Papers of the Archaeological Institute of America

Google

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project

to make the world's books discoverable online.

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover.

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the

publisher to a library and finally to you.

Usage guidelines

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing tliis resource, we liave taken steps to
prevent abuse by commercial parties, including placing technical restrictions on automated querying.
We also ask that you:

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for
personal, non-commercial purposes.

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the
use of public domain materials for these purposes and may be able to help.

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for in forming people about this project and helping them find
additional materials through Google Book Search. Please do not remove it.

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner
anywhere in the world. Copyright infringement liabili^ can be quite severe.

About Google Book Search

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web

at |http: //books .google .com/I

papers of \\t ^rt^^ologital Institnle of America.

CLASSICAL SERIES.

I.

REPORT ON THE INVESTIGATIONS AT
ASSOS. i88i.

By JOSEPH THACHKR CLARKE.

SQitb an Spptntii,

CONTAINING INSCRIPTIONS FROM ASSOS AND LESBOS,

AND PAPERS

Bv W. C. LAWTON AND J. S. DILLbJt

BOSTON:

PUBLISHED HY A. WILLIAMS AND CO.

LONDON: N. TRUBNER ANI> CO,

1882.

Dfu

University Press:
John Wilson and Son, Cambridge.

■L^i^n STA^^rnrn jr. jjivERSuy.

ARCH^OLOGICAL INSTITUTE OF AMERICA.

'ExtcnXxvt Committee, 1881-82.

CHARLES ELIOT NORTON, President.

MARTIN BRIMMER, Vice-President.

FRANCIS PARKMAN.

W. W. GOODWIN.

H. W. HAYNES.

ALEXANDER AGASSIZ.

WILLIAM R. WARE.

HENRY L. HIGGINSON, Treasurer.

E. H. GREEN LEAF, Secretary,

CONTENTS.

Pacb

Preliminary Report of the Investigations at Assos dur-
ing THE Year i88i. By Joseph Thacher Clarke ... i

APPENDIX.

I. Inscriptions found at Assos and Mitylene .... 133

II. Notes on BunArbashi and other Sites in the Troad,
BY William C. Lawton ; including Notes on the
Map of the Acropolis of the Bali Dagh, by C.
Howard Walker 143

III. The Geology of Assos. By J. S. Diller 166

IV. Notes upon the Geology of the Troad. By J. S.

Diller 180

\

LIST OF PLATES.

■^

Pagb

1. Plan of Assos i

2. Topographical Sketch of Acropolis after Excava-

tions 29

«

3. Topographical Sketch of Plateau with Stoa,

Theatre, &c 35

4. Topographical Sketch of Gymnasium 40

4«. Map of Aeolic Mysia and Lesbos 48

5. Section of Acropolis 52

6. View of Acropolis 52

7. Plan of P'loor of Temple 80

8. Plan of Temple, restored 85

9. Drum of Column, and Fragment of Corona .... 88

10. Outline of Echinus and Neck of Column 89

11. Section of Temple Order 91

12. Head of Lion from Gutter 94

13. Roof of Temple, restored 95

14. Front of Temple, restored 100

15. Relief from Epistyle of Hercules and Centaurs . . 107

16. Relief from Epistyle of Two Sphinxes 111

17. Relief from Epistyle of Lion and Boar 113

18. Relief from Epistyle of Hind-quarters of Lion . . 114

19. Relief from Epistyle of Fragment of a Sphinx . . 115

20. Relief from Epistyle of Fragment of a Centaur

with Fore Legs of a Horse 116

21. Metope Relief, Man pursuinc; a Woman 117

22. Metope Relief, Two Warriors 117

23. View of Mosque and Tower 122

24. Door of Mosque 123

■ • •

viil LIST OF PLATES.

Page

25. Mosaic Pavement from Gymnasium 124

26. Wall of Polygonal Masonry 125

27. Portal in Western Wall 125

28. Tower at Northwest Gateway 125

29. Section of Cemetery, restored 126

30. EXEDRAS from CeMETERY, RESTORED 1 27

31. Receiving Tomb 126

32. Receiving Tomb, Plan and Section 127

33. Sarcophagus 127

34. Sarcophagus, restored 127

35. Bridge on the Satnioeis 128

36. Turkish Port and Mole 131

37. Coin of Assos 131

I. Bronze Tablet, with Inscription on Accession of

Caligula 133

II. Acropolis of the Bali Dagh from the North . . . 149

III. Acropolis of the Bali Dagh from the South . . . 151

IV. Map of Acropolis of the Bali Dagh 153

V. Trojan Plain from the Bali Dagh 156

VI. Trojan Plain from Hissarlik 162

VII. Geological Map of Assos 166

\"

/

' '/

h

^ Q

[i.
J

D

-\\

1

\ \

PRELIMINARY REPORT

OF THE

INVESTIGATIONS AT ASSOS

DURING THE YEAR 1881.

THE following account of the first year's work of the
expedition to Assos sent out by the Archaeological
Institute of America must be prefaced by a reminder that
the time has not come for a thorough and conclusive publi-
cation of the results achieved. It is evident that descriptions
of monuments but recently discovered, and in part still hidden
beneath the earth, will be extended, and possibly corrected,
as the studies upon the site advance. Indeed, many 4X)ints
are touched upon in this Report only to indicate the direc-
tion and scope of the work. After the termination of the
investigations, it is hoped to present the. full results in a
monumental volume upon Assos and the Southern Troad.

The present Preliminary Report is divided into two parts.
The first — more or less introductory — contains a notice of
the visits to the site by travellers and archaeologists during
the past century, and an account of the present expedition.
The second treats of the geographical conditions of the region
and their influence upon the development of Assos during
antiquity, of the history and topography of the city, the ar-
chitectural monuments investigated, and the sculptures and
inscriptions discovered.

2 ARCHAEOLOGICAL LWSTITUTE.

A description of the temple and its important reliefs is
given in detail. The account of other buildings is less full, a
consideration of many points already determined being re-
served for the report upon the fortification walls, theatre,
stoas, gymnasium, etc., which will be prepared after the close
of the work of the second year.

An appendix is furnished by the geologist of the expedi-
tion, Mr. Joseph Silas Diller, containing the results of his
special studies at Assos and excursions in the Troad.

The last sparks of Greek civilization, the various phases of
which had for twenty-four centuries been exhibited at Assos,
were extinguished by the Latin conquest of Constantinople.
The establishment of the Genoese principality of Lesbos was
soon followed by the inroads of the Turks. Assos was de-
serted and forgotten. Its ruins are to-day a nameless append-
age to the squalid village of Behrim,^ perhaps so called from
one of the emirs serving under the conqueror Orkhan. Once
the most important city of the Troad, it is now represented
by a hundred miserable dwellings. Its commercial prosperity
declined with the failure of the agricultural energy which

^ The orthography of personal and geographical names in the present Report
requires a word of explanation. Turkish names — as derived from an alphabet
wholly distinct from the English — are rewritten according to their sound, the
letters having in every case the value peculiar to them in English. It is impos-
sible, however, by any combination of letters, to convey the sound of the Z — the
sharply aspirated /T, like the German cA in acA — which occurs in the name
Behr^m; and it is to be observed that the final « so frequently employed
(Pademle^, Sazle^, Narle^, etc.) approaches in character the French «, or Ger-
man u^, a vowel not known in English.

The Greek spelling of Greek names has been adopted whenever the word has
not, by long use, become fully Anglicized; that is to say, changed m pronuncia-
tion. As the English alphabet provides two letters for the Greek Kd-mroj the c
has been employed as the more familiar (Corinth, Acropolis, etc.), except in
cases where the true sound is not thereby conveyed : namely, before r, /, and r,
when the Jt has been substituted. As no English word ends in /, the final ai is
transformed to ae, according to the universal usage of our tongue.

INVESTIGATIONS AT ASSOS, 1881. 3

once had produced upon the plains of the Satnioeis the finest
wheat known to the Persian court. The land became a
stronghold of Mahometan fanaticism. The austere and
bigoted character of the Turks of the Troad was remarked
by early travellers,^ and it is still uncomfortably evident. It
is true that no open attempt is now made in times of peace
to persecute unbelievers ; but in their presence there is a
lowering constraint on the part of the men, while women
hasten from the sight of an infidel, or, crouching behind some
shelter, shield the terrified children with their skirts.

Strange as it may appear to those acquainted with the
mixed population of the more southern coasts of Asia Minor,
it has been only within late years that Greek settlers have
been able to gain a foothold in the Southern Troad. At
the important port of Babi-calessi ^ there is but one Greek
merchant ; at Behrim only one magazine • is Greek ; and in
the interior the number of Christian inhabitants is very
small. That the time is rapidly approaching when the land
will be held in great part by Greeks, no one can doubt who
has observed the progress of that people, and the melting
away of the Turkish population.

The Gulf of Adramyttion is dominated by Ivalee,^ opposite
the northern coast, and distant from it but two hours' sail.
The modern history of this city well illustrates the position of
. the two races, and foreshadows the development of the Troad
in the near future. The inhabitants of the Turkish town of
Ayasmat* totally destroyed Ivalee during the War of Inde-
pendence, confiscating the olive orchards and the vineyards

1 Michaud et Poujoulat, work quoted below, p. 9.

^ The promontory of Lecton is known to the Turks as Bab^ ; the town at its
extremity as Babk-calessi, «>., Babk-castle, from the considerable fortifications
and garrison there maintained.

• On the site of the ancient Heracleia.

* On the site of the ancient Attea, or Attalia.

4 ARCHJSOLOGICAL INSTITUTE.

in its neighborhood. To-day, however, nearly forty thou-
sand Greeks inhabit Ivalee, and not one Turkish family;
while for miles around the city all the land is again in the
possession of Greeks. Ayasmat, on the other hand, has
dwindled to a squalid village of twenty or thirty huts, with a
Moslem graveyard more than a mile long.

The sparsely populated northern coast of the Gulf of Adra-
myttion lies off the line of the marine traffic, which the for-
mation of land and sea has led into fixed courses in this part
of the Mediterranean. The steamers which constantly ply
between Constantinople and Smyrna seek escape in the Chan-
nel of Mytilene from the high winds which disturb the open
iEgean, and pursue their sheltered course along the island ;
and travellers commonly pass through the strait without giving
much attention to the steep and sterile volcanic plateau, which
rises toward the sea as a wall, enclosing the isolated valleys
where trickling streams maintain a luxuriant verdure through-
out the long heats of summer. The smallest coasting vessels
are seldom forced to make the northern coast of the gulf at
any point east of Babi. Some twenty years ago the Aus-
trian steamers stopped at Bab^-calessi ; but this route was
abandoned, from the lack alike of freight and of passengers.
The annual crop of valonia (the cups of the acorn of Quer-
cus cegilops) and the occasional surplus of wheat grown in the
alluvial plain of theTouzla are exported by native merchants.

Though Edremit (Adramyttion), at the head of the gulf,
has remained a populous town under the Turks, the com-
mercial prosperity which it enjoyed under the rule of the
kings of Pergamon has been wholly lost. During the cen-
turies in which great thoroughfares existed from Pergamon to
the Hellespont by way of Adramyttion, the distance of eight
kilometres between the city and the sea and the lack of an
adequate harbor were not obstacles that prevented the city

INVESTIGATIONS AT ASSOS, 1881. 5

from having a thriving trade ; * the port continued to be a con-
siderable emporium as late as the time of the Latin princes, but
under Turkish rule it had become almost entirely deserted by
the middle of the last century. The products of the fertile
land in the vicinity of Edremit now pass into the hands of the
ever busy Greeks, and are carried to Ivalee by the ten thou-
sand camels of this Vilayet, being thus still further removed
from the northern coast of the gulf.

So little have these waters been frequented by well-manned
European vessels, that even in our days the nooks of the Gulf
of Adramyttion have been among the last resorts of Greek
pirates, — sharing notoriety, in this respect, with the shores
of inhospitable Amorgo.

In short, the isolation of the Southern Troad, by reason of
the configuration of the land and the peculiarities of its inhab-
itants, was so complete, that at the beginning of this century,
when the present Renaissance of Greek thought and art was
far advanced in Attica, and when even the neighboring plain
of Troy was familiar to us from the reports of many travel-
lers, all our knowledge of Assos was restricted to the im-
perfect description given by Count Choiseul-Gouffier in his
"Picturesque Voyage." ^ The Count had made his first jour-
ney to the Levant in 1776; his appointment, eight years
later, as Minister Plenipotentiary of France to the Porte,
gave him an exceptional opportunity for the completion and
extension of studies which, though in many respects of narve
inaccuracy, were of great value in calling the attention of
European scholars to sites previously unexplored.

* Compare Acts of the Apostles^ xxvii. 2.

* Vd^age Pittoresque de la Grhe, tome second, taris, 1809 J PP* 86-88. In
1819 Choiseul's map received some corrections and additions founded upon
the observations of M. Dubois, who had been sent to the Troad in the preced-
ing year by M. de Choiseul-Gouffier. The first volume of the Voyage Pittoresque
was published in 1782.

6 ARCHAEOLOGICAL INSTITUTE.

M. de Choiseul gives a strange and confused plan of the
city/ and a wonderful restored view of the site. The letter-
press is better than the illustrations, and affords a compilation
of the remarks of ancient authors bearing upon the subject,
so thorough as to suggest the work of a literary assistant.
The erroneous assumption of three temples at the foot of the
Acropolis is not surprising, being evidently based on the pecu-
liar formation of the stoa plateau, with its terraces at either
end. It is worthy of remark that this first account with all
its shortcomings yet shows its author's appreciation of the
striking situation of the city, which has not failed to kindle
the admiration of every subsequent traveller. '' Paide vilies**
says the author, ''jouissent (Tune sitiiaticn aussi hcurcuse, aussi
magnifique que celle d'Assos ; l imagination des plus habiles
artistes tie sauroit alter au-deld des tableaux, si riches, si impo-
sans, quelle devoit jadis pr^sentir de toutes parts''

But though the detailed plan and restoration of the city,
given in the " Picturesque Voyage," were fanciful and incor-
rect, the accompanying maps of the Troad were long the chief
source of information for that important part of Asia Minor,
being even reproduced with but few alterations as late as the
publication of Mauduit's book upon the Troad.^

The influence of the "Voyage Pittoresque" is evident, from
the fact that nearly half of the travellers who have subse-
quently visited and described Assos have been French, that
the only extended investigations upon the site were made at
the expense of the French Government, and that the cele-
brated reliefs of the temple were finally obtained by the
Louvre, and transported to France on a national vessel.

Eight years before the appearance of Choiseul-Goufficr's

* And yet the author congratulates himself that, ** Le plan qu'offrc la planche
IX. a ^t^ lev^ avcc exactitude."

^ Dicotevertes dans la Troade. Paris. 1844.

INVESTIGATIONS AT ASSOS, 1881. 7

volume Assos was mentioned by M. Olivier,^ in a book which
gives much information concerning the condition of the Troad
during the last decade of the eighteenth century ; but the
author did not land at Behrim, contenting himself with ex-
amining the coast from his vessel.

That eminent authority updn the topography of ancient
Greece, Colonel Leake, visited Assos in the year 1800. His
short notice of the site was first published in 18 17 in the
continuation of Walpole's " Memoirs relating to European
and Asiatic Turkey," ^ and several years later appeared in his
own " Journal."^ This writer, whpse extended travels and great
erudition give his opinion decisive weight, considered the re-
mains of Assos to present the most perfect idea of a Greek
city that is anywhere to be obtained.

Dr. Hunt saw the ruins one year after Leake ; his report
was the first to be printed,* though not till sixteen years
after his visit. Hunt's accurate and detailed account of the
theatre is particularly valuable, and his description of the
temple, the porticos with their inscriptions, the antique
edifice used as a Turkish bath, etc., cause wonder that the
ruins above ground should have remained in so perfect a
state of preservation so late as the beginning of the present
century, and regret that the excavations advised by him should
not then have been undertaken. Well might they have been

* Voyage dans P Empire Ottoman ^ P Egypte et la Perse, Fait par ordre du Gouv-
emement pendant ies six premieres ann^es de la Ripublique^ par G. A. Olivier.
Paris. An 9. Vol. i., chap. xxv.

2 Travels in Various Countries of the East; being a eontinuatiofi of Memoirs
relating to European and Asiatic Turkey. Edited by Robert Walpole. London.
1820.

8 Journal of a Tour in Asia Minor^ with comparative remarks on the ancient
and modern geography of that country. By William Martin Leake. F. R. S., etc.
London. 1824.

* Memoirs relating to European and Asiatic Turkey^ edited from manuscript
journals, by Robert Walpole. London. 1817. Number VL Account of Dr.
Hunt and Prof. Carlyle.

8 ARCH^OLOGICAL INSTITUTE.

" repaid by the discovery of many valuable works of art," had
they been prosecuted before the lamentable destruction of
later years.

Hunt was succeeded by Von Richter, whose interesting
sketch, valuable especially in its description of the walls, was
published seven years after, in a book which is the best monu-
ment to one who found an untimely grave while in the midst
of his Oriental investigations. Von Richter visited Assos in
June, 1816. It was upon this journey that he caught the
fever which left him scarcely time to relate his observations
in his journal, published by Ebers.^

Philip Barker Webb's studies upon the Trojan Plain were
extended to Assos, and were first printed in Italian, in the
"Biblioteca Acerbi," in the volumes for June and July, 1821.2
It was through him that attention was first called to the inter-
esting geological character of this volcanic region.

At a later date the vicinity was explored and described
from a scientific point of view by the eminent Russian geog-
rapher Tchihatcheff,^ whose routes upon the map, given to
illustrate his itinerary, show him to have visited Assos in
1847 ^"d 1849. It is greatly to be regretted that the fourth
part of Tchihatcheff's great work, which was to be devoted to
the statistics, politics, and archaeology of the country, should
never have appeared. His most interesting results, if not
wholly lost, have thus been too greatly delayed to be of full
service to science.

^ Otto Friedrich von Richter. Wallfahrten im Morgenlande. Aus seinen
Tagtbuchern und Brief en dargesUllt von Johann Philip Gustav Ebers. Berlin.
1822.

2 Better known in a later French edition : Topographie de ia Troade. Paris.
1844. Webb complains, in the preface to the republication, that the studies had
not attracted due attention in their original form. They had meanwhile been
translated into German ; but this work docs not seem to have appeared in a large
edition, as it is rare and little known, notwithstanding its importance.

8 Asie Minenre, description physique, statistique et arckiologiqtu cU cette contrie.
Par Pierre de Tchihatcheff. Paris. 1S53-1869.

INVESTIGATIONS AT AS SOS, 1881. 9

By far the best description given by any traveller is that of
Prokesch von Osten, whose most admirable book of " Oriental
Notes," ^ justly led to the author's preferment to high official
position. The letter relating to the ruins of Assos is dated
at Mytilene, July, 1826; in it the author speaks of the remains
as the best preserved of all between the Propontis and the
Ionian coast. Apart from the interest of the general account,
the technically correct descriptions and accurate measure-
ments of monuments, then still in a comparatively perfect
state, are of a value to the present investigations which may
be estimated from the fact that the given dimensions of the
theatre and fortification walls, for instance, are not only more
trustworthy, but more complete, than those in the pretentious
work of the later French expedition under Texier.

After a lapse of thirty years, when the writings of Choiseul
and Olivier had become antiquated, the attention of the French
was again called to the ruins of Assos by the Oriental cor-
respondence of Michaud and Poujoulat.^ These companions
were separated at Babi, — Poujoulat going on horseback to
Behr^m, while their coasting vessel, upon which Michaud re-
mained, ill of a fever, was driven from the insecure port at the
cape by a storm of wind. Poujoulat's description of his jour-
ney to the ruins of Assos is graphic ; but his understanding
of the antique was inadequate and led him into absurd mis-
takes, a number of which will be mentioned later on.

The admirable survey of the northern coast of the Gulf of
Adramyttion, made by Commander Copeland of the English
Navy, is dated in 1834.^ Upon it the position of Behrim is

1 Denkwiirdigkeiien und Erittncrungen aus dan Orient^ vom Ritter Prokesch
von Ostcn. Aus Julius Schneller^s Nacklass hcrausgegeben von Dr. Ernst Miinch.
3 B'dnde. Stuttgart. 1836-37.

2 Correspondance d'* Orient, Par M. Michaud, de I'Acad^mie fran9aise, et M.
Poujoulat Vol. iii. Paris. 1834. Lettre Ixix.

' Charts of the English Admiralty, No. 1665. Mytilene Island.

lO ARCH^OLOGICAL INSTITUTE,

accurately designated, though the independent character of
the volcanic peak is overlooked.

It was in June, 1835, five years after the visit of Poujoulat,
that Charles Texier, commissioned by Guizot, who was then
the French Minister of Public Instruction, to study the antiqui-
ties of Asia Minor, examined the ruins. The results of his
expedition were most luxuriously published, at government
expense, in three immense volumes, in the second of which
are the plates and letter-press concerning Assos, the illustra-
tions being restricted to the fortifications and the temple of
the Acropolis.^

Unfortunately, as more recent scholars have frequently had
occasion to remark,^ the facile architect and director of the
expedition had le g^nie de V inexactitude. Texier's detailed
topographical plan of the city is hardly creditable as a sketch
from memory. The given measurements, though expressed in
the smallest fractions of the metric system, are often wholly
fictitious, the restorations largely imaginary. Even were the
present expedition to do no more than accurately to determine
the points treated with such unworthy carelessness by Texier,
it would render a definite and valuable service to archaeological
science.

By the successive labors of Poujoulat, Huyot, and Texier,
the reliefs of the epistyle Ikd metopes of the temple, which

1 Description de PAsie Mi n cure, fait par ordre dit Gouvernemcnt fran^aise de
1833 h 1837 et ptibiiie par le Ministirc de V Instruction pubUquc. Par Charles
Texier. Deuxiemc partic, deuxi^me volume Paris. 1S49. The eminent archi-
tect and archaeologist, Huyot, who had visited Assos about 181 7, and made draw-
ings of the remains, is said to have directed the attention of Texier to Assos, and
to the reliefs which lay exposed upon the sides of the Acropolis. Huyot is said
by Clarac to have attempted to earn,' off the sculptured blocks.

2 Referring to the account of Old Symrnn, given by Texier, Dr. Hirschfcld
says : " Ixiider muss dieselbe beinahe als werthlos Ix'zcichnet werden ; dcnn die
elegant gezeichneten Formen entsprechen der Wirklichkeit in keiner Weisc."
(Compare the paper by Dr. Curtius in the Abhandluni^en der berliner Akadcmie,
1872.) The remarks of M. Perrot upon the plates concerning Pessinunt arc even

INVESTIGATIONS AT ASSOS, 1881. II

lay exposed upon the summit of the Acropolis and its south-
eastern slope, appear to have become regarded as due to
France; and the well-known archaeologist, Raoul-Rochette,
having succeeded in obtaining a formal grant of the blocks
as a gift of the Sultan Mahmoud II. to the Louvre, they
were removed in 1838. Through these remarkable archaic
works of sculpture the attention of every scholar of Greek
antiquity and art has been attracted to Assos. Three publi-
cations^ have made them familiar to those unable to study the
originals, or the casts exhibited in European and American
capitals.

Shortly before the reliefs were loaded upon the brig " La
Surprise,** of the French navy, they were seen upon the site
by Sir Charles Fellows, in whose interesting " Journal " ^ there

more to the point, as illustrating Texier's manner of dealing with a subject in
every way comparable to Assos : " Le plan donn^ par M. Texier ... est une
mauvaise plaisanterie. II donne des noms h. tout, il indique la disposition \\\i6-
rieure de tous les Edifices jusque dans leurs moindres details; il ne vous fait pas
grace d'une colonne, quand, de son propre aveu, il n*a pass^ li que quelques
heures, et s'est born^ ^ noter, du haut de I'acropole, la situation relative des dif-
f^rents amas de decombres qu'il apercevait dans diff^rentes directions." (Lcttre
de M. Perrot h, M. Renicf. Bidlettino deW Instituto di Corrispondenza archeo-
logica. 1861. VIII, Agosto.) A full review of Texier's shortcomings in regard
to Assos would here lead to too great length ; a number of points will, however,
be mentioned in the consideration of the temple. *

1 In lithographed plates, with two pages of inadequate text, by M. F. de
Wittc, in Annali deW Instituto di Corrispotidensia archeologica. Volume tredi-
cesimo. Roma, 1842 ; and in Monumenti inediti pubblicati deir Instituto, etc.
III. Roma e Parigi, 1839-43. In the second volume of Texier's Description de
r Asie Mineure, referred to above, and in Clarac's MusSe de Sculpture^ antique et
moderne ; ou Description historique et graphique du Louvre et de toutes ses parties y
etc. Tome II., seconde partie. Paris, 1841. Texier's engravings are the best
representations of the sculptures, though they do not include all the reliefs.
Clarac's text gives a detailed account of the removal of the blocks, and of the
sawing to which they were subjected to prepare them for the walls of the Assos
Room in the Louvre. A full review of these publications is reserved for an
essay on the temple sculptures, which is to appear among the papers of the
Institute.

•^ A Journal written during an excursion in Asia Minor. By Charles Fellows.
London. 1839.

12 archjeological institute,

are drawings of the most prominent blocks, as well as a good
general description of the ruins.

He was followed by another English traveller, signing him-
self "G. R. L.," who contributed a short but well-written notice
of Assos to the •* Gentleman's Magazine" in 1842.^

The geographical studies of Dr. Henry Kiepert and Prof.
A. Schoenborn in the Southern Troad were made at about
this date ; they will be referred to below in the consideration
of the maps of the land.

In 1842 Professor Purearitis, of the University of Athens,
published a slight account of the ruins in the iVea Tlavloapa?
interesting only as a proof that the destruction of later years
had not then begun, — the seats of the theatre still being in
perfect preservation.

The next account was printed by Mr. Pullan, in 1865, in a
work which, so far as it refers to Assos, is a partial translation
of Texier's text, illustrated by lithographic reproductions of
the French engravings.^

Mr. Abbot, of the Foreign Office, visited Assos subse-
quently to Mr. Pullan ; his admirable report has been recently
printed.*

At the time of Mr. Abbot's visit, in November, 1864, a
work of systematic destruction was going on. The Turkish
Government were employing a considerable detachment of
soldiers to displace and carry from the ruins the largest and

* This paper was considered worthy of translation and republication by Ger-
man geographical journals.

^ In the number of that Athenian periodical for February i, 1862. The account
was reviewed, and in part reprinted in the Afitt/ieilungen aus Justus Perthes^
j^ographiscker Anstalt^ iiber wichti^e nate Erforschunfren auf dent Gesammtgebiete •
der Geographic . Von Dr. A. Petcrniann. Gotha. 1862.

* The Principal Ruins of Asia Minor ^ illustrated and described. By R. Popple-
well Pullan. London. 1865.

* Handbook for Travellers for Turkey in Asia, Fourth edition. London.
1878.

INVESTIGATIONS AT AS SOS, 1881. 1 3

best hewn stones. The material thus obtained was shipped
to Constantinople, and used, it is said, in the construction of
the new docks of the Arsenal at Top-haneh.* The auditorium
of the theatre, which less than twenty years ago remained
almost uninjured, was by this vandalism transformed to an
enormous quarry, the seats being piled one above another in
indescribable confusion. The chief entrance gate of the city,
one of the finest known monuments of Greek military archi-
tecture, — previously in such good preservation that it in no
wise seemed a ruin, — was in part carried away, in part wan-
tonly overthrown. Blocks spoken of as part of a Doric temple,
which had long passed for that of Augustus, were at the time
of Mr. Abbot's visit ranged side by side on the path leading
to the sea, ready for shipment.

It appears from the present aspect of the site that this
destruction was carried on for some months. The work was
undertaken as though all the remains of the city were to be
carried away ; a road was built down the most regular decliv-
ity of the hill for the transport of the stones upon rough
sledges, so that the making of a way for the reliefs taken from
the Acropolis by the present expedition was greatly facilitated.
The overthrow and removal of these stones must have been
the most severe blow ever experienced by the ruins of Assos.
The lime-burners of the Middle Ages had destroyed every
vestige of marble to be found upon the surface ; that the re-
maining monuments of volcanic stone should so very recently
find a similar fate is indeed deplorable. The carved archi-

^ The present writer has twice examined the arsenal docks, and indeed the
entire water-front of Toi>haneh, on one occasion in company with the geologist,
Mr. Diller ; but no blocks of the characteristic trachytes of Assos could be dis-
covered. Most of the stone used in their construction is a grayish limestone,
evidently taken from antique buildings, though not from any ruins of the Troad.
If the material obtained from Assos found its way to Top-haneh at all, it must
have been used for foundations beneath the water.

14 ARCHAEOLOGICAL INSTITUTE,

tectural fragments, which still thickly cover the city enclosure,
only indicate the great relative wealth of the site.

The misfortune of Assos should stimulate archaeological
investigation in lands suffering under the Turkish Govern-
ment. The insufficiency of previous investigations, like those
of Texier, is keenly felt. Our knowledge of the remains at
Paestum, for instance, or even at Athens, is already such that
their total destruction could not wholly deprive us of their
lessons. But in Assos, as in countless sites of Asia Minor,
the case is otherwise ; when their monuments have been so
demolished that restoration is not possible, the loss to science
is irreparable.

During the last season of Dr. Schliemann*s excavations at
Hissarlik, that energetic explorer, accompanied by Dr. Vir-
chow,^ visited Assos while on a journey through the Troad ;
and during the past year Dr. Schliemann again visited the
site, to the pleasure of the agents of the Institute who were
then engaged upon the preliminary survey.^

It was in June, 1879, that the present writer, with his
companion, Mr. Francis Henry Bacon, visited the site for
the purpose of investigating the remains of the temple of the
Acropolis, — a monument of the greatest importance in the
history of the Doric style. The observations made during a
limited stay were presented, somewhat in the form of a review,
in the First Annual Report of the Archaeological Institute of
America.^ The paper concluded with a recommendation of
the site as a promising field for more extended investigations.
There was indeed no reason to anticipate such brilliant discov-
eries of treasure as rewarded the excavators in Cyprus, at

* Beitrdge zur iMttdeskunde der Troas, von Rudolph Virchow. Aus den Ad-
handiungen der kg/. Akadtmie der Wissenchaften zu Berlin. 1879.

^ Reise der Traas im Mai 1881. Von Dr. Heinrich Schliemann. Mit einer
Karte. Leipzig. x88i.

« Notes on Greek Shares By Joseph Thacher Clarke. Pp. 145-163.

INVESTIGATIONS AT ASSOS, 1881. 15

Hissarlik, or at Pergamon ; but though the prospect of such
novel, and in great measure accidental, results was lacking
from the outset, the important additions to our knowledge of
antiquity made during the past year, and presented in the
following Preliminary Report, cannot fail to be considered
as eminently satisfactory.

The determination of the Institute to undertake the
exploration of Assos was announced in the Second Annual
Report,^ in May, 1881 ; but the preparations had begun long
previously.

Owing to the inclemency of the winter and early spring in
the Troad, it was planned not to undertake active operations
before the beginning of April ; and from the same considera-
tion it was evident that excavations would have to be sus-
pended by about the first of November. Nausiclides^ re-
marked of the country of the Hellespont that " it had no
spring and no friends," and although the reason he gave, —
that no truffles were there found, and no fish of the kind
called yXavtcia-KOf;, — may be deemed insufficient for such a
depreciation, it is true that the Troad is much more inclem-
ent during the winter months than the neighboring islands
of the ^gean, or the thickly settled tracts of the continent
which border the Caicos or the Hermos. The different char-
acter of the winter in the Dardanelles and in Smyrna is
surprising. The Troad is midway between the lands of
soft Ionian skies, where secure from frost the pink blos-
soms of the almond appear during the first days of Feb-
ruary, and those high and sterile plateaus of northern Asia
Minor, where the winters last eight months, communica-

^ Second Annual Report of the Archaological Institute of America. Cam-
bridge. 1 88 1.

^ In Athenxus, ii. 6a

1 8 ARCHAEOLOGICAL INSTITUTE,

duty would be obliged to carry their bulky merchandise
over a rugged plateau, by a path which, though winding
so as to make the distance half as far again as the coast
line, rises, near Arablir, to a height of nearly five hundred
metres. It is natural that in the face of such difficulties every
insignificant landing-place conveys the products of its vicinity
directly to Smyrna, — the slow and difficult voyages of the
small coasting vessels thus employed unfavorably aflfecting
the development of commercial resources on an adequate
scale. The failure of the Austrian steamers to maintain a
communication with Babi-calessi, before referred to, was
more owing to this hindrance of trade than to any absolute
unproductiveness of the Southern Troad.

On arriving at Smyrna, the goods of the expedition, as con-
sisting solely of scientific instruments and personal property
not intended for sale, were permitted to enter the country free
of duty, after the opening of every package and the payment
of heavy incidental fees. As all means of farther transpor-
tation directly to Behrim were lacking, it was necessary to re-
ship the property to Mytilene. After a constant attendance
upon the officials at Smyrna for nearly a week, a tesker^h, or
grant of free entrance to Mytilene and Behrim, was procured.
On arrival at the island an objection was made to some ir-
regularity of form in the document, which was in fact a pre-
text to enforce, by the delay of two weeks necessary for return
mails, the payment of the eight per cent ad valorem levied as
entrance duty upon all merchandise. This was at a time —
the 9th of April — when great despatch was requisite in
order to bring the surveying instruments into the field, and
the present consular agent of the United States at Myti-
lene, Mr. Phottion, gave a personal bond that the answer
expected from the chief of the custom district at Smyrna
would bear out our assurance that the goods had been de-

INVESTIGATIONS AT ASSOS, 1881. 1 9

clared wholly free of duty. After a second examination we
were permitted to remove the cases from bond to the tempo-
rary quarters of the Expedition, where they were unpacked.

The surveying instruments were soon after carried in a
small sail-boat to Assos, and the actual work upon the site
began upon the 19th of April.

When the agents of the Institute were believed to be thus
out of reach, the custom officials of Mytilene made immediate
and unceremonious demand for the sum which could be levied
upon the cases that passed through their hands, by estimating
their value at an excessive rate. In the absence of all expedi-
tious communication with Behrim, there seemed to be no pos-
sibility of evasion on the part of the bondsman. But, unfor-
tunately for this well-conceived plan, it so happened that the
writer, unknown to the officials, had returned immediately
from the site to Smyrna on other business, and on receiving
telegraphic news from Mr. Phottion, was enabled to protest
against payment, pending the obtainment, by a further ex-
pense of time and money, of a direct order and reprimand
from the Smyrna headquarters. The friendly assistance of
Consul B. O. Duncan was efficient in this as in other junc-
tures.

During a preliminary visit to Assos no available dwelling
for the party could be found, either in the squalid village
above or in the four buildings at the foot of the cliff. It was
hence not advisable to transport at once the whole outfit of
the expedition from Mytilene to Behrim. To keep up a com-
munication between the two places until it might prove pos-
sible to settle definitely at Assos, a row-boat was bought in
Smyrna, towed to Mytilene, and there rigged with sails. It
was not until the middle of May that two large rooms could
be secured in the chief valonia magazine at the port. In the
meantime the first comers slept and stored the instruments in

20 ARCHAEOLOGICAL INSTITUTE,

a room in the granary of a kindly disposed Greek merchant,
K. Agichristo, to whom the members of the Expedition owe
many subsequent favors. Much time was at first lost by the
many voyages to and from Mytilene, distant about forty-eight
kilometres from Assos. The passage became more and more
difficult as the season advanced, owing to the prevalent and
increasing northerly winds, — the Etesians of the ancients,^ —
which blow during the whole summer. On the 27th of July
all connection with the island was severed.

The survey was by this time well advanced. A base line of
five hundred metres, running from east to west, had been ac-
curately measured in the river valley upon a sandy reach ; and
another, from north to south, was laid out on the street of
tombs, — the only tract of the high land intervening between
the stream and the sea where so long a level could be
found. From the stations thus fixed the triangulation ad-
vanced, the calculations being constantly compared with
direct measurements. The rugged character of the ground
rendered the choice of stations difficult, and greatly increased
their number, — it frequently being necessary independently to
determine points distant but a few metres in plan.

The only interference offered by the inhabitants, to whom
of course such a survey was incomprehensible and suspicious,
was the systematic destruction of station pegs, which were
almost always pulled up during the night. Recourse was had'
to engraved marks upon stones, so heavy as not to be easily
displaced. The complex triangulation being thrice repeated,
the map may be relied upon as accurate.

The transit employed was, if anything, too delicate and

^ The great imp(#tance of these winds upon the development of the lands
bordering the northeastern Mediterranean can hardly be appreciated by those
who have never lived in the Levant. A pleasant characterization of the Etesians
is given by Curtius, in his Griechische GeschichtCy bd. i. cap. i.

INVESTIGATIONS AT ASSOS, 1881, 21

cumbrous for field work. It was otherwise with the light
levelling instrument, which suffered severely from once being
overthrown by the wind. The exact determination of the
various heights, and the final decision of the question con-
cerning the curvatures of the temple stylorate, to be referred
to below, have thus to be reserved for the second year's
work.

An even greater part of the preliminary investigation than
the survey consisted in a thorough examination of the ruins
remaining above ground, — the purpose and relation of the
hewn stones gradually becoming evident by continued study
and comparison of the confused heaps of rubbish.

With the gradual completion of this work the impatience
of the explorers increased for the long-promised earadek}
or official grant, which was to allow the commencement of
digging. Permission to undertake investigations at Assos
had been definitely granted to the Archaeological Institute of
America by the Porte, through the Turkish Minister of Public
Instruction, as early as the autumn of 1880. A further assur-
ance that the document setting forth the right of excavation
was at the immediate disposition of the agents had been
required and received before the departure of any members of
the Expedition from America. But notwithstanding repeated
requests made during the winter by the American Lega-
tion in Constantinople, and even a vigorously worded note from
the Secretary of State, the earadeh was not forthcoming until
far into the summer, — months after the explorers were on
the site. And before digging could be begun under its
sanction, the document had to be presented in due form to

* An earadeh is a document given by one of the ministers of the Turkish
Government, as distinguished from a firmhn^ which is a grant dependent ulti-
mately upon the Sultan. A request to undertake excavations, like all other
matters concerning the antiquities of the Ottoman Empire, is referred to the
Ministry of Public InstructioxL

22 ARCH^OLOGICAL INSTITUTE,

the Pasha of the Dardanelles, as governor of the district in
which Behrim is situated, and to the Kymacihm of Iradjik,
as the pearest local authority.

Before these formalities had been fulfilled, even the survey
was liable to interruption. Indeed, any appearance of the
explorers upon the site before being in possession of the
earadeh was discouraged by a number of residents long
familiar with the usages of the Turkish Levant. But ex-
treme care was taken to avoid all display and intrusion upon
the villagers ; and by the time the official permission to exca-
vate arrived at Assos, the greater part of the preliminary
investigations had been accomplished.

The excuses advanced to account for this delay in granting a
document promised to the official representatives of the United
States by the Turkish Government, day by day for six months,
are so characteristic as to deserve notice. Behrim, it was
said, was situated in the Vilayet of Broussa, and the governor
of that province was cited as being strenuously opposed to
the undertaking, by foreigners, of any archaeological researches
within his jurisdiction. He was reported to have thrown all
manner of difficulties in the way, — averring that the roads
were impassable, and that commissioners were unable to pro-
ceed from Broussa to Behrim to ascertain whether public or
private interests were liable to be interfered with by the pro-
posed diggings in the vicinity, etc. While plans were being
matured to overcome this opposition, a remarkable telegram
from the Sublime Porte was received at the Conac of Midhat
Pasha, inquiring if the village of Behrim were not under his
jurisdiction, and in the Vilayet of Smyrna. It thereupon
appeared that the site was not, and never had been, com-
prised in the province of Broussa, but, like Chanac,^ the

^ The city of the Dardanelles ; situated somewhat to the southwest of the
ancient Abydos (Point Nagara).

INVESTIGATIONS AT ASSOS, 1881. 23

chief town of the district, was directly dependent upon Con-
stantinople.^

The Governor of Broussa, whose opposition had been so de-
tei:mined, and against whose will it had seemed so inadvisable
to the Porte to grant an earadeh, had nothing whatever to do
with the matter, and in all probability had never even heard
of it. After the exposition of these facts further evasions
were not attempted. The permission to excavate was at last
granted.

Another hmdrance to the advance of the investigations,
severely felt during the summer months, was the shipwreck
of the vessel which had on board the household outfit of the
party. Having left Boston at the end of January, the barque
" Fame " discharged her cargo upon St. Thomas, to which
island she had been driven from her direct course to Smyrna.
The goods of the Expedition not spoiled by salt water were
reshipped, but did not reach Smyrna until the end of June.
The presence of an agent of the Institute was required there
to attend to the legal determination of the general average
necessary before the goods could be unloaded, to conduct
similar troublesome negotiations with the customs officials to
those described above, and to forward the cases, obtained
after great delay, to Behrim.

In returning from this unpleasant detention the writer was
enabled, by the hospitality of his friend Dr. Carl Humann, to
study in Pergamon the various methods of excavation which
had been proved by long experience to be best adapted to
the peculiarities of the country. The Expedition is under
great obligations to Dr. Humann for his effective furtherance
of the work by sending to Assos, at a later date, a small body
of picked men who had been in his service since the first

1 The division of the Vilayets of Asia Minor is evident from any good map;
as, for instance, from Kiepert's well-known General Chart of the Turkish
Empire.

24 ARCHAEOLOGICAL INSTITUTE.

brilliant success of the excavations at Pergamon, by his liberal
transfer of tools adapted to the usage of the laborers, which
those brought from London were found not to be, and by the
loan of a powerful winch.

By the first of August all preparations for digging had been
made, and all requisite formalities complied with ; on the sixth,
the work at last began. It was at first difficult to obtain la-
borers on account of the scanty population of the land and
the inhospitable character of the little village. The natives,
moreover, are indolent. A well-known Ottoman proverb af-
firms, that "it is better to serve without pay than to stroll
without purpose ; " but most of the Turks in the Southern
Troad are evidently of the opinion that absolute idleness is
preferable to either. Those of the hundred and fifty male in-
habitants of Behrim who are willing to work at all are busied
in the grain-fields bordering the Touzla, or follow their rest-
less goats over the neighboring mountains in search of
verdure.

The first laborers to arrive upon the site were Greek quarry-
men from Stypsis, a village upon the slopes of Mount Elias,^
near the northern coast of Mytilene. Later on came Greeks
from Ivalee, from the villages on the north of the Adramyt-
tian Gulf, and from various parts of the island of Mytilene.
Those who had served at Pergamon were natives of Lemnos.
Greeks and Turks were employed side by side, working in
perfect harmony, and even with some spirit of emulation.

The Fast of Ramazin, which occurred during August, de-
prived us of the services of nearly all the Moslems. As during
that month no believer may touch food or drink from sunrise
to sunset, the Turks are wholly unfitted for severe or continued
exertion.^ When the night is occupied in eating, drinking,

* The ancient Mount Lepcthymnos.

* The strict observance of the Ramazan is particularly severe upon the lower

INVESTIGATIONS AT ASSOS, 1881. 25

and mutual congratulations that the long hours of privation
are over, the day can be spent only in sleep and inaction. On
the conclusion of this distressing period, however, the Turks
of neighboring villages came in numbers to be engaged on the
excavations, and were particularly valuable as forming a link
between ourselves and the inhabitants of the vicinity, — to
explore which was among the purposes of the Expedition.

As a general rule the Greek proved a more diligent and
intelligent laborer than the Turk. There were, however, note-
worthy exceptions in favor of the Moslems, especially in the
case of some discharged soldiers, who had been subjected to
the privations and discipline of late campaigns.

The men were paid at the uniform rate of one-half a medjid
(about forty-one cents) a day. This sum is a trifle larger than
the average given to navvies upon the Smyrna railroads ; but it
was found that the best workmen, when employed at graded
wages, were in the end the cheapest, and the comparatively
small staff needed at Assos was made efficient and trust-
worthy by weeding out all but capable and diligent men.

The number of laborers employed never exceeded thirty-
five, averaging about twenty-six during the last half of the
work. The hours of labor were from half-past five in the
morning until th6 same time in the afternoon, including two
hours' recess, — a half-hour for breakfast, and one and a half
hours at noon. A short siesta at the time of the sun's great-
est blaze seems to be a necessity of the climate. The duty of
the superintendent, beside the oversight of the trenches, com-

classes when it occurs during the summer, the long parching days provoking
intolerable thirst, and the least exertion in the fields causing exhaustion. The
lunar month devoted to the Fast naturally occurs in every season of the year dur-
ing the course of thirty years. Its effect upon land and people is pitiable ; it is
astonishing that human beings can subject themselves to such abstinence. The
precepts of the Fast are carried out in the austere Troad with scrupulous fidel-
ity. As the Turkish word for smoking unfortunately signifies " to dritU: smoke,"
the believers are deprived even of that incomparable solace.

26 ARCH^OLOGICAL INSTITUTE.

prised the economy of the laboring forces. The proportion
of pickaxes, shovels, and wheelbarrows in use varied from day
to day according to the nature of the ground, and much de-
pended upon a wise adjustment. Though the exertion of the
pickers in the stony earth was greater than that of the other
laborers, their task was for some reason considered to be
more honorable ; and the older and better men were not easily
induced to handle a shovel, much less to trundle a barrow.
An esprit'de-corps, a spirit to which the modern Greeks greatly
owe their advance, was soon developed, resulting in a system
which otherwise it would have been difficult to introduce,
and impossible to maintain. Under its influence the inde-
pendence and marked individuality of the laborers proved to
be decidedly favorable to the work. Quarrels and drunken-
ness were unknown.

It is the custom in the Levant that large bodies of laborers
should be abundantly supplied with drinking-water by the con-
tractor. As all the trickling springs of the village cease to
flow by July, the supply had to be brought from the half-
stagnant river below in large earthen jars slung upon either
side of an ass, much like the amphorce 7iasitcrncc of the
ancients.

The food of the men was that which has supported the work-
ing classes of the land from the earliest ages of Hellenic civili-
zation. Bread was prepared by the Greek baker of the port
in the same manner, and the loaves were of the same shajje
as in the fifth century, b. c. White goat's-cheese and onions,
or leeks, were eaten with it, — a\<f>iTa, o-^ov he Kpofiva koI
rvpov} — while the rich black olives, — pvaaX koX SpvireTreU,^
— so preferable to the green fruit exported to northern lands,
took the place of meat.

^ Plutarch : On the Glory of the Athenians in War and Wisdom^ § vi.
2 Athcnaeus, li. 56, and iv. 137.

