
condition. The ornaments consist of inverted lotos-buds above the panels, 
rows of dots down the sides of the panels, and rays shooting up from the 

Both the subjects on the panels are of considerable interest : the obverse 
represents Herakles bringing Kerberos out of Hades, the reverse, two heroes 
playing at draughts in the presence of Athene, both being familiar subjects 
on b. f. vases. The former was not previously represented on any vase in the 
Brit, Mus., but the latter occurs fairly often (see Vase Cat. ii. p. 27 and 
also E 10). 

On the obverse of our vase Herakles, with club in hand, short chiton, 
and lion's skin tied over his head, hauls the two-headed Kerberos along with 
a chain. He moves towards the left, and looks back at his prize, the two 
necks of which are adorned with purple collars. A richly-curling mane runs 

' Meiatera.^ p. 66, Nos. 54, 65. 

294 H. B. WALTERS. 

down the back of the monster. On its further side is seen Hermes, 
with face to right, but feet to left. His hair is looped up behind in 
the KptD^vXo'i} and he is attired in the usual style, with petasos, chlamys, 
and high boots with large tags in front. The petasos is painted white 
with a purple brim. The locality is here indicated by a Doric column 
with white capital, which is artistic short-hand for the palace of Hades, and 
within the palace is Persephone, holding a sceptre capped with a pome- 
granate ; she stands away from the scene, but looks back at Kerberos. She 
wears a chiton and himation. 

On the reverse is a plain block of stone ^ in the centre, before which 
stands a statue of Athena with face to left, the left hand up-raised, and a 

Fia. 5.— Amphora: Heroes Playing at Deauqhts. 

spear couched in the right. This position of the goddess is new to the type ; 
she is usually placed on the farther side of the block. The two heroes are 
draped, not armed, but each has two spears in the left hand, and the one on 
the right hafi a sword. Behind each is a shield, the one on the left being 
inscribed 0X0, while the other has a tripod as device; and above each shield 
is a helmet. The space above is filled in with branches. The warrior on the 
right holds an object between his fingers, presumably a rrea-ao'i ; the same 

^ Studniczka in Jahrh. d. Arch. hist. (1896), has been pointed out by Blinkenberg {Ath. 

P- 248 ff. Mittheil. xxiii. (1898), p. 9), that it really re- 

2 I have refrained from styling this object an presents the table on which a board was marked 

' altar,' as it has been hitherto called, since it out for playing the game of ^irl weVrt ypafM/xiy, 



action occurs on B 193 in Brit. Mus., but no pessi are here visible on the 
table between them as is usually the case.^ 

The subject of Herakles and Kerberos occurs on no fewer than 36 
black-figured vases, as against 6 red-figured, and 7 of later date. Several 
lists of them have been made, the first by Gerhard in his Auserlesene Vascn- 
hilder, ii. p. 157: he gives there 13 examples; supplemented by Conze in 
Ann. dell' Inst. 1859, p. 398, who adds 11 more. Dr. F. J. Schneider in his 
Zwolf Kdmpfe des IleraJcles, p. 45, gives a more complete but somewhat 
inaccurate list, which has been partly corrected by Hartwig in Jahrhuch vii. 
(1893), p. 158. As Hartwig does not give a revised list in full, I have thought 
it advisable to do so here, pointing out where possible the variations in the 
type. Several of the vases in Gerhard's list are only vaguely and superficially 
described, and even now it is possible that there may be one or more 
duplicates ; but as far as possible every item has been identified. 










A. With Euiystheus in pithos ; no palace of Hades ; K. has three heads. 
Cervetri Louvre 679 Mon. delV Inst. vi. 36 

Cervetri Caatellani — Bull. delV Inat. 1869, p. 249 

I. 23 











B. No palace of Hades 









St. Petersburg 





or Eurystheus ; K. has two heads 







Inghirami, Vaai Filt. ii. 

Gerhard, Auaerl. Vaaeiib. ii. 

97 3 
Bull. delV Inat. 1878, p. 178 
Gerhard, Auaerl. Vcuenb. ii. 

Bull. deW Inst. 1865, p. 145 

Ann. dell' Inst. 1877, p. 125 
Inghirami, Vasi Fitt. i. 40 
Gerhard, Auserl. Vascnb. ii. 

[K. has three h§ads] 
Bull. delV Inat. 1847, p. 28 

1. 1 



























^ Another interesting variant of the ' Brett- 
spiel ' type is given by a Cypriote-Attic oinochoe 
from Poli in the Museum at Nicosia (Myres, 

Cyprus Mus. Cat. p. 84, No. 1603); behind 
each player is an attendant warrior. 








1 ULLIXTlciN. 



rt'Br.K ATiiiNs. 










IJ. Rochet tc, Mov. In6l 









Hull. delV Inst. 1839, p 









Bull. dclV Inst. 1869, p. 






271 (Collign- 









209 (CoUign- 





C. With palace of Hadea 





Brit, Mu3. 

J.H.S. xviii. p. 295 






Bull, dell' Inst. 1869, p. 



II. D 





SA 267 

Jahrbtich d. Arch. Inst. 
(1893), p. 157 








Musco Oregoriano, ii. pi 

Cajiipana Cat. iv. 507 














[K. has two dogs' and 





snake's head] 














Gerhard, J. V. i. 40 








Bull, dell' Inst. 1840, p. 








Gerhard, A. V. ii. 131 








Arch. Zeit. 1859, pi. 125 



[K. has one heaa.] 

A. Without Eurystheus or palace of Hades. 




Chiusi ? 


Naples (Bour- 


Jahrbuch, 1893, p. 162 
Jahrbuch, 1893, pl. 2, p. 165 
Jahrbwh, 1893, p. 160 


B. With palace of Hades. 





Durand Coll. 311 

Jahrbuch, 1893, p. 163 
[K. has one head] 


III. APULIAN VASES (Representations of Underworld). 
[K. has three heads.] 








Brit. Mus. 



Vorl. E. 6, 1 



„ E. 2 




„ E. 6, 5 




„ E. 3, 2 


Munich . 


,. E. 1 




„ E. 3, 1 

49 1 Kelebe 

|Tenea(?) I Berlin | 2882 I Coll. Sabouroff, i. 74 

26 & M 


21 & N 

298 H. B. WALTERS. 


Lekythos from Greece, G J in. high, with black figures on white slip ground 
(Fig. 7). The style is late and careless ; there appear to have been purple 
accessories, but they are faded away. The shoulder of the vase is left in red, 
and palraettes are painted on it in black ; round the top of the design runs 
a band of maeander. The black varnish is very poor in quality ; the handle 
has been repaired. 

The subject represented is one usually associated with the earlier b.f. 
period : the combat of Herakles and Geryon. Our example presents several 
features in which there is a marked deviation from the ordinary type, as will 
be seen from the description. Herakles kneels on his left knee to the left 
and discharges an arrow from his bow at the monster, which is separated from 
him by a rock, on which grows a tree. On this rock Herakles has laid his 
lion's skin or some piece of drapery. The figure of Geryon is represented 
in the traditional manner (T/3€t9 dvBpe<; dXk'q\oi<i Trpoa-exofJ'evoi). One body 
(apparently the middle one) has fallen forward wounded ; he still clings to his 
spear, but his helmet has fallen off. The other two are armed with spears, 
and wear helmets and cuirasses. Behind Geryon are the figures of the 
herdsman Eurytion and his dog Orthros, the latter seated on the ground, the 
former stooping forward. Eurytion wears a pileus or conical leather cap, and 
carries a chlamys over his left arm. He is armed with a spear and sword. 
Under the handle appears the figure of Athena with hands raised, one of 
them holding a spear, round which twines her serpent. 

The types of Herakles and Geryon have been collected by Klein, 
Euphronios (2nd edn.), p. 59, and those on Brit. Mus. vases in the Catalogue, 
ii. p. 17 ; nor do there appear to be any additions to make to the list. The 
one that approximates most nearly to our example appears to be the r.f. kylix 
illustrated by Klein on p. 81, but only in our vase is Herakles turned towards 
the left.^ Again, nowhere else but on our vase does Herakles kneel even 
when using the bow (which is also seen on Munich 407 and Brit. M|us. B 442) ; 
the attitude in this case seems to be borrowed from the east pediment of 
the Aegina temple, where Herakles appears as a kneeling archer. 

Another new feature is the position of Eurytion and the dog, who are 
usually represented as lying wounded or dead; and finally we note the rock, 
which often indeed forms the centre of a scene on b.f. vases (cf. the Brit. 
Mus. Troilos hydria B 324) but never otherwise occurs in a Geryon-scene ; in 
fact Herakles is generally too close to Geryon to use any weapon but a club 
or sword. It may be that the painter put in a rock here to give the idea of 

' Also in Noel des Vergers, Mnirie,^ pi. 88. 





Panathenaic amphoriskos, 3^ in. high, with black figures on red ground 
and white details (Fig. 8). This little vase can hardly be regarded as 
belonging to the class of Panathenaic amphorae, although it has all their 
characteristics in miniature. It was presumably a child's plaything, as many 
vase-paintings seem to indicate that toy vases were popular in the Greek 

Fig. 8. — Panathenaic Amphoriskos. 

Small as it is, it is by no means devoid of interest. In date it must be 
quite late, perhaps of the 4th cent. B.C., to judge by the free and careless 
execution. The figure of Athena is of the usual type, but on the reverse is 
what appears to be an entirely new subject for a Panathenaic amphora, viz. a 
runner in the torch-race (XafnraSrjBpofita). This subject is common on 
late r.f. and Graeco-Italian vases of the fifth-fourth century (e.g. F 59 and E 
389 in Brit. Mus., and Tischbein, Hamilton Vases, ii. PI. 58). The runner 

1 F 101, E 627, E 634, E 635, kc. in Brit. Mus.; also Jahn in Ber. d. siicha. Oesellsch. 1864, 
p. 243 H. 


wears a large wreath with several upright crests over the forehead, puiuted 
in white. 

An almost identical vase has lately been acciuired by tiie (Jassel 
Museum.^ Instead of the torch-runner, an atlilete of Polycleitan type is 
depicted on the reverse. 

H. K Waltljis. 

* Jahrbuch, xiii. (1898), Anzeigcr \>. 192,fig. 11. 


The coin of Sicyon, of the obverse of which a drawing by Mr. F. 
Anderson (made over a photograph) is given here, has been twice pubHshed 
both times by Professor Percy Gardner.^ 

Statek of Sicyon (British Museum). 
Enlarged 2 diameters. 

It is a stater of the fourth century B.C., of the usual types : 
Obverse : t. E Chimaera to r. 

Reverse: Dove flying r. ; behind it, over the tail, a small bow. 
The whole in an olive-wreath. Concave field. Slightly double- 
struck on both sides. 

Weight 188 grains (12182 grammes). 
The inscription on the obverse, which lends special interest to this piece, 
is unique among adscititious inscriptions upon Greek coins,'^ not only in its 
elaborate character, but in the manner of its execution. Such inscriptions 
are in other cases graffiti, scratched with a point ; this is pricked into the 
metal with a pointed instrument. 

^ Numismatic Chronicle, 1873, p. 183 (PI. 
VII. Fig. 5, from a drawing by F. Lees) ; Brit. 
Mas. Catal. Peloponiiesus, p. 41, no. 65, (PI. 
VII. 26, autotype). 

' They have been collected by F, Lenormant, 

Eemie Numism. 1874-77, pp. 325 f. To his 
list add the coins of Pheneus, J.H.S. xvii. p. 83, 
and Corinth, Rev. Num. 1898, p. xliii. and 
B. M. Catal, Corinth, nos. 8, 131, 226. 


Professor Gardner, in publishing the inscription, reads it 


He adds that the T of the third word may be a r, the A of the fourth woni 
is indistinct, and that ' at the end of that word is a mark which might stand 
for I, although I believe it merely to indicate the end of the inscription, there 
being a similar mark at the end of the first word.' These difficulties are, 
however, small in comparison with that connected with the meaning of 
iXKera'! (or rather e\«eTa9, if it is an adjective formed from the verb ?\k€iv). 
None of the explanations connecting the word with eXKCiv seems to me entirely 

Under these circumstances a further examination of the inscription 
seemed worth the making. My results are as follows. 

In the first place, as will be seen from the drawing, the initial letter of 
the inscription is most probably that which stands under the chimaera's tail, 
behind the left hind leg. * I make no doubt that the letter is T and not r , 
which could hardly be represented with the same number of points in both 
strokes, even in a carelessly punctured inscription (which this is not). But if 
it is T, its position in regard to the other letters can only be explained by its 
being the first letter of the inscription. In beginning the dedication, it was 
natural to hold the coin so that the type stood the right way up. But the 
WTiter found out his mistake when he came to make the second letter. If on 
the other hand this T is not the first letter in the inscription, its position is 
much less explicable. 

The inscription offers no further difficulty until we come to the first 
letter of the second (outer) circle. This is read by Mr. Gardner as A. AH that 
is visible is A. It might be supposed that the lower parts of the two legs of 
A were omitted owing to want of room. But this supposition is excluded by 
the fact that in other cases of confined space the punctures are earned over 
the edge of the coin, as may be plainly seen in the letters KE behind the tail 
of the chimaera. Hence there seems little doubt that the letter in question is 
A and not A. 

After the letter N comes the last sign in the inscription ; Mr. Gardner con- 
jectures this to be a stop of the same kind as occurs at the close of the word 
ARTAMITO?. But while the latter stop is made with two strokes of the 
instrument, efifecting marks much larger than any others in the inscription, 
the sign at the end is made in exactly the same way as the letters. It is 
safe, therefore, to suppose it to be | and not a stop. 

Thus read, the inscription becomes 


which I interpret raf; 'ApTdfiiro^ ra? e(\) A(a)KeB{aC)/j,ovi. 

* 'Drawer of the bow,' 'deliverer from formation ; and the termination -tm would at 
trouble,' 'helper in childbirth.' It is also any rate have a passive force, 
questionable whether i\KfT6^ is a possible 

304 G. F. HILL. 

The omission of the second letter of the preposition is of course quite in 
order.^ But the omission of the vowels in the word KaKehalfxovL certainly 
gives pause. The first A could, however, easily escape, owing to the A 
immediately preceding it ; the A I must have been sacrificed for reasons of 
space. It was more necessary to preserve the dative termination than the 
vowels in the middle of the word.^ 

Abbreviaiion by syncope is excessively rare in Greek before Byzantine 
times.3 As M. Perdrizet has shown,* fiaXio<i on the now famous tile of King 
Nabis is not an abbreviation, but an imperfect rendering of the rapid pro- 
nunciation of the Doric form ^ahiKio<; as /9aA\eo9. But the form /3au9 which 
occurs on tetradrachms of Smyrna ^ at the beginning of the second century 
B.C. is an undoubted instance of syncopated abbreviation. The form ^acrcrrf^ 
has been quoted from a papyrus of the time of Euergetes IL,^ but it is so 
carelessly written that it can hardly count as evidence.^ 

On Greek coins of the Imperial period, especially at the beginning of 
third century, it is common to find the word AYTOKPATHP abbreviated 
AYTKP or AYTKPA.^ This form occurs so often that it can hardly be due 
to a mere blunder. In the CB which is sometimes found instead of 
CeB(o<rT09), the loss of the € may perhaps be explained by its likeness to C. 
Forms such as AOVKlC for AovKto<i, on the other hand, are probably neither 
abbreviations nor blunders, but, like HMIOBEAIN, anticipations of the later 
Greek terminations -t?, -iv.^ After the /3av9 of the coins of Smyrna, the 
earliest instance of syncopated abbreviation known to me in official inscrip- 
tions is the monogrammatic form of KAP for Kaiaap which Imhoof-Blumer 
has described from coins of Chalcedon and Byzantium struck at the 
beginning of the Empire.^*' But many methods might have been allowed in 
monograms which were unusual in ordinary writing. 

These notes are sufficient to show that the syncopated method of 
abbreviation existed, though sporadically, at an earlier period than is 
generally supposed. 

* Cp. TOi(X) AaKtBaifiovlo[is] (undoubtedly the osterreieh. Gymnasien, 1891, p. 673 ff. For a form 

right restoration) Olympia Inschr. 252. In Hke KN for Kvv6<Tovpa (p. 709), even if the 

other cases the double \ is written : iK AaKeSai- restoration were certain, would hardly count. 

fiova, Olympia, no. 171 ; i\ AaKfiaifxovi, 4 JSTumism. Chron. 1898, p. 5. 

Meister, Or. Dial.- Inschriften, 4430. But its 5 j^ g Mionnet, iii. p. 190, no. 917; Supp. 

omission is in accordance with the rule which yj p 3Q2 ^q. 1391. 

gives us Kar6v, vor6v for nhr t<Ji/,. itJt t6v, and 6 Qrenfell, An Alexandrian Erotic Fragment 

other single writings of double letters. ^.^^ jj^ 24. 

2 In AF*TAMT$, which occurs in an in- 7 Mr. Kenyon points out that there is room 

scription on a metal vessel (Hoffmann, Gr. for more than trtr, and that the word appears 

Dial.-Inschr. 1600 ; Purgold, Arch. Zeitung to be fiaaiXivarts very cursively written, 

xl. (1882) p. 393), the first ^ is probably a E.g. at Perinthus, Berlin Beschr. d. ant. 

omitted by a mere accident. The epithet pre- MUnzen, i. p. 214, nos. 41, 43. 

ceding should perhaps be completed [A I C]- ^ J.H.S. 1897, p. 82; Jannaris, Historical 

EPAT5 • for the worship of Artemis at Greek Grammar, %iO\, ZQI. 

Aigeira see Pans. vii. 26. 2 f. " Joivnml LUernatioruil d'Arch. Numzsm. 1. 

^ No instances are given by J. Simon, Ahkiir- (1898), pp. 15 i. 
zungen aiif gr. Inschr. in the Zeitschr. f. d. 



It remains only to admit that it is impossible to ascertain to which of 
the many goddesses named Artemis in Lacedaemon this coin of Sicyon was 
dedicated. If, however, the dedication was pricked on the coin before the wor- 
shipper came to Lacedaemon, the want of closer definition does not seem 

Since the above remarks were put into type, the inscription with which 
they are concerned has been interpreted in yet another way by Professor O. 
Rossbach.2 In most of the preliminaries to an explanation we are agreed ; 
as, for instance, in the identification of the initial letter, and in the interpre- 
tation of the final word as a dative. The I7th letter, however, he takes to be 
7, and his transliteration is accordingly 

Ta? 'ApTa/xtT09 Ta? €7 KeSfiwvi. 

This interpretation has the one great advantage of dispensing with the 
abbreviation which I have assumed. On the other hand, two considerations 
lead me to adhere to the interpretation I have proposed. In the first place, 
as Prof. Rossbach himself admits, the place-name KeBfiwp is entirely unknown. 
The invention of this name is not entirely justified by any philological 
probability it may possess, or by the addition it makes to our list of 
sanctuaries of Artemis. Secondly, the form A for 7, although by no means 
impossible in Peloponnesian alphabets, is much less common than r. The 
probabilities are therefore in favour of the value which has hitherto been 
given to the sign in this inscription. 

G. F. Hill. 

* At the shrine of Zeus Kasios in Corcyra a 
stamp with the name of the god, Aibi (mono- 
grammatically written) Kaalo, was impressed on 
dedicated coins (Brit. Mus. Catal. Thesaaly to 
Aetolia, p. 158, nos. 616-632, including coins 

of Lacedaemon and Cnidus, as well as of Corcyra 
itself). It would seem that the stamp in this 
case was provided by the temple authorities. 

2 Berliner Philolog. Woehenschrtft, 20 Aug. 
1898, p. 1063. 



The following inscriptions are some of the epigraphical results of three 
journeys in Eastern Asia Minor. Tlie first two of these, in which Prof W. M. 
Ramsay, Mr. D. G. Hogarth, the Rev. A. C. Headlam, and Mr. J. A. R Munro 
took part, were made in the summers of 1890 and 1891. the third, which was 
organised by Mr. Hogarth and in which I took part myself, was made in the 
summer of 1894. The geographical results of these expeditions have been 
published in the Proceedings of the Royal Geographical Society in the form 
of a paper by Messrs. Hogarth and Munro, entitled ' Modern and Ancient 
Roads in Eastern Asia Minor' (R.G.S. Svppl. Pcqjcm, iii. pp. 643-739) and a 
paper by myself ' A Journey in the valley of the Upper Euphrates ' (Geo- 
grajyhical Journal, viii. pp. 318-335 and 453-474). 

The inscriptions are arranged under the places where they were found 
without reference to their probable dates or the language in which they 
are written. It would have been possible to classify them either in 
respect of age or of subject, but they do not fall readily into any such 
divisions. Comparatively few of the inscriptions are without interest, but 
special attention may be drawn to Nos. 1, 2 which are interesting as belong- 
ing to the historical CiHcian Gates, to No. 14 (copied curiously enough in 
London) from Samosata which adds another to the remarkable series of 
inscriptions on the monuments of the Nemrud Dagh and neighbourhood, and 
to the legionary tiles from Sadagh which establish the identification of this 
place with Satala, for which epigi'aphical evidence has hitherto been wanting. 
No. 45 a fine inscription of Justinian and No. 34 which gives a fixed point 
on the frontier road along the Euphrates north of Melitene are also worthy 
of note. 

I am indebted to Prof Ramsay, Mr. Hogarth and Mr. J. G. C. Anderson 
for many suggestions which are not always acknowledged in the text. 

In every case the initials of the copyist and the year in which the copy 
was made are given above the inscription. 


No. 1. D.G.H. 1894. Ciliciati Gates. 

IMPC ■ V Imp. C[aesar Marcus A]u- 

RKLIV8 VS relius [Probus ? Pi]us 

FELIXINVIC AVG. Felix Invic[tus] Aug[u] 

STVS . O .stus [P]o[ntifex Maximus] 

VIAMIN viam [latiore tramite fecit ?"| 

OBOII "O{p)oi K[i\iKcoi'] 

S .... 


The inscription, which is cut on a panel in the rock at the narrowest part 
of the Gates, is the same as Le Bas-Waddington, No. 1520 and C.I.L. iii. 228 
where Mommsen restores vi\am latiorc [tramite from AMIATIORE). It was 
copied by Mr. Hogarth with tlie aid of an improvised ladder made of 
two fir poles and pieces of cord. With an ordinary ladder it would 
doubtless be possible to decipher the whole o'i the inscription, which is fairly 
well preserved. 

No. % D.G.H., V.W.Y. 1894. Ibid. 

I I . AM (vi)am 

EI OPILIS e[t pontes] (a) P(y)lis 

VSC . . . ILXAV . . SM usq[ue ad A]l(exan)[dri](a)m 

EXIN . . CR . CAES"^ . IVI. ex in[te]gr[o] (r)est[i]tui[t]. 

On a pillar cut out but not detached from the living rock on the 
modern road about 100 yards below (on the south side of) the Gates. It 
is the same inscription as that given in Le Bas-Waddington, No. 1519 and 
C.I.L. iii. 227. I owe the interpretation of this and the foregoing inscription 
to Prof. Ramsay, who read, at the end of the first, OP 01 . . AlKtON, and the 
second, from viam onwards, in full as here restored. The new inscription 
enables us to restore the end of Le Bas-Waddington, no. 1495 : usque A[lexan]- 
dria[m e]x in[teg]ro [rest]i[tuit]. 

No. 3. D.GH., V.W.Y. 1894. Missis (Mopsuestia). 




















X 2 

308 V. W. YORKE. 








Mofo-eo9 ifjb vda-t 7rpo\r][^(f)0€l<i] vtto Tpv(f)Q)vo<{ tov a8€X(f>ov, 7n<TT€v<ra<; 
avTto Tr)v TTpaaiv t&v yeco^py^Lcov TrdvTcov elKoaaerla^;, \^ov\he ttotc Xoyia- 
T€u<Ta<i avTOv /cal d['ir](a)[T]r]6el<; vtto avrov Kara nravra «a[t] ^irj Svvdfi€vo<i 
dyiv 7r/J09 avrov to irpdyfia, XeiTro/u-ei/o? kuI rfj e07/^[e/>/]ft) Tpocpfj kol irapa 
fila Tayvrepov TeXeicop tov ^lov, iiriKaXovfiai kutcl Tpvipwvo^ tov dB€X(f)ov 
fMov Koi TWj/ TeKvcov avTov Tov<i ivovpaviov^ 6€ov<; kol tov<; KaTayjdovt.ov'i, 
KoX Trdaav dpav Koi Xvaaav '^oXcodijvat avTol<; iv oX(p tc3 ^itp avTtov koI Ta 
lepa fjLT) i^ov avTm Troifjaac kuto firjheva Tpoirov, firjhe oaTOvv fiov a-aXevaai 
e(x) TOV fivrjfiaBiov t9 tov alcava r) Tpv(f)cova rj dXXov Tivd fxrjSe i^atpaviaai ti 
T(ov ev Tc5 fivr)fiei(p, fXT}8e to irapaKei^evov fiot ifi^oXdSiv opv^at Tiva, fieveiv 
8e ifiol dxipeov i<f>' oXov to ea-oiOev t(ov a-vvKecfiivoyv Xidoiv, e/cTO<? el fir] eav 
^dyva fiovr) OeXija-rj 77 o-i;[i/]o-«?7[y]69 fiov. rjv koX eav Tf9 dBiKija-jj rj ifie Tt9 
^Xdyjrr) a8t/ca)9 K€)([o]Xco/jLevot, avTo[l<;] yevovVTo ol \^a^vT\o\\ deoi.. 

On a stone lying close to the road from Adana to Missis, about five 
minutes before entering Missis. It was copied by V. Langlois and is 
published in Waddington and Le Bas, No. 1499, but as their copy is far from 
perfect and the inscription is interesting, I have given our copy in full. 

The Greek is now quite straightforward, the only doubt being the 
interpretation of ifi^oXdhtv (line 21). Waddington and Le Bas take this 
to be for efifioXdBtov the diminutive of €fi^oXd<; ' a grafted tree.' Can it, 
however, be an adverb, expressing violent rvfi^wpv^^a ? Mova€o<; and 
dxipeo*; are for M.ovaalo'i and dKipauo^. 

In line 24, we may restore <TvvaK'T]vo<i = contubernalis with certainty. 

The letters at the bottom of the inscription are undecipherable ; 
they possibly may have recorded the infliction of penalties for violation 
of the tomb. 

No. 4. D.G.H. 1894. Ibid. 

. . Toh . . . o(()oNirTnEAo 


av]€Oev Biov 7roXe/xo<<? crrpaTtcoTav. 
Ar]fi7jTpio<i 'A<T/f[X,7;7ri]a[6]ou(?) VaiM YleTpoyvup {M)evi'Tr7j(^) tcG dBeXtfx^ 
Kai Toif; yovevai fMvtj/jLrjii ■y^dpiv. 

The first part of the inscription is evidently metrical. MeviTnrov, the 
genitive, seems to be written for Mei/tTTTry, the dative, as often. Demetrius 
and Menippus were sons of Asklepiades(?) ; Menippus became a soldier and at 
missio honesta received the citizenship as usual, and took the name C. 
Petronius, perhaps after his commander. 

No. 5. D.G.H. 1894. Ibid. 

^TVSa^NTeN 1 



re\H\' AV99 

[Va]lentiniano et V(a)lenten(o) et [Grjatiano (m)axi[mis] vic[t]ori[b]u(s) 
8e[mper] Augg .... 

On a column, much defaced, in a graveyard below the village on the 
right bank of the Jihan. It is evidently a milestone. For the form Valenteno 
for Valenti cf. C.LL. viii. 10352 where Valentino occurs. A copy of this 
inscription is given by Heberdey and Wilhelm in their work Eeisen in Kilikien 
{Wiener Benkschriften\\..-^.\^),hui as it diflfers slightly from the above I 
have thought it advisable to publish our copy. 

No. 6. D.G.H., V.W.Y. 1894. Ibid. 



ayio^ 6 ^[fo]?, 07409 la-^vp6<i, ayio^ addvaro^, 
6 <Tr\avp^(i}6\<; hi ^fid^; eXerjaov rifid^. 

310 V. AV YOEKE. 

On two fragments of a long marble block lying in a graveyard on the 
right bank of the Jihan (Pyramus). The second half of the inscription has 
also been copied by Davis {Life in Asiatic Turkey, p. 67); he has AIHMAC 
in the second line. This form of the Tpiadytov was introduced by Peter the 
Fuller at Antioch (he died 477). It was afterwards adopted by the Syrian 

No. 7. V.W.Y. 1894. Yarso^vat. 

All Aa 

No. 8. 

No. 9. 

MEnCTU)! f^eyia-Ttoi 

A////////HAIAMATPljJ A[vp]'nX^a Marpw- 

NAZOJCAeAYT va ^coaa iavr- 

HKAIAYPHAIOJ fj kuI AvprjXicp 


QJANAPIMN oi dvSpl p,v- 

HMHCXAPIN »;>?/<? xapty 

Nos. 7 and 8 are on cippi in the courtyard of a private, No, 9 on a 
sarcophagus lying close to the mosque. Yarsowat is a small town showing no 
sign of antiquity about 12 miles on the road from Missis to Osmanieh. The 
inscriptions may denote an ancient site. 

No. 10. D.G.H., J.A.R.M. 1891. Sis (Flaviopolis ?). 
In the courtyard of the Serai; badly cut and defaced, readings very 





'AvTiTraTpa) nao-t[/c]p(aTr;9) or TlaatlcfiijKltj or -09] 
dfjL e[T]at/3€('a(?) [ecr](T77)(7e t- 
(c5) v{^) avTov Ttjv Kude- 
B^pav evvoLa<i -^dpiv. 

In line 3 we have the genitive t]ov vo(v) as often. 
In line 2, the last five letters are apparently CETCT 


No. 11. D.(J.H. 181)1, ]hi,l. 

Cippus in a small cenieteiy m the town ; very taint. 



"Etou? e/io-' -[6 helva] 
Tr)v] Kaa-TTjv yvveKa avTov 

€V€Ka] fJLVt]fir)<i. 

Sis was formerly identified by Ramsay with Flaviopolis {Hist. Geogr. 
p. 385). The era of this city was ad. 74 which will make the date of 
the inscription A.D. 319, if the identification, which he is now inclined to 
abandon, is correct. 

No. 12, D.G.H. 1891. Marash (Germanicial). 
Cut on the rock over tomb-doors in a street. 

(1) eVeVvj/YXlA Ev <:€v>^|rvxli = el)'A. 
rA0OKAlA yadoKXia. 
OYAeiCAG ovSeU dd- 

ANATOC dvaro^. 

(2) eVv^VXHC Evylrvxv( = €i) X- 
/llll/llMHOy [a\(ol)^ri. ov- 
////////A0AN B€U]dddp- 
//////TOC ajro?. 

No. 13. J.A.R.M. 1891. IbU. 

Fragments of stele, round at the top, broken below and chipped to 
right, built into the wall of a private house ; the rest is built in close by, but 
the writing is hidden. 

HEnEriHKDn Oe^ iirnKotp 


BAPNAIDY Bapvaiov 

CTPATHTDC (TTparr^yot 

EYPHNANEL evpa>v dvia- 

THrENVriEPT Trjffev virep T- 

[?79 crwrripiaii eavrov k.t.X], 

312 V. W. YORKE. 


No. 14. D.O.H., V.W.Y. 1895. Samsat (Samosata). 







I have indicated the corner of the stone round which the inscription 
runs with crosses XX. 

o<? 8t«at09 e7rt<^aj/^[? (fytXopcofiaioq Kal 
<f>iXiX\7}v 6 [e]« /9a<xt\e[G)9 Mt^paSarou ku- 
Xi^viKov Kal (3a(7iXi(Ta7]<i A[aoBtKr)^ Bedfi (f>iX- 


5 a8€]\(f>ov T^<f €K ^acriXeaxi ' Ai>t[i6)(ov Beov 0i- 
\o]fiijropo^ KaXiviKOV tovt[o (va-e/Ssia yvoofiijii 
€]fMrj'i vofiov re Koivt)<i evcrel^eiaf a-e^ofievo^ r- 
cL TrdvTa vpovoiai hai^6vw\v XiO€iai<; uneSei- 
^ eV iepai<;. €70) irdvTcov dy[a6a)v ov fxovov ktt)- 

10 aiv ^e^aioTarrjv dWh K[al dirokavcrtv ijhiarrjv 
avd^pdavoi^i evoficcra T^[i' evcre^eiav, rrjv avrrjv 
T€ Kpitriv Kol 8vvd/j.€(i)^ €[vTV')^ov<i Kai y^prj(Tea)<{ fia- 
Kapta-TTJf; alriav ea-)(ov, [irap' 6\ov re rov 0iov oi- 
<f)dr}v dirdcri ^a<ri\eta<; €fj,[i]<; kuI <f)v\aKa TnaTord- 

15 rrjv Kol [rjepi/rti/ afxeifirirov [rjyov^€vo<; rrjv bcrio- 
Ttjra. At' ct Kal kivSvvov^ fi€[yd\ovf; irapaSo^ox: 
Bc€<l)vyov Kal irpd^ecov 8v<T€[X7ri(TTCov evfirj'^dveof; 
iweKpaTrfaa Kal fitov 7ro\ueTo[i)9 fiaKapi<TTfo<i eirXr}- 
pcodrjv. €70) Trarpoaiav ^acn\€l[av irapaXa^obv dvo 

20 At09 T€ 'ripofidcrSov xal 'A7r6\\[<wi/o? Mi6pov 'HXi'ov 'Ep- 
fiov Kal 'Aprdyvov 'Hpa/cXeof? ["Apeox?, Kal ttoit]- 
<rd]/jt€vo<; 7ra\acd<i 8vpdfi€(o<; [kuI rvyr}^ v€a<t rij^ e- 
firi<i rjXiKccoTip deoiv fieydXa)\y Ttjv dp')^alav TCfirjv 
iv Updi re Xideiai fiid<; 7repio[ . . . hai^oacv ovpa- 

25 vio(,<i -^apaKrrjpa fjLop<f)rj(: efiri<i [eTn^Koocf; avvOpovo- 
V €i<i he^ik^ rrapea-rija-a, fi€[ifir}fia Sixaiov ^vXda- 
<T<ov dOavdrov <f)povri8o<i 

On one side and the back of a broken slab of black basalt in the 
possession of H. J. B. Lynch, Esq., of 33 Pont St., London. This slab was 
brought to England some years ago by Mr. Lynch's father ' from the banks 
of the Euphrates near Samosata.* On the front of the slab a human figure 
is carved in relief. Only the upper part of the figure is preserved ; it is 
turned to the left with right hand extended and wears a radiate crown. 
The whole slab measures 2 ft. 7 in. in height, 1 ft. 2 in. in breadth and 9 in, 
in width. The depth of the relief in which the figure is carved is at the 
forehead IJ in. and at the breast 3 in. 

From the character of the inscription and relief, the material and the 
general correspondence in measurement, there can be no doubt that this stone 
belongs to a series of reliefs, one of which has been found already at the 
village of Selik near Samsat (Samosata) and is published in Humann and 
Puchstein's book {Reisen in K.A. p. 368, ff.). The subject of this relief is 
very similar to that of one of the reliefs found on the Nemrud Dagh in which 
King Antiochus of Commagene is represented in converse with Heracles 
whose hand he grasps, and it has an inscription which is almost word for 
word the same as a portion of the long inscription which Humann and Puch- 
stein found on these monuments. Hpnce they infer that a second series of 
reliefs on a much smaller scale with the same subjects and inscriptions was 
put up by Antiochus at or near Samosata, for it is not certain that the 
monument at Selik is in situ. The inscription on Mr. Lynch's stone in the 

314 V. W. YORKE. 

same way corresponds almost exactly with the beginning of the long inscrip- 
tion on the Nerarud Dagh ; it is written in the same manner, that is, partly 
on the side and partly on the back of the slab, and the figure of the relief is 
turned with extended hand as if in converse with another figure. By the 
analogy of another of the reliefs belonging to the large monuments this 
figure may be identified almost certainly with Apollo, who wears here as 
there a radiate crown (Humann and Puchstein, op. cit. taf. xxxviii. 2). The 
only discrepancy between this slab and the slab found by Humann and 
Puchstein is in the measurement of the depth. Our slab measures 9 inches, 
theirs 22 inches, but there is no reason why the slabs should have been all of 
the same thickness. 

The restoration of the inscription presents little difficulty. Lines 1-6 
are the same as the beginning of the inscription on the Nemrud Dagh (H. 
and P. op. cit. p. 272, lines 1-7) and lines 9-19 of our inscription correspond 
exactly with lines i. a. 10-24 of theirs. The rest of the inscription can be 
easily restored with the exception of line 19. Mr. Hogarth suggests 7r6/3io[;^?)<?], 
Mr. G, F. Hill irepiolhov] {i.e. a stone relief filling one circuit of the monu- 
ment). The expression 6t9 5efta<? = 'in converse' does not occur on the large 
inscription, but doubtless there was a relief on this monument representing 
Anliochus and Heracles grasping hands as on the Nemrud Dagh. 

No. 15. D.G.H., V.W.Y. 1894. Ihid. 

I/I// D\nPCCBCeBAUJ\CJP/l///l 
. . . ov 7rp(e)o-/3(efTo£>) 'S,€^{aa-Tov) dvTiaTplaTijyov] .... 

Cut in large letters on a big squared stone built into the ka/^ at 
Samosata. Unfortunately the name of this ' legatus Auguati ' is missing. 

No. 16. D.G.H. 1894. Ibid. 

3Y<j> Aee^^'i A 


On a pedestal in the court of an Armenian house. 
No. 17. D.G.H., V.W.Y. 1894. iftirf. 

o- o- M [I]. (O). M. 

LEG X7l Leg[io] XVI. 

F <y F F[lavia] F[irma] 

On a cippus in a private house. 

No. 18. D.G.H., V.W.Y. 1804. Ihui. 


Legion(i)s XVI. F[laviae] F[irmae]. 

Retrograde. On a tile shown to us. The Legio XVI. Flavia Firma was 
brought into existence by Vespasian, and is already known to have been 
stationed here (v. C.I.L. vol. vi. p. 1404, and Mommsen's Provinces, vol. ii. 
p. 119). 

No. 19. D.G.H. 1894. Kiakhta. 

opnep . e . 





o{v)vep i- 

009 6Teu^[a- 
ro ^(Ofiov. 


Hexameter. On an altar in a private house. Philadelphus dedicated 
the altar to one of the kings of Commagene, meaning the reigning king to 
be understood, while he adds a dedication at the end to all the kings (past 
and present). 

No. 20. V.W.Y., D.G.H. 1894. Udd. 

At the Roman bridge on the column nearest the left bank of the Kiakhta 
Chai, almost completely erased. It is on the same column as that on which 
the name of Julia Domna is inscribed (v. C.I.L. iii. suppl. 6714). 


The letters are all so doubtful that it is useless to hazard any conjecture 
as to the Emperor referred to, but the existence of the inscription is 
important, as it points to the probability of this bridge being the work of an 
earlier Emperor than Septimius Severus. It is improbable that the erasure 

316 V. W. YORKE. 

can be that of Geta's name, for, as Humann and Puchstein have shown, there 
must have been a fourth column belonging to the bridge on which his name 
would naturally have been inscribed. The copies of all the inscriptions 
belonging to the bridge, with the exception of the above, which was not 
noticed by Humann and Puchstein, are to be found in the Corpus {C.l.L. iii. 
suppl. 6709-6714). In 1. 13 of 6714 we read MVNIFICENTISSIMVM. 

No. 21. V.W.Y. 1894. Perrin (near Adiaman, the ancient Perre). 

tYXE ^I^X(«^) 


MAPOANHC Mapodvr)^. 

TAYTATAP ravra yap. 

On a recess cut in the rock ; to the right of the inscription are the 
figures of a reclining man and sitting woman roughly sculptured in the rock. 
It is one of many tombs cut in the rock which abound on the site. There 
are probably many more inscriptions to be found on these, but time was short 
and my investigations were curtailed thereby. ravTa yap = for this (is what 
they are). Prof. Ramsay compares ovk ijfirjv iyevo/nrjv. ovk eao/jLar ov 
fiiXei fioi. 6 ^io<i ravra (Ramsay's Cities and Bishoprics of Phrygia, i, p. 700 
and B.C.H. 1884, p. 233). 

No. 22. 

".W.Y. 1894. Ihid. 







eeof OY 




The name yiaix^oyea or Ma/A/Qoyea)? = Ma/x,y9o7ato9 is interesting.^ 
Mabug or Mambug was the native name of Hierapolis in Syria, which became 
in Greek Bafi^vKt}. The modern name is Membidg. 


No. 23. D.G.H. 1891. Shahr (Comana Cappadociae). 
On a marble mural tablet finely engraved. 

E A M E r I Z T\-/l/l @€f Mey larr) 

THZXnPAZKA T?79 x^R^*^ '^*" 

//////// X A P I Z//// T^ €v]xapca[rcav 

lOTAinXMH M]i0para)XM[': 

1 The name [Mja/B/Bo-yalos has been suggested by Ramsay as a restoration in a Syrian 
inscription (Wright and Souter in Pal. Explor. Fund, Quart. St. 1896, p. 58). 



The name MidpaT(o')^fir)<; occurs in another inscription copied at the 
same place and published by Waddington {v. B.C.H. vii. 135). He draws 
attention to the interest of the name. The goddess is evidently Md 
{v. Strabo xii. 53o). 

No. 24. D.G.H. 1891. Ihid. 

On a cippus of fine limestone in a yard near the top of the town. 


Ba<rtX.€a 'Ap^€\a[oi' 
(f>i\67raTpip TOP 
KTicnrjv Kol <T(OTijpa 
6 Si]fji,o<i. 

This is evidently an inscription in honour of Archelaus, King of 
Cappadocia from B.C. 36 to A.D. 17. 

No. 25. D.G.H., J.A.R.M. 1891. 
Stele in the house of Badiler. 



Kol M.i6pr)<; Tcapa- 
/Set TWfc Trarpi. 

No. 26. D.G.H., J.A.R.M. 1891. Ihid. 

On a small marble stele now in the hammam at Hadjin, nine hours 
from Shahr (Comana), but brought from the latter place according to native 

CEMEIPAMI2 2€/i,etpa/zt9 

DBAPZANDY ' Xpi]o^ap^dvov 

DAIiJNIATHA 'A7r]o\(i)vca rfj d- 

HTHBYrA ya7r]r)T^ dvya- 

MAlAANAnNC Tp]l ISta - - 

The last word, which is quite clear on the stone, may possibly be meant 
as a contraction for dvaTrava-afievrj. 

No. 27. W.M.R. 1890. D.G.H. 1891. 
Pedestal in a wall in the main street. 



318 V. W. YORKE. 

Ilov/3\.(iov) AcKivvUov) KopvrjXiov 
OvaXepiavhv rov €7ri<f)ave(T- 
rarov Kac'crapa 'leponoXet- 
TOiv T} /3ov\'r) Kal 6 8rjfio<;. 

The name is that of the elder son of the Emperor GalHenus, and 
corresponds with the form given in C.I.L. viii. No. 2383 and C.I.L. iii. Suppl. 
No. 6956. For a discussion of the names of the two sons of Gallienus 
V. C.I.L. viii. p. 1051, and Prosopographia, s.vv. An inscription of the father 
has already been found in this district (v. B.C.H. vii. p. 132). 

No. 28. D.G.H. 1890. Ihid. 


HAYBIOC 'Hgy/9to9 

AAMAOY Aa/^a Ou- 

APNATti) dpva TO) 

fTATPlMNH irarpl fivr]- 


No. 29. W.M.R. 1890. Ihid. 

A YP H A4 D A UPA A^p. 'HXcoBcopa 

P A I A Ml/ TIJ To]p8iavS TO) 

L Y NC P I TiK J\ I a]avvKpiT(p Kal 

KPKIILiAAEA xpv<^''^ «^€^- 

4) uj M rsH M h: X A 4><^ f^^vM^ %«- 

P t N pcv. 

For the same name in this district v. B.C.H. vii. 137, Nos. 18 and 19. 

No. 30. D.G.H. 1891. Ihid. 

On a marble stele. The inscription runs round the head of a large 


-|- ''Ei^^a KaraKiTC 6 Tfj<; fiaKapia<; fivij/xrjf; Bap^;^o<?. 

No. 31. J.A.R.M. 1891. Ibid. 
In the main street. Broken R. 



fieydXov \pi(TTo . . . 

^'IX'*"' {B)p6vov ap'^ieprjutv + 

I am unable to explain the monogram. 

No. 32. J.A.R.M. 1801. Seraijik 

In a cemetery near Seraijik. Quadrangular altar or base ornamented 
with wreath.s. 

AAerACZ€YCO^ M€ya<; Zev^ Ov- 

PANIOCeiOYAI pdvio<:' FJovXi- 

OCKATAKCAe^ o<: Kara KeXev- 

ciN y^ — V eec o-iv 0€o[v. 

j Wreath] 

The inscription is the same as that published in B.C.H. vii. p. 147, said 
to have been copied at Hutchbil ; this copy however is an improvement. 

No. 33. D.G.H., V.W.Y. 1894. Malatia {Melitene). 

\ h\>% iS [I]mp. [Cae]s. 


F LI ^ kl /////'^'^/ ^^- J"li["s Constantius] 

f T Fl \lfHI/ffl^^^^ ®*^ ^^' ^t"^^"^ Constans] 

P P F F M /\'/^/y/^f^ P.P.F.F. Ma[ximi Victores] 

SE/APER'6Y^ Semper Aug[usti 

On a pillar, certainly a milestone, forty minutes on the road from 
Malatia (Eski Sheyr) to the bridge of Kirkgeuz Keupru. The stone is 
defaced above and hacked away below. It is probably the same inscription 
as that copied by Fischbach and published in the Carpus {G.I.L. iii. Suppl. 
6893), though comparatively few of our letters agree with those of his copy. 
The milestone may have belonged to either the Melitene-Sebastea or Melitene- 
Satala road, perhaps more probably to the former, as what appears 
to be a milestone was seen on the line it must have taken in the plain 
of Hassan Badrik by Brant {v. Ritter, Erdkwnde, x. p. 863), and we found 
no milestones on the line most probably followed by the road Meliteue- 



Armenia Minor. 

No. 34. D.G.H., V.W.Y. 1894. 

Close to the remains of a Roman bridge over the Kara Budak, a short 
distance from its junction with the Euphrates on the road from Divrik to 
Kemakh, 5 hours 40 minutes after leaving the village of Zimarra, 1 hour 
30 minutes before reaching the village of Hassan Ova. 

flMP-CAES-L-Mo r 
i-I^S^VC PIVSFcpoNT {^^ 

Imp. Caes, C, Mo[es]. 

Traianu[s D]esio Pio Fe- 

li(ci) ? Aug Pius Fc Pont- 

i(f)ex Maximus P(rinceps) O(ptiraus) po(n)tem s- 

u[p]. flumini Sabrina- 

e (res)tituit per C. Va[l. T- 

erjtullum 1. Aug. pr. pr. 

In the first four lines there seems to be considerable confusion between 
the nominative and dative, Traianus is followed by Desio, etc. In the third 
line we have Pius Fc. repeated after Pio Felics in the second line and the 
inscription generally is the work of a poor Latin scholar. Attempts at 
correction (felic. for felius, pontific. for pontifex) may afterwards have been 
made. Where we should expect to find restituit we have a strange word 
poltituit or potlituit, every stroke of which is plain on the stone. 

For the title of P. O. (princeps optimus) in connection with Trajan 
Decius, compare C.I.L. ii. 4958. 