INVESTIGATIONS AT ASSOS, 1881. 27

The food supplied by Behrim and its vicinity is scanty and
monotonous. Beside fowls the only meat obtainable is the
stringy flesh of goats, and occasionally mutton ; no vegetables
whatever are grown, with the exception of onions, tasteless
gourds, and the so-called dolmaSy or meljinas {Solanum 'melon-
gena Z.), in appearance similar to our egg-plants, but immeas-
urably inferior. Fine figs are grown in the few valleys where
the burning sun does not parch the scanty soil, but these and
pomegranates are the only fruit. Goat*s-milk is seldom to be
had fresh, the most rational manner of eating it in this climate
being in a curdled 'condition (yaoirt), or made into an acrid,
chalk-like cheese. Of fish, cuttle-fish, and octopods there is,
however, an abundance ; and the bread baked by the natives
is excellent. We could not accustom ourselves to the snails
and sea-urchins eagerly sought by the workmen during holi-
days. Wild honey was occasionally brought from the neigh-
borhood, and reminded us of the appearance of the bee upon
the coins of the ancient city.^

The Greek islanders appear to have retained more Hellenic
characteristics than the inhabitants of the Morea, and their
modes of life, in primitive simplicity, present in many ways a
commentary upon the usages of the ancients. The most radi-
cal change has of course resulted from their Christianization.
During August and September, four holidays of Greek saints
interrupted the work. A further disturbance was caused by
the heavy rains of October, three week-days being lost on that
account during the first half of the month.

The heat of the midsummer sun, reflected by the sea upon
the southern slopes of the arid volcanic cliffs, was intense.
As there are no marshes, there is no malaria at the rocky port
or at the village of Behrim; but so great were the' heat and
fatigue that only one of the eight Americans who were at the

^ Sestini : Lettere numismatic he . . . cflrUinuoMiime, part. viiL p. 33.

28 ARCH^OLOGICAL INSTITUTE,

• site from time to time during the summer wholly escaped
from fever.

The laborers gradually deserted the work with the advanc-

*r* ing season, until on the first of November hardly one-third of
the entire number remained. Till the end of September the
men had slept in the open air, upon the flat housetops. The
interiors of the coffee-houses and khans were, it is true, unin-
viting dormitories, but the continued exposure in a climate
more rigorous than that of Mytilene caused much suffering
from colds and rheumatism.

The first digging in the soil of Assos for the purpose of
archaeological investigation was on the summit of the Acrop-
olis. The prospecting trench struck immediately upon the
steps of the archaic temple, which once crowned the great
natural altar. Neither walls nor columns remained in posi-
tion to mark the site, and the earth which hid the founda-
tions had accumulated to a depth averaging one and a half
metres. The first adequate description of Assos published,
that of Dr. Hunt,^ remarks that of the "temple which stood
on the citadel, parts of the shafts remain on their original site,
so that a person conversant with ancient architecture might
easily trace the plan and different details." Texier, on the
other hand, describes the summit of the Acropolis as covered
at the time of his visit with ''grandes constructions militaires
mocUrncsr It thus appears that the final levelling of the ruins
took place during the first third of this century (1801-1835),
and the accumulation of debris must, in the main, date from
that time. The comparatively recent removal of the lower
parts of the columns from their original positions is evident
from the fact that the channelled blocks, roughly built into
the Turkish walls marked FF upon the plan, Plate 2} on the

* Page 126 of the work referred to above.

3 The topographical sketch, Plate 2, exhibits the condition of the Acropolis

INVESTIGATIONS AT ASSOS, 188L 29

south and west of the citadel, are almost exclusively lower
drums.

It is possible that the uprising of the Greeks in their strug-
gle for independence may have led to the construction of the
Turkish fortification, the recent date of which is proved by
its irregularity and the entire lack of the mortar which was
lavishly used in mediaeval masonry. The lime-kilns had ex-
hausted such marble as was to be found upon the site before
the present century. Behr^m, it is true, could never have
been directly exposed to a concerted attack of the insurgents ;
but the proximity of the island of Mytilene, with its predom-
inant Greek population, may reasonably have induced the
Turks to erect defensive works on the strongest natural for-
tress of the Southern Troad.

The upper part of the columns must have been overthrown
and rolled down the steep sides of the Acropolis at a time
when the stumps of the shafts were still standing. Several
of the smaller drums were dug out from the reservoir before
the stoa ; others, hollowed at one end, have long served the

upon the termination of the year's digging. The walls in black are mediaeval,
and remain to a height of at least three metres above the ground.

A A Cemented wall, in which the fragments of the sphinxes from the eastern
front of the temple were found.

B Position of the bowman relief when discovered.

C Position of the relief of the three centaurs fitting upon the bowman.

D Position of the unbroken metope.

E Position of the sphinx from the western front, found upon the surface, and
apparently overlooked by the French in 1838.

FF Turkish walls, built without mortar, and containing many blocks of the
temple.

G G Towers and magazines of late rubble masonry.

//// Remains of early fortification walls of carefully jointed polygonal ma-
sonry.

J J Capitals placed upon edge as a rampart.

K K Pits and trenches dug by the Expedition.

L MO P Chutes used in the removal of earth.

N Position in which the relief of the lion and boar, and the hind quarters of
the lion were found.

30 ARCHjEOLOG/CAL ixstitute.

inhabitants of the village as mortars for crushing coflFee. The
cella wall, of which not a block is recognizable, was probably
removed at a far earlier period by builders covetous of its
evenly squared stones. The skeleton of columns and entab-
lature may then have stood in much the same condition as
those of the temples of Segesta and Aegina

The soil which buried the temple foundations contained no
ancient coins, and had evidently collected during the recent
occupation of the summit by Turkish constructions. It was
traversed by a complex of roughly built walls, piled up of small
stones without mortar, — in every way similar to those of the
neighboring village huts. No blocks of the temple super-
structure remained upon the stylobate.

The entire exposure of the foundations was at once under-
taken. A steep slope upon the east of the Acropolis was ex-
amined; and as no antique remains of importance were found
to exist upon the native rock, the earth carried off in wheel-
barrows was thrown over that brink, Z, Plate 2. As the work
advanced, a second chute, M, was prepared upon the western
side.

None of the sculptured epistyle blocks, which we eagerly
desired to find, were met with during this digging; but there
was much interest in tracing the plan of the building as it
emerged from the debris which so long had covered it. To
preserve from injury the upper surfaces of the stylobate and
pavements, a layer of earth a few centimetres thick was left
upon them until the very last, — thus preventing all scratching
and chipping by the iron wheels of the barrows.

When all was swept and the blocks carefully washed, the
position of eighteen of the pteroma and two of the pronaos
columns became evident by the slight weathering of the stylo-
bate surface which had occurred while the blocks were still
in place. The almost microscopical traces left by some of

INVESTIGATIONS AT AS SOS, 1881. 3 1

the shafts only displayed the outline of the channellings by
the sharp side-light of the rising or setting sun in a cloud-
less sky.

The effect of the rain upon the stucco priming, undoubtedly
once employed throughout the structure, was evident from
the grayish discolorations of those joints and clefts into which
the lime was precipitated. The clouds of sharp volcanic dust
driven upon the building by the north winds have not been
without effect ; the surfaces once protected, like the stand-
points of the columns, appearing slightly in relief when inves-
tigated by an artificial flame on still, dark nights. By the
diffused light of day these infinitesimal projections were not
visible. The coarse material of which the temple was built
was, however, not favorable to the preservation of delicate
indications of this nature.

The site of the cella walls was recognized by the delicate
incised lines traced by the Greek master-builder upon the
stones of the stylobate, to mark the position of the first up-
right blocks. Upon either end of the building pits were sunk
to the native rock to study the lower courses of the stereo-
bate, and in places where the pavement of the pteroma was
broken away its bedding was similarly examined. A detailed
account of the results thus obtained is reserved for later pages
of this Report.

Some of the marks made upon the temple plan after the
destruction of the cella walls and roof are of a curious inter-
est. Upon the pavement of the northern pteroma there is
the trace of an exploded shell, which is hard to account for,
since the last signal struggle known to have affected Assos —
namely, the invasion of the Southern Troad by Orkhan and
his emirs — was before the introduction of cannon. Upon
the foundation stone at the southwestern corner of the cella
wall the peculiar squares necessary for the old game of morris

32 ARCH^OLOGICAL INSTITUTE.

were found deeply engraved. Near the temple at a consider-
able depth a number of hand-stones for grinding wheat were
unearthed and carried off with delight by one of the old men
of Behrim, who put them at once into the primitive service
for which they had been prepared centuries before.

As before said, no coins older than the last century were
met with in uncovering the stylobate ; but in trenches dug in
other parts of the Acropolis and on the levels of the lower
temple-steps various Byzantine moneys indicated the relation
of the different strata, and illustrated the gradual advance of
the destruction. The only coin of precious metal, an electron
of the reign of Michael VIII. (Palaeologos) — 1261-1282 a. d.
— was found within the citadel enclosure, north of the temple.

That coins or ornaments of precious metal would be se-
creted by the laborers, notwithstanding the constant super-
vision, was naturally to be assumed in view of the notorious
tendencies of modern Greeks. To obviate so far as possible
the chance of such a loss to the investigations, the intrinsic
value of every piece of gold or silver was offered as a premium
to the finder, in addition to his regular wages. By this arrange-
ment little was to be gained by theft from the trenches.

The reliefs from the temple, the discovery of which, to-
gether with the stylobate, form the most important result of
the year's work, were chiefly found in the lower courses of
the fortification masonry which enclosed the inner citadel.
The blocks of the sculptured epistyle and the metopes re-
maining above ground were, as already mentioned, removed
to Paris in 1838. From the accounts given by Hunt, Rich-
ter, Prokesch von Osten, and Fellows, it appears that these
reliefs, with few exceptions, lay scattered upon the south-
western slope of the Acropolis, where they had evidently
been thrown on the destruction of the late ramparts in which
they previously had been embodied. The descriptions of

INVESTIGATIONS AT ASSOS, 1881. 33

these travellers were fully borne out by the interesting testi-
mony of one of the old men of Behrim, who remembers when
a youth to have seen the sculptured blocks lying upon the
surface, and to have watched the operations of the sailors in
carrying them to the sea-shore.

The search of the French was thorough. Only one frag-
ment was found by the present Expedition upon the declivity ;
namely, the second sphinx from the western front, which lay
face downward at the spot marked E, Plate 2. This block,
though overlooked by Raoul Rochette, may have been seen
by Texier, who in his restoration correctly drew the shaft
upon which the fore-paws of the recumbent animals are sup-
ported, — a feature not evident from any of the reliefs in the
Louvre. As will be seen in the detailed consideration of the
sculptures, this sphinx accurately fits upon its mate now in
Paris, and could not have been purposely left behind.

The two blocks forming the far more beautiful sphinxes of
the eastern front were found in the wall. A A, at the north-
west of the Acropolis. This mass of masonry, from its relation
in plan and bonding to a Turkish semicircular tower abutting
upon it, as well as from the employment of mortar in its more
careful construction, is proved to be of comparatively early
date. The lime which covered and preserved the features of
the archaic heads had become so hard that the stones could
only be loosened from their beds by iron wedges and sledge-
hammers. The broken metope and the small fragment on
which were the front legs of a centaur were found in the
vicinity.

The two important blocks of the bowman and centaurs, the
most valuable discovery of the year, were met with late in the
season, in the foundations of the rampart at the southwestern
angle of the citadel. They were not embedded in mortar, and
were lying near each other at a depth of 1.5 metres below the

3

34 ARCH^OLOGICAL INSTITUTE,

present surface of the earth. The lion and boar relief and
the hind-quarters of a lion were similarly situated near the
western steps of the temple, N,

The complete metope did not form part of the enclos-
ing wall, but lay buried in the accumulated soil at the north-
east, D.

Prokesch von Osten mentions, among the sculptured blocks
remaining upon the surface at the time of his visit in the
year 1826, a metope with a figure of "Amor, seated and hold-
ing a bow," — more probably the archer of Heracles. This
relief was not taken to Paris, but though the most careful
search was made it could not be found.^

The peculiarity of the lateral blocks of the triple epistyle, —
the step upon the back, — made them readily recognizable while
only partly exposed ; and it was with almost feverish anticipa-
tion that stones of this shape were turned over. The pro-
portion of plain blocks was great, both among the lintels, —
of which a considerable length had been taken from the site,
while all the unsculptured inner epistyle remained, — and
among the metopes, of which only those upon the fronts
appear to have been decorated. The delight of discovery was
less frequent than the check of disappointment.

A terra-cotta antefix was found in a crevice at the south-
eastern corner, deeply buried, as it must have been one of
the first parts of the temple to be overthrown, — if, indeed,
the archaic roof-tiling to which it appertained was not re-
placed by a restoration during the flourishing ages of the city.
The lion's head from the corner gutter lay in the deep soil at
the north of the temple.

During a great part of August the work of digging upon
the Acropolis was impeded and made unpleasant by the high

1 This block is particularly described in the Wiener Jahrbuch^ 1832, ii. An*

INVESTIGATIONS AT ASSOS, 1881. 35

north winds, — the Etesians already mentioned. The wind
was so heavy at times as to render it difficult to stand upright
upon the walls of the exposed summit, where any disturb-
ance of the dry soil by pick or shovel raised blinding clouds
of sand and lime-dust into the air. The men made wary
by experience at Pergamon had provided themselves with
spectacles of gauze ; the less fortunate raw hands wept pain-
fully, and aggravated the ill by rubbing their eyes with gritty
sleeves. The distress was sufficient to reduce the number of
laborers ; and it was a relief when, by the end of August, ma-
terial had been obtained for preliminary study, and the work
could be temporarily transferred to the sheltered southern
slopes of the lower town.

Little earth had accumulated upon the commanding terrace
before the stoa. The bowlders and debris washed down from
the heights above had formed a slide across the colonnade,
almost wholly filling up the great basin in front of it, but thus
caught as by a moat, left the open place almost bare. The
general arrangement of the long colonnade remained so plainly
evident that even an unscientific traveller could comprehend
from the ruins the appearance of the original structure.

Although the broad flight of steps which must have served
as an ascent from the theatre, the parapet, and in places parts
of the terrace itself, are missing, yet the general disposition
of the public buildings in the vicinity could be determined
with no great difficulty. Compare Plate 3.^

1 The topographical sketch, Plate 3, exhibits the condition of the stoa, theatre,
and adjacent buildings upon the termination of the preliminary excavations made
during the year. The walls in black are antique, and, with exception of those
of the theatre scene and of the building at T', remain nearly to their original
height. The trial pits and trenches dug by the Expedition are designated by
asterisks.

A A Narrow subterranean passage leading to the stoa plateau.

B Steps in position.

C Fountain niche, before which is the vaulted dstem of accurately jointed
polygonal masonry.

36 ARCHAEOLOGICAL LWSTITUTE.

The first trial pits revealed the arrangement of the en-
trance to the stoa, showing the parapet between the columns
upon the extreme east and the end wall. Trenches were dug
through the mass of earth and stones which filled the hall,
and the lower drums of the inner range of columns were
found to be still in position. Several shafts were sunk to
the bottom of the great reservoir, exposing its accurately
jointed pavement. Its basin was in great part filled with the
blocks of the colonnade and of the buildings which once stood
upon a higher level. Even drums from the summit of the
Acropolis were found here, as has been before mentioned.
Near its western end were the outlet and conduit, which led
the water from it to a lower basin upon the level of the ter-
race Q Q, Plate 3. The stone channel was in admirable pres-
ervation, even the water-box and lead-pipe of a late Byzantine
restoration remaining undisturbed.

In connection with this work a preliminary examination
was made of the rectangular foundations at the west, and of

L> Stone lintels forming the ceiling of the subterranean passage. Broken, bat

in position.
£ E Mcdixval walls, thoroughly excavated. Among these ruins were found

the bronze inscribed tablet and marble inscriptions.
F F Columns in position.
G G Vaults, of Roman or Byzantine period.
//// Modern Turkish enclosing walls.
J J Doorway jambs in position.

K K Subterranean vaulted chamliers l>eneath either end of the theatre auditory.
L L Balustrade of orchestra, and lower seats in position.
M Remains of wall and gateway.

N Ruins of a building restored by Texier as a " N)'mphaeum."
O O Turkish enclosures used as goat-pens.
P P Pavement of the place before the stoa in position.
Q Q Subterranean water conduit, leading to the lower terrace.
R K Greek retaining walls of heavy masonry.
S Greek foundation wall, with water-pipes.
T Foundations of a rectangular building, possibly a temple.
U Mediaeval remains on Greek foundations.
V Ruins of a Byzantine church.

INVESTIGATIONS AT ASSOS, 1881. 37

the considerable remains of a building at the east, of the great
public place. Trenches determined the position of the walls,
and the general plan of these structures; but the limited
excavations could do no more than reconnoitre the field. No
carved fragments or inscriptions were met with upon the
terrace itself ; but on digging at a lower level upon the east,
among the mediaeval walls E E, the debris was found to con-
tain important antique remains. A number of inscriptions
came to light which must originally have stood as upright
slabs on the pedestals of trachyte still remaining upon the
parapet above.

Thus encouraged, we had all the earth lodged in the angle
formed by this lower terrace removed, and *the subterranean
passage leading to the place before the portico freed from
debris. The ceiling beams of this passage had been broken,
and, falling in, had filled the space with their fragments ; the
bearings, however, remained in position upon the lateral walls,
illustrating the peculiar notched system of their jointing.
The steps at the eastern end apparently owe their present
awkward arrangement to a Christian reconstruction.

Close to the lower entrance, at Q Plate 3, there was found to
have been an important fountain, probably supplied from the
great reservoir before the stoa, and standing in immediate
connection with a vaulted cistern. The marble slab and
trough which once filled the niche had disappeared, only small
fragments of the latter being recognizable among the rubbish
in the vicinity ; but the general arrangement of the water-
works could be easily traced.

The cistern, as will be seen, is remarkable for the accuracy
of its polygonal masonry as well as for certain peculiarities of
plan. Some six or seven metres below the surface, the earth
with which it had partly been filled was found to contain some
fragments of inscriptions and various water-jugs of Byzantine

38 ARCHAEOLOGICAL INSTITUTE.

form. Upon the southern side of the passage late walls, E E,
had been built against its enclosure. Among their ruins were
found all the inscriptions published in the present Report, and
a considerable number of fragments which it is hoped to com-
plete by excavations upon a still lower level. Nearly all the
marbles had been shattered by their fall from the parapet
before the portico.

The inscribed bronze tablet appears to have owed its excep-
tional preservation to long service as a fire-back, of which it
bears traces ; the intrinsic value of so large and thick a sheet
of metal would otherwise have led to its destruction.

The badly-built mediaeval walls had been thickly plastered,
and in many places a debased painted decoration was distinct
upon them. The various enclosures were without doubt at
some time occupied as houses and shops, the last inhabitants
of the southern town naturally retreating to the sheltered
slopes near the great public place.

As indicated by Mr. Abbot's account of Assos, written at
the time of the systematic removal of hewn stones from the
site, that work of destruction nowhere produced more lament-
able results than in the theatre. In place of the almost per-
fect monument seen by previous travellers, there now remains
little more than a hollow in the steep hill-side. The upper
seats have been torn away, the lower are covered with rubbish.
The orchestra is filled with earth ; of the stage only the lower
walls exist.

Prospecting trenches uncovered the seats for several tiers
above the balustrade which separated the spectators from the
orchestra. The foundations of the scene were also followed
out. Here the debris had accumulated to a depth of more
than two metres, the space having been used during the Mid-
dle Ages for dwellings, as was evident from the remains of
household fires, the bones and tusks of wild boars, shards of

INVESTIGATIONS AT ASSOS, 1881. 39

barbarous kitchen utensils, etc. Of the pavement of the
orchestra no traces whatever could be found. The only
decorative sculpture met with was a Hermes upon the west-
ern termination of the balustrade.

In connection with these preliminary studies at the stoa
and theatre, some attention was devoted to the great struc-
tural masses in the vicinity. A number of pits were dug
upon the lower terrace in front of the western half of the
stoa plateau. The walls of a Christian church, V, had made
it evident before the beginning of the excavations that the
later Byzantine occupation had greatly altered the level of
this terrace and the plans of the buildings upon it. A lit-
tle digging showed that a thorough removal of the consid-
erable accumulation of earth would be necessary before any
adequate understanding of the complex constructions could
be obtained. At a depth of from two to four metres the
pits revealed antique pavements, water-pipes, foundation
walls, and even the bases of columns in position, — the fur-
ther investigation of which, on account of the extent of the
work, we found ourselves obliged to reserve for another
year.

The case was similar with the interesting remains of a
portal, M, belonging to a building at the extreme east of
the upper plateau. The massive lintel blocks, fallen from
their position, were not to be moved without the help of the
winch, which did not arrive until after the completion of the
work at this point.

The ruins of an enclosure of considerable extent, within
the city walls and at the southwest of the Acropolis, attracted
the attention of several of the earlier travellers, — notably of
Prokesch von Osten, to whom we owe an admirable descrip-
tion of the state of these remains at the time of his visit.
A fragmentary inscription upon the epistyle blocks of a sur-

40 archjEological institute.

rounding colonnade, published by Richter,^ has led to the
supposition that the edifice was dedicated to Augustus, and
even that it was a temple to that deified monarch. Though
its real character is still far from certain, the building will be
referred to as a gymnasium in the present Report, some of
its features indicating this designation.

The outline of a polygonal apse was plainly visible above
the ground, by the side of the footpath which leads from the
village to the sea. ( Compare Plate 4, ) Within this the
accumulated soil proved to be a little more than one metre
in depth, while outside the pavement of the street was not
reached until seven metres below the surface. This made
it clear that the building bordered, toward the south, upon
the parapet of a terrace, and lent weight to the supposition
that the portico observable upon the principal thoroughfare
of the city stood in connection with its inner court, notwith-
standing the great difference in level.

Both apse and portico were freed from earth. Within the
enclosure trial pits determined the position of the gate of the
gymnasium at the northwest, and of one column of the more
important portal at the northeast. A marble stylobate and
the carefully jointed slabs of a broad pavement were found
within the colonnade, which extended at least upon the north-
ern half of the rectangle. One shaft remained in position,
and additiohal epistyle blocks, bearing the carefully cut letters
of the inscription before referred to, were found at no great
depth.

Near the marble steps were various remains belonging to
a monument of small dimensions and lavish Diadochian or-
namentation, — the marble gutters carved with lions' heads,

^ More generally accessible in Boeckh, Corpus Inscriptionum Graearum^
No. 3569. The inscription will be referred to at greater length in the descrip-
tion of the building itself.

INVESTIGATIONS AT ASSOS, 1881. 41

broken cornice blocks and mouldings being so incomplete as
to afford no guide to the original purpose or appearance of
the structure. This state of destruction had been brought
about by the systematic burning of the stone, the blackened
walls of a mediaeval lime-kiln standing directly beside the
stylobate.

The floor of the late building, of which the apse formed the
termination, proved to be a marble mosaic. This was fol-
lowed by a trench to a length of more than thirty metres.
The border of the pattern was nearly intact, but the centre
appeared in great part broken away. Within the limits of
this building, which seems to have been a sacred and possibly
originally a forensic basilica, were found various fragments of
Byzantine decoration sculptured in relief, bearing the cross,
palm branches, etc.

To these four sites within the city walls, to which more
or less attention was paid during the year, — namely, the
Acropolis, stoa, theatre, and gymnasium, — is to be added the
street of tombs outside the fortifications. By similar pre-
liminary excavations the general disposition of the terraces
was here determined, and a number of sarcophagi, exedras,
and vaulted tombs were examined. The stone pavements
were covered with fine earth, washed down from the heights
of the Acropolis, varying in depth from half a metre to three
metres. In many places the slabs were still in position. All
the sarcophagi had been opened and despoiled in former
times. The heavy lids of some had been lifted off, and lay
upon the ground next to the enormous coffers ; others had
been broken into from the side.

An amusing instance of the ignorant rapacity of the riflers
is presented by a small solid sarcophagus, which once served
as the decoration of some mausoleum. The attempts made
to pry off the lid are evident from the rough chiselling of

42 archjEological institute.

the fictitious joints, and the disappointment of the treasure-
seeker is shown by the spiteful battering of the sides, which
were not to be broken into like those of the hollow chests.
Choiseul-GoufEer relates that some years before his visit the
heavy rains, washing away the accumulated earth, had ex-
posed a sarcophagus which had never been plundered. All
the inhabitants of the village assembled, and the coffer was
broken open in the presence of an official ; it was found,
however, to contain no treasure, and the human remains,
with the household utensils buried with them, were flung
away.

Many sarcophagi entirely buried beneath the soil were
found during the excavations of the past year, but none which
had not been opened. Trenches were dug around a number
to expose the carv^ed ornamentation of their sides, and two
exedras were wholly freed from earth. Two vaulted tombs
of interesting construction were also excavated, both having
been stripped of their facades and choked with debris. Two
imperfect inscriptions upon large marble slabs were found
during this work, but it was not possible to decipher them
before the recovery of further fragments.

Toward the close of the season trial pits were dug in the
river-bed to trace the foundations of the ancient Greek bridge ;
but this interesting investigation soon had to be relin-
quished because of the rapid rise of the stream. During
half the year the Touzla is almost stagnant ; but the broad
sandy reach which intervenes between the narrow summer
bed and the high water mark, and upon which the piers in
question stand, is overflowed by the heavy rains of Oc-
tober.

Of the eleven weeks during which excavations were carried
on, six were spent upon the Acropolis in uncovering the tem-
ple plan and in investigating the late fortification walls in

INVESTIGATIONS AT ASSOS, 1881. 43

search of reliefs. The staff of men was comparatively small
during these first weeks, — the news that employment was to be
obtained on the site travelling slowly. The pits and trenches
dug at the stoa, and the clearing of the earth in the subter-
ranean passage and Byzantine rooms beneath, occupied two
weeks. Scarcely six days each could be devoted to the re-
maining sites, — the theatre, gymnasium, and street of tombs.
All the digging carried out in the lower town can count for
little more than a preliminary investigation.

It was perhaps a disadvantage that the work of the year
had so to be planned that its results should present, so far as
possible, an independent study of the city. The undertaking
of a second campaign was by no means certain.

Towards the end of October, some time before the date
fixed upon for the suspension of the work, digging was
brought to a sudden close by official interference. By one
of the laws relating to the antiquities of the Ottoman Empire,
it is required that excavations undertaken at a distance from
towns so great as to render supervision by local authorities
difficult shall be watched over by a governmental commis-
sioner, whose salary is to be paid by the investigators. Not-
withstanding the restrictive clause, this law is enforced in
whatever neighborhood the work is carried on, even in popu-
lous cities like Bergama ^ and Tersoos.^

Upon the granting of the iradeh, the Minister of Public
Instruction, Kiameil Pasha, stated, in reply to a direct ques-
tion, that before the commencement of digging, and during
any considerable suspension of the work, the presence of the
commissioner upon the site would not be required.

A week before the arrival of the first laborers due notifica-
tion was sent to the local authorities of Iradjlk, and a com-
missioner was obtained, to whom was paid the maximum

1 The ancient Pergamon. ' The ancient Tarsos.

44 ARCHAEOLOGICAL INSTITUTE.

salary customarily allowed in similar undertakings (at Troy, Per-
gamon, etc.). The presence of this guard at the scene of the
excavations proved to be a mere formality, and the amicable
relations of the Expedition with the gentleman appointed to
the post, Mehmet Effendi, member of the council of Iradjik,
were perfect.

It became evident that, on account of the advancing season
and gradual desertion of the men, excavations would have to
be suspended for the winter at the end of three months, — on
the 6th of November. On the 1st of October notice to that
effect was submitted to the local authorities, and formally
accepted. It is w^ith pleasure that the obligations of the
Expedition to Shefkfet Bey, Kymacihm of Iradjik, are here
acknowledged; the familiarity of that gentleman with the
French language, and his liberal views, the result of residence
as attach^ of Turkish embassies in various European capitals,
made intercourse with him personally agreeable, and assured
his favorable consideration for our work.

About two weeks before we proposed to close the excava-
tions a Turkish office-seeker, of a type familiar in the ante-
chambers of the Sublime Porte, arrived at Assos, stating that
he had been appointed commissioner to the Americans at
Behrim, by authorities above the Kymacahm in power. He
at once demanded excessive travelling allowances, and main*
tained that his salary, — in amount thrice the generous sum
before paid, — was to be continued throughout the winter,
whether work were carried on or not. The new-comer pre-
sented no credentials whatever, but, on referring the ques-
tion of his official character to Shefk^t Bey, assurance of his
direct dependency upon the Pasha of the Dardanelles was
given.

To accede to such excessive demands was out of the ques-
tion ; to accept the new official would be to give a precedent

INVESTIGATIONS AT ASSOS, 1881. 45

for all manner of extortion in the future. Still it was requisite
to procure in person an endorsement of the Minister of Pub-
lic Instruction upon the iradek, which could be displayed to
the Kymacihm, and, if need be, to the Pasha of the Darda-
nelles. This direct appeal to the eventual arbiter of all ques-
tions relative to the prosecution of excavations in territory
under Turkish rule was wholly successful, after the usual
delays attending the transaction of business at Stamboul.
The would-be commissioner retired from the scene without
even collecting his expenses. He had gained nothing, and
the probability of similar attempts at extortion had been
greatly diminished for the future ; but meanwhile the work
had been stopped, and the enforced close of the excavations
was vexatious.

It was fortunate that the heavy sculptured blocks from the
Acropolis had been brought down to the magazine at the port
early in the season, for at the end of the year laborers enough
did not remain to perform this task expeditiously.

It is well known that all pictorial representations are an
abomination unto the Moslem ; on this account it proved ne-
cessary to remove the reliefs from the reach of the Turks
as speedily as possible. The villagers of Behr^m gradually
became too closely attached to the interests of the Expedition,
by the friendly and unobtrusive bearing of its members, and
by the material profit derived frqm the work, to make any hos-
tile demonstration ; but the wilder peasants and herdsmen
who came to the site from time to time were not always well
disposed. The mosque of Behr^m is the only place of worship
for miles around, and the inhabitants of the neighborhood
frequently assemble, in festive attire and high spirits, to listen
to the droning intonation of the Imam. After the excavations
had been transferred from the Acropolis to the lower town the
visitors always crowded, on Friday afternoons, to the exposed

46 ARCHAEOLOGICAL INSTITUTE.

foundations of the temple, and twice raised the heavy carved
blocks which had been left, face downwards, on beds of fine
earth, setting them up as targets for stones. Although this
stoning was rather the result of wantonness than of malice,
and prompt intervention allowed no time for noticeable dam-
age, the occurrence caused a constant fear that so long as the
sculptures remained upon the ground they might be defaced.
The slightest injury would have been irreparable, and until
the means of transport' were obtained a watch was stationed
to guard the discoveries.

Among the articles soon aft^r brought from Pergamon was
a sledge, which had been built by Dr. Humann for the purpose
of removing heavy stones from the mighty citadel of that
royal town to the roads practicable for wagons. Upon this
the reliefs found by the present Expedition were securely
bound and dragged down the steep slopes of the Assos
Acropolis to the sea, by the whole gang of workmen.

It has been mentioned that the track formed by the Turkish
soldiers in their work of destruction was utilized in the prepa-
ration of the road for the sledge ; yet there still remained,
especially in the upper course, many gullies to be filled up,
and enormous blocks of the thickly strewn ruins to be thrown
aside. The road descended in a tolerably direct course from
the summit of the Acropolis to the port ; but so great was the
exertion required, that the transport of the smallest sculp-
tured blocks could not be effected in less than two hours and
a half Like the laborers represented upon Egyptian and As-
syrian reliefs as moving gigantic statues, the men at Assos
pulled upon either side of two long and heavy ropes, while
the weight was started from behind by levers ; and, as was
customary five thousand years ago, shouting and the clapping
of hands formed an obligatory accompaniment. Facilitated
as it was by the steepness of the track, the noisy exciting

INVESTIGATIONS AT ASSOS, 1881. 47

work afforded an almost childish amusement, and was usually
reserved for the end of the day.

After the close of the excavations Messrs. Bacon, Diller, and
the writer remained upon the site until the 1st of December.
The results of the work were added to the map, the buildings
unearthed were measured, and the preparation of the present
Report and of the geological appendix to it began.

The excavations proposed for the second season have been
carefully considered, and it is with pleasure that the long and
uninterrupted work, to begin in March, 1882, is looked for-
ward to. The delays and difficulties experienced in the past
year, and the requisite preliminary survey, restricted the dig-
ging to one third of the time which it is hoped actively to
employ during the coming campaign. The exertions and ex-
periences of the first season are full of value for the second ;
the broad foundation of the investigations at Assos has
already been laid ; it is comparatively easy to add elaborate
details to the general plan of the city. The expense of outfit
and installation must always be one of the chief items in the
cost of explorations in so distant and inhospitable a land.

It is believed that four weeks* further digging will suffice
thoroughly to complete the studies upon the summit of the
Acropolis ; the amount of time and attention required by the
other sites will become evident as the work advances. Upon
all sides there are important and interesting questions await-
ing solution ; and in the deep slides of earth, such as have
been formed between the stoa and the base of the Acropolis,
and directly above the theatre, remains of antiquity may be
brought to light, of the existence of which there can as yet
be no conception.

Chief among the problems reserved for solution during the
present year, in extent as well as in interest, will be those con-
nected with the fortification walls built at various periods of

48 archjEological institute,

the city's history; and notably the outer enclosure^ whicfi,
though known only from one of Texier*s inadequate plates,
has long been famed as the finest existing monument of
Greek military engineering.

The outlines of the sketch map of iEolic Mysia and Lesbos
here given are derived from the accurate coast surveys of the
English Admiralty. The charts consulted were those of the
Dardanelles, No. 2,429, surveyed by Graves, 1840, Spratt,
1855, and Wharton, 1872; of the entrance to the Darda-
nelles, No. 1,608, surveyed by Spratt, 1840; and of Mytilene
Island, No. 1,665, by Copeland, 1834 The last includes the
northern coast of the Gulf of Adramyttion. The course of
the Satnioeis and the position of the ruins of the Southern
Troad have been determined by an independent compass
triangulation, made by the present Expedition, — in chief part
by its indefatigable geologist. The ancient towns have been
added from the descriptions of their sites given by scientific
travellers of the past century, from the references of ancient
authors, especiall)^ of Strabo and Pliny, and in some few
instances from the authority of the most eminent archaeolo-
gists who have written upon the topography of Asia Minor, —
Forbiger* and Cramer.^

The map is here given only as indicating the general feat-
ures of the land during antiquity. No attempt has been made
to display the important relations of mountain and plain. A
map on a larger scale, embodying all the observations of the
Expedition, and complete, so far as possible, in respect to
modern and mediaeval as well as ancient geography, is re-

1 Handbuch dcr alien Gcop^apkie^ aus den Quellen bearbdtet von Albert Forbi-
gcr. 2 Bande. Leipzig, 1842, 1S44.

2 A Geo^rraphUal and Historical Description of Asia Minor ^ with a map, by
J. A. Cramer. In two volumes. Oxford, 1832.

s

Q

<

CO

o

H

INVESTIGATIONS AT ASS OS, 1881. 49

served for publication with a projected essay upon the topo-
graphy and topographical history of the Southern Troad.

The best existing map of Asia Minor is that of von Moltke,
von Vincke, and Fischer, published in Berlin in 1844, and
accompanied by a memoir relative to its construction.^ The
eminent geographer Dr. Henry Kiepert edited this map from
the surveys of the gentlemen mentioned, who were Prussian
officers temporarily in Turkish service. Its scale is i to
1,000,000, and it includes, besides the whole of Asia Minor,
Armenia, Kurdistan, and Azerbijan.

An outline map, scale i to 3,000,000, is given by Tchi-
hatcheff, as an illustration of his great work upon Asia Minor
before referred to.^ It appears in two forms, as a colored
geological chart, and as an indicator of the routes followed by
the traveller during different years. A small portion of the
northwestern corner of Asia Minor is also included in the
official military charts of the Austrian Geographical Institute,
that numbered P 14 of the Central European Series giving
the greater part of the Troad, scale i to 300,000.^

For those desirous of closely following the geological inves-

1 Title of map : Karte von Kleinasien, enttuorfen und gezeichnet nach den neus'
ten und zitverldssigsUn Quellen ; vortiiglich nach den in den Jahrett 1838-39, von
Baron von Vincke, Fischer, und Baron von Moltke, Majors (j/V/), im k. Preuss.
Generalstabe, und 1841-43, von H. Kiepert, Prof. A. Schonborn, und Prof. K.
Koch, ausgefuhrten Recognoscirungen. In vi. Blattern. Redigirt von Heinrich
Kiepert. Berlin, 1844.

Title of text: Memoir tiller die Cofistmction der Karte von Kleinasien und
tiirkisch Armenien, von v. Vincke, Fischer, v. Moltke und Kiepert ; nebst Mit-
theilungen iiber die physikalisch-geographischen Verhaltnisse der neu erforsch-
ten Landstrichc. Redigirt von Dr. H. Kiepert. Berlin, 1854.

Another map of the land, on a still more generous scale, i to 400,000, intended
to embody the results of all the recent surveys of the interior, is in preparation
by Dr. Kiepert, who proposes also to publish his itineraries in the Troad during
1841 and 1842, on a scale of I to 100,000, which cannot fail to prove a most
important addition to our knowledge of that country.

^ See ante, p. 8, note 3.

* Published in 1878 by R. Lechner. Vienna.

4

50 ARCHAEOLOGICAL INSTITUTE.

ligations made during the past year, this Austrian map will
be found the most serviceable of those hitherto published,
being on the largest scale, and giving with reasonable accu-
racy the position of nearly all the Turkish villages mentioned
by Mr. Diller.

The most recent map of the Troad is that prepared by
Professor Ernst Ziller, of Athens, and Carl Heise, carto-
grapher of the Royal Prussian survey, for Dr. Scbliemann's
Travels in the Troad during 1881.^ It is almost beneath
criticism, — being without scale, or degrees of latitude and
longitude, and so incorrect that, for instance, the outline of
Lesbos is drawn without its two great gulfs !

While the land of Europe is invaded on all sides by water,
the general character of the enormous Asiatic continent is that
of compactness, and its coast-line is comparatively short. Still
the favor of fortune which formed the long peninsulas upon
the northern shores of the Mediterranean, and so signally
advanced and assured the commerce and civilization of Greece
and Italy, was not wholly withdrawn from that part of Asia
Minor bordering on the -^gean. It has been remarked that
the waves of that sea seem to have a peculiar power of pene-
trating and dissolving parts of the land upon which they beat,
forming islands, peninsulas, and capes by this dissolution, and
creating a disproportionately long coast-line, with many gulfs
and nooks favorable to primitive marine intercourse.

All Asia Minor turns its back upon the steppes and deserts
of the interior continent, no considerable river running to the
east, and the Troad is separated from inner Mysia by rugged
and uninhabitable highlands. If Asia Minor appears reluc-
tant to belong to the great continent, the Troad unequivocally
opens its arms to Greece. The iEgean, from the earliest

^ See anttf p. 14, noU 2.

INVESTIGATIONS AT ASSQS, 1881. 51

ages of marine intercourse, while seeming to divide has really
united the opposite shores, and the water-way to Europe has
been more practicable than the overland journey to the inner
countries.

The Troad is the portion of Asia most nearly allied to
Europe. Its eventful history tells of successive coloniza-
tion by Phoenicians, Carians, Leleges, and finally by iEolic
Greeks. It was conquered successively by Crcesus and Cyrus;
it was among the earliest of Roman possessions in Asia ; it
often changed hands in the struggles between the Byzantine
Greeks and Latins, and at length it submitted to and sank
under the blows of Seljukian and Ottoman invaders.

Leavmg out of account the unsubstantial realm of ancient
Ilion, Assos appears to have been in ancient times the most
populous and flourishing city of the Troad. It was, moreover,
the chief citadel of the land.

Towards the close of the tertiary period an extended vol-
canic upheaval revolutionized the northern coasts of the Gulf
of Adramyttion. Two flows of trachyte — forming craters,
dykes, and plateaus — covered the original limestone so
completely that it is only in small and isolated patches
that stratified deposits remain upon the surface to display the
former geological condition of the land. A crest, rising to a
height of five hundred metres, was thrown up along the coast
from Antandros to the promontory of Lecton. The Satnioeis,
second only to the Scamander among the rivers of the Troad,
rises only six or eight kilometres from the Gulf, but, hemmed
in by this continuous range, does not reach the iEgean until
after a course estimated at not less than seventy kilometres.

At the point where the Satnioeis most nearly approaches
the coast of the gulf, the intervening strip of land is but one
and one-half kilometres broad. It was here that the crater ol
a volcano formed the Acropolis of Assos. Situated between

52 ARCHAEOLOGICAL INSTITUTE.

Stream and sea, rising steeply to a height of more than two
hundred and thirty metres, and wholly isolated from other
peaks, the cone is one of the most prominent features of the
country.

The inclination of the land between the port and the sum-
mit is represented by Plate 5.^ The average height of the
surrounding plateau is about that of the terrace occupied by
the theatre ; all above this may be considered as the elevation
of the Acropolis.

The crater was choked by the second and final flow of
trachyte, — the stone which has had signal influence upon the
topography and architecture of Assos. This material cleaves
naturally to vertical and horizontal joint planes, and it is often
difficult to distinguish the surfaces thus formed from those
hewn by the hand of man during the systematic quarrying
from the cliff.^ The sides of the Acropolis assumed the char-
acter of a vertical rampart, which reaches the greatest height
in a double tier on the south and west. The view of the
Acropolis from the northwest, Plate 6,^ shows its cliffs, which

^ It is to be remarked that those not accustomed to judge the proportions of
topographical sections will be naturally inclined to undervalue the steepness
indicated in Plate 5. The elevation is not exaggerated, contrary to the usual cus-
tom of introducing two scales, and making that of the vertical dimensions from
twice to ten times as great as that of the plan.

■•* Compare the remarks on the second trachjle of Assos in the geologrical
appendix.

^ This view (Plate 6) is taken from a spot near the road which leads to the
northwest from the point shown on the edge of the topographical plan of the city,
Plate I, as the site of " ruins." The grain-fields of the foreground have in great
part been reclaimed by the villagers since the writer's first visit to the site. Be-
yond them are the overthrown sarcophagi of the street of tombs, before the
l)rincipal gate of the fortification walls. The ramparts can be traced from the
re-entering angle to the declivity on the southwest of the Acropolis, and their
outline is evident as far as the low towers which mark their extent upon the
north. The transverse divisicm wall is seen greatly foreshortened. At the left
of the summit are the semicircular Turkish bastion, a mediseval tower on Hel-
lenic foundations, and the early Christian church now serving as a mosque. Be-
neath these follow the houses of Behr^un.

INVESTIGATIONS AT AS SOS, 1881. 53

are also indicated in their full height, upon the south, by the
section Plate 5.

Naturally steep upon all sides, and rendered still more
secure by a judicious scarping of the rock, the summit be-
came wholly impregnable by the construction of enclosing
walls. The limited circuit was easily to be defended, while
the enclosed area was still of sufficient extent to accom-
modate an adequate garrison. A fissure in the rock of
the lower step forms a natural well, and the supply of water
was still further assured by the excavation of deep cis-
terns at this point. It was with truth that Strabo^ re-
marked that Nature and Art had united to make Assos
a stronghold.

The view from the Acropolis is magnificent. At the north,
beyond the Turkish village, the land descends rapidly to the
alluvial plain formed by the Satnioeis. The river emerges
from a rugged and confined gorge, and, winding through the
green fields, is lost to sight in the dense oak forests of its
lower course. The great volcanic plateau, which separates
the stream from the sea, extends to the west, rising above
Lecton to a height even* greater than that of the isolated cra-
ter of. Assos. At the south, occupying nearly half of the
horizon, lies the Gulf of Adramyttion, stretching from the
little port, in the extreme inner nook, which bears its name,
to the open ^Egean, north of Cape Sigrion. Beyond this

t

narrow channel is "the noble and pleasant island" of Lesbos,
the pearl of ^Eolic lands. At the foot of Lepethymnus the
promontory and citadel of Methymna is relieved against
the majestic mountain which glows with constantly changing
light and color, as the seasons of the year and the hours of
the day advance. In the far distance, directly south of Assos,
rises the peak of the Mytilenian Olympus. At the east tower

^ Strabo, xiii. 610.

54 ARCHAEOLOGICAL INSTITUTE.

the heights of Ida, the domes of Gargarus and Cotylos, and
on a lower level Mt. Alexandria, famed for the judgment of
Paris. Upon every side scenes of Greek legend and history
are presented to that powerful second-sight of the lover of
antiquity which sees the busy life of former ages where now
remain but trackless plains and desolate ruins. In all Greek
lands, from Sicily to Cilicia, no Acropolis is more favored than
that of Assos, few more beautiful.

The primitive races of the Mediterranean coasts everywhere
built their towns upon such eminences or at the foot of them ;
and this citadel thus directly upon the sea, and yet secure from
piratical attacks, must have been occupied by the first inhab-
itants of the Troad. Thucydides,^ indeed, remarks that in the
most ancient times cities were founded at a considerable dis-
tance from the sea, in ordef that they might not be surprised
by the sudden descents of pirates ; but that after the advancing
civilization had brought immunity in this respect, a situation
directly upon the shore was preferred. The inland positions
of Troy, Athens, and many other cities near the iEgean must
have been determined by such considerations of safety. At
Assos, however, the high plateau and inaccessible Acropolis,
though close upon the shore, were easily defensible, so that
from the first its inhabitants were secure while they enjoyed
the benefits of proximity to the sea, as well as the advantages
afforded by the neighboring river and the fertile alluvial plains
formed by its waters.

The volcanic range, before mentioned, descends steeply
upon the entire northern coast of the Adramyttion Gulf,
nowhere affording a natural shelter either of roaidstead or
of port. The building of a mole, midway between the inner
end of the Gulf and the promontory of Lecton, provided a
refuge most welcome to the voyagers on the way from the

1 Thucydidcs, i. 7.

INVESTIGATIONS AT ASSOS, 1881. 55

city of Adramyttion, gr from the natural harbor of Heracleia,^
to the great marine highway of the Hellespont, while it secured
to Assos the monopoly of the commerce arising from the ex-
port of the produce of the Southern Troad and the import of
foreign merchandise required by that land.

The history of the mole would be the history of the mate-
rial prosperity of the city. When a storm washed away the
upper part of the breakwater two years ago, it was the first
care of the native merchants to patch it up with heaps of
small stones, — temporizing with the fate which threatens the
entire destruction of the port by silting up the shallow basin.
Thus while the existence of Assos was primarily determined by
the strategic advantages of its citadel, the further growth of the
city was due to the commerce attracted by it as the only con-
tinental port upon the Gulf of Adramyttion. It was its mole
that made Assos the chief mart of the Troad, notwithstand-
ing the fact that the area of the arable land of the Satnioeis
valley is much less than that of the Scamander, with its
broad-stretching plains. Assos was the sole emporium of the
southern country, with the exception perhaps of a limited dis-
trict in the immediate vicinity of Lecton. The later artificial
port of Adramyttion at the end of the gulf was separated
from the valley of the Satnioeis by the heights of Ida, and,
deriving its exports mainly from the fertile Theban plain,
can never have materially interfered with the commerce of
Assos.

Miserable as is the present village of Behr^m, it still in
great measure maintains the commercial relation to the inte-
rior that during antiquity rendered Assos the chief mart of
the land south of the Scamander. The port is always crowded

1 The magnificent harbor here formed by the group of islands known to the
ancients as Ilecatonnesi has in recent years secured the growth of the flourishing
town of Ivalee, referred to on p. j.

56 ARCHAEOLOGICAL INSTITUTE.

with coasting vessels, — seldom less than eight, often more
than twenty, lying within the mole. Communication by their
means is regular between Behr^m and Smyrna, Mytilene,
Ivalee, and Molivo.^ In fair weather live-stock is daily carried
across to the opposite island. The merchandise most exten-
sively exported is valonia, the district in which Behrim is
situated producing greater quantities of this valuable tanning
material than any other province of the Ottoman empire.'
Long trains of camels bring the valonia from all parts of the
interior to the port, where it is stored in the magazines and
slowly loaded upon the boats. In the busy season seventy or
more camels may sometimes be counted on the narrow strip of
land between the cliff and the sea.^

The port at Baba might seem a dangerous rival to Behrim.
being fairly protected by the gigantic blocks of the mole men-
tioned by Strabo, lying nearer to Europe, and not situated
under the lee of far-stretching cliffs ; but it has only a trifling
commerce. Babd-calcssi, though strongly fortified, and a con-
siderable centre for certain Turkish manufactures,* is too dis-

1 The ancient Mcthj-mna.

- According to statistics given to the writer by I^vantine merchants, the an-
nual i)roducti()n of the district of Mytilenc, Iradjik, Eancdch, amounts to 140.000
cantars, — the cantar being theoretically equal to 56.1 kilograms (123.7 pounds
avoirdupois). The most extensive forests of the valonia-oak in the Troad are
in the Toii/.la Valley, and dependent upon the i)()rt of Behram.

3 Ikhrhm has lost much of its strategic significance by the extermination of
the pirates, who so lately troubled the shores of the Adramyition Gulf, and by the
present security of the land from marauders, — both resulting from the general
advance of civilization in the Levant. The village itself is, probably, not much
larger than it was during the last century ; but the port has become of much
greater imp<.rtance. Three of the four magazines at the water's edge were built
during the last twenty years; the largest of them, which served the Kxpedition
as a dwelling, being not yet two years old. In kSi6, at the time of V(jn Kichter's
visit, three vessels lav within the mole ; to-dav the number would average sixteen
or eighteen. This increase of commercial activity indicates a gradual ameliora-
tion of the interior country, evident from other considerations.

* The cutlery of Uaba-calessi has a far-spread repuUtion, especially its silver-
handled knives of peculiar fashion. ,

INVESTIGATIONS AT ASSOS, 1881. 57

tant from the three fertile plains of the Touzla to possess
much export trade.

The commerce of Behrim, which with this exception re-
mains the only sheltered port on the coast of the Troad south
of the Hellespont, is relatively petty enough. Pasturers of
herds were never willing servants of Demeter ; and now that
the Turks — a people by nature nomadic, and possessed with
a supreme contempt for agriculture — have dwelt in the land
for over four centuries, the fields bring forth but a small frac-
tion of what they might be made to produce by thorough
cultivation. The invincible repugnance of the Turks to till-
ing the soil is a cl^racteristic of the greatest political and
economical importance, perhaps even the point of greatest
moment in their inevitable national decline.

The gradual destruction of the forests of the Troad has
been followed by parched summers and stormy winters. The
streams disappear in the dry season, to flood and devastate
their banks during the rainy months. The accumulated soil
has washed away from the volcanic highlands, exposing barren
crests of rocks, and covering the humus not within the reach
of freshets with beds of sand and gravel. Only a small frac-
tion of the once arable land is tilled at all, and the country
which formerly exported grain is now barely able to supply its
own demands, — though supporting perhaps the fourth, per-
haps but the tenth part of its ancient population. A horn-
of-plenty upon the coins of Assos once indicated the fertility
of its territory ; * the symbol would most certajnly now be
inappropriate. The area occupied by the city proper, within
the line of fortifications, appears never to have exceeded
one-half a square kilometre, fifty hectares,^ — a small surface,

1 Several examples of the cornucopia upon coins of Assos are given by
M'lonnct, Descn'pti^m de MidatlUs antiques ^grecques et romaines. Vol. ii. Paris,
1807.

- About one hundred and twenty-four English acres.

58 ARCHAEOLOGICAL INSTITUTE,

indecil, compared with the extent of modern towns. Still the
number of inhabitants assumed for Athens, Ephesus, or Syra-
cuse, at the time of their greati^st power, stands in small rela-
tion to the crowded population of existing capitals.

rho limits to the growth of Assos, fixed by the natural forma-
tion of the hmd, were not less marked than the advantages of
its silo. The position of the city upon a promontory divided
from Inner Mysia deprived it of any extensive political influ-
cuoo» like that long enjoyed by Pergamon. The port, upon a
j;ulf which retreats from the regular marine highways of the
V^riont, never couKl assume the character of a commercial cen-
in" for the gvHxIs of other countries. Its position was not
such as creatovl a ix^pulous city upon the barren Tenedos at
iho nunuh of the Hollesjx^nt. or concentrated the nautical
av'livity of the ARhijK'lago at Delos in antiquity, — at Syra in
ihv^ piosout century.

S^vuix^ within the unrix'alled ramparts provided by Nature
an*J ArJ : nos:U\l around the archaic Doric temple of the
Ax w^jv^lis, so hij;h alx^ve the sea as to lose the noisy cries
ot the buN\ Hi:U^ jv^r;, — ancient Assos may be imagined as a
staul an^l ouJoilv vvmmorci.il :ow:\ Tenacious of ]ong-estab-
l\>hvNl UNa>;os, and consorva:i\x^ in its interior and exterior
j\s;tu N. 1: is t>^ such a \x cl) vrv'.orovi ov.stence :ha: all the indi-
canons atK^tNUNl h\ inNCup;;o-s an^: ruKic iric^r-uments point.

Vho )M^J^^n o! As^iA^ has ^vc:". v^nei ar.i evenifal, but

ti\^^^^ ?ho u.\tin,v) ^^^n^;;:^^:^s .^! :^v^ '..^vr.. ilrcjfeiy referred to,
\,r,Vo\ \v\x>j\s^ ;)^An a. ;;\\\ a^^: >o*nv ro: r^^c^.v.intc'i in detail
V\ .;;\x u !^j \x\;^)x \\ IN *,M\^\;>'c t'M: :>e r^.rriciins. the
tv.N; k;\.N\xn >x ^1 u\'^N »;> ;!>o \x,;u^:n ,^t t"^* V^ri*. v'"cC'M:ii«d a
*,;n^; xNi \,;, ^ »r,^isM ^,u^x x' a^ *%>.n' I'.Wx/. :," :".,: ex: ir-scve trade

t;a/.;;>s J^^^t^ tv.xxv tsxi\ haiXxlAi o^^x^ ;* * K.t :>>i^ prccaiDcnt

INVESTIGATIONS AT ASSOS, 1881, 59

Citadel of Assos, rising directly above the sea, must have
been among the first sites to attract the colonization of these
marine adventurers. The remains of a fortification enclosure
of most primitive polygonal masonry exist upon a height a
few hundred metres to the west of the port, termed by the Ex-
pedition the " Seaward Acropolis," and have not been disturbed
by an occupation of that site during the historical ages, and
evidently antedate the Greek colonization of the land.

From the Hecatonnesi to the Hellespont no shelter what-
ever is provided by natural indentations. Of the three moles,
which have been built on the southern and eastern coasts of
the Troad to supply this pressing need, — namely, those of
Assos, Lecton, and Alexandria Troas, — the truly gigantic
blocks at Lecton may possibly be the most ancient, as that
cape is an important turning-point of the winds, and often a
port of unwelcome detention ; but the first building of a break-
water at Assos cannot be referred to a much later date.*

The rough and piratical Carians in great measure kept step
with the Phoenicians in the pursuit of the profitable commerce
of the Euxine, and they too colonized the Troad ; in all proba-
bility occupying the same stations, as they are known to have
done on the shores of the inland sea.

The prehistoric population of the Troad seems to have been
driven from the land in the earliest historical ages by that
branch of the Thracians known to Strabo, and to all later
antiquity, as Mysians. The people to whom this geographi-
cal denomination was applied were of the same stock as the
Leleges, who at the period described by the Homeric poems

1 The location of an " ancient mole,** at Point Sivrijec (a slight projection of
the land near the site of Polymedion), is one of the extremely rare mistakes of
the chart of the British Admiralty, No. 1,665, referred to above, p. 9. The
peculiar formation of a natural reef at this point gave rise to the error. An
extensive consideration of the ruins of Polymedion, discovered by the present
Expedition, will form an interesting chapter in a future Report.

()0 ARaiA'lOLOGICAL INSTITUTE.

oriMipit*(l the northern coast of the Adramyttion Gulf. The
hloulilu'ation of Lclcj^os and Carians, referred to by Strabo,'
ap|u\ns inaihnissihlo ; l)ut traces of a preceding Carian occu-
pation of the Troad. such as the names of towns, may nat-
urally have Ihvu retained by the former people.

It is an opinion not hitherto advanced, which seems to the
wnloi v'ajMblo \>{ supjH^rt, that Pedasos, the capital city of
thv* I \ lo:;os» tiu* town sackoti bv Achilles,^ is identical with the
laioi Anmvv The 1 .olo^os» fauKxl as nangators and pirates,
ndubitwl the Southern ri\ud at the time of the Trojan war,
tvu\j; N|v*lvvM\ ol bv Uomor as Uving uix>n the coast.^ This
x;a;on\otu is v\M\tumv\l by Stralnv who describes the province
x^t ;^o I oU\;;\*s as oxtonvlin^; fivm Lecton to Ida,* and again
x^xjsv:ar.\ N:a:os that thoy jvssessovl the country around
\vxx\N.* l:» ;>,o n:s: iviss.i;;x'' v^l the Iliad bearing: uDon the
v::\ V.I >',:o>;u^*\ l^IatxNs is s;v^kcn v^t as livir.iT -bv the banks
^^i .>x* x^; ' xs^ s. i^** suv*.^ Tvv.as.vs '* I:: rhe sOvX^ni :he king

^ \ ■ »V » » ~ . • . •

\t,x\ -tx V \^ .;,v.v ;>x* Ni:.' ,v.>. *: :> rvisor^rle t? look 2t

V ". • •

^i ■« ^ • iX , ». . ■ V . •. 1 X . N> J; C\ • ^ .> . », ■ >. ^ctiift -»

.V. " . V ■ ;x' s:- i.',: :.r*^ -.--ir. Aa.

ft »

>. V *»

' " X »

V .. ft • . ■ ^ • ,.-..▼-,' /h^:.\

k »

^^ ■ ■ •".■*■

INVESTIGATIONS AT ASS OS, 1881, 6 1

almost direct proof that the citadel at this point, which by
nature commands the Southern Troad, served as the Lelegian
as well as the Greek capital is further offered by the fact, that,
in following the Satnioeis from the Halesian Plain of its delta
to the headwaters of the rugged interior, no other site oc-
curs to which the epithets alircLvof; and alirrjeL^i could be
applied. The Acropolis of Assos is thereby described with
that truth to nature characteristic of the poet, whose thor-
ough acquaintance with the Troad is evident in all his local
descriptions.

The relation of the names Pedasos and Assos seems con-
firmatory of this conjecture ; and the often remarked lack
of all direct mention of Assos in the Homeric poems is
explained by it, — an omission the more surprising as the
citadel is so conspicuous a feature of the land. In reading
the Iliad in the Troad, one is readily inclined to believe the
scholiast's tale that the poet resided at the Trojan Kenchreae
while composing his work, and to doubt his blindness at the
time.

Strabo^ mentions a town in the inner country of Halicar-
nassus named Pedasa, surrounded by a tract known even
in his day as Pedasis ; and it appears not impossible that
the occurrence of the name in the native land of the Cari-
ans may point to the designation of our city as a relic of
early Carian occupation of the Troad. The termination
a<7<709, aaaay or i<r<ro^, taaa^ signifying town,^ retained in the
names of several cities of Mysia and Lesbos (besides Pe-
dasos or Assos, Lyrnessos, Caressos, Prepenissos, Corybissa,
Thebassa, Eressos, Antissa, Larissa, etc.), is extremely com-

1 Strabo, xiii. 6ii.

' Dr. Fligier, Beitrdge tur Ethnographie Klein'Asiens und der Balkanhalh-
insei, Breslau, 1875, derives this termination from the Sanscrit, and points to
its occurrence in almost all the lands famed in ancient history, from Dacia to
India.

62 ARCHAlOLOGICAL institc/te.

mon in Caria and the neighboring tracts (besides Pedasos,
lassos, Halicamassos, Mylassa, Halmylessos, Milessos, Ades-
SOS, and Tymnissos, that is, the city of Tymnos, a Carian hero,
in Caria ; Pelmessos, Sagalessos, Carmylessos, Acalissos, and
Habessos, a name of Antiphellos, in Lycia ; Colobrassos, Saga-
lassus, Tarbassos, Aarassos, Termessos, Pednelissos, and Sel-
gessos, the ancient name of Apamea, in Pisidia ; Ariassos and
Termessos in Cabalia ; Coropassos, Adopissos, and Pirnissos
in Lycaonia, and many others). In many of these cases the
independent significance of the prefix is recognizable, so that
it is conceivable that it might be dropped off as in the case of
Assos.

In the passage last referred to, Strabo speaks of Pedasos as
not in existence in his time ; but his failure to identify it with
Assos may be compared to his fallacious argument concerning
the site of ancient Troy, and his refusal to admit the identity
of the primitive Chrysa with the town bearing that name at a
later day.^

Strabo^ quotes the passage from the Iliad in which Pedasos

1 Dr. Schliemann, in his recently published book of Travels already referred to,
p. 14, note 2, as well as in a paper previously read before the Anthropologica]
Society of Berlin, which appeared in the Augsburger Allgemeine Zeitung^ iden-
tifies Assos with the Homeric Chrysa ; remarking : " ich glaube dies um so
mehr, als, nach der Ilias (i. 431), das alte Chrysa cinen Hafen hatte, der ihm
auch von Strabo (xiii. 612), zugeschrieben wird, wahrend an der ganzen nord-
lichen Kiiste .des Golfs von Adramytteion Assos der einzigc Ort ist, der einen
solchen hat " (p. 23). That Chrysa was situated upon the Gulf of Adramyttion
seems an assumption at variance with the shortness of the voyage of Odysseus,
which appears to have been made, from Troy to Chrysa and back, in one of the
poet's days. In this view the account would well agree with the identification
of ancient and modern Chr}'sa, assumed on the accompanying sketch map. At a
point of the coast near that site (the modern village of Kinlacle^) a small cove,
constantly sought by fishing boats, provides good anchorage for vessels of no
great draught, and, in most winds, fair shelter. Homer's description of the
landing-place and the anchoring is better applicable to this spot than to one pro-
vided with a breakwater. Strabo, in the passage referred to, in regard to the
harbor merely repeats the words of Homer.

« Strabo, xiii. 584.

INVESTIGA TIONS A T ASSOS, 1881. 63

is said to have been sacked by Achilles,^ in connection with
the piratical expedition of the hero to Lesbos, during which
Thebe and Lymessos, also upon the Gulf of Adramyttion, and
Chrysa, near Lecton, were ravaged. He speaks of Pedasos as
in the country " opposite to Lesbos," and, if weight be at-
tached to this testimony, the city can hardly be elsewhere
placed than at Assos.

The importance of the Southern Troad in the progress of
the arts during pre-historical ages is indicated by the Greek
legend of the Dactyls upon the heights of Ida, rich in the
metals employed by those primitive artisans, whose names —
Kelmis, Damnameneus, and Acmon; that is, hammer, tongs,
and anvil — designate cunning workers in iron and bronze,
This personification points to the empaistic art of the Phoeni-
cians, — an art which appears to have been practised in several
mining lands exposed to the influence of that people, as Crete
and Rhodes (Telchinae). The significance of the convention-
alized relief-sculpture upon the archaic temple of Assos, as
affected in their style by the Asiatic overlaying of wood-
carvings with sheets of beaten metal, will be referred to else-
where.

One of the most important and interesting chapters of the
early history of the Troad and of Assos to be filled out by
future researches is that relating to the influence of the great
Mesopotamian civilization upon the coast lands of the iEgean.
— an influence of subtile and far-reaching character, affecting
alike the politics and the art of the early Asiatic Greeks.

The recorded history of the Assyrians in the Troad consists
of a few scattered passages in Greek writers, — the cuneiform
inscriptions, hitherto deciphered and published, affording no
direct information concerning a land which appears to have
been beyond the borders of the Mesopotamian Empire even at

^ Iliad, XX. 90-92.

64 ARCHAEOLOGICAL INSTITUTE.

the time of its greatest extent. It is not probable that West-
ern Mysia was ever subjugated by the Assyrian monarchs, not-
withstanding the assurance of Diodorus ^ that the Troad and
the shores of the Hellespont were conquered by Ninus. Stra-
bo^ mentions walls in Tyana^ and in Zela,* said to have been
built by Semiramis, which make it evident that the conception
of an Assyrian occupation of Asia Minor was entertained in
the later ages of Greek antiquity.

While, however, we may doubt the fact of the actual incor-
poration of the Troad in the Mesopotamian Empire, it yet
appears undeniable that that powerful state exerted a consid-
erable political influence upon all the countries of Western
Asia, possibly even demanding a regular tribute from those
upon the northern coasts of the ^Egean. This view is borne
out by a passage in Plato's Laws,^ where the Trojans are
spoken of as counting upon the support of the Assyrian Em-
pire, ** of which Troy was a portion." And Diodorus gives a
tradition that the Assyrians, who at the time of the appear-
ance of the Greeks under Agamemnon before Troy are said to
have maintained their supremacy throughout Asia for a thou-
sand years, sent a considerable contingent to the assistance
of King Priam.'^ These passages, if taken literally, are indeed
of little historical value ; but, like most such legends, they
have a basis of truth.

From the cuneiform inscriptions we learn that the realm of
Tiglath-Pileser I. extended, before the end of the twelfth
century b. c, to the shores of the Mediterranean ; that the

^ Diodorus, ii. 2.
2 Strabo, xii. 537. 559.
' The present Kiz, or Killis Ilissar.

^ The present Zilleh has retained the ancient name of its site almost un-
altered.

* Plato. Laws, iii. 22.

^ Diodorus, ii. 22.

INVESTIGATIONS AT ASSOS, 1881. 65

great commercial cities of the Phoenicians, those early colo-
nists of the Troad whose influence was so constant and
extended, paid tribute to Assur-nazir-pal as a conqueror in
870 ; that Shalmaneser II. visited the shores of the sea in-
cluded in his realm in 859 ; and that his successor, Vul-
nirari III., visited these provinces in 803 b. c. The cele-
brated stele sent by King Sargon to Cyprus in 709, now
in the British Museum, attests the subjection of that power-
ful island, which was in so many respects the cradle of
Hellenic culture. The Assyrian account of the expedition
of Sennacherib to the Persian Gulf in 697 is particularly in-
teresting, when the vessels built by Syrian and Phoenician
workmen were manned by sailors chosen from the seafaring
nations inhabiting the coasts of the iEgean, and notably by
lonians. The Assyrian king could even contest the maritime
supremacy of the Mediterranean with the fleet of the Greeks,
winning a decisive victory on the coast of Cilicia, at a date not
far from 690 b. c. The naval conquests of Tyre, at that time
the greatest mercantile city of the world, and the conquest of
northern Egypt, made by Assur-bani-pal,^ must have spread
the fame and influence of the Assyrians to the most remote
lands of the sea. So extended was this pre-eminence by the
middle of the seventh century that even the Lydians sent
tokens of submission to the Mesopotamian despotism. Sardes,
the Lydian capital, was less than two hundred kilometres
distant from Assos.

The peculiar importance and interest of the Assyrian influ-
ence consists in its bearing upon the advancing civilization
and art of the Asiatic, and through them of the European,
Greeks, rather than in any direct political ascendency. It is
hoped that the recovery of the archaic temple, and more espe-

^ Assur-bani-pal, 668 to 626, known to the Greeks as Sardanapalos.

5

66 ARCHAEOLOGICAL INSTITUTE.

cially of portions of its sculptured decoration, by the present
excavations, may add somewhat to our knowledge of the
development of the Doric style and of the early Greek stone-
carving, which stood in undeniable relationship to the artistic
spirit and methods of Mesopotamia.

The Southern Troad, once occupied by Leieges and Thra-
cian Mysians, may be considered as sharing in some degree
the aspirations and advance of the ethnographically allied
Hellenic race. It was wholly and forever united to those
interests by the ^Eolic colonization of Assos. In the latter
half of the eleventh century the ^Eolian Greeks possessed the
neighboring islands of Lesbos, Tenedos, and the Hecaton-
nesi. The commanding site of Assos, famed for its strategic
and commercial advantages, appears to have been occupied
by them about the same time.

It is not strange that the Gregk settlers of Assos should
have been reputed a colony of Methymna,^ close as is the
intercourse which the city is destined by nature to maintain
with that opposite port. Methymna, the home of Arion, and
at one period the chief city of Lesbos, retained in its name a
reminiscence of the Ionian colonization of the island, which
had preceded that of the iEolians. It is the site upon the
northern coast of Lesbos, naturally corresponding to the
Acropolis of Assos in the Troad ; and, as offering similar
advantages, must have been occupied from the earliest ages.
The strait which separates the island from the continent is
only ten kilometres broad, the distance between Methymna
and Assos less than twenty. On calm days the passage is
often made by row-boat ; the winds prevalent during the
greater part of the year, though heavy, are regular, and sel-
dom raise a dangerous sea in so confined a channel.

^ Myrsilos, quoted by Strabo, xiii. 6x0.

INVESTIGATIONS AT ASSOS, 1881. 67

This easy communication by water tended to connect Assos
more intimately with iEolic Lesbos than with the neighbor-
ing lands of the Scamander, to which the roads are rugged
and difficult. In primitive and lawless ages the sea is always
safer than the land ; no ambush or unforeseen difficulty
need be feared upon the narrow strait, which was overlooked
from the citadel of either town. The low houses at the south
of the castle of Molivo are visible from the port of Behrim
and from the Acropolis; and on clear days it was possible to
note the departure from the island of the little boat which
weekly brought across the eagerly awaited mail of the Expe-
dition.

The iEolians gradually Hellenized the tracts of the conti-
nent chosen for their settlements, apparently without any long
warfare with the previous inhabitants, to whom they were in
some degree ethnographically related. Some force was doubt-
less at first required, but the final results must have been
mainly due to the superior activity and intelligence of the
Greeks, who stood in much the same position to the Mysians
of the tenth and ninth century b. c, as do their descendants to
the Ottomans of the present day.

The iEolians appear to have acquired by degrees many
traits of the original inhabitants of the continent, — even as
the modern Greeks are in many ways affected by certain
Turkish peculiarities of manner and speech.

Having become wholly Greek, Assos advanced in power and
prosperity until it possessed an extended tract of the surround-
ing country, and was itself able to found the colony of Gargara
upon a spur of the Ida range, twenty kilometres at the west.
Though Assos may never have rivalled the greatness of the
cities of the mother island, it was intimately connected with
Methymna and Mytilene, at a time when they represented the
highest contemporary advance of Hellenic civilization. When,

•68 ARCHAEOLOGICAL INSTITUTE.

after an existence of nearly five centuries, Assos» in 560 b. c,
fell into the hands of the Lydians, it is spoken of as the
strongest and most important city of the TroacL

The Lydians took up the thread of Oriental domination
where it had been dropped by the Assyrians. Their influence
is of particular interest in the history of the civilization and
art of Assos.

In the last half of the seventh century b. c, with the ascen-
sion of the dynasty of the Mermnadae, the Lydians revolted
from the yoke of Mesopotamia. The politic Gyges allied himself
with Psammitichus in overthrowing the rule of Assur-bani-pal
in Egypt ; and though Ardys, son of Gyges, after the invasion
of Lydia by the nomadic Cimmerians, tendered submission to
the Assyrian monarch, the land did not again fall under the
declining power of the Mesopotamian monarchy.

Concerning the independent development of the Lydian mon-
archy we have only the authority of Greek writers, who offer
a history rather copious than consistent. Gyges seems to
have dreaded the advancing civilization and political power of
the Greek settlements of the coast, and is said to have con-
quered a great part of Mysia, including the shores of the
ndlespont ; so that the Milesians, the most influential Greeks
of Asia, were obliged to request the permission of the Lydians
to found Abydos, in the Troad.^ One of the chief sources of
the wealth of Gyges, Alyattcs. and Croesus was reported ^ to
be a mine situated between Pcrgamon and Atarneus,^ almost
within sight of Assos. The expansion of their power upon all
the coasts of the iCgean is evident from many such accounts.

It is even possible that Assos had been subjected to the
direct rule of the Lydians at an earlier date than that assumed.

^ Strabo, xiii. 590.
2 Strabo. xiv. 6S0.

' Atarncus is identified with the present landing-place Deekelte, from whence
the tools brought from Bcrgama (Pergamon) to Bchr^m were shipped.

INVESTIGATIONS AT ASSOS, 1881. 69

Croesus was appointed satrap of Adramyttion and the Theban
Plain during the lifetime of his father, and his jurisdiction
may well have included the neighboring cities upon the Gulf.
Modern authorities believe this event to have taken place
twelve years before Croesus became king.^ Adramyttion
itself, named after Adramytus, another son of Alyattes,^ was
known to later ages as a settlement of the Lydians of this
period.*

The Lydians, at least in the early ages of their history, were
without an independent literature and art.* Their conquest
destroyed the political independence of the land, but does not
seem to have interfered with the intellectual development of
the Asiatic Greeks.

The artistic activity and progress of the Greeks on the Spo-
rades, as well as in the chief cities of the main land, notice-
able during the second quarter of the sixth century b. c, may
in good measure be attributed to the fostering interest of the
Lydian dynasty, and particularly of Croesus. The building of
the Artemision at Ephesus and of the great temple at Miletus
owed much to the proverbial wealth and generosity of this
monarch.

Unhappily the sovereignty of Croesus was not of long dura-
tion. Fourteen years after his accession to the throne the
Lydian Empire fell into the hands of Cyrus. The Troad,
under the name of Phrygia upon the Hellespont, became a
satrapy of the Persian Empire. So rude and unlettered a
people as were the Persians of that age could have had little
intellectual influence upon the countries thus transferred to
their rule.

1 See Bachr*s note on Herodotus, i. 45.

* Aristot. in Stephan. Byz. • Strabo, xiii. 613.

* The inventions of minted money and of inns for travellers were attributed
to the Lydians. See Herodotus i. 94.

^ Arrian, i. 12; Xenophon, iii. 2, and iv. i ; Diodorus, xviii. 5.

70 ARCHMOLOGICAL INSTITUTE.

The fall of Croesus did but change the master by whom a
certain proportion of the produce of the land was levied, the
internal administration remaining almost unaltered. It is a
noteworthy fact that the collectors of the tithes were more
frequently Greeks than Persians. That the tribute was often
oppressive there can be no doubt ; but this was apparently
rather owing to individual exactions of the agents than to
unreasonable demands on the part of the Persian monarch.
The entire tax required from the Hellespontians of the south-
ern coast, Phrygians, Asiatic Thracians, Paphlagonians, Ma-
riandynians, and Syrians (/'. e, Cappadocians),^ — namely, three
hundred and sixty talents yearly, — does not appear excessive.
Assos must have been too long accustomed to dependence
upon foreign rulers to feel that exasperation at the supremacy
of the Persians which, in Greece, led to the later victories of
Salamis, Platoea, and Mycale.

After these signal defeats the Barbarians were driven
from the Asiatic coasts of the -^gean. Herodotus concisely
states,^ that before the invasion of Xerxes there were Persian
governors in Thrace and on the Hellespont ; and that these,
with the sole exception of Mascames, in Doriscus, were after-
wards driven out by the Greeks. The resistance of the
fortified Sestos was an exception deemed worthy of especial
remark.^

It is probable that the towns of the Troad were freed by
the fall of Byzantium (477 b. c), if, indeed, the Persians re-
mained in the land after their decisive defeat at Mycale
(479 B. c). To maintain communication open between the
iCgean and the Pontus, it must have been of primary im-
portance to assure the freedom and fidelity of the Troad.

The rapid growth of the Athenian state led to its alliance

^ Herodotus, iii. 90. 2 Herodotus, vii. 106.

* Herodotus, ix. 114, iiS.

INVESTIGATIONS AT ASSOS, 1881. 71

with nearly all the cities of northwestern Asia Minor, and
probably with Assos, although this name does not occur in
the remarkable inscription which, dating from between 440
and 436 B. c, records many of the cities belonging to the
confederation. Neandria, Kebrene, Lamponia, and even the
colony of Assos, Gargara, are on the list ; and Assos itself can
hardly have been omitted. The object of the union was to
carry on the warfare with the Persians, who were finally
forced to the convention commonly known under the decep-
tive name of the " Kimonian peace," at a date subsequent to
449 B. c. By this treaty, whether tacit or written, the freedom
of the cities upon the coast was fully secured ; no Persian
vessels were allowed upon the ^Egean, and no armaments
within a certain distance from the sea.

With this security Assos may well have had a monumental
renaissance, similar to that of Athens, if upon a smaller scale.
Thasos, near the Trojan coast, offers a striking example of
the material advance made by the Grecian states of the north-
ern iEgean during the decades immediately following the de-
feat and expulsion of the Persians. Darius had deprived the
island of its fleet and razed its city walls ; but only twenty-
five years later, at the time of its revolt from Athens, Thasos
was armed by a strong maritime force, and fully protected by
fortifications.

The part taken by Assos during the Peloponnesian war is
difficult to determine. Its position between the contending
cities of Antandros and Mytilene was certainly not favorable
to peace.

Before the end of this unhappy contest between the Greek
states the Lacedaemonians had assured the return of the Per-
sian despotism to the coasts of Asia Minor, by their infamous
treaties with Darius II. (412 b. c). The Troad did not pass
wholly into the hands of the Barbarians for more than half a

72 ARCHAEOLOGICAL INSTITUTE.

century, being at first subject to the oligarchical govemnient
instituted by Lysander irpmediately after the battle of Aegos-
potam (405 B. c).

Even after the peace of Antalkidas (387 b. c), which deliv-
ered many of the Greiek cities of Asia Minor to the Persians,
a certain banker, Eubulus, maintained himself as master of
Atarneus and Assos independently of the authority of Artax-
erxes. On his death the eunuch Hermeias, a former confiden-
tial servant of Eubulus, succeeded to power over these cities*

Concerning the reign of Hermeias we have fuller informa-
tion than of any other period of the immediate history of
Assos, which is the more fortunate as the city then appears
to have been one of the chief seats of Greek refinement and
learning. Hermeias, a scholar of Plato, and himself thfe au-
thor of a work (now lost) upon the immortality of the soul,
attracted to Assos his fellow-pupils Xenocrates and Aristotle,
the latter of whom was related to him by marriage. Aristotle
lived in Assos for three years,^ and we still possess the mag-
nificent paean composed by him in honor of his benefactor.

Hermeias maintained the independence of Assos until the
year 345 b. c, when he was betrayed by a Persian general,
Memnon (or, according to Diodorus, Mentor), who, under pre-
tence of effecting a reconciliation between the governor and
Artaxerxes III., invited Hermeias to an interview, and sent
him, ignominiously sewed up in the skin of an ox, to the Per-
sian capital, where he was crucified.^ The general thereupon
sent letters, bearing the impression of a seal belonging to the
unfortunate Hermeias, to the cities maintaining allegiance,
stating that the sovereignty had been amicably delivered over

1 Compare Fabridus, Bibl. Gr iii. pp 203, 495, etc ; also Blakeslcy*s Life 0/
Aristotle, pp. 35, 44.

2 Strabo, xiii. 610, and Diodonis, xvi. 52, relate the fortunes of Hermeias, the
former giving the most detailed account of the visit of the philosophers to
Assos.

INVESTIGATIONS AT ASS OS, 1881. 73

to Artaxerxes. Assos again passed into the hands of the
Persians without a struggle. ^

The state had preserved a partial independence for six dec-
ades, and was not long to remain under the rapidly declining
power of the Barbarians. At the time of the fall of Hermeias,
Alexander the Great was of age to receive the instruction of
the fugitive Aristotle. Only eleven years afterwards all Mysia
was freed by the battle of the Granicus (334 b. c). From
Arrian we learn of the Hellenic reorganization of Phrygia
upon the Hellespont after the astounding successes of the
conqueror. But the varying political fortunes of the province
need not be here recounted, as it passed from hand to hand
during the disturbed period of the Diadochi.

Of more concern in the history of Assos was the occupa-
tion of the Troad by the Gauls. The fertile valleys of the
Scamander and Satnioeis were separated only by the narrow
Hellespont and the easily navigable Thracian Sea from these
barbarous tribes, who established themselves in the Cherso-
nesus and Macedonia after the death of Alexander. The
Troad was exposed to the special ravages of the Trocmae,
who for a time settled upon the Acropolis of the later Ilion.

The repulse of the Gauls was due to the rising state of
Pergamon, to which Assos was united in the year 241 b. c.
Eumenes and Attains, refusing tribute, drove the wild tribes to
the coasts of the Hellespont, where they continued their rav-
ages until expelled from Ilion by the inhabitants of Alexan-
dria Troas, and finally defeated in a pitched battle near Arisbe
(216 B. c), after having occupied the land for more than sixty
years.

Sharing the fate of the powerful monarchy of Pergamon,
upon which so much light has lately been thrown by the
excavations at Pergamon itself, Assos passed by bequest of
Attalus HI. to the sovereignty of Rome in 133 b. c. It was

74

ARCHAEOLOGICAL INSTITUTE.

during the period of Roman dominion that the greater part
of the lower town of As§os, now in ruins, was built, the long-
continued peace favoring the extension of the commerce upon
which its existence depended.

A number of the coins of Assos, Adramyttion, and Per-
gamon, preserved in the numismatical collection of Munich,
bear the counter-stamp of an owl, which appears to have been
given them during this period to regulate the value of the
different mintages and to facilitate their circulation through-
out the province. The owl was naturally chosen as a com-
mon emblem, the worship of Athena having been predominant
in the cities mentioned.

During the wars of the Romans with Mithridates, that
ruler occupied Pergamon, the Romans being dislodged from
Adramyttion and possibly also from Assos {^Z to 85 b. c).

Mytilene remained in a state of constant revolt between
the first and second Mithridatic wars, and the situation of
Assos must have led to constant disturbance during those
years. Upon the whole, however, the powerful domination of
Rome secured a long period of tranquillity to the city.

Assos seems to have become Christian at an early date,
perhaps in some measure as a result of the visit of St. Paul
and St. Luke, while on their way from Alexandria Troas
to Mytilene,^ but more probably from the proximity of the
seven churches of Asia, the influence of which was felt
especially at the north. The disciple of St. Peter or St.
John, St. Ignatius, — that great upholder of the prerogatives
of the clergy, — dwelt for some time in the Troad. Marinus,
Bishop of the Troad, was present at the first CEcumenical
Council of Nica^a (325 a. d.), and in the lists of the third
council of Ephesus (43 1 a. d.) occurs the name of Maximus,
Bishop of Assos.

' Acts of the ApoitUsy xx. 13, 14.

INVESTIGATIONS AT ASSOS, 1881. 75

The church militant, with the support of the infamous Con-
stantine, destroyed many monuments of the earlier Greek civ-
ilization in every part of the country. If the temple of Assos,
which had arisen with the freedom of Hellas from Oriental
despotism, remained intact until the age of Theodosius, it had
then little chance of further escape, — the imperial edicts
ordering the closing of all fanes, and permitting any persons
to carry off the hewn stones of their walls, to be used in the
building of dwellings.

The exposed Troad suffered from nearly every blow in-
flicted upon the declining Empire of the East. Under Latins,
Byzantine Greeks, Franks, Seljukian and Ottoman Turks the
Acropolis of Assos was exposed to many attacks, and it is not
surprising that the ruins show its fortifications to have been
levelled to the ground again and again. Assos, like all the
cities of this land, was thus gradually reduced to a miserable
village.

Asia Minor was long exposed to the destructive incur-
sions of the Moslems. The authority of the emperors in
the land was little more than nominal after the beginning of
the eleventh century, and in 1080 the Seljukian Soliman
occupied all the cities of the Troad. The unity of God and
the mission of the Arabian prophet were preached in the
Byzantine church, which had been built with the stones
of the archaic Greek temple of Assos. The history of the
three centuries which intervened between the first appearance
of the nomadic Turkish tribes and the settled establishment of
the Ottoman power presents a wearisome repetition of inva-
sions and occupations.

The unreasoning multitudes led by Peter the Hermit passed
by the land, not inflicting directly upon it the destruction and
misery which everywhere followed in their track. The oppor-
tunity created by this disturbance was improved by the crafty

76 ARCHAEOLOGICAL INSTITUTE.

Alexius, who, in enlarging his empire (1097 A. D.), added to it
the Troad, which had been wholly estranged from the Chris-
tians for a period of seventeen years. Asia Minor was recov-
ered to the banks of the Maeander, and the Seljukian Turks
driven forever from the Troad, to which the Christian ele-
ment was again introduced by colonization from Europe.

The region was more immediately afFected by the passage
of the third crusade (1189 a.d.), — the Emperor Barbarossa
crossing into Asia from Callipolis to Lampsacus, and trav-
ersing the land with the last Christian army which has accom-
plished that feat.

In the contentions between the Franks and Greeks at the
beginning of the fourth crusade (1204 A. D.), Adramyttion was
taken by Henri de Hainault, brother of the Emperor Baldwin.
The extreme sectarian aversion felt between the branches of
the Christian church, and still shared by the Levantines o£
to-day, prepared the way for the final triumph of Moham-
medanism.

Exhausted by continual struggles, the Troad fell irrecover-
ably into the hands of the Ottoman Turks in the beginning of
the fourteenth century. The conquest was finally achieved by
Orchan ; but his predecessor, Osman, had defeated the Greek
fleet at Lemnos in 1288, and soon after had occupied Yenisheri,
near the ancient Sigeion.

It is not plain whether Assos was at any time subject to the
Gattilusii, the Genoese Princes of Lesbos, who obtained their
power in the year 1355 a. d., and, besides holding Lesbos,
Tenedos, Ainos, and the four Thracian Lslands, appear to
have occupied some points of the Trojan coast. Lesbos
maintained an administrative independence until 1463 a. d.,
though it had been tributary to the Turks for almost a cen-
tury previous. One of the hard conditions enforced upon
the Gattilusii by Sultan Mahomet II. was the responsibility

INVESTIGATIONS AT ASSOS, 1881, 'j-j

for all marine damages affecting Turkish vessels upon the
Asiatic coast opposite Lesbos. The tract specified by the
historian Ducas as subject to this condition extended from
the river Crimac^ to Behrim, and this is the first mention
of the Turkish town upon the ancient site of Assos.

The district and civil government of the Troad, which have
remained unchanged in all fundamental respects, were insti-
tuted by Orchan and his brother Ala-Eddin. The subsequent
advance of the Ottoman power into the heart of Europe could
have had no influence upon the Asiatic provinces beyond
insuring their freedom from the miseries of invasions and
sieges.

The long-continued quiet could not bring prosperity to the
Southern Troad, deserted by its Christian inhabitants. Un-
der the enervating yoke of the Turks the sparsely populated
country languished in lethargic repose, severed from all inter-
course with Europe until the advent of the scientific travel-
lers and archaeologists of the past century.

For convenient reference in the study of the development
and decline of the city, — as iHustrated by the monuments, —
the chief periods of the history of Assos may be grouped
under the following dates : —

Pre-historic occupation of the Troad by Semitic,
Phoenician, and Carian colonists ....
Pedasus (Assos) capital of the Leleges . . before looo b. c.

Date commonly assumed for the beginning
of the Trojan war, and sacking of Pedasus by
Achilles : B. c. 1193.
Growth of the ^olic colony .... about 1000 to 560 B.C.
At the close of this period, Assos the most im-
portant city of the Troad.

The influence of Assyria felt by all the lands
of the Eastern Mediterranean, from the age of

^ The ancient Caicos.

78 ARCHjEOLOGICAL institi/te.

Tiglath-Pileser (1120-1100) until that of Assur-
bani-pal (668-626).

Lydian conquest 560105498.0

First subjugation to Persia 549t0 479B.c

Assos a semi-independent state 479 10345 B.C.

The influence of Athens paramount before
405 B. c. (battle of Aegospotami) ; after that date,
establishment of an oligarchy by Lacedaemon.

The rule of Hermeias, at the close of this
period, particularly worthy of attention.

Residence of Aristotle in Assos (348-345 b. c).

Return of Persian ascendancy, prepared by

Lacedaemonian treaties with Darius II., 412 B.C.

Second subjugation to Persia 345 to 334 b. c.

Rule of Alexander the Great and his followers . 334 to 241 b. c.
Invasion and occupation of the Troad by the
Gauls from 288 until 216 B.C. (battle of Arisbe).
Assos embodied in the kingdom of Pergamon . . 241101338.0.

Empire of Rome 133 b. 0.10330 a. d.

Assos exposed to the ravages of the Goths
during their second and third expeditions (264
and 269 A. D.).

Early Christianization of the cit}', and conse-
quent destruction of the monuments.

Empire of Byzantium 330 to 1080 A. D.

Period of continual decline.
Occupation of the Troad by Seljukian Turks . . 1080 to 1097 a. d.

terminated by the first crusade.
Continuation of the Byzantine empire by Greeks

and Latins 1097 to ab't 1330

The Troad in the hands of the Franks from
1204 until 1224 A. D.

Gradual advance of the Ottoman Turks ; vic-
tory of Osman at Lemnos, 1288 A. D.

Final occupation of the land by Ottoman Turks, about 1330

The village of Behrkm, upon the site of Assos,
visited by Choiseul-Gouffier, a. d. 1785.

INVESTIGATIONS AT ASSOS, 1881, 79

The volcanic crater of Assos formed a majestic natural altar
peculiarly adapted for a Greek acropolis. The irregular cone
is divided, as by a terrace, into two steps, which are in plan
so eccentrically related that their fortification walls are united
upon the east. The area of the inner enclosure contains very
nearly 3,cxDO square metres. The summit, which is not a perfect
plane, rises to the highest point at the extreme northwestern
corner. (See Plate 2). The altitude was determined, in the
lack of a level, by repeated barometrical readings to be 234
metres above the sea.

Of the most ancient fortification walls of this inner citadel
only a vestige remains at H, displaying carefully jointed
polygonal masonry of comparatively small stones. From
the position of these blocks it appears that, at least upon
the southern side, the area of the Acropolis has rather been
contracted than extended by the later occupants. The mediae-
val and Turkish ramparts are too rough to deserve particular
attention ; cut stones were employed only for the sill and jambs
of the western gate, still in position. Hastily built of broken
blocks embedded in thick layers of mortar, all the masonry
bears evidence of the frequent demolition which the citadel
has sustained. In digging -around these enclosures a num-
ber of skeletons were brought to light, with broken weapons,
spear-heads, knives, etc. All remained as they had fallen
during the attack or defence of the stronghold, with the
rubbish of which they were covered.

Upon the summit no ruins of ancient buildings were dis-
covered other than those of the temple. How the northern
half of the enclosure was occupied in ancient times is not as
yet evident. The transverse trench at the north, shown upon
Plate 2, though exposing the native rock throughout its course,
struck upon no walls antedating the Middle Ages. The sur-
face of the cliff was so uneven and inclined, that if the existence

8o ARCHAEOLOGICAL INSTITUTE.

of any antique structures whatever be assumed upon the
northern half of the Acropolis, it is apparent that they must
have been founded upon a terrace of earth which has long^ since
been washed away.

At the south the volcanic rock presented a more even
surface; and, by the help of quarrying and filling out with
courses of masonry, a level of considerable extent was secured
as the site of the chief sanctuary of the city. In all the
wonderfully picturesque lands inhabited by the Greeks, no
site of a building was more imposing and beautiful than
that of the temple of Assos. The peak rose so steep, that,
standing within the peribolos of the fane, one could look down
into the holds of the vessels in the port beneath, and so
high that the foundations of the temple were at an elevation
half as great again above the sea as are the finials of the .
slender spires of Cologne above the Rhine, or the apex of the
great pyramid of Gizeh above the Nile.

The constructive details of the temple of Assos, though
wisely planned and carefully executed, were, from the nature
of the material employed, not of the delicacy observed in the
limestone structures of Attica. The carving was bold and
effective, but somewhat blunt in the smaller members ; the
jointing was perfectly close but irregular.

If is a peculiarity of this building that the cliff Itself was al-
lowed to remain as the stereobatc wherever this was possible,
— in two instances, indicated by asterisks upon Plate 7, even
rising to the level of the naos pavcMncnt, and serving directly
as the foundation of the cella wall. A great part of the peri-
bolos enclosure was made by smoothing the summit of the
crater, as is evident from the plan ; the rock forming almost
the whole of the northern and more than half of the western
bed. Upon the south and southeast the rock here and there

I

rises to the level of the lower step, these points being indicated

INVESTIGA TIONS A T ASSOS, 188L 8 1

on the plate by asterisks. The paving slabs which occupied
the interstices have remained only at the north of the fane,
the destruction of later ages having reached a greater depth
upon the south 'and east. The natural rock was, however,
not permitted to form the stylobate or the lower step, it being
here quarried to the level of the surrounding plane.

At the southwestern corner of the building the depression
in the rock, to be filled with a substructure of masonry,
was particularly deep. A pit sunk at this point to a depth of
1.6 metres showed the even and carefully jointed courses to
project slightly, like the well-known foundations beneath the
southern steps of the Parthenon. (See the section upon
Plate 7.) A firm bedding for the steps, whether cut from
the native rock or formed by a substructure of masonry, was
thus carefully insured. Notwithstanding the many earth-
quakes which are known to have affected Assos, the entire
crepidoma of the temple has remained unshaken.

The two steps were formed of blocks varying in length
from I to 3.2 metres of a nearly uniform thickness of 0.28
metre. The lower course was brought into position by
knobs left upon the^ exposed faces of the stones even after
the completion of the building. Next to the lateral surfaces
of contact, — upon the exposed front and upper edges of the
blocks of both steps, — there were also left thin {o.ooi m.) and
narrow (002 m.) projecting fillets, to obviate, in as far as pos-
sible, the chipping and defacement of the joints. The oblit-
eration of such legitimate technical makeshifts was contrary
to the spirit of Greek workmanship.

The stones were bonded together by iron clamps, cast in
lead ; no system was observed in this connection, either one
or two clamps being employed for each joint at irregular dis-
tances from the front edge of the step. The length of the
stylobate blocks, at least upon the remaining sides, was not

82 ARCHAEOLOGICAL INSTITUTE,

determined by the width of the inter-columniations, the posi-
tion of the columns in relation to the joints being entirely
irregular. Beneath the shafts, the fillet on the edges of the
stones was removed.

Where the pavement of the pteroma and pronaos did not
rest immediately upon the native rock, its foundations were
not constructed of the courses of masonry deemed necessary
for the steps. In three places where the paving blocks of
the pteroma were missing, the natural surface of the cliff,
uneven anil untooled, was exposed by the excavations at a
depth of from 0.6 to 0.8 metre. Upon the plan, Plate 7, this
rock is indicated by daggers. It was* covered with chips of
trachyte, evidently resulting from the carving of the building
blocks. Upon the firm bed thus provided there rested rec-
tangular paving slabs averaging 0.18 metre in thickness.
The system of jointing observed in the pteroma was irregular,
though transverse blocks with a width of about 0.57 metre
were common.