The name of the legatus Augusti, though it cannot be restored certainly 



from the inscription, is new (v. Liebenain Furschungen, p. 119). Possibly we 
should read Tertullianum. 

The bridge, the restoration of which is here recorded, must have belonged 
to the Satala-Melitene road (see my paper in Gmf/raph. Journ. viii. 467-8). 
The name given to the Kara Budak by the Romans, as far as it can be 
deciphered from the inscription, seems c\irioual} enough to be the same as 
that given by them to our own Severn. 

No. 35. D.G.H. 1894. 

In a garden, opposite Pingan on the bank of E\iphrates, belonging to 
Mesardurian Bartolomeus. It is not far from the spot where the inscription 
of Ala II. Ulpia Auriana was found {Arch. Epigr. Mitth. Oester. 1884, 
p. 239.) 



No. 36. Sadagh (Satala). On tiles. D.G.H., V.W.Y. 1894. 
(1) Picked up. 

^ (Leg.) XV. Apol. 


J O ^ A VXg 

(2) In threshold of house of Mehemet Suleiman. 

ViV"/ ) 3 J Leg. XV. A 

(3) In house of Hadji Hassan. 

(4) Ibidem. 

L wiiiiimm L 


322 V. W. YORKE. 

(5) Ibidem. 


Legio XV. Apollinaris was stationed at Satala (v. Noiit. Dignit. Orient. 
cap. XXXV.) and the discovery of these tiles places the identification of Sadagh 
with Satala, which has been generally accepted, beyond doubt. Other in- 
scriptions of the same legion have been found at Carnuntum in Pannonia (v. 
Arch. Upigr. Mitth. Oester. vol. v. pp. 208 f. f.). 

No. 37. V.W.Y. 1894. Ibid. 
In the graveyard. 

eNGAAEKATA evBdSe kotu- 

KlieOMAKA KLTe 6 fiaKa- 

PIOCANAPeAC pio'; ' AvSpia^ 

KEANAreNOJC Ke avayevtoa- 

K ////T e C e Y C a C Ko[v]Te^ ev^aa- 

TevnepeMOY ■^^ ^^^'^p ^VoO. 

dvay€V(ocrKovT€<i is for dvaytyvcocrKovre'i. For the formula compare 
No. 44 from Nicopolis. Many similar epitaphs exist at Sadagh, but those 
that we were able to decipher are of little or no interest. 

No. 38. D.G.H. 1894. Ibid. 

. . IVS AMMON C. Iul]ius Ammon- 

IVSE . CIVLI ius e[t] C. Iuli[us 

RVS . . . . SF . AT Rus[ticu]s f[r]at[res. 

No. 39. V.W.Y. 1894. Ibid. 

On stone built into the roof of the mosque. Apparently the end of an 
inscription. Broken above and to left. 

i\/\ (^ J\ (J I K O R ' matrejm castror(um). 

The first letter of the inscription is evidently an R from which we must 
infer a vowel between R and M, and though the tail of the intermediate 
letter is quite plain on the stone and it should be I, it is best on the whole to 


restore it as E (perhaps F for E by error). Mater Castrorum is doubtless 
Julia Domna. 

No. 40. D.G.H., V.W.Y. 1894. Ibid. 

lu room of house of Bal Oglu Hassan. At the bottom of an oblong slab 
4 ft. high, broken to r. The inscription is in a very bad light, and being built 
into the living room of a house, it was impossible to have it moved. Con- 
sequently the readings are doubtful. 

E C V N DA/^/z/RlMlE 

Line 1. L may be R. 
and F. 

Line 4 : there may be another letter between R 

Mr. Hogarth suggests 

[S]ecunda [A]rme[niaca Const-J 
antia Justini[ana] 
Mi[lia] perfecta ca[str] 

I have not however been able to find other instances of the formula ' milia 
perfecta,' and Justiniana is an unlikely name for an ala, at least none of the 
known alae take their names from so late an Emperor. 

Armeniorum is an alternative to Armeniaca. There was an ala secunda 
Armeniorum at the Oasis Minor in Egypt {Not. Dign. xxv. A 9). For 
Armenians in the Roman army, v. Arrian {eKra^i^; 29). 

No. 41. D.G.H., V.W.Y. 1894. Ibid. 

On an altar built into the wall of the house of Suleiman Selim 

GEN CoL Gen(io) Col(oniae) 

LE • FECiT Le(gio) fecit 

. . )NSF ... s(acris) f(aciundis)? 

On another side of the same below a wreath. 
COR Cor(ona) 

Y 2 


V. W. YOl^Klv 

This appears to indicate that Satala was a colony, a fact not known from 
other sources. That it may have been a colony and yet have struck no coins 
is not without parallel, v. Ramsay Hist. Geogr. p. 284. 

To these inscriptions of Satala may be added one of the Emperor 
Aurelian found there by Taylor in 1868 {Journal R.G.S. xxxviii. p. 288). 
which may be restored from his copy as follows: Imp. Caes. L. Dom 

Au(r)[eliano] P[io] (F)[elici] Invicto Aug[usto] Pontif[ici] Ma[ximo] 

[P]ar[thico] Ma[ximo The letters ARMA of his copy may also be 

C]ar[pico] Ma[ximo] as in another inscription (Orelli 1029). I draw attention 
to the inscription as it has not found its way into the Covpus. 

We also found a milestone in the village, but it is in such bad condition 
that little more than the letters IMP could be made out. 

No. 42. J.A.R.M. 1891. Purk (Nicopolis). 
+ THAEKATAKITA10THCMA ^V^^ KaTdKirat 6 rri^ fia- 

TrePlM€H4S€An"IAA + 

Kap.ia'i fjbvij/j,rj<i "EwTrfpiKO^ 
69 t6v wpia-jxevov avr^ va- 
pa %{eov) ■)(^p6vov ^CQ)<ra<i to tto,- 
cTi Koivov Ka-TrapeTYjTov tov 
^iov iripa^ ehi^aro ttjv irav- 
KO(Tp,iov tP]<; dvaaTda€(o<i 
nepifievoyv iXniSa + 

No. 43. D.GH. 1891. Ibid. 

Altar-stele now used to form the altar of the lower church, 

OniOCMAP "Oinoii Map/co? ^i(oaa<; KuXax; 

KOCBICiJCAC evae^oiv MapKO^'O-mo^; 'AySyXa? 

KAAUJCeVC vi6<i dvecrrrjatv /3[(u]/U0i/ iJivr)\^tir]<i 

eBOJNMAPK Xaptf-] 







No. 44. D.G.H. 1891. Ihid. On a small stele very rudely engraved. 

+ € N e A A € K///// -(- 'Kvddh^ K[ard- 

Kl TAIHK//// Kirai'\\^.{,- 

P I U) T r € to I pi(9) (•0)T(p)ea)[9 or pt(o9) rea)[p7/oy]. 

K ONTcCcY KovT€<i ev- 

XAC eAine pi ^ ao-(9at Tre^l 

A T T n C atiT'^9. 

A similar formula occurs in one of our inscriptions from Satala (No. 37), 
and also in Lycaonia, see Ramsay's note in Jahreshefte des Oest. Inst, i., 
Beiblatt, p. 95. 


No. 45. D.G.H., J.A.R.M. 1891. Kejiut. 

Limestone slab in the graveyard, ornate letters deeply and very carefully 
cut between ruled lines. The slab appears to have been let into a wall. 
Native Armenians reported that it had been brought from Sivri Tepe, where 
are ruins of a church, two hours in an easterly direction from Kejiut. 

'r oPACKs///^AYMA/i^^^CArAeorK'<^ lAO 

lO^CTIN ianoc*aToyc;toca''tokpat6op 


eEOA///////0YTO^^NA02* KOMKT^TCJN 

^- 6pa<i K€ 6avfid[^]i<; dyadov K€ </>tXo- 
')(^piarov SeaTrorov (fyiXoTifiiav. 
^lovo'TCPiavo'i Xvyova-Toi; avTOKpdriop 

326 V. W. YORKE. 

aveyipe . a . 8e cnrovBfj Ke Trpovoia 
®€oh[o(ri,'\ov To(v) evB6^(ov) /co/4r;T(o9) t(ov 
KadoiaiovfjLevoav) 8o/jL{eaTiK(ov) Ke Oeiov KovpaTopo<i 
e]7ri crcoT'rfpia tmv kavrov olKiqrSiv + 

In line 5 either dviyipe ravra or av. ra TfjSe may be suggested. 

Mr. C. H. Turner has supplied the restoration of line 7. Theodosius 
was comes devotissimorum domesticorum and curator of the sacred buildings 
(cf. Novell. Tiberii, c. 1, 2, 4 : ol ivho^oTaTot koX fieyaXoTrpeiriaTaTOi Kovpd- 
Tcopet; have charge tmv Oeicov oiKtov). 

No. 46. D.G.H., J.A.R.M. 1891. 

Kavsa, i.e. Thermae Phazemonitarum (Strabo) in the wall of the mosque, 
mai'ble stele, broken to left. 




^vfi<f>ai aKocr/xTjTOi'i ivl Scofiacri vaieraovaaL 
' A'x^do/jievai to irdpoidev iT[i]]p[e]o{v) ayXaov [vScop-] 
"Hx^^ro 8"'H</)e(o-)To?- vvv 8' ooTracre 'lo^iva avracf 
'H^yefiovoov o-y^ dpiaTo<;, eV evTv/croi<; OaXafioLaiv 
AvcnTTOvoi^ vvfMtpatac koXov (TTe(f)o<i, b(f)pa /cat avral 
'I/z]6(p)Tai9 {p)i^a><rLv dyaWofievai ^apiTecr<T(i)v. 

For previous publications and discussions of this epigram, see Hubert in 
Rev. Arch. xxiv. p. 308, Rubensohn, Berl. Phil. Woch. 1895, col. 380 and 603. 
The present copy would suggest irrjpeov in v. 2, and confirms Wilhelm's 
suggestion of lfiepTal<i in v. 6 (cf. Anth. Pal. ix. 669). 

No. 47. D.G.H., J.A.RM. 1891. 

Kavsa, in the mosque wall. A late stele made almost illegible by the 
scratchings of a native decipherer. 



p oisk ^miwintrin\o h'^o 

















"Ito\o<? t,r)aa<i 0iov 
nyadov K€ Tp67ro[v 
opdov fce [vov fio]i 6vTo[<i 
v€aviaKo[v] ^Xdov etV 
yfjpa<;. ^(o{v) [i]TroiT)aa 
ifiavitfi ^6<rt[»'] <l>v- 


(Tvv TTj yvveKi iJ,o\^v 
Aofivr) E0HC. e^rjaa 
Ka\a)<; k€ €(v)t€KV(o^. 

e^dpa^a K{aTOfjLvv)(ov fie- 
(T)a rr]{v 'q)^€Tepav KaTd[^de- 
aiv firfSiva eiravv^e 
K€ ^repov Ki)8o<t 
KaTa(d)lvG' eav 8e ti^ 
inai/v^T) Ke ere- 
pov KrfSo<i Karadfi 
Scoat Te3 rafii(fi 
7rpo(nifiov ^ B<I> 
eSoj/ce dcre^ia[^ 

The inscription seems to have been written by Italos, who records the 
preparation of the tomb for himself and wife, and the erection of a aTTjXr) in 
his memory, and adds the usual curse against violators of the tomb. For the 
corrupt form of Karofivvcov compare Ramsay's Cities and Bishoprics, i. pt. 2, 
p. 734, etc. The last line probably gives the name of the violator and the 
amount of penalty paid, but it is impossible to restore it satisfactorily. 



The past year has not been one of startling discoveries. The effect of 
the disastrous war has been a paralysing one and the difficulties in the way 
of archaeological work have been great. While the political situation was so 
complicated, new undertakings were almost out of the question : it is to be 
hoped, however, that the report on 1898-9 will be able to speak of Crete as 
having been thrown open to the scientific world. The most interesting 
archaeological event has been the foundation of a new Austrian Institute, 
under the guidance of Prof. Benndorf in Vienna. Dr. A. Wilhelm, the well- 
known epigraphist, and Dr. W. Reichel, whose brilliant essay on Homeric 
Armour has won for him a prominent place among archaeologists, are 
permanently stationed at Athens, and a building is contemplated in the near 
future. Dr. R. Heberdey will be stationed at Smyrna and Dr. Kalinka 
at Constautinople, so that the new Aiistrian Institute will be a powerful 
agency for the discovery and preservation of Hellenic antiquities. The old 
publication ' Archdologisch-epigraphische Mittheihmgen aus Oesterreich' is now 
brought to a conclusion, and is replaced by the 'Jahreshefte des osterreichischen 
Archdologischen Institutes,' of which vol. i. has appeared. The contemplated in- 
ternational congress of archaeologists at Athens, which the war of 1897 rendered 
out of the question, was announced for Easter 1898 to synchronize with the 
celebration of the fiftieth anniversary of the foundation of the French School, 
but this year as before practical difficulties caused its postponement. The 
adjourned festivities at the French School took place as arranged. If the 
congress is to assemble at a future date, it will be necessary that arrange- 
ments should be made for it long enough beforehand to ensure its success. 

The work of the German, American, and British Schools has 
continued on the usual lines. During the early months of the year Dr. 
Dorpfeld took a six weeks' trip to Egypt and briefly recounted his 
impressions at a meeting of the German School on March 31. His remarks 
about the Temples of Philae in connexion with the contemplated water- 
storage reservoir at Assouan were exceedingly just. It was hoped, that the 
protest of British scholars would have had effect in causing the abandonment 
of any plan, by which the island would be submerged at all. It appears, 
however, that according to the modified plan Philae will still be covered every 
season with a few feet of water. It cannot be too strongly insisted, that this 
means destruction equally with the old plan, only that the process will be 
more gradual.^ The utility of the scheme is clear and no greater benefit 
could be bestowed on Egypt, but even if the expense of building two 

1 See (with reference to Luxor and Karnak) Sa-ndstcinmatcrialsderTempelhautenaufPhilar,' 
L Borchanlt, ' Berlrht iihcr die Corrosion des Sitzungsber. der Berl. Akad. 1898. 23. 


reservoirs instead of one be incurred, it is surely better than tliat the British 
occupation of Egypt should sufifer under a deserved imputation of needless 

The first thing that has struck the eye of the visitor to Athens this 
spring has been the solid scaffolding on the west front of the Parthenon. 
The suspension of the work of repairs has been due not merely to the war 
but to the difficulty of securing sufficiently large blocks of marble from 
Pentelicus to replace the shattered parts of the architrave. Several blocks 
when brought were found to be useless and are lying on the Acropolis. A 
new company has however been formed, which is beginning to quarry on the 
Marathon side of the mountain, and it is hoped that it will shortly be in a 
position to supply blocks of the size and strength. It has been 
rather a shock to the eye for visitors to see this scaffolding, but there is 
reason for thankfulness that the serious earthquake of 1894 did not bring 
down the whole west front. At any rate the work in hand will prevent any 
further destruction of the great fabric. 

The excavations on the north slope of the Acropolis in the previous year 
resulted, it will be remembered, in the complete clearing of the caves of 
Apollo and Pan and the uncovering of an interesting way of access to the 
Acropolis, while the chief actual find was the inscription which finally settles 
the date of building of the Temple of Nike. Part of the construction laid 
bare by this excavation under the caves is now regarded by Dr. Dcirpfeld as 
belonging to the north enclosing wall of the Pelargikon, which in this direction 
extends just beyond the Clepsydra so as to bring it within the fortification of 
the Acropolis and its slope. The wall in question abuts on the citadel rock 
just under the western of the two caves. Another wall uncovered which runs 
down northwards towards the town he regards as part of the so-called 
wall of Valerian The most interesting find made this year is an inscription 
containing part of a yfr7](f>ia-fia proposed by Alcibiades. As it has been 
published with commendable promptitude by M. Kavvadias (Ejih. Arch. 1898, 
1-2.), in whose charge the excavations have been, it is not necessary to say 
more than that it is similar to C.I.A. iv. 1. p. 18, No. 61a, an inscription found in 
the Asklepieion and containing a decree which confirms the arrangements of 
the aTparrjyoi with the people of Selymbria. Thucydides (viii, 23) tells us 
that the Athenians occupied Clazomenae, while the anti-Athenian party 
retired to Daphnus, and that (viii. 31) Astyochus made an attempt on 
Clazomenae later. This decree ratifies the arrangements of the generals with 
the Clazomenians, who occupied Daphnus. Either therefore the anti- Athenian 
party made their submission and received terms, or the Spartans were 
successful in expelling the Athenian party to Daphnus, with whom the 
Athenian generals then made a covenant. The decree supplements the 
narrative of Thucydides, though not to the extent of making the whole 
series of events quite clear. It must have been proposed during Alcibiades' 
stay at Athens in 408, In the ruined Byzantine church of the Seraphim, 
which had been covered by the debris from the Acropolis excavations, other 
inscriptions were found, including an architrave block inscribed KvuiKtov. 

330 a. a HTCHARDS. 

Work has also been progressing at the Stoa of Attalos under the charge 
of M. Mylonas, with a view to its complete clearing. A beginning was made 
at the south side, and on the west side traces were found of an old road. The 
chief finds include an inscription of the fourth century B.C. referring to Zeu? 
(fypdrpcoi; and 'AOijvr], a portrait-head (probably of one of the Pergamene 
royal family), and a small female head of the best period. 

Thirdly the Greek Archaeological Society has been excavating in the 
peribolos of the Olympieion, the work being in charge of M. G. Nikolaides. 
Since Mr. Penrose's work, nothing has been done, and the large precinct has 
never been thoroughly investigated. It was also thought that some of the 
standing columns required strengthening. The whole foundation of the 
temple has been dug round and the stylobate has been found to consist of 
three steps, the upper of marble and the lower two of poros. The lowest is 
only half the width of the other two on the north, south and west sides, and 
was apparently only used as a step on the east side, though smoothed and 
polished and so intended to be visible. In many places the whole structure 
has been removed. A drain for the purpose of carrying off the rain-water from 
the temple was discovered running along the outside.^ The foundations of the 
remaining columns have been strengthened, and in the precinct by the north 
wall, a number of inscriptions and sculptured fragments of Roman times have 
been found. South of Kallirhoe on the Ilissus M. Skias has found a few founda- 
tions which appear to belong to the Ionic temple by the Ilissus of Stuart and 
Revett : and Dr. Dorpfeld regards this as the temple of Artemis Agrotera. 

The excavations of the German Institute in Athens have also been 
continued. After laying bare the dp-x^aia dyopd Dr. Dorpfeld's chief object 
has been to investigate the extent of the later Agora by discovering, if pos- 
sible, some of the buildings which surrounded it. Beneath the so-called 
Theseion on its east side, it will be remembered that he had already found two 
buildings, one of which, a quadrangular hall, may well have been the famous 
Stoa Basileios. Anyhow, in these buildings he sees with great probability 
the western side of the Agora (Ath. 3Iitt. xxi. 458, xxii. 225). This season 
he has attempted to fix the position of the south side of the Agora on the 
north slope of the Areopagus, and in an open space by the little church of 
the Prophet Elias actually came upon the corner of an old Greejt building 
running east and west, surrounded by several later walls. There is great 
probability that this is part of the foundations of one of the buildings on the 
south side of the Agora, perhaps the Metroon or the Bouleuterion. If, as 
Dr. Dorpfeld thinks, the so-called Stoa of the Giants also marks a side of the 
Agora, the dimensions of the latter are nearly ascertained. At the same 
time an excavation on a piece of land belonging to M. Kalliphronas, the 
demarch of Athens, higher up on the north slope of the Areopagus, resulted 
in the discovery of six graves of the Dipylon period. These graves dated 
from a time when the whole of the Areopagus lay outside the fortified wall 
of the Polis, i.e. the Acropolis and its western and south-western slope. Two 

* A recent incorrect newspaper report seems to refer to the finding of the basis of the cultus-statue. 


iron swords, an iron knife, and a bronze lancp-head were found with charac- 
teristic geometric vases of the earlier Dipylon period. The interest of the 
discovery lies in the fact that traces were found of both burying and burning 
the corpses, so that it is no longer possible to hold that cremation was a mucli 
later practice than burying. Both seem to have existed side by side at a very 
early period. No success attended Dr. Dorpfeld's efforts to locate the 
Thesmophorion and the Eleusinion. He has no doubt that the latter .sanctuary 
must have been situated on the west slope of the Acropolis in the great bend 
of the carriage-road, and south of the Amyneion discovered by himself, but 
hardly even foundations remained traceable on this spot. Similarly while he 
still maintains the identity of the temple of Demeter Kore and Triptolemos 
seen by Pausanias ' above the Enneakrounos ' witli the Thesmophorion, holding 
that the Pnyx had the same relation to it as the Boideuterion to the 
Metroon, no remains were discovered in the spot where he looked for them. 
Some time was devoted to clearing the great subterranean aqueduct between 
the theatre of Herodes and its terminus beneath the Pnyx. To the investi- 
gation of the water-system and its numerous side ramifications, one of which 
seemed to extend to the Acropolis itself, he has devoted much attention, and 
pronounces several of the channels and receptacles to be older than the great 
aqueduct. In order to avert the falling in of the rock over the aqueduct, a shaft 
thirteen metres deep was sunk from above at a point half way to the theatre 
of Herodes. Dr. Dorpfeld will before long be able to publish ample 
details and plans, which will make us intimately acquainted with the water- 
system of ancient Athens. Later on in the spring, he commenced an excava- 
tion on a piece of ground north of the Kolonos Agoraios and just the other 
side of the railway. Here a new house was being built, and before its 
foundations were laid a part of the great street leading from the Dipylon 
gate to the Agora was unmistakably discovered. When the cutting was 
made for the extension of the Piraeus railway to the Place de la Concorde, it 
must have gone through this road, but nothing was observed at the time. It 
is also contemplated to make trial diggings on the Kolonos itself round the 
' Theseion,' in the hope of obtaining further light on the identification of this 

Mr. C. N. Brown of the American School has been investigating 
the outside of the Acropolis with a view of finding inscriptions. His re- 
searches carried on with considerable risk to life and limb have been rewarded 
by the discovery of a number of unknown or lost stones. Dr. Cooley of the 
same school has been investigating the traces left by the pedimental sculptures 
of the ' Theseion,' and it is announced that Dr. Sauer will shortly follow up 
his similar work on the Parthenon with a dissertation on this subject. 

Not much change in the Museums of Athens is to be reported. The 
sculptures of one of the pediments of the old Athena temple, the subject 
being Athena in the Gigantomachy, have been set up in the Acropolis 
Museum, and will surprise those who have not seen them both by their 
unexpected completeness and also by the complete contrast they offer to the 
Aeginetan marbles, from which they cannot be far removed in time. In the 

332 G. C. RICHARDS. 

smaller muse\im a sample of the architectural members of this temple has 
also been mounted, which will give a good idea of its external appearance. 
The National Museum every year is getting into more complete order, but 
discoveries corae so quickly, that the process of mounting and exhibiting 
everything is necessarily slow. The collection of bronzes, increased by those 
from the Acropolis and Olympia, has received a noteworthy addition in the 
shape of an archaic bronze nude statue of Poseidon, found by a fisherman in 
pieces near Dombrena and the site of the ancient Kreusis. In spite of the 
restorations, which have unfortunately been coloured so as to resemble the 
bronze, the general effect is very striking. The feet are perfect, the left leg 
being advanced. The head is bearded and much resembles the Zeus-head 
from Olympia. The arms are broken off; the left was raised, but as the 
weight of the body rests on the left leg, probably did not rest on a sceptre, 
but may have brandished a trident. The lowered right arm perhaps held a 
tunny-fish. The head has the hair carefully incised with parallel lines 
starting from the crown. There is a plain circlet and two rows of fourteen 
forehead curls. The eyes are hollow. The beard is wedge-shaped and has 
the small interior wedge below the lower lip, as in the case of the bronze 
head (of a strategos ?) from the Acropolis. The nipples were like the eyes, 
specially inserted. The total height of the figure is about 1*18 metres. 
The thickness of the bronze proves its genuineness — if proof were necessary. 
At Eleusis, M. Skias has been digging for some years on the south slope 
of the Acropolis hill and in the neighbourhood. He has now published the 
interesting results of his work. The importance of his discoveries is that he 
found the layers of succeeding ages undisturbed. Immediately over the 
layers of Mycenaean sherds, he found Geometric pottery containing some of 
the rude ware with scratched patterns like that found at Aphidna, and argiies 
from his finds that the Dipylon art must have been developed elsewhere and 
imported into Attica full-blown. He is strongly of opinion that it came in 
from north Greece with the Dorians. The practice of cremation seems to 
have been the rule at Eleusis. The same enterprising archaeologist is also 
reported to have found traces at Eleusis of a sanctuary of Asklepios earlier in 
date than those at present known. At Megara Dr. Dorpfeld has been 
investigating the topograph}' and water-system of the ancient town in 
company with M. Stambolas, a native who has studied Megarian antiquities 
and contemplates writing a monograph on the subject. The most interesting 
conclusion, at which he has arrived, is that which takes the converse view to 
LoUing's with respect to the sites of Nisaea and Minoa. Lolling regarded 
the low hill called Paleokastro with the mediaeval tower as Minoa, and 
placed the Acropolis of Nisaea on the much higher hill crowned by a small 
chapel of St. George to the east. Dr. Dorpfeld is, however, confident that 
the reverse is the case. On the latter hill he has discovered a wall which 
he thinks is probably part of the fortifications erected by the Athenians on 
Minoa. To this view Prof. S. P. Lambros, in making known a boundary 
stone inscription from Megara of the fifth century with Ato? MiXi^io 
U.av(f)v\o at a meeting of the German Institute, has given liis support. 


The American school contimuMl its W(jrl< iit CJoriiith this season and the 
excavations lasted for about three months ending on June 12. The main 
result was the excavation of the Peirene of Pausanias, the site of which is now 
definitely fixed at a spot about 100 yards directly south of the Platia of the 
modern village. Hitherto there had been two views, according to one of which 
it was the spring on Acro-Corinth (so Stiabo, viii. p. 379, a view mentioned 
by Pans. ii. 5, 1) now covered by a Turkish well-house, while the other 
identified it with the ' bath of Aphrodite ' so-called, below the village of Old 
Corinth in the direction of Lechaeum (see Frazer, Paus. vol. iii. p. 24). Prof 
Richardson writes to me as follows : ' There can be no doubt of the identity. 
It has six chambers with natural ruck covering, but back, front and sides 
architecturally equipped, and so fits exactly the phiase of Pausanias olKi'jfiara 
(TirrjXaloi'i Kara ravrd (ii. 3, 3). The fa(^ade which was in two storeys, had 
the marble revetment mentioned by Pausanias, as is evidenced by holes and 
by masses of fragments of thin plates of marble. One of these fragments was 
inscribed PI RET. Besides the adjustment, which dates from the time of 
Pausanias, an older as well as a later adjustment is plainly discernible. Two 
rock-cut channels, traced to an aggregate of about 300 yards, still bring water 
down from the direction of Acro-Corinth past the two ends of the fai^ade 
supplying the modern village. In front of the facade which faces north, was 
uncovered a semi-circular building closely connected with it, which is likely 
to be the Trept'/SoXo? of Apollo 7rpo9 t^ Ueipi^vr) of Pausanias (ii. 3, 3). The 
agora may now be located within very narrow limits just south of Peirene. 
The old temple appears now to be the temple of Apollo.' (It at first seemed 
likely that the old temple was to be identified with the vaol deoip 6 fiev 
Ai09 6 B' 'Aa-KXrjTTiov (ii. 4, 5), since the division into two separate parts 
has been proved by Dr. Dorpfeld. In the light of the new identifica- 
tions it seems natural to point to the old temple of Athena on the Acropolis 
of Athens and to conjecture that the western cella at Corinth also was used 
as a treasure-house). ' Two more columns of this temple, the seventh and 
eighth on the south side, reckoning from the west end, were found lying, just 
where they fell outward. A flight of poros steps was found leading from 
the temple down in the direction of Peirene, and a flight of marble steps, 
quite imposing, leading up past the west end of Peirene towards the agora. 
West of this latter flight of steps, a long building, probably a stoa, with walls 
standing to a height of from eight to twelve feet, was brought to light, 
closing the valley on its side towards the temple. There were also found six 
large marble statues without heads, a great quantity of old Corinthian ware 
mostly fragmentary, one whole geometric amphora, two interesting bronze 
figurines and a great many Roman inscriptions mostly fragmentary, as well 
as several Greek. Among the latter the most interesting is one of the age 
of the Dveinias-inscription,' (which was also found near Old Corinth by 
Lolling Athen. Mitt. i. p. 40, PI. I.) ' and another of uncertain date, but as late 
as the imperial times, reading \ rCOTHEBP (Tvv]ayQ)yT) 'E^p[aia)p' The 
American school is greatly to be congratulated on the success of a campaign, 
where great patience and considerable outlay were required. Another 

334 O. C. RICHARDS. 

season's work will add Corinth to the number of those Greek sites, whose 
topography has been practically elucidated. 

The Austrian school undertook an excavation on the site of the temple of 
Artemis at Lousoi in North Arcadia, near to the village of Sudhena, on the 
slopes ofMt. Khelmos and not very much further from the town of Kaldvryta. 
The interest attaching to this site was increased by the mention of it by 
Bacchylides in the ode in which he recounts the legend of the healing of tiie 
daughters of Proetus (x. [xi.] 96, 110 Blass) : clearly it must have been, 
therefore, a well-known sanctuary in the fifth century B.C. Unfortunately, 
Drs. Wilhelm and Reichel did not meet with the success that was hoped, as 
the site had already been plundered by private persons. Even in such an 
out-of-the-way spot as this, this danger has to be faced by intending 
excavators, and it is but few who are lucky enough in Greece to find an 
actually virgin site. Little else is to be noted from the Peloponnesus in the 
way of excavation or discovery during the past year. The Museum at 
Olympia has yielded up its bronzes to the National Museum at Athens, but for 
want of a resident ephor has not yet been completely arranged. The museums 
at Sparta, Megalopolis, and Tripolis are becoming more and more worthy of 
visit. At Argos, however, failing a separate building, few things are allowed 
to remain in the Demarchy, and the sculpture fragments from Bursian and 
Rhangabe's excavation at the Heraion, as well as the American finds, have 
been transported to Athens. It appears that the Athena statuette, the 
importance of which has been emphasized by Mr. Cecil Smith, has not yet 
been taken to Athens from Patras, and the fine mosaic in the square at 
Patras should certainly not be left covered up : if it cannot be taken to 
Athens, the citizens of Patras ought to take proper care of it in their own 
town, but as they have allowed their Roman Odeion to be stripped of nearly 
every fragment of marble it possessed, it is clear that they are not to be 
trusted with any antiquities, and that all should be removed to Athens if 

In North Greece, owing to the unsettled condition of things after 
the war, very little has been done. Mr. Arthur Hill of Athens is endeavouring 
to have the Thessalian marble quarries worked once more, which it appears 
are by no means exhausted. In the course of the hasty works, which were 
thrown up for the protection of the Greek army at Thermopylae, when a 
Turkish advance was imminent, the remains of an ancient watch-tower ^ were 
discovered guarding a path over Callidromos, possibly the very path by which 
Ephialtes led the Persians, and one kilometre south of the warm springs an 
ancient cemetery was discovered, as to which particulars have not come to 
hand. But the archaeological centre of interest on the mainland has been 
Aetolia. Considerable interest has been directed lately .to this little-explored 
country, and particularly to the site where all recent authorities have agreed 
to place Thermon, namely at the Palaiobazari of Kephalovryso, east of Lake 
Trichonis. The latest discussion of the question is to be found in Mr. 

' This was described as a small Doric temple in the newspapers at first {Ath. Mitt. xzii. 1-2, 229). 


Woodhouse's Aetolia, where it is conclusively settled ; and it only remained 
for excavation to confirm his view. This has now been done. The Greek 
Archaeological Society has been digging in the sacred enclosure or Altis, and 
found the great hall of assembly, numerous statue, and inscriptions, but 
the destruction wrought by Philip V. seems to have been very thorough, for 
apparently the site has remained undisturbed since then. It is, however, 
possible that subsequent digging will result in the find of some, at any rate, 
of the statues which were dedicated here in such numbers. Everywhere, 
however, in Aetolia there is little actual ' plunder ' to reward the excavator, 
so far as present experience goes, and if Thermon proves a blank, 
no other site is as likely to tempt future workers. In another 
part of Aetolia Dr. Herzog and Dr. Ziebarth conducted on behalf 
of the German Institute a short excavation this spring, in the little theatre 
at the kastro of Irene, an hour and a half from Mesol6nghi, the site 
generally identified with New Pleuron. Unfortunately there was not a single 
inscription found which in any way proved this identification to be true. 
Nevertheless, the conclusion, accepted by Mr. Woodhouse in his book on 
Aetolia (p. 119), may still be regarded as highly probable. It will be remem- 
bered from Mr. Woodhouse's description and plan that this, the smallest of all 
Greek theatres preserved, abuts closely on the town wall just where a 
quadrangular tower juts out from it. As the distance from the furthest point 
of the orchestra circumference to the wall is only ll'lS metres, it was natural 
for Mr. Woodhouse to disbelieve in the existence of a proskenion : neverthe- 
less Dodwell was right, for the stylobate of the proskenion with one door in 
the centre has been discovered by the German excavators. From the back of 
this to the town- wall is only a distance of 1"94 metres. The floor of the 
orchestra is the natural rock slightly levelled ; one better seat was discovered 
and fifteen rows of ordinary seats. Dr. Herzog thinks that owing to the 
small dimensions of this theatre it has an important bearing on the stage 
question, and certainly it must have been difficult to see the actors on an 
elevated stage from the lower seats of such a theatre as this. The evidence 
supplied by this theatre will clearly have to be carefully examined when 
published. No further light was thrown by the excavations on the curious 
constructions described and photographed by Mr. Woodhouse (p. 121) and 
locally called crraU (pvXaKah, and whether they were corn magazines, cisterns, 
dungeons, or what purpose they served, must remain an unsolved problem. 

At Delphi little or no work has been done this year. It is announced 
that the Athenian banker, M. Syngros, has undertaken to build a museum 
here as he did at Olympia, and as the temporary building is more than full 
already, and not by any means waterproof, this is good news, though it is to 
be hoped that the building will be unobtrusive. With the finishing of the 
work at the Stadium, the chief part of the French School's great work is over, 
and in the Bulletin (1897, Sept.-Oct.), we have three provisional plans giving 
some idea of the vastness of the work. It remains to pull down the chapel 
of Hag. Elias built over the foundations of the Amphictyonic avvihpiov, which 
are plainly visible at the north-west end of the site on the spur overlooking 

336 G. C. RICHARDS. 

new Kastri, and to clear the ground south of the Arachova road, where 
stood the gymnasium and other buildings mentioned by Pausanias (x. 8, 6). 

The chief scene of archaeological activity in Greece Proper this year has 
been the islands. The important vase find at Aegina has been made known 
by Dr. Pallat's publication in Ath. Mitt, xxii., pt. 3. Whether Aegina ever 
possessed a local industry which rivalled those of Corinth, Athens, or Chalcis, 
is an interesting question. Dr. Pallat does not deny to Aegina a local fabric, 
and thinks it lasted on in spite of foreign importation. That there was a 
fabric of vessels for household use is proved by Herodotus, but Dr. Loeschcke 
denies a separate art-fabric to Aegina, and Dr. Hoppin takes this view too. 
The so-called proto-Corinthian ware seems to have been manufactured in the 
Argolid, and it is this which is so largely represented in the Aeginetan find. 
At the same time, it must be plainly confessed, that this question cannot yet 
be regarded as settled. The publication of the American finds at the Heraion 
will do much to throw light on these early vase fabrics. The work done by 
Drs. Hiller von Gaertringen and Dragendorff in the island of Thera does not 
fall into this year, but Dr. Tsountas excavated last summer in Paros, Naxos, 
and Despotiko. He found there a quantity of pre-Mycenaean ware resembling 
that of Pelos, some cornelian beads, and one piece of gold jewellery in the 
shape of a stephane. His excavations bear out the view that the early 
incised ware is contemporary with the Cycladic marble figures, and is distinctly 
prior to the painted pottery. 

This summer an interesting work has been going on at Paros. It 
will be remembered that early in 1897 a fragment of the Parian Chronicle 
was discovered in a piece of ground close to Paroikia, the present 
capital of the island of Paros. The stone ^ contained the account of the 
years from the death of Philip to 299 B.C. (thus supplementing the 
Oxyrhynchus papyrus which contains the chronology of the period 355-315 
B.C.), and as a period of 19 years separated it from the Ashmolean marble and 
the complete stone went down at least to 264 B.C., it was natural to hope 
that the remainder might be found. Dr. Rubensohn has been conducting 
excavations on the site of ancient Paros this summer on behalf of the German 
Institute. Though the Parian Chronicle has not been further enlarged and 
only a lucky accident may bring further fragments to light, these excavations 
have been singularly successful. The Temeuos of Asklepios has been found 
close to the sea containing a spring, a quadrangular court with an altar in the 
centre, many architectural fragments and inscriptions, and a round archaic 
basis of Parian marble on which an inscription contains the name of 
Mikkiades and the letters Phoib — . That this is the Chiot artist and that 
the statue was a dedication to Apollo seems extremely probable. In fact 
another of the early ' Apollo ' figures was also found complete with the 
exception of the feet. Of this statue we hear that the archaic smile is very 
pronounced and that its execution is comparable for care with that of the 
Apollo of Tenea. It is to be hoped that the circumstances of the find may 

' Ath. MUt. xxii. 1, 2, p. 183 (Krispi and Wilhelm). 


throw some light on the question of the identification of this early nude 
male type. Dr. Rubensohn has also laid bare the foundations of the town 
walls on the three sides facing the land, which are preserved in part to the 
height of a metre, and pronounces them to be of fifth century construction. 
He has also identified a sanctuary, which he discovered on a neighbour- 
ing height, as the temenos of Aphrodite, and on a site underneath the 
summit another as that of Eileithyia. Finally it i.s satisfactory to learn that 
a museum has been formed for the reception of the Parian finds and that the 
inhabitants have brought to it antiquities previously in their possession. If 
this sort of thing had happened twenty or even ten years ago in various parts 
of Greece, how much barbarous destruction of ancient monuments would have 
been avoided. A museum has also been started at Vathy in Samos, The 
excavation of the Heraion by Dr. Sarre, to whom permission has been granted, 
has apparently not yet taken place. Lastly, there has been the second 
season of the British School's work in Melos. The work will be described 
elsewhere ; it can here be said without exaggeration that in Phylakopi a 
second Troy has been found. Nowhere else, except at Hiss^rlik, have the 
Mycenaean and pre-Mycenaean strata been discovered in so little disturbed 
a condition. Three distinct strata representing different settlements have 
been found practically undisturbed, and the finds of pottery especially will 
when worked up throw much light on early Aegean civilization. It may 
be fairly said that no excavations of the past year in Greece have been 
more interesting or valuable. 

It remains to say a few words about archaeological work in Asia Minor. 
Besides the explorations of Prof. Kbrte in Phrygia, and Mr. Anderson in 
Galatia, there have been two important excavations. The Austrian Institute, 
under the guidance of Prof. Benndorf, has been conducting excavations at 
Ephesus, of which a provisional report has appeared from Prof. Benndorf 
and Dr. Heberdey, first in the Anzeiger dcr philosophisch-historischen Klasse 
dcr Kais. Akademie der Wissenschaften in Wien, 1897, n. v,-vi., 1898, n. 
vii.-viii., and repeated in the first number of the Jahresheffe of the 

The first work was carried on in some ground close to the scene of 
J. T. Wood's labours, on the north and west of the Artemision, outside the 
wall which marks off the ground belonging to the British Museum. On the 
north side no fragments of the temple-building were found, on the west it was 
hoped that the altar might be discovered. This was not the case, but at a 
distance of about 55 to 60 metres from the west front of the temple a 
pavement of polygonal marble blocks, with sculptured fragments and potsherds 
of the sixth and fifth century, was brought to light. Benndorf thinks it 
likely that the altar lay nearer the temple under the earth heaps thrown up 
by Wood's excavation. Similar trial excavations on the presumed site of the 
Lysimachean city, Avhich were superintended by the lamented Humann, 
showed that digging on a large scale would be abundantly justified. From 
April to December, with a respite in the two hottest months, the work was 
carried on, chiefly under the supervision of Dr. Heberdey. The 7l) m^tre square 
H.S. — VOL. XVIII. ^ 

338 G. C. RICHARDS. 

marble-paved court surrounded by a colonnade with halls and rooms opening 
upon it, and entered by a propylaion on the east side, could be nothing but 
the Hellenistic agora, which seems to have been mainly destroyed by the 
Goths in 263 A.D. and only partly rebuilt subsequently. The theatre is 
pronounced to show three periods of construction, one of the time of the 
foundation of the city, one rebuilding about 150 A.D., and later restorations 
A round monument on the adjacent hill, of which the architectural members 
were found in such nximbers that it could be restored in plan, is conjectured 
to be a trophy of victory, possibly of the Ephesians at Kyme over the 
pretender Aristonicus in 133 B.C. Important finds have been made, including 
a bronze statue of a nude youth, over life-size, which seems to be an original 
of the later Attic school, and a free reproduction of the Attic work of the 
5th century representing an athlete anointing himself, of which copies exist 
in the Uffizi, Louvre, and Vatican, a marble statue of a seated naked boy 
with a duck, suggesting the works of Boethos, a group in black basalt 
representing a sphinx tearing a nude youth, which can be paralleled with the 
centaurs by Aristeas and Papias of Aphrodisias in the Capitoline Museum, 
and a beautiful bronze incense-vessel in the shape of a candelabrum. The most 
interesting inscription mentioned is a letter of M. Aurelius and L, Verus 
ordering that the statues of previous emperors shall not be remodelled to 
represent themselves. 

Since 1895 the important excavations of the Berlin Museum at Priene 
have been progressing, first under the supervision of Karl Humann of Per- 
gamon fame, and since his death of Dr. T. Wiegand, his successor in his post. 
They have resulted in the laying bare of another Pompeii. Priene lies on the 
lower terraces of Mykale and is crowned by a lofty Acropolis, which could only 
be reached from the town by a steep path and on the south falls away so 
abruptly that no wall of defence was required. The town is divided by a net- 
work of streets crossing at right angles into about seventy nearly equal 
rectangles. The only departure from this absolute regularity occurs in the 
agora, where more open space was required. The difficulty of laying out a 
town so regularly on such an uneven hilly site must have been enormous, yet 
no labour of cutting into the rock or terracing up was avoided, in order to 
preserve this absolute uniformity of plan. Below the Temple of Athena, so 
well-known from the Dilettanti Society's publication and the excavations of 
Pullan, in a distance of 30 metres at least 1000 cubic metres of rock had to 
be removed to give the main street from the west gate a more practicable 
slope upwards to the agora. The streets are carefully paved with blocks of 
breccia, but there are no raised pavements at the side. Down the centre of 
the street generally runs a fresh water channel connecting with the private 
houses and fountains at the street-corners. Each of the rectangles surrounded 
by streets is divided into four parts, each quarter being the superficies of a 
dwelling-house. The whole town must have been built at one period and 
according to a single plan, which was no doubt conceived in the age of 
Alexander the Great. It is interesting to see how the traditions of Hippodamos 
were carried on in Asia Minor. The plan of the private house is generally 

ARCHAEOLOGY TN (JllEECE, 1897-8. 339 

that of a rectangle with a lour sided court in tlie centre surrounded by pillars 
with the rooms opening into it on all sides. The front to the main street is 
a blank wall, the door that gave access to the court is in the side street. 
The decoration resembles what is called the first style at Pompeii, i.e. real 
architectural ornament consisting of half-columns, triglyph friezes and cornices, 
with actual sculptured additions, applied to the surface of the wall. Of the 
public buildings discovered, apart from the temple of Athena, the chief were 
the Asklepieion, a rectangular but theatre-like building, probably a place of 
assembly or Bouleuterion, the Prytaneion, and the Theatre. In the latter 
the stage buildings are wonderfully preserved. The front row of seats, a plain 
marble bench with back, is interrupted by five marble thrones. If the circle 
of the proedria be continued it touches the front of the proskenion. An altar 
was actually discovered, not in the centre of the orchestra, but close to the 
front row of seats. It bears an inscription of the third century B.C. The 
complete publication of these extraordinarily interesting discoveries will be 
eagerly awaited. 

Q. C. Richards. 

z 2 

Some Corrections and Additions. 

(See vol. xvii, p. 396 ff. and xviii., x>' 81^.) 

In my second article on Phrygia, Vol. xviii., pp. 101 and 1U9, mention is 
made of some Latin inscriptions which were reserved until the stones should 
be re-examined. I had the opportunity of seeing them again this summer, 
and I now publish them, together with a few corrections and additions to 
both papers. 

Vol. xvii., p. 401, 1, 18. Read ' the small river nearest hut 07ie io Sarai 
Keui ou the east ' ; see the inset map, PI. IV. 

P. 418 ff. This inscription is discussed by A. Schulten in a paper entitled 
Libello dei coloni d\tn demanio iniperialc in Asia in Mitth.dcs Instituts, Rom. 
Ahtheil., 1898, pp. 231-247. My restorations, which had to be made very 
hurriedly and with an inadequate knowledge of the sjDecial literature of the 
subject {e.g. of the important inscription of Skaptoparene, which throws so 
much light on our document^), were merely tentative suggestions; but 
it seemed better to publish the inscription at once for the benefit of 
scholars than to hoard it up until an exhaustive commentary could be written 
on it. But while Schulten's restorations are generally a great improvement 
on mine, they frequently pay scant regard to the conditions ; e.g. in lines 14 
and 15 he restores twenty-one and thirty-four (or more) letters, where the 
stone can hardly have had more than thirteen and fifteen. The main part of 
the inscription, 1. 4 ff., is engraved on a sunk panel and the lines were there- 
fore of equal length. 

I should state that the reproduction given on p. 418 does not quite 
accurately represent my coj^y in some slight details ; and I take this oppor- 
tunity of giving an improved reproduction of some portions of the inscription, 
made directly from an impression. I also add some notes on the text and a 
few corrections, which a careful re-examination of the stone has rendered 

L. 1. The names were no doubt erased, as Hulsen supposes (cqh Schulten, 
p. 233 n.). They seemed to me to have been merely worn away, but erasures 
not infrequently have this appearance. 

2. PER was doubtless intended, but R is certainly not on the stone. The 
stone-cutter did not understand Latin. 

rublished ill Zfl. der Savujnysti/tung f. Rechtagesch., xii. p. -246. 


Hiilsen's suggestion proconsule v(ir) c(larissimu3) is probably correct. 
[It occurred also to Mr. F. Haverfield.] 