The level of this pavement was not so high as the general
level of the stylobate by 0.015 metre ; and this sinking, taken
in connection with the irregular character of the jointing, seems
to point to the original existence of a flooring of cement. The
stones of the pavement abutted in places directly upon the ver-
tical surface of the wall, as is the case for instance next to
the southern antae ; but more frequently the slabs did not
meet the irregular foundations of the wall, and the considerable
interstices thus remaining could not well have been otherwise
filled than by the cement generally employed in primitive
Doric constructions. It is natural that no vestiges of such a
thin layer of stucco should have survived the exposure of the
pavement to the weathering of fifteen centuries, and its occu-
pation as the floor of mediaeval and Turkish dwellings.

Upon the rear of the building the pteroma pavement has

/Nl'ESTfGAT/ONS AT ASSOS, 1S81. 83

rbeen entirely carried off ; upon the front only the course of
stones next to the upper step is missing. Those following
show a projection in the axis of the entrance, 2.7 metres wide,
the purpose of which is not evident, and to which no great
Importance can be attached in view of the irregular character
of the jointing.
Within the pronaos, sinkings at A A, Plate 7, expose a lower
foundation, which appears to have supported pedestals natu-
rally to be assumed in that situation.

The beddings of the deor-jambs are cut upon the lower sill,
which makes evident the width of the portal and the thickness
of the wall between pronaos and naos.

I The interior pavement of the enclosure is preserved in some
vestiges of a mosaic formed of cubes of black and white mar-
ble. Enough of this remains to insure the restoration of the
design, the return being fortunately preserved upon a frag-
ment at the northwest. A border of bands and the broad
Greek wave ornament enclosed a field of diamond pattern.
This mosaic rectangle probably occupied that part of the
^_ naos, before the sacred figure and the bema, which was open
^H to the worshippers ; its area corresponds, in relative extent, to
^P the similar spaces in the plans of the great temple of Zeus at
Olyrapia and of the Parthenon. It is impossible to determine
the age of the mosaic, but it may be supposed to date from a
late restoration. The inner pavement of the sanctuary was
naturally that part of the building first worn away and most
easily replaced. The stones of the mosaic were laid in a floor

»of cement, which remains to a considerably greater extent
than the pattern. Beneath this the entire area of the naos
was covered with fine earth, which in part appears to have
accumulated during the occupation of the site by dwellings,
in part is evidently the original bedding of the floor.

The foundation stones of the cella walls were in position

^

84 ARCHAEOLOGICAL INSTITUTE.

throughout their extent, with the exception of two blocks
next to the northwestern corner. These stones were of
irregular shape and size, brought to a plane upon the upper
surface, to receive the imposed masonry, but otherwise rough
and unhewn, since they were hidden from sight, upon their
inner edges by the pavement of the naos, upon the outside by
the cement floor of the pteroma and pronaos. Upon these
blocks, and upon the two exposed surfaces of the natural rock
before mentioned as sharing their functions, the outer line of
the cella wall was engraved in its entire extent. The temple
crepidoma, thus characterized technically as well as ideally as
an d/3a|, was directly employed by the Greek master-builder
as a drawing-board. On the plan, Plate 7, these delicate inci-
sions are given in broken lines, being distinguished from the
measurements in line-dot, and the traces of weathering at the
bottom of the columns in dots. The lack of this engraving
upon the interior points to a less careful execution of the
inner surface of the wall, which probably bore a coating of
stucco.

The thickness of the walls of the antae was indicated by
these lines. In the lack of similar evidence for the lateral
walls of the naos these may reasonably be assumed as of equal
thickness to the division walls between naos and pronaos.
Examples of this manner of construction, where the enclosing
walls are thinner than the free-standing antae, though com-
paratively rare, are still not wanting among the peripteral
Doric temples hitherto known.

The position of the foundation stones and the engraved
lines upon them display an exceptional feature of the plan ;
the cella was wholly without an epinaos, the plain wall of its
rear being carried across the west at the same distance from
the steps as upon the sides.

The two columns of the pronaos in antis stood upon square

INVESTIGA TIONS A T ASSOS, 1S81.

85

slabs, considerably larger than the adjoining paving stones,
beneath which foundations of masonry probably descend to
the native rock. These shafts, protected by the ceiling of
the broad front pteroma before them, were but little exposed
to the weathering, and the position of their lower drums is
distinguishable only by microscopical traces. Ten columns
upon the northern side and eight upou the south have left
more distinct marks, from which the number and position of
the lateral shafts are evident. It is particularly unfortunate
that the lack of the stylobale upon both ends has rendered it
impossible to ascertain the various widths of the inter-colum-
niations of the front and rear. With this single exception,
which has been hypothetically made good according to the
striking analogy of the Theseion, the restored plan, Plate 8,
is accurately determined from the remains.

Not one stone was found in position above the stylobate.

The restoration of the supcrstmcture was consequently a

}rk of considerable difficulty, requiring the most careful

fch for important blocks. The drums of the columns, scat-

l.Upon all sides of the Acropolis and built into the enclos-

86 archjeological institute.

ing fortifications, varied in length from a6 to i^ metres. To
ascertain the height of the shaft several hundred measure-
ments of these blocks were necessary, their comparative
shortness being unfavorable to the investigation.

A difference of 0.02 metre was observed in the lower
diameters, but the great number of bottom drums rendered
the given average trustworthy. The twenty capitals remain-
ing upon the site allowed a similar calculation for the upper
diameter of the shaft, of which the individual variation was
nearly as great. By measuring each diameter of the inter-
mediate drums eight times from arris to arris, the proportion-
ate diminution of every truncated cone was ascertained.

The results thus obtained, contrary to expectation, averaged
exactly the same for upper as for lower drums ; thus proving
that the columns were without the entasis, which would have
required a considerably greater diminution above than below
one-third the height of the shaft. This lack of entasis is
perhaps explicable by the small dimensions of the temple
and the hard and coarse nature of the material of which it
was built. According to the statement of Durm,^ the col-
umns of Corinth, which are in other respects similar to those
of Assos, are also without entasis, and it is possible that
this refinement was not generally introduced until a more ad-
vanced period in the development of the Doric style. It has
been mentioned, that, owing to an injury to the levelling instru-
ment, the question of the curvature of the horizontals could

1 Josef Durm, Die Baukunst der Griechen, des Handbuches der Arckitik*
tur zweiter Theil : Darmstadt, i88i ; p. 63. The author evidently refers to
original investigations, as the older authorities upon the ruins of Corinth —
IJlouct, Expedition de Morie, vol. iii., and Stuart and Revctt, Antiquities of
Athens ^accessible to the present writer only in a translation) — do not refer
to the entasis directly. Krcll, Geschichte des Dorischen Styls, on the other
hand, states that an entasis existed, but whence his information is derived is
not stated.

INVESTIGATIONS AT ASSOS, 1881. 87

not be definitely determined during the past year ; but so far
as the observations went the stylobate appeared perfectly
level, — any deviation as great as o.oi metre would have
been readily recognized by the reversed readings. This ap-
parent neglect to counteract the optical deceptions of math-
ematically exact lines agrees entirely with the omission of an
entasis, which was designed for a similar object to that of
the curvature.

The proportionate diminution determined by the difference
between the lower and upper diameters of the shaft fixed the
height of the column. The given dimension can hardly vary
more than 0.08 metre from the truth.

The lower surface of the bottom drum generally displayed
the slot cut for the centre peg by which it was turned upon
the customary lathe. In some instances this sinking had
been obliterated by the shortening of the block. For if the
total height of the several drums intended to be fitted together
to form a shaft was found before their erection to be too great,
it was at the base alone that a decrease could be effected, —
the surfaces between the drums requiring the steadying
centre presently to be described, and the juncture with the
capital, like all the intermediate joints, not allowing any
change of diameter.

The upper surface of the lower drum, and both planes of
every one superposed (with the exception of the uppermost, on
which the capital rested, where the slot of the turning cen-
tre peg remained), showed a hole cut for a cylindrical pin of
wood about 0.045 n^etre in diameter, which served as an axis
for the g^nding of each stone upon the one next beneath. In
the perfected Doric buildings of Attica this pin was enclosed,
and worked in cubical boxes of the same material, cemented
into the opposite drums with red lead. In the temple of
Assos the solicitude for accurate juncture had not been

88 ARCHAEOLOGICAL INSTtTl/TE.

carried so far, the woodea axis bearing directiy upon the
stone, in the centre of which a cylindrical hole was cut to
receive it.

As can be seen from the sketch, Plate 9, the plane surfaces
of the drums were so tooled as to present points of contact
only in a concentrical band, about 0.1 metre bioad, upon
their edges, according to the practice universal in all Greek
architecture of good period.

^ Plate 9,

LThe shafts of the peripteros had sixteen channels, those of
the pronaos eighteen. It is an inexplicable and unique ar-
rangement of the channelling upon the columns of the temple
of Assos that arrises, not hollows, were in the axes of the
plan, and in line with the faces of the abacus. This peculi-

^m

m^

INVESTIGATIONS AT ASSOS, 1881. 89

arity was evident from the weathered marks of the lower
drums upon the stylobate, as well as from the termination of
the channelling upon the necking of the displaced capitals.

It is evident that with their eighteen channels the shafts of
the pronaos presented a hollow in the line of their lateral axes
better fitted to receive the transverse bars of the grille^ cus-
tomarily employed as a barrier between pteroma and pronaos,
than the sharp edge of an arris. Still, it should be remarked
that upon the single drum of eighteen channels which was
found during the excavations, no traces of such a metallic
barrier were to be detected.

Plate 10.

From the lower surfaces of the capitals it appears that the
juncture between them and the upper drum of the shaft
formed an incision. The channellings, as is shown in the
outline of the necking and echinos, Plate 10, were terminated

90 archjeological institute.

by direct intersection with the lower annulet The three
annulets projected in nearly horizontal planes, in some m-
stances the first slanting slightly upward from the shaft,
while the two following were almost imperceptibly under-cut
The outline of the echinos is of great vigor and. beauty,
the upper termination, hidden from the eye, being generally
treated as a straight line, meeting the lower surface of the
abacus at an angle of forty-five degrees. The variations of
the individual capitals are chiefly noticeable in the diameter
of the surface adjoining the upper end of the shaft, and in the
width of the abacus. The height of the necking is one of
the most constant dimensions of the structure.

Not one surely recognizable block of the cella wall remains
upon the site. The identification of the stones composing
the most ancient portion of the neighboring Byzantine church
as belonging to the walls of the temple is more than probable,
but leads to no result. Among the blocks lying near the
temple was one which may prove to be the inner lintel of the
naos door, and another which seems like a fragment of a capital
of one of the antae ; but this remains to be determined. From
the marks upon the foundation stones it is evident that the wall
throughout its extent was without projections in plan, and
hence probably plain upon its surface.

The epistyle beams, as in the Parthenon, were triple, — an
exceptional number for so small a construction, the entire
member measuring only 0.82 metre in thickness. It is ex-
actly as broad as high, while the epistyle of the Parthenon,
of more than double the absolute dimensions, is one-third
again as broad as high. The middle beam did not occupy
the entire height of the epistyle, the outer blocks being so
thickened upon the upper half as to meet above the block
between them. (See the section, Plate 11.) It is difficult
to advance a satisfactory explanation of this peculiar con-

I.vyEST/GATIOA'S AT ASSOS, 1831.

9»

struction. The saving effected in the weight of the facing
blocks was more than counterbalanced by the additional
labor required to cut stones, naturally splitting to parallel and
rectangular planes, in-
to the irregular shape
thus determined ; and
the difficulty of aasur-
. ing exact joints upon
the soffit was rather
increased than les-
sened by the duplica-
tion of the surfaces of
contact.

The outer face of
the epistyle, being
sculptured with reliefs
requiring an architec-
tural frame, was bor-
dered upon the bottom
by a band which is not
found in any other
Doric building. Tae-
nia and regulae were
of comparatively
slight projection, the
latter being without
tninnels. The plain
epistyle blocks with-
out lowerborder.found
during the investiga-
tions, probably be-
longed to the inside.

Thai both the outer

92 archjEological institute.

and inner beams of the epistyle were dowelled to the upper
surfaces of the capitals is evident from the swallow-tail sink-
ings upon the ends of their soffits. These marks, occurring
upon the fronts of the sculptured epistyle, present the most
conclusive proof that these remarkable blocks were above the
columns and inter-columniations, and not upon'the cella wall,
where a projection as great as that of the abaci could not
have been provided by any continuous moulding.

The edges of the triglyphs were under-cut so as to afford
a reveal, into which the thin slabs of the metopes, whether
sculptured or plain, could be slid from ^bpve. A remarka-
ble variation is noticeable in the width of the triglyphs, which
appear of two dimensions, — 0.52 and 0.56 metres. That the
narrower blocks were situated above the columns is evident
from the fact that the corner triglyphs, three of which were
fortunately found, were of the smaller width. Like the outer
epistyle, the sculptured metopes were provided with a base
band. The slightly projecting band which crowns triglyphs
and metopes was of the same width over both members, thus
forming a continuous line along the upper part of the frieze.
The metopes were further terminated by a narrow and delicate
Lesbian cyma, — a crowning and connecting member similar
to the astragal occupying this position upon the Parthenon.

The mouldings indicated by Texier as existing above the
frieze are wholly imaginary. The increased projection of the
cornice arising from their introduction would have given to
the corner mutules a disproportionate width, which would
have been without a parallel in the style. Corner blocks of
the corona proving the non-existence of the moulding were
found among the ruins, while the constructive impossibility
of interposing a continuous band of trachyte only 0.106
metre thick between the mighty stones of the Doric en-
tablature is evident from the French restoration itself.

INVESTIGA TIONS A T ASSQS, 1881. 93

The general arrangement of the cornice was such that the
corona blocks extended from centre to centre of the triglyphs.
Reposing directly upon the frieze, the stones were so cut as
to provide a bed of nearly two-thirds the thickness of the
entablature. Upon the ends of the blocks U-shaped grooves,
like those noticeable in Selinus, iEgina, and other Doric
sites, were cut to receive the ropes by which they were lifted
to their positions. (See the fragment of a corona block shown
on Plate 9.) An exact jointing was secured by restricting the
surfaces of contact to a band upon the edges, like the concen-
tric bands on the drums of the columns.

The soffit of the corona was so divided that the mutules
above the metopes upon the side of the building were only
about three-fifths as broad as those over the triglyphs.
Upon the front the greater width of the inter-columniations
increased this proportion to seven-eighths ; and a similar in-
crease was noticeable next to the corners of the sides, where
the triglyph was not in the axis of the column. Like the
regulae, the mutules were without trunnels (guttse).

Behind the triglyphs and metopes there probably was
placed a plain backing as an interior frieze, upon the upper
surface of which, not occupied by the corona blocks, reposed
the ends of the pteroma ceiling-beams. A coffered stone,
which possibly belongs to this part of the temple, is built into
the wall of the Byzantine church, with its soffit outward. As
' may be seen from the sketch, Plate 23, it is so high above the
ground that it was not possible to measure it without ladders,
which were not at hand. The block is evidently less than
2.43 metres long, and consequently shorter than the clear span
of the pteroma ceiling, — so that, if it be assumed to belong to
the temple at all, it indicates a complicated system of trans-
verse beams.

One of the three corner corona blocks which were found

94

ARCH/EOLOGICAL LVSTJTUTE.

1

gave the approximate angle of the gable, while the discoverj'
of one fragment of the inclinecJ gable corona determined (he
reveal of the tympanon and the character of its border. The
soffit of this important cornice was under-cut in the usual
proportion, the projection being separated from the upright
tympanon veil by a Lesbian cyma.

The entire structure thus far considered was built of the
second trachyte of the Acropolis. In the lion's head of the
corner gutter the first appearance of another material is noted,
that gargoyle being of a lighter and softer stone than the tra-
chyte, — like it of volcanic origin, but stratified by the action
of water. The upper half of this fine head (Plate 12) was one

of the most interesting fragments discovered during the year.
It displays, even more strikingly than the sculptures of the
epistyle, the round and flat, yet sharply detailed, forms peculiar
to the empaistic work abundantly produced upon the eastern
shores of the Mediterranean during the ages of Phcenician
influence. The form of the teeth, the ribbed roof of the
mouth, the angular furrows which suggest the whiskers upon
the upper lip, — in short, every detail of the head shows a

INVESTIGATIONS AT ASSOS, 1881.

95

power of animal characterization which corresponds with the
masterly treatment of the lions and boars of the reliefs.

The general dimensions of the crowning gutter (sima) upon
the fronts were evident from this gargoyle, and the charao

PlMTE 13.

tet of the side cornice was similarly determined by the dis-
covery of lower tiles (imbrices) and an antefix. The general
arrangement of the roof was thus shown to resemble closely
that of the Parthenon and of the temple of Rhamnus.

The terra-cotta tiles bore the black glaze observed upon
such remains at Olympia, Argos. and Mycenas, but not, so far
as the writer is aware, upon any of the older examples of Sicily.

96 ARCHAEOLOGICAL INSTITUTE.

Only small fragments of the lower plates were found, and no
remains whatever of the joint-tiles (calypteres). The upper
end of the former was provided with a projecting band to hook
on to the timbering of the roof. In the detailed restoration of
the roof (Plates 1 1 and 1 3) the lower tiles have been assumed
to extend from axis to axis of the mutules, with an ordinary
width of 0.61 metre. This dimension is not great, compared
with the tiles elsewhere employed for Greek temples, which

#

often measure 0.8 by i.i metre, and in some instances even
attain a length of 1.2 metre. The width of the tiles at Assos
appears to be determined by the existence of antefixes, which
could hardly have been otherwise situated than above each
mutule. In all Doric temples the tiles appear to have rested
directly upon the rafters, there being no cross slats ; the as-
sumed agreement with the mutules would confirm the sup-
posed derivation of the entablature and roof from a wooden
prototype.

The broken terra-cotta antefix which was found, displayed
the rich red-and-black of the archaic Doric polychromy, and
showed the form of these terminations of the joint-tiles. No
information was obtained in regard to the acroteria, the ridge
and joint tiles, and the terra-cotta gutter of the fronts.

The following table presents the chief dimensions of the

building : —

Metres.

Length of lower step 30.885

Breadth " " 14.585

Allowance for width of each step . . . average 0.275

Length of stylobate 30-335

Breadth " " ......... 14-035

Exterior of cella, length 22.360

" " " breadth 7.965

Enclosing walls of naos, thickness • . . about 0.6

Walls of antae, thickness % . . 0.660

Door of naos, breadth • 1-650

t< u «

(( (( <( (< (( (i ((

«

INVESTIGATIONS AT ASSOS, 1881. 97

Interior of naos, length about 17. 865

*' " " breadth about 6.765

Total width of pteroma, sides and rear .... 3.035

*' " " " front, before antae . . • 4-940

Columns from centre to centre, sides .... 2.449

front corners, assumed about 2.5

2.7
Lower diameter of shaft t . • . • . average 0.915

Upper " " " " 0628

Height of steps * " 0.280

" " column calculated about 4.78

" shaft " "4.3

" capital average 0.480

" " epistyle . *. • 0.820

" " frieze members 0.784

"* " corona 0.416

Total height of order, from pavement to upper

surface of corona . . . calculated about 7.36

In comparing these dimensions with the intention of re-
cognizing the unit of measure employed in the building, it is
noticeable that the width of the side and rear pteroma is as
nearly as possible one-tenth of the length of the stylobate.
This relation of the most important divisions of the plan is so
strikingly exact as to exclude the assumption of a coincidence.
It is hence extremely probable that a system of decimal feet
was employed, or that 3.0335 metres contains an entire num-
ber of the original unit of measure.

If the plan be supposed to be 100 feet long, and the pteroma
10 feet, a foot of 0.30335 metre would result, — a dimension
varying but very slightly from the Greek foot as determined
by M. Aurds (0.307 metre),* by Don Vasquez Queipo (0.30864
metre),* and by Boeckh (0.30821 1 metre). ^ In this case the

^ jStude des Dimensions du Grand Temple de Pastum^ p. 4. Paris. 1868.

* Essai sur les Systimes mitriques et monitaires des anciens Peuples, Paris.
1859. Tome i., p. 387.

• Deduced from Boeckh*s estimate of the Roman foot by Charles Eliot Nor-

7

98 ARCH^OLOGICAL INSTITUTE.

thickness of the cella wall would appear as 2 feet, of the antse
walls as 2.2 feet, the width of the naos door as 5.5 feet, the
lower diameter of the column as 3 feet, etc. A suggestion,
perhaps more plausible, has been made by my friend Richard
Bohn, architect of the excavations at Pergamon, that the
dimensions were respectively 9 and 90 feet, of a consequent
length of 0.337 metre. The breadth of the naos interior would
thus appear as 20 feet, its length as 53 feet, etc.

For those not accustomed to the metric system it may be
stated that one hundred English feet equal 30,479 metres, or
less than six inches more than the length of the stylobate.

To serve in comparison with the useful table compiled by
Krell, in his Geschichte des Dorischen Stils, the proportions of
the temple of Assos may be given as follows : —

Distance from axis to axis of the side columns, measured

by halves of the lower diameter 5.3^

Width of the side and rear pteroma remaining between the
inner side of the peripteral columns and the cella
wall, measured by the lower diameter 2.31

Semper's norm ' for the sides

16.0

10-45 + 4-40= 14.85

{column 61.0

epistyle 10.5

frieze lo.o

corona 5.3

Height of column in lower diameters 5.23

Proportion of height of capital to height of column . . i : 9.96

ton, in The Dimensions and Proportions of the Temple of Zeus at Olympia^ in the
Proceedings of the American Academy of Arts and ScienceSy Boston, 1877, p. 15a

1 Compare Gottfried Semper, Der Stil, etc. Miincheny 1863, vol. ii., p. 411. If
three inter-columniations, from axis to axis of the columns, be taken as the base
of a rectangle, the side of which is equal to the height of the order, — calculated
from the upper edge of the stylobate to the summit of the corona, exclusive of
the gutter, — the normal proportion of plan and elevation, or as it is concisely
termed the " norm " of a temple, is graphically represented. When expressed
in figures, one-half the lower diameter of the shaft serves as the unit, the dimen*
sions of the column and entablature being given separately.

INVESTIGATIONS AT ASSOS, 1881, 99

( peripteros 16
Number of channelhngs : jpronaos 18

Number of annulets 3

Number of necking incisions i

Proportion of height of capital to width of abacus . . . i : 2.5

Proportion of width of abacus to space between the abaci

of the side i: 1.04

Proportion of the height of abacus to the height of echinos

and rings i : 0.95

Height of capital divided by one-half the upper diameter

of shaft 1.63

Width of abacus divided by one-half the upper diameter

of shaft 3.72

The general untrustworthiness of the Descriptian de PAsie
Mineure has already been referred to. The description of
the temple of Assos, presented in that work, appears almost
worse than valueless.

The remains now unearthed show the Orientation of the
building to have varied considerably from the east to the
south ; Texier places it thirty degrees to the north of its true
direction. The two steps are increased to three upon the
French elevation, to four upon the fronts of the plan. The
disposition of the plan given in the fine steel engraving, with
its double dipteral ranges of columns upon the east, and the
epinaos in antis upon the west, must have been conceived by
the ingenious author after his return to Paris. The width
of the building is given on the plan as 23, on the elevation
as 13 metres. The excessive, sack-like entasis of the shafts,
which has given rise to many wild theories, did not exist. The
striking arrangement of the channel arrises in the axes of the
building was overlooked, whilg important members, which
never existed, were added to the entablature, these being,
with unparalleled effrontery, scaled to the millimetre, as if
accurately measured ! The projecting mouldings inserted be-
tween frieze and corona are wholly at variance with the char-

lOO ARCHAEOLOGICAL INSTITUTE,

acter of the style. It is needless to multiply illustrations of
this manner of reporting scientific investigations.

Such being the character of Texier's account of the temple,
and it having been, up to this time, the only published source
of information, the result of the work of the present Expedi-
tion, as recorded in these pages, may fairly rank as the direct
recovery of one of the most important monuments of the
Doric style, that noblest and first-born offspring of Greek
architectural genius. (See Plate 14.)

The temple of Assos is the only known Doric peripteros
in all Asia Minor, with the exception of the fane of Athene
Polias,* recently excavated at Pergamon, which was built at
a much later period. The historical interest of the temple
is evident from the attention devoted to it by every writer
upon the development of Greek architecture and sculpture,
even while the building has been most imperfectly known.

Its age appears more accurately to be determined from its
architectural characteristics than from the archaic but pro-
vincial reliefs sculptured upon its epistyle. It is the writer's
belief that the building of the temple of Assos is to be referred
to that activity spoken of in the historical sketch as affecting
all the lands of the northern iEgean immediately after the
battle of Mycale and the expulsion of the Persians. The argu-
ments for this view can be little more than indicated in this
preliminary account. While the temple shows many signs of
having been built during a period of development previous to
the canonical determination of every detail of the style, yet its
general disposition, — especially in the decisive points of the
axes of the plan, and the relative dimensions of the eleva-
tion, — is far more advanced than that of the archaic Sicilian
temples.

The provincialism of Asia Minor during the first half of

1 Der Temptl der Athena Polios zu Pergamorif von Richard Bohn. Berlin. 1881.

INVESTIGATIONS AT ASSOS, 188L lOI

the fifth century b. c. amply accounts for the appearance of
primitive traits in the temple of Assos at a time when the
architecture of Attica had reached its full development. The
Asiatic provinces, which for six decades had suffered from
Lydian and Persian occupation, bore in the year 475 b. c. a
somewhat similar artistic relation to European Greece to that
which the eastern shores of the Adriatic bore to the western
during the later ages of the Roman Empire. It was not to
be expected that the full advance displayed in the temple of
./Egina, or the Theseion, should be shared by contemporaneous
buildings in Mysia.

Even from this point of view, however, the temple of Assos
must be classed as one of the more primitive examples of that
phase of the Doric style, designated by Semper as the " fully
developed archaic." The only other peripteral temples in
which the epinaos is known to have been omitted are those
extremely ancient monuments at Selinus, designated as the
temples C, D, and 5, and the fragmentary remains near Cadac-
chio upon the island of Corfu. The epinaos, unknown in the
primitive temple in antis, seems to have had no purpose con-
nected with the service of the temple, there being no entrance
through it to the naos, so that its introduction may be regarded
as a concession made to the formal symmetry of the edifice at
a time when the general arrangement of plan was still under-
going development.

The constructive character of the temple of Assos and the
irregularity of its details show that the building antedates
the time when the entire fane, down to the most inconsider-
able members, was laid out according to a systematized canon.
The individual variations, noticed in its different parts, seldom
occur in later buildings, but are sufficiently common in archaic
temples. The variation of 0.09 metre in the lower diameters
of the columns of the great temple of Zeus at Olympia is pro-

I02 ARCHAEOLOGICAL INSTITUTE.

portionately greater than that of those of the temple of Assos.
The sculptured epistyle is entirely exceptional in the Done
style, and points to Oriental reminiscences of great antiquity.
The awkward division of the entablature in such a manner as
to form fractional mutules above the metopes finds a striking
parallel in the most ancient temple, C, of Selinus.

The example of the columns upon Cape Sunion proves that
the occurrence of the sixteen-channelled shaft is not neces-
sarily an indication of great age, — as might be supposed from
its appearance in Selinus and Olympia, — being at Sunion,
as perhaps at Assos, due to the great elevation of the temple
above the sea and neighboring plains. Still the abnormal ar-
rangement of the arrises in the axes of the plan is so primitive,
that, taken together with the number of the channels, it affords
an indication either of the remote date of the shaft or of a
provincialism in its treatment, the exact effects and limits of
which are difficult of determination. At all events the aesthetic
object and the technical significance of the channelling were
imperfectly appreciated at Assos.

The outline of the capital, always of great importance in
the determination of the age and relative position of monu-
ments of the Doric style, indicates a decided advance upon
the forms of what has been called the lax archaic period. The
Sicilian apophyge has been entirely given up, the simplest pos-
sible juncture of necking and annulets being effected. On the
other hand, the great angle formed by the lower echinos with
the shaft is still retained, and the general character of the curve
is similar to that of the oldest capitals. The annulets are not
as yet organically connected with the echinos. (See Plate lO.)

Arguments tending to prove the primitive character of the
temple might be derived from the lack of trunnels upon the
regulas as well as upon the mutules, from the extremely blunt
forms and slight projection of all the band mouldings, nota-

INVESTIGATIONS AT ASSOS, 1881. 103

h\y of the taenia, and from the archaic colors displayed by the
terra-cotta antefix.

The most trustworthy and conclusive results are to be ob-
tained from a comparison of the table given above with the
known proportions of other Doric temples. The height of
the column expressed in lower diameters shows the temple
of Assos (5.23) to stand between the temples of Athena
(4.27) and of Artemis (4.29) at Syracuse, of the temple at
Corinth (4.32), of the temple {D) at Selinus (4-50), etc., on the
one hand ; and the Theseion (5.62), the Parthenon (5.47), and
the temple of ^Egina (5.30), on the other. The relation of
Semper's norm for these buildings is particularly interesting,
but the statement of it is of too great length to be given here.
A further comparison of proportions, leading to the same re-
sult, is best made between the heights of epistyle, frieze, and
corona, the width of the pteroma, and the relative diminution
of the shaft.

A remarkable similarity of absolute dimensions is notice-
able between the temple of Assos and the Theseion. In the
Theseion, for instance, the breadth of the stylobate is 13.816
metres, in the temple of Assos 14.035 metres; in the former
the breadth of the cella upon the exterior is 7.928 metres, in the
latter 7.965 metres. The number of columns upon front and
sides, the orientation south of east, in neither case necessitated
by the configuration of the ground, and even the exceptional
reduction of the steps to two, are in both temples the same.
While the plans of the archaic monuments of Selinus show
a helpless irregularity of general arrangement, the temple of
Assos presents a developed disposition of parts attainable only
after many experiments.

From the situation of the pronaos columns, exactly in the
lateral axis of the third shafts of the side, the existence of
transverse lintels above the pteroma is rendered almost cer-

I04 ARCHAEOLOGICAL INSTITUTE.

tain. As may be seen from Plate 8, the pteroma before the
pronaos was thus provided with an independent ceiling in the
entire breadth of the building, the space being thus separated
from the side colonnades, and characterized as a vestibule, —
in so small an edifice an arrangement of great advantage to the
general composition. From a review of the plans of peripteral
Doric temples, it is evident that such an advance could not have
been made until a comparatively late period. This peculiarity,
observed elsewhere only in the Theseion and in the temple of
Sunion, is one of the strongest arguments against the assump-
tion that the temple of Assos was built before the Persian wars.
The relative depth of pronaos and naos ; the width of pteroma
and cella ; the arrangement of gutters, gargoyles, and ante-
fixes ; the black glaze of the tiles, — all point to a perfection
of the Doric style not to be expected on the coasts of Asia
Minor earlier than the date which has been assumed for the
construction.

That the temple of Assos, the chief building of the city,
was consecrated to Athena there can be but little doubt, in
view of the invariable occurrence of the head of that deity
upon the obverse of all the coins of Assos, excepting those of
the later Roman Empire, when the portraits of the rulers were
substituted. This supposition is rendered almost a certainty
by the mention of ** the pure virgin whom our Fathers wor-
shipped '* upon the bronze inscription, discovered during the
past year, which records the oath of allegiance taken by the As-
sians upon the accession of Caligula. The worship of Athena
was universal throughout Mysia. Even in the Homeric legend,
her temple was the principal sanctuary of Troy ; she was the
patron of Pergamon, Adramyttion, and other cities in the
neighborhood, as well as of Assos. The only other Doric
temple of Asia Minor, — the building upon the Acropolis of
Pergamon, which has already been referred to, — was, like

INVESTIGATIONS AT ASSOS, 1881. 105

the far more ancient temple of Assos, dedicated to Athena
Polias.

It is one of the most interesting, and at the same time
one of the most complicate, problems of classical archaeology to
determine in what measure Greek art, which in so short a time
rose to such marvellous perfection, was founded upon the
architecture and sculpture of older and foreign races.

A most valuable addition to our knowledge of archaic and
advancing sculpture has been made during the past year by
our discovery of eleven fragments forming eight complete
reliefs of the carved epistyle and metopes of the temple of
Assos. These sculptures illustrate, as does no other series of
connected works, the gradual Hellenization of Oriental types
and artistic methods.

A thorough consideration of these discoveries must neces-
sarily treat in detail of the seventeen fragments (thirteen
reliefs) of the series which were removed, in 1838, to Paris,
and is reserved for another publication of the Institute.^
The sculptures from Assos in the Louvre have for over forty
years attracted the attention of archaeologists and historians
of art, and the literature concerning the significance of their
representations, and their artistic character, has become too
extensive to be fitly treated here. Even the story of their
removal, by Raoul-Rochette, is an instructive illustration of
archaeological conditions during the first half of this century,
and of the remarkable career of Reshid Pacha, the powerful
and cunning minister of Sultan Mahmoud II.

* The published illustrations of the sculptures from the temple of Assos, now
in the Louvre, have been referred to on page 11, note i; they have been
reproduced by wood-cuts in several histories of Greek art. In a future vol-
ume the blocks in Paris will be shown in new drawings from the originals,
uniform in size and treatment with the drawings made from the newly discovered
relief.

I06 ARCH^OLOGICAL INSTITUTE.

The original position of the sculptured epistyle has been
determined during the past year to have been upon the perip-
teros, and the new blocks go far to disprove the assumed
explanation of the representation upon the principal relief in
Paris as the combat of Proteus with Menelaus (Texier), or
with Aristaeus (Clarac), rather confirming the interpretation
of. it as a scene from the myth of Heracles, either that hero's
struggle with Triton (Friedrichs, Overbeck), with Nereus
(Brunn), or with the marine monster to which the daughter
of the Trojan king Laomedon was exposed, — the last iden-
tification being suggested by the writer, because of its local
character.

The reliefs were indeed published by Texier as carved
upon the epistyle of the peripteros ; but the only argument
advanced in support of this fact, — namely, that no fragment
of an unsculptured epistyle was to be met with among the
ruins, — has been proved to be a misstatement,^ and it was
more natural from aesthetical considerations, and the analo-
gies of other temples, to assign the zophoros to the cella wall.
That this, however, was not its position has been made evident
by the additional thickness given to the top of these lintels,
and by the swallow-tailed dowel-holes upon the ends of their
soffits, where the metallic clamps fastened the stones to the
upper surfaces of the projecting abaci. In the case of all the
Assos sculptures in the Louvre, these indications have been
effaced by the sawing of the stones to thin slabs, that they
might the more readily be attached to the wall of the
Museum.

The statement in the Description de VAsie Mineure^ that
the thickness of the reliefs was uniformly equal to one-half
the lower diameter of the peripteral shafts, must be regarded

* Compare the writer's Notes on Greek Shores^ p. 1 53.

111

aa

V/

iliii^

Mm

INVESTIGATIONS AT ASSOS, 1881, 107

^liberately false ; it was thought necessary to prove the
'tion of the blocks upon the columns, and the tendency
Sexier to distort facts in favor of his theories has been
^dy pointed out.

*rhe identification of Heracles in the enigmatical struggle
ith the marine monster appears more than probable from
be combat of that hero, which is represented upon the most
interesting relief found during the past year (Plate 15). The
Scene illustrated is that episode from the legend of the Doric
hero, which, not figuring as one of his greater labors, is usu-
ally connected with his expedition in pursuit of the Eryman-
thian boar.^ On that journey Heracles came to the cave of
the centaur Pholos, who offered friendly hospitality, — placing
roast meat before his guest, while he himself was contented
with raw. With this meal the hero required wine ; and as
Pholos, the son of Silenos, had received a jar directly from
Dionysos, with instructions that it should be kept until the
advent of Heracles, this was forthwith opened. The mountain-
roaming centaurs of the neighborhood, who in Greek legends
always appear untamably maddened by wine, perceived the
broaching of the attractive liquor from its odor, and rushed
upon Heracles, armed with clubs and stones. The hero drove
away, with firebrands, those who came nearest, and continued
the contest in the forest with his bow. The struggle was ren-
dered difficult by Nephele (the cloud), the mother of the cen-
taurs, who poured down torrents of rain, so that Heracles could
hardly stand upright upon the slippery earth, while his four-
legged opponents were not thereby discomforted. The bravest
of the centaurs were at last killed, and the rest pursued to
Cape Malea.

This identification of the relief is supported by the analogy

* The story of Heracles and Pholos is told at length by Apollodorus, Diodorus,
mud Tsetzes. It is mentioned by many other ancient writers.

I08 ARCHAEOLOGICAL INSTITUTE.

of similar representations upon vases.^ It is interesting to
note that this same scene was one of the subjects of the re-
liefs upon the throne of Amyclae,^ and apparently one of those
also upon the chest of Kypselos at Olympia,* — these two
most celebrated examples of ancient decorative sculpture thus
giving evidence of the extent and popularity of the legend
of Heracles and Pholos at an early period of Greek artistic
activity.

The appearance of the bowman Heracles without lionVskin
and club is not uncommon in archaic representations. Upon
our relief the stooping position of the carefully running archer
may perhaps be referred to the slippery ground, which plays so
important a part in the legend. The front foot of Heracles
is flat upon the ground, which is not the case with cither of
the centaurs.

The fragmentary figure at the extreme left of the compo-
sition probably represents lolaos, the relative and constant
companion of Heracles. lolaos is indeed not mentioned by
ancient writers as taking an independent part in the hunt of
the Erymanthian boar or the combat with the centaurs ; but
in all ages of Greek story he appears so inseparably connected
with Heracles that the proverb irpo^ hvo ovB' 6 'HpaKXryi^ sug-
gested the support of lolaos, who was even directly venerated
as his irapao'Tdrr)^!^ Upon five of the seventeen known an-
tique illustrations of this centauromachia, lolaos appears at

1 The representations of Heracles in the cave of Pholos and in combat with the
centaurs have been collected in the Compte Rendu de la Commissim fmpiriaU
d" Archiologie de St. Peter sbourg, 1873, ^Y Stephani, in a most learned paper
entitled Erkldrung einiger VasengemcUde der kaiserlichcn Ermitage, The most
interesting of the seventeen examples which illustrate the combat outside the
cave is a sarcophagus in Rome, published in Mon. pubbl. dalP Inst. Arch. l855,pl.
19; fourteen are vase paintings.

^ Pausanias, iii. iS.

• Pausanias, v. 19.

♦ Plato, Phaedon, 89 : " Heracles is not a match for two."

* Plutarch on Brotherly Love.

INVESTIGATIONS AT ASSOS, 1881. 109

the side of Heracles, — this being notably the case upon the
sarcophagus in Rome, referred to in a preceding note.

The legend of the hospitable reception in the cave of the
centaur particularly relates that Pholos and Heracles were
drinking the tempting wine from great cups^ when interrupted ;
and as an indication of the original cause of the combat lolaos
holds in one hand a drinking-vessel, raising the other as in
encouragement. The appearance of this figure seems almost
a reminiscence of the ever-present followers of the victorious
Assyrian monarchs upon Oriental reliefs.

Judging from the width of the inter-columniations, and the
position of the middle regula, this epistyle block is, upon the
upper surface, preserved in its entire length. Were it not for
this, and for the fact that the sculptured representations were
limited to the fields of the separate lintel blocks, it might be
more natural to assume that the fragmentary figure was that of
the centaur Pholos, — to whom the attribute of the drinking-
vessel would more directly appertain. But it is hardly possi-
ble that the body of a horse could have found room upon the
left of the epistyle relief. The appearance of two upright
human figures is certamly better in artistic effect than if
Heracles had been wedged in between the greater bulk of
centaurs upon either side, without reference to a symmet-
rical composition of the opponents.

Heracles bends his bow against three centaurs, who hasten
away with brutish gestures of fear, throwing their arms wildly
into the air and running so closely together that each over-
steps the hind legs of the one before him. The foremost bears
upon his shoulder a rude club, similar to those observable

* Mr. Sidney Colvin, in his excellent paper on " Representations of Centaurs
in Greek Vase Painting," Journal of HtUenu S'ttdUsy i88o, remarks that the earliest
literary allusion to the story, a fragment of Sicsichoros, in Athenaeus (xi. c. 99),
describes the huge cup, handed to Heracles by Pholos, as (rxi/^etop . . . hi-KM
HnfUTpoif Cft TpiXdywoy,

no ARCHAEOLOGICAL INSTITUTE.

upon the reliefs of three centaurs in Paris ; the one nearest
the hero seems to hold a stone in his uplifted right hand, —
these being the weapons with which, according to the legend,
the centaurs carried on the combat.

The great peculiarity of these centaurs is that they are rep-
resented in that highly archaic combination of man and beast,
in which an entire and perfect human being is joined to the
body of a horse. It is this form of a centaur which is described
by Pausanias in his account of the chest of Kypselos, which has
been before referred to. The appearance of centaurs with hu-
man fore-legs is sufficiently common, especially in the case
of Chiron, upon early painted vases and engraved gems. But
the only examples in which it has hitherto been observed in
sculptures are two small archaic bronzes, — a figurine found
by Ross near the Parthenon, and a relief recently unearthed
at Olympia.^ So fine an example of its occurrence in mon-
umental stone-carving as the present relief is wholly un-
paralleled.

All the centaurs upon the blocks of the epistyle and upon
one of the metopes, taken from Assos to Paris, show the im-
proved form of the beast, with the four legs of a horse. This
remarkable occurrence of both species upon the same build-
ing is probably due to the execution of the reliefs by different
sculptors, whose artistic conceptions were as various as the
degree of technical ability displayed in their work.^ The

* Archaologiscke Aufsdtze von Ludwig Ross. Erste Sammlung. Leipzig,
185$, Taf. vi. The author speaks of the little figure as " half a span long,"
and refers to similar discoveries in Etruscan graves. See Mon, ined. ii. plate
29. A woodcut of the figurine from Athens is given by Perrj' in his Greek
and Roman Sadpture. London, 1882, p. 102. Compare also Archceologiscke
Zeitttvg, 1881, Drities Heft, where O. Puchstein illustrates, plate xii., a vase
from Cyrene, on which are represented in the same scene centaurs of b<fth
kinds.

2 Die Ausgrabungen zu Olympia. Band Hi. Ubersickt der Arbeiten und
Funde^ vom Winter und Friijahr^ 1877, 1878. Herausgegeben^ von E. Curtius,

platb 16. Txa^neni of

INVESTIGA TIONS A T ASSOS, 1881. 1 1 1

decorations of the temple of Assos were evidently not under
a masterly superintendence, like that which assured so perfect
unity to the sculptures of the Parthenon.

The figure of Heracles (Plate 15) is so similar in certain
anatomical peculiarities to the hero represented as struggling
with the sea-monster, upon the most important relief in Paris,
that the blocks may reasonably be assumed as the work of
one hand.

Of all the sculptures of Assos discovered by the present
expedition and in the Louvre, the magnificent sphinxes
(Plate 16) are by far the best preserved, they alone having
been taken from a hard bed of mortar, which had long saved
them from weathering. The carving of this relief is of a
delicacy and vigor comparable to the best works of fully
developed Greek art. Throughout the body the firm muscles
and yielding cushions of flesh are indicated with an apprecia-
tion of natural forms which shows a distinct advance beyond
the art of Mesopotamia, successful as were its representations
of animals, while the decorative character of the composition
is maintained by the admirable outline of paws, wings, and tail.
The heads are of that archaic type familiar in Attic sculptures
dating near the beginning of the fifth century b. c. The eye,
though shown nearly in profile, is still too large ; the corners
of the mouth drawn up to a meaningless smile.

The Egyptian derivation of the sphinx is more evident than
is elsewhere the case upon Greek works, by the closely fitting
head-dress, welted upon the forehead and falling stiffly behind
the ears. The origin of the sphinx, which appears so often
in the early legends and art ^ of Hellenic lands, is a vexed

F. Adler, und G. Trcu. Berlin, 1879. See also Curtius, Das archaische Bronu'
relief aus Olympia. Berlin, 1880.

1 The appearance of decorative sphinxes may be in the following publica-
tions : IngWrami, Vasi fittili^ pi. 279, 308; Micali, Storia^ pi. 4, 7, 11, 16, 17,
28, 29 : Mus, Borb. vol. xiii. pi. 57 ; Mon. pub. dalV Inst. Arch.^ vol. ii. pi. 18, and
many others.

112 ARCH^OLOGICAL INSTITUTE,

question upon which the sculptures of Assos may throw
important light. Following the most thorough writer upon
the subject, Jaep,^ the belief has hitherto been common that
the sphinx was an independent creation of the Greek myth ;
still the Egypto-Phoenician character of the settlement of
Boeotian Thebes by Cadmus, and the first appearance of the
monster at that place, seems too plain an indication to be
easily explained away. The present relief certainly disproves
the assumption of Voss ^ that the Greek sphinx, like the Egyp-
tian, originally had no wings, — not receiving them until the
age of the great Attic dramatists, — which theory had already
been made extremely improbable by Gerhard.^ But the in-
fluence of Mesopotamia is known to have had a most direct
bearing upon the artistic concef)tions and methods of the
Asiatic Greeks, and winged combinations of human heads
and animal bodies are common in the decorative sculpture
of Assyria.

The dimensions of this relief, the architectural symmetry of
the composition, arid the existence of a similar relief for the rear
of the temple, prove it to have decorated the lintel above the
central inter-columniation of the front. The couching griffin,
or sphinx, appears from the reverse of all the earlier coins of
Assos to have been the emblem of the city. The representa-
tion of these animals above the entrance and upon both fronts
of the chief fane of Assos, in exactly the same conventional
attitude as upon the coins, and in a duplication which is the
fundamental principle of the coat-of-arms, makes the assump-
tion of its heraldic significance more than probable. Curtius*

^ Die griechische Sphinx, eine Mythologische Abhandliing. Von Dr. G. Jaep.
Gottingen, 1854.

^ Voss, Mytholo^fische Briefe, vol. ii.

• In the Abhaftdiunf^en dcr konigl. Akademie des Wissenschafien zu Berlin. 1 839.

* Ueber IVappengebrauch und Wappenstil im gricchischen Altertkum^ von
E. Curtius. Abhandlungcft der konigl. Akademie der Wissenschaften zu Berlin^
1874. Among the illustrations of this interesting paper there are several

• .

INVESTIGATIONS AT ASSOS, 1881. 113

has pointed to the emblematic character of the two crows of
Crannon, the winged sow of Clazomenae, the double head of
Tenedos, the goat of Elyros, and other types. The two axes
carved above the portals of Mylasa appear also upon the coins
of that town. The most striking example of such a civic coat-
of-arms is presented by the well-known relief above the Gate of
the Lions at Mycenae, — the most ancient monumental sculpt-
ure of Europe, — the design of which is comparable in every
respect to the sphinxes carved upon the temple of Assos.
The heraldic animals at Mycenae, like those at Assos, are
separated by an upright shaft, by which firm division a certain
decorative character is obtained, not held to be sufficiently
evident from the symmetrical repetition of the lions and
sphinxes, and the conventionalized treatment of their atti-
tudes and bodily forms. The erect position of the lions was
determined by the triangular tympanon for which they were
composed, even as the long, low extent of the sphinxes resulted
from the proportion of the epistyle block : in principle the two
reliefs are entirely similar. It is above gates and entrances
that such figures are particularly in place, as the custom of
the Middle Ages bears witness. The employment of emblems
was general and varied in Greek antiquity. Of all the animals
chosen, with the exception perhaps of the constantly occurring
lion, none was better adapted for a municipal symbol than the
sphinx, — a mysterious creature of supernatural force, wisdom,
and ever-blooming youth.