3. The fourth letter is probably a cursive d. The same form 

of letter stands for h in dabit (1. 26, see below), and the /< of ^ 

revocahit is represented by a similar letter. !D f^ 

The latter of the two enigmatical letters at the end of the K^ 

line is not M but A (which has become blurred in the repro- tz i 

duction). ^ — 

4. The second 6 in EYCEBEI is of semi-circular form. ^ ^ 

5. Read ErAEkT. [Schulten is wrong in supposing that uj \^ 
the K was omitted by the engraver and then inserted above the ^^^ O 
line.] V— ' Z^ 

J< ^ 


7. Read MOtT€ANWN. A re-examination of the stone ^ ^ 

revealed traces of a letter between O and T, and the impression ^ -^ 

shows it to be in all probability a T. The space is narrow, and t., ^ 

evidently the engraver had omitted it at first and then inserted it. ^ |^ 

This improved reading confirms the correction kolv(ov T)oTT€av(ov. ,, ^ 

10. Read TWsl, as suggested in the note (p. 420). -^ 3 

15. MITE ; there seems to be no ligature between M and I. ,Z^. ^ 

20 fin. As I suspected, when editing the text, I have omitted H 

a syllable. The correct reading is BOACANf. Py \^ 

24. The last letter is certainly E. O i!!!y 

26. The outline of the blurred sixth letter is <^, which ^ \0 

probably stands for h, as in revocahit. Mr. Haverfield tells me he iD 

has come across a similar case. The form /^ which occurs on ^ '^ 

coins, e.g. of Olbasa {B.M. Catalogue of Lycia, etc. pp. 229, 230) is ^ ^ 

more intelligible. Q^ i 

30. The first letter is € (it cannot be 0). "X ^ 

31. Read EPmoYCeAI. LU ^ 
The following articles may be recommended to the atten- ^ \H 

tion of those who are interested in the Imperial Domains : — ^ UJ 

Toutain, L'inscription d'Henchir Mettich, in the M6moircs pr6- '^ Q 

sent4s par divers savants a I'Acad. des Inscr. et Belles-Lettres, t. xi. .< ^ 

1897.1 ^ 2 

Cuq, Le colonat partiaire, in the same volume. ^ ^ 

Beaudouin, Les grands domains dans I' Empire Romain d'apr^ v>_ -r- 

des iravaux r^cents in the Nouvelle Revue histm: du droit, 1897, pp. ^ — 

543 ff. and 673 ff. ^ ^ 

Schulten, Die lex Manciana in Gottingen Ahhandlungen, 1897. ^ 

P. 424, No. 22. A certain Optimus was proconsul of the pro- ^^^- 
vince of Asia in 250-1 A.D. He is described a^ proconsul apud Asiam (or apud 
Asiam provinciam) sub Decio imperatore (249-251 A.D.) in the Acta /S'. Maximi 

' Pjiuted also in Nouv. Revue hist, du droit, 1897, p. 373 ff. 



Martyris (Ruinart Acta Sincera, p. 157), and in the Acta S. Martyris Petri, 
Andreae, etc. (Ruinart, p. 160), and Waddington assigned him to 250-1 A.D., 
inasmuch as the proconsul of the preceding year is known {Pastes den prov. 
Asiat., No. 176). Dr. Dessau has recently (]uestioned the authority of these 
Acta, and expressed the opinion that the proconsul's name is corrupt and that 
the real name was perhaps Aristus, which became Optimus in the Latin 
translation of the Acta {Prosopographia imp. Rom. p. 437). But our inscrip- 
tion shows that there is no reason to doubt the name and distrust the Acta 
on that score. We are indeed tempted to identify the proconsul of the Acta 
with the Flavins Optimus of the inscription. But, as Prof. Ramsay has 
shown in the Expository Times (August 1898, p. 496) the identification is not 
possible; for Fl. Optimus 'bears the title 8ia<jr)/j,6TaTo<;,pcrfcctissiimis, which 
marks liim as belonging to a lower grade of governors, and to a later era. In 
250 A.D. the governor of Asia had the rank XafivpoTaro^i, clarissivius ; and 
the inscription must be referred to the fourth or fifth century, when Asia had 
been broken up, and Meiros was part of Phrygia Salutaris, administered by a 
praeses perfcctissimus.' [He goes on to show, with much plausibility, that 
apud Asiam, with the variants ajmd Asiam prov., in Asia civiiate, apud 
Amhiensem provinciam, is probably a corruption of ajmd Apiam.] 

Vol. XVIII., p. 87, No. 24. The most probable restoration of the name 
of the KUTotKia is perhaps ["0]X-/3o(9, a proposal first made to me by Prof. 
Ramsay and recently repeated by Dr. Korte in a letter. The name is well 
known in Cilicia Tracheia, and appears in Pamphylia or Lycia (Steph. Byz. 
8.V. Olhia). I had introduced this restoration at first, but I afterwards 
cut it out, as there are other possibilities and certainty is, therefore, not 

P. 101, 1. 14 f. The inscription in question is now in the village of 
Oktchular. The letters are mostly quite clear, and I did not re-examine the 
stone this year. My copies of 1896 and 1897 agree. 

Fines loci q[u- 
evi ex pialudc 
conduci covf[ir- 
5 iii\ateio ? vicano\i'- 
w]?7i PoIyntc7io[r - 
um Hermioni\o- 
r(^u)m [Moragocom- 
es? . . 



The interpretation of tlie first line was suggested by Mr. Ilavcrfield, tliat 
of 1. 7 f. is due to Prof. Ramsay, amfirmatiu in 1. 4 occurred to all three of us. 

Another inscription, exceedingly worn, now at the village Alp Arslau, 
seems to be a companion stone. The difficulty of deciphering it is greatly 
increased by the roughness and irregularity of the engraving. 

I L(((((\\Hi,y\\\c//., 


Fines [luci quern 
ex p]a[/]ii[(fe .... 

v]icanoo'U7a [Folynte- 
5 iwrum Hermi\oneo- 
rum Mo[r]agoc[om- 
csl condv[x]e[runt} 

The following remarks embody suggestions made by Prof. Ramsay. The 
palus is clearly the marshy part of the valley beside Oktchular, the lowest 
point in the plain, where in winter the water of the land-locked ova collects 
and forms a lake, which becomes a marsh again in spring and dries up entirely 
in the height of summer (cf. CB. ii. p. 747). It seems probable that we 
have here the boundary stones of another Imperial Estate, stretching from 
Oktchular to Alp Arslan and containing two or three vici{Caesaris), the first 
of which is Polynta and the second the Hcrmo-kome of the Tekmorian Lists 
(Sterrett, Wolfe Exp. 375, 10, Ramsay, Hist. Geog. p. 412), while the third is 
uncertain. The inscriptions seem to refer to some kind of union of these 
vici. For the other Imperial Estates in Asia Minor whose existence has 
been proved by Prof. Ramsay, cf. Schulten, Rom. Mitth. I.e. pp. 221-231, 

P, 109, 1. 8. The inscription is cut on the rocks on the hill-side, just at 
the point where a causeway and a low bridge carry the road over the marsh.^ 
It contained four lines. A second trial enabled us to read the name Fronto 
quite clearly, 



mill I I im^^^'''^ " 

C. Caristlanlilitls 

Fronto cos.(l) . . . 
. c]lusu[ml . . . . 

. . . aperuit. 

^ My first copy has ERVN, which is clearly 
right (and may be traced on the impression). 
- In the map (Plate V.) the hills (Kara-kush 

Dagh) are inaccurately represented ; the slopes 
run down quite close to Armudli and they touch 
the north-east edge of the marsh. 


It is unfortunate that this inscription is so badly worn. Gaius Caris- 
tanius Fronto was Jegoitis Av.g. pr. ;r/'. of the province Lycia-Pampliylia in 
the reign of Domitian (Steirett, Epigr. Journ. No. lOH: cf B.C.H. l&SG, 
p. 46, No. 2). At that time he was of praetorian rank. After his consul- 
ship he evidently became jyroconsv! of Asia. 

In the fragmoTitavy state of the inscription, it is impossible to say what 
was the precise nature of the work undertaken by Fronto. It may be noted, 
however, that while the enormous marsh which now fills the centre of this 
jtlain is almo.'st entirely the result of Turkish neglect, there was probably 
always a certain amount of marsli near the point where the inscription is 
engraved ; for there is no possible exit for the large volume of water which 
Hows down into the plain from the copious springs at its .south-west edge 
(below Geneli). Perhaps, then, the accumulation of water at this point had 
made an impas.«able barrier, and Fronto opened a new road over it. 

P. 112, No. 52. In 1. 8 read Tr,v ar,v olKTpo[7ci]jr)i'. In 1. 9, probably 
761/67; hi fjioi. In 1. 10 there appears to have been one letter, apparently a A. 
between (piXoi and edcvro ; rpo4)r]e<; seems certain. 

P. 113. No. 53 his. M. Franz Cumont points out to me that the two 
symbols are anchors (cp. Kraus, Rcalcncyc. der Christ. Altcrthinn. s.v. Ancora), 
thus adding one more confirmation of the theory (already well proved) that 
the formula eaTai avrw irpo'i top 6eov is Christian. This seems to be the 
only example of the anchor known in Phrygia. 

P. 123, 1. 8 from the foot. Delete 'perhaps also in yiaiKidi'T)<i,' etc. 
MaiKiavrj is the Latin Maeciana, as Prof. Ramsay points out in Classical 
Review, 1898, p. 342, n. 2. 

P. 128. A letter from Prof. Ramsay asking me not to publish his note at 
the end of my second article failed to catch me, as I had left England for the 
East. His note was written under the impression, derived from a too hasty 
perusal of my proof-sheets, that his view in Hist. Gcogr. differed from mine 
both about Kaballa and about Tzibreli-tzemani. In reality he inclined to the 
same conclusion about the latter place a? that which I had reached ; see his 
note -f* on p. 359. 

J. G. C. Anderson. 


It is with personal reluctance that I, like Mr. Grundy, again take up the 
argument. It is probably, however, better for the reader if the two side.s 
thrash it out. The main reason why so many of the full-dress debates of 
archaeology, on the Xoyeiov for instance, or the ' Old Temple,' are still 
obscure for the non-combatant, is because the protagonists seem to get bored 
with each other's arguments, and pass tliem over in silence as self-evident 
fallacies. The result is endless, indecisive .summarizing by those not imme- 
diately concerned. 

In the present number of the Jf)urnal, p. 234, Mr. Grundy expresses hi.s 
astonishment that the walls on Pylos and 8j)hacteria, as photogi-aphed in 
Plates VII , IX. and X., of the preceding number present no more definite 
marks of date. This is surely a failure to recognize the conditions of the 
problem. It is a question of "What went ye out for to see ? " Did Mr. Grundy 
expect for the TraXaiov epvfia a master-piece of Mycenaean splendour, 
a Lion's Gate or a Tirynthian Gallery ? Did he hope that an obliging 
Athenian sailor had marked wall L on Pylos with an o = cu ? The remains 
exactly answer to Thucydides' description (see J. U.S. xvi. pp. GO, n. 40), and 
the fact that wall BB of the iraXatov epv/xa (Figs. 2 and 8) strikes Mr. 
Grundy" as the §ame in character as wall L (Figs. G and 7), though less 
perfect, is an evidence /or my identification and not against it. Thucydides 
uses the same distinguishing word for both ; XiOatv XojdSrjv TreTroLrjfiivov 
in iv. 31, 2, XoydBrjv <^epovT€<; Xc6ov<; in iv. 4, 2. 

Mr. Grundy then, if his theory of Cumberland sheepfolds be right, can 
only say that any such wall as Thucydides describes would defy dating. 
This he should have said two years ago, and not only after seeing my 
photographs. What he means by now. saying (p. 234) that he has " seen " 
the walls marked Figs. 2 and 8, it is difficult to determine. He has not been 
to Pylos since 1895. We have both hitherto assumed that he then either 
did not see the walls, or mistook them for the .stratification of the limestone 
rock (C.E. Nov., 1896, p. 371. Feb., 1897, p. 2). Is it possible that this is 
not so, and that Mr. Grundy realized that the remains which Dr. Schliemann, 
Mr. Bosanquet, and Mr. Crowfoot have accepted as the TraXaibv epv/xa were 
walls, but thought them so undoubtedly modern that they were not only not 
worth arguing about, but not even worth mentioning ? Th« second altern- 
ative is more damaging to him than the first. 


Must we acquiesce, liowever, in the theory of sheepfolds ? Certainly 
not. Wo must consider under what circumstances walls of this XoydSrjv 
character, yet eight feet tliick as wall L, two metres as wall BB, three metres 
as wall C (J.Il.S. xviii. pp. 153, 156) could have been made. It is certain in the 
first place from their position as well as their thickness that one and all of 
them were military walls, and not houses or sheep shelters. This does not 
carry us far, but it clears the ground. In the second place it shows want of 
discrimination to say of the walls on Pylos and Sphacteria that " either 
might belong to any age " (p. 235). They must be treated separately. The 
case for Pylos is a strong probability.^ For Sphacteria it is a practical 

To take Pylos first, then, what military operations could have taken 
place there of which wall L would be a normal result ? I have already 
applied this test to later classical and mediaeval times {J.H.S. xvi. pp. 66-67), 
and will not multiply proof. If it be urged that the building of wall L in 
the fifth century B.C. would have been as abnormal as the building of it at 
any subsequent time, I can only reply that Thucydides tells us that the 
Athenians built abnormally, and why they did so, and that it is indeed an 
example of ' another man of the same name,' to turn from an actual 
abnormal event for the occurrence of which you have detailed reasons 
given, to an hypothetical one of which you know absolutely nothing. That 
the survival of an early rough wall is no difficulty has been convincingly 
proved by Mr. Bosanquet (J.H.S. xviii. p. 156). 

The War of Independence, and the historical fact of the siege of Palaeo 
Kastro (our Pylos) in 1825 is, indeed, another matter. 

It is true that the Greek Insurgents were without knowledge of 
building, and therefore somewhat in the position of the Cumberland 
shepherd. It is true, too, that they were given to erecting temporary cover 
in the shape of ' tambouri.' 

It is possible also, though not probable, that their line of defence was 
here.^ But though they may have used wall L, it is almost inconceivable 
that they built it. In the first place so serious a piece of work could 
scarcely have been carried through without constant use of the masses of 
squared stones that stood near at hand. Secondly, ' tambouri ' were flimsy 
cover for marksmen, not great continuous walls eight feet thick. But — 
final, and, I think, conclusive reason — Bory de St. Vincent {Rdation, p. 155) 
saw • un tambour ' of 1825 during his visit to Pylos in 1829. And it is Bory 
who was convinced of the very ancient character of wall L. {J.H.S. xviii. 
p. 156). 

But if it is difficult to imagine any circumstances in which wall L on 
Pylos could be so naturally built as by Demosthenes, the evidence for identi- 
fication is quite another matter when we turn to Sphacteria. On 
Sphacteria it is no longer merely a question of a style of building normal 

' See J.H.S. xvi. p. 68, n. 47. de St. Vincent, Atlas, Plate IV. Rilation, 

' See James Emerson Diary, \k 149. Bory p. 155. 


only at one or two epochs. We have a j^Tound plan which is conceivable only 
under particular conditions, and we possess detailed information as to military 
operations in that very period of history where style helps ua least. None of 
the Sphacteria walls could have been built durin<,^ the Greek War of Inde- 
pendence. We have definite first-hand information in the pages of Collegno 
and Millingen as to what was done and what was not done on Sphacteria right 
through the siege of New Navarin up to the day when Ibrahim attacked the 
island, and the Greeks fled headlong before the charge of the Arab infantry. 
Collegno in particular was often on the island, ' cosi gaiae favorita/ turning in 
relief from the squalor and confinement of the besieged town to the ' delizie 
della mia isola ' (p. 46). The Greeks put one or perhaps two small batteries 
in position on the South Point, facing New Navarin, and another at the 
principal — probably the Panagia landing-place. It is certain that they built 
no fort on the North Peak. But was it the Turks who built the fort ? ' It 
was here,' says Mr. Grundy (p. 235), ' that the Egyptians established their 
batteries in 1825 in the attack on Pylos.' Pylos ' was maintained by the 
insurgent Greeks for six weeks against the assaults of Ibrahim Pasha's force ' 
(p. 234). One would hardly imagine from this language that the Egyptians 
were in possession of Sphacteria for less than two full days before Pylos, that 
is Palaeo Kastro, capitulated. Nor that the evidence for their having 
during these two days established any batteries at all on this spot, is so 
shadowy as to be safely neglected.^ As a matter of fact, it concerns my 
argument not at all whether they did so or not. Even if they did post a 
battery on the summit, it is inconceivable that the walls of Mr. Crowfoot's 
plan (J.H.S. xviii. p. 152, Fig. 10) were built by them to protect it. Their 
ground plan precludes the possibility of their being built for an a,ttack on Pylos. 
On the actual summit, the point which does command Pylos, the main frontage 
of wall BB is to the west and not to the north, to the island itself, and not to 
Pylos. The north wall of the Hollow (C) fronts north-west, but is in a worse 
position for commanding Pylos than dozens that could be chosen further 
west. The south wall of the Hollow (D) is meaningless. 

We can put aside then all question of the War of Independence. As 
certainly can we disregard mediaeval or later classical times.^ The only use 

1 The evidence for all thiri would be of little from Collegno, Jourdain (pp. 170-172), who is 
interest here. I have, however, examined the worthless, Gordon (pp. 202-205), who was not 
question with some minuteness, and submitted in Greece at the time, Gouin [L'Egypte mi xix. 
the details to the criticism of several scholars, SUdc, pp. 382-384), Prokesch-Osten (i. pp. 
including Mr. R. C. Bosanquet. In brief, I 353-4), Mendelssohn-Bartholdy (ii. pp. 354- 
foUow primarily the evidence of three eye- 355), Tiicoupis (iii. pp. iJ05-206) and Finlay 
witnesses, Collegno {Diario, pp. 42-62), Grasset (vi. pp. 359-303). The difference between the 
{a^d Emerson, pp. 172, 175) and (with slight Greek and West European Calendars has worked 
exceptions) Millingen (Memoir, pp. 290-310). havoc in the dates. The right dates are mid- 
Emerson is also, with some qualifications, a day Sunday, May 8th (our reckoning), for the 
good source, writing his Diary (pp. 136-151) fall of Sphacteria, and early morning Tuesday, 
from accounts given by a day or May 10th, for the capitulation of Pulaeo- Kastro. 
two after the events. The secondary authori- ^ The battle of Navarino in 1827 gives no 
ties I. have had access to are Pecchio (apud scope for wall building on the north. There is 
Emerson, pp. 109-115), who heard the story mention (Finlay, vii. pp. 17-18) of Turkish 


then likely to have been made of the summit was for a watch-tower, and it 
was pointed out by me in my original article {J. U.S. xvi. p, 63, n. 27) that 
Leake saw such a tower, and by Mr. Bosanquet (J.H.S. xviii. p. 156) that traces 
of its presence still exist. The ground plan of our fort, however, is incon- 
testably not that of a watch-tower. It is a stronghold of an early half- 
civilized people, either pirates themselves, as suggested by Mr. Crowfoot 
(J.H.S. xviii. 153), or irepiKTiove^ in fear of pirates, as suggested by Mr. 
Bosanquet {Ih. p. 156). It was built to face an attack by land with short- 
range missiles, and it guarded against surprise by climbing. The resources 
of its defence were adequate to meet the resources of attack in those early 
days. And then alone could it have been built. Of its fitness to be the 
iraXaiov epvfia of Thucydides* narrative I have already said enough (J.H.S. 
xvi. pp. 59-63. a.E., Feb. 1897, pp. ]-2). 

Mr. Grundy, however, has suggested (p. 235) that the fort has not got a 
plan at all, that the walls do not make up a homogeneous whole. As regards 
the south he may have been misled by the fact that Mr. Crowfoot, as stated 
in the text (J.H.S. xviii. p. 154), did not see wall D, and hypothetically located 
it too far north. To doctor his plan would have deprived it of all value as 
evidence, and to wait till he again could visit the spot would have indefinitely 
delayed publication. The defence on the south is however, in point of fact, 
unbroken. Wall D does not guard the approach from the interior of the 
island at all. It was built to bar the gorge to climbers. It is the south-east 
corner of Avail BB that prevented approach to the south of the hollow 
from the interior (J.H.S. xvi. p. 60). It is possible there was once a wall 
connection for the yard or two which separates the south-east corner of BB 
from the precipice. It is possible that a yard or two of cliff has broken 
down. But there is no difficulty or break of connection. As things stand 
you could only pass within sword distance of the wall. 

Mr. Grundy, however, would answer that the north wall of the hollow 
' shows manifest disconnection ' with wall BB (p. 235). By this he does not 

guDS on the south of Sphacteria, but no men- act of folly of whicli neither Frank nor Vene- 

tion at all of the nortli. If indeed it was at lian would be guilty, an anomaly in tl:e history 

any period thought necessary to bar the entry of Medijeval War. They would' at least have 

to the Sikia Channel from the side of the island, worked as far as possible on the Barbican prin- 

tlic point chosen would for a certaintj- not have ciple, and continued the western walls down to 

been the summit, but the slightly lower hiil on the foot of the Sikia Channel, 
the west, which stands between it and the sea. In point of fact the short range and small 

Not only therefore is a modern date impossible, power of artillery up to 1572, when Neo Kastro 

but the possibilities of a medifeval one are con- (New Navarin) was built, and Palaeo Kastro 

siderably limited. Our walls could at that sank into insignificance, would have rendered 

period only have been built by a force liolding the fortification of the north of Sphacteria an 

I'ylos to prevent an enemy from commanding unnecessary precaution. The great point was 

it. But not only are the arguments as to style to have as small a line as possible open to 

of building as valid here as for wall L, but the attack by land. A breach in the south wall of 

fortification of a small detached outwork, too Palaeo Kastro, which could only be reached by 

weak to resist any foicc that could hope to sea, would be as nothing compared with the 

attack the Great Castle, without a water supply, exposure of a large part of the garrison to 

without means of communication, would be an annihilation by land. 


merely mean that the evidence for the existence of remains of the connecting 
wall is not conclusive. He means, as his further words show, that the 
builders of wall BB would not have built C at such an auglo. They would 
have run it to meet the north-east corner of BB, so that tliis odd receding 
angle would never have been necessary. 

Yet it is this very point which gives as clear an indication of date as the 
XoydBrjv character of the walls. Where was the gate oi' the fort ? Clearly, 
as Mr. Bosanquet suggested to me many months ago, at the apex of this 
angle. It was a principle with the makers of early fortifications that the 
approach to a gate should be covered by two walls, that the enemy should be 
open to attack from both sides. The builders of tiie naXaiou epvfia were 
doing with their simple resources what was elaborately provided for at Tiryns 
and Mycenae.^ That the principle was not put out of date by the knowledge 
of flanking towers we sec from the pre.sence of such towers in both these 

On other points I have little to say. Mr. Grundy now (p. 228) lays far 
greater stress than he did originally (J.I/.S. xvi. p. 11 par. 1, p. 12 par. 4) on the 
physiographical evidence for the dating of the southern sandbar. Nothing 
but the decided opinion of an experienceil geologist would warrant us to 
treat this as a case where historical and arclueological argument must bow 
before a law of nature. We have no such opinion. Meanwhile it must not 
be forgotten that the Boidia Koilia sandbar was admittedly formed first. 
This being so, Mr. Grundy cannot expect us to treat as a scientific certainty 
for the fifth centuiy B.C. the following remarkable combination of unprovable 

(1) The lagoon must have been navigable right up to Boidia Koilia, so 
that engines could be landed there. 

(2) The southern sandbar must have reached to within about 200 yards 
of the south-east corner of Pylos. 

(3) No sand could have drifted to this south-east corner. 

In regard to the south-east corner Mr. Grundy dwells (p. 233) on the 
misleading character of photographs to those who have not first-hand know- 
ledge of the ground. He should notice that Mr. Bosanquet had such first- 
hand knowledge {J.H.S. xviii, p. 158), and that nothing could be more definite 
than the confirmation which he gives on this point to my views. The reason, 
of course, that Demosthenes' ships were not destroyed (p- 232), is the one 
given us by Thucydides (iv. 9, 1), that he protected them by a stockade. 

On p. 233 n. 1 Mr. Grundy makes another of his unfortunate charges 
of misunderstanding. I can only ask the reader who cares to take the trouble, 
to compare Mr. Grundy's remarks in CM., Nov. 1896, p. 372 col. 1, my 
answer in C.B., Feb. 1897, p. 2. col. 2, p. 3 col. 1, his quotation of them 
C.B., April, 1897, p. 156 col. 1 /in. and col. 2, and my quotation of that quota- 
tion in J.H.S. xviii. p. 149. He will see that the misunderstanding is not my 

* Schuchhardt's Schliemann, Hag. Tram., pp. 103, 104, 105, 132, 138, 298. Frazer's Pau- 
mnia.t, Vol. ii. pp. 100, 221. 


fault. But he will also sec that we have both been guilty of obscurity in 
describing this south-east corner. It should be thought of, not as two distinct 
slopes, but as a shoulder sloping two ways, to the sandbar on the east, to the 
Sikia Cliannel on the south. A glance at Figs. 1 and 4 (J.If.S. xviii. 
Plates VII. and VIII.) will extricate the reader from a muddle into which our 
language may easily have led him. 

Ronald M. Burrows. 







Abbreviation in inscripliuTis, xviii. 304 

Abretteno?, Zeus, xvii. 289 

Achaeans, late entry into Greece, xviii. 

xxxlv. Fair-haired, xviii xxxv. 
Acropolis on coin of Anazarbos, xviii. 162 
Adriania (Mysia), xvii. 290 
Aegean Script, xvii. 327 ; relation with 

Aegina, antiquities from : Mycenaean bronze 

knife, xvii. 65 ; Mycenaean marble pyxi.s, 

ibid. ; Mycenaean pottery, xvii. 77 ; find 

of vases, xviii. 336 
Aegis of Atliena, xvii. 314 ; on coin of 

Soloi, xviii. 165 
Aegium (Achaea), coin, xvii. 82 
Aeschylus, archaeological points in Eu- 

menides, xviii. xlii. 
Aetos, site, xviii. 83 
Africa, North, vine-mosaics, xviii. 69 
Agia Paraskevi (Cyprus), bronze-age ne- 
cropolis, xvii. 134 
Agon on coin, xvii. 80 
Agora at Athens, xviii. 330 ; at Melos, xvii. 

Aidos = Aetos, xviii. 83 
Aigeai (Cilicia), coins, xviii. 161 
Akcheler, nr. lake Apollonia (Mysia), inscr., 

xvii. 269 
Aksakal, fifteen miles from Panderma 

(Mysia), inscr., xvii. 274 
Aksheher, xviii. 113 
Al Baladhuri, xviii. 182 
Alcibiades, psephism of, xviii. 329 
Alexander of Macedon at Plataiai, xviii. 43 
Alp Arslan, inscr., xviii. 343 
Al Tabari, xviii. 182 
Al Ya'kubi, xviii. 182 
Amaltliea and Zeus, xvii. 82 
Amasia taken by Arabs, xviii. 193 
Amazon, wounded, various types, xviii. 142 
Ammonius, quoted in de Sublimitate, xvii. 

Amompharetos at Plataiai, xviii. 54 


-VOL. X.VII1. 

Amon-cult, degradation under Ptolemiea, 
xviii. 239 

Amorgos, Mycenaean gems from, xvii. 70 ; 
oil-press in, xviii. 216 

Amorion, Arabs at, xviii. 183, 192, 204 

Amphiaraos, departure of, xviii. 290 

Amphilochos at Mallos, xviii. 163 ; at 
Tarsos, xviii. 179 

Amphilochos at sacrifice of Polyxena, xviii. 

Amphitrite of Melos, xviii., xli. See also 

Amphorae, types from Graeco-Phoenician 
tombs, Larnaka^ xvii. 160 

Anava-Sanaos, xvii. 413 ; inscr., xviii. 90 

Anazarbos, coins, xviii. 161 

Anchor (Christian symbol), xviii. 344 

Androcrates, heroon near Plataiai, xviii. 38, 

Animals, combinations of, in early art, xvii. 

Antignotos, sculptor, xvii. 325 

Antiochia (Pisidian), Arabs at, xviii. 183 f. 

Antiochia (Syrian), in the Arab wars (641- 
750), xviii. 183 f. 

Antiochus I. of Commagene, xviii. 312 

Antiphates at sacrifice of Polyxena, xviii. 

Aphrodite, Homeric Hymn to, xviii. 23. 
Riding on bull, xviii. 165 ; of Melos, xviii. 
xli. ; of Praxiteles, on coin of Cnidue, 
xvii. 89 ; A. and Eros playing morra, 
xviii. 129 ; with bird, xviii. 132 

Apollo, an Achaean deity, xviii., xxxv. ; 
Homeric Hymn to, xvii. 241. On Nemrud 
Dagh reliefs, xviii. 314 ; ? on votive terra- 
cotta relief, xvii. 316 ; on coins of Lamos. 
xviii. 163 ; of Selinus-Traianopolis, xviii. 
164 ; of Pompeiopolis, xviii. 167-169 ; 
LykeioB or Tareeus on coina of Taraoe, 
xviii. 172-174 
Apollonia ad Rhyndacum, inscr., xvii. 269 
Apollonia-Sozopolis, inscr., xviii. 95 
ApoUonieron, Apollonos-hieron, xviii. 89 

Apotripi spring, xviii. 37 

.A A 



Aquaiousha, xviii., xxxiv. 

Arabs in Asia Minor (ad. 641-750), xviii. 182 

Aratos, portrait, xviii. 167 

Arcadian League, federal buildings confined 

to southern half of Megalopolis, xviii. 18 
Archaeology in Greece, 1897-8, xviii. 328 
Archelaus of Cappadocia, inscr. in honour 

of, xviii. 317 
Archenioros, death of, xviii. 270 
Arkut Khan, inscr., xviii. 122, 127 
Armenia, Arabs in, xviii. 190 f. ; A. minor, 

inscriptions, xviii. 320 
Armudli (Kinnaborion), xviii. 109 
Artemis of Aigeira (?) dedication to, xviii. 

304 ; A.-Hecate, relief from Eraed, xvii. 

281 ; A. in Lacedaemon, dedication to, 

xviii. 302 
Artemisium, battle of, xvii. 212 
Asia Minor, Arab invasions (a.d. 641-750), 

xviii. 1H2. Eastern : road-system, xvii. 

22 f. ; inscriptions, xviii. 306 
Asopos, River (Boeotia), xviii. 56, 236 
Asopos, River (Phrygia) = Gumush Tchai, 

xvii. 405 
Assar Keui, nr. 1. Simav (Phrygia), inscr., 

xvii. 288 
Assarlar, inscr., xvii. 283 
Athena, on votive terra-cotta reliefs, xvii. 

307 ; Ergane, xvii. 309 ; Polias, xvii. 310 ; 

Promachos, xvii. 312 ; Dedication, xvii. 

310. See also Uapdivos. 
Athenian version of battle of Plataiai, xviii. 

Athens, excavations, 1897-8, xviii. 329 
Athlete, head of, belonging to Dr. Nelson, 

xviii. 141 
Atli Hi8sar = Sibidounda, xviii. 105 
Attouda, site and inscr., xvii. 398 
Augusta (Cilicia), coin, xviii. 162 
Augusteum at Constantinople, xvii. 109 
Austrian Institute, foundation, xviii. 328 
Azara (Azari Keui), inscr., xviii. 113 


B, form in Latin epigraphy, xxiii. 341 

Bacchylides, illustrations to, xviii. 267 

Baddelu (Phrygia), inscr., xvii. 288 

Baharlar, inscr., xviii. 87 

Balat (Mysia), inscr., xvii. 290 

Balia Maden (Mysia), inscr., xviii. 293 

Baljik Hissar, xviii. 103, 106 

Balki Keui, inscr., xviii. 119 

Balukiser (Mysia), inscr., xvii. 292 

Basil, campaign of 876-7, xvii. 34 f. ; of 872, 

Basilica at Constantinople, xvii. 1 1 1 
Basmaktchi Yaila, inscr., xviii. 92 
Beyje (Mysia), inscr., xvii. 279 
Bithynia, coin, xvii. 84 
Boreas and Oreithyia on a late Attic vase, 

xviii. 136 
Boul^ of Laodicea, xvii. 408 

Brasidas, rocks of, xviii. 148 

Bria (Phrygia), site and inscr., xvii. 415 

British Museum, antiquities in : 

Mycenaean : gold fibulae (Cyprus), xvii. 
bronze knives (Suria and Aegina), xvii. 
64, 65 ; chisel (Suria), xvii. 65 ; celt 
(Suria), xvii. 65 
marble pyxis (Aegina), xvii. 65 
engraved gems (various sites), xvii. 65 f. 
pottery (various sites), xvii. 71 f. 
B.f. vases : B 160, xviii. 274 ; acijuired 

1893-1898, xviii. 280 ff. 
Bronze statuette (Eros playing morra), 

xviii. 131 
Gold ring (Aphrodite and Eros), xviii. 

Coins, xvii. 78 f. 
Bronze-age necropoleis in Cj-prus : Agia 
Paraskevi, xvii. 134 ; Kalopsida, xvii. 
138 ; Laksha tu Riii, xvii. 147 
Bull : led by two men, on Mycenaean gem, 
xvii. 70 ; goddess riding on, xviii. 165 ; 
bull-Bacchos, xviii. 166 
Bulladan = Apollonieron, xviii. 89 

C (see also K) 

Caecilius, rhetorician, xvii. 193 

Caeneus and the Centaurs, on Harrow vase, 

xvii. 294 ; the meaning of the legend, 

Caesareia (Cappadocia), roads from, to east, 

xvii. 22 f. ; to south, 28 f. ; to Cilicia, 29 

f. ; from Sebasteia to, 31 
Calabria, Mycenaean gem from, xvii. 69 
Calymna, Mycenaean gem from, xvii. 68 ; 

Mycenaean pottery, 75 ; oil presses, xviii. 

210 f. 
Canal of Xerxes, xviii., xxxviu. 
Cantharos, unusual in pottery, xviii. 288 
Cappadocia, Byzantine campaigns in, xvii. 

33 ; inscriptions, xviii. 316 
Carian oil-presses, xviii. 209 
C. Caristanius Fronto, xviii. 343 
Catagusa {KaTayova-a), xvii. 310 
Celt, bronze, from Suria, xvii. 65 
Centaurs and Caeneus, xvii. 294 ; spirits of 

the forest, 300 
Cerigo, vase with inscribed linear characters, 

xvii. 349 
Chalcidice ? coin of, xvii. 79 
Chalcoprateia (Constantinople), xvii.. 111. 
Chelidonia-Diniae (Kara-dilli), xviii. 104 
Chepne, nr. lake Manias (Mysia) inscr. xvii. 

Chisel, bronze, from Suria, xvii. 64 
Choban Tepeh (Tread), tumulus at, xvii. 

Chrysippos, statue and bust on coins of 

Pompeiopolis, xviii. 167 
Chrysorhoas, R. at Hierapolis, xvii. 413 
"Churches, Three," Melos, xvii. 122 



Cicero, quoted in de Siiblimitate, xvii. 195 

Cilicia, inHcriptions from, xviii. 306 ; coin- 
types of Cilician cities, xviii. 161 ; Cilician 
Gates, inscriptions, xviii. 306 

Clariashnus, xviii. 342 

Clazoinenae, psephisni of AlcibiadeB con- 
cerning, xviii. 329 

Cnidus, coin, xvii. 89 

Coins acquired by B. M. (1887-96), xvii. 
78 ; of Cilicia, xviii. 161 

Coloneia, Arabs at, xviii, 184 

Coloni on Imperial estates, xvii. 421 ; xviii. 

Colossae, inscr., xviii. 90 

Comana (Cappadocia) inscr., xviii. 316 

Coniniagene, inscr., xviii. 312 ; passes from 
Melitene to, xvii. 30 

Constantinople, Arabs near, xviii. 184; topo- 
graphy of Imperial quarter, xviii. 109 

Corinth, excavations, xviii. 333 

Corybantes, xvii. 90 

Cotys, portraits of, xvii. 321 

Cresilas, Amazon attributed to, xviii. 142 

Crete, meeting point of Thraco-Phrygian and 
Libyan elements, xvii. 372 ; connection 
with E^rypt and Libya, xvii. 362 ; oil- 
presses in, xviii. 210 ; Mycenaean gem 
from, xvii. 67 ; script, xvii. 327 ; com- 
pared with Libyan and Tuareg alphabets, 
xvii. 385 

Crimson, prophylactic, xviii., xliv. 

Croesup, story in Bacchylides and vase- 
paintings, xviii. 267 

Crusade, first (1097), xvii. 39 

Curator reipublicae, xvii. 403. See also 

Cyprus, Mycenaean objects from, xvii. 63 ; 
excavations (1894), xvii. 134 

Cyrene ? coin of, xvii. 79 


D, form in Latin epigraphy, xviii. 341 

Danaides as well-nymphs, xvii., xxxv. 

Death and the Horse, xviii. 1 

Delphi, statues of two Fates at, xviii., xliii. 

Demeter, Homeric hymn to, xvii. 49 

Devrent, xviii. 84 

DiadumenoB heads, xviii. 144 

Dictaean Cave, libation table, &c. from, xvii. 

Diniae. See Chelidonia. 
Dionysus, Homeric hymns to, xvii. 48 ; 

xviii, 28, 31 ; bearded and horned, xviii. 

165 ; bust on coins of Augusta, xviii. 162 ; 

with Corybantes, xvii. 90, See also 'Apxi- 

ffoKxos, Borpvs, Bptiatvs, 2t}Tdvfios, Tpttrrj- 

piKus, 4>Xc(ut' 
Dioscuri, Homeric hymns to, xviii. 31 
Doghan Assar, inscr., xviii. 118 
Dolichenus, xviii. 170 
Doriscus, site, xviii., xxxvii. 


Earrings, silver, from bronze-age tomb 

(Cyprus), xvii. 136 
Ecclesia personified, xviii. 161 
Edinjik (Mysia), inscr., xvii. 275 
Egypt, " Mycenaean " pottery from, xvii, 74 ; 

early connection with Crete, xviii, 362. 

See also Ptolemies 
Eiresiae in Homeric Hymn to Apollo, xvii. 

Eleusis, excavations, xviii. 332 
Eldesh, inscr., xviii. 123, 125, 127 
Ellesler, inscr. xviii. 115 
Elpis on coins of Anazarbus, &c., xviii, 162 
Emed, inscr., xvii. 281 

Ephesus, coin with UpanrfixT], xvii. 87 ; ex- 
cavations, 1897/8, xviii. 337 
Erinyes, xviii., xlll. 
Eros playing morra, xviii. 129 f. 
Eski Manias, inscr., xvii. 276 
Eteocles and Polyneices? on b.f. cantharos, 

xviii. 290 
Eteocretes, xviii., xxxiv. 
Euboea, coins with facing horsemen, xvii. 

Euphronius, Theseus cup, xviii. 276 ; 

Caeneus vase of his school, xvii. 297 
Euposia, Aurelia, xviii. 76 
Evjiler, inscr., xviii. 93 
Ezara. See Azara 

Fates and Furies in Aeschylus' Eumenides, 

xviii., xlll. 
Female element in Osiric cycle, xviii, 240 ; 

succession in Egypt, xviii, 263 
Festival coins of Pompeiopolis, xviii. 166- 

Fetishism in Greece, xviii. xxxv. 
Fibulae, gold Mycenaean, from Cyprus, xvii. 

Fish in Melos Mosaic, xviii, 71 ; fishing, 

methods of, xviii. 72 : with trident, xvii. xl. 
Flaviopolis = Sis? xviii. 311 
Fort, ancient, on Sphacteria, xviii. 161 
Fronto, C. Carietanius, xviii. 344 
Furies in Aeschylus' Eumenidea, xviii, xlll. 


Gaia? present at rape of Oreithyia, xviii, 

Galaxidi, Mycenaean ^m from, xvii. 6S4 
Gargaphia ppring xviii. 36, 235 
Gazelle in Melos Mosaic, xviii. 71 
Gedil Keui inscr., xviii. 113 
Gems, engraved, Mycenaean, from various 

sites, xvii. 65 f. ; from Larnaka, xvii 152, 

158. See also Seal -stones 



Geometric decoration of Melos Mosaic, xviii. 

George Cedrenus on Nika riot, xvii. 104 
Germaniceia (Marash), road from Caesareia 

to, xvii. 28 ; from Sebasteia to, xvii. 31 ; 

inscriptions, xviii. 311 
Geryon. See Heracles 
Getchid Keui, inacr., xviii. 119 
Qeune, ruins near, xviii. 82 
Geurness Keui, inscr., xviii. 112 
Giaour 01uk = ron;AcX((n'a, xviii. 107 
Gisza, in inscr., xviii. 113 
Glass tesserae in Melos Mosaic, xviii. 66 
Graeco- Phoenician type of lamp (Kalopsida, 

bronze-age, Cyprus), xvii. 139 ; sanctuary 

(Larnaka, Kamelarga), xvii. 164; tombs 

(do.), xvii. 152 
Graeco-Roman tombs at Larnaka, xvii. 162 
Gryphon and Apollo, xviii. 269 
Gymnasiarch, functions of, xviii. 97 


Hades, association with horses ? xviii. 1 
Hadriani (Mysia), xvii. 290 
Hadrianopolis-Sebaste, site and inscr., xviii. 

Hammamli Keui, above Cyzicue, inscr., 

xvii. 275 
Hecate, in Homeric Hymn to Demeter, 
xvii. 51 ; associated with Men in N. 
Phrygia, xvii. 283. See also Artemis, 
Helisson, river dividing Megalopolis, xviii. 

Hellenistic tombs at Larnaka, xvii. 162 
Heracles, last sacrifice of, xviii. 274 ; H. and 
Geryon on b. f. lekythos, xviii. 298 ; H. 
and Kerberos, xviii. 269 ; on b. f. vase in 
B.M., xviii. 292 ; list of representations 
on vases, xviii. 296 ; H. and Nemean 
lion, on votive terra-cotta relief, xviii. 
317 ; strangling-motive, xviii. 274 
Heraion near Plataiai, xviii. 50 
Hermes, Homeric Hymn to, xvii. 252 ; at 

Pheneus, xvii. 83 
Hermocome 1 xviii. 342, 343 
Herodotus on Artemisium and Thermopylae, 
xvii. 212 ; on Salamis, xvii. 230 ; on the 
Greeks at Plataiai, xviii. 33 
Hestia, Homeric Hymns to, xviii. 31. 
Hierapolis (Phrygia) and its villages, xvii. 

Hierophant, portrait of, from Melos, xviii. 

Holmoi (Plirygia), xviii. 109 
Homeric Hymns, Text of, xvii. 45, 241 ; 

xviii. 23 
Homonoia Sebaate, xvii. 84 
Horse, association with Death I xviii. 1 
Horseman to front on coins, etc., xvii. 81 
Hydra, Mycenaean gem from, xvii. 68 

Ibn Al Athir, xviii. 182 

Ibn Wadhich, xviii. 182 

Ikonion taken by Arabs, xviii. 198 

Ilghin, inscr., xviii. 123, 125 

Imperial Ei^tates in Asia Minor, xvii. 421 ; 

xviii. 340, 342 f. 
Ionia, coin with two lions, xvii. 87 
Ipsos. See Julia. 

Iris i head, on Lampsacene coin, xvii. 85 
Isauria, Arabs in, xviii. 183 (note), 189, 194 

Ishakli, inscr. xviii. Ill 
Isis, importance in Osiric cycle, xviii. 240 
' Island' on field of Plataiai, xviii, 37, 50 f. 

John the Lydian, on Nika riot, xvii. 94 
John Malalas, on Nika riot, xvii. 95 
Julia-Ipsos, xviii. 110 
Jupiter Dolichenus, xviii. 170 

K {see a/.s-o C) 

Kaballa, site, xviii. 120, 128, 344 

Kadi Keui, inscr. from, xviii. 86 

Kadmos, R., xvii. 407 

Kallateboa, xviii. 86 

Kalopsida (Cyprus) bronze-age settlement 

and necropolis, xvii. 138 
Kapros, R. =Geuk Bunar Su, xvii. 406 
Kara Budak = Sabrina, xviii. 320 
Karadilli = Khelidonia, xviii. 104 
Karadja-Oren (Holmoi), xviii. 109 
Karoura, site, xvii. 398 
Kassiiba Kirmasti (Mysia), inscr., xvii. 277 
Kavsa, inscr., xviii. 326 
Kebsud (Mysia) inscr., xvii. 291 
Kejiut, inscr., xviii. 325 
Kenaion, sacrifice un, xviii. 275 

Kerberos, number of heads, xvii. xlii. See 
aha Heracles. 

Kestelek (Mjsia), inscr. xvii. 278 

Khitab Al'Lyun, xviii. 182 

Khoma (on the Ak I'agh), xviii. 94 

Kiakhta, inscr., xviii. 315 

Kidramos site (Budjak Keui), and inscr. 
xvii. 396 

Kinnaborion, site, xviii. 109. 

Kithairon, Mt., Greek position on, before 
Plataiai, xviii. 35 

Knife, bronze, from Suria, xvii. 64, fr. 
Aegina, 65 

Kotchash (Hadrianopolis), xviii. 117, inscr. 

Krobylos, xvii. 90 

Kronos on coin of Tarso?, xviii. 178 

Kynosarges, stelae from, xvii. 174 



Laksha tu Riu (Cyprus) : bronze a^u 
necropolis with Mycenaean va3C8, xvii. 

Lamos (Cilicia), coin, xviii. 163 

Lanipe, inscr., xviii. 94 

Lampsacus, coins, xvii. 85 

Laodiceia ad Lycuin, the rivei;* of, xvii. 
404 ; inscr. 408 

Laodiceia in Syria, burnt by Romans, xviii. 

Larnaka, inscriptions from, xvii. 171 ; 
Graeco-Phoenician and Hellenii^tic tombs 
at Turabi Tek6, xvii. 152 ; Graeco- 
Phoenician sanctuary at Kaiiielarg^, xvii. 
164 ; excavations at Batsalos, xvii. 170 

Lausus, Palace of (Constantinople), xvii. 

Legio XV. Apollinaris at Satala, xviii. 321 

Legio XVI. Flavia Firma at Samsat, xviii. 

Lesbos, oil-presses in, xviii. 217 

Libation-table, inscribed, from Diktaean 
cave, xvii. 350 

Libyans in the Nile valley, xvii. 366 ; con- 
nection with Greeks, xvii. 375 ; Libyan 
elements in Cretan culture, xvii. 372 ; 
alphabets compared with Cretan and 
Aegean script, xvii. ^85. 

Lions, two, confronted, on electrum coin, 
xvii. 87. Lion and bull, coin-type of 
Tarsos, xviii. 175 

Longinus, Dionysius, Treatise on the Sub- 
lime ascribed to, xvii. 176 

Lopadium (Mysia) inscr., xvii. 271 

Lousoi, excavation, xviii. 334 ; Proetidaeat, 

xviii. 271. 
Lycia, coin, xvii. 90 
Lycos river, xvii. 405 ; bridge over, xviii. 89 

Lysias, xviii. 107 


Macedon, Philip II., coin of, xvii. 79 

Maenads and satyrs dancing, on Nikosthenes 
vase, xviii. 292 

Mahmud Assar, inscr., xviii. 122, 124, 126 

Malatia. See Melitene 

Males, descent through, xviii., xxxv. 

Mallos (Cilicia), coin, xviii. 163 

Mantinea, area compared with that of 
Megalopolis, xviii. 20 
Mantinean basis, xvii. 120 

Marash. See Germaniceia 

Marcellinus on Nika Riot, xvii. 92 

Marriage of brother and sister in Egypt, 
xviii. 244 

Mater castrorum, xviii. 322 

Maurus, Terentianus, xvii. 205 

Megalopolis, a double city, xviii. 15 ; com- 
pared in area with Mantinea, xviii. 20 

Megara (Greece), exeavation.s, xviii. 332 

Melampus and Proetidae, xviii. 271 

Meleager, death of, xviii. 26!) 