The relief of the lion and boar (Plate 17), when compared
with the Heracles, shows the far greater ability to deal with ani-
mals than with human forms, which is peculiar to the sculpt-
ures of Mesopotamia, and to that early artistic activity of
the Asiatic coasts which stood in close relationship to it. It

examples of this duplication of animal types to form the coat-of-arms of a city, as
in the coins of Marion, Delphi, Argos, and of some town in Lycia.

8

114 ARCH^OLOGICAL INSTITUTE.

is of course not supposed that these two reliefs were the work
of one hand ; the marked superiority of the hunting-scenes
is evident throughout all the sculptures of the temple, and
does not need to be argued from any single example. While
the human figures of Plate 15, of the metopes Plates 21 and 22,
and of the Heracles and banqueters in the Louvre, are so
helplessly designed and executed as to compare most unfavor*
ably with the nearly contemporary gable groups of ^gina, —
the sphinx (Plate 16), the boar (Plate 17), and above all the
hindquarters of the lion (Plate 18) are, on the other hand,
works of admirable art.

The legs and tail of the boar (Plate 17) are characterized
with great truth. Though seized by the lion, the animal has
not lifted his head from rooting, the attack in the rear not
seeming to cause him much disturbance. The hind-legs are
set to withstand the push of the burrowing snout ; the tail
hangs limply upon the broad flank, as if in indication of
hoggish enjoyment. Along the back runs a fin of bristles,
terminating sharply between the ears. It is unfortunate that
the head and fore-legs have been much injured ; in some meas-
ure the details of this part of the body may be determined by a
comparison with the lank boar of the Louvre metope, although
that relief is, throughout, in a state of preservation hardly
superior to the most battered parts of the present block.

Though thus familiar with the appearance and action of
wild boars, which have always abounded in the neighbor-
hood of Assos,^ it is evident, from certain errors of form,
that the sculptor had never seen a lion. The latter ani-
mal is less well carved in detail, while an Oriental stiffness

* Enormous boars from the Mysian mountains devastated the land during an-
tiquity. See Herodotus, i. 36. The animals are numerous to-day, especially upon
the heights of Ida; the villagers, whose fields arc upn)oted by the beasts, arc
glad to beat the bush for hunters armed with effective weapons.

INVESTIGA TIONS A T ASSOS, 1881. 1 1 5

makes his action appear listless and unconcerned. The head
is turned upon the side in such a manner as to show it directly
from above ; the ears being almost equidistant from the out-
line of the neck. The long tail hangs nearly straight, the ex-
tremity being so turned as to make the tassel stand upright.
Throughout this relief the sculptor has displayed a certain
humor, which makes up for the ungraceful carving and the
ignorance of leonine peculiarities.

The magnificent figure of a lion (Plate 18) is of an entirely
different character, and is evidently the work of another hand.
The beauty of outline, force, and delicacy of the muscles ; the
action of the limbs and swing of the tail, — in short, every de-
tail of this block displays a fine mastery of animal form and
action. In point of technical execution, the sculpture of this
relief is hardly inferior to the masterpieces of the Theseion
and Erechtheion.

It is particularly to be regretted that the situation of the
block, after the overthrow of the temple, exposed it to the
action of standing water and frosts ; the surface is not cor-
roded, but the raised parts of the stone have, in several places,
been split from the background. The upper part of the left
leg and the end of the tail have thus been entirely lost, while
the flank is in fragments. The lion bears upon his back the
legs of a deer.

The extent to which the individual characterization of the
various animals was indicated in the details is evident from a
comparison of the tails of the heraldic sphinxes, the boar, and
this lion.

The fragment of a sphinx (Plate 19), discovered upon the sur-
face, has already been spoken of as possibly having been seen
by Texier. It fits accurately upon the epistyle relief of a sphinx
now in the Louvre. In workmanship it is greatly inferior to
the corresponding carving of the eastern front, and, like all

Il6 ARCHAEOLOGICAL INSTITUTE.

the sculptures which were not covered with earth, it is badly
weathered. The disproportionately long fore-legs resulted
from the distance left between the breasts of the animals, and
determined a low angle for the uplifted paw against the central
support, which, with the gap between the opposite heads, was
fatal to the decorative character of the composition. The dif-
ferent termination of the separating shaft is no improvement
upon that of the eastern front.

The fragment of an epistyle relief (Plate 20) is of interest, as
showing the occurrence of centaurs with horses lore-legs side
by side with the more archaic form referred to above. Upon
the right of this block, as upon Plate 15, a narrow fillet is
noticeable, the purpose of which was to separate the sculpt-
ures upon adjoining lintels, — often designed upon a different
scale, and of a different character.

From these fillets, as well as from the triple epistyle con-
struction, it is evident that the reliefs were carved upon the
ground during the erection of the building and before the
placing of the lintels. This was without doubt also the case
with the metopes, which were slid in behind the projecting
edges of the triglyphs. This accounts for the independent
and disconnected appearance of reliefs set in. juxtaposition
after their execution, as well as for the number of different
hands employed upon the stone-cutting. The sculptors were
limited to a much shorter time than if the decorations had
been carved after the completion of the edifice.

In the restoration of the temple front (Plate 14) the two
longest reliefs in Paris occupy the corners, to which they are
assigned on account of the exceptionally large spaces between
their regulas. The position of the recumbent sphinxes, above
the central inter-columniation, has been determined with greater
certainty. The relief illustrating the combat of Heracles and
the centaurs thus appears to have been above the second inter-

Fto^inerLi of Temple I^Te-lc^e.-Aoeca

Tem-ple >l^ope .J^^^sam

INVESTIGATIONS AT ASSOS, IS81.

117

columniation. and has consequently been placed there. As
no other block known to have belonged to this front has been
found, this last scene has been reversed to fill out the epistyle.

During the excavations a number of plain metopes were met
with, and as only five sculptured metopes are known, — Arae
being in Paris, and two having been found by the present
Expedition, — it is probable that those upon the sides of the
temple were not ornamented with carvings.

The one complete metope relief (Plate 21) represents a man
pursuing a woman, — a time-honored subject, difficult to indi-
vidualize, which may. perhaps, from the analogy of the other
sculptures, be referred to the myth of Heracles. The female
figure is crowded helplessly into the side of the field ; her
arm must have touched the edge of the triglyph, which pro- I
jected, from constructive reasons already considered, about
0.01 5 metre over the metope.

It is possible that the decoration was thrown thus out of
centre, by a diminution on one side of the width of the slab,
after the carving. The stones, sculptured in the workshops,
probably did not always fit exactly into [he interstices left
between the iriglyphs. In the planning of a frieze of many
blocks, the allowance for the joints is usually under-calculated,
especially in so coarse a material.

The fragmentary metope (Plate 22) shows two warriors in
combat, — the one of which the body has been prcseri'ed
drawing a short sword from a sheath held in his k-fl hand.
His loins are girded with a cloth, — this being the only indica-
tion of drapery upon any of the reliefs, with the exception of
that of the female figures upon the relief in Paris, the struggle
of the hero and sea-monster.

The primitive clumsiness of these metopes, when compared
with the representations of the lion (Plate 18) and the sphinxes
(Plate 16), instructively illustrates the very different degrees of

Il8 ARCHAEOLOGICAL INSTITUTE,

ability and artistic advance displayed by the stone-cutters em-
ployed contemporaneously upon the decoration of the building.
This excellence of the best of them is a weighty argument
that the construction of the temple does not date from an
epoch more remote than the termination of the Persian war.

The archaic character of the reliefs is due to local provin-
cialism, as well as to the antiquity of the work : in deter-
mining the proportion in which these two influences are to
be estimated by the historian of art, the appearance of so
great an individual variation gives great weight to the former.
With the sculptures, as with the architecture, it is evident
that the most advanced characteristics must be held as the
true indication of the age of the monument, rather than the
traits that exhibit primitive conceptions and technical in-
ability. The high perfection exhibited by Plates i6 and i8
by no means points to a greater antiquity than, for instance,
that of the sculptures from the gables of the temple of
./Egina, with which the semi-barbarous decorations of Assos
may be supposed to be contemporary. The sculptures of the
temple of iEgina, like its architectural peculiarities, record
an independent advance beyond the most immediately pre-
ceding works. Such an advance is not to be expected on
the northern coast of Asia Minor during this historical pe-
riod, and could not have been instantaneously shared by a
land so recently freed from the long occupation of Oriental
barbarians.

The metopes from the temple of Assos, here presented, are
certainly far more inferior to the sphinxes, lions, and boar
of the epistyle than are these latter to sculptures of European
Hellas, referred to the third decade of the fifth century b. c. ;
and yet the greatest variation possibly in contemporary sculpt-
ure cannot be assumed to be displayed in these decorative
reliefs.

INVESTIGATIONS AT ASSOS, 1881. 1 19

Though previous writers have been able to judge of the date
of the sculptures of Assos only from the blocks in the Louvre,
without knowledge of the architectural arguments derived from
the plan and elevation of the temple now given for the first time,
it is proper to call attention to the fact, that, with the exception
of Clarac, the most eminent authorities on the history of Greek
art by no means share the views here advanced, but assign to
these works a considerably greater age. Dr. Brunn, whose un-
equalled knowledge of the style and artistic relations of antique
sculptures gives his opinion the greatest weight, has not found
the discoveries here published to alter his former belief, that
the construction of the temple of Assos was previous to the
sixty-fourth olympiad, b. c. 524. Mr. Sidney Colvin, in the
essay before mentioned, remarks that it is impossible to date
the reliefs later than the sixth century. Of the older writers,
Fricdrichs even attempts to prove, from the absence of the
lion's skin as an attribute of Heracles, that the sculptures
were carved before the end of the seventh century B.C. ; while
Overbeck, on the other hand, thinks it doubtful if they are
older than the sixtieth olympiad, b. c. 540.

The temple reliefs of Assos may be considered as the most
important link in the chain connecting the carvings of the
early civilizations of the East and the unequalled sculptures of
Greece. It is only by defining the position of such works
that the application of the historic method to the study of
intellectual and artistic growth can be of value. Archaeo-
logical investigations can in no wise give a more direct and
practical assistance to the architecture and sculpture of to-day
than by indicating the path followed by the early Greek artists
in their progress toward supreme excellence.

The Oriental and transitional character of the reliefs, evi-
dent from the pre-eminence of the animal forms, is even more
apparent in the reminiscences of the cmpaistic work of Phoe-

I20 ARCHAEOLOGICAL INSTITUTE.

nicia, that great mediating power between the sculpture of
Mesopotamia and the primitive attempts of Hellenic art
The Homeric epics constantly point to the Syrian coast as
the home of skill in sculpture and metal work ; and it is not
surprising that the Greeks of Asia Minor, being immediately
exposed to this influence, should retain traces of the art of the
hammer even as late as the end of the Persian occupation.

The proceeding of the Phoenician artisan was to make a
model of wood for the relief, or sculpture in the full round, as
the case might be, upon which sheets of metal were secured,
and finally beaten to the shape of the carving beneath. This
method of work was long practised, and, its products being ex-
ported in all directions, was of the most widespread influence.
It is natural that the peculiar forms resulting from the techni-
cal properties of beaten sheet-metal should determine a style
which is recognizable even in stone carvings, when these were
the creations of sculptors familiar with works of this kind. All
the prehistoric monuments of Greece bear traces of this influ-
ence ; and it appears in the archaic and provincial reliefs of
Assos, recent as these are when compared with the treasures
and tholos of Mycenae. It is most noticeable in those sculpt-
ures which are least developed in artistic respects ; the sphinxes
and the hindquarters of a lion betraying no traces of it, while
the characteristic metallic forms are strikingly evident in the
struggle with the sea-monster, the banquet, in the metopes
found during the past year, and in the lion's head from the
corner gutter.

The figures upon these last reliefs offer, in general form
as well as in detail, analogies to the primitive vase paintings
of Phoenicia.^ This empaistic character of the sculptures of

* Compare Raoul Rochctte, in the Journal da Savans^ Avril, 1835; De Witte,
Catalogue Durand^ Introduction, and other passages relative to the question;
Ch. Lenormant, Cours d^Hiitoire Ancienru: Introduction h PHistoire de CAsie
Occidentalet etc.

INVESTIGATIONS AT ASSOS, tSSl.

Assos explains the striking similarity noticeable between
them and the most ancient bronze works of Etruria, — espe-
cially the important reliefs from a chariot found at Perugia
now preserved in the Glyptothek of Munich, and the figures
from Cervetri, published by Grifi.

Not only the detailed forms of the decoration of the temple
of Assos, but its position upon the building, point to the pro-
totype of a work of hammered metal, and in this respect it
appears of direct importance to the history of the early archi-
tecture as well as the sculpture of the Greeks. The reliefs
upon the epistyle, the principal constructive member of ihe
entablature, warrant the conjecture that the timbering of an-
cient Asiatic fanes was overlaid with sheets of metal, as is
known to have been frequently the case with the columns and
walls>

The wooden roof and ceiling of the original Hellenic cella I
appear in the temple of Assos already translated to the un-
varying stone forms of the Doric frieze and cornice, with the
exception alone of the trunnels, which seem not to have been
regarded as of canonical importance. The great peculiarity of
the entablature, — namely Ihe decoration of the epistyle, a func-
tional lintel never sculptured in the perfected Greek styles, —
appears to be a provincial imitation of the empaistic overlaying
customary in the architecture of neighboring lands.

The importance of so remarkable a monument to the early
history of I^Iellenic art is evident.

It is not the purpose of this first Report to treat in detail
of the city walls of Assos or the monuments of the lower
town. Much, indeed, has been ascertained to which no refer-
ence can be at present made ; for even were the full pub-

» A refcienee to this empaistic character of the reliefs of Asso* i< made bj
Semper in Dtr Stil. etc. Ziveilc Audage, MUnchcn, ig7& Bd. i. p. 406.

122 ARCHAEOLOGICAL INSTITUTE,

lication of the partial results already obtained considered
desirable, there does not now remain time for the comparative
studies to which it would lead. The season is approaching
when the trenches are freed from the frosts of winter, and
the active work of the second season is at once to be begun.
Still, in order to indicate the scope of the investigations, a
number of illustrations are given which require a brief ex-
planation.

In descending from the upper step of the Acropolis, re-
mains of Hellenic fortifications are met at the northeastern
extremity of the lower level. The enclosures at this point rise
to a height of one metre above the present surface of the
ground, being of a heavy masonry of equal courses, apparently
of about the same character and date as the extensive city
walls.

These ramparts must have been overthrown at a compara-
tively early age, for they appear as the foundations of a square
tower of good mediaeval masonry (Plate 23), which has been
filled by the kitchen ddbris and ashes of successive occupants
to a height of not less than eight metres above its floor. The
door of this structure seems to have opened upon the platform
of the ramparts which enclosed the lower Acropolis ; and as
this wall has been demolished, there is now no accessible
entrance. The corner of the tower has been broken into by
the Turks, at which point the stratified debris is exposed.

The Byzantine church, now serving as a mosque, is sepa-
rated from the tower by a narrow passage, and, as may be
seen from the plan (Plate 2), stood outside the fortifications.
This is the building — ^"un ancien temple de forme elegante,
moiti^ carr6, moiti^ conique" — which appeared so remarkable
to Poujoulat, who maintained that " la religion musulmane
nous a ainsi conserve dans son int6grit6 premiere un monu-
ment appartenant aux beaux dges de la Gr^ce."

INVESTIGATIONS AT AS SOS, 188L 123

Such an amusing conception is not necessary to make the
church of interest to the investigator ; its importance as
being built of stones from the wall and ceiling of the Doric
temple has already been indicated. The site of the building
has been planed from the top of a prominent cliff, the columns
of the vestibule standing directly upon the native rock. The
greater part of the edifice is Byzantine, its age being perhaps
determined by the inscription upon the lintel of the door
(Plate 24). The vaulting of the dome, which appears upon the
exterior as an octagon, may be that of the original construc-
tion, although the pendentives are Turkish stalactites, dating
Irom an extensive restoration, which greatly altered the ex-
ternal appearance. The narthex must have been almost
entirely rebuilt at this time, its graceful arches being of the
pointed form peculiar to early Ottoman architecture. The
building is situated so directly above the village that the
minaret which the Turks elsewhere added to Christian
churches was not necessary. The bare interior was at first
zealously guarded from the visits of giaours, but during the
second season there will probably be no difficulty in making
the necessary measurements for detailed plans and sections.

The publication of the monuments which appear in the
topographical sketch (Plate 3) is wholly reserved for the next
Report. Before the stoa plateau, directly above the theatre,
there extended a second edifice, provided like the upper colon-
nade with reservoirs. No excavations whatever have been
made in the deep earth at this point, and the general arrange-
ment of the complex of buildings must be accurately deter-
mined before any consideration of details can be of value.
The elaborate descriptions of the theatre given by Hunt and
Prokesch-von-Osten, taken in connection with the points de-
termined by the trial pits sunk here during the first season,
establish the arrangement of auditory and scene, at least in

124 ARCHAEOLOGICAL INSTITUTE.

plan ; but the connection of the theatre with the terraces
above it is uncertain, and all these structures of the central
town, which seem to date from a contemporary rebuilding,
are too closely related to admit of their being separately
deecribed.

The excavations at the gymnasium have not led to results
which could as yet justify a thorough consideration of this
edifice (Plate. 4), which, in its frequent antique restorations
and involved original disposition, presents many unsolved
problems. A detail of the extensive mosaic discovered in the
basilica hall is given in Plate 25. Formed of various-colored
marble cubes, of careful workmanship and interesting design,
this pavement must have covered a space not less than three
hundred square metres. The border, of which every division
presents a different pattern, has remained intact in the greater
part of its length exposed by the trenches, the centre having
unfortunately been almost entirely broken away. Remains
of another mosaic were found in the enclosure, and it is
hoped that the continued examination of this site may lead
to interesting discoveries.

It has already been stated that one of the chief tasks of the
second season will be the thorough study of the fortifications,
of Assos. The importance of these unrivalled monuments of
Greek military engineering is so great that were their publi-
cation to be the only result of the Expedition, the undertaking
would be amply repaid. Not only are the planning and con-
struction of the ramparts, towers, portals, and posterns of in-
terest in each case, but traces of successive enclosures, dating
from different ages, illustrate the growth of the city in extent
and power, giving information such as is afforded by no other
remains of antiquity.

The most recent Hellenic fortifications, which alone have
been known from the Description de VAsie Mineure, notwith-

INVESTIGA TICNS A T ASSOS, 1881. 1 2 5

standing the lamentable injury lately done to them by the
Turks, are still in a wonderfully perfect state of preservation.
The remark of Texier appears hardly an exaggeration ; in
places the walls -' seem rather a commenced and unfinished
work than a ruin." Throughout their entire extent, — a length
of over three kilometres, — these fortifications are built with
unvarying care, being skilfully so planned as especially to pro-
tect the points by nature most exposed to the attack of a besieg-
ing enemy. The greater part of the circuit can be traced ; it is
only at the north of the Acropolis, near the precipitous descent
from the village to the river valley that the position of the wall
is uncertain.

The rectangular blocks, exactly jointed, are laid without mor-
tar in horizontal courses of equal height, bonded from face to
face by headers. This regular masonry is at times built upon
and on the face of the polygonal walls of an older period, as is
shown by Plate 26, which represents a breach at the extreme
west. The principle of the vault is employed in one of the
towers, but not in any of the gate-openings where circular and
pointed blind-arches are cut from the horizontal courses, — as
at Ephesus, Thoricos, Messene, etc., — or where the edges of
the projecting stones form an oblique transition to a compara-
tively short lintel, as in the portal, Plate 27. This opening,
marked A upon the topographical plan (Plate i) is in the
transverse division wall, which runs from the Acropolis cliff to
a re-entering angle of the outer fortifications. The northern
and southern enclosures of the city were connected only by
this narrow passage, in the jambs of which the bolt and pivot
holes of the heavy doors are visible.

The chief gateway of the northwest upon the ancient road
leading to Lecton and Alexandria Troas is flanked by enormous
towers, one of which is shown in Plate 28. The view is taken
towards the Acropolis, the northwestern corner of which, show-

126

ARCH^OLOCICAL INSTITUTE.

ing the height of the lower step, appears in the distance From
the clift' descends the before-mentioned transverse wall with
the portal, Plate 27.

Outside of the fortifications are seen the vestiges of the sar-
cophagi and sepulchres which bordered the street of Tombs
The plan of this extensive cemetery appears on a small seal*
upon the map of the city ; its section, looking to the north, \i
given by Plate 29. All the antique structures upon this sketch'
restoration have not been determined by the limited excava-
tions undertaken at the site, but the general arrangement 0:
the terraces is accurately indicated.

The funeral monuments were placed upon the edges of thre<
levels, which, rising above the principal road, extended to th(
foot of the city enclosure. The broad passages left free fron
sarcophagi must have served as a promenade and place o
assemblage for the inhabitants of the crowded town ; thi

I

Plate 32. Plan ani> Section op Receiving Tomb

INVESTIGATIONS AT ASSOS, 1881.

127

destination being shown by the attractive arrangement of broad
steps and public seats from which the magnificent sunset pano-
ama of the ve p a n and far st et h ng gulf could be enjoyed.
Two such exed a the one of sem c cula he o her of rect-
angula plan a e p csented n e eva on by P a e 30

Plate 33.

The tombs are of every variety of form and disposition, from
vaulted receiving sepulchres, like that shown in Plates 31 and
32, to free-standing sarcophagi, — one of the most interesting,
but by no means one of the best preserved, of which is illus-
trated in its present condition and original appearance by Plates
33 and 34. The carving upon this latter chest, although badly
weathered, is so characteristic in design, that if its shattered
sides are discovered during the future investigations, it will
well repay transportation, notwithstanding its great weight.

Farther from the gates of the city are mounds of debris,
which mark the situation of extensive monuments, so hope-
lessly overthrown that an understanding of their construction

128 ARCHAEOLOGICAL INSTITUTE.

was not possible without excavations, for which the first year
allowed no time.

One branch of the road which passes the street of Tombs
continues directly to the north, crossing the Satnioeis at a
point indicated upon the map of the city, Plate i. Here were
discovered considerable remains, which afford the only known
example of an ancient Greek bridge, Plate 35. The structure
is certainly the only existing instance of a work of this kind
in which the principle of the lintel, so tenaciously adhered to
previous to the age of the Diadochi, has been consistently
carried out. The fact that the Greeks seldom attempted the
execution of monumental works of engineering, such as were
so often undertaken by the Romans, made wooden bridges
much more common than those of stone, even in such impor-
tant positions as the passage between Aulis and Chalkis, where
a bridge connected the island of Euboea with the mainland.
Of these timbered constructions there remains, of course, not
a vestige. All the stone bridges occurring in Greek lands are
of vaulted form,^ and must be referred to the late epoch of
the Roman occupation, as in the instances of the triple pass-
age over the river Pamisos, between Andania, Megalopolis, and
Messene, and the single arch over the Eurotas, near Sparta.
The projecting horizontal courses of the foundations on the
road between Pylos and Methone may be of considerable age ;
but, as in every known example, the upper portion of this struct-
ure, built with wedged-shaped stones, dates from a mediaeval
restoration.

At Assos, on the other hand, the ruins show the bridge to
have maintained its original form unchanged as long as it was

^ Cell, liiuerary oj Greece^ ctc.^ London, iSio, mentions two examples of small
constructions above rills with a horizontal termination, at Phlios, and near
Mycenae, on the road to Nauplia ; but the former appears to have been a mere
opening in the fortifications of the town, and the latter is a formless mass of
small stones, the age of which is extremely doubtful. Neither can be spoken of
as a proper bridge.

Plate 35. Briixie ox the Satnioeis.

INVESTIGA TIONS A T ASSOS, 1881. 1 29

in use. Upon the southern bank of the stream, above the high-
water mark, the stone beams of the platform are still in posi-
tion. The piers are in plan of elongated diamond-shape,
and extend upon either side slightly beyond the bridgeway
to a length of 3.6 metres. The masonry of these supports
consists of large blocks, carefully jointed, and is particularly
remarkable for the system of combing by which the action of
the current is resisted. The detail of pier is given on Plate 35.
The joggles cut twice upon each course made it impossible to
displace any stone by lateral pressure without entirely over-
throwing the heavy pier, which presented a minimum width to
the stream.

Upon these admirable foundations was laid a platform of
stone lintels, in length about three metres from centre to cen-
tre of the piers. Four beams were placed side by side to
provide a passage amply broad for the needs of ancient traffic.
Wagons can never have been extensively employed in the
rugged Troad. The lintels were bonded together by swallow-
tailed dowels of wood, in the manner universal in the Hellenic
architecture of the fourth century b. c. Seventeen piers, thus
connected, are known to have extended from the southern bank
to the present summer bed of the river, where the last traces
were examined- Upon the northern bank are the remains of a
heavy abutment. The midsummer work of the second season
will determine whether the piers and horizontal stone beams
were continued across the deeper water, or a lighter-timbered
structure spanned the thirteen and a half metres remaining
between the abutment and the last foundation which could be
observed after the October rains.

It is certain that the course of the stream has not changed at
this point, which was by the nature of its banks particularly
well adapted for the site of a bridge. Only a short distance
MbovCf the sandy reach, overflowed by the stream, is several

9

130 ARCHMOLOGICAL INSTITUTE.

hundred metres broad ; while below, in the Halesion Plain, the
arches of a Roman bridge are so far from the present bed
that the water cannot be seen from the ruins.^

A peculiarity of the Assos bridge is that it did not cross the
river at right angles, but followed the general direction of
the road. The axes of the piers were, however, parallel to the
current ; and hence the lintels were in plan placed obliquely
upon their foundations.

The winter bed of the stream, near the bridge, appears to
have been entirely paved in antiquity, probably as an approach
to the summer bed, where the water is still drawn for the
use of the village of Behr^m during the dryest months, as
already described. The greatest disadvantage of the site of
Assos must always have been its lack of fountains ; and the
reservoirs and cisterns built for the- collection and distribution
of the rain-fall are of an importance not elsewhere accorded to
them by the Greeks.

In crossing the plateau again, the road, after passing the
southern limits of the city fortifications, descends so abruptly to
the port that the houses are seen almost directly from above.
The climb from the sea to the city enclosure is the steepest
and stoniest conceivable ; the break-neck position of Assos
was notorious even in antiquity among a people who found
nothing remarkable in the elevation of the Acrocorinthos or
the Acropolis of Segesta. Stratonicos, an Athenian musician
and poet, noted for his witty and sarcastic remarks, a number
of which are preserved by Athenaeus, applied to it the line of
the sixth book of the Iliad, —

" hwov W, C)t K€^ Baaaov okiBpov ndpaff (kt/oi, —

playing upon the adverb. Surprisingly little has been done to
ease the natural difficulty of the ascent. Here and there are
fragments of ancient paving-slabs and polygonal retaining-

1 Compare the mention of the altered position of the Touzla at this point in
Ml. Diller's geological appendix.

INVESTIGATIONS AT ASSOS, 1881. 131

walls ; but so much has been washed away from the declivity
that for the greater part of the way one is obliged to scramble
up the side of the natural rock. If the trodden path be deserted,
the climb has to be performed with hands as well as feet. ' Beasts
of burden take an easier, round-about way, indicated on the
map of the city.

The mighty blocks of the ancient mole are shown by Plate 36,
as seen beneath the clear water. Results of interest are hoped
from the further examination of this extensive work of engineer-
ing, as well as from the similar remains at the east Both
these dykes seem to have sunk bodily, as by an earthquake, to
a depth averaging about two metres below the surface of the
water. The preservation of the masonry is in places excellent,
and on calm days the posts cut upon the coping, to which the
ancient vessels were moored, can be distinctly recognized.

The Turkish mole, piled up of small stones and gravel, shel-
ters but about one-fifth of the original enclosure. The little
open boat at the extreme left is the Myzethra, by which all the
communication of the Expedition with the outer world has been
maintained. The large magazine on which the flag is hoisted
is our home ; in two rooms of its upper story are the simple
household effects, the surveying and measuring instruments,
the many books and drawings, with which in the present year
the work at Assos is to be carried on. May it result in fortu-
nate discoveries, as well as in the thorough investigation of
ruins already known, but not yet properly studied.

JOSEPH THACHER CLARKE.

APPENDIX.

I.

INSCRIPTIONS FOUND AT ASSOS IN 1881

I.

This inscription, engraved on a bronze plate (0.54 X 0.38 m.), of which
a facsimile is given in the plate opposite (Plate I.), contains a decree of
the town of Assos, passed on the accession of the Emperor Caligula in
37 A.D.

'EttI {nrdrcDV Tvalov *AK€ppa)Viov Upokkov Kat
Faiov HovTLOV UeTptavCov Hiypivov.

SP'T^^tcr/ia ^Aaauav yvdfiy tov Bijfiov.

"Ettci ^ Kar* cyp^^v ttSutw &yOpwfjroL^ iXirurOtiau Taiov
KoiO'apos T€pfiavucov %€fia(rTOV i^yc/xovux icai-i/vycXrcu,
ovSfy Sk /JL€TfX3v \apa9 cvpi7ic[c]K 6 Koa-fioqy ircura 8i 7r6\is
fcat Tray Wvm hri rrjy rov Btov oif/iv lairevKeVf o)? Sy rev

*TE8o^€V tq PovXq Kol Tots TTpayfuircvofLCi^cHf irap* ripAV
F<ofuuoL9 Kal Ty ^fup Tcp 'AcrcriW icaraora^Tvoi irpto'
fitiay he rtay irptjimay kou Apumav 'Puifuutav re ical "EAXi;-
vtay rrjv hrr€v(ofji€injv koX (rvvri(rOrja'Ofi€vrjv avrf
10 i€ri&rf<rofifvrjy re l;(Civ 8ia fivi^firf^ koX Jo/Sc/xovias

n/v V'oAii', KoBifi Kol avTOf fLcra rov Trarpof T€pfiavucov
hrifias TTpiOTU)^ rg iwapxtufL rrj^ ^/xcr^a? iroAcws

134 ARCHAEOLOGICAL INSTITUTE.

"Ofivvfiev Blol awrrjpa koI $€ov Kaurapa Sc^oorov koX rrfv
15 irarpujv dyvrjv irapOivov cvvoi/crciv Taua Kourapi '^tPaxr-
TO) Kot rep crvfiTram ouca> avrov, koI ^cXov9 re KpLvtiv
OU9 ^^ avTos irpoojLpriTaL kcI 1x0 pov^ 0U9 Av avros irpo/9o[A-]
Ai^au £topicoi)(riv /xev rjplv cv cii;, iff^iopKowrtv 8c ra cvay-
[tui].

TIpecfievTal iTrqvyeiXavTo Ik rS)v ihCtov

Faio^ Ovdptoi ToLov vio^f OvoXtivlo, KaoT09>
20 *Epfio<f}dvrj^ ZijjitKov,

Kttjto^ Hururrpdrov,
Al(r)(pitav K.aXXx<f>avoy^f
*ApT€/uit3<i)pos ^iXofiowroVf
otrtvc? *cai vn-lp r^ Taiov KatVapos ^cfiaarov Fcp/Liavucov
25 crarnypta? €v(dp.€V0L Atl KaTrtTwXwi) tBvcav t^ t^ ttoAc-
(JS ovofiaru

IN THE CONSULSHIP OF GNAEUS ACERRONIUS PROCLUS
AND GAIUS PONTIUS PETRONIUS NIGRINUS.

^ Decree of the Assians by Vote of the People.

Since the supremacy of Gaius Caesar Germanicus Augustus, for which
all men have hoped and prayed, has been proclaimed, and the world has
known no bounds to its delight, and every city and every nation is eager
to behold the face of the God as the greatest delight which the present
age can offer to mankind, —

Be it enacted by the Senate, and the Roman merchants established
among us, and the People of Assos, that an embassy be appointed from
the first and best Romans and Greeks to meet and congratulate him, and
to entreat him that he will hold our city in remembrance and under his

INVESTIGATIONS AT ASSOS, 1881. 135

protection, even as he himself promised when with his father Germanicus
he first entered upon the government of our city.

Oa/A of the Assians,

We swear by the Saviour and Deity Caesar Augustus, and by the pure
Virgin whom our fathers worshipped, that we will be faithful to Gaius
Caesar Augustus and all his house, and that we will consider those our
friends whom he shall prefer, and those our enemies whom he shall declare.
May it be well with us if we are true to our oaths, and may it be other-
wise if we are false to them.

These offered themselves as ambassadors at their own expense : —

Gaius Varius Castus, son of Gaius, of the tribe Voltinia.
Hermophanes, son of Zoilus,
Ctetus, son of Pisistratus,
Aeschrion, son of Calliphanes,
Artemidorus, son of Philomusus,

who also invoked Jupiter Capitolinus for the safety of Gaius Caesar
Augustus Germanicus and made sacrifice in the name of the city.

II.

This inscription, copied by Mr. W. C. Lawton, was found on two frag-
ments of stone, September 5 and 6, 1881, in Assos, at the eastern end of
the Stoa plateau, in a narrow passage which ran close to the edge of the
parapet and was probably tfie chief outlet thence to the lower town. It
contains a decree, passed by some town whose name is lost, giving a
crown and a vote of thanks to the town of Assos for sending judges or
referees to decide certain lawsuits, and giving the same distinctions to the
judges themselves. The upper part of the inscription, with most of the
preamble, is lost. Inscription No. 3568 /", in Boeckh, Corpus Inscript.
Graec. vol. ii. p. 1128, contains a similar vote of thanks sent by the town
of Peltae to Antandros : Boeckh assigns this document to the third cen-
tury before Christ. No facsimile of the present inscription, giving forms
of the letters by which its date could be determined, has been received.

136 ARCHAEOLOGICAL INSTITUTE.
MIA • • •

AHM02*AINHTAITA2KA

T0I2KAA0I2KAIArAe0I2T0NANAPQN

nAPAriNONTAIANAPE2A2;iOITOTAHMOTEIA.

5 nAPX0r2ANErXAPI2TIANAEA0XeAITHIB0TAH •

AHM0IEnHNH2eAIT0NAHM0NT0NA22I0NEniT-

• • • • SaXEXEinP02HMA2KAI2TE*AN0T2eAIArT0NENT0IS

• • • T0I2AI0Nr2I0I2ArAHTnNTHinPflTHIHMEPAIXPr20I2TB

• ANniEniTnAn0STEIAAIAIKA2TA2KAA0r2KArAe0T2KA •
10 • PAMMATEAEnHNHSeAIAEKAlT0T2AIKA2TA2T0r2nA •

PArEN0MEN0r2EXEAA0NAeHNAr0P0rAATIM0NKAE0M0P

• 0TKAI2TE*ANn2AIEKATEP0XArT0NXPT2ni2TE*AN0NEn •

• • • • ITA2MENAIAAIKA2AITnXAIKnNI220KAIAIKAI02TA2
T2AIAn0IIANT02T0rBEATI2T0rrnAPXEINAEATT0 •

15 MB0TAHNKAIT0NAHIM0NnP0T0I2METATAIE

PATHAPXEIXA • • • • • KAinP0SEN0r2TH2n0AE02HM0N2TB
*ANni2AIAE • AITOrrP • • • TEAMEAArXPONMEAArXPOreAAB
PfiI2TE*AXniEniTftnAPA2X • 2eAITHNKAeArT0NXPEIAXMETA
nA2H2*IA0TIMIA2TH2TEAXArr • AIA2T0N2TE*AN0NTHNEn.

20 • • • 2 • • n0H2A2eAIT0T2Ar0N0eET • 2T0TM0r2IK0TINAAEKA.
A22I0IEIAH21}2fXTHNTETONAXAP • • KAAOKAPAeiAXKAITHN
T0TAHM0TETXAPI2TIANAIPEenNAinPE2BErTA20ITINE2A*IK0

MEN0inP02ATT0T2EnE E22INTETHN • OTAHrKAITONAH

M0XT0TE*H*I2MAAn0An20T2INATT0I2KA • • • 0*ANI0r2INTH •

25 TETnXAXAPOXKAAOKArAeiAXKAITHXETXOIAX • NEXOMEN
nP02T0XAnM0XATTnXKAinAPAKAAE20T2IXA2210T2KAinA
PArT0I2n0H2A2eAITHXAXArrEAIAXTnX2 • E*A
NftXTn0T0rKATA2TAeH20MEX0rAr0X0eET0T • OT
M0r2IK0TArftX02nP0X0H2AIAEIXAKAIT0>I'H*I2MAAXA • • •

30 *HIEI2TnAnXAieiXHXKAIAXATEeunAPATT0I2EXTfiIE
ni*AXE2TATl}IT0nftinPE2BETTAIHPHeH2AXKAE0MH
AH2HriA2Ar0P0TAXA2;Ar0PA2AI0XT2I0T

INVESTIGATIONS AT ASSOS, 1881. 137

/ita

S^/xo9 ^oJyrjfrai rhs Ka . . •

Toi^ icaXoi9 KoX dyaOoii rtuv dv8/>ujv

TTopayCvtmnrtu av3pc$ a^toi tov Sijfiov ctS-

5 .... [oTC9 Ttp^ v^irdp^ova'av €V)(apurTULVf BeBo^Oai rj povX.'ff
. . . [xal rw] Si/fup hrqvijtrOai tov Brjfiov tov 'Aaauov arl t[^ •
. . . tvyoiijf. ^^v ^€1 irpo^ VH-^^y '^^ aT€<f>avov(rO(u avrbv iv Toi^
.... Tot? AioioxrtoiS) avXi^Twi' tJ irpwrg '^fJ^^p^ Xpvo-^ otc-
[^]avw, ^i ra> aTrooTctAot StKuoras KoXovq KayaOov^ '^^'V'l

10 [y]/Ki/x/xar€oi, hnjv^<rOat 8^ kuI tov? Sucooras tov^ wa-
fyaytvofiiyovg, 'E;(cAaov *A9rjvay6pov Aan/xov KAco/xop*
[yjou, Kflu OTCc^voxrai iKdT€pov avroiv 'Xpfvo'di aT€<f>dv<o, CTrfl]
[rep] ra9 ficv 8ia8i#ccur(U rcui' Sticbii' [urcosj ical Stx(U(i)f> ras
[8c 8iaA]u<rat ciTro Travros tov jScXt/otov, v7rdp\€iv Sc avrofi?]

15 C'^]/* Povkfiv KoX TOV l^fioy TrpioroK ftCTct Ta ic-

p<£ • v7rap;(Civ 8*[avT]^[vs] icai trpo^fvov^ t^ ttoXco)? ruiMV o'TC-
^V(p(r(U 8c [fjal Toy yp[a/x/uui]rca, McAay^pov McAay;(pov, ^oAc-
p<3 OTC^ovfp, ^i T^ 7rapaax[€]tr^cu t^ ica^* avrov )(p€iay fX€Ta
TraxTTfi ^tXori/ua9 * t^ tc ai^yy[€]\ia9 twv otc^kwv t^v €t[i-]

20 [oTcurtv] TrorjaajirOat T0V9 dywiyoOera^ tov /jloxhtikov. *Iva 8^ 'fttC*]
"Aa'O'LOi ci8i/(r<iKrtv Tiyv tc twv av8p[a>v] KokoKoyaOiav koX t^v
TOV Si^fjLov cvp(apurrtav, cupc^vot 7rpc(r)3cvTas otrivcs d^tico-

/xcvoi irpos avTOvs ^ttc o-tV tc r^ [i^]ovX^ kgu tw 8^

/LMW TO TC \lrrj<f>urfjia oltto^uhtovo'iv avrois xafi dir]o<^vun)o'iv ti}[v]

25 TC Twv dv8p(tfy KoXoicaya^tiav icai t^v cwotav [17] v ^o/tACV

TTpoc TOV B^fJLov avTwVy KOI TropoxaAco'ovo'a' 'A(r<riov9 Kal ^o-

p' avTOiS TTOiyoxur^at t^v dvayycAxav twv ©"[tJc*^-

vciiv wro TOV KaTaoTaOrfcofiivov dytavoOerov [j^ov

fiovaucov dytavo^' irpovorjaui 8c iva koi to ilnjffHO'fia dvafypa-J

30 ^ *^[0 ^J^Xiyv XiOivrp/ Kal dvaT€$rj Trap* avrois ^ T<p ^-
in^vcoraTip tothj). npccr/Jcvrat '^p-qOrfcav KXco/uii^
S79 'HyuMrayopovy *Ava^aydpas Aiowo'iov.

138 ARCHAEOLOGICAL INSTITUTE.

. . . [That] the people may appear [duly grateful (?)] to noble and good
men [and that] men may come to us who are worthy of the people, knowing
the gratitude which is in store for them, be it enacted by the senate [and
the] people, that the people of the Assians be thanked [for the good-will
which] they have for us, and be crowned with a golden crown at the . . .
Dionysia, on the first day of the flute-players, inasmuch as they have sent
us good and honorable judges, together with a clerk; and further, that the
judges who came to us, Echelaos, son of Athenagoras, and Latimos, son of
Kleomorgos, be thanked and be crowned each with a golden crown, inas-
much as they gave judgment in some of the suits [equitably] and justly,
and settled others amicably in the best possible manner ; that they have
[access to] the senate and people the first after the sacrifices, and that
they be consuls of our city; further, that the clerk Melanchros, son of
Melanchros, be crowned with a wreath of leaves, inasmuch as he has per-
formed his duties with all zeal ; and . . . that the overseers of the musical
contest be charged with the proclamation of the crowns. And in order
that the Assians may be made aware of the excellent character of these
men, and of the gratitude of our people, be it further enacted that ambas-
sadors be appointed who shall go to them and [thank] their senate and
people, and deliver to them this decree, and shall make known to them
the good character of these men and the good-will which we have for
their people, and shall invite the Assians to make proclamation of the
crowns in their own city also, through the overseer who may be appointed
to superintend the musical contest, and to see that this decree be cut upon
a stone pillar, and set up in the most conspicuous place in their city.
Kleomedes, son of Hegiasagoras, and Anaxagoras, son of Dionysios, wer^
appointed ambassadors.

Line 8. a\t\-r\rSiv t^ irpt^rri rifi^pf, : cf. Acschines in Ctcs. § 45, KriplrrteBai. rots
Tpay(f)lio7s, and the spurious decree in Dcmosth. Cor. § 118, ^toyvtriois rpay<i»9ois
KouvotSy with the corrupt expression in the spurious indictment (ibid. § 54), Aioyv-
<rlots rpayto^uy rfi Kaiyy.

Line 12. 2TE*ANnN is the stonecutter's mistake for 2TE*ANm.

Line 13. I22n must be a mistake for 12X12 or OSIHS-

Line 16. Perhaps for inrapx*iv 5* avrovs koX irpo^^yovs we should read itrdp-
X'iv 5f yfv4(rQai irpo^tvovs.

Line 23. The word here needed seems to be iiraiptarovai, which might be
spelled itrtv^tTovai'y but Mr. I_^wton reports that the fifth letter is circular (8, O,
or n), and the copy from the stone gives the ending E22IN, but with only the
first 2 certain.

Phonetic spellings, as r^^i fiov\r)v (1. 15), roy ypafifiaTta (1. 1 7), povXrfy
Kai (1. 23). will be noticed ; as also occasional omission of 1 in HI and QI,
and careless insertion of I after H and Q.

INVESTIGA TIONS A T ASSOS, 1881. 1 39

III.

This is a fragment of a decree of the Roman period, entitled ir^pi rov
iiff KaBivTatrBai TrpuKTopas. We have chiefly the preamble, of which the
last lines are imperfect. The inscription has the late forms C and /i
for 2 and O, and omits I entirely in HI and QL

AOrMAnEPITOTMHKAeiCTAEGAinPAKTOPAC
rNOMHBOTAHCTEKAIAHMWAAXONTONAO
rMATOrPA*«NEnANeOTCTOTEPMOrENOTC
EPMOrENOTCTOTEnANGOTCKPATHCINEI

5 KOTTOTMENELeEOC . . EHEIAHOKOINOCAnAN
TONEKHPOrONONETEPrETHCTI X KA X NEIKA
LI C CrXAHA CINOI CA AAOI CETEPrETITHNH A
TPIAAKOCMONTOEATTOrrENOCENnANTIKAI
PfiENAEIKNTMEXOCTHNElCTHNnATPIAAET

10 NOIANKAITHCHMEPONHMEPABEBOTAHTAI

NOMOGETHCEICTONAIOXA . . TAECTHNAITH.
KOINHEETEPrECIAtKAiniKP . . MErAAOT*OP

TIOTTHNHATPIAAKOr OUANAAEXO

MENOCTHNTOXnOA KTOPONHPA

15 SINAEAOXGAITH HMOKAITOIC

nPArMATETOM OMAIOICEHH

NHLGAIMENXT TONAP . .

TAAErOXTA

TAKAAAICT

20 niKE*A

LTPA

Ar . OMO

THNKATOPe

nPAKTOP

25 SENTK

TOTT

TO

I40 archjEological institute.

^oyfJLa ircpi rw firi KoBioTaaOai vpateropa^
TvitifAf) povX^ re icoi S-qfiovy Xaxovngy 80-
yfiUToypdifHav *EiirdvO(ns rov "EpfioycFOVs
"EpftoycFous TOv "Eiroi^ous KpaTi7<r«yci-

5 Kov TOV Mcvco'^ca)9 . *Eir€id^ 6 KOtvos diraK-
TCDi/ ^K irpoyovdjv €v€pyerqi Ti. KX. Ncuca-
(rt9 crvv airouriv ol^ dXAocs eucpycri t^ ira-
rpiSa, Koa-fiAtn/ ro iavrof ycvo?, ^ irayrl koI"
pcj €v8€iicn;/xcvo9 r^ cts rrp^ TrarptSa cv-

10 vocav Kal r^ ai^fjxpov fffiipa ptpovXrjfrax
vofJioOerq^ €« rov auova . . rot ^s rrpf Xtny •

KOii^ €V€py€fTia^ KoL vucp fA€yaXov 4^opr

riov TTiv irarpiZa kov os dvoSc^o-

ficvo9 TTfv Twv ?roA[iTiic(uv irpa] KTopaif irpa-

15 fiv, ScSop^^at r5 [j3ovA^ re kcu to) Sy/J/ud ical rots
7rpay/xarevo/>i[ci'Oiff vap* rifuv 'Pjwfuuots ^Tny-

vr^SuLL fi€v T TOV ap[ur-]

ra Xeyovra

ra KaAAarr[a]

20 7ruc€<t>a

orpa •

ay • ofJLO

rrjv KOTopO

TTpOKTOp

25 (tVLK

TOVT

TO •

IV.

A SEPULCHRAL inscription, found, Sept. 12, 1881, on a large trachyte
block at the beginning of the street of tombs in Assos.

nonAifiioTAPim

nonAIOTTiniANIHNSIZ
AKTIAAI

JNVESTIGA TIONS A T ASSOS, 18S1. 1 4 1

UmrXuf Ovapuf P. Vario

n<nrXi[o]v vUa 'Avt^is P. F. Aniensis

'AicviXq, Aquilae

If the copy is correct, we have the genitive UonXitov. ^Ainrjvais repre-
sents the Latin genitive Aniensis.

V.

This inscription, of the Roman period, is the touching tribute of a
Lesbian youth, named Anaxeos, to the memory of his dog Parthenope.
Similar epitaphs of animals may be found in the Anthol. Palat vii. 207,
208, 211, 212. A figure of the dog in bas-relief is cut upon the stone
above the inscription. The stone was found in Mytilene, in the autumn
of 1880, by workmen who were digging the cellar of a mill.

nAPe€NOnHNKTNAeA^'€ NANASeOCHCTNAerPEN

TATTHNT€Pna)AHCANTIAIAOTCXAPITA

eCTAeAONCTOPFHCAPAKAIKrCINcoCNTKAIHAe

€rNOTCOTCATPO*€ICHMAAEAONX€TOAB

eCTOAOPa)NXPHCTONnOIOT*IAONOCC€nPOeTMa>C

KAIZa)NTACTePrOIKAINEKPOXAM*I€nOI

UafiOeyoTnjv Kvva Odj(l/€y 'Ava^cos, ^ (rwdOvpey^

TavTrjv rtpirtiikrfi dKri8i3ovs \dpiTau
*Eot' iOXov <rropyrf^ apa ical tcvtriv^ & w koI ^8c

£woi;9 ovcra rpo^ct crrjfia \€\ov\€ ro8c.
*£« t6^ opSiVy )(p7f(TTbv iroLov <^tXov, 09 C€ wpoOvfJuoi

Kai (juiiVTa aripyoi, koI v€Kp6v d/tAt^cTrot.

Parthenope his dog, with whom in life
It was his wont to play, Anaxeos here
Hath buried : for the pleasure that she gave
Bestowing this return. Affection, then,
Even in a dog, possesseth its reward,

142

ARCH^OLOGICAL INSTITUTE.

Such as she hath who, ever in her life
Kind to her master, now receives this tomb.
See, then, thou make some friend, who in thy life
Will love thee well, and care for thee when dead.

XI. G. C.| Jr.

VI.

Found at Mytilene, in April, i88r

HPOCXAIPe

ZHCACeTHEMHNeCiA

HMGPACKe

II.

NOTES ON BUNARBASHI AND OTHER SITES

IN THE TROAD.

By WILLIAM C LAWTON.

HMEI2 AB KAEOZ OION AKOVOMEN OYAE TI lAMEN.

PRELIMINARY NOTE.

[The identification of the site of Homeric Troy has long been a subject
of animated controversy among those scholars who believe that the Iliad is
a more or less literal account of events which actually happened, or that it
has at least a considerable foundation of fact. In 1785-86 Lechevalier*
explored the Troad, and identified Bundrbashi as the Ilios of Homer.
Since his time other archaeologists have advocated the claims of Chiblak'
and of Atchi-kieui ; ' but their theories were never widely accepted, and
seem finally disproved by the investigations made lately upon these sites
by Dr. Schliemann. The dispute now, therefore, lies between the rival
pretensions of Bundrbashi and Hissarlik, which latter place is recognized
by the common consent of most archaeologists of note as the Hellenic
Ilium, the so-called ** Ilium Novum." The inhabitants of Ilium maintained
a tradition that the Trojan Ilios had not been destroyed completely by
the Achaeans, and had never ceased to be inhabited. They even pointed
out in their city many features which had survived the ruin of its famous
predecessor. We cannot, however, allow much weight to local traditions
of this character, which rest often upon very weak foundations — as in
Italy to-day, many towns are abandoning their good old names, some
of which are Hellenic, and older than the name of Rome itself,* to adopt,
often upon quite insufficient grounds, those of Roman municipia.

* Lechevalicr : Description of the Plain of Troy. London, 1799.

* Clarke. — Philippe Barker- Webb : Topographie de la Troade Amienne et
Modeme. Paris, 1844.

* Ulrichs: Reisen und Forschungen in Griechenland^ 184a

* Fran9oi3 Lenormant : La Grande Grhe. Paris, 1881.

144 ARCH^OLOGICAL INSTITUTE,

It is established by Professor R. C Jebb^ that "in the belief of the
ancient world " — except of the people of Ilium, who were influenced origi-
nally, doubtless, by a natural inclination to magnify the importance of their
native city, and except, too, of Alexander the Great and the Romans, whose
acceptance of the tradition of the I Hans was uncritical, and actuated by
motives of self-interest — Homeric Troy "had ceased to be inhabited
when it was sacked by the Achaeans, and its site had ever afterwards
remained desolate. This was not an accidental detail of the ancient
tradition, but a capital and essential feature. If so much of Troy had
been spared that the old inhabitants could continue to occupy it, the ten
years' siege would, in the feeling of the old world, have ended with an
abject anti-climax. The gods who had fought for the Achaeans would
have been robbed of their due triumph over the gods who had fought for
the Trojans."

Thus the ancients did not believe that the Hellenic Ilium occupied the
site of Troy. It is, however, entirely possible that the Hellenic Ilium,
which was probably founded centuries after the destruction of Troy, —
perhaps as late as the reign of Croesus,^ — and long after all tradition of
its exact site had disappeared, may have been established, unintentionally
and unknown to its founders, upon the accursed spot.

Criticism of the text of Homer affords arguments apparently strong
in favor alike of Hissarlik* and of Bundrbashi.* The question must
therefore be decided, if at all, by excavation.

The great extent of Dr. Schliemann's work at Hissarlik is well known.
Whatever bearing his discoveries may have upon the Iliad, the unearthing
there of six (or more *) cities buried one beneath the other, is an archaeo-
logical acquisition of the highest importance ; and the pottery and the
metallic implements and ornaments found in the four lower strata of
debris, form, with those of Thera, the earliest material that we have
for the study of primitive Greek civilization.' At Bundrbashi the only
archaeological investigation of any extent that has been made is that

1 Journal of Hellenic Studies, ii. I, — " Homeric and Hellenic Hium" All who
are interested in the subject should read this important article. Cf. Mr. W. J.
Stillman's letter on the ''Site of Homertc Troy,'' in the Nation of May sth, 1881.

2 Professor Jebb : loc. cit.

* Schliemann : Ilios; Professor A. H. Sayce : Jmtrnal of Hellenic Studies, i.,
" Notes from Journeys in the Troad and Lycia " ; fimilc Bumouf : Mhnoircs sur
r Antiquity, " Troie " ; Virchow and others.

* Nicolaides : Topography and Strategy of the Iliad ; W. J. Stillman ; Curtius
and others.

* Professor A. H. Sayce.

« M. CoUignon : VArchSologie Grecque. Paris, 1881. Cf. the study on " Troie!*
in the work of M. Bumouf above referred to.

INVESTIGA TIONS A T ASSOS, 1881 . 1 45

of HeiT von Hahn in 1864.^ But his work was too incomplete to pro-
duce results of importance.

In September of last year three members of the Expedition of the Insti-
tute at Assos — Messrs. Diller, Wallcer, and Lawton — made a summary
examination of the Trojan Plain, with the view especially of determining
the desirability of undertaking further investigations upon the Acropolis
of Bundrbashi, or elsewhere in the neighborhood. The observations of
the party and the conclusions arrived at by them have been embodied by
Mr. Lawton in the following Report upon iheir excursion. — T. W. L.]

During the autumn of 188 1 a party was sent from Assos to the Trojan
Plain, to examine the little Acropolis of the ** Bali-dagh," above Bundrbashi
village, and to report on the desirability of continuing the excavations of
Von Hahn. The ruined classic city now called Chigri was visited on the
way northward. This site is very little known, and the determination of
its ancient name might perhaps aid in the solution of some of the problems
of classical geography which await us in the Trojan Plain proper. A few
notes on other famous localities and much-debated questions have been
added. The Troad has been so seldom visited that it is hoped that the
testimony of unprejudiced eye-witnesses will be of some interest. Little
of what we tell is new ; but we have tried to see with our own eyes, and
not to quote at second-hand. The ascent of Ida was made in October
by a party on foot, who skirted the whole northern shore of the Gulf of
Adramyttion, and, ascending from the town of the same name, returned
through the Plain of Beiramitch.

Most of the topographical notes are to be credited to Mr. Diller. The
drawings, and the plan of the city walls on the Bali-dagh, with the ap-
pended explanations, are the work of Mr. Walker. We are under great
obligations to Mr. Frank Calvert for his hospitality and courtesy, and
also for his most instructive guidance. Dr. Schliemann, with his usual
kindness toward students, placed most freely at our disposal his rich
library of works on the Troad.

CHIGRI.

Midway on the journey from the Gulf of Adramyttion to the
Hellespont is the little Turkish town of In^, built on the west
branch of the M^ndereh, just above its junction with the main stream.
For several miles before In6 is reached the road runs northward

* V^on TTahn : Ausgrahutti^n attfder homerischen Pergamos. Leipzig, 1865.

10

146 ARCHAEOLOGICAL INSTITUTE.

with the river, crossing it at short intervals. In May there was a
brisk stream, more than ankle deep and a dozen metres wide ; but
in October the bed was quite dry. From this river road is seen
prominently on the left the long ridge on which Chigri lies. At In^
there are tolerable khans and an excellent locanda, and the Greek
inhabitants are intelligent and courteous. A visit to Chigri should
be made from here in a single day, as the Turkish villages nearer
the mountain can provide no tolerable accommodation.

The ancient city now called Chigri, identified by Mr. Calvert with
the classic Neandreia, is magnificently situated on a plateau more
than five hundred metres above the sea. The walls extend along
the ridge for over a kilometre and a half, and are to a great extent
still standing in good condition. The courses of stbne are some-
what less regular in their lines than the best work at Assos, and
occasionally lapse suddenly into polygonal. The thickness of the
wall where we measured it was 3.20 metres. The general structure
was the same as at Assos, each side of the wall being neatly faced
with smoothed stones, while inside the stones were left rough. The
interval between the inner and outer faces of the wall was filled up
with small stones.

The ground within the walls is approximately level, but with a
considerable rise towards the northern end, as well as at the south
end near the little Acropolis. Large rocks lie scattered over the
surface, and the soil is as a rule. very scanty. No hewn stone is
seen, and in general little except the walls recalls the fact that a
city once stood here. It would seem that the ground was never
fully occupied (perhaps no very massive buildings were erected),
that no later settlement came to accumulate debris above the Greek
remains, and that the storms of twenty centuries have washed the
hill almost bare of all traces of human habitation. Greek coins
are often found here by the Turks of the village just below, who
pasture their flocks within the walls.

The Acropolis is merely a precipitous hillock, covered with great
fragments of natural rock piled high upon each other in the wildest
confusion. From this point we obtained our first good view of the
Trojan Plain. Just west of this elevation is a great gate in the city
wall. Its jambs are still standing, and in one of them is a neatly
cut slot, in which the hinge of the gate may have rested. Between

INVESTIGATIONS AT ASSOS, 1881. 147

this gateway and the Acropolis was merely a re-entering angle of
the wall ; but on the other side the entrance was commanded by a
large square tower. Looking outward from this gateway, the visitor
sees above, on the left, a curious bit of wall on the side of the
Acropolis, in which two large rocks have apparently been utilized
in situ.

Just outside the gate are half a dozen shallow graves, each lined
with four rough slabs of stone. They were excavated by Mr. Cal-
vert, who found in them pottery, which in his opinion forms a link
between the art of the Lydian city of Hissarlik (the sixth, accord-
ing to Dn. Schliemann's present numbering) and archaic Greek
work. Most remarkable are the terra-cotta figures of an Egyptian
or Assyrian type. Many of the vases are of a dark gray clay, and
similar in form to those found at Hissarlik. Some scarabaei were
found ; but these are supposed to be imitations. The fact that
these graves were unrifled tends to strengthen the impression that
this site was not occupied by later races.

If we can form any judgment from the contrast between these
little graves just outside the great gate of Chigri and the magnifi-
cent street of tombs, crowded with exedrae and sarcophagi, in the
corresponding position at Assos, we can infer that here there was
never much display of wealth and splendor.

FROM INE TO BUNARBASHI.

We first saw the Mendereh, by general consent identified with the
Homeric river Scamander, —

in September, at a point an hour's walk north of In^. It was run-
ning with a swift clear stream, half a metre deep at most, and half
a dozen metres in width, winding among the banks of sand that
fill its broad winter bed. Fish three or four inches long were
abundant. The plain is here about two kilometres wide, and was
at the time of our visit covered with maize. Further north the
wooded hills close in, and for several hours the road follows the
curves of the river around their bases. At last the path seems to
be crowded down into the sand at the very brink of the river ; and

148 ARCHAEOLOGICAL INSTITUTE,

after leading wearily around a few long sweeping curves, it abandons
the valley and strikes over the hills towards the left. Presently
a crest is reached, and suddenly the Plain of Troy appears, ex-
tended at our feet. The minaret of Bun^rbashi is already in sight
below, and in descending towards it a glimpse is caught, above us
on the right, of the tumuli upon the Bali-dagh. We are passing
downward among the valonia oaks over the ground which Leche-
valier covered with the lower town of his enormous Ilios. On the
left, not far away, the regular cone of the Ujek Tepeli (supposed
to be the tumulus erected by Caligula) is a prominent landmark,
standing on a considerable elevation upon the western eilge of the
plain. Before us the M^ndereh winds mile after mile through the
level plain towards the distant strip of blue water, beyond which
rise the islands of the Northern -^gean.

He who is fated to spend a night at Bundrbashi would perhaps
do best to test the hospitality of ** Zachariah's chiflik," the country
house of a rich Christian Albanian on the edge of the village. In
the strong enclosure, within which the stables and servants* quarters
are built against the wall on three sides while the veranda of the
house forms the fourth, he will find a reminiscence of Odysseus'
palace, or of the enclosures throughout the Iliad, around which the
lions are perpetually roaring and watching their chance to leap
over the walls. He will be welcomed as Odysseus was, not by
faithful old Argos, but by the dogs of Eumaios, —

ol fX€V KfKXriyOVTfS fTTtbpafXOV

until the master appears, and

Tovs fi€v ofioKXriaas trtvtv Kvvas aWvbis SXXjj
nvKurjaiv \i$d8faa^,^

It is often wondered at that man's most faithful comrade has
fared so ill in our proverbial expressions ; but the pedestrian in the
East who, after being set upon at intervals all day by the wolfish

1 " And suddenly the noisy dogs saw Odysseus :
With a yelp they flew at him.
. . . shouting, dispersed the dogs in all directions
With well-aimed stones." — Odyssey xiv. 29.

i

.^MWliI

^r

mw'^

mi

15»

<mmm-

The AcnuPOLls of the Bali Dach .

INVESTIGATIONS AT ASSOS, 1881. 149

shepherd dogs, comes at nightfall into the village only to be at-
tacked by a yelping pack at the gate where he seeks shelter for the
night, and who has seen and heard the filthy, maimed, blood-stained
brutes that go howling in droves throughout the night in the larger
cities of the Orient, will not wonder that to the Mussulman " dog "
is the strongest expression of loathing and contempt.

THE BALI-DAGH.

The Acropolis of the Bali-dagh, identified by many writers since
the time of Lechevalier as the Pergamos of Priam, is a striking emi-
nence (rising some hundred and forty metres above the Hellespont)
at the southern extremity of the Trojan Plain, overhanging the left
bank of the M^ndereh just where that river breaks from the moun-
tains and fiows out into the plain. As it is the last height on the
west bank of the stream, it commands an unbroken view down the
whole length of the plain. The M^ndereh fiows around it on three
sides, at the base of steep cliffs. The only easy approach is over a
comparatively narrow neck from the northwest. Its position will
be clearly understood from the accompanying map and drawings.
Plates H., in,, IV.

The walls which appear on the map were traced out and partially
laid bare by Von Hahn, who, accompanied by Messrs. Schmidt and
Ziller, of Athens, excavated here for three weeks only in May, 1864.
He employed at first five laborers, afterwards twenty-two, and for
the last three days thirty-six. His report is a small pamphlet of
thirty-three pages, in the form of two letters to Mr. Finlay, the
historian of Greece.

The walls are constructed chiefly of the crystalline limestone of
which the mountain is formed. Indeed, immediately under the
northwest corner of the city wall is a quarry of considerable size cut
into the precipitous side of the hill. In this quarry is an old wild
fig-tree, which bears excellent fruit. It will be noticed that on the
south side the walls are not traceable. The native rock here is very
scantily covered with soil. Occasional remnants of light house-
walls appear, and others which seem to supp6rt a terrace of earth.
In the whole eastern part of the city the rock strata of the hill
appear often on the surface. In the middle and western portions
there seems to be a considerable depth of soil.

150 ARCH^OLOGICAL INSTITUTE,

Von Hahn's work consisted of little more than running trenches
along the outer side of the walls, which have consequently at
present the appearance of structures built to sustain the mass of
earth within ; but this is doubtless because, since the abandonment
of the city, soil has been washed down from the highest jjortion of
the Acropolis, and accumulated against the inner side of the walls.
In fact, it must be remembered that until excavations were made
here the whole surface of the hill seemed to be a natural slope,
and only the weather-worn upper course of the wall appeared on the
surface, scarcely to be distinguished at first glance from natural
boulders.

The space within the fortifications was less than two hundred
metres long by about one hundred metres wide ; and within these
compact limits lies almost everything of interest in connection
with the site. On three sides, as we have said, there is a steep
descent of limestone cliffs towards the river. Passing to the north-
west along the plateau, we notice foundations in the form of two
tangent circles, built of small rough stones carelessly laid together.
The middle of each circle is slightly depressed. The circular walls
rise very slightly above the surface. Virchow supposed these to
be in the agora of the city. Von Hahn suggests the possibility of
their having been threshing-floors, — though rather large for the
purpose, considering their situation so far from the plain. At the
point of tangency of these two circles Von Hahn dug down nearly
a metre, but the character of the wall was the same at the bottom
of his trench as above. It can hardly extend much deeper at any
point, as the rock comes to the surface close by. South of these
foundations, near the edge of the plateau, is visible one corner
of the foundation -wall of a considerable building. The stones
are rather large, and unhewn. Just beyond, an embankment runs
across the plateau at its lowest point, whence there is a slight rise
towards the city as well as towards the tumuli.^ This embankment
is composed of earth and small broken stones, and may be the re-
mains of a rude wall which at one period marked the limit of the
city. Turning to the left we find ourselves upon a spur running

* Three prominent tumuli alonj^ the slope of the Acropolis, which are referred
to in all descriptions of Bunarbashi. They have been identified by some of the
most zealous advocatesof'Lechevalier's theory as the tombs of Trojan heroes.

INVESTIGATIONS AT ASSOS, 188L 151

up the stream of the Mdndereh, towards which it presents a con-
tinuation of the steep rock face of the Bali-dagh proper. The soil is
very scanty, and large rocks project from it on all sides. Never-
theless, Mr. Calvert has discovered and opened here an ancient
cemetery. The bodies were placed in enormous earthen jars (m^ot),
and these were laid on their sides in the interstices of the rocks
and covered with earth. In these jars was found some pottery
of the fifth and fourth centuries b. c.

If, on the other hand, we turn northward upon passing the embank-
ment, we shall go down a very regular slope, which brings us to the
plain by the river side. Along the edge of this slope we see in suc-
cession the three tumuli, and, lower down, many considerable heaps
of stones. The tumuli are themselves mere heaps of stone, in two
cases mingled with earth. Two of them have been opened with
very meagre results. They do not exhibit the structure described
in Iliad xxiii. 255-56, and exemplified in the Tomb of Tantalos
near Smyrna, the i'omb of Andromache near Pergamon, and many
similar structures, —

TopvcixravTo di arjfia^ OffitiKid rt irp-t^oKovro
a/i^t nvpriv. tiBap bi x^^^ *''^* yaiav cj^fuai/.*

Close to the uppermost tumulus is a rudely circular excavation in
the solid rock, which may have been an ancient quarry. Half way
down to the second tumulus is another such quarry some twelve
metres across. Close to the south side of the lowest tumulus is a
circular wall, rising somewhat above the surface, and made of much
larger stones than the two upon the plateau. This is perhaps the
substructure of a tumulus which was never finished, or from which
the earth has been quite washed off.

In the stone heaps Mr. Calvert showed us that the line of a wall
could occasionally be traced, though disguised by the toppling over
of its upper portion.'^ These may therefore have been house-walls.
Here, again, any hope of fruitful excavation is frustrated by the

' " They rounded off the burial mound, and built a sustaining wall around it ;
then they poured libations upon the bankcd-up earth."

* Cf. Viollet-lcDuc, Ilistoire de T IlaNtation Ilumaiue^ chap xv., " Les P^lasgesy^
and illustrations. Remains of similar circular house-foundations have been
found elsewhere in Asia .Minor, in Greece, in Italy, in Spain, etc.

152 ARCHAEOLOGICAL INSTITUTE.

rock of the hill, which appears in step-like layers all the way down
the slope.

The descent from the Acropolis straight towards the village of
Bun^rbashi is rolling and gradual. The village is about a kilometre
and a half from the Acropolis. On these slopes we failed, like other
recent visitors, to find pottery or any other trace of human occupation.
The soil is tolerably fertile, and gravel occasionally appears.

The haste with which Von Hahn worked may in part explain the
fact that he found few coins or relics of any kind. Near the pro-
jecting terrace was unearthed a headless terra-cotta figure six
inches high, of fairly good workmanship. This gave him the impres-
sion that a shrine or small temple may have stood here, as these
statuettes were common votive offerings. By an ancient grave small
bits of stucco were picked up, and also fragments of tiles of good
workmanship which formed the covering of the grave. Two simple
black-glazed lamps and fragments of a white pavement were found.
Only sixteen coins came to light, of which four were identified as of
Mytilene, three of Sigeion, two of Abydos, one each of Alexandria-
Troas, Ilium, and Arcadia, all dating from the third or second century
B.C. These were found "at no great depth." At the northeast
of the square well-laid foundation-wall were found standing the
slumps of two weather-worn unfluted columns, 40 centimetres in
diameter, and respectively 120 and 90 centimetres high. A few clay
waterpipes and tiles also came to light here. No inscription of any
sort was discovered. If excavations are undertaken at Bundrbashi,
it can hardly be with any hope of startling and brilliant discoveries.
Von Hahn's experience has shown that the tangible return is likely
to be small. Yet his work was only half done, and at some time
ought to be completed. The popular interest in this region is a
legitimate and desirable one. If an expedition should bring away
nothing but a more accurate knowledge of this beautiful country, it
would not have been sent in vain ; while a simple inscription giving
a clue to the name or age of the ancient city would be of the highest
interest and value. The walls ought to be laid bare both inside and
out, and the original level on which they were built accurately deter-
mined. The principal buildings which have already been discovered
ought to be carefully cleared out. A series of pits should be sunk
to determine with exactness the amount and character of the de'bris

INVESTIGA TIONS A T ASSOS, 1881. 1 5 3

accumulated within the walls. This could be done at a moderate
expense, and a satisfactory judgment could then be formed as to
what might yet remain to be done.

The earth could be disposed of easily by shooting it over the steep
rocky cliffs, and there would be no danger of thus covering places
which must afterwards be excavated. Tolerable quarters could
probably be obtained in the neighboring village. An abundance
of trained labor — thanks to Dr. Schliemann — can be secured in
the vicinity.

To sum up, then : The thorough excavation of the Acropolis of
Bundrbashi might give interesting results. If it finally laid the ghost
of Lechevalier*s Troy it would help the cause of peace as well as
that of classical geography; but it would probably be by no means
a rich field in the ordinary sense, and we should hesitate to urge
its claims while so many sites in Asia Minor and in Greece proper,
whence rich archaeological returns are certain, are yet untouched by
the spade.

[Notes on the Map of the Acropolis of the Bali-Dagh.

The notes in quotation marks are taken substantially from Von Hahn's Report.

" Beginning at the northwest corner of the Acropolis outside the walls ;

*M is a quarry 7 metres deep and 15 metres in diameter.

'*B, The wall 7? is composed of blocks averaging .60 X -60 X *2o, well
cut and joined, resting on a projecting ledge which advances .05 metre,
and is .15 metre high. The wall is of a yellowish stone, probably vol-
canic, radically different from the stone of the Bali-dagh and neighboring
hills."

The foot of this wall is covered with ddbris ; indeed the accumulation
of ddjris is greater here than elsewhere, and makes it exceedingly difficult
to trace any definite plan of the walls at this point. The exposed surfaces
of the yellowish stone are disintegrating wherever found upon the Bali-
dagh.

** Between the wall ff and the bastion D E \s 2. passage C, 1.40 metre
in width, with side walls of the stone of B. The size of the blocks varies ;
largest, 55 X i. 90 metre. Two pilasters [of which nothing now remains]
were found at the entrance, but no signs of a gate.

" Above the sides of this passage appear three courses of projecting
blocks. Each successive course approaches its corresponding opposite

154 ARCH^QLOGICAL LWSTITUTE,

course more nearly than the next below ; in this way undoubtedly was
formed the roof of the passage. The walls below these courses are i
metre in vertical height. There is no trace of walls beyond the angle
shown on the plan."

This passage is now nearly filled with ddbris. We were unable to trace
the walls around the angle, and the three roofing courses have entirely
disappeared.

Above were numerous rough walls recently built, possibly to protect
the sheep and goats pasturing here.

D, E ■'* These remains, of what was apparently a bastion, consist of
irregular blocks of about .50 X i metre, with rough and projecting split-
face surfaces. The joints are well cut. '1 he stone is from the quarry near
at hand."

These bastion walls are now about 2 metres in height ; the angles of
the wall are carefully cut, so that a margin, perhaps 8 centimetres
deep, is left smooth upon each face of the angle. The work is very similar
to that upon the great gate at Assos.

The wall 6' is of the same character and workmanship.

/% G. *' At the open space between E and G four rough steps led up to
a wall F^ of which, however, only three stones remain in place. Along
the lower edge of these stones run three grooves with rounded edges."
Remnants of the wall appear to extend behind G ; but Von Hahn did not
wish to destroy G in order to ascertain their exact disposition. The work
and materials of B and F are similar ; the walls, Z>, £", and C, Von Hahn
thinks later additions.

/'"and the four steps are now covered with earth. Von Hahn's ditch
has filled but little. G is, from the bottom of the ditch, 3.36 metres in
height.

//. " The north wall, //, extends in a curved line towards the east, follow-
ing the curve of the hill, varying in direction from 98° east at its juncture
with the terrace G, to 1 10*^ east at the angle of/. It is built of smaller
blocks than Z>, E, and G, more oblong than square, with rough faces and
excellent joints. It is apparently of a different period from the bastion
Z>, A', and the terrace 6^."

Only the upper course of this wall is now visible above the soil ; it
appears never to have been excavated.

/, K. " The wall now comes forward and inclines at an angle of nearly
45°, which inclination continues eastward. We could find no gate be-
tween /and A', only fragments of possible walls."

The inclination in the wall running east is much less than that of the
return wall.

The corner is built upon natural rock, which here comus to the surface.

INVESTIGATIONS AT ASSOS, 1881. 155

The disposition, thickness, strength, and inclination of the walls would
seem to suggest that they formed the base of a tower guarding the gate L.

The stones have joints well cut ; the courses follow the curve of the hill,
which here descends rapidly.

The walls upon each side of the gate L are in some places built upon
the natural rock ; the joints are excellently, cut, the beds being absolutely
level.

The passage into the city can be traced only a short distance, because
of the ddbris of later light walls.

The east side of the Acropolis is covered with a confused mass of
walls, of the age and use of which Von Hahn formed no conjecture. He
could trace the city wall but little farther. Doubtless the Acropolis limits
varied at different times. All attempts to follow the south wall were vain,
until the southwest corner was reached ; here a fine polygonal wall, largest
stone I metre in height, was found. X^ K Von Hahn thought this wall
the oldest upon the hill. These walls incline at an angle of 69^. Von
Hahn considered them the foundation of the city wall proper. From here
eastward, the inner or lining wall is the only enclosure of the Acropolis.
At this point the mass of d(5bris is very great. Von Hahn considered Z
the finest wall found. It has four courses, each 45 metres high : each
course projects beyond the one above it ; the surface of the courses
has an inclination of 85°. — C. Howarx) Walker.]

THE BUNARBASHI RIVER.

A short distance southwest of Bunarbashi are the springs called
the " Forty Eyes." They are found in the old crystalline lime-
stone near its junction with the tertiary limestone. In September
— the driest month of the year — they were pouring out an abun-
dance of cold, pure water, forming a swift and clear stream, along
the banks of which grow thickets of rushes and willows. Turtles
and frogs were abundant. This stream, slightly augmented at
times by the surface drainage of the hills west of the Bali-dagh,
forms a series of marshes along the western edge of the Trojan
Plain, and what is left of it passes off at last through an arti-
ficial channel cut for it between the heights of Sigeion and Ujek
Tepeh, to Besika Bay. Its natural course was traced by Forch-
hammer by the old channels, which are still filled when the river is
at its highest. He shows that it formerly emptied into the present
M^ndereh just above Yeni Slier. It must have been a mere thread
of connection between swamps, in a part of the plain unfit for mili-

156 ARCHJSOLOGICAL INSTITUTE,

tary operations or human dwellings ; and it does not appear to be
alluded to at all in the Iliad.

In this poor little rivulet Lechevalier recognized the PaBvSiyr^i^
2Ka/jLav6po9, " deep-eddying Scamander." It is a valid objection to
this, that it would make the whole twenty-first book of the Iliad, if we
attempt to identify exactly the scenes of Homer, utterly meaningless.
The fact is that the Sxa/iAvSpos is throughout Homer 6 irora/uk,
the great river of the plain, — that stream which, however its lower
course may have changed, must have been for ages sweeping around
the Bali-dagh on its way from Gargaros to the Hellespont. We will
quote here one illustrative passage : —

Tav €$v€a TToXXa vtau &jro Koi KktaidoMf
fs nediov irpoxiovro ^KOfidvbpiov avrap vir6 x^o»y
afiff^aXeov Kovdfii^f nodav airrau re koi iinro»v,
itrrav d* cV Xtifiavi ^KOfiaubpii^ av6fp6iVTi
fivpioij oaaa rt <f)vXka kcu avBta yiywvrat &pQ?-

THE PLAIN.

The walk of fourteen kilometres from the Bali-dagh to Sigeion
should be taken once by every student of the Iliad, though he
may find it wearisome, and possibly monotonous. It will be heavy
walking over the ploughed land and through the endless fields of
maize ; but he will remember that under the very walls of the city
Homer speaks of TrcStoto TrvpoKfiopoiOy and Athene, striving with Ares,
hurls at him, —

\idou . . .

K(ip.fVou iv frcdto), ptKava Tprj^vv tc pt'yau Tf,

Tov p avhpfi npoTipoi diaav (p.pfvai oZpov dpovprj^f —

^ "The many tribes poured forth from ships and huts
Into the Scaman-lrian plain. The earth
Groaned fearfully beneath the feet of men and horses.
And in the blooming Scamandrian mead they stood
Countless as are the leaves and flowers of spring."

f/iad ii. 464.

* "A stone

That lay upon the plain, black, rough, and huge,

Which men of earlier days had set, to be

The cornland's bound.'*

///ac/ XXI. 403.

Plate V.

INVESTIGATIONS AT ASS OS, 1881. 1 57

from which it appears that the battle-field had long before been
under cultivation. The pedestrian will come upon flocks of sheep
and goats and herds of grazing kine. The bones of these animals
were constantly found in the excavations made by Dr. Schliemann.
On the night just mentioned, the Trojans drove out from the town
p6a^ KoL L<l>ia firj\a for their evening meal. Apollo is mentioned as
tending the ctXtVoSa? tkucaq fiov^ of Laomedon, during his year of
servitude, but it was, —

Beside the river channels are growing irrcAcoi tc Kat ircot rjSk
fivpiKou, and while pressing through the thickets the traveller is
often held fast, like poor Adrestos, by the thorny tamarisk. There
is no lack of marshes like that where Ulysses lay, —

Kctrh fuonrfia irvKvi,
hv liovcucas Koi Aor.

If the traveller starts up a heron, as we did near the " Forty
Eyes," it will recall to him the night when Diomedes and Odysseus
set forth to visit the Trojan camp : —

rotcri dc deftly ^k€U ipo»^thv ryyvs Sioio
UaXXas ^A$i]vairj' rot d* ovk iSov 6<f>BdKfioi(n
ruKTa d«* 6p<f>v(urjVf SXKit Kkay(avTos ufcovcrov.'

On such soil common sights and sounds are full of classic asso-
ciation. The eagle we roused up was an omen from Zeus, and
the flock of great birds, —~)(tivS>v rf yepaywv rj Kvicviav hovXixpBeCpwVy —
flying in circles and working their way up to a higher level, was a
Homeric simile written in the sky. Every step onward leads to
the conviction that the writer of the Iliad knew well the plain that
was the scene of his heroes* struggles. It was no dreamland like
Phaiakia, but the very ground beneath our feet.

SIGEION.

We sit upon the promontory in the cool sea-breeze and study
the lovely panorama spread before us. The frame of the picture,
at any rate, is unchanged, —

^ " And Pallas Athene sent a heron for them,

Close to their path on the right. They saw it not with their eyes
Through the murky night, but they heard its cry."

I/taJ X. 274.

158 ARCHAEOLOGICAL INSTITUTE.

o^cai §v (OtkritfOa kol at K€V rot ra fi(firi\]j,
.... *FXkri<nrovrou in l\B\i6fvra irktovaa^

Out of the blue Mgezx\ rise Tenedos, Lemnos, Imbros, and, soar-
ing high above Imbros, the stately peak of Samothrake, — the
watch-tower of Poseidon, — where

ou5' oKaofTKOKLriv €i)(€ Kp€i<i)v ^ E'/o (ri}(B<ap^

Koi yap 6 davfid((i)v ^aro 7rr6\(ii6v re fidx^v T€

v^oO eV dKpoToTTjs Kopv(f)^s Scifiov v\rji(rarjs

Spri'uciris ' €v6fv yap i(f>jivfTo ndaa p,(U "idrj,

(fmivero de Ilpidp.010 jioXis Koi urj€s ^A^aicitv. ^

«

Far away to the southeast we can descry the seat of his mightier
brother, Zeus, —

eV* aKporarrji Kopv(j)i}S iroXimldaKos "ifi^r-

These are the magnificent limits of Homer's ** mythological back-
ground," as Virchow well expresses it.

Nearer at hand, we can trace the line of heights about the plain,
and see the river descending from the far-away water-gap by the
Bali-dagh. Opposite us stands Rhoiteion ; and stretching from it
towards us is the low sandy shore where the Greek ships lay, —

Koi nXrjaav &7rda'rjs
rjXovos ar6p,a paKpbvj ocrov <rvv€€pyadov uKpai.*

The sandy spit of Koum Kaleh runs out boldly, and looks like a
recent encroachment of the land upon the sea. The hypothesis of
Strabo, that in Homer's day a deep bay extended inland between

^ *• You shall see if you will, and if you care for that,
. . . shii^ sailing on the fishy Hellespont.'*

I//aif ix. 359.
2 *• The wide-rtiling Earth-shaker kept no blind watch :
For, wondering at the war and strife, he sat
High on the topmost crest of woody Samothrake ;
Thence all Ida was in sight.
And Priam's city and the Achaians' ships."

/Ir'ad xiii. 10.
' " And filled the broad mouth of all the coast
Within the promontories' bounds."

I/iat/ xiv. 35.

INVESTIGATIONS AT ASSOS, 1881, 159

the promontories, seems to be finally disposed of by the learned

and searching essays of Mr. Calvert and Dr. Virchow. We are

unable to see any allusion to such a bay in the Iliad. In the chief

passage on the subject, from which we have just quoted, the poet

says, —

upwxro i^f r

At the beginning of the Twenty-third Book he says i^a? tc icai
'EAAijcTTTovToi' iKovTo. Truc, the expression 6aKaxT(rq^ koXttov also
occurs ; but its most natural and literal meaning is " the bosom of
the sea." Certainly that is the meaning of the same expression
in xviii. 140, where Thetis bids her sisters

bvT( BaXaa-OTjs €vpia leJXirov,
o^o^fvai T€ yipovff akiov kcu doifiara narpos,^ —

for we do not imagine that any scholar would venture to place the
home of Nereus in this hypothetical " bay of the sea " !

Decisive evidence on this subject would be the discovery of some
human monument of undoubted antiquity near the present shore-
line, or of remains of such character is to mark clearly a different
shore-line further inland. A search for the wall built by the Greeks
in the Seventh Book has been suggested. Apart from the difBculty
of deciding where to seek it, it should be remembered that it was
constructed hastily, in a single day and on a sandy shore, —

7jp,os b* oCt up TTfi) 70)S, iTi d' d/i^tXv/ci; vv(t
T^fios tip dp(fH TTvp^v KpiTos ^yp€To \a6s *A)(aici>¥f

dva-fTo d* fftkLos, TcrcXcoTo 5* ipyov *A;(aicI)»',^

its Utter destruction being meanwhile promised by Zeus (vil. 459-
463), and afterwards described with more elaboration than its erec-

* " Plunge into riie broad bosom of the sea,

To behold the Ancient of the deep, and your father's halls."

* " And ere yet day was come, but twilight lingered,

A chosen band of Achaians arose about the p>Te.

The sun set, and the Achaians* work was done.*'

liia/i vii. 433-65.

i6o archjEological institute.

tion (xii. i-ii). But, indeed, against all attempts to use the Iliad
as a history or an itinerary, there is an earnest warning in the line
we have made our motto.^ In the text Thucydides read, the wall
seems to have been built when the Greeks first landed ; for the sup-
position that the historian wrote carelessly, with only a vague recol-
lection of the Homeric account, is surely inadmissible. cn-ciS^ M
dff)i.KOfi€ioi f^dxo ^»(poiTrf(rav {8^X.ov 8c • to yap €pvfia r^ OTpar<nr«(S<p owe

But where then is Troy ? The distance from Ilios to the ships
seems pretty accurately fixed by an abundance of accidental evi-
dence. Dr. Schliemann has treated this question so exhaustively
in his f/ws, that it is needless to pile up quotations upon it. Perhaps
the clearest single passage is that where Idaios starts — probably
from the agora before Friam's palace — at dawn, to carry his
message to the ships, and is back again by sunrise, —

The action waits until his return, —

oi d* cor* ftp ayopjj TpS>is Koi Aapdav(a>pcr,
ndvTft 6fiTiy€p€t£, »ror»dry/i€vot, &fm6T &p* TkBtn
'idaior* 6 b* dp* ^XB€j —

' The chief arguments of Forch hammer, Virchow, Calvert, and Schliemann
may briefly be summarized thus : —

(i) By comparison with the effect of other rivers of greater power, like the
Nile and Ganges, it appears that all the alluvium the M^ndereh brings down
could not build the coast-line out many furlongs in three thousand years.

(2) The current of the Hellespont is strong enough to sweep away any
deposit.

[Beyond the line of shore of to-day; but not if there was a bay. In this case,
the shore-line might have been built out till it met the current, when the process
would cease. — \V. J. S.]

(3) The crumbling vertical banks of the Asmdk mouths and lagoons show that
the sea is rather encroaching tiian losing ground.

(4) If there were any considerable permanent deposit of alluvium, the first
result would have been the filling up of the great lagoons of the lower plain.

(5) The forts at Dardanelles and Koum Kaleh were built respectively about
four and two centuries ago ; but no growth of the shore has occurred at those
points since their erection. They still front directly upon the sea.

2 " And when on their arrival they had won a battle, — as it is plain they had,
else they could not have built a wall of defence for their ships." — Thuc. i. 11.

But

INVESTIGATIONS AT ASSOS, 1881. i6l

and a few lines later, —

*HcXior /Acv rircira viov irpoaifiaikk^v dpovpas,^

There are passages which indicate that the poet, in imagination
at least, saw such a general picture as. we are studying from Sigeion.
Take, for instance, the passage where the gods /ac^ ofiiXov . . . rjkvOov
&ySpu)v: the battle was then raging near the shore, and Athene
urges on her beloved Greeks from close at hand, —

aras 6re fxtv napa rdcfipop opvicrrfu T€ixtos (KT6sf
SXKoT €ir* aucrdoav ipibaimav ficucphv dvT€U

a^€ S* "Aprfs (T€po»$€», fptfJ^vjj XatXcnri laoSf
d(v KOT aKpoTarris TroXccay Tpa>€<r<ri iccXcvd^y,
aXXore Trap Si/xocyri Btav cVt KaXXiKokayQ'^

There is one passage which seems to have been preserved from
the oldest traditionary lore to aid us in our search for the site of
Troy,-

KrltTO't df ^pdaviriv, cVcl oH ir<o *lXiO£ Ip^
cV iTtbi^ iTfvSkurrot irSkis yttpovav dvOpoiinoVt
aXX' €ff vtraptias ^Ktov ircXwidoKov "idr^s,*

* " At dawn Idaios went to the hollow ships.

The Trojans and Dardanians in the agora
Sat all assembled, waiting for the coming

Of Idaios ; and he came

. . Then the sun was just beginning to shine on the fields."

I/iad vii. 381-421.

* " Sometimes standing by the moat outside the wall,

Sometimes on the resounding promontories, she shouted afar.
And Ares, on the other side, roared like a black hurricane,
Shouting shrill orders to the Trojans, sometimes from the Acropolis,
Sometimes running along the Simois to Kallikolone."

I/tai/ XX. 49-53.

* " He built Dardania ; for holy Ilios,

The city of mortal men, was not yet founded in the plain.

But they yet dwelt on the foot-hills of many-fountained Ida."

Iliad TUX. 216.

[If we arc to take the Iliad as our literal guide, might not this passage refer

simply to the change which took place, — as in most Hellenic cities, — when the

Acropolis was flearcd of dwellings and left, except in case of necessity of war,

sacred to the gods, and occupied only by their temples, while the city proper was

built beneath in the plain ? — T. W. L.]

II

1 62 ARCHAEOLOGICAL INSTITUTE.

Now the city on the Bali-dagh (Bundrbashi) could hardly be
more suitably described, at least as it appears from the lower plain,
than on " the foot-hills of Ida." It may, then, be Dardania ; but
hardly the citadel of Troy.

But where, then, shall we look for Ilios ? Not hopelessly over
the flat expanse before us, for the city certainly had an Acro-
polis high enough to overlook the plain, since from it the gods
often watched the battle. There seems to be but one possible site.
From the distant boundary line of hills a long ridge descends
towards us, dividing the plain into two river valleys. This ridge
ends, within a few miles of us, in a little eminence, of which the
name is familiar enough, for around it has raged, if not the glorious
struggle of Homer, at least the second Trojan war — of words ! It
is in the plain, for the plain sweeps nearly around it. And yet we
cannot resist a feeling of disappointment as we say, " What 1 only
that little brown hillock ? "

HISSARLIK.

We shall not attempt anything like a history or a description 'of
Hissarlik, because Dr. Schliemann, in his exhaustive work liios^
has already given to the world an account of the site and of his
indefatigable labors upon it. These enormous trenches will for
ages be a monument to Dr. Schliemann's energy and perseverance.
We have found his descriptions of all portions of the Troad most
accurate and complete, and his thorough familiarity with the Iliad
leaves but scanty gleanings for those who follow him.

The only rest for the eye amid the desolation of Hissarlik is in
the steadfast line of Greek wall along the top of the trenches.
Striking architectural fragments from the Hellenic or Roman Ilium
are lying about in the trenches or in the heaps of debris. Every
lover of Greek art must desire that search should be made for the
ruins and the remaining sculptures of the Apollo temple, after find-
ing by chance so magnificent a metope as that of Helios conduct-
ing his four-horse chariot. To allow the earth to accumulate above
the probable resting-place of its fellows, without searching for them,
seems like almost too exclusive devotion to prehistoric discovery.

Whatever opinions may be held about the earlier occupation of
this site, it must be remembered that here, without doubt, stood for

i

li

1

E

J-

c

i

s

E

CKO

Plate V

INVESTIGATIONS AT ASSOS, 1881. 163

*
many centuries the citadel of the Hellenic Ilium. Hither Xerxes and
Alexander came to honor the memory of the heroes of Troy, and
hither the Romans cam^ to shower favors on the people from whom
they were proud to claim descent. Such memories are surely honor
enough for the little hill, whatever be the fate of its legendary claims.*
We repeatedly saw Gargaros from Hissarlik, and it would doubtless
be visible from the " town chief's " doorway if the later accumulations
were entirely removed from the hill. We were not so fortunate as to
get the sunset view of Athos, of which travellers often speak.

THE SCAMANDER.

From various passages in the Iliad it is clear that the Scamander
and a lesser stream, the Simois, united between the city and the shore,
and flowed into the Hellespont. The Scamander, moreover, passed
very near the city walls. The Simois, no doubt, was usually dry, and
the battle often raged in its dusty bed, —

2(fu$€(£, o$i iroXXa fiodypta Koi rpv^oActat
Kamrea-ov €v kovitjo'i, koI ^fiiBioav ytpos ay^pny.^

This is not now the condition of things. The marshes to the north-
east of Hissarlik are indeed drained by a stream, the Doumbrek, which
may do duty for the Simois, but it is met below the city only by an
unimportant stagnant creek called the Kalifatlf AsmAk. The M^ndereh

^ The question, what belief the Greeks of historical times entertained in
regard to the site of the Homeric city, is beset with great difficulties and con-
tradictions. No reader of Dr. Schliemann*s book can rest fully satisfied with
his treatment of the deliberate conclusion of Strabo, the striking rhetorical ex-
pression of Lycurgus, and the ode of Horace, all of whom agree in this state-
ment at least, that Priam's city was left utterly desolate, and never occupied
again. A full and impartial discussion of this question will be found in Pro-
fessor Jebb*s article on the ** Homeric and Hellenic Ilium " in the Journal of
Hellenic Studies for April, 1881.

2 " The Simois, where many shields and helmets

Fell in the dust : and the race of godlike men."

Hiail xii. 22.

The beautiful reminiscence of these lines in Virgil reads as if his text were

different : —

" ubi tot Simois conrepta sub undis

Scuta virum galeasque et fortia corpora volvit"

III.

THE GEOLOGY OF ASSOS.

By J. S. DILLER.

THE topographical isolation of the hill at Assos is apparent
from many points of view in the southern part of the Troad,
and its natural advantages as the site of an ancient fortified city
were very great. Its form may be described, in a general way, as a
truncated cone, the base of which at the eastern and western sides
is drawn out into comparatively unimportant ridges. Upon the south-
ern side it descends very abruptly by several terraces and high cliffs
to the sea. To the northward the slope is more gentle to the river,
which is only 1.5 kilometre from the coast. The river at this
point has an elevation of 100 metres above the sea-level. The
Acropolis of Assos is the highest point south of the Touzia (Satni-
oeis) river between, Coslou-dagh 7 kilometres to the eastward, and
the great plateau about the same distance in the opp>osite direction.
According to the measurements of the present expedition it rises
234 metres above the sea. The low truncated conical form and
the bold cliflFs upon the seaward slope are best seen from the west^
the point from which the view (Plate 6) was taken.