Melissa, site, xviii. 106 

Melitene, roads fiom Caesareiu, xvii. 22 ; 

pa.S8e8 into Commagene, xvii. 30 ; Arabs 

in, xviii. 184, 190 f. ; inscriptions, xviii. 

Melos, inscriptions, xvii. 1 ; Mycenaean 

gem, xvii. 68 ; oil-press, xviii. 215 ; 

' Aplirodite' of M. xviii., xli. ; excavations 

on bite of 'Three Churches,' xvii. 122 ; 

agora, xvii. 131 ; hall of the Mystae and 

mosaic, xviii. 60 
Men and Hecate in N. Phrygia, xvii. 283 
Meros (Plirygia) site and ir.dcr. xvii. 422, 

xviii. 341, 342 
Mesopotamia, frontier vars (641-750), xviii. 

Metropolis, roads to Synnada. xviii. 101 
Mikhalich (Mysia) inscr., xvii. 271 
Mikkiades, name of, on basis at Paros, xviii. 

Miletopolis (Mysia), xvii. 271 
' Minotaur' scheme in early art, xvii. 370 
Missis (Mopsuestia), inscr., xviii. 307 
Moeris, lake, reclaiming of, xviii. 251 
Moimul (Mysia) inscr., xvii. 281 
Monandry in Homer, xviii., xxxv. 
Moni near Amathus, fibulae from, xvii. 63 
Mopsuestia, inscr., xviii. 307 
Moragocome ? xviii. 342, 343 
Morra, the game, xviii. 129 
Mosaic from Melos, xviii. 60 
Moses, quoted in Treatise on the Sublime, 

xvii. 194 
Mucius (Q.) Scaevola, xvii. 277 
Mycenae, gem from, xvii. 69 ; Mycenaean 

antiquities in B.M., xvii. 63 ; Mycenaean 

vases from Lakshc\ tu Riii, xvii. 151 
Mysia, inscriptions, xvii. 268 
Mystae, Hall of, at Melos, xviii. 60 ; societies 

in Greece, 78 
Mytilene, coin, xvii. 86 


Narlinar (Mysia) inscr., xvii. 279 

Necklace described in Homeric Hymn to 
Apollo, xvii. 244 

Necropoleis in Cj'prus : bronze-age, at Agia 
Paraskevi, xvii. 134 ; Kalopsida, xvii. 138, 
140 ; Lakslii, tu Riu, xvii. 147 ; Graeco- 
Phoenician and Hellenistic, Larnaka, xvii. 

Nemean Lion. See Heracles 

Nemrud Dagh, reliefs and inscr. from, xviii. 

Neoptolemos sacrificing Polyxena, xviii. 285 

Nika riot, xvii. 92 

Nike, head of?, on Lampsacene coin, xvii. 
85 ; temple of N. Apteros, inscr. xviii. 

Nikopolis (Armenia Minor), xviii. 324 

Nikosthenes, b. f. kyathosby, xviii. 292 



Octagon (Constantinople), xvii. 112 

Oil-presses, xviii. 209 

Oiuan = Lysias, xviii. 107 

Oinoe besieged by Archidamos, xviii. 225 

Okeanids, names of, xvii. 60 

Oktchular, inscr., xviii. 342 

Olympieion (Athens), excavations, xviii. 330 

Omar Keui, nr. Panderma (Mysia), inscr., 

xvii. 273 
Omphalos, Apollo on, xvii., xll. 
Onchestus in Homeric Hymns, xvii. 247, 

Optimus, proconsul of Asia, xviii. 341 
Oreithyiaand Boreas, xviii. 136 
Osiris-worship under the Ptolemies, xviii. 


Palaeo Kastro in 1825, xviii. 346 

Pallas, winged, xviii., xlili. 

Pan, Homeric hymn to, xviii. 29 

Panathenaic amphoriskos, xviii. 300 

Panderma (Mysia), inscr., xvii. 274 

Paroreios, Phrygia, xviii. 110 

Paros, discoveries in, xviii. 336 ; Parian 

Chronicle, new fragment, xviii 336 
Parthenon, repair of, 1897/8, xviii. 329 
Paschal Chronicle on the Nika Riot, xvii. 

Pausanias at battle of Plataiai, xviii. 36 f. 
Peirene (Corinth), excavation, xviii. 333 
Peiresiae in Homeric Hymn to Apollo, xvii. 

Pelasgians, xviii., xxxlv. S. 
" Peloponnesian " vases, xviii. 283 
Perfectisnimus, xviii. 342 
Perrin, inscr., xviii. 316 
Perseus on coins of Tarsos, xviii. 174 
Phagres, site, xviii , xxxvii. 
Phazeraonitarum Thermae, inscr., xviii. 326 
Pheneus, coin, xvii. 83 
Philae, threatened ruin of temples, xviii. 

Philemon (I), portrait on coins of Pompeio- 

polis, xviii. 168 
Philip II. of Macedon, coin, xvii. 79 
Philomelion, inscr., xviii. Ill 
Phocaea, coin, xvii. 89 
Phradmon (?), head by, xviii. 145 
Phrygia, explorations in, xvii. 396 ; xviii. 

81, 340 
Phrygo-Lydiiin frontier, xviii. 81 
Pictographs, Cretan, xvii. 331 
Pingan, inscr. xviii. 321 
Pisa (Phrygia), site and inscr., xviii. 114 
Plataea, the Greeks at, xviii. 33 ; Thukydi- 

des' description of the siege, xviii. 218 ; 

importance of the place, xviii. 225 ; criti- 
cism of Woodhouse s theory of the battle, 

xviii. 235 
Pleuron (?), New, xviii. 335 

I'oemanenum (My.sia), xvii. 271 

Polycephalous monsters, xvii., xU. 

Polynta, xviii. 342, 343 

Polyxena, sacrifice of, xviii. 284 

Pompeiopolis. See Soloi 

Pompeius on coins of Pompeiopolis, xviii. 

Pontus, inscr., xviii. 325 
Poseidon, Pelasgic, xviii. xxxv. ; archaic 

bronze statue from Dombr^na, xviii. 332 ; 

P. of Melos, xviii. xli. ; with Amphitrite 

at Tenos, xviii. xli. ; his trident, xvii., 


Praetorium (Constantinople), xvii. 113 
Presses for olive-oil, xviii. 209 
Priene, excavations, 1897/8, xviii. 338 
Priest-class under the Ptolemies, xviii. 238 
Procopius on the Nika Riot, xvii. 93 
Proetos, daughters of, xviii. 271 
Province of Asia, awtdpiov, xvii. 277 
Ptolemies, their policy of marriage and suc- 
cession, xviii. 243 ; women in their age, 
xviii. 238 
Purk (Nicopolis), inscr., xviii. 324 
Pylae Ciliciae, xviii. 307 
Pylos and Sphacteria, xviii. 147 ; Thucydi- 
des' description of the operations, xviii. 
218, 228 ; criticism of Burrows' theory, 
xviii. 233 ; of Grundy's, xviii. 345 
Pyxis, marble, from Aegina, xvii. 65 


Quadriga seen to front on coins and vases, 
xvii. 81 


Red, prophylactic significance of, xviii., 

Rhaescuporis, King, xvii. 322 
Rhodes taken by Arabs, xviii. 187, 189 
Ring, gold, from Lampsacus with morra, 

xviii. 129 ; in B.M. with Aphrodite and 

Eros, xviii. 132 
Road-system of Eastern Asia Minor, xvii. 

22 ; Royal Road, xvii. 41 
Romanus IV , campaign of 1068/9, xvii. 36 
Royal road, xvii. 41 
Russiler (Jlysia), inscr., xvii. 290 

Sabrina, river = Kara Budak, xviii. 320 

Sadagh. See Satala 

Saujilar (Phrygia), inscr., xvii. 286 

Sala, site of, xviii. 81 

Salaniis, battle of, xvii. 230 

Samosata (Samsat), inscr., xviii. 312 

Sanaos. See Anava 

" Sardanapalus, monument of," xviii. 169 

Saria. See Suria 



Satalft, inecr., xviii. 321 ; a colony, xviii. 

Satyrs dancing, on Corinthian vases, xviii. 

287 ; with Maen&cls, on Nikosthenes 

kyathoB, xviii. 292 
Seal-Btones (Cretan), with linear characters 

and fif^urep, xvii. 329. 346 ; with early 

pictography, 331 ; with conventionalised 

pictographs, 334 ; early prism-seal from 

Karnak (Egypt), 362. Net (iIho Genia 
Sebaste Hadrianopolis, site and inscr. xviii. 

Sebasteia taken by Arabs, xviii. 193 ; roads 

radiating from, xvii. 31 
Selene, Homeric hymn to, xviii. 31 
Seleucia ad Calycadnum, coin, xvii. 90 
Selinda, site (Selind), xviii. 115 
Selinus-Traianopolis, coins, xviii. 164 
Seraijik, inscr., xviii. 319 
Shahr, inscr., xviii. 316 
Sibidounda, xviii. 104 
Siblia, in inscr., xviii. 93 
Siege- operations as described by Thucydides, 

xviii. 218 ; a new feature in 5th century, 

xviii. 219 
Sikyon, coin with dedicatory inscr., xviii. 

Simav, lake (Phrygia), inscr., xvii. 285 
Siren on oenochoe from Aegina, xviii. 281 
Sis (Sision), road from Caesareia to, xvii. 29 ; 

inscr., xviii. 310 
Sivri Tepe, inscr., xviii. 325 
Slavs, city of the, xviii. 194 
Smyrna, coin with /3awf, xviii. 304 
Soloi-Porapeiopolis, coins, xviii. 165 
Sozopolis. See Apollonia 
Sparta, Mycenaean gem from, xvii. 69 
Sphacteria. See Pylos 
Stasioikos of Cyprus, coin, xvii. 165 
Stelae from Kynosarges, xviii. 174; grasping 

of the wrist on Attic Stelae, xviii. 133 
Sublime, treatise on, xviii. 176 ; its author- 
ship, 189 
Suria (nr. Karpathos), bronze implements 

from, xvii. 64 
Sybrita, coin, xvii. 83 
Symbols (conventionalised), on Cretan picto- 

graphic seals, xvii. 339 
Synnada, roads from Metropolis, xviii. 101 
Syracuse, Thucydides' description of the 

siege, xviii. 218 
Syria, frontier wars (641-750), xviii. 203 

Tachtali, near Brusa, inscr., xvii. 268 
T&citns, Dialogua compared with de Suhlimi- 

tate, xvii. 200 
Tarsos, coins of, xviii. 169 
Tash Keui (Mysia), inscr., xvii. 289 
Tavshanli (Mysia), inscr., xvii. 279 
Taz Kiri plain, xviii. 91 

Tchuushji Keui (Ilj^hiii lake), xviii. 124, 

Tchigil, inscr., xviii. 119 
Tchivrili-tchemani, xviii. 121, 12h, 344 
Tegeans at Pialaiai, xviii. 41 
Tekke Keui, inscr., xviii. 127 
Tembrion, Imperial Estate at (inscr.), xvii. 

Tephrike, road from Sebasteia to, xvii. 

Terentianus, addressed in de Subliiiiitate, 

xvii. 195, 203 
Terra-cottaH, votive, from Lamaka, xvii. 
164 ; reliefs in Acropolis Museum, xvii. 
Theatre of New Pleuron (1), xviii. 335 ; of 

Priene, xviii. 339 
Thebe on vase of Assteas, xviii. 271 
Theodoridas, son of Daistratos, dedication 

by, xviii., xli. 
Theodorus, quoted in de Sublimitate, xvii. 

Theodorus Lector on Nika Riot, xvii. 94 
Theophanes on Nika Riot, xvii. 101 
Thermae Phazemonitarum, inscr., xviii. 326 
Thermon, excavations, xviii. 335 
Thermopylae, battle of, xvii. 215 
Theseus and the ring, xviii. 276 
Thracian colonists at Apollonia-Sozopolie, 

xviii. 96 
Thraco- Phrygian elements in Cretan culture, 

xvii. 372 
Thucydides, his detailed account of sieges, 

xviii. 218 
Thymbrion, xviii. 116 
Thynnaros, hero of Synnada, xviii. 103 
Tiryns, Mycenaean pottery from, xvii. 75 
Tombs. See Necropoleis 
Totemism in Greece, xviii. xxxv. 
Traianopolis. See Selinus 
Tralla, site of ? xviii. 83, 85 
Trapezopolis, site and inscr, xvii. 401 
Tree-spirits : Caeneus and Centaurs, xvii. 

Trident of Poseidon, xvii., xl. 
Tripolitan oil-presses, xviii. 209 
Trophimus, M. Marius, hierophant, xviii. 

Tuareg alphabet compared with Cretan and 

Aegean script, xvii. 385 
Tumulus of Choban Tepeh (Troad), excava- 
tion, xvii. 319 
Tyana taken by Arabs, xviii. 191, 192 
Tyche on coins, seated, xviii. 161, 163, 166, 

179 ; adoring, 166, 167 
Tyriaion, inscr., xviii. 121 


Ulubad (Lopadium), inscr., xvii. 271 
Uluborlu, inscr., xviii. 95 



VALENTENUS = Valens, xviii. 309 

C. Fa[/](en«s) \_Ter]tullus or \^Ter]tullianm. 

legatus Aug. pr. pr., xviii. 320 ^ 
Victor Tonnenuensis, on the Nika Riot, 

xvii, 94 
Vine-mosaics, xviii. 69 
Votive figures from Laruakn, xvii. 164 ; 

terra-cotta reliefs in AcropoliH Museum, 

xvii. 306 


WALLSonPylos and Sphacteria, xviii. 234, 345 
Warrior, departure of (b.f. cantharos), 

xviii. 289. Warriors playing with ntara-oi, 

xviii. 294 ; preparing for battle (b.f. 

kylix), xviii. 291 ; combat over fallen 

warrior, xviii. 289 
Women in Ptolemaic Egypt, xviii. 238 
Wreath adorned with human heads, on coins 

of Tarsos, xviii. 179 
Wrist, grasping of, as greeting, xviii. 133 

Xeres, march of, xviii., xxxvu. 

Yaka Sinek, inscr., xviii., Ill 

Varsowat, inscr., xviii. 310 

Yemishli (Phrygia), inscr, xvii. 289 

Yendin Keui, inscr., xviii. 118 

Yenije, u. of Assar Kale (Mysia), inscr, 

xvii. 293 
Yenije Keui. nr. Emed (Mysia), inscr., xvii. 

Yenije Keui, nr. Panderma (Mysia), inacr., 

xvii. 273 

Zacharias of Mytileue, continuation of, on 

the Nika Riot, xvii. 95 
Zarukas (Cyprus), late bronze-age site, xvii. 

Zeus, xvii. 416 ; Homeric Hymn to, xviii. 

30 ; Z. and Amalthea, xvii. 82. See also 

Abrettenos, Knrat/Sarar, Kpa/.t^'rjcor, Mtyt(T- 

Tos, Ovpdvios, ndvbrjfxot, StoTTjp 
Zibatra, xviii. 208, note, and Corrigendum 
Zonaras on Nika Riot, xvii. 104 


^AyKiarpflu, xviii. 72 
'Adpaarrja (games), xvii. 410 
"Adpaarot, MtjTijp, xvii. 400 
Abpiavtav ^ovKt] Koi Sfjpos, x\ ii. 290. 
'Adrjvalr] Tlapdivos, xviii., xliil. 
[Aly]fipaTis (Artemis), xviii. 304 
'A\(^ay8fiavr) (Tarsos), xviii. 173 
'Ain-io;^«a (games), xvii. 410 
'AvT<uv€ipiapi) (Tarsos), xviii. 173 
unr)p.r), (iTrrjpT) Upd, Xvii. 87 
anodo^^fluv, xviii. 88 
dnopoipa in Egypt, transfened from temples 

to Queen Arsinoe, xviii. 251 
'Airniavo3V to KXipa, xvii. 419 
'Apayovr)voi, xvii. 419 
"Apecor vtLKai, xvii. 275 
'Apxi^aKxos (Dionysos), xviii. 78. 
'Apxtdaa-adpa, xviii. 79 
'Apxi^ox^KoXos., xviii. 79 
'AT[r]aXiy [<f>v\\rj at Laodicea, x\ ii. 408 
'ArriKal bpaxpai, xvii. 414, 415 
'ATTowSe'toi' 6 bffpoi, xvii. 399 
<i<^j;^a>s', xvii. 245 

Ba/SvAcoj/ior, xvii. 173 

^aKxdov, xviii. 78 

BaKYot, xviii. 78 

/3aX»jj/=king, xviii. 96 

/SoAXetJ/ vopov, xvii. 242 

^avi=^a{aiKt)vi on coins of Smyrna, xviii. 

Bdx;^fta, xvii. 91 
Bei/ercoi/ (f)v\fi, xviii. 98 
B(t/i/ct Sotjvwv, xvii. 284 
Bojj^o'r (Perseus), xviii. 175 
BoTpvs (Dionysos), xviii. 78 
BovKoAot, xviii. 79 

^ovXfVTciv (Pv\ri 'ArraKis (Laodicea), xvii. 408 
Bp(ia-€V{ (Dionysos), xviii. 78 
[Bp]tat»3s, xvii. 416 

rovvKkicria, xviii. 107 

Aela 2€^a(TTa oiKOVfifviKa, XVV. 410 
Ata, xvii. 410 
HiaoTjpoTaroi, xviii. 342 
SiKTVfia, xviii. 72 


iiaypirai, xviii. 123 
8op((TTiKan', Koprft Tuv, wiii. 32G 
fiojf, xvii. 54 

'Ent/i)S, fdvoi, xviii. 24 

'E0p[aio)v (Twlayayrj, x\'iii. 333 

(Iprjvapxrjs, xviii. 123 

flafXaariKii, xvii. 410 

(l(Tnov8((i)v = awovBaiuv, xviii. 126 

'EKK\i][cria] porsnnified, xviii. 161 

'EKoVa, xvii. 283. See iwnipr], Sar^pv. 

tXtf, xviii. 1 

(p^oSddiv, xviii. 308 

(^(Tacrrai (Laodicea), xvii. 408 

tnapxiai, y', xviii. 181 

fnfjKoos, 0f6s (Mara.'}!), xviii. 311 

*mp€'Ki]Tij%, xvii. 402, 4f'3 

(iriTpoTroi 2(j3a(TTS>p, xvii. 399 

fnK^nvfararoi Ka'tcrap, xviii. 161, 318 

fpaviarai, xviii. 78 

'Epivvfs, xviii., xlU. 

tpvfia, 7ra\ai6i>,on Sphactcria, xviii. 152, 234 

(vytinis TTio-Toi 6(o<l)i\i'ts MaKfbopfs (Aigeai), 

xviii. 161 
fipvXiT], xvii. 261 
EiipvavaKTiiav y xviii. 216. 
€v\d, xvii. 83 

'H/ijo'/3<Xn', xvii. 83 
'HpoKXtia (games), xvii, 410 

eea Mf-y/oTi; (Ma), xviii. 316 
QtoSalata, xviii. 75 
d(o(pikf'ii. See (vytvus 
dtaacJTai, xviii. 78 
6vpa of tombs, xviii. 95 

'UpanTjpij, xvii. 87 

'If ponoXurai : Comana, xviii. 318 ; Hiera- 

polis, Plirygia, xvii. 411 
l(po(f>avTr]t, xvii. 14 
'iXiadris for 'OiXid8>7r, xviii. 286 
\^l<ToXvp]nia, xvii. 410 

B B 



Ka6ocriovfji€Poi dofifariKoi, xviii. 326 

Kanpos (liver), xvii. 405 

Kapnto, iii\, xviii. 134 

Kurdyfiv, xvii. 310 

Karat^arar ZfJf, xvii. 8, 9. 'j^arai^drr^i, 

xvii, 9 
xaTot/ci'a, 17 Mftprji/wi;, Xvii. 423 
KUToiKot (V ['0]Xj3otf ? xviii. 87, 312 \ 
KaTwp, xviii. 28 

Kfbpcov (suggested place-name), xviii. 305 
Kfaapiavoi, xvii. 420 
xfi = Kai, xviii. 91 
Krjpfs 'Epivvts, xviii. xlii. 
KicrTa(f)6pos, xviii. 79 
K\ip.a TO ' AnmavaVf Xvii. 419 
«XiTof, xviii. 9 
K\vr6nco\os, xviii. 1 
kKvtos, xviii. 1 

KOivbs dfipos Mo^taviop ^orjvuu, xvii. 419 
Ko\u>voi, xviii. 96 

KofXTfs tS>v Kado(riovp.fv<t)v Bo/if ariKwv, xviii. 326 
KovTuxTis, xviii. 72 
KoTVf ^naikfvs, xvii. 322 
Kovpiirtopy GtloSf xviii. 326 
Kpaivttv, xvii. 263 
Kpap.y\rr)voi Zfv'f, Xvii. 293 
xparrjpiapxos, xviii. 79 
KvpTfia, KvpTTj, xviii. 72 
Kupai of Hierapolis, xvii. 411 
Kdpdpxai, xvii. 412, 420 

Aap.Trair}8popia on Panathenaic amphoriskos, 

xviii. 300 
^apTTporaros, xviii. 342 
"XiKvatfiopos, xviii. 79 
XoytoT^r at Trapezopolis, xvii. 403 
AvKoKairpos (river), xvii. 405 
AvKos (river), xvii. 405 

*08/ii7 *'C*»i xvii. 50 

oiKia, o?(cof = liall of .AFystac, xviii. 80 

<HKovp(VT)v, fl(T(\a<rTiKu (U Tf)v, xvii. 410 

oiKovptviKa, xvii. 410 

["OjX/Soif ?, KiiroiKot fv, xviii. 87, 342 

OXw/iirta, xvii. 410 
(tpo^wpioi 6fo\ 2e,3a(rrot, xvii. 284 

Opovoin 2f ^aa-TTj, xvii. 84 
nvoTaa-Tov, xviii. 27 
'Qparla (tribe), xvii. 292 
opytwvfs, xviii. 78 
opyi7, xvii. 55 
*Opoi K(X(KCi>c, xviii. 306 
Ovpavios ZtiJf, xviii. 319 

Haiavoi, MaV, xvii. 283 

navappofiov, xvii., xH. 

ndvbrjpos, Zfvs, xvii. 289 

Trapa^vXaKts, xvii. 411 

7rapa<f>v\aKiTai, xvii. 412 

IlapBivos, xxiii., xliii. 

ndrpcop Trjs TroXfo)?, xvii. 403 

narpaos (Perseus), xviii. 177 

nacpitj (Aplirotlite), xvii. 291 

TJtpratTrjpiKoi aywi/ff, xvii. 410 

TTfpi^aipiot, xvii. 16, xviii. 79 

TToraoi, xviii. 294 

TfiaToi. See fvytpui 

-nokos in compounds, xviii. 3 

npoKadeCopfPTf (Tarsos), xviii. 173 

npoDTdpj((i)p TTjs irokfcjs, xvii. 402 

TrpcuTavXor, xviii. 96 

TrpoiTOKtiiprjTrjs, xvii. 292 

npvravis rfjs (f>v\fis, xvii. 408 

TTvXrfHoKos, xvii. 253 

nvpfjpa, xviii. 209 

7r(o\ti(T$ai, xviii. 3 

TriXof, xviii. 1 ; in compounds, xviii. 3 

Ma (goddess of Coniana), xviii. 317 
Ma^^oyaios, Map^oya'ios, xviii. 316 
MaKeSover. See tvytPt'is 
Map 6 Ilaidvos, xvii. 283 
Mfyas Zevs Ovpdvios, xviii. 319 
Mtyicrros Zfiin, xvii. 403, xviii. 310 
Mtiprfpcip KaToiKia, xvii. 423. TroXtf, xvii. 