Although the rocks in the vicinity of Assos are of great variety,
yet, with the exception of a conglomerate composed chiefly of marl
and fragments of limestone, they are all trachytes. They are,
however, not all of the same age, nor were they extruded in the
same manner. According to differences in age the various modifi-
cations may be grouped under three principal trachytes, which in
general appearance are quite distinct from one another. For con-
venience of description these trachytes will be named, beginning
with the oldest, the first, second, and third respectively. Besides
the tertiary conglomerate and the three trachytes already men-
tioned, there is also a volcanic conglomerate having a very irregular
distribution, and composed of trachytic fragments. In respect to

INVESTIGA TIONS A T ASSOS, 1881. 167

age it stands between the first and second trachytes. The lime-
stone conglomerate to which reference has been made is older than
the third trachyte and younger than the second, upon which it rests.
These rocks, beginning with the oldest, will be described in the
order of their occurrence.

FIRST TRACHYTE.

This trachyte is one of the most abundant in the immediate
vicinity of Assos, and yet from the fact that it leaves few fragments
upon the surface it appears to be quite rare as compared with that
which forms the Acropolis. It is exposed in two large areas, one
south and the other northwest of the Acropolis, connected by a
narrow band extending across the hill in a southeasterly direction.

The prevailing color of this trachyte is purple, but it is frequently
modified so as to become yellowish or reddish purple, or even brick
red. In the compact and uniform ground-mass are imbedded nu-
merous minute feldspars never exceeding two milUmetres in length,
and generally not half that size. They are either opaque white or
glassy, and never so prominent as to greatly modify the color of
the rock. Some of the feldspars are distinctly striated, but the ma-
jority of them are too small to determine with an ordinary lens.
There are small quantities of variously colored accessory minerals
scattered in the ground-mass, and others which are frequently
found in cavities or crevices. Among the latter hyalite is the most
common, occurring in beautiful botryoidal forms.

Of all the trachytes in this region no other preserves so well the
peculiarities of its surface at the time of eruption. The upper
portion is frequently very cellular and ropy, like that of modem
lava. The cells are of all forms and si?es, but are generally elon-
gated in such a way as to show the direction of motion when the
trachyte was extruded. They are sometimes drawn out in large
curves, indicating the manner in which the molten mass rolled down
the steep slope. A yellowish-colored substance lines many of the
cells, and they decrease in size and number downwards to a distance
of several feet from the surface, where the trachyte becomes very
dense. The direction of motion is frequently indicated also by a
stream-like arrangement of the porphyritic crystals of feldspar. Oc-
casionally there are imperfectly developed joint planes parallel to

1 68 ARCH^OLOGICAL INSTITUTE.

this fluidal structure, and more frequently there is an irregular
columnar structure at right angles to the slope.

The elongated cells and other marks which indicate the former
fluidity of the first trachyte occur in all parts of the area in which
this rock is exposed, and it is important to notice that these lines
of fluidal structure point to the Acropolis as a common source from
which the trachyte has proceeded.

The form of the hill of Assos, taken in connection with the facts we
have just noticed, together with the composition and distribution of
the volcanic conglomerate to be hereafter considered, make it evident
that the site of Assos was once the crater of an ancient volcano,
from which proceeded most of the volcanic rocks in its immediate
vicinity. It is probable that there are other ancient volcanic craters
in the Southern Troad, but as far as the explorations of the present
expedition have extended, the eruptions, excepting those at Assos,
have been through large fissures.

VOLCANIC CONGLOMERATE.

The term conglomerate cannot be properly applied to all of the
rocks considered under this head, for some of them are fine ashes
the separate particles of which cannot be perceived by the unaided
eye. However, the rocks are with few exceptions well-defined
conglomerate, and the exceptions are so intimately associated with
the conglomerate both in origin and distribution, that all must be
considered under one head.

The conglomerate is one of the most varied and by far the most
irregularly distributed formation in the vicinity of Assos. It occurs
chiefly upon the seaward slope in small areas varying greatly in
shape, and rests directly upon the irregular surface of the first
trachyte. The small patches are simply the remains of a once
more or less continuous sheet of fragmental material filling the
depressions in the old trachyte and hanging upon the steep slopes
of the hill.

In its most common constitution the conglomerate consists of
numerous fragments of trachyte of various sizes up to half a metre
in diameter. The light-colored groundmass which generally fills
the interstices is sometimes nearly wanting ; in that case the rock

INVESTIGATIONS AT ASSOS, 1881. 169

consists of reddish and yellowish cinders thrown together entirely
unarranged, in the manner in which they accumulate about the cra-
ters of active volcanoes. The fragments are usually light-colored,
distorted and fitted into one another as if they had fallen and fused
together when in a somewhat plastic state. The scoriaceous variety
of the conglomerate is common in the immediate vicinity of Assos,
and although the conglomerate frequently occurs in other parts of
the Troad, this variety seldom appears. The fine material which
constitutes the groundmass is generally ashes, and varies greatly in
amount, from th6 merest trace in the conglomerate of cinders to a
rock in which it is the sole constituent. The finest ashy materials
are usually quite bright colored, either red or brown, and contain
occasionally a few scattered fragments of black scoria. Sometimes,
although completely uniform in color, it is made up entirely of small
light scoriaceous fragments like some of that at Arthur's Seat, near
Edinburgh, and about the recently extinct volcanic crater near Ro-
landseck, on the Rhine. The fragments in the conglomerate about
the hill of Assos are wholly trachytic, and in all cases where it has
been possible to identify them they have belonged to the first
trachyte. Several doubtful fragments of other rocks have been
found in the conglomerate, but from the fact that they cannot be
identified they are relatively unimportant.

Upon the seaward slope near the port is a small area of conglom-
erate, in the light-colored groundmass of which are imbedded
numerous very light, small, cellular, fibrous white fragments. This
rock, although rare at Assos, is of common occurrence among the
stratified deposits of the surrounding country. It varies somewhat
in color, and considerably in the size of its fragments, but is always
light and porous, closely resembling some of the tufa of the Brohlthal,
in Germany. The material of the conglomerate is not rounded and
water worn, but has been thrown together in a manner entirely unlike
the arrangement such materials would assume under the influence
of water.

That the conglomerate is composed of fragments of the first
trachyte and rests directly upon it cannot be doubted, for many
exposures in the cliffs by the sea, where the conglomerate is most
fully developed, show the relation of the two formations very

170 ARCHMOLOGICAL INSTITUTE.

The lower part of the conglomerate, where it rests upon tbe cel-
lular trachyte, is coarse, and composed wholly of cinders. The
amount of fine ashy materials increases in the upper part of tbe
formation until the large fragments entirely disappear, and the rock
is composed wholly of fine ashes.

There are several excellent exposures, wtiich, besides showing the
conglomerate resting upon the first trachyte, exhibit small masses
of the latter overlying the former. One of these outcrops upon tbe
seaward slope is represented in the adjoining figure (Fig. 1). The

portions of trachyte which overlie the coarse conglomerate are
always small, — very small indeed, as compared with the underlying
mass.

It is evident from the relation of the conglomerate to the first
trachyte that the eruption of the bulk of tbe latter took place from
the old crater beneath the Acropolis before the formation of the
conglomerate ; and it is equally apparent, from the composition and
distribution of the conglomerate, that it is of volcanic origin, and
was thrown out from the same crater. The ejection of the conglom-
erate, doubtless, followed closely the extrusion of the trachyte, in
fact even before the flowing out of the trachyte had completely
ceased. Moreover, in the earliest part of the eruption of the frag-
ments, a very coarse material was ejected, and finally the volcanic
energ}' spent itself in showers of ashes. It seem.s probable that at
the time of the eruption of the conglomerate the crater was about
as high above the sea level as at present, that is about two hundred
metres, for the conglomerate shows no trace of the arrangement it
would have assumed under the influence of water.

INVESTIGATIONS AT ASSOS, 1881. I 71

SECOND TRACHYTE.

Of all the rocks found in this vicinity there are none of more
general interest than the one we are now about to consider. It is
the celebrated " Sarcophagus Stone " of Assos, and was used not
only for the city walls, but also for nearly all the important build-
ings within them. The temple, with its many sculptured parts, was
built of it upon a bold acropolis of the same rock. The second
trachyte is the most abundant one occurring in the immediate vicin-
ity of Assos, although it is perhaps not the most abundant trachyte
in the Troad as a whole. It forms the Acropolis proper, extending
to the river upon the north and northeast, and to the sea upon the
southeast. Westward from the Acropolis is a large area extending
from the river to the sea, but separated from the Acropolis by a nar-
row band of the first trachyte and conglomerate. Besides the two
large areas already referred to, there is a small one upon the cliffs
by the port, where the rock is much fractured and generally of a
yellowish or greenish color.

The second trachyte is commonly of a gray, light gray, or purplish
gray color, and has prominent porphyritic crystals of feldspar, which
sometimes attain a length of eight millimetres, but usually only half
that size. Some of the large porphyritic crystals are opaque white,
but most of them are clear and glassy, and of the latter a very few
appear to be striated. Among the larger crystals are numberless
small white crystals of feldspar, varying from i to 1.5 millimetre
in length, which, notwithstanding the presence of other minerals,
gives the prevailing light color to the rock. The groundmass, which
is usually only a small portion of the whole, is gray or purplish
gray ; it has apparently a fine granular porous structiire, and the
porphyritic crystals are so numerous and irregular that the fracture
of the rock is uneven. The whole aspect of the formation is quite
granitic, and this resemblance is increased by its containing a vari-
able quantity of small crystals of mica and other iron-bearing min-
erals, the alteration of which sometimes produces small pits and
stains.

The second trachyte has two well-developed sets of joint planes,
which have determined the development of the peculiar topographi-

172 ARCHAEOLOGICAL INSTITUTE.

cal features of the Acropolis. One set of planes is for the most
part approximately horizontal, and the other nearly vertical. The
former divides the rock into distinct layers, and thus gives rise to
the small terraces and steps so common about the upper portion
of the Acropolis. The layers into which the formation is sepa-
rated vary in thickness from less than ten centimetres to several
metres, and assume the appearance of distinctly bedded rocks of
sedimentary origin. There seems to be some connection between
this peculiar jointing and a certain concealed structure in the mate-
rial. At some places, where the formation is massive and a few
joints are opened, there are upon the weathered surface elevations
and depressions closely resembling those developed in a weathered
sandstone composed of thin layers of different degrees of durability.
It is evident, also, at several localities that the longer axes of the
larger feldspar crystals are not only approximately parallel to one
another, but also to the joint planes.

Although this jointing is seldom exactly in a horizontal plane,
excepting about the southwestern portion of the Acropolis, yet the
deflection is never great, and it is interesting to notice that in gen-
eral the deflection is such as to cause the layers to slope away from
the Acropolis. Although the parallel arrangement of the cr)'stals
is not a very common or prominent character, and the quaquaver-
sal dip of the layers not without exceptions, yet they are sufli-
ciently marked to suggest some connection between the jointing
and the direction of motion of the trachytic lava at the time of
its eruption.

Besides the joitit planes already referred to, there is another set
nearly vertical. Where these joints are few, they divide the rock
into larg^ blocks ; but where abundant, irregular columns are pro-
duced. The columnar structure is best developed in the bold cliffs
of the Acropolis, facing the sea, but there is no approximation to
the regular columnar structure so prominent in the trachyte of
Wolkenberg, in the Seven Mountains. The cliffs are well shown in
the view of the Acropolis from the west, Plate 6.

The jointing results nt many places in strewing the surface with
innumerable massive boulders. In the region west, north, and
northeast of the Acropolis, where the second trachyte occupies
large areas, the surface is completely covered with large fragments
and ledges overgrown with dwarf oaks.

INVESTIGATIONS AT ASSOS, 1881. 173

Occasionally, where this trachyte is in contact with older rocks,
instead of separating into large blocks, as is usually the case, it
breaks into many small angular fragments, and appears like a mass
of breccia. Several small areas of this formation are very decep-
tive, on account of the fact that where considerable decomposition
has taken place along the many small fractures, the rock closely
resembles a conglomerate with subangular pebbles.

Although this trachyte is for the most part considerably altered,
it generally preserves its appearance of durability. In rare in-
stances, however, it is altered almost to a white micaceous clay, and
at other times disintegrates, forming a grayish micaceous sand.

The relation of the second trachyte to the first is made evident
by a number of facts. It contains distinct fragments of the first
trachyte, which must have been picked up by the second at the time
of its eruption. These pieces are not numerous, but yet they are of
such a character as to leave no doubt concerning their identity and
signification. Small portions of oth^r rocks are quite frequently
enveloped by the second trachyte, especially near its junction with
older formations, and some of these fragments are very inter-
esting.

It is evident that by the erosion of the second trachyte a consid-
erable portion of the first trachyte has been brought to the surface.
Southwest of the Acropolis is a narrow band of the first trachyte
extending northwest across the hill, and separating the two large
areas of the second trachyte. This belt lies upon a steep slope
directly beneath the high cliffs of the Acropolis, and there is abun-
dant reason in the structure and topographical relations for believ-
ing that the trachyte of the Acropolis was once connected with that
of the large area to the westward.

Beneath the cliffs of second trachyte, a short distance southwest
of the Acropolis, a long tongue of the first trachyte extends far to
the northwest, and there can be no doubt that this area also has
been exposed by the wearing away of the overlying formation.

That the second trachyte is of more recent eruption than the first
is made evident, also, by their relation to the volcanic conglomerate.
At the western base of the Acropolis, the trachyte of which it is com-
posed rests directly upon the ashes associated with the volcanic con-
glomerate. Near the port the small mass of second trachyte plainly

174 ARCH^OLOGICAL INSTITUTE.

overlies the coarse conglomerate, composed wholly of fragments of
the first trachyte. This enables us to understand why the conglom-
erate is associated with Uie first trachyte only ; it reposes upon the
first trachyte, and is covered by the second.

It is evident, therefore, notwithstanding the fact that the line of
contact between the first and second trachytes is not exposed, that
their relative age is fully established by other phenomena. It is
perhaps well to notice here that in the Troad the lines of contact
between two eruptive rocks, or between one which is eruptive and
another of sedimentary origin, are rarely exposed. They are always
lines of weakness, and the adjoining rocks are so disintegrated as to
afford little evidence concerning their relative age. It is different,
however, when the rocks are metamorphosed, for then the lines of
contact frequently become durable.

The second trachyte is on the whole uniform, and beyond an
occasional streamlike arrangement of the crystals does not show a
prominent fiuidal structure. Its topographical relations, however,
leave no doubt as to the point from which it proceeded. It slopes
away in all directions from the Acropolis, and the imperfect colum-
nar structure has a corresponding inclination. The thickness of this
trachyte varies greatly, no doubt, but in some places upon the slope
north of the Acropolis it certainly reaches thirty metres. In the
Acropolis the trachyte rises about twenty metres above the top of
the old crater from which the first trachyte was extruded.

From the fact that the second trachyte in the vicinity of Assos
proceeded from the Acropolis, and that the Acropolis, itself com-
posed of it, rests directly upon the point from ^hich the first and
second trachytes must have issued, it appears that when the eruption
of the second trachyte ended the crater was completely closed,
and since then the volcano has been extinct. A somewhat similar
example may be seen at Arthur's Seat, near Edinburgh.

That there was not a great interval between the eruption of the
first and second trachytes is made evident by the fact that much of
the scoria upon the surface of the flow was not removed from a
steep slope by erosion before the extrusion of the second trachyte
occurred. The closing of the vent by the second trachyte enables
us to understand why it was not succeeded, as was the first, by a
volcanic conglomerate.

INVESTIGATIONS AT ASSOS, 1S81. 175

The second trachyte was the only one used for making sarcophagi,
or having any connection whatever with the burial of bodies at
Assos. It seems most probable, therefore, that it was the stone
known in antiquity as the "Lapis Assius," or " Sarcophagus Stone."
It was reputed to be a good medicine for certain diseases, and to
have the peculiar property of consuming within forty days the bodies
buried in it. It is impossible to conceive how it came to be con-
sidered as having such wonderful properties.

It has been supposed that the trachyte, being an eruptive rock,
was in those ancient days still highly heated. But it is evident from
the rocks associated with the second trachyte that since its eruption
it must have been long beneath the sea, and subsequently long
exposed above the sea before the region was inhabited by man, so
that there is no probability whatever that the sarcophagus stone was
still hot within the historical period. The geological changes which
have taken place upon the hill of Assos since the founding of the
Greek city, nearly 3,000 years ago, are entirely inappreciable when
compared with the great changes which took place in the long
period between the eruption of the second trachyte and the habita-
tion of the site by man.

The second trachyte is an excellent building stone, and nearly
all the important edifices within the city were constructed of it.
It is not only very durable, but even when altered it preserves its
original shape with remarkable distinctness. Unlike many other
rocks, it rarely crumbles upon the surface, and yet its coarseness
unfits it for the sculpturing of delicate forms. Its warm gray color
compares favorably with the dull-colored sandstones so commonly
used for buildings in America. The only other stones used at Assos
for building besides marble were a few blocks of conglomerate in the
theatre and of the first trachyte for wall filling.

MIDDLE TERTIARY.

A short distance east of the Acropolis is a small exposure of
rocks, which in the immediate vicinity of Assos are very poorly
represented. Elsewhere along the southern coast of the Troad they
are extensively developed, and will be more fully considered in the
second part of this Report.

176 archjEological institute.

The formation is chiefly an incoherent conglomerate, consisting
for the most part of light-colored fragments of limestone. These
are imbedded with more or less of the first trachyte in a whitish
marly groundmass, which is sometimes free from pebbles, and
appears like a soft sandstone. Some of the calcareous fragments
are very hard and heavy ; most of them, including a few pebbles
from metamorphic rocks, are subangular, varying in size up to
twenty centimetres in diameter. The thickness of the whole mass
is not over five metres, and it is about 225 metres above the sea
level.

The best exposures are at the east end of the Turkish cemetery,
where the formation appears to lie upon the second trachyte. These
deposits contain no good evidence of their age, but they are closely
connected with others further eastward, the relations of which to
the other rocks are easily determined. The conglomerate at the
cemetery is not distinctly stratified, but the same formation near by
is plainly arranged in strata. We may therefore feel sure that the
deposit was made under the influence of water.

According to the researches of Tchihatcheff, the sedimentary de-
posits, a part of which we are considering, were placed provisionally
in the middle tertiary, and thought to be of fresh-water origin. But
few fossils have been found in this formation, yet it is hoped that
those secured by the present Expedition, in connection with some
already collected by others, may be sufficient to determine the age
of the formation more definitely. It is the upper portion of the
middle tertiary that rests upon the second trachyte at the Turkish
cemetery ; and it appears probable, from facts which will be here-
after mentioned, that the first and second trachytes were extra-
vasated shortly before the close of the middle tertiary period. The
disturbance at the time of the eruption of these trachytes did not
result in unconformability between the different members of the
formation. It was at the close of the period in which the great
masses of the third trachyte were extruded, that the whole of
the Southern Troad was raised above the sea.

INVESTIGATIONS AT ASSOS, 188U 177

THIRD TRACHYTE.

The third trachyte, which appears to be the prevailing rock in the
southern part of the Troad, is represented in the immediate vicinity
of Assos by. an area southeast of the Acropolis so small that it
scarcely appears upon the map.

It is usually dense, and of a reddish or purplish-brown color.
The groundmass, as in the first trachyte, forms the greater portion
of the rock. In it are imbedded numerous small crystals of feld-
spar, many of which are glassy, while others are opaque white and
irregular in outline. A few small flakes of mica are scattered
throughout the rock, and apparently also a few grains of quartz.
The formation is frequently cellular, but not because of the expan-
sion of gases, as in the first trachyte. The cells are elongated and
irregular in outline, having rough surfaces, as if produced either by
the decomposition of minerals or by the flowing of the mass at the
lime of its extrusion. These cavities are frequently of considerable
size, especially where the trachyte contains many fragments arranged
parallel to a well-marked fluid al structure. Associated with this
trachyte is a very interesting glassy rock, containing more or less of
a black substance quite like obsidian in its general aspect, but dull,
softer, and breaking easily into small pieces. Occasionally the
formation is almost wholly composed of this vitreous material, con-
taining opaque white crystals arranged in parallel lines.

The relation of the second and third trachytes is not so readily
determined as that of the first and second. The superposition of
the third trachyte upon the second was clearly seen at a locality

Fig. 2.

I. Third Trachyte. II. Second Trachyte.

about one kilometre east of Assos. At this place the fluidal struc-
ture of the third trachyte is well developed. The annexed figure

12

178 ARCHAEOLOGICAL INSTITUTE.

(Fig. 2) illustrates what may be seen in the locality mentioned. The
fluidal structure in the third trachyte is represented by the short
lines. This rock appears to have been once continuous across the
depression in which the second trachyte is exposed. The line of
contact could not be found even after several hours' digging in the
disintegrated rocks.

The relation of the two trachytes to each other is, however, more
certainly indicated by their relation to the middle tertiary deposits
of the Southern Troad. The second trachyte, as already noted, is
older than the latter portion of the middle tertiary formation, while
on the other hand, a short distance east of Assos, the third trachyte
distinctly overlies the same deposits, and must, consequently, be of
more recent origin.

This trachyte, when developed so as to influence the topography,
gives rise to surface features very different from those of the other
trachytes. Looking east from Assos, several low, rather irregular
ridges will be seen extending in an easterly and westerly direction.
In form these ridges closely resemble the trap ridges of the Con-
necticut Valley, being very steep, with cliffs facing the sea, while to
the northward the slopes are gentle. These ridges are formed of the
third trachyte, which, like the trap rock of the Connecticut Valley,
has been extruded through great fissures between the strata.

ALLUVIUM.

The Touzla River, north of Assos, flows in an alluvial plain, about
five kilometres in length by two kilometres in greatest breadth.
The soil is fertile, and generally cultivated. By the river bank
the brownish sandy loam extends to a depth of one metre and a
half, and rests upon a bed of gravel on a level with the present
bed of the river. The loam contains numerous very small Gastero-
pod shells, and is exposed upon the surface of the greater portion
of the plain. The latter does not rise more than about two metres
above the present bed of the Touzla.

SUMMARY.

In summarizing what is known of the geology of Assos and
its vicinity, it may be stated that, as compared with some portions

INVESTIGATIONS AT ASSOS, 1881. 179

of the Troad, the formations are quite recent. It seems probable
from facts which will be mentioned hereafter that the oldest rocks
found at Assos were formed towards the close of the middle tertiary
period. The hill of Assos was then a volcano. From its crater
issued the first trachyte upon the irregular scoriaceous surface, on
which succeeding showers of cinders and ashes were deposited. It
seems probable from the general appearance of the conglomerate
and the absence of stratification, that the volcano was sub-aerial,
rising at least to a height of two hundred metres above the sea
level.

The eruption of the first trachyte and conglomerate was followed
after a comparatively short interval by another eruption, which
brought to the surface the second trachyte and completely closed
the crater. There appears to have been no great eruption of gases
connected with the extrusion of the second trachyte ; since this was
brought to the surface the volcano at Assos has been inactive,
although later eruptions have occurred in the neighborhood.

During the latter part of the middle tertiary period the extinct
volcano was almost, if not altogether, submerged. At the close
of the period an upheaval took place by which the southern part of
the Troad was raised perhaps to its present elevation.

Atmospheric agents have since been active in tearing down the
formations, and the topographical features resulting from the ero-
sion are those previously determined by the peculiar structure of the
rocks. By a long process the deep valley, the plain of the river, the
high cliffs, the terraces and the steep slopes of the hill were formed,
until finally the present surface was developed and the foundations
of Assos were laid.

IV.

NOTES UPON THE GEOLOGY OF

THE TROAD.

By J. S. DILLER.

AMONG the numerous works written upon the Troad, there are
but few which consider its geology. Of these, the oldest is
that by P. Barker Webb, first published in the Bibliothcca Ita/iana,
but better known in its French translation as Topographs de la
Troaile issued in 1844.

The most important work is that of Tchihatcheff, who travelled
through the Troad in 1847 ^"^ I849, ^"^ ^ ^^^ years later pub-
lished a series of volumes upon Asia Minor. Four of this series are
devoted exclusively to geology and palaeontology.

Among the more recent contributions is Virchow's Beitrdge zur
Landeskunde dcr Troas, an excellent paper upon the Anterior Troad,
especially upon the Plain of Troy.^

Unfortunately the present Report is written under such circum-
stances that the writer is unable to consult the geological literature
upon the Troad, or to compare the collections of rocks and fossils
made by the Expedition with those already identified.

The following notes are based upon observations made in ex-
cursions from Behr^m (Assos). All the region embraced within
a four hours* journey from that place has been quite thoroughly
explored, but elsewhere the boundaries of the various formations
have not been fully determined.

The general map of the Troad, as well as the geological map of
the same region, both of which are in course of preparation, are not
yet ready for publication. In these notes reference will be made to
Mr. Clarke's sketch map of i^olic Mysia and Lesbos, Plate 4'*.

The rocks of the Troad are of many varieties, and their relations
so complicated that the distribution of them is very irregular, and

^ Sec references to these works in the preceding Report, pp. 8, 14.

INVESTIGATIONS AT ASSOS, 1881. i8l

requires for its determination a great amount of labor. Among the
formations of sedimentary origin are those which have been highly
metamorphosed, as well as unaltered rocks in various stages of dis-
location, and others also which have suffered no change whatever
since their deposition.

The eruptive rocks are of yet greater variety, embracing serpen-
tines, basalts, trachytes, granites, and also conglomerates of volcanic
origin. Before the relations of these formations can be conven-
iently described, it will be necessary to consider a few of the leading
features in the topography of the Troad.

TOPOGRAPHY OF THE TROAD.

The rivers of the Troad may be considered in four groups. The
first embraces the M^ndereh (Scamander) and all its ramifica-
tions ; the second includes the small rivers which carry the water
from the western slope into the -^gean ; the third, or Touzla
group, drains a long, narrow area south of the M^ndereh ; and the
fourth conveys the water of the southern slope into the Gulf of
Adramyttion.

Of these groups, that of the M^ndereh is the largest and by far
the most important. It drains the whole pf the central part of the
Troad, and gathers nearly as much water as all the other rivers
combined. As it touches one side of all the divides which deter-
mine the other groups, its gathering ground has a more or less
circular outline, and is surrounded upon all sides by rugged moun-
tains, through which the river breaks its way to reach the sea. This
topographical arrangement naturally divides the river basin into two
parts : a great central portion, including the large area washed by
the principal tributaries of the M^ndereh, and a portion along the
coast, separated from the other by the mountains through which
the river has cut its way towards the Hellespont. Each part is
distinct from the other, and contains a great plain. The beautiful
Plain of Troy, having a length of fourteen kilometres and a width
varying from three to five kilometres, extends from Koum Kaleh,
near the site of ancient Sigeion, to the mouth of the Thymbrios.
Between the Trojan Plain and Ednedeh, which occupies the site of
Scamandria, the river passes through a deep gorge cut in the meta-

1 82 ARCHACOLOGIC^ INSTITUTE.

raorphic rocks. This defile is picturesque, especially in the portion
nearest Bundrbashi, where its steep sides have many cWSts of
gray crystalline limestone. Towards Ednedeh the lower and more
gently sloping hills are composed of serpentine and trachyte.
From several kilometres below Eanedeh to beyond Beiramitcb,
near Curshunlou-tepeh, the site of Kebrene, the valley of the Men-
dereh has an extensive (Samonian) plain. It is long, comparatively
narrow, and bordered, especially upon the south, by low undulating
hills, which from a distance appear to be a part of the plain itself.
An excellent view of this region, and in fact of the whole Troad,
may be obtained from Chigri-dagh, upon the summit of which are
the extensive ruins of Neandreia. From all sides of this large
plain the tributary streams enter the M^ndereh. The largest of
these flows in from the south at Ednedeh and is separated from
the Touzla by a low divide, upon the southern side of which the
flourishing village of Ivadjlk is situated. Most of the tributaries
during the latter part of the summer are completely dry, and the
M^ndereh itself is reduced to a mere brook, which sometimes
wholly disappears in the limestone gorge below Ednedeh. It is in
the fountain head of Mount Ida that the persistent streams arise,
and were it not for the water supply of that mountain all the rivers
of the Troad would disappear during the dry season. All of the
brooks along the western coast and the southern coast as far east
as Chfpuee, about five kilometres southeast of the ruins of Gdrgara,
are without water during a large part of the year. Further east-
ward, however, the small streams are full of clear cold water from
the slopes of Caz-dagh, and furnish excellent facilities for irrigating
the great olive forests of that region.

The Touzla River, anciently known as the Satnioeis, has a quite
remarkable valley, in which are found three alluvial plains. All of
these, excepting the Halesian Plain at its mouth, are smaller than
those of the Mendereh. The river itself is peculiar in flowing for
many miles nearly parallel with the southern coast, which, in the
vicinity of Behrkm, it approaches within 1.5 kilometre. Of its
source in the western portion of the Mount Ida range very little is
known. After flowing for some distance between high rugged
mountains, the river enters the plain of Ivadjlk, which is northeast
of the site of Lamponeia, upon Coslou-dagh. This plain is long.

INVESTIGATIONS AT ASSOS, 1881. 183

narrow, and fertile. Along the northwestern base of Coslou-dagh
the river flows through a deep gorge. The pinnacled slopes of
coarse angular conglomerate at this place give a peculiarly wild
aspect to the scenery. The river then enters the broad fertile
plain from which the ancient Assians derived their supplies ; turn-
ing northwestward, it passes another deep defile, about eight kilo-
metres in length, before reaching the great Halesian Plain of the
western coast.

Judging from the distribution of the streams, one would naturally
suppose that there was but little system in the arrangement of the
mountains of the Troad. This impression is only heightened by a
casual study of these highlands, but when their geological structure
is fully known, they will be found to be a closely related and ex-
tremely interesting group, the diversity in the arrangement of which
is due to differences in structure and origin.

Mount Ida, or Caz-dagh (Goose Mountain), as it is known to
the Turks, is the chief mountain of the peninsula, and reaches a
considerable height above the timber line. Viewed from the great
Plain of Edremit, it appears to be a low cone upon a small but lofty
plateau. Such is apparently the case from other positions, for the
present summit is only a small portion of the rim of a great dome
which once formed the top of that grand mountain. The arrange-
ment of the spurs and ridges connected with Caz-dagh is peculiar,
and can be fully understood only when the geological structure of
that group is better known. It is certain, however, that none of the
parts which properly belong to Mount Ida extend beyond the great
Plain of Beiramitch, or further west than Dlkeleh-dagh, upon a spur
of which (Cojaykia-dagh) are situated the remains of ancient
Girgara.

The divide between the valley of the Touzla and that of the
Bahchahle^, which is the largest tributary of the M^ndereh, is low,
and the topography so misleading that the position of Ivadjik, the
largest town in the southern part of the Troad, is, upon most maps,
incorrectly represented. The watershed south of the one just men-
tioned, separating the valley of the Touzla from the sea, between the
sites of Gargara and Lamponeia, is comparatively low and broken,
thus completing the semicircle of plains and low hills which mark
the topographical as well as the geological limits of Mount Ida.

184 archjEOlogical institute.

The long, narrow, mountainous belt separating the Touzla from
the Gulf of Adramyttion upon the south, has many varied and
interesting features. The eastern portion of the southern coast is
bordered by a long, narrow, fertile plain at the foot of Caz-dagh,
the many fountains of which furnish abundant water for irrigating
the extensive olive-groves. Further westward, in the vicinity of
Gdrgara (from Sazle^ to Addtepeh), the plain is displaced by bold
cliffs and deep ravines facing the sea.

The extensive walls of Lamponeia are upon Coslou-dagh, the
form of which furnishes a connecting link between that of the
great plateau west of Behrim and the small sharp ridges further
eastward. The plateau which- ends in the bold promontory at Bab4-
calessi (Lecton) is separated from Coslou-dagh by lowlands out of
which rises the imposing Acropolis at Behrim.

Upon the western coast, north of the mouth of the Touzla, is a
narrow, undulating plain, widening to the northward, and covered
for the most part by extensive forests of valonea oak. From the
lower portion of the Touzla Valley towards the site of Neandreia,
the whole country is elevated, supporting numerous peaks, and de-
scending upon all sides abruptly. The height decreases somewhat
to the northward, until the prominent serrated ridge of Chigri-dagh
is reached, while upon the western coast the bold limestone cliffs of
Sac4r-kyah form the most noticeable geographical feature in that
part of the Troad. Further northward the rounded hills decrease
in size, Cdrah-dagh alone rising to a considerable height above the
Trojan Plain.

METAMORPHIC ROCKS.

*

The metamorphic rocks are widely distributed in the Troad, and
have been found to occur in six distinct localities. Some of the
areas occupied by them are very small. This is especially true of
one at Lldjah, near the western coast, and two in the southern part
of the Troad, within nine kilometres of Behr^m, Out of the fourth
and somewhat larger tract rises the prominent summit of Sacir-
kyah, the high cliffs of which, facing the yEgcan, may be seen from
all points along the coast. The fifth is more interesting and exten-
sive ; it occurs in the hills north of Chigri-dagh, includes the rocks

INVESTIGATIONS AT ASSOS, 1881. 185

of C^rah-dagh, and crosses the M^ndereh between the Trojan Plain
and the Plain of Eanedeh and Beiramitch.

The small patch of metamorphic rocks about nine kilometres
north-northeast from Behrim consist chiefly of massive crystalline
limestone, usually white. It forms the cliffs of a gorge along the
small stream flowing from Ealesfdhkee into the Touzla, and is asso-
ciated with mica schist, a portion of which is quite calcareous.
There are at least sixty metres of limestone overlain by the schist,
dipping 11° in an easterly direction. These are in turn surmounted
by the tertiary conglomerate, containing many fragments of the
strata upon which it reposes.

Northwest of Behr^m about nine kilometres, near Golfdl, a small
exposure of metamorphic limestone and schists occurs in the Valley
of the Touzla. This locality is encircled by mountains of trachyte.
Upon the right bank of the stream, by the road from Behr^m to
Golfil, rises a hill composed chiefly of schists. A light-colored
quartzose and ferruginous mica schist overlies massive gray crys-
talline limestone, which upon its weathered surface is very irregular.
The strike of the schist is S. 70° E., its dip 30° northerly, and the
thickness of the mass about sixty metres. In the lower part of the
hill it varies from a light to a bright green color, frequently has an
unctuous feel, and consists of soft, flexible, but inelastic laminae.
The chloritic and talcose schists overlie limestone and quartzite,
both of which have occasionally a well-marked schistose structure.

The area about Sacdr-kyah, near the western coast, a short dis-
tance northeast of the site of Larissa, contains a very thick, massive
limestone, which forms the bold cliffs of the mountain. Associated
with this are thinner crystalline limestones, interstratified with greatly
disturbed schists. These are well exposed west of Sacdr-kyah, on
the road from the village of Tavacle^ down to the sea-coast. The
path from the base of the mountain to Kioiiseh-der^ssee crosses a
ridge of limestone, and affords one of the finest views to be obtained
along the i^gean. Near Eski Stamboul, in the Lldjah Valley, is
a small exposure of highly contorted schists, from which issue the
several hot springs of that locality.

In the vicinity of Cdrah-dagh the metamorphic rocks occupy a
large territory, extending from the rugged peaks near the base of
Chigri-dagh, northeast across the M^ndereh, towards the Sea of

1 86 archjEological institute.

Marmora. The strata of that region are greatly disturbed, highly
altered, and intimately associated with old eruptive rocks, so that it
is very difficult to determine their exact boundaries. The road
from Eanedeh to Eski Stamboul, passing through the flourishing
villages of BurgAz and Yayicle^, crosses the formation near its
southern limit. About two kilometres east of fiurgdz the rocks
and soil are bright red and yellow, while a short distance further
west the gray limestone forms a fertile tract covered with valonea
oak. Near 'the village the limestones and schists are greatly dis-
turbed by intrusive granite. Upon the road towards Yayicle^, after
passing over a small area of rocks which probably belong to the
tertiary formation, the vertical schists again appear, and continue to
the outskirts of the village.

The deposits in which the deep gorge of the M^ndereh, south of
the Trojan Plain, has been cut, belong to the metamorphic group.
Between Ednedeh and Bunirbashi, after following the river for three
kilometres, the path turns to the west over comparatively low round
hills of trachyte and serpentine, then, returning to the river, enters
the defile in the massive gray crystalline limestone which continues
to the plain of Troy. Near Bunirbashi it forms Mount Daydeh and
Bali-dagh, the latter of which is supposed by some to be the site of
ancient Troy. The limestone occasionally contains a great deal of
quartz, in cavities and veins penetrating the rock in all directions.
In the upper portion of the valley of the Kemar (Thymbrius) River
are good sections of the metamorphic rocks, showing a dark mica
schist and a light greenish schist, probably chloritic, interstratified
with large layers of limestone occurring in frequent alternations
throughout a great thickness.

Of all the areas of metamorphic rocks in the Troad there are
none larger or more interesting, at least topographically, than that
of Mount Ida. The altered strata of that locality first appear along
the southern coast in a deep ravine betweeg Moussooradle^ and
Aracle^, where the greenish schist lies beneath the tertiary forma-
tion. At the head of the ravine, about six kilometres from the sea,
upon the beautiful limestone summit of Cojakia-dagh, are the ruined
walls of ancient Girgara. Associated with the gray limestone and
the schists, which in some places are well-marked, evenly bedded
mica schists, is a ferruginous quartzite forming the pointed summit

INVESTIGA TIONS A T ASSOS, 18S1. 187

of Dikele^-dagh. These rocks continue eastward in the high moun-
tains at some distance from the coast to near Edremit, where they
reach the sea. In the vicinity of Papazle6, upon the river of the
same name, they form the impregnable Acropolis on which are the
ruins of Antandros. This Acropolis is an excellent example of
what might be called insular erosion in the formation of valleys.
The two branches of the rapid stream flow for some distance above
their junction in deep parallel gorges. About one kilometre above
their confluence the watershed between the two ravines has broken
down, leaving this wonderful Acropolis completely isolated, and
bounded on all sides by immense cliffs.

From the plain of Edremit the conical summit of Mount Ida
seems to rest upon a very elevated plateau, the southern slope of
which is furrowed by deep ravines and bold spurs descending to
the sea. From Edremit the ascent requires eight hours. The road
at flrst winds across the sandy plain, upon the edge of which are
exposed white, gray, and black crystalline limestone, associated with
various schists. Leaving the beautiful village of Zytinle^, the path
ascends one of the spurs, which is composed at its base of greenish
schist and gray or yellowish limestones. The former is greatly con-
torted, and is the prevailing rock. Its strike is ap{Jkrently at right
angles to the coast, so that the spurs and ravines are parallel to the
general strike of the formation of which they are composed. The
schist upon the southern slope varies from a true mica schist to one
containing a large proportion of hornblende. Occasionally consid-
erable feldspar is present, and produces many small white spots
upon the weathered surface. Smooth surfaces polished by friction
at the time the rocks were dislocated are common. Sometimes the
strata are slightly gneissoid, and their fractures lined with epidote.
The slopes of Mount Ida are covered by extensive pine forests, which
are the chief source of timber in the Troad. The bare rocky top ex-
tends far above the timber line, especially upon the eastern side, and
is known to the Turks as the Chiplak, — a term which is very con-
veniently used when reference is made to the whole of the treeless
upper portion of the mountain. Northeast of the Chiplak, about
the head-waters of the Zylinle^ River, the black hornblende schists
are abundant, and dip away from the summit. The same is true
also upon the northern slope of the mountain, where the beds de-

1 88 ARCHAEOLOGICAL INSTITUTE.

scend towards the liead-waters of the M^ndereh. This arrangement
will be better understood from an examination of the topography and
structure of the Chiplak. Its form is, so to speak, a decapitated dome,
with its highest point. Mount Gargaros, near the northwestern edge.
Once, doubtless, the dome was complete, but now its summit has been
carried away by erosion, and instead of being convex, it is concave,
quite like a volcanic crater. Surrounding the depression upon the
north, east, and south sides, is a rim, which has been broken away
towards the southwest by the head-waters of the Monasteri River.
The stratification is well-marked, and the structure plainly visible.
The upper portion of the Chiplak is composed of three distinct
strata, the lowermost of which is a coarsely crystalline white lime-
stone, weathering light gray and appearing in the depressed centre.
Upon this rests a gneissoid hornblende schist, which forms the
greater portion of the rim. The summit. Mount Gargaros, in the
northwestern part of the broken circumference, is composed chiefly
of talcose schist containing veins of fibrous minerals, and rests upon
the rocks already mentioned. Upon the rim are five peaks, all of
which rise a considerable height above its lowest portions, and may
be reached by a good path from Gargaros in about half an hour.
The view from the Chiplak is extensive, and extremely interesting.
It embraces all of the historic region of the Troad and the adjoin-
ing portions of Europe and Asia Minor. They are spread out at
the feet of the observer as if upon a great map, and more than
repay him for the trouble and fatigue he must endure in order to
reach that celebrated spot. The descent from the rim is not steep
at first upon the east and southeast, but upon the north it is abrupt.
The larger portion of the slope is occupied by a variety of schists,
among which hornblende schist prevails. It is sometimes almost
completely composed of large crystals of hornblende, and is inter-
stratified with actinolite schists and limestones. The latter near
the summit are coarsely crystalline, but further northward in the
great limestone belt they are finer grained. It is from this belt,
which is nearly midway between the top of Gargaros and Evjilir,
that the source of the M^ndereh issues. The limestone forms
very high cliffs, which, owing to the peculiar position of the strata,
appear to have a 'columnar structure. From the base of one of
these cliffs are numerous springs, gushing forth as if the whole

INVESTIGATIONS AT ASSOS, 1881. 189

mountain were filled with water and just beginning to burst. The
rains increase the size of the streams so much, that the cave from
which the main spring issues cannot be examined in all seasons.

The metamorphic rocks continue to near the base of the moun-
tain, where they are replaced in the more gentle slopes by those
which are granitic. The distribution of the strata and their posi-
tion, so far as observed, seem to indicate that although the beds are
sometimes considerably disturbed, Mount Ida is quite a simple anti-
clinal, with a very short axis extending east and west, — so short,
indeed, that its summit in structure is approximately a dome.

TERTIARY.

The tertiary formation in the Troad occurs chiefly along the coasts,
but also in the interior. Many of the areas are small, and they can
be most conveniently considered as parts of two large tracts, one of
which borders upon the Hellespont and the iEgean, while the other
occupies the interior and the shore of the Gulf of Adran^ttion. It
may be that the rocks of these two regions belong to different
periods of deposition, but there can be no doubt that both were
formed during the tertiary age. This subject can be discussed to
better advantage hereafter, when the fossils collected by the present
Expedition have been identified, and the works of other observers in
the Troad can be consulted.

The chief exposure along the southern coast extends from Coslou,
eight kilometres east of Behrkm to the vicinity of Avjildr, which is
not far from the site of^Aspaneus. In the neighborhood of Papa-
zle6 (Antandros) the narrow strip of tertiar}' is interrupted by a con-
siderable mass of granite. Between Coslou and Aracle^, which is
upon the coast south of Girgara, a broad belt of tertiary strata
extends northward across the Valley of the Touzla into that of the
M6ndereh where it expands so as to reach from near Ednedeh to
Beiramitch, a short distance west of the site of Kebrene. This
area is broken across by trachyte upon the watershed between the
Touzla and the Bahchahlee, which is the largest southern tribu-
tary of the M^ndereh.

The most complete section of this formation that may be obtained
at one exposure, occurs up>on the sides of the deep ravine at Aracle^.

igO ARCHAEOLOGICAL INSTITUTE.

The lowest strata of the group are reddish shales and conglomerate,
containing well-rounded pebbles of quartzite and other metamorphic
rocks. Upon these rest thin-bedded greenish sandstones, inter-
stratified with yellowish shales, some of which are calcareous,
altogether having a thickness of about two hundred and forty
metres. They form the lower, most gentle part of the slope, above
which rise the great cliflfs of the overlying massive siliceous lime-
stone. This is usually pale-yellowish colored, soft, light, and porous
as if it had been thoroughly leached. Frequently it contains earthy
black spots or nodules, and occasionally well-defined small crys-
tals. Specimens from some distance beneath the surface effervesce
in acid, but upon the weathered surface the acid is immediately
absorbed without effer\'escence. It is massive, has a thickness of
about one hundred and thirty metres, and forms prominent cliffs,
in which are caves of considerable size. The u|>per strata of the
section, consisting of thin limestones, shales, and tufas, having a
thickness of many metres, are not exposed at Aracle^ but crop out
further westward in the neighborhood of Coslou and Behrkm.

The conglomerate at the base of the series is exposed at a num-
ber of places between Sazle^ and Narle^ Near the latter place,
upon the slope towards Papazle^, it is ven* coarse, composed chiefly
of pebbles of granite, with some from the metamorphic rocks to the
northward. The fragments are all angular or sob-angular, and
appear to hax-e been mo\*ed only a short distance from tbeir source.
In the bottom of a ra\-ine at Ahmijah. it crops out with large
n?unvi pebbles of altered stratx and has a greater tfaickDess than
further east at Araclei6. By the sea, beneath the eie\*ated Tillage
04' Ssul^. ten kiloaetres west of Aracle^. the congioiBerate is not
so cvvirse ; it is as^joclated with a great deal of deep red sandstone ;
reaches i^> swa:est thickness, about one hundred and scventr-
n\-e m^rtres : and in the absence of the massive limcstocie above,
forms a prc^n^ inert ridge. A:l of the pebbles of this detrital fomia-
::orL >n^ far jls^ i: is known, were derive>i fr«?cD the metamoqphic
rocks or the c^Iier erurcn^es^ Frjun^^^ts of mch3rte or
have no: berrn f.N;:nc anvwbere :n the tow^r scrara cc tiie
ujv^n :he <ou:>em 5edbo*iri.

An i^ola^^i vxircron c: r'-e sirara. in :Sf 'o"«er r^irt of the
contra abosX e^t kiJccaeres aonheasi ot B^.ri?a xipCMi the load

INVESTIGATIONS AT ASSOS, 1881, 191

to Ivadjfk. Red clays and thin-bedded yellowish limestone are
associated with sandstone and conglomerate. The last contains
pebbles of quartzite, besides many fragments from the underlying
schists.

The most extensive exposure of the shales and sandstones of the
lower part of the series is between Chipne6 and Addtepeh, upon
the coast, six kilometres southeast of the ruins of Gdrgara. In
that locality the gray, greenish, and yellowish sandstones and shaly
grits form the lower hills, separating the bold limestone cliffs of
Addtepeh from the prominent ridges of the same calcareous stratum
further westward. These beds are greatly disturbed, and are the
source of the hot springs at the Lfdjah (hot baths) of that region.
They crop out also at Narle6 and Avjildr, but have not been
seen further eastward.

The massive limestone near the middle of the series is an inter-
esting and perplexing rock. It so resembles in general appearance
the trachyte, with which it is intimately associated about Chipnee
(south of Gdrgara) and Demearjee-kioy, that special care needs to
be taken in determining its boundaries. It reaches the sea at
Ahmdjah, and continues in detached masses along the coast for
nine kilometres, forming high cliffs separated by profound gorges.
These topographical features are a result determined by the position
of the strata, for each ravine is upon a gentle anticlinal, while the
broad, shallow, synclinal structure preserves the soft limestone
within it. This structure is most plainly seen at Addtepeh, which
is situated upon the narrowest and most completely isolated syn-
clinal. Its short axis extends northeast and southwest, and it
presents bold cliffs to the northwest and the sea. The anticlinal
at its western base is broad, and the strata much more disturbed
than the tertiary strata elsewhere. The axes of the gentle folds in
the tertiary formation of the southern coast are short, and either
nearly at right angles to the general trend of the shore line, or else
extend northeast and southwest. These disturbances are doubtless
accompanied by faults, for upon the coast south of Demearje^-kioy
the massive beds of the conglomerates are found abutting directly
against beds of yellowish limestone in another part of the series.