Miji/, xvii. 283 
Mo{^)faPa)P 1 {MoTT€apStv) 2ot]p5>p Srjpos KOipui, 

xvii. 419, xviii. 341* 
MovKifia, xvii. 276 
MovaSiP 6p€TTTfip, xvii. 268 
fjLvarai, xvii. 14, xviii. 78 
pv(TTap\iai, xvii. 275 

NciKOi "Aptwy, xvii. 275 
'N(iico7ro\(iTT)s, xvii. 275 
p(oi, oi (Attuda), xvii. 399 
puyftap, xvii. 59 

S in Melian epigraphy (l), xvii. 5 

'PiKPovadai, xviii. 288 

^aparjpmp 6 d^pos, xvii. 414 

2f/3aoTa o'lKovpfPiKa {Ada), xvii. 410 

2(^aaTT) 'Opopoia, xvii. 84 

2fQaoToc, 6po^d>pioi 6foi, xvii. 284 

^Tjrdpdos (Dionysos), xviii. 78 ^ 

lor]pu>v Efvi/fi, xvii. 284 

2oT]va)P Mo^tavup (ToTTtaPcop ?) koivos dfjpos, 

xvii. 419 
[arf](f)avr](f)6pos, xvii. 414 
(TToid, xviii. 88 
(TTpardpxqi, xvii. 419 
(TTpaTTjyos tnl Tfjs x<^P'^^) xviu. 123 
(TvvaKrjvos, xvill. 308. 
lardpT) 'EKdrrj, XVii. 284 
2mtt]p, Zevr Miyta-ros, xvii. 408 
liOT^prj 'EKaTTj, xvii. 284 
^uTtjpia, xvii. 276 

Taplas, xviii. 401 
Tipufip 8to TptOf xvii. 48 
TfpniKtpavvos, xviii. 3 



ToTTfctt'oi, xvii. 420, xviii. 341 

'l'paiTf(i>no\iTu)v ^ovXt], xvii. 403 

Tf)d(p()s — Td(f)f)os, xviii. (J3 

Tfjffudovs, Tjiffiivduvs, Tp€fMipdia, xviii. G2 

rpiaiva, xvii. (xli.) 

rpioSoirrla, rpiubovi, Xvii. (xl.) 

TiHfTrjpiKus AtoVuo-oy, Xvii. 14 

'VXoro/iof, xvii. 56 
vTTofipvxias, xvii. 257 
inoTdfifov, xvii. 56 

* lepreseuteJ liy O, xvii. 89 

<P6uvoi 1)11 vase with deulh of Meleager 

xviii. 2()<) 
♦Xt'un CDioiivsos), xviii. 78 
♦/juT/nof, Zfiis, xviii. 330 
<^vX(iKiru(, xvii. 412 
<Pv\ii Ufi/er<of, xviii. 98 

XpvatjXdKaros, xriii. 3 

4' ill Melian epiginphy, xvii. 5 


Somtg for i^t |pr0motion of Hellmic Slu&its. 

I. The objects of this Society shall be as follows: — 

1. 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 
of 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 archaeological and topographical interest. 

III. To organise means by which members of the Society may have 
increased facilities for visiting ancient sites and pursuing archaeological 
researches in countries which, at any time, have been the sites of Hellenic 

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. 


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. 

9. Three Members of the Council, provided not more than one of 
the three present be a permanent ofificer of the Society, shall be a 

10. 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, Ofificers, and Auditors for the ensuing year 
elected, and any other business recommended by the Council discussed 


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 

16. The President and Vice-Presidents shall be appointed for one 
year, after which they shall be eligible for re-election at the Annual 

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 of those present. 

b 2 


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 ;^I5 15^., entitling compounders to be 
Members of the Society for life, without further payment. All Members 
elected on or after January i, 1894, shall pay on election an entrance fee 
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 thereof 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 

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. 



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. 

n. 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. 

in. 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, ii 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). 

VIT. That the Society's books (with exceptions hereinafter to be 
specified) be lent to Members under the following conditions :— 

(i) 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 :— 

(i) That all requests for the loan of books be addressed to the 

(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. 


(5) All expenses of carriage to and fro shall be borne by the 


(6) All books are due for return to the Library before the summer 


IX. That no book falling under the following categories be lent out 
under any circumstances : — 

(i) 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 


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, 

XL 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. 
Mr, Talfourd Ely. 
Prof. Ernest Gardner. 
Prof. Percy Gardner. 
Miss Jane Harrison, LL.D. 
Mr. Walter Leaf, Litt.D. 
Mr. George Macmillan {Hon. Sec). 
Mr. Ernest Myers. 
Mr. J. L. Myres. 

Mr, Arthur Hamilton Smith. {Hon. Librarian). 
Mrs. S. Arthur Strong, LL.D. 
Sir E. Maunde Thompson, K.C.B., D.C.L. 

Assistant Librarian, MiSS Fanny JOHNSON, to whom, at 22, Albemarle 
Street, applications for books may be addressed. 

SESSION 1898— 1899. 
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 : — 

Thursday, November 3rd. 

Thursday, February 23rd. 
Thursday, April 27th. 
Thursday, June 29th (Annual). 
The Council will meet at 4.30 p.m. on each of the above days. 




PROFESSOR R. C. JEUD. I.itt.D., D.C.L.. LL.D., M.P. 


PROK. S. H. BUTCHER, Litt.D., LL.D. 






MR. D. B. MONRO, LittD., LL.D., Provost of Oriel 

College, Oxford. 

PROF. H. F. Pl.LIIAM, Preiident of Trinity College 

MR. F. C. PENROSE, F.R.S., D.C.L., LL.D. 
MR. J. E. SANDYS, Litt.D. 
PROF. R. Y. TYRRELL, Litt.D.. D.C.L. 





MR. A. B. COOK. 









MR. G. F. HILL. 









MR. J. A. R. MUNRO. 





MR. G. H. RENDALL, Litt.D. 







Hon. Treasurer. 


Hon.' Secretary. 


Assistant Secretary. 


Hon. Librarian. 


Assistant Librarian. 


Acting Editorial Committee. 


Consultative Editorial Committee. 

and Mr. D. G. HOGARTH (ex officio), as Direciorof the British School at Athens. 

Auditors for 1898-99. 







Officers and Committee for 1898-1899. 

(S ^airman. 
Prof. R. C. Jebb, Litt.D., D.C.L,, LL.D., M.P. 

Ili»-(t ^airman. 
Mr. J. E. Sandys, Litt.D. 


Mr. J. G. Frazer, LL.D. 
Prof. Ernest A. Gardner. 
Mr. Henry Jackson, Litt.D. 
Mr. M. R. James, Litt.D. 
Prof. W. Ridgeway. 

Mr. E. E. Sikes. 

Mr. Arthur Tilley, 

Mr. a. W. Verrall, Litt.D. 

Mr. C. Waldstein, Litt.D. 

Pon. Seerelarg. 
Mr. Arthur Bernard Cook, Trinity College. 



HIS MAJESTY THE KING OF THE HELLENES, d Af. le Secretaire du /foi des 

Hellenes, Athens, Greece. 
Prof. Friedrich August Otto Benndorf, The University, Vienna. 
Sir Alfred Biliotti, K.C.B., H.B.M. Consul for Crete. 
Prof. D, Comparetti, Istituto di Studii Superiori, Florence. 
M. Alexander Contostavlos, Athens 

Prof. A. Conze, Kaiserl. Deutsches Archaeologisches Institut, Comelius-tr., II., Berlin. 
Mr. George Dennis c/o Lloyds Bank, Limited, 16, St James's Street, S.IV. 
Prof. Wilhelm Ddrpfeld, Ph.D., Director of the German Archaological Institute, Athens. 
Monsieur L'Abb^ Duchesne, Ecole Fran^aise, Rome. 
Monsieur P. Foucart, 13, Rue de Tournon, Parts. 
Prof. Adolf Furtwangler, The University, Munich. 
Monsieur J. Gennadius, 21, Hyde Park Place, IV. 

His Excellency Hamdy Bey, Keeper of the Museum of Antiquities, Constantinople. 
Prof. W. Helbig, Villa Lante, Rome. 
Monsieur HomoUe, Director of the French School, Athens. 
Monsieur P. Kavvadias, Ephor-General of Antiquities, Athens, Greece. 
Prof. A. Kirchhoff, The University, Berlin. 
Prof. U. Kohler, The University, Berlin. 
Prof. S. A. Kumanudes, The University, Athens. 
Prof. A. Michaelis, University, Strassburg. 
Prof. E. Vtitrstn, Instituto Archeologico Germanico, Monte Tarpeo,Rome. 


♦ Original Members, "t" Life Members, 
The other Members have been elected by the Council since the Inaugural Meeting. 

Abbott, Evelyn, Balliol College, Oxford. 
fAbercrombie, Dr. John, 23, Upper Wimpole Street, W. 

Abram, Edward, i, Middle Temple Lane, E.C. 

Adam, James, Emmanuel College, Cambridge. 

Adams, Miss Mary G., 43, Campden Hill Square, Kensington, W. 

Agnew, Philip L., 18, Gloucester Square, IV. 

Ainger, A. C, Eton College, Windsor. 

Ainger, Rev. Canon, Master's House, The Temple, E.C. 
fAinslie, R. St. John, The School, Sedbergh. 

Alford, Rev. B. H., St. Lukd's Vicarage, Nutford Place, U' . 

Allbutt, Professor T. Clifford, M.D., F.R.S., Chaucer Road, Cambridge. 

Allen, T. W., Queen's College, Oxford. 

Amherst, Lord, Didlington Hall, Brandon, Suffolk. 

Anderson, J. G. C, Lincoln College, Oxford. 

Anderson, J. R., Lairbeck, Keswick. 

Anderson, Prof. W. C. F.(Council), Firth College, Sheffield. 

Anderton, Basil, Public Library, Neivcastle-on-Tyne. 
*Antrobus, Rev. Frederick, The Oratory, S.W. 

Apostolides, S., 

Archer-Hind, R. D., Trinity College, Cambridge. 
tArkwright, V^.,Adbury 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., New 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. 

Ball, Sidney, St. John s Colles^c, Oxford. 
fBarlow, Miss Annie E. F., Greenthorne^ Edgworth, Bolton. 
Barlow, Mrs., 10, Wimpole Street, W. 

Barnewall, Sir Reginald A., Bart., 23, Cliveden Place, Eaton Squat e, S. Vl. 
Barnsley, Sidney H., Pinbury, near Cirencester. 
Bar ran, J. N., VVeetwood, 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. 
fBeaumont, Somerset, Shere, near Guildford. 
Beebee, M. J. L., New Travellers Club, 97, Piccadilly, W. 
|-Benn, Alfred W.,70, Via Cavour, Elorence. 

Benschoten, J. C. van, Wesleyan University, Middletown, Conn., U.S.A. 
Bennett, S. A., Audley House, Richmond, Surrey. 
Bent, Mrs. Theodore, 13, Great Cutnberland Place, W. 
Bevan, E, R., 

Bickford-Smith, R. A. H., 98, Palace Gardens Terrace, W. 
Bienkowski, Prof, von P., Retoryka, 15, Krakau. 
fBikelas, Demetrius, 50, Rtte de Varenne, Paris. 
Bishop, Major Tuke, 2E, Albany, Piccadilly, IV. 

Blomfield, Sir A. W., A.R.A.,6, Montagu Place, Montagu Square, IV. C. 
Blomfield, Mrs. Massie, Port House, Alexandria, Egypt. 
Blumenfeld, Ralph Drew, 64, Cheyne Court, Chelsea, S. IV. 
Bodington, Prof. N., Principal of the Yorkshire College, Leeds. 
Bond, Edward, M.P., Elm Bank, Hampstead, N. W. 
Bosanquet, Rev, F. C, T., The Hermitage, Uplyme, Devon. 

Bosanquet, R. Carr (Council), Rock Hall, Alnwick, Northumberland. 

Bosdari, Count Alessandro di, 20, Grosvenor Square, IV. 

Bougatsos, Christos Ch., Alexandria, Egypt. 

Bousfield, William, 20, Hyde Park Gate, S. IV. 

Boyd, Rev. Henry, D.D., Principal of Hertford College, Oxford. 

Boys, Rev. H. A., North Cadbury Rectory, Bath. 

Braraley, Rev. H. R., The Precentory, Lincoln. 

Bramwell, Miss, 73, Chester Square, S. W. 

Branteghem, A. van, 29, Queen Anne's Gate, S. W. 

Brinton, Hubert, Eton College, Windsor. 

Broadbent, H., Eton College, Windsor. 

Brock, Mrs., 115, Adelaide Road, South Hampstead, N.W. 
*Brodie, E. H.fGrasendale, Malvern. 

Brooke, Rev. A. E., King's College, Cambridge. 

Brooke, Rev. Stopford A., i Manchester Square, W. 

Brooks, E. W., 28, Great Ormond Street, W.C. 

Brooksbank, Mrs., Leigh Place, Godstone. 

Brown, Horace T.. F.R.S., 52, Nevern Square, South Kensington, S IP. 

Brown, Prof. G. Baldwin, The University, Edinburgh. 

Brown, Walter, 2, Albert Square, Great Yarmouth. 
*Bryce,The Right Hon. James, D.C.L., M.P., 54, Portland Place, W. 

Buck, G. M., City of London School, E.C. 

Buller, Lady Audrey, 29, Bruton Street, W. 

Bulwer, Sir Henry, K.C.B., 11, South Street, Park Lane, W. 

Burge, Hubert M. University College, Oxford. 

Burgh, W. de. University Extension College, Reading. 

Burnet, Prof. J., l, Alexandra Place, St. Andrews, N.B. 

Burrows, Prof Ronald, University College, Cardiff. 

Burton, Sir F. W., 43, Argyll Road, Kensington, W. 

Bury, Prof. J. B., Trinity College, Dublin. 

Butcher, Prof. S. H., Litt.D., LL.D. (V.P.), 27, Palmerston Place, Edinburgh, 


fBute, The Marquis of, K.T., St. John' s Lodge, Regents Park, N.JV. 

Butler, Arthur J., IVood End, IVeydndge. 
* Butler, The Rev. H. M., D.D., Master of Trinity ColUge, Cambridge. 
Buxton, F. W., 42, Grosvenor Gardens, S.W. 
Buxton, Mrs. Alfred W., 32, Great Cumberland Place, VV. 
By water, Prof. Ingram (V.P.), 93, Onslow Square, S. IV. 
fBywater, Mrs., 93, Onslow Square, S. IV. 
Calcutta, The Right Rev. the Lord Bishop of, T/te Palace, Calcutta. 
Calvert, Rev. Thomas, 
fCalvocoressi, L. M., Messrs. Ralli Bros., Mellor's Buildings, Exchange Street East, 
Campbell, Rev. Prof. Lewis (V.P.), 33, Campden House Chambers, W. 
Campbell, Mrs. Lewis, 33, Campden House Chambers, IV. 
Capes, Rev. W. W., Bramshott, Liphook, Hants. 
Carapdnos, Constantin, D^put^, Athens. 
Carey, Miss, 13, Colosseum Terrace, Regent's Park, N. W. 
*Carlisle, A. D., Haileybury College, Hertford. 
Carlisle, Miss Helen, Houndhill, Marchington, Stafford. 
fCarmichael, Sir T. D. Gibson, Castlecraig, Doiphinton,N.B. 
tCarr, Rev. A., Addington Vicarage, Croydon. 
Carr, H. Wildon, 25, Cumberland Terrace, Regent's Park, N.W. 
Carter, Prof. Frank, McGill University, Montreal. 
Carthew, Miss, 15a, Kensington 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. IV. 
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. 
fChawner, G., King's College, Cambridge. 
•fChawner, W., Master of Emmanuel College, Cambridge. 

Cheetham, J. C M., Eyford Park, Bourton-on-the- Water, R.S.O., Gloucestershire. 
Cheetham, J. Frederick, Eastwood, Staleybridge. 

Christian, J. Henry, 18, Devonshire Place, Portland Place, W. 

Christian, Rev. G., Redgate, Uppingham. 

Christie, Miss Annie, i, Westbourne Terrace, W. 
♦Christie, R. C, Ribsden, Bagshot, Surrey. 

Churchill, E. L., Eton College, Windsor. 

Clark, Charles R.R., cjo E. P. Warren, Esq., 18, Cowley Street, Westminster, S. W. 
t Clark-Maxwell, Rev. W. Gilchrist, Clunbury Vicarage, Aston-on-Clem, Salop. 

Clarke, Joseph Thacher, 3, College Road, Harrow, N. W. 

Clarke, Somers, 22, Whitehall Court, S.W 
fClauson, A. C, 12, Park Place Villas, Paddington, W. 

Clay, C. F., 38, Great Ormond Street, W.C. 

Clerke, Miss Agnes, 68, Redcliffe Square, S. W. 

Cobbold, Felix T., The Lodge, Felixstowe, Suffolk. 
*Cobham, C. Delaval, H.B.M. Commissioner, Larnaca, Cyprus. 

Colby, Rev. Dr., 12, Hillsborough Terrace, Ilfracombe. 

Cole, A. C, 64, Portland Place, W. 

Colfox, William, Westmead, Bridport. 

Collins, Miss F. H., 3, Bramham Gardens, South Kensington, S. W. 

Colvill, Miss Helen H., Overdale, Shortlands, Kent. 

Colvin, Sidney (V.P.), British Museum, W.C. 

Compton, Rev. W. C, The College, Dover. 

Connul, B. M., The Yorkshire College, Leeds. 
•Constantinides, Prof. M., Coundouriotes Street, Munychia, Peiraeus, Athens, 

Conway, Sir W. M., The Red House, 21, Hornton Street, W. 

Conybeare, F. C, 13, Norham Gardens, Oxford. 

Cook, Arthur Bernard (Council), Trinity College, Cambridge. 

Cookson, C, Magdalen College, Oxford. 

Cookson, Sir C. A., C.B., H.B.M. Qon^yxX, Alexandria. 

Corbet, His Honour Eustace K., Native Court of Appeal, Cairo. 

Cordery, J. G., C.S.I., 63, Goldington Road, Bedford. 

Corgialegno, M., 21 , Pembridge Gardens, W. 

Courtenay, Miss, 34, Brompton Square, S. W. 

Courtney, W, L., 53, Belsize Park, N. IV. 

Cowper, The Right Hon. Earl, K.G., Panshan^er, Hertford. 

Craik, George Lillie, a, West Halkin Street, S. W. 

Crewdson, Wilson, The Barons, Reigate. 

Crooke, W., Westleigh, Arterberry Road, Wimbledon^ S.W. 
tCrossman, C. S., The College, Winchester. 

Crowfoot, J. W.,, Bishop's Hostel, Lincoln. 

Cruickshank, Rev. A. H., 'The College, Winchester. 

Cust, H. J. C, St. fames' Lodge, Delahay Street, S. W. 

Cust, Lionel, The Crescent, Windsor. 

Cust, Miss Anna Maria, 63, Elm Park Gardens, Fulham Road, S. W. 

Cust, Miss Beatrice, 13, Eccleston Square., S.W. 

Dabis, Miss, Holloway College, Egham, Surrey. 

Dakyns, H. G. (CoxxncW), Higher Coombe, Haslemere, Surrey, 

Danson, F. C, B., Liverpool and London Chambers, Liverpool. 

Darbishire, B. V., Trinity College, Oxford. 

David, Rev. W. H., Kellf College, Tavistock. 

Davidson, H. O. D., Harrow, N.W. 
fDavies, Prof. G. A., University College, Liverpool. 

Davies, Rev. Gerald ?>., Charterhouse, Godalmtng. 

Delamarre, Jules, 4, Impasse Royer-Collard, Parts. 

De Saumarez, Lord, Shrubland Park, Coddenham, Suffolk. 

Dickson Miss Isabel A., Dunnichen 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. 

Draper, W. H., 52, Doughty Street, W.C. 

Drummond, Allan, 7, Ennismore Gardens, S. W. 

Duchitaux, M. V., 12, Rue de VEchauderie, d Reims. 

Duhn, Prof, von. University, Heidelberg. 

Duke, Roger, 8, Neville Terrace, Onslow Gardens, S. W. 
t Dunham, Miss, 37, East Thirty- Sixth Street, New York. 

Dunlap, Miss Mabel Gordon, c/o Messrs. Brown, Shipley & Co., Founder's Courii 
Lothbury, E.C. 

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. 

Edmonds, C. D., Emmanuel College, Cambridge. 

Edwards, G. M., Sidney Sussex College, Cambridge. 

Egerton, Mrs. Hugh, 11, Tite Street, Chelsea, S. W. 
fEgerton, Sir Edwin H., K.C.B., H.B.M. Minister, British Legation, Athens, Greece. 

Egerton, Miss M., Whitwich Hall, York. 

Eld, Rev. F. J., Polstead Rectory, Colchester. 
t Ellis, Prof. Robinson, Trinity 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. 

Eumorfopoulo, N., i, Kensington Park Gardens, W. 

Evans, A. J. (V. P.), Ashmolean Museum, Oxford. 

Evans, Sir John, K.C.B., D.C.L., F.R.S., Nash Mills, Hemel Hempstead. 
t Evans, Lady (Council), Nash Mills, Hemel Hempstead. 
Eve, H. W., 37, Gordon Square, W.C. 
Ewart, Miss Mary A., 68, Albert Hall Mansions, S. H^. 
Fanshawe, Reginald, 53, Canynge Road, Clifton, Bristol. 
Farnell, L. R., Exeter College, Oxford. 
Farrar, Rev. Canon A. S., Durham. 

Farside, William, Thorpe Hall, Robin Hoods Bay, Yorkshire. 
*Fearon, Rev. W. A., D.D., The College, Winchester. 
Fenning, W. D., Haileybury College, Hertford. 
Field, Rev. T., Radley College, Abingdon. 
Fisher, H. A. L., New College, Oxford. 

Fisher, Miss Edith S., 21, Chapel Street, Belgrave Square, S. W. 
fFitzmaurice, Lady Edmond, 2, Green Street, Grosvenor Square, W. 
Flather, J. H., 52 Bateman Street, Cambridge. 
Flower, Wickham, Old Swan House, Chelsea Embankment, S. W. 
t Forbes, W. U.,Balliol College, Oxford. 
Ford, The Right Hon. Sir Francis Clare, G.C.B., G.C.M.G., 17, Park Lane, W. 
Fowler, Harold N., Ph.D., Western Reserve University, Cleveland, Ohio, U.S.A. 
* Fowler, Rev. Professor, President oj Corpus Chris ti College, Oxford. 
Fowler, W. Warde, Lincoln College, Oxford. 
FVanklin, 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), i, Airlie Gardens, Campden Hill, W. 
fFreshfield, Edwin, LL.D., 31, Old Jewry, E.C. 
Freston, Henry W., Parkfeld, Prestwich, Lancashire. 
Fry, Right Hon. Sir Edward, Failand House, Failand, near Bristol. 
*Fry, F. J., Eversley, Leigh Wood, Clifton. 
Fullerton, W. Morton, Rue Vignon,Paris. 
tFurley, J. S., 10, College Street, Winchester. 
Furneaux, L. R., Rossall School, Fleetwood. 
Furneaux, Rev. W. M., Repton Hall, Burton-on-Trent. 
Gardner, Miss Alice, The Old Hall, Newnham College, Cambridge. 
t Gardner, Prof. Ernest A. (Council), Radnor Cottage, Sandgate. 
*tGardner, Prof. Percy, Litt.D. (V.P.), 12, Canterbury Road, Oxford. 
Gardner, Samuel, Oakhurst, Harrow-on-the-Hill. 
Gardner, W. Amory, Groton, Massachusetts, U.S.A. 
Gamett, Mrs. Terrell, 3, Queen Annis Gate, S. W. 
Gatliff, Hamilton, 11, Eaton Square, S. W. 
Geddes, Sir W. V>., Principal of the University , Aberdeen. 
Geikie, Sir Archibald, F.R.S., 10, Chester Terrace, Regenfs Park, N. 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., Ridgjield, Wimbledon. 
Gonino, Miss G., 90, Warwick Street, Warwick Square, S. W.