The upper beds of the series, consisting of thin limestones, sand-
stones, and shales, with tufas and conglomerates made up entirely

ARCHMOLOCrCAL INSTITUTE.

of volcanic debris, are not exposed
east of Sazled The limestones
are usually thin-bedded, yellowish
or gray ; sometimes soft and marly.
They arc the only beds of the
whole scries upon the southern
coast in which fossils have been
found. These fossils, chiefly small
Gasteropods, occur in considerable
numbers at a few localities, but the
range in species is not great. Most
of them have been obtained from
a little exposure upon Coslou-dagh,
about seven kilometres northeast
of Behr^m. The horizontal marly
beds, having a ihickncss of eight
metres, contain numerous large frag-
ments of trachyte, and are complete-
ly surrounde by volcanic rocks.

The small ciilcraps of tertiary
rocks, enveloped by trachyte and
volcanic conglomerate, are numer-
ous in liie souiliern portion of the
Troad, and the relations of the two
formations are for the most part

distinctly ii
r^m

of these i
lowing sec
the relatio
locality.

ndicated. East of Beh-
kilomelres are several
xposures, and the fol-
ction (Fig. 3) represents
»ns of the rocks in iliat
The lowest limestone
^ i (I.) is siliceous and minutely ool it-
^ i ic, containing in its upper por-
^ '^ tion numerous fossils. Small Gas-
^ _; teropods are most abundant, and
- widely distributed in the strata.
The small lamellibranchiate mol-
lusk which is so abundant in the
I g limestone of the Trojan Plain and
-J £ at Eski Stamboul occurs in a thin
$. J layer near the middle of this lime-
^ _■ stone. Adikeof trachyte(II.)5ep-
arates the lowest limestone from
the second (III.), which has a thick-
ness of about sixteen metres. It

INVESTIGATIONS AT ASSOS, 1881. 193

is of a gray color, rather soft and oolitic, containing numerous small
Gasteropod shells. It dips northerly under an angle of twenty de-
grees, the strike being parallel to the general trend of the southern
coast. Over this lies a bed of light-colored tufa and ashes (IV.),
which is succeeded by the second dike of trachyte (V.). The third
limestone (VI.), having a thickness of only two metres, is soft, light-
gray, and marly, containing small Gasteropods, like the ones in the
inferior beds. This calcareous stratum is overlain by at least thirty-
five metres of greenish conglomerate, sandstones, and shales (VII.).
The conglomerate alternates frequently with the sandstone, and
contains numerous cellular and compact fragments, apparently iden-
tical with the first and second trachytes at Behr^m. The upper
bed is a greenish sandstone, upon which reposes a large stratum
of tufa (VIII.), about thirty metres in thickness. It is composed
chiefly of very light, soft, white fibrous fragments, in a light-colored
groundmass, containing also a few small pieces of trachyte. The
tufa at this place shows no evidences of stratification, but elsewhere
similar detritus is definitely arranged. At the top of the section is
a large dike of trachyte (IX.), which in the first part of the present
Report has been designated the third trachyte. The interposed
dikes of trachyte are of the same kind, and both have 'distinct
fluidal structure dipping northerly, parallel with the stratification
in the adjoining rocks. This trachyte is not represented among
the pebbles in the fragmental rocks of the section, — a fact which
indicates that the volcanic rocks are not overflows contemporaneous
with the deposition of the formation in which they occur, but are
subsequent injections after the deposits were complete. The dislo-
cation and distribution of the stratified rocks is incompatible with
any supposition but that which regards them as older than the
eruptive formation with which they are associated.

Further westward the amount of volcanic dcJbris in the sedi-
mentary beds increases. Four kilometres west of Behr^m, by the
sea, is exposed a coarse conglomerate, with a small proportion of
fine detritus, having in all a thickness of at least sixty metres. The
fragments are well rounded ; a few are of compact trachyte ; many
of quartzite and other metamorphic rocks ; but the majority of
limestone, apparently like some of that belonging to the tertiary
formation. This sedimentary deposit appears to be overlain by

194 archjEOLOG/cal institute.

trach3'te, above which crops out a section composed wholly of vol-
canic debris distinctly stratified. The beds consist chiefly of ashes,
usually of a gray color, alternating with layers containing numerous
large round fragments of trachyte, like that beneath. The upper
bed, six metres in thickness, is of reddish-brown and bright-red
ashes, upon which rests a mass of trachyte. A well-defined colum-
nar structure is developed in the bright-red ashes along its junction
with the overlying formation, but the same structure does not
appear in the trachyte. The thickness of the volcanic sediment at
this exposure is at least forty metres. The trachyte occurring near
the middle of this section is apparently the same as that called the
first trachyte in the part of this Report referring to the geology of
Assos, while the one at the top of the section is more closely related
to that of the Acropolis at Behrkm.

Small outcrops of stratified volcanic debris belonging near the
top of the tertiary formation are numerous in the southern part of
the Troad, and show conclusively that the tertiary strata occupied
the whole surface of that region before the great eruption of trachyte
occurred. Many of these exposures are of special interest, but can-
not be noticed without expanding this Report far beyond its proper
limits. Let it be sufficient to mention one more outcrop, which
is remarkable on account of the fossils and lignite which it con-
tains. It is only a few hundred metres from the ^gean shore, near
Point Deviy, about ^vt kilometres east of Bab4-calessi. The
exposure is at the foot of the steep slope and high cliiTs of trachyte
which rise abruptly to the plateau. Half a score of years ago the
locality was explored by means of a horizontal drift, eight metres
long, in the hope of finding valuable coal. The lignite is lean and
earthy upon the surface, but occasionally there are thin laminae of
good quality. Its thickness where greatest is 2.5 metres, but is sub-
ject to sudden variations, and it may be traced along the base of
that clilT for a distance of fifty metres. The associated rock, both
above and below, is gray limestone, containing many fossils, appar-
ently diiTercnt from those found at other localities. A thickness of
more than fifty metres of limestone is exposed ; its general strike is
parallel to the adjacent coast, and it dips northerly about twenty-
five degrees ; but near the basaltic rock and trachyte, both of which
occur in the immediate vicinity, the position of the strata is such as

INVESTIGATIONS AT ASSOS, 1881. 195

to indicate that they were dislocated by the extrusion of the eruptive
rocks.

By the path leading from the old excavation to Baba-calessi
there are excellent exposures of distinctly stratified rocks, com-
posed wholly of volcanic debris and fragments of eruptive for-
mations. These strata are greatly disturbed, being occasionally
nearly vertical. They are evidently older than the trachyte, which
forms the mass of the plateau, and with the limestone and lignite
apparently belong to the same series as the stratified rocks east of
Behrkm.

The distribution of the tertiary about the great plain of the M6n-
dereh between Ednedeh and Beiramitch (near Kebrene) has not
been completely determined. Upon the road from Beiramitch to
Ivadjfk the grayish compact limestone crops out near the former
place, and closely resembles that along the southern coast near
Behr^m not only in general appearance, but also in containing the
same fossils and being very oolitic. -At one locality good specimens
of pisolite were found scattered upon the surface. The limestone
is frequently earthy or marly, and contains small pebbles of other
rocks. The strata are nearly horizontal, and they crop out over a
large territory of low rounded hills and ridges along the side of the
plain southeast of Ednedeh. The tertiary formation in the valley of
the M^ndereh is separated from that in the Touzla Valley and the
southern coast by a mass of trachyte, but within the narrow belt of
this eruptive rock, which is younger than the tertiary strata, there
a»e small exposures of the latter, and there can be no doubt that the
deposits of the two large areas in question were once connected. It
is a general fact, observable throughout the southern portion of the
Troad, that wherever the trachytes are found in contact with the
tertiary beds the latter are considerably disturbed, and it is evident
that the dislocations arc due to the intrusion of the eruptive rocks.

The tertiary bordering upon the Hellespont and the western coast
is very fossiliferous, and in this respect appears to be different from
that which occurs in the interior and along the southern coast. In
the vicinity of the Trojan Plain and the Dardanelles it has been
studied recently by Virchow, Calvert, Neumayr, and others whose
works the writer is unfortunately not able to obtain at this time.

An excellent section of these rocks is exposed in the steep cliffs

196 ARCHAEOLOGICAL INSTITUTE.

facing the Hellespont, just north of Reu-kioy. The metamorphic
and eruptive rocks which limit the tertiary formation south and
southeast of the Trojan Plain form irregular mountains, extending
from Carah-dagh, west of the valley of the Mendereh, north-
east to the Hellespont. At the base of these mountains the tertiary
beds form a low undulating plateau, the strata of which, generally
horizontal, gently rise towards the northeast, until in the neighbor-
hood of Chanac-calessi their dislocation is quite marked. Out of
the horizontal strata of fossiliferous limestone has been cut the
depression occupied by the Trojan Plain, and upon one of the spurs
(Hissarlik) projecting into the plain from the east are the celebrated
ruins of Troy.

Excellent exposures of a part of this series of rocks occur along
the Valley of the Kemar. The lowermost stratum of the group
appears to be a marly conglomerate, containing fragments of
serpentine and other altered rocks. Sometimes it is a quite com-
pact limestone, but generally it is soft and light colored, having a
thickness of about fifteen metres. Upon this horizontal stratum
rests another, composed chiefly of red clay, containing many peb-
bles, but occasionally it is a regular conglomerate of mica-schist frag-
ments mixed with those of other metamorphic rocks. Overlying
these strata upon both sides of the valley is a thick layer of basalt,
which, in the vicinity of the chiftlik of the American Consul (Mr.
Frank Calvert), is itself overlain by red clay and shelly limestone.

At the northeast base of Bali-dagh, near Bundrbashi, the same
calcareous conglomerate which occurs in the Kemar Valley, appears
to rest unconformably upon the crystalline gray limestone. The
soft pebbly bed is composed chiefly of fragments of the limestone
upon which it reposes, but contains also numerous pieces of
serpentine, and is distinctly overlain by basalt.

At the northwestern extremity of the " Forty Eyes,** near Bunar-
bashi, the conglomerate again occurs, and is composed of large
angular fragments of the crystalline limestone, upon which it lies
unconformably. At this locality it is overlain by soft sandy strata.

The marly and sandy horizontal beds which form the prominent
cliffs facing the ^gean at Yeni-share extend southward along
the undulating coast, covered by extensive forests of valonea oak.
The ruins of Eski Stamboul are upon a soft shelly limestone,

INVESTIGA TIONS A T ASSOS, 1881, i g 7

which appears to be connected with that like it about the Trojan
Plain. That the tertiary formation around the Plain of Troy is
connected with that in the vicinity of Eski Stamboul is rendered
very probable, not only by the similarity of the limestones in the
two localities both in general aspect and fossil contents, but also by
the fact that northeast of Eski Stamboul, about seven kilometres in
the neighborhood of Yayicle6, there is a coarse conglomerate, the
horizontal beds of which are composed of granite and crystalline
limestone pebbles, with those of other metamorphic rocks, and rest
directly upon the strata from which they were derived. This con-
glomerate appears to occupy the same position as that at the base
of the tertiary strata in the neighborhood of Bundrbashi.

South of Eski Stamboul one kilometre, the granitic rocks of
Chigri-dagh advance westward and reduce the tertiary to a nar-
row belt by the sea ; but further southward, about the supposed site
of Larissa, it expands and forms a series of flat-topped hills. The
strata are generally horizontal, but sometimes they have a gentle
dip and contain many fossils, among which is a small Ostrea, A
fine exposure of the coarse conglomerate at the base of the terti-
ary beds, as well as the granite and metamorphic rocks from which
it was derived, may be seen upon the road leading from the sea to
Tavaclee, which is situated high upon the slopes of Sacar-kyah.

The tertiary formation, continues along the western coast to within
four kilometres of Baba-calessi. Just north of the mouth of the
Touzla the trachytes advance westward from Touzla-dagh, and
again reduce the tertiary to a mere strip; but south of the low
projecting ridge of trachyte about the great Halesion Plain the
tertiary rocks are extensively developed. Near the sea, opposite
Touzla, the small tertiary ridges extending across the plain are
composed for the most part of very fossiliferous limestone, some
of which is compact, but generally soft and marly. The overlying
limestone consists wholly of finely comminuted shells, and dips
seaward. It has very distinct ripple-marks, with occasional cross-
bedding, and must have been deposited in shallow water. Beneath
this compact limestone the strata are soft, containing numerous
small Gasteropods and other molluscan forms. One stratum is
composed wholly of oyster shells. Lower down in the series occurs
a conglomerate containing many fragments of trachyte, some of

Io8 ARCHAEOLOGICAL INSTITUTE.

which closely resemble the oldest of the three trachytes at BehritnL
The whole section exposed in the plain has a thickness of about
ninety metres.

Upon the eastern edge of the plain, close to the village of
Touzla, in the immediate vicinity of several hot springs, occurs
a remarkably beautiful section of highly-colored strata, composed
almost wholly of volcanic debris. The base of the series of rocks
exposed at this place is a conglomerate of scoriaceous fragments of
trachyte. This is succeeded by frequent alternations of strata con-
taining coarse and fine sediment, which ranges in size from parti-
cles of clay to fragments nearly half a metre in diameter. Many of
the larger pebbles are of a light-colored tufa which occurs in the
neighborhood, and is used for making millstones. The layers have
all varieties of red and yellow color, and present a wonderfully
beautiful as well as unique appearance. They are distinctly folded,
and small faults are of frequent occurrence. These highly-colored
beds have a thickness of about thirty metres, and doubtless owe
their extraordinary appearance to the presence of the hot saline
springs by which they are surrounded.

No fossils have been found in these strata, but their position, as
well as composition, makes it ver}* probable that they belong to the
tertiarw

Upon the road between Kioulacle^ (Chrysa) and Bab4-calessi,
about two kilometres from the former place, the tertiary beds may
be seen in contact with the trachxte. The strata are marly, light
colored, sandy, and pebbly, containing distinct fragments of trachj'te
and metamorphic rocks. Near the sea the beds are horizontal,
and continue in that attitude eastward to the neighborhood of
the trachMe, where thev are suddenlv disturbed and thrown into
a vertical position.

Fi^. 4 is a representation of the structure in that locality.

It is n.n known certainly to what portion of the western coast
tertiary- the strata containinir the trachvte fragments belong. It b
evident, however, that the conglomerate containing these pebbles is
beneath at least sixrv metres of comoact and marlv limestones, in
which are tound many fossils- It cannot be doubted, therefore,
that while some of the tracr.vte is voun^cr than the tertiary of the
western coast, another portion was extracied lor.^ before the close
of that formation.

INVESTIGATIONS AT ASSOS, 1881,

199

In comparing the tertiary strata of the southern coast and the
interior with those bordering upon the Hellespont and the ^gean,
it is to be remarked that there is an apparent difference in the num-
ber and character of their fossils. While the latter may be said to

Fig. 4.

be characterized by the abundance of fossils, among which the most
prominent and numerous are bivalve mollusks, the other appears to
be distinguished by its paucity of organic remains, most of which
are small univalve moUusks. It is probable, however, (hat some of
the species are identical in the two faunae, and that their difference
arises rather from unlike conditions than a want of agreement in
the time of deposition.

There appears to be no essential difference in their relation to
the trachytes. It is evident that while some of the trachytes are
younger than the tertiary rocks of both regions, there are others
older than the upper strata of the scries in each of the two terri-
tories ; however, upon the western coast the trachytic fragments
occur apparently lower down in the series, and the rocks generally
are somewhat less disturbed than those along the coast of the Gulf
of Adramyttion.

Notwithstanding these differences there are some important points
of agreement. In both regions the tertiary beds come in contact
with the metamorphic rocks, and the lower stratum is a conglomerate
derived directly from the altered strata upon which it rests.

The occurrence of lignite near the shores of the Hellespont,* as

* Its occurrence northeast of Lapsake^ has been described by TchihatchcfF.

2CX) ARCHAEOLOGICAL INSTITUTE.

well as along the southern coast east of Babd-calessi, and probably
also in the interior,* indicates that the strata in which it is found in
all the localities mentioned are essentially of the same age.

The rocks of both areas occur at elevations high above the sea
level, and make it evident that a great change in the configuration
of the country has taken place since the period of their deposition.
The distribution of the tertiary rocks shows clearly that they were
formed before the Hellespont existed, and suggests that what is now
the peninsula of the Troad may then have been several islands. It
has been shown by the observations of others that the water in
which the strata were deposited was either fresh or brackish.

ALLUVIUM.

The alluvium of the Troad occurs chiefly in the plains already
noticed in describing the river valleys. Two of the plains are along
the M^ndereh, and of these the Plain of Troy has been fully de-
scribed by Professor Virchow, in his excellent work entitled Beitrdge
zur Landesknnde der Troas.

Of the three along the valley of the Touzla only the Halesian
Plain by the sea is of considerable importance. It is extensive and
fertile, and is nearly divided into two parts by the low ridges of
tertiary several kilometres west of Touzla. The old Roman bridge,
which once spanned the river where it breaks across these ridges,
now stands upon a level plain about two hundred and thirty metres
from the present river bed. The amount of filling around it, by
which the surface was brought up to the general level of the plain,
appears to have been at least two metres. The detritus near the
ancient structure is generally very fine, but contains some gravel,
and is like that upon other portions of the great plain, whose sur-
face is about two metres above the bottom of the Mdndereh.
Were it not for the bridge one would not be likely to suspect that
formerly the river bed had been at that place. It is an interesting
example, sliowing that great changes have occurred within the last
two thousand years.

* Good specimens of lignite were shown to the writer at E4nedeh, and were
said to have been collected within a two-hours* walk from that place ; but their
possessor could not be induced to disclose more definitely the locality of his
treasure.

INVESTIGATIONS AT ASSOS, 1881. 2OI

The changes which have taken place in the Halesian Plain are
recorded in such a way that even the most sceptical cannot doubt
them, and are important when considered in connection with those
said to have occurred in the Plain of Troy. Although the gravel
beds and succession of deep pits containing the stagnant pools of
the Kalifatlf Asmak, together with the well-marked banks of a large
stream, are proofs that the Scamander once flowed close to the foot
of Hissarlik, yet they are not nearly as impressive evidences of recent
changes as the presence, in a level plain, of a large bridge far from
the stream which it once must have spanned.

ERUPTIVE ROCKS.

A large portion of the rocks of the Troad are eruptive, and their
distribution is extremely irregular. The trachytes are by far the
most abundant, and occupy an extensive area towards the bold
promontory of Babd-calessi. Granitic rocks stand next in abun-
dance and importance as topographical determinants, while the
basaltic rocks, and probably also the serpentines, although widely
distributed, do not extend over large districts.

SERPENTINE.

The serpentine of the Troad has been found only in the north-
western portion south of the Trojan Plain in the vicinity of Cardh-
dagh, where it is intimately mixed with the limestones and schists
of the metamorphic series. Upon the road from Ednedeh to
Bundrbashi, about four kilometres from the former, a path turns
to the westward, and after passing several considerable elevations
of conglomerate and trachyte, ascends the low rounded conical hills
of serpentine near the base of Cardh-dagh. The rock is usually of
a deep green color, but varies, becoming bluish or reddish, and
contains small but distinct crystals of a lamellar mineral supposed
to be diallage. It is much stained by oxide of iron, and presents
many fibrous, smooth surfaces like slickensides. Upon a fresh
fracture the rock is usually dull greasy, and occasionally the promi-
nent foliated cr}'stals give it a porphyroid structure. It weathers
reddish brown, and in general has a very ancient aspect. An

202 ARCHjEOLOGICAL INSTITUTE,

imperfect columnar structure is occasionally present, but was not
seen fully developed anywhere ; the rock for the most part being
much fractured and decomposed.

Some good exposures of the serpentine occur along the Kemar
River, about five kilometres beyond the Plain of Troy. At that
locality it is compact, and intimately associated with the schists and
limestones, through which it appears to penetrate in the form of
irregular dikes. However, the rocks are so much disturbed that its
relations are not easily determined. According to Mr. Frank Cal-
vert,* the American Consul at Dardanelles, the serpentine occurs in
distinct dikes, cutting the crystalline limestone.

The age of the serpentine is definitely shown by its relations to
the metamorphic rocks and the tertiary. That it is younger than
the former strata is evident from the fact that it cuts them in the
form of dikes. Its occurrence as pebbles in the conglomerate at
the base of the tertiarj^ series of that region is equally positive
evidence that its eruption took place before the deposition of the
conglomerate commenced.

GRANITIC ROCKS.

The granitic rocks of the Troad are widely distributed, but the
single outcrops are generally small. The largest of them is that
east of Beiramitch, near the head-waters of the Mendereh- Quite
an extensive mass occurs also about Chigri-dagh, the site of Nean-
dreia, and two smaller expvosures may be found along the southern
coast near Papazlee and Avjildr. At the latter locality the rock is
coarsely granitic, consisting chiefly of amphibole and feldspar, with
a smaller but yet considerable proportion of black mica and quartz.
The hornblende occurs well cr\-stallized in forms frequently one
cenriaietre lone, and half as broad. The feldspar, usually well
cr^sraHircd, is occasionally distinctly striated. Fragments of the
mica schis: which occurs in the mountains a short distance north of
this locality are enveloped by the granitic rock, which must there-
fore be more recent than those of ihe metamorphic series.

^ Mr. Fra~V Ci*. ^rt. the An:.v'ca.r. C,^-"<u' j: Pjircjn^!'e?, is rcnr famiUar
w :>. :!^.' ji.*.'-.:^- -^: ;'.e ar.:^::.r rr-\i.u ir..: tv^ hini the writer is izKiebCed for
\juu-i*-lc j:is<;5:xr-oe while eximinin^ the rovik^ c: thit re-^^ioo.

INVESTIGATIONS AT ASSOSy 1881. 203

The granitic rocks in the neighborhood of Papazle6 and Narle6
are like those just east of Avjilir. Both exposures are at the foot
of Mount Ida, and form low rounded hills, whose gentle slopes are
occasionally covered with micaceous sand, resulting from disinte-
gration. Near Narlee the coarse conglomerate at the base of the
tertiary series contains many fragments of the underlying granite, —
a fact which is conclusive evidence that the latter rock was extruded
before the deposition of the tertiary commenced.

Upon the northern side of Mount Ida, between Curshunlou-
tepeh and the source of the M^ndereh, the rocks present a similar
appearance and composition. In the coarsely crystalline portion
hornblende is always abundant, but the amount of mica varies
greatly, being at times apparently absent from the unaltered rock,
while in the weathered portions it is occasionally nearly as abun-
dant as the amphibole. The rocks are generally coarsely crys-
talline, much disintegrated, and contain distinct fragments of
metamorphic schists, but near their contact with the latter they are
finely crystalline, containing quartz, feldspar, and mica in equal
proportions, and apparently no hornblende. The relation of this
fine granite to the coarsely crystalline rock has not been deter-
mined. It occupies a narrow belt upon the gentle slopes at the
foot of Mount Ida, without entering as an essential member into
the mountain structure.

The irregular serrated ridge of Chigri-dagh is composed of a
granitic rock which is not so coarsely crystalline as that of either
of the other districts. It forms the low uneven plateau extending
west and southwest from Chigri-dagh to the heights close by the
sea, where it is limited by a narrow belt of tertiar)'. The rock con-
sists of quartz, feldspar, and mica, with some amphibole and occa-
sionally large prominent crj'stals of feldspar, sometimes attaining a
length of two centimetres and a thickness of five millimetres. It
has evidently been regarded as a trachyte by Tchihatcheff in his
extensive works upon Asia Minor, while by Webb it was considered
as a granite. The rock is completely crystalline, and is usually
quite different from any of the trachytes of the Troad. However,
it is variable, and intimately associated with light-colored compact
rocks, whose relations have not yet been fully determined. Near
Chigri village, and also upon the eastern slope of the mountain

204 ARCHAEOLOGICAL INSTITUTE,

towards EAnedeh, the granitic rocks are penetrated by dikes of a
soft, highly altered, light-colored, slightly porphyritic rock, which
appears to belong to the trachyte. In the vicinity of Eski Stam-
boul the cr}'stalline rock has suffered considerable disintegration,
but is frequently compact, containing few or many porphyritic
feldspars, which appear to have no striations.

North of Chigridagh, in the neighborhood of Burgiz, the granitic
rocks occur as irregular dikes cutting the metamorphic strata, which
are greatly disturbed. The same phenomena may be observed
near Tavaclee (near Larissa), about eight kilometres southwest of
Chigri-dagh.

At the last locality, as well as seven kilometres northeast of Elski
Stamboul, the conglomerate, at the base of the tertiary deposits,
contains numerous fragments of the granitic rocks of that region.
It is evident, therefore, that the rocks of Chigri-dagh are more
recent than those of the metamorphic series, and older than the
tertiary strata along the western coast, and, moreover, it appears
that all of the granitic rocks of the Troad are of the same relative
age.

TRACHYTES.

The trachvtes of the Troad occur chieflv in the southwestern
portion, where they occupy a large area, extending from the south-
em cv^jst between Rihi calessi and Coslou, north across the Valley
of the Tv^uila and the hi^h irregular peaks of Touzla-dagh, Kazik-
da^h, Cavakniagh, and Cxzniagh, lo Einedeh, and the plateau of
grarjitio rocks about Chigri-dagh^ An irregular arm of trachyte
from the lar^i^e mass extends eastward upon the watershed between
:he chief southern branch of the Mendervh and the Touzla, and
fo::v.> :ho low, brcud mountain car.evi Diydeh-dogh. Several small
ocLuheo. areas occur alo:^.g the southern coos: m the neighborhood

of IV.r.car^eeki.n, ChipiHV i^south of Gar^raraV and Kizil-ketchily,

• * •. .

It :> e\ viert :h.-.: .n the \is:i"::v of RehTirn there are at least
three tr,:.h\:eN, v\.::V':r;: rv>t or*v ::> i:x^r.c'"a* a::^>?aranoe but also in
ACe, I: :> r.-t ;vs>:h\\ howi wr, at -orese":. to >e:\arate the various

ers TroacL The) \an i^rcatly :n u^er^::t rsk::> ci ibc region, and

INVESTIGATIONS AT ASSOS, 1881. 205

it is very probable that rocks which are here included under the
trachytes when they are better known will be classed among other
groups.

The trachyte which in The Geology of As so s has been called the
first trachyte, occupies a large portion of the area between Behr^m
and the great plateau further westward, as well as a considerable
district about the base of Coslou-dagh towards the east. Its color
is usually dark-purplish, but varies greatly. The compact uniform
groundmass contains varying quantities of small porphyritic crystals
of feldspar, a few of which have the characteristic striae of plagio-
clase, but orthoclase is by far the most abundant. The ground-
mass usually contains a small quantity of minute scales of mica and
other dark-colored crystals, some of which are probably hornblende.
The upper portion of the trachyte is frequently cellular and scoria-
ceous, like the surface of a modern lava-flow, and can often be
recognized among the pebbles ol the tertiary conglomerate of the
western and southern coasts, — a fact which clearly indicates that
it is one of the oldest trachytes, and yet it occasionally occurs also
in the position of the most recent rocks of Its kind. About four
and a half kilometres northwest ot Behr^m the trachyte distinctly
overlies the ashy beds at the top of the tertiary series, and must be
younger than the beds upon which it reposes.

In the vicinity of Balabahny, upon the plateau directly north of
the site of Polymedion, a trachyte occurs containing numerous
small but distinct crystals of mica and many thin tabular, glassy
crystals of orthoclase, some of which attain a length of eight milli-
metres. The crevices of this rock are often coated with beautifully
colored chalcedony. It is much lighter colored than the first tra-
chyte at Behrkm, and does not appear to have an extensive distribu-
tion. The same trachyte occurs near Baba-calessi, where the crystals
are so small that if plagioclase is present it cannot be recognized
with a hand-lens. A fresh fracture shows only a small quantity of
the accessory minerals, but upon a weathered surface they are more
distinctly seen ; the small black crystals of mica and greenish horn-
blende occasionally give to the rock a peppered appearance

Upon the north side of the Touzla, similar rocks appear near
Gulfdl, about ten kilometres northwest of Behrhm, and extend east-
ward, occupying most of the area immediately north of the river as

206 ARCHJEOLOGICAL INSTITUTE.

far as Ivadjfk. At Pashd-kioy, however, which is directly north of
Behrkm six kilometres, the rock, although similar in its general
appearance to the trachytes already noticed, is essentially diflEerent
Its few porphyritic feldspars are for the most part plainly striated,
and the crystals of hornblende, much more abundant than the min-
ute scales of mica, sometimes attain a length of five millimetres,
and are more prominent upon a fresh fracture than the feldspar.
This grayish rock appears less siliceous than the ordinary trachytes,
and is not abundant in the Troad, although it occurs at intervals as
far north as Chigri-dagh.

The trachyte designated in the first part of this Report as the
second trachyte, has a wide distribution, and appears to cover con-
siderable districts. It extends only a short distance east and west
of Behrim, and is then replaced by other rocks of the same kind.
Comm#nly its color is light gray, with many irregular milk-white
spots, indicating the presence of numerous crystals of feldspar.
These vary greatly in size, appearing in tabular form sometimes ten
millimetres long and eight millimetres in width. The large crys-
tals are comparatively few, but they are surrounded by innumerable
smaller ones, whose limits upon the rough fractured surface of the
rock are not distinctly outlined. Within the groundmass, which is
irregularly cellular, are numerous small crystals of black mica, and
probably a few of hornblende, with small quantities of other acces-
sory minerals. The crystals are so much fractured that the kind of
feldspar is not easily determined. All of the larger ones may be
orthoclase ; the smaller ones, bearing even indistinct striae, are rare.
The granular and porous structure of the groundmass gives to the
rock a rough, angular fracture.

This trachyte does not form any important topographical feature
south of the Touzla, excepting the Acropolis of Assos, at which
place it appears, from facts already presented in the preceding
paper, to have been extruded from a volcano before the depo-
sition of the tertiary strata of the southern coast was completed.
There is evidence also, but as yet not conclusive, that, at another
place three kilometres west of Behr^m, this trachyte came up in the
form of a dike and overflowed the ashy strata at the top of the
tertiary.

Among the high mountains north of the Touzla this trachyte

INVESTIGATIONS AT ASSOS, 1881, 207

forms Cavak-dagh and Kazik-dagh. It is of a pale-reddish color,
with numerous orthoclase feldspar of less size than those in the
Acropolis trachyte at Assos. Further north, near the plateau of
granitic rocks about Chigri-dagh, the color is gray, and not so
coarsely granular as that in the southern portion of the Troad.

It is in the neighborhood of Ednedeh, however, that this trachyte
has its most pronounced form. There the tabular crystals of ortho-
clase are large', frequently sixteen millimetres long and fourteen
millimetres wide. They are usually clear and glassy, and are sur-
rounded by a granular gray ground mass, containing innumerable
small white feldspars, apparently orthoclase, besides small quanti-
ties of mica and hornblende.

It should be remarked that the determination of the kind of
feldspar, by means of a small lens, is in most cases very unsatisfac-
tory, for the crystals are generally small and much fractured, so that
the presence or absence of the characteristic striae is not easily
discovered. It is certain, however, that the large crystals of this
trachyte are orthoclase, and that some of the crystals in the
trachytes already noticed are plagioclase.

The trachyte near Einedeh containing the large crystals of ortho-
clase closely resembles in general appearance the Drachenfels
trachyte in the Seven Mountains, upon the Rhine, while that already
described as occurring at Pashi-kioy appears like the trachyte of
Wolkenberg in the same region. The prominent orthoclase crystals
are frequently arranged so that their tabular surfaces are approxi-
mately parallel, — a phenomenon which has been noticed in the
trachyte at Behr^m also, but in neither case is it true for the greater
part of the rock.

The trachyte named in the first part of this Report the third
trachyte, is extensively developed south of the Touzla, but does
not reach far to the northward. The groundmass is usually
brownish or reddish-brown, and contains, besides minute flakes of
•mica and small grains of quartz, numerous crystals of feldspar, a por-
tion of which appear to be orthoclase, but are generally too small
to be determined with a pocket-lens. Although the rock is some-
times compact, it is generally more or less cellular between the
irregular laminae which mark the fiuidal structure. The laminae are
occasionally drawn out so as to produce distinct bands of different

2o8 ARCHJSOLOGICAL INSTITUTE.

colors continuous for a metre or more, such as may be seen in the
felsites of Marblehead Neck, north of Boston. The fluidal structure
usually consists of a streamlike arrangement of the small porphyritic
crystals and pebbles, as well as the elongated irregular cells, and
small darker and lighter portions of the ground mass.

At the base of the dikes of this trachyte, especially where it lies
upon fragmental rocks, is commonly found a pebbly rock containing
more or less of a soft, black, brittle vitreous substance, which is
usually arranged in elongated parallel patches corresponding in
position to the fluidal structure in the overlying trachyte.

A portion of the first trach3^te has been frequently found scoria-
ceous, but the same phenomenon has not been observed in con-
nection with the second and third trachytes. The last, being so
intimately associated with the ashy strata at the top of the tertiary
formation along the southern coast, is frequently full of fragments
which it picked up at the time of its eruption. Some of the inclu-
sions evidently belong to the first trachyte, but the majority of them
cannot be identified.

The third trachyte is one of the chief topographical determinants
along the southern coast. It forms the bold ridge of Coslou-dagh,
upon which the ruins of ancient Lamponeia are situated. The north-
ern slope of the mountain is gentle, but upon the south it presents
high cliffs towards the sea. At its eastern extremity the trachyte
rests directly upon the upper portion of the tertiary formation. The
strike of the underlying strata is parallel with the general trend of
the mountain, approximately east and west, and the dip is northerly,
corresponding to the fluidal structure in the superimposed trachyte.
The slope of the sheet of trachyte is in some places so gentle, that
it forms a small plateau upon the mountain top. This peculiar
feature furnished an excellent site for a large city, where the exten-
sive Cyclopean walls of Lamponeia are found.

There can be no doubt that the prominent ridge of Coslou-dagh
owes its position to a large dike, and was formed in much the same
manner as Mount Holyoke and Mount Tom of the Connecticut
Valley. West of Behr^m, about eight kilometres, the great plateau
begins and extends to Babd-calessi. Although several varieties of
trachyte are found in that region, the prevailing one closely re-
sembles the third trachyte at Assos, and occurs in extensive dikes,

INVESTIGATIONS AT ASSOS, 1881. 2O9

the gentle dips of which, like that at Coslou-dagh, determine the
existence of the plateau. That the plateau is made up of a series
of dikes, or overflows, which gently dip to the northward, can be
seen upon the plateau itself, where the dikes occasionally form cliffs
facing towards the south, as well as at its eastern extremity, where
they overlie the tilted tertiary strata.

In the vicinity of Demearje^-kioy, about twelve kilometres east of
Behr^, occurs a peculiar light-colored trachyte. Enclosed in the
fine groundmass of this, are numerous glassy crystals of orthoclase,
and some apparently of quartz. The ordinary accessory minerals
are almost entirely wanting.

The trachytes of the Troad are frequently much altered, and it is
often difficult to obtain good hand-specimens. They generally pre-
serve their form, notwithstanding their alteration, and rarely crumble
like the granitic rocks. Of all places where these alterations occur
there is perhaps none more interesting than that found in connec-
tion with the hot springs at Touzla (IVagasae), where the trachytes
have a great variety of bright colors, like the sedimentary rocks
which they have displaced.

The first and second trachytes at Behrkm are among the oldest
in the Troad, and flowed, as has been shown in the first part
of this Report, from a veritable volcanic crater before the close of
the period during which the tertiary strata of the southern coast
were deposited. Later the same trachytes appear to have reached
the surface through long fissures. The third trachyte, which was
erupted through fissures only, was doubtless extruded after the
tertiary strata were deposited, and most probably as one of the
closing events of the period when the land was raised above the
sea level.

CONGLOMERATE.

At many places in the Troad the trachyte is so intimately asso-
ciated with a conglomerate of the same material, that it is scarcely
possible to map the two separately. They are mixed in the most
complicated fashion, and it is often difficult to determine their
relations.

Excellent exposures of the conp;lomerate occur in the cliffs by the
port of Behrim. It is here composed chiefly of cinders apparently

14

210 ARCHAEOLOGICAL INSTITUTE.

fused together into an irregular lumpy mass, as described in the
preceding paper. A similar conglomerate, composed wholly of
red cinders, occurs along the coast about eight kilometres east of
Behr^m, and also to the westward, but is not of common occurrence
elsewhere. Near the small village of Sdnobar, three kilometres
southwest of the ruins of Lamponeia, the coarse fragmental rock
contains, besides scoriated stones, others which are compact, and
quite unlike those occurring in the volcanic conglomerate about
Behram. It is in the gorge of the Touzla, however, by the northern
base of Coslou-dagh, that the finest exposures of this formation are
to be found. It is composed of fragments of all sizes heaped
together indiscriminately, and cemented in some places as if by
fusion. The stones are usually reddish or black, coarse, compact,
and angular, and show no signs whatever of erosion. Cinders are
rare at this outcrop. It forms the steep slopes of the gorge in which
the river fl9ws between the plain of Ivadjfk and that of Behram.
The surface of the rock is extremely rough, and exhibits a marked
tendency to form sharp pinnacles and columns. The dark -colored
fragments are frequently magnetic, and appear to belong to the
basaltic rocks, although the trachytes (so called by all observers in
the Troad) occasionally affect the magnetic needle, and render it
difficult to obtain correct bearings in the ordinary way.

In the high cliffs by Babd-calessi occurs a cindery conglomerate
closely resembling that at Behrim, and appears to rest upon the
trachyte with which it is associated. The same is true in part of
that in the Touzla Valley at the base of Coslou-dagh, but in the
same region also, near the western end of the mountain, the trachyte
distinctly overlies the conglomerate.

Among the mountains north of the Touzla and in the vicinity of
Ivadjik and Sapandje^ there are extensive areas of fragmental rocks,
everywhere intimately associated with the trachytes and the tertiary
strata. Their relation to the latter is in some localities difficult to
discover. The conglomerate occurs at many places, composed of a
great variety of volcanic debris, differing widely in size and weight,
and yet there may not be the slightest trace of stratification. More-
over, in the same neighborhood, at an equal height above the sea,
distinctly stratified l)eds of similar volcanic material, belonging to the
upper part of the tertiary, may be found.

INVESTIGATIONS AT ASSOS, 1881, 21 1

The facts seem to indicate that what has been proved true at
Behr^m may be true also of the whole of the region occupied by the
trachyte, viz. : that the earlier eruptions of trachyte were accompanied
or closely followed by great showers of cinders and ashes. A part of
the fragmental material thrown out from craters or fissures may have
fallen in water and become stratified ; but it seems to be more prob-
able that the land was subsequently submerged and most of the
fine material stratified, while the larger portion of the coarse was not
re-arranged.

The fact that the conglomerate is distinctly overlain by trachyte is
positive evidence that there were eruptions of the latter subsequent
to the formation of at least a part of the former. It is very probable
that the conglomerate is not all of the same age, but nothing has as
yet been observed to indicate that any part of it is younger than the
third trachyte, which forms Coslou-dagh and the plateau south of the
Touzla.

BASALTIC ROCKS.

Rocks belonging to the basalt group are widely distributed in the
Troad, but always occupy comparatively small areas. One of the
largest tracts is between Sazle^ and Demearje^-kioy, about fifteen*
kilometres east of Behrkm. The rock is dark colored, excepting
where considerably weathered, in which case it is yellowish gray. It
has a well-marked columnar structure, and evidently tilted the ad-
joining tertiary limestones at the time of its extnision. Occasionally,
near Houssen-fake^ the rock is cellular, but generally compact, while
near the coast, south of the trachyte which divides this area into two
parts, it is frequently amygdaloidal and of a greenish color. The
amygdules are usually chalcedony, but this substance may be enveloped
in calcite, or the whole amygdule may be calcareous. The greenish
groundmass, sometimes granular, contains numerous small cr)'stals of
feldspar, besides other crystals of dark-colored minerals. The rock is
generally much fractured, and contains many seams of calcite.

The manner in which this basaltic rock has disturbed the adjoining
tertiary strata clearly indicates that the former is younger than the
sedimentary rocks with which it is assoc iated. Its relation to the
trachyte, however, is not easily determined. The trachyte of that

212 ARCHAEOLOGICAL INSTITUTE.

locality is isolated from the great mass further westward, and forms
the rugged hills between KyalAr and Ahmdjah. The hills are ap-
parently composed of large dikes of trachyte, dipping northward and
presenting cliffs towards the sea. Southwest of Demearjee-kioy about
two kilometres, the trachyte, with its usual strike and dip, cuts directly
across the area of basaltic rocks as if it had been forced up through
them in reaching the surface. Moreover, upon the south side of the
trachyte it appears to overlie the basaltic rocks.

Along the coast directly south of the area described, irregular dikes
•of basaltic rocks may be seen penetrating the tertiary strata. The
same phenomena may be observed in the neighborhood of Aracle^,
south of the site of Gargara. Small exposures occur also in the
vicinity of Tactd-kioy (Astyra) and Zytinle^, hear Edremit. At the
former locality the hot springs appear to owe their origin to the
presence of the basaltic rocks from which they rise.

Upon the left bank of the Bahchahle^ River, about fifteen kilo-
metres southeast of EAnedeh, at the head of a plain rises the ma-
jestic hill called Sapandjei-tepeh. It is formed of basaltic rocks
containing numerous small grains of olivine. The columnar structure
in the rock being well developed and nearly vertical, the slopes are
very steep, and for the most part perpendicular cliffs. Upon the east-
em side, however, where the columns are much contorted, the approach
to the summit is not difficult. This prominent hill, rising close to the
river and standing at the head of a fertile plain, must have furnished
an excellent site for an ancient city ; and the traveller is disappointed at
not finding fragments of pottery or ruins upon the summit.

At the southern base of Curshunlou-tepeh, the site of ancient
Kebrene, by the right bank of the M^ndereh, is a small plateau of
basalt containing many small crystals of feldspar and bright grains of
olivine. This area appears to be quite large, extending west across
the river into the hills south of Beiramitch.

The largest exposure, however, which has yet been mapped within
the Troad is between Bundrbashi and the valley of the Kemar (Thym-
brios) River, at the southern end of the Plain of Troy. The rock is
usually compact, containing numerous grains of olivine, but other min-
erals are not prominent. Occasionally it is very cellular and amygda-
loidal. The round and elongated amygdules are of calcite, which forms
also numerous irregular veins. In the valley of the Kemar the basalt

INVESTIGATIONS AT ASS OS, 1881. 213

distinctly overlies about fifteen metres of marly conglomerate and six
metres of red clay, both of which are horizontal, and appear to belong
to the tertiary formation. Near the mouth of the river the same
basalt is overlain by horizontal red clay and shelly limestone, which
appear to be younger than the rock upon which they rest.

While it is evident along the southern coast that the basaltic rocks
are younger than the greater portion of the tertiary strata of that
region, it may be true that they were extruded before the highest beds
of that series were deposited, for the basaltic rocks are not known to
pierce those beds anywhere in the Southern Troad.

SUMMARY.

In briefly summarizing the results derived from the observations
described in this preliminary Report, the rocks of the Troad may
be divided into two groups. The first contains the metamorphic
schists, together with their associated eruptive rocks, the granites
and serpentines. In the second are placed the tertiary strata, the
trachytes, and the basalts. The members of the former are very
ancient and highly altered, while those of the latter are compar-
atively new and fresh. The long interval of time which must have
elapsed between the formation of the sedimentary rocks of the two
groups has no representative among the deposits of aqueous origin
in the Troad, but in other parts of Asia Minor not far distant the
series is more complete.

The oldest rocks of the Troad are an extensive series of coarsely
cr)'stalline limestones interstratified with micaceous and hornblen-
dic schists. They constitute the basis upon which and out of which
the framework of the Trojan peninsula has been developed.

They are the chief mountain-forming strata of that region. The
great mass of Mount -Ida is composed wholly of them, and along
the western coast they give rise to the prominent peak called Sacir-
kyah.

The structure of Mount Ida appears to be a comparatively simple
anticlinal, with so short an axis, extending east and west, that the
upper portion of the mountain is approximately a dome.

The position and distribution of the crystalline schists and lime-
stones indicate that, in the early stages of its development, the

214 ARCHAEOLOGICAL INSTITUTE.

peninsula of the Troad was probably represented by several islands,
which furnished the detritus for subsequent formations.

The extrusion of the peridotic rocks, from which the serpentines
are derived, and the granites occurred some time during the long
interval between the deposition of the metamorphic series and the
beginning of the miocene.

The most important topographical feature formed of the old
eruptive rocks is the peculiar irregularly serrated ridge of Chigri-
dagh, whose rough granitic slopes are the chief landmark in the
Northwestern Troad.

The tertiary strata of the western coast are separated from those
of the interior and the shore of the Gulf of Adramyttion by a broad
belt of trachyte, within which, at intervals, are numerous outcrops of
the same strata extending west to within a short distance of Babd-
calessi. This fact makes it very probable that beneath the sheet of
trachyte which has been spread over the surface of the stratified
rocks, the latter are connected so as to form one great area border-
ing the entire coast of the Troad, and occupying a considerable
portion of its interior.

The occurrence of deposits of lignite at various places through-
out this area, as well as the apparent identity of some of the fossils
and the similar relations of the strata upon both sides to the divid-
ing trachyte, make it probable that the stratified deposits of the
entire area are essentially of the same age. Those along the
shores of the Hellespont have been shown by other observers to
have been deposited in fresh or brackish water during the miocene
period.

The eruption of the trachytes commenced shortly before the
close of the miocene, first, at least in one case, from a crater, and
finally through large fissures. The greatest eruption occurred after
the completion of the miocene deposits, and most likely as one of
the closing events of that period, when the peninsula of the Troad
was, for the first time in its essentially finished form, raised above
the water.

The extrusion of the trachytes was accompanied by great show-
ers of cinders and ashes, which furnished not only the sediment
out of which the upper strata of the miocene were built, but also
the material for the unstratified volcanic conglomerate so intimately
mixed with the trachytes.

INVESTIGATIONS AT ASSOS, 1881. 215

The peculiar drainage of the southern part of the Troad is due to
the great east-and-west dikes of trachyte of which Coslou-dagh and
the plateau south of the Touzla are composed. .

The basaltic rocks were extruded either during the latter part of
the miocene or after its close, and their presence has not materially
modified the topography of the country.

The Halesion Plain, near the mouth of the Touzla^ has been sub-
ject to a considerable change, in the position of its stream, within
the historical period (two thousand years).

University Press : John Wilson and Son, Cambridge.

^^^HV-^

■ iijyiJiiiJiiiiiijyiUiii 9BI

^B 3 bias PIS 'iPB 751 iM^l

PF

k

Y.l
Stanford University Library

Stanford, California

In order that others may use this book,
please return it as soon as possible, but
not later than the date due.

